Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Idarubicin (IDA¹) and related anthracycline antibiotics are among the most powerful anticancer drugs (e.g., Weiss, 1992; Toffoli et al., 2000). However, the irreversible cardiotoxicity of anthracyclines limits their therapeutic use (e.g., Keefe, 2001). A variety of mechanisms have been suggested to explain the cardiotoxicity of anthracyclines. Thus, evidence has been provided that the intramyocardial formation of secondary alcohol metabolites is an important contributor to cardiotoxicity of anthracyclines like doxorubicin and IDA (Olson and Mushlin, 1990; Cusack et al., 1993; Minotti et al., 1995; Forrest et al., 2000; Minotti et al., 2001). IDA is reduced to its C13-dehydroderivative idarubicinol (IDOL) by cytoplasmic NADPH-dependent aldo-keto or carbonyl reductases (Loveless et al., 1978; Forrest and Gonzalez, 2000). New insights were obtained on cardiac carbonyl reductase by investigating doxorubicin metabolism in an in vitro human heart system (Licata et al., 2000) and in transgenic mice with heart-specific expression of human cardiac carbonyl reductase (Forrest et al., 2000), respectively. Although the evaluation of efficient inhibitors of anthracycline reductases has been suggested as one potential strategy to optimize cardioprotection (Minotti et al., 1998), quantitative information on the effect of carbonyl reductase inhibitors like rutin (Forrest et al., 2000) and phenobarbital (Behnia and Boroujerdi, 1999) on the kinetics of alcohol metabolite formation in the intact heart is lacking.

The amount of IDOL generated from IDA has been previously determined in the isolated perfused rat heart (Platel et al., 1999; Kang and Weiss, 2001). However, the cardiac pharmacokinetics of IDOL (i.e., its formation, transport and disposition in the intact heart, is still poorly understood). One aspect of studies on intact organs is that direct measurements of intramyocardial drug concentrations are difficult to perform; the functional characterization of cardiac IDOL kinetics and its localization requires a mathematical model. We recently developed a compartmental model that is capable of describing the uptake and disposition kinetics of IDA in the isolated rat heart for a 10 min infusion profile (Weiss and Kang, 2002). The concentration-dependent uptake, along with the temperature-dependent transport (Kang and Weiss, 2003) suggested the presence of a specific mechanism that could be described by Michaelis-Menten-type kinetics.

The present study was undertaken 1) to extend this model by inclusion of the production and disposition kinetics of IDOL, 2) to analyze venous outflow curves of IDA and IDOL following infusion of IDA, 3) to investigate the effect of the carbonyl reductase inhibitors rutin and phenobarbital on the conversion of IDA to IDOL in the heart, and 4) to apply this model to previously published data to examine the influence of P-glycoprotein (P-gp) inhibitors verapamil and amiodarone on the myocardial kinetics of IDOL (Weiss and Kang, 2002). The combined model approach uses kinetic analysis of both IDA and IDOL kinetics to estimate the parameters describing the formation and disposition of IDOL in the heart. Since IDOL is more hydrophilic than IDA, its efflux kinetics is of special interest in view of a facilitated retention in myocardium.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. Idarubicin was purchased from Pharmacia and Upjohn (Erlangen, Germany) and rutin and phenobarbital from Sigma-Aldrich (Deisenhofen, Germany). Idarubicinol was kindly donated by Pharmacia and Upjohn. All other chemicals and solvents were of the highest grade available.

