INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

AS POWERFUL NUCLEOPHILES, 2,3-butanedione monoxime (BDM) and related oximes act as "chemical phosphatases" and so may exert biological effects by altering protein phosphorylation. The ability of BDM to impede calcium-dependent contraction of cardiac (35) and other muscle is now well documented, although the mechanism remains controversial (reviewed in Ref. 27). Inhibitory effects have been described both on the calcium transient through impaired ion flux across the sarcoplasmic reticulum and sarcolemma and directly on the myofilaments. However, the primary effect of BDM in muscle appears to vary with concentration, temperature, and species.

In calcium-activated myofibrils, BDM shifts the equilibrium distribution of cross bridges from "strong" to "weak" attachment states (15, 20) and reduces myofibrillar ATPase activity (16), consistent with a stabilization of the M**:ADP:P_i intermediate state of the cross-bridge cycle (15). This negative inotropic effect has been exploited to suppress acute injury arising from myocardial hypercontracture at reperfusion (10, 26), reduce cutting injury (22), and lengthen the storage time for hearts in cardioplegic solutions (31). However, deleterious effects of BDM on cardiomyocyte recovery were evident in an in vitro model of reperfusion (3).

As cytosolic ATP levels fall to a threshold, rigor complex formation activates cross-bridge cycling and myosin ATPase independent of calcium (5), whereas, at cell level, contracture coincides with abrupt depletion of residual cytosolic ATP (4). A positive feedback loop whereby rigor activation of actomyosin accelerates ATP depletion has been suggested to produce the striking temporal association observed between contracture and erosion of transsarcolemmal sodium and calcium gradients in individual poisoned or hypoxic cardiomyocytes (1). On this basis inhibition of rigor should conserve ATP and delay or attenuate loss of ion homeostasis. However, although BDM indeed attenuates ATP and phosphocreatine depletion in working myocardium during metabolic inhibition (13), hypoxia (23), or ischemia (33), this may be secondary to reduced energy demand. When ATP resynthesis was limited in arrested heart during ischemia (14, 32) or in quiescent cardiomyocytes during hypoxia or metabolic inhibition (3, 18, 29) at physiological temperature, high-energy phosphate levels declined at least as rapidly in the presence of BDM as in its absence. Thus either rigor inhibition conferred no benefit or, alternatively, that benefit was offset by an ATP-wasting effect of BDM. Therefore, we studied in parallel how BDM affects rigor and myofibrillar ATPase activity in isolated quiescent cardiomyocytes at 37°C and investigated its impact on ATP metabolism and cell survival during metabolic inhibition.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Isolation of rat ventricular myocytes. Ventricular myocytes were isolated from male Wistar rats (200-300 g) by Langendorff perfusion with collagenase (Worthington Biochemical, NJ) as described previously (2). The isolate contained predominantly rod-shaped cells (>60%) and was maintained at 37°C in medium 199 (ICN/Flow, Thame, UK) under 5% CO₂ in air for use within 8 h.

Solutions. Experiments with permeabilized cells were conducted in a standard buffer comprising (in mM) 30 MOPS, 3.5 EGTA, and 1 dithiothreitol (DTT). ATP (sodium salt), MgCl₂, and KCl were added to give defined MgATP concentrations, 0.5 mM free magnesium, 3 nM free calcium, and 180 mM total ionic strength after pH had been adjusted to 7.10 at 37°C with KOH (6). Chemicals were of the highest quality commercially available from BDH (Poole, UK), Calbiochem (Nottingham, UK), or Sigma (St. Louis, MO). All solutions were prepared in fresh double-distilled water purified by reverse osmosis (Millipore), except for digitonin, fura 2, and some inhibitor stocks dissolved in DMSO.

