Abstract
The clinically approved cardioprotective agent dexrazoxane (ICRF-187) and two of its hydrolyzed metabolites (a one-ring open form of dexrazoxane and ADR-925) were examined for their ability to protect neonatal rat cardiac myocytes from doxorubicin-induced damage. Dexrazoxane may protect against doxorubicin-induced damage to myocytes through its strongly metal-chelating hydrolysis product ADR-925, which could act by displacing iron bound to doxorubicin or chelating free or loosely bound iron, thus preventing site-specific iron-based oxygen radical damage. The results of this study showed that whereas dexrazoxane was able to protect myocytes from doxorubicin-induced lactate dehydrogenase release, neither of the metabolites displayed any protective ability. Dexrazoxane also reduced apoptosis in doxorubicin-treated myocytes. The ability of dexrazoxane and its three metabolites to displace iron from a fluorescence-quenched trapped intracellular iron-calcein complex was also determined to see whether the metabolites were taken up by myocytes. Although ADR-925 was taken up in the absence of calcium in the medium, in the presence of calcium, its uptake was greatly slowed, presumably because it formed a complex with calcium. Both of the one-ring open metabolites were taken up by myocytes and displaced iron from its complex with calcein. These results suggest either that the anionic metabolites do not have the same access to iron pools in critical cellular compartments, that their uptake is slowed in the presence of calcium, or, less likely, that dexrazoxane protects by some other mechanism.
Dexrazoxane (ICRF-187; Zinecard; Fig. 1) is clinically used to reduce doxorubicin-induced cardiotoxicity (Hasinoff, 1998; Hasinoff et al., 1998). There is now considerable evidence to indicate that this toxicity may be due to iron-dependent oxygen free radical formation (Meyers, 1998) on the relatively unprotected cardiac muscle. Dexrazoxane likely acts by diffusing into the cell and hydrolyzing (Hasinoff, 1994a,1994b, 1998; Hasinoff et al., 1998) to its ring-opened metal ion-binding form, ADR-925 (Fig. 1), which has a structure similar to that of EDTA. ADR-925 may then either remove iron from the iron-doxorubicin complex (Buss and Hasinoff, 1993) or bind free or loosely bound iron, thus preventing oxygen radical formation. ADR-925 binds Fe2+ and Fe3+ with formation constants of 1010 and 1018.2 M–1, respectively, compared with formation constants of 1014.3 and 1025.1 M–1 for EDTA (Huang et al., 1982; Diop et al., 2000). Thus, dexrazoxane, which is easily permeable to cells (Dawson, 1975), can be considered a neutral prodrug analog of EDTA.
We previously showed in isolated hepatocytes (Hasinoff et al., 1994), in a rat model (Schroeder and Hasinoff, 2002), and in humans (Schroeder et al., 2003) that dexrazoxane was rapidly metabolized to the one-ring open compounds B and C, and ADR-925 (Fig. 1). This metabolism may be due, in part, to the ability of dihydropyrimidinase, which is present in the liver and kidney, but not the heart, to enzymatically hydrolyze dexrazoxane to B and C, but not to ADR-925 (Hasinoff et al., 1998). We also recently showed that dihydroorotase, which is present in all three of these organs, enzymatically hydrolyzes B and C to ADR-925 but does not act on dexrazoxane (Schroeder et al., 2002). We also previously used a neonatal cardiac myocyte model to compare the ability of dexrazoxane and deferiprone to protect against doxorubicin-induced cytotoxicity (Barnabé et al., 2002). Given the rapid appearance of these metabolites in plasma, we decided to examine whether these metabolites could prevent doxorubicin-induced damage to isolated neonatal rat myocytes and, thus, whether they might be useful as drugs. To accompany these studies, we also investigated the uptake of dexrazoxane and its metabolites in myocytes by following the fluorescence dequenching of an intracellular iron-calcein complex (Barnabé et al., 2002; Esposito et al., 2002; Hasinoff, 2002; Kakhlon and Cabantchik, 2002).