Perfused Rat Heart. The experiments were performed by using an isolated perfused rat heart preparation previously described (Kang and Weiss, 2001). Briefly, adult male Sprague-Dawley rats, 300-350g, were anesthetized with sodium pentobarbital (50 mg/kg, intraperitoneally). Following the onset of general anesthesia, heparin (500 IU) was injected into the tail vein, and a cannula is bound into the trachea for ventilation. The chest was opened and an aortic cannula filled with perfusate was rapidly inserted into the aorta to prevent ischemia. Retrograde perfusion was started with an oxygenated Krebs-Henseleit buffer solution, pH 7.4, containing NaCl (118 mM), KCl (4.7 mM), CaCl₂ (2.52 mM), MgSO₄ (1.66 mM), NaHCO₃ (24.88 mM), KH₂PO₄ (1.18 mM), Glucose (5.55 mM), and Na-pyruvate (2.0 mM). It was continueously bubbled with 95% O₂ to 5% CO₂ and maintained at 37°C. The pulmonary artery was incised to allow outflow of the perfusate. Coronary perfusion was initiated through a short cannula in the aortic root and maintained at a constant pressure of 60 mm Hg in a single pass way by the Langendorff technique. The flow was controlled by a roller pump. A latex balloon was placed in the left ventricle and connected to a pressure transducer line. The balloon was inflated with water to create a diastolic pressure of 5 to 6 mm Hg. After stabilization, the system was changed to constant flow condition maintaining a coronary flow of 9.5 ± 0.4 ml/min. The hearts were beating spontaneously at an average rate of 300 beats/min. Coronary perfusion pressure, the left ventricular pressure, and heart rate were measured continuously, and a physiological recording system (Hugo Sachs Elektronik, March-Hugstetten, Germany) was used to monitor left ventricular systolic pressure (LVSP), left ventricular enddiastolic pressure (LVEDP), maximal and minimal values of rate of left ventricular pressure development (LVdP/dt_max and LVdP/dt_min). Left ventricular developed pressure (LVDP) was calculated as LVDP = LVSP  LVEDP. The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (National Institutes of Health Publication 85-23, revised 1996). Prior approval was obtained by the Animal Protection Body of the State of Sachsen-Anhalt, Germany.

Experimental Protocol. After the stabilization for 20 min, 0.5 ml of IDA (1 mg/ml) was infused for 1 min in the absence (n = 5, Kang and Weiss, 2001) and presence of the carbonyl reductase inhibitors, rutin (10 µM, n = 5) or phenobarbital (100 µM, n = 5). The doses of rutin and phenobarbital were below the threshold values that lead to changes in the measured cardiovascular effects. Outflow samples were collected every 10 s for 3 min, every 30 s for the next 7 min, every 60 s for the next 10 min, and every 5 min for the next 60 min (total collection period 80 min). These samples and hearts were assayed for IDA and IDOL by high-pressure liquid chromatography as previously described (Kang and Weiss, 2001).

Modeling. We started data analysis using the compartmental model, which describes IDA disposition for the 10-min infusion experiment (Weiss and Kang, 2002) to the outflow data of IDA measured for the injection of 0.5 mg IDA in 1 min. Since only the kinetic data were fitted, the model could be simplified by replacing the (intracellular) saturable transport process from Compartment 2 to Compartment 3 by a passive one [i.e., rate constant k₂₃ (Kang and Weiss, 2003)]. The structure of the model and the parameters accounting for the disposition of IDA were then fixed in fitting an extended model to the outflow concentration profile of the formed IDOL. Thereby, we used the ADAPT II software package (D'Argenio and Schumitzky, 1997) and a modeling methodology that has been described in detail previously (Weiss and Kang, 2002). Since especially for nonlinear systems the information content of one experiment may be insufficient to resolve the parameters into a unique set of most probable values, we took advantage of the fact that three data sets stem from different but related experiments, the modeling function of which shares one or more parameters. Thus, simultaneous nonlinear regression was performed using three data sets, the average data of the control group (Kang and Weiss, 2001) and those of the rutin and phenobarbital groups. The parameters described the kinetics in the control group and factors f_i accounted for a potential change in parameters P_i due to the treatments (i.e., in the model of the treatment groups, model parameters P_i were replaced by f_i P_i). All possible combinations of factors were tested with the aim to describe treatment groups by a minimum number of factors (i.e., free model parameters). The model selection was made according to the following criteria. Any model showing a noninvertible Fisher's information matrix was discarded as nonidentifiable (Landaw and DiStefano, 1984). The comparison between the different fittings to a set of data were performed by means of visual examination of the distribution residuals and Akaike Information Criterion (Ludden et al., 1994) and fractional standard deviations of parameter estimates (FSD). We considered an FSD > 50% to indicate that a parameter has not been estimated with sufficient statistical certainty.

(1)

(2)

(3)

(4)

(5)

(6)

View larger version (18K): [in this window] [in a new window]   Fig. 1.   The kinetic model for the description of cardiac formation and disposition of IDOL generated from IDA (indicated in bold). First order rate constants are denoted by kij (k1m and k2m are the formation rate constants of metabolite). The Michaelis-Menten like transport processes are characterized by parameters Vmax,ij and Km,ij.