Rigor and myofibrillar ATPase activity in suspensions of permeabilized cardiomyocytes. Cells (2 × 10⁵) harvested at 20 g for 40 s from 1.5-ml aliquots of the cardiomyocyte suspension were resuspended in an equal volume of intracellular-type medium (ICM-1) preequilibrated to 37°C, which comprised (in mM) 164 KCl, 3.5 EGTA, 1.02 MgCl₂, and 30 MOPS-KOH, pH 7.10. After 3 min, suspensions were cooled to 2°C over the next 5 min with occasional gentle trituration to prevent settling, and then cells were transferred to ice-cold ICM-1 supplemented with digitonin (2 µmol/mg protein, 0.1% wt/vol) for a further 5 min. Finally, permeabilized cells were washed free of excess detergent in four changes of an equal volume of fresh, cold ICM-1 and were allowed to settle before the supernatant was aspirated.

137148

Rigor in permeabilized attached cardiomyocytes. Cells were transferred from medium 199 to an extracellular-type medium [ECM; comprising (in mM) 125 NaCl, 2.6 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 10 HEPES-KOH, pH 7.40] supplemented with 10 mM glucose and 0.3 mM EGTA. After 5 min of incubation at 37°C,the cells were harvested and resuspended in thermoequilibrated intracellular-type medium (ICM-2) with (in mM) 140 KCl, 10 MgCl₂, 2 EGTA, 2 ATP, 10 phosphocreatine, and 35 HEPES-KOH, pH 7.00. Digitonin was added to 0.01% (wt/vol), and then, after 1 min, cells were washed three times in an equal volume of ICM-2 without detergent. A portion of this suspension (6 × 10³ cells) was transferred to a thermoregulated chamber (volume 125 µl) maintained at 37°C on the stage of an inverted microscope (Nikon), and the permeabilized cardiomyocytes were allowed to attach to the untreated glass floor for 2 min before superfusion was begun. Rigor was then induced by reducing concentrations of ATP and phosphocreatine in the ICM-2 superfusate to 10 µM and 2 mM, respectively. The ×100 image of the cells (with 5-10 cells per field of view) was relayed via charge-coupled device camera to a video recorder and TV monitor so that shortening could be tracked off-line relative to initial cell length as marked on the monitor screen. Onset and duration of rigor was measured for cells individually. When BDM was used, it was added to the cells 2 min before induction of rigor and its effects were compared with same-day controls.

Rigor and cytosolic free calcium in intact, attached cardiomyocytes. Cells were transferred to ECM supplemented with glucose (10 mM) and calcium (1 mM), exposed to 2 µM fura 2 (Molecular Probes, Eugene, OR) in the dark for 15 min at 37°C, and then transferred to dye-free medium for 30 min to allow fura 2 deesterification to reach completion. Fura 2-loaded cells were placed in the superfusion chamber on the microscope stage at 37°C and allowed to attach for 2 min before superfusion was begun. Cytosolic free calcium was measured at 33 Hz in individual, rod-shaped cardiomyocytes with a dual-excitation microfluorimeter (Cairn, Sittingbourne, UK) by exciting intracellular dye at 340 and 380 nm (bandwidth 10 nm), measuring fluorescence at 400-470 nm with a low-noise photomultiplier (Thorn EMI, Ruislip, UK), and calculating free calcium by a standard method (11). The cell was simultaneously imaged under red light. After stable resting calcium was measured from the cell over 5 min, glucose in the superfusate was replaced with 2-deoxyglucose (5 mM) and either cyanide (2 mM) or carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP; 2 µM) to inhibit or uncouple mitochondrial respiration, respectively.

BDM. BDM (50 mM) was freshly dissolved in ICM-2 on the day of use, the pH readjusted, and the solution protected from light. Consistent with the observation that BDM affects osmotic strength only marginally (27), we found that replacement of BDM with equimolar sucrose or mannitol caused none of the parameters measured to deviate significantly from control values.

Protein. Protein content of cell suspensions was measured against bovine serum albumin standards using the Bio-Rad assay.