Materials and Methods
Materials. DF-15 medium [with 7.5% (v/v) fetal bovine serum and 7.5% (v/v) horse serum], fetal bovine and horse serum, penicillin, streptomycin, and fungizone were obtained from Invitrogen (Burlington, ON, Canada). Trypsin, collagenase, and DNase were from Worthington Biochemicals (Freehold, NJ). The black plastic 96-well plates with clear bottoms used in the fluorescence plate reader studies were obtained from Corning Glassworks (Corning, NY). Unless specified, other reagents were obtained from Sigma-Aldrich (Oakville, ON, Canada). Stock calcein (Molecular Probes, Eugene, OR) solutions were prepared in water and stored dark in the cold over the chelating resin Chelex (0.2 g/ml), which was found to greatly increase its stability. HBS buffers were likewise stored over Chelex to reduce levels of adventitious iron. Metabolites B and C were prepared from NaOH-hydrolyzed dexrazoxane and separated by reverse-phase C18 and then ion-exchange column chromatography as described (Schroeder et al., 2002). As determined by high-performance liquid chromatography, B and C contained less than 1.0 and 1.6 mol% ADR-925, respectively. The ApoAlert DNA fragmentation assay TUNEL kit was from BD Biosciences Clontech (Palo Alto, CA). Nonlinear least-squares curve fitting was done with SigmaPlot (SPSS Science, Chicago, IL). The errors shown were least-squares calculated S.E. values. Where significance is indicated, an unpaired t test was used with p < 0.05 considered as significant.
Myocyte Isolation and Culture. Ventricular myocytes were isolated from 1- to 3-day-old Sprague-Dawley rats as described (Barnabé et al., 2002; Hasinoff, 2002). Briefly, minced ventricles were serially digested with collagenase and trypsin in Dulbecco's phosphate-buffered saline /1% (w/v) glucose at 37°C in the presence of DNase and preplated in large Petri dishes to remove fibroblasts. The preparation, which was typically greater than 90% viable by trypan blue exclusion, yielded an almost confluent layer of uniformly beating heart myocytes by day 2. For the LDH release experiments, the myocyte-rich supernatant was plated on day 0 in 24-well plastic culture dishes (5 × 105 myocytes/well, 750 μl/well) in DF-15 to yield an almost confluent layer of beating heart myocytes by day 2. On days 2 and 3, the medium was replaced with 750 μl of fresh DF-10 containing 10% (v/v) fetal bovine serum. To lower the background LDH levels, on day 4, 24 h before the drug treatments, the medium was changed to DF-2 and again on day 5 just before the addition of drugs. The animal protocol was approved by the University of Manitoba Animal Care Committee.
LDH Determination and Drug Treatments. LDH released into the myocyte growth medium was determined as previously described (Barnabé et al., 2002; Hasinoff, 2002). In all experiments at the end of the 3-h doxorubicin treatment, the myocytes were washed (two 20-min incubations followed by replacement with fresh medium) with fresh DF-2 medium containing 0, 20, or 100 μM dexrazoxane, B, or ADR-925, respectively, and maintained in this medium for a further 72 h. Starting on day 6 after plating, samples (80 μl) of the myocyte supernatant were collected every 24 h for 3 days after treatment with doxorubicin and/or dexrazoxane, B, or ADR-925. The samples were frozen at –80°C and analyzed within 1 week. After the last supernatant sample was taken, the myocytes were lysed with 250 μl of 1% (v/v) Triton X-100/2 mM EDTA/1 mM dithiothreitol/0.1 M phosphate buffer (pH 7.8) for 20 min at room temperature. The total cellular LDH activity, from which the percentage of LDH was calculated, was determined from the activity of the lysate plus the activity of three 80-μl samples previously taken. The LDH activity was determined in quadruplicate, in a kinetic assay in a 96-well plate in a Molecular Devices Corp. (Sunnyvale, CA) plate reader. The initial velocity of the LDH-catalyzed reaction of NAD+ with lactate to produce NADH and pyruvate was determined by measuring the rate of increase in absorbance at 340 nm at 25°C. The assay buffer contained 2.4 mM NAD+ and 290 mM sodium lactate in 28 mM Tris buffer (pH 8.8).