(7)

Statistics. The outflow data are presented as mean ± S.D. Statistical significance was assessed with an analysis of variance followed by Student-Newman-Keuls test using the SigmaStat program (SPSS Science Inc., Chicago, IL). A value of P < 0.05 was considered statistically significant. The likelihood ratio test (Huet et al., 1996) was used to determine the significance of parameter changes in the nested models due to the presence of metabolism inhibitors (rutin and phenobarbital) or P-gp inhibitors (verapamil and amiodarone).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Kinetics of IDOL Formation and Disposition. The outflow concentration of IDOL generated by the heart after the 1-min infusion of 0.5 mg of IDA was much lower (peak approximately 3 orders of magnitude) than that of the parent drug (Fig. 2, A and B). The IDOL outflow profiles display a characteristic double peak shape; after the initial peak at the end of IDA infusion, the curves decay rapidly within 10 s before reaching the second peak at about 10 min (Fig. 3C). This behavior implies the assumption of an additional site of metabolism (Compartment IDA₁) besides the expected formation of IDOL in cardiomyocytes (Compartment IDA₂). The minimal model (Fig. 1) allowed a reasonable fit of IDA and IDOL outflow data (Fig. 3, A and C). As previously shown for IDA (Weiss and Kang, 2002; Kang and Weiss, 2003), the re-uptake of IDOL is characterized by a saturable, Michaelis-Menten type process (maximal transport rate V_max,IDOL, 3.2 nmol/min; apparent Michaelis constant K_M,IDOL, 0.24 nmol). The parameter estimates are summarized in Table 1.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Panels A and B, IDA and IDOL outflow profiles in hearts for a 1 min infusion of 0.5 mg of IDA in the presence of phenobarbital (100 µM) and rutin (10 µM) compared with control (mean ± S.D., n = 5 in each group). Panels C and D, recovery of IDA and IDOL in perfusate and amount in heart at 80 min in hearts perfused with buffer (control), phenobarbital and rutin (mean ± S.D., n = 5; *, P < 0.05; **, P < 0.01, compared with control). Control data from a previous publication (Kang and Weiss, 2001).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Model fits for the mean IDA and IDOL outflow profiles for a 1 min infusion of 0.5 mg IDA of the control, phenobarbital and rutin group obtained by simultaneous nonlinear regression. A and B, IDA (expressed on a log-scale); C and D, IDOL.

Effects of Rutin and Phenobarbital on Pharmacokinetics of IDA and IDOL. In the presence of metabolic inhibitors, the outflow concentration of IDA and IDOL is reduced during the terminal washout phase (Fig. 2, A and B). The average outflow recoveries and residual amounts in the heart at 80 min for IDA and IDOL in the absence and presence of rutin (10 µM) or phenobarbital (100 µM) are depicted in Fig. 2, C and D, respectively. Rutin increases the residual amount of IDA in the heart at the end of experiment ~1.4-fold (P < 0.05). Phenobarbital and rutin induced a decrease in residual amount of IDOL to 47% (P < 0.05) and 32% (P < 0.05) of control, respectively. Rutin also decreased IDOL recovery into perfusate by 69% (P < 0.01).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Model simulation of the effect of phenobarbital and rutin on the time course of total amount of IDOL in the heart (solid line) and the amount sequestered into Compartment 3 (dashed line). Note that the predicted amount of IDOL also contains its metabolites formed in myocard.