Statistical analysis. Data sets were compared with the use of Wilcoxon-Mann-Whitney and unpaired t-tests by using Astute software (DDU Software, University of Leeds, Leeds, UK) and taking significance as P  0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

BDM inhibition of rigor. BDM (50 mM) impeded development of rigor in cardiomyocytes as measured after 40 s at 37°C over a range of MgATP concentrations (Fig. 1). The effect was dose dependent (IC₅₀ ~13 mM at 56 µM MgATP; Fig. 1, inset). At the individual cell level, this inhibition was expressed as a slowing of the shortening process (Fig. 2) and a reduction in its extent (Fig. 3A). However, even at 50 mM, the agent did not reverse established rigor in cells that had already undergone contracture after 40 s at 32 µM MgATP (data not shown).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   In permeabilized cardiomyocytes in suspension, 2,3-butanedione monoxime (BDM) inhibits rigor developed over 40 s at 37°C, where rigor is defined as cell contracture to a length-to-width ratio <3 (see MATERIALS AND METHODS). Data points are means ± SD (n = 4-8). Inset: BDM inhibition of rigor is dose dependent and is shown at 56 µM MgATP. Bars are means + SD (n = 4).

View larger version (11K): [in this window] [in a new window]   Fig. 2.   BDM (0-20 mM) slows rate at which rigor contracture progresses in attached, permeabilized cardiomyocytes exposed to 10 µM ATP and 2 mM phosphocreatine. Data points are means ± SD [n = 3 except for control (0 BDM), where n = 22]. * P < 0.05, ** P < 0.01 vs. control.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effects of BDM on rigor in intact cardiomyocytes poisoned with 2-deoxyglucose and either cyanide (CN) or carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP), with extent of rigor shortening (A) and time of rigor onset (B) shown. At 20 mM BDM, CN-treated cells shortened by 28 ± 2%, but time to rigor onset was not significantly different from that in absence of BDM. Bars are means + SD (n = 3-7). * P < 0.05, ** P < 0.01 vs. control.

Impact of BDM on cell energy status and cytosolic free calcium. Intact cells exposed to 2-deoxyglucose and either cyanide or FCCP underwent contractures of similar extent (Fig. 3A), followed by similar rises in cytosolic free calcium (data not shown), but showed different rates of deenergization, as indicated by onset of contracture (Fig. 3B). We could not measure temporal changes of cytosolic ATP directly in individual cells because BDM interferes with the only method available, namely microinjection with firefly luciferase (4). However, because BDM did not affect onset of rigor in permeabilized cells superfused with low concentrations of ATP (not shown), onset of rigor in intact cells could be used as an index of ATP depletion to a threshold value. On this basis it was evident that BDM accelerated ATP depletion in cells poisoned with cyanide but attenuated ATP depletion in cells uncoupled with FCCP (Fig. 3B). The essential difference between these conditions is that cyanide prevents mitochondrial respiration and hence oxidative phosphorylation, whereas FCCP, by uncoupling mitochondria, activates a powerful intramitochondrial ATP sink as the F₀F₁-ATP synthase reverts to an H⁺-pumping ATPase to counter depolarization of the inner membrane.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   BDM limits extent of contracture in cells poisoned with FCCP and 2-deoxyglucose but does not conserve cytosolic ATP. Rapid washout of BDM (50 mM) after rigor in continued presence of FCCP does not allow further shortening; cell length remains at 81 ± 7% (n = 6; upper dashed line), indicating that cytosolic ATP is already too low to support resumption of cross-bridge cycling. Lower dashed line indicates extent of rigor contracture in control cells (59 ± 4%, n = 6), whereas hatched area denotes mean duration of shortening event. Data are compiled from experiments on a series of 8 cells.