Calcein Loading of Myocytes and Displacement of Iron from the Fluorescence-Quenched Intracellular Calcein-Iron Complex by Dexrazoxane, B, C, and ADR-925. Calcein was loaded into attached myocytes 6 to 10 days after plating in 96-well plates (125,000 myocytes/well, 200 μl of medium/well), essentially as described (Cabantchik et al., 1996; Zanninelli et al., 1997; Barnabé et al., 2002; Hasinoff, 2002). Briefly, myocytes were incubated with 125 nM calcein-AM (Molecular Probes), the cell-permeant acetoxymethyl ester of calcein, for 5 min at 37°C in serum-free DF-0 medium, followed by three changes of medium at room temperature to remove extracellular ester. The kinetic fluorescence measurements were conducted on a BMG Labtechnologies Inc. (Durham, NC) Fluostar Galaxy fluorescence plate reader (λex of 485 nm, λem of 520 nm, 30°C) equipped with excitation and emission probes directed to the bottom of the plate. To reduce background fluorescence from DF-0 medium, the medium was changed to HBS buffer (100 μl) a couple of minutes before the addition of the drug. After initial baseline fluorescence intensity data were collected, dexrazoxane or its metabolites were added to the attached myocytes by pipetting stock drug solution into the well and gently mixing the solution with a pipette. The increase in fluorescence as the drug displaced iron from the fluorescence-quenched trapped intracellular calcein-iron complex was recorded relative to untreated controls as a function of time. The initial velocities (v) for the fluorescence change occurring upon the addition of chelators were calculated by linear least-squares fits of the fluorescence-time data over the first 20 min for dexrazoxane and 50 min for the metabolites. The neutral iron chelator deferiprone (Fig. 1), which rapidly enters cells and displaces iron from its complex with calcein, was used a positive control (Zanninelli et al., 1997; Barnabé et al., 2002).
Solution Kinetics of the Reaction of ADR-925 and B with Fe2+-Calcein. The displacement of Fe2+ from its complex with calcein by ADR-925 and B was followed in the fluorescence plate reader at 30°C in HBS buffer containing 1 mM ascorbic acid and 0.1 mg/ml catalase. Because the Fe2+-calcein complex is rapidly oxidized to Fe3+-calcein under aerobic conditions (Breuer et al., 1995), ascorbic acid was used to maintain the iron in its ferrous state. The catalase was used to decompose H2O2 formed from oxidation of Fe2+-calcein and prevent the oxidation of calcein to a nonfluorescent product (Petrat et al., 2002). The Fe2+-calcein complex (1 μM) was formed by adding FeSO4 to calcein in the well of the 96-well plate, mixing, and waiting 10 s for the reaction to complete. The chelator was then added, and the fluorescence was followed with time. The initial velocities were calculated from the first 0.7 to 1.5 min of the reaction.
TUNEL Assay of Doxorubicin- and Dexrazoxane-Treated Myocytes, Epifluorescence Microscopy, and Image Analysis. To determine whether doxorubicin treatment of myocytes induced apoptosis, a TUNEL assay was carried out along with a nuclear propidium iodide counterstain to determine the percentage of apoptotic cells as per the manufacturer's directions. The terminal deoxynucleotidyl transferase catalyzes incorporation of fluorescein-dUTP at the free 3′-hydroxy ends of fragmented DNA. During late-stage apoptosis, cellular endonucleases cleave DNA between nucleosomes. The wet-mounted fixed RNase and anti-fade-treated cells were imaged on a Zeiss Axioscop 2 MOT epifluorescence microscope with fluorescein and propidium iodide filter sets, respectively. The percentage of TUNEL-positive cells in 20 randomly chosen fields (approximately 125 cells/field) per treatment was determined from counting the number of TUNEL-positive cells in the green plane relative to the total number of cells in the red plane.
Results
Doxorubicin-Induced LDH Release from Cardiac Myocytes Preincubated with Dexrazoxane, B, and ADR-925. As shown in Fig. 2, a to c, a 3-h treatment with 1.5 μM doxorubicin, which was followed by washing doxorubicin off and maintaining the myocytes in doxorubicin-free medium, resulted in a significant increase in the cumulative amount of the LDH released, compared with untreated myocytes at all three times up to 72 h. The release of the cytosolic enzyme LDH from myocytes is commonly used as a measure of doxorubicin and other drug-induced damage (Hershko et al., 1993; Barnabé et al., 2002). The data of Fig. 2a also show that preincubation of the myocytes with either 20 or 100 μM dexrazoxane for 3 h before doxorubicin treatment resulted in a significant reduction in doxorubicin-induced LDH release similar to what we observed before (Barnabé et al., 2002). The data of Fig. 2, b and c, show that preincubation of the myocytes with either 20 or 100 μM B or ADR-925 for 3 h before doxorubicin treatment did not result in any significant decrease in doxorubicin-induced LDH release. In similar experiments carried out using 0.6 μM doxorubicin, preincubation with either 20 or 100 μM B also did not reduce doxorubicin-induced LDH release (data not shown).