IDOL Pharmacokinetics after 10-min Infusion of IDA: Effects of Verapamil and Amiodarone. Figure 5, A and B, shows the fits of average outflow concentration-time profiles for IDOL obtained from a 10-min infusion of IDA (0.5 mg) in the absence (n = 5) and presence of verapamil (1 nM) and amiodarone (1 µM, n = 5), respectively. As for the 1-min infusion, the model (Fig. 1) allowed a reasonable prediction of the outflow concentration of the generated IDOL. Note that the peak concentration of IDOL (Fig. 5A) is by a factor of 0.004 less than that of IDA (Weiss and Kang, 2002). A comparison with Fig. 3C shows that the initial peak becomes less pronounced if the input rate decreases; it nearly disappears for 10-min infusion of IDA (Fig. 5A). Both verapamil and amiodarone lead to a parallel downward shift. As previously shown for IDA, the saturable re-uptake of IDOL is also enhanced by P-gp inhibitors: verapamil (Fig. 5A) and amiodarone (Fig. 5B) increase V_max,IDOL by a factor of 1.6 (P < 0.01) and 1.2, respectively. The model parameters characterizing the formation and disposition of IDOL in the 10-min experiments are listed in Table 2.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Model fits for the mean IDOL outflow profiles for a 10 min infusion of 0.5 mg of IDA of the control, verapamil, and amiodarone group obtained by simultaneous nonlinear regression. The corresponding IDA data have been published previously (Weiss and Kang, 2002).

View this table: [in this window] [in a new window]   TABLE 2 Parameter estimates from simultaneous fitting of mean outflow IDOL data after 10-min infusion of IDA in hearts of control, verapamil-(1 nM), and amiodarone- (1 µM) treated groups The corresponding IDA data have been published previously (Weiss and Kang, 2002).

Effect of Rutin and Phenobarbital on IDA Pharmacodynamics. Rutin (10 µM) and phenobarbital (100 µM) do not show any significant effect on IDA-induced cardiac performance (including its negative inotropism) except a change in coronary vascular resistance (CVR), which is depicted in Fig. 6. The maximal vasoconstrictive effect of IDA is significantly reduced by rutin (22.3%, P < 0.05) and also the secondary increase in CVR during the washout phase is inhibited (P < 0.05). Phenobarbital, in contrast, potentiates this secondary phase of IDA induced vasoconstriction (nearly 2-fold at 80 min, P < 0.05)

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effect of phenobarbital and rutin on the IDA-induced increase in coronary vascular resistance CVR (mean ± S.D., n = 5 in each group).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Pharmacokinetics. From the model independent results it follows 1) that about 2 and 8% of the IDA dose were metabolized to IDOL in the 1- and 10-min infusion experiments, respectively, and 2) that the reduction of the residual amount of IDOL in the presence of 100 µM phenobarbital was greater than the effect of 10 µM rutin (Fig. 2D). It is conceivable that because of nonlinear (concentration-dependent) cardiac uptake of IDA, the formation of IDOL is reduced when the same IDA dose is infused with higher rate (leading to higher input concentrations). Note that the total recovery is not complete since the formation of nonpolar products, aglycones, represents a second major metabolic pathway of IDA metabolism (about 30% of dose not recovered as IDA) (Loveless et al.,1978)

Pharmacodynamics. Here we are concerned with the effects of rutin and phenobarbital on the IDA-induced change in CVR. It has been suggested that the vasoconstriction produced by IDA may be due to an inhibition of nitric oxide (NO) synthesis (see Kang and Weiss, 2001 and references cited therein). In principle, our observation that rutin reduced the IDA-induced increase in CVR and nearly abolished the secondary rise in CVR during the washout phase (Fig. 6) is in accordance with the coronary vasodilator effects of this compound (Schussler et al., 1995). However, the fact that in the present study rutin was used in a concentration that did not affect baseline coronary flow and that the IDA-mediated vasoconstriction has been attributed both to a reduction of nitric oxide production and increased inactivation by the formed superoxide (Garner et al., 1999) suggests that this effect of rutin can be explained by its free radical scavenging action. This mechanism was also supposed for the reversal by melatonin of the doxorubicin-induced coronary vasoconstriction (Liu et al., 2002). The potentiation by phenobarbital of the secondary vasoconstrictor response, on the other hand, could be explained by an inhibition cGMP-mediated endothelium-dependent and independent vasorelaxations (Gerkens, 1987; Terasako et al., 1994). Interestingly, the initial rapid vasoconstriction (which parallels the IDA outflow concentration profile) was not affected by phenobarbital. In view of the low fraction of IDOL (2%) generated from IDA, a significant contribution of IDOL to the pharmacodynamic effects of IDA is very unlikely in these experiments. Thus, the observed effects of rutin and phenobarbital on IDA pharmacodynamics cannot be explained by their inhibitory action on cardiac IDOL formation. Notably, no influence of rutin and phenobarbital on the time course of IDA-induced negative inotropism could be detected.