BDM effect on ATP hydrolysis in permeabilized cells. In parallel with measurements of cardiomyocyte rigor at low ATP concentration, we studied how BDM affected ATP hydrolysis in permeabilized cells. At 37°C, but not 2°C, cell-dependent ATP hydrolysis in the presence of inhibitors of major non-myosin ATPases was stimulated by BDM (Fig. 5), although BDM neither interfered with the P_i assay nor dephosphorylated ATP directly. However, the effect was abolished when ATP was replaced with adenosine 5'-O-(3-thiotriphosphate) (ATPS; data not shown), suggesting that the ATP hydrolysis induced by BDM involved protein dephosphorylation to which thiophosphate esters were resistant. The possibility that BDM enhanced net ATP hydrolysis by opposing kinase-mediated protein phosphorylation was tested directly using the general serine/threonine kinase inhibitor H-7 and the tyrosine kinase inhibitor genistein. Together, H-7 and genistein returned net P_i release essentially to control values in the presence of BDM (Fig. 6), despite having no effect in its absence. More specific inhibitors of myosin light-chain kinase and protein kinase C (PKC), kaempferol and chelerythrine, respectively, also inhibited BDM-dependent ATP hydrolysis significantly. Finally, neither kinase inhibitors nor ATP replacement with ATPS affected inhibition of rigor by BDM.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   BDM promotes ATP hydrolysis in permeabilized cardiomyocytes, despite attenuating rigor contracture induced at 1 mM MgATP. Myofibrillar ATPase activity was measured as cell-dependent Pi release from ATP in presence of 50 µM each of cyclopiazonic acid, oligomycin, and ouabain, as described in MATERIALS AND METHODS. Other conditions were as follows: 1 mM MgATP, 0.5 mM free Mg2+, and 3 nM free Ca2+, with total ionic strength 180 mM, pH 7.10, 37°C. Data points are means ± SD [n = 5-6 except for control (0 BDM), where n = 10]. * P < 0.05, ** P < 0.01 vs. control.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Protein kinase inhibitors attenuate ATP hydrolysis promoted by BDM at 1 mM MgATP in presence (+) or absence () of BDM (50 mM), kaempferol (1 µM), chelerythrine (10 µM), 1-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H-7; 10 µM), and genistein (50 µM), with other conditions same as in Fig. 5. Bars are means + SD [n = 4-6 except for control (BDM), where n = 10]. * P < 0.05, ** P < 0.01 vs. control.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

BDM inhibits actin-myosin interaction (15, 20, 24, 35) and reduces myofilament sensitivity to calcium (12). Although the agent decreases the efficiency of energy coupling in cardiac contraction (8), it has been used successfully as a negative inotrope to prevent tissue hypercontracture at reoxygenation or reperfusion until calcium has been restored to diastolic levels (10, 26). The present study addressed the impact of BDM on energy status of cardiomyocytes at 37°C. Besides inhibiting development of rigor, BDM showed three other major effects: 1) it accelerated deenergization under conditions that restrict ATP regeneration, 2) it promoted promiscuous dephosphorylation of phosphoproteins at serine/threonine and tyrosine residues, and 3) it inhibited adenine nucleotide movement across the mitochondrial inner membrane. Although BDM impeded hypercontracture of cardiomyocytes at removal of metabolic blockade, it provided no ATP-sparing effect nor postponed ionic derangements.

BDM slowed and reduced the extent of rigor contracture in intact cells, whether they had been subjected to respiratory inhibition or uncoupling. BDM also slowed rigor induced at low ATP concentration in the myofilament array of permeabilized cells but could not reverse contracture that had already taken place. Inhibition of development, but not reversal of contracture, by BDM (13, 18, 24) is consistent with stabilization by BDM of weakly bound cross bridges that contain ATP hydrolysis products at the expense of strongly bound (force-producing) states (15, 20).

BDM exacerbates the rate of decline of ATP in cyanide-treated cells. Although cytosolic free ATP cannot be measured directly in individual cardiomyocytes exposed to BDM, onset of contracture marks ATP depletion to the threshold at which rigor is initiated. On this basis we found that BDM at high concentration (50 mM but not 20 mM) accelerated deenergization in cells in which ATP resynthesis was prevented. This is consistent with the observation that BDM greatly reduced both extramitochondrial and mitochondrial ATP in ischemic guinea pig heart at 35°C, but not at 4°C (14), indicating that ATP depletion promoted by BDM is not only concentration dependent but also strongly temperature sensitive.