Kinetics of Displacement of Intracellular Iron from Its Complex with Calcein by ADR-925, Metabolites B and C, and Dexrazoxane in Myocytes. As shown in Fig. 3a, the addition of various concentrations of ADR-925 to attached calcein-loaded myocytes in HBS resulted in increases in fluorescence intensity consistent with the removal of iron from the trapped intracellular iron-calcein complex. Upon the addition of 50 to 1000 μM ADR-925, the initial rate of the fluorescence change increased with an increase in the ADR-925 concentration as seen previously (Hasinoff, 2002). A plot of the initial velocities (v) as a function of ADR-925 concentration (Fig. 3b) suggested that ADR-925 displacement of iron from its complex with calcein was a saturable process. A nonlinear least-squares fit of the initial velocity data v to the Michaelis-Menten equation, v = Vmax · [ADR-925]/(1 + Km · [ADR-925]) gave a Km of 103 ± 26 μM.
Because ADR-925 binds Ca2+ and Mg2+ with formation constants of 106.9 and 105.1 M–1 (Huang et al., 1982), respectively, in vivo, in plasma the ADR-925 is likely complexed to Ca2+ and to Mg2+, to a lesser extent. Thus, to determine the effect of Ca2+ and Mg2+ in the medium on the ADR-925-induced fluorescence dequenching of the intracellular iron-calcein complex in attached myocytes, the experiments were repeated in the presence of Ca2+ and Mg2+ at concentrations typically found in plasma (Sheppard and Kontoghiorghes, 1993). As can be seen from Fig. 4a, the initial rate of calcein fluorescence dequenching by 500 μM ADR-925 was greatly decreased (6.6-fold) by the addition of 2.5 mM Ca2+ to the HBS medium, but not by the addition of 1.0 mM Mg2+. A control experiment (Fig. 4b) was carried out with deferiprone, which has a low affinity for Ca2+ and Mg2+ (Sheppard and Kontoghiorghes, 1993). As previously seen, deferiprone in HBS caused a rapid fluorescence dequenching of calcein (Barnabé et al., 2002). However, in contrast to the ADR-925 results, in the presence of a mixture of Ca2+ and Mg2+, the deferiprone-induced fluorescence dequenching was only slightly affected. Similar results with both ADR-925 and deferiprone were seen with calcein-loaded myocytes in Hanks' buffer containing 1.25 mM Ca2+, 0.81 mM Mg2+, 20 mM HEPES (pH 7.4). Probenecid is well known to inhibit the uptake of organic anions into cells (Tsuji et al., 1990). To determine whether uptake of dianionic ADR-925 (Fig. 1) was inhibited by probenecid, ADR-925 fluorescence dequenching of calcein-loaded myocytes was determined in the absence and presence of 1 mM probenecid. The prior addition of probenecid to HBS had no effect on ADR-925-induced fluorescence dequenching of myocytes.
Because significant levels of the intermediate metabolites B and C are found in plasma after the administration of dexrazoxane to both rats (Schroeder and Hasinoff, 2002) and humans (Schroeder et al., 2003), the ability of B and C to displace iron from the trapped intracellular iron-calcein complex were also examined. As seen from Fig. 5, b and c, both B and C were able to dequench calcein fluorescence in myocytes in a concentration-dependent manner, although at rates that were, respectively, 5.1- and 3.9-fold lower than ADR-925 at a 500 μM drug concentration. The rate at which dexrazoxane dequenched calcein fluorescence (Fig. 5a) was much lower than either B, C, or ADR-925 (maximally 7.4-fold less) and, as previously seen, did not measurably vary with the dexrazoxane concentration (Hasinoff, 2002).
Solution Kinetics of the Displacement of Fe2+from Its Complex with Calcein by ADR-925 and Metabolite B. The kinetics of the displacement of Fe2+ from its complex with calcein by ADR-925 and B were studied in solution under reducing conditions, as would exist in the cell, to see whether these reactions might be rate-limiting and, thus, whether they could affect the observed rate of fluorescence dequenching observed in myocytes. As shown in Fig. 6a, upon the addition of ADR-925 to Fe2+-calcein, a rapid (t1/2 of 2.5 min at 100 μM chelator) increase in fluorescence resulted, due to the displacement of Fe2+ from its complex with calcein. Under conditions where ADR-925 was present in large excess, the fluorescence increased exponentially with time (Fig. 6a). To obtain kobs, the observed pseudofirst-order rate constant for the reaction, the observed fluorescence changes were nonlinear least-squares fitted to a three-parameter exponential equation: in which F is the fluorescence at time t, ΔF is the total change in fluorescence, and F0 is the fluorescence at time 0. The plots of kobs as a function of the ADR-925 concentrations are shown in Fig. 6b and increase with an increase in chelator concentration. However, the reaction of B with Fe2+-calcein was much slower (t1/2 of 10 min at 100 μM B) and could not be accurately fit to eq. 1 because it displayed both a fast initial phase and a slower second phase. To obtain a kobs for the first initial reaction, kobs was calculated from v/ΔF, which assumed that the fast initial reaction was a first-order process. The values of kobs initially increased and then leveled off at about 100 μM B (Fig. 6b). In the presence of 2.5 mM Ca2+ in the medium, the kobs for the reaction of ADR-925 and B with Fe2+-calcein decreased from 1.4- to 3.4-fold, between 100 and 1000 μM ADR-925. However, the presence of Ca2+ in the medium increased the rate at which B displaced Fe2+ from its complex with calcein. The kobs increased from 3.0- to 4.5-fold, between 50 and 1000 μM B. A kobs was also determined at 200 μM EDTA in the absence of any added Ca2+. The kobs for EDTA under these conditions was 0.53 min–1 compared with 0.43 min–1 for ADR-925.
TUNEL Assay of Doxorubicin- and Dexrazoxane-Treated Myocytes. To determine whether doxorubicin induced apoptosis in myocytes and whether dexrazoxane was able to prevent doxorubicin-induced apoptosis, a TUNEL assay was carried out on myocytes. As shown in Fig. 7, myocytes that were treated with doxorubicin for 3 h, washed, and TUNEL-assayed 72 h later showed a significant (p < 0.001) increase in the proportion of TUNEL-positive myocytes compared with untreated controls. Myocytes that were pretreated with 100 μM dexrazoxane for 3 h, then treated with doxorubicin for a further 3 h, and washed with medium containing dexrazoxane showed a significant (p < 0.001) reduction in the proportion of TUNEL-positive myocytes compared with doxorubicin-treated myocytes. Thus, dexrazoxane acted as an antiapoptotic agent. Dexrazoxane was not able to reduce the proportion of TUNEL-positive myocytes to control values (p < 0.001). This result is consistent with the LDH results of Fig. 2a in which LDH levels of dexrazoxane-treated myocytes did not reach levels of control myocytes. Dexrazoxane-treated myocytes also showed a significant (p < 0.001) increase in the proportion of TUNEL-positive myocytes compared with untreated controls. This result might be due to the fact that dexrazoxane is also a strong inhibitor of DNA topoisomerase II (Hasinoff et al., 1995) that can induce apoptosis in K562 cells (Hasinoff et al., 2001). The high percentage of TUNEL-positive control apoptotic myocytes is likely a result of the collagenase/trypsin tissue dissociation treatment used to obtain the primary cell culture.
Discussion
The concentrations of dexrazoxane and its metabolites that were used in this study were well within pharmacological levels achieved when it is used as a doxorubicin cardioprotective agent (Hochster et al., 1992; Schroeder et al., 2003). The peak plasma concentrations of dexrazoxane, ADR-925, B, and C in humans dosed at 1500 mg/m2 are 211, 30, 19, and 9 μM, respectively (Schroeder et al., 2003). The concentrations of doxorubicin used in these studies were even smaller (1.5 compared with 12 μM) than the plasma concentrations seen clinically at the end of a 60 mg/m2 doxorubicin infusion period (t1/2β of 1.8 h) (Hochster et al., 1992). Thus, the doxorubicin concentrations and the time over which the myocytes were exposed to doxorubicin were also in a pharmacologically relevant range.