BDM promotes protein dephosphorylation. BDM acts as a noncompetitive inhibitor in preparations of myosin ATPase from skeletal muscle (15, 20) and reduces myosin ATPase activity in cardiac muscle (8). However, although it attenuated rigor in permeabilized cardiomyocytes, ATP hydrolysis was increased considerably (Fig. 5) unless ATP was replaced with ATPS. This suggested that promotion of cell-dependent ATP hydrolysis by BDM might involve phosphatase activity against which protein thioesters generated by kinases were resistant. Such an interpretation is strongly supported by our observation that kinase inhibitors reduce or prevent BDM-dependent P_i release from ATP. Both serine/threonine kinase and tyrosine kinase inhibitors were effective and, in combination, restored ATP hydrolysis in the presence of BDM to control levels. Thus it would appear that at 37°C, but not at 2°C, BDM dramatically shifts the equilibrium between protein phosphorylation and dephosphorylation in cardiomyocytes, eliciting a compensatory increase in kinase activity at the expense of ATP. This might contribute to the ATP depletion produced by BDM in ischemic heart (14). Protein dephosphorylation may also underlie the osmotic fragility (3) and susceptibility to hypercontracture (19) induced in cardiomyocytes by BDM against which phosphatase inhibitors afford protection. Nevertheless, we found no evidence that putative changes in protein phosphorylation affected inhibition of rigor by BDM.

BDM impedes adenine nucleotide transfer across the mitochondrial inner membrane. A crucial observation in this study is that BDM accelerated ATP depletion in cardiomyocytes subject to respiratory inhibition but conserved ATP in cells poisoned with respiratory uncoupler. Mitochondrial uncoupling increases ATP hydrolysis by F₀F₁-ATPase in opposition to the fall in mitochondrial membrane potential. In the absence of BDM, FCCP induced rigor much faster than did cyanide, indicating that if respiratory inhibition caused mitochondrial depolarization (25) under our conditions, then it resulted in far less F₀F₁-mediated ATP hydrolysis than did FCCP. However, these rates of cyanide- and FCCP-induced deenergization were equalized in the presence of BDM, although the agent has no direct effect on F₀F₁-ATPase (21). The most direct explanation is that BDM restricted access of cytosolic ATP to F₀F₁-ATPase, consistent with the suggestion that adenine nucleotide exchange across the mitochondrial inner membrane is inhibited (21). Thus, in the presence of BDM, nonmitochondrial sinks for ATP predominated, potentially including futile cycling of protein phosphorylation. On the basis of its simultaneous, mutually reinforcing effects on ATP synthesis and expenditure, BDM at high concentration would therefore be expected to hasten deenergization of cardiomyocytes under hypoxia and hinder their reenergization at reoxygenation, with the latter probably masked by low energy demand during contractile paralysis. This is consistent with what was observed during metabolic inhibition (Fig. 4): although BDM reduced contracture, it rapidly depleted cell energy reserves. This also explains why BDM does not postpone or attenuate the rise in cytosolic free calcium. We conclude that, in quiescent cardiomyocytes at 37°C, BDM initiates ATP wastage that negates any energy-sparing effect of rigor inhibition.

BDM use in cardioplegia. BDM is likely to be far more specific at low temperature because its negative inotropic action increases (16), whereas its effects on mitochondrial ATP synthesis and protein phosphorylation may diminish. We observed no BDM-induced ATPase activity in permeabilized cells at 2°C, and cold storage of rat hearts with the general kinase inhibitor staurosporine in addition to BDM did not enhance survival (9). In contrast to the situation at physiological temperature, the energy-sparing effect of BDM also outweighs inhibition of ATP resynthesis at 4°C (14). BDM has been found to improve myocardial survival of ischemia during cold storage (9, 31) and of reperfusion at 37°C (10). However, the data presented here suggest that prolonged exposure of myocardium to BDM at physiological temperature should be avoided under circumstances in which the capacity of the tissue to resynthesize ATP is limited, for example, as in bypass surgery.