The rapid rate of displacement of Fe2+ from its complex by calcein by ADR-925 in solution (Fig. 6b) in either the presence or absence of Ca2+ suggests that this reaction was not likely to be rate-limiting when ADR-925 was added to calcein-loaded myocytes. However, the slower rate at which B displaced Fe2+ (Fig. 6b) suggests that this reaction partly limited the kinetics of fluorescence dequenching in calcein-loaded myocytes in the absence of added Ca2+ (Fig. 5b). We previously showed that B and C are also chelators that are capable of displacing Fe3+ from its complex with doxorubicin (Buss and Hasinoff, 1993). The similarity in the kobs values for the ADR-925 and EDTA displacement reactions in solution suggests that there was a common rate-determining step that was most likely the breaking of an Fe2+-calcein bond. The Ca2+-induced decrease in rate of displacement of Fe2+ from its complex with calcein in solution is likely due to the requirement that the Ca2+-ADR-925 complex dissociate before it can complex Fe2+ released from calcein. We previously showed that Ca2+ binds to B (Buss and Hasinoff, 1997), although likely much less strongly than to ADR-925. The reason for the Ca2+-induced increase in rate of reaction of B with Fe2+-calcein is not known but may be due to a faster dissociation of the weaker Ca2+-B complex.
Our recent pharmacokinetic studies showed that dexrazoxane was rapidly metabolized in rats (Schroeder and Hasinoff, 2002) and in humans (Schroeder et al., 2003) to the one-ring open compounds B and C and ADR-925 (Fig. 1). Given the rapid metabolism of dexrazoxane in vivo, we tested the hypothesis that circulating B or ADR-925 in plasma might be the active form of dexrazoxane that protects myocytes from doxorubicin-induced LDH release. However, as shown in Fig. 2, neither B nor ADR-925 protected myocytes against doxorubicin. The inability of ADR-925 to protect myocytes from doxorubicin-induced damage is in accord with the inability of ICRF-198 (the racemic form of ADR-925) to protect hamsters from acute toxic effects of the doxorubicin analog daunorubicin (Herman et al., 1985). The reason that both B and ADR-925 do not protect may be due to their being anionic species that enter cells more slowly than neutral dexrazoxane does (Dawson, 1975).
To determine whether the lack of uptake of these anionic metabolites prevented them from protecting myocytes, B-, C-, and ADR-925-induced fluorescence dequenching of the iron-calcein complex as they were taken up in myocytes was determined. However, the results of Fig. 3 show that there was a saturable uptake of ADR-925 into myocytes. The fact that ADR-925 uptake was saturable suggests that dianionic ADR-925 may have been taken up by an anion transport system that was, however, not inhibited by probenecid. However, in the presence of Ca2+ concentrations similar to those in plasma, the rate of entry of ADR-925 was greatly reduced. This probably occurred because ADR-925 can form a complex with Ca2+ (K of 106.9 M–1; Huang et al., 1982). Although the presence of 2.5 mM Ca2+ did decrease the rate of the reaction of ADR-925 with Fe2+-calcein in solution, this decrease was not wholly consistent with the large decrease in rate of fluorescence dequenching of calcein in myocytes that was caused by the addition of Ca2+ to the medium (Fig. 4a). However, the reduced rate of uptake of the Ca2+-ADR-925 complex may, in part, be responsible for the inability of ADR-925 to protect myocytes from doxorubicin.
The results of Fig. 5 show that B and C were taken up by myocytes more rapidly than ADR-925. However, the uptake of B did not result in its protecting myocytes from doxorubicin (Fig. 2). Together, these results suggested that these anionic metabolites did not have the same access that dexrazoxane had to iron pools in critical cellular compartments. Doxorubicin has been shown by fluorescence microscopy to localize in the mitochondria of myocytes (Swift and Sarvazyan, 2000) with doxorubicin-induced dichlorofluorescin oxidation occurring close to the mitochondria (Sarvazyan, 1996). Cardiac mitochondria are also a prominent site of injury by doxorubicin (Gianni et al., 1983; Sokolove and Shinaberry, 1988). We have shown that dexrazoxane reduces doxorubicin-induced oxidation of intracellular dichlorofluorescin and prevents mitochondrial damage (Hasinoff et al., 2003). Thus, dexrazoxane may be protecting mitochondria because it is neutral and permeable to mitochondria, whereas the anionic metabolites B and ADR-925 are not. The intracellular Ca2+ concentration is so low that the ADR-925 would not be present as its neutral Ca2+-ADR-925 complex. The calcein in myocytes is located mainly in the cytoplasm and, thus, our experiments do not give us any information on the displacement of iron from the mitochondria. The results of Fig. 7 showing that dexrazoxane reduced doxorubicin-induced apoptosis are consistent with its ability to prevent daunorubicin-induced apoptosis of myocytes (Sawyer et al., 1999).
An alternative, although less likely possibility, is that dexrazoxane did not act by preventing iron-based oxidative damage, but through its ability to catalytically inhibit DNA topoisomerase II (Ki 13 μM) (Hasinoff et al., 1995). It is, however, unclear how preventing doxorubicin from targeting topoisomerase II could protect myocytes. Dexrazoxane is clearly able to prevent oxygen free radical damage inasmuch as it has been shown that dexrazoxane can protect myocytes from hypoxia-reoxygenation and other drug-induced free radical damage (Hasinoff, 2002; Hasinoff et al., 1998).
The fact that dexrazoxane only very slowly removed iron from its complex with calcein in myocytes (Fig. 5a) is explained by the fact that at the pH and temperature of this study, it is only slowly hydrolyzed to its one-ring open intermediates B and C (Fig. 2) (t1/2 of 14 and 29 h, respectively) and then to ADR-925 (t1/2 23 h) (Hasinoff, 1994b). We previously showed that ferrous ion strongly promoted the ring opening of B and C to ADR-925 (Buss and Hasinoff, 1995). Given that the intracellular environment of the cell is highly reducing, free or loosely bound intracellular iron would be expected to be largely present in the ferrous state. Thus, by reacting with ferrous ion, the strongly chelating ADR-925 could be formed intracellularly more quickly. From the kinetics of the formation of B and C (Hasinoff, 1994b), and assuming rapid ferrous ion-promoted ring opening, it can be estimated that a 3-h preincubation with 100 μM dexrazoxane could yield an intracellular ADR-925 concentration of 20 μM. The free or loosely bound iron concentration in cells is not known with certainty but has been estimated using calcein to be 1.3 μM in mouse leukemia cells (Picard et al., 1998). Thus, assuming that similar concentrations of free iron were present in myocytes, a 3-h preincubation with dexrazoxane would produce more than enough ADR-925 to chelate all of the free iron in the myocytes.
In conclusion, this study showed that whereas dexrazoxane was able to protect cardiac myocytes from doxorubicin-induced damage, its B and ADR-925 metabolites were not. Thus, these metabolites are unlikely to be useful on their own as antioxidant drugs. The ability of B, ADR-925, and dexrazoxane to displace iron from intracellular trapped iron-calcein complex suggests that these metabolites were taken up into the myocyte and bound iron. The fact that the these anionic metabolites did not protect myocytes from doxorubicin suggests either that the metabolites do not have access to the same cellular compartments that neutral dexrazoxane has, or that they are taken up too slowly to protect, or, alternatively, that dexrazoxane protects myocytes through its ability to inhibit topoisomerase II through some unknown mechanism.
Footnotes
-
This work was supported by the Canadian Institutes of Health Research, the Canada Research Chairs Program, and a Canada Research Chair in Drug Development for B.H.
-
ABBREVIATIONS: DF-x, Dulbecco's modified Eagle's medium/Ham's F-12 medium with 50 mM HEPES (pH 7.4) (x is v/v % serum); HBS, HEPES/NaCl buffer (20/150 mM, pH 7.4); kobs, pseudofirst-order rate constant; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick-end-labeling; LDH, lactate dehydrogenase; t1/2, half-time; t1/2β, beta phase pharmacokinetic elimination half-time; v, initial velocity; λex, λem, fluorescence excitation and emission wavelengths, respectively.
- Received February 23, 2003.
- Accepted May 20, 2003.
- The American Society for Pharmacology and Experimental Therapeutics