- Email: [email protected]
Modulation of Rigor and Myosin ATPase Activity in Rat Cardiomyocytes
J Mol Cell Cardiol 30, 1349–1358 (1998) Article No. mc980703
Modulation of Rigor and Myosin ATPase Activity in Rat Cardiomyocytes Mary T. Stapleton and Ashley P. Allshire Department of Pharmacology and Therapeutics, University College Cork, University Hospital, Cork, Ireland (Received 26 January 1998, accepted in revised form 31 March 1998) M. T. S A. P. A. Modulation of Rigor and Myosin ATPase Activity in Rat Cardiomyocytes. Journal of Molecular and Cellular Cardiology (1998) 30, 1349–1358. Ischaemic myocardium undergoes calciumindependent contracture at millimolar tissue ATP, though in actomyosin solutions ATP must be reduced to micromolar before rigor complexes form. This contracture is associated with myosin ATPase activity that may contribute to tissue de-energization. Here we used isolated rat cardiomyocytes permeabilized with digitonin to analyse in parallel how rigor and myosin ATPase activity are modulated by metabolic conditions that develop during ischaemia. At pH 7.1 and 37°C rigor and myosin ATPase showed co-ordinated bell-shaped dependence on ATP concentration over 3–1000 l. Rigor, but not myosin ATPase, was inhibited by acidosis (pH 6.2), indicating reduced efficiency of cross-bridge cycling, while both parameters were stimulated by ADP (Ζ1 m) and unaffected by inorganic phosphate (Pi, 30 m), AMP, Mg2+, lactate or inhibition of adenylate kinase with diadenosine pentaphosphate. Combined acidosis and high ADP inhibited rigor, while Pi attenuated the enhancement of rigor by ADP. Thus, rigor complex formation activates myosin ATPase in the intact myofilament array, modulated by ADP, Pi and acidosis in the ranges that occur in ischaemia. There was no evidence that adenylate kinase might attenuate falling ATP/ADP ratio at the myofilaments. In combination these effects are sufficient to resolve the apparent discrepancy between ATP concentrations triggering rigor in actomyosin and onset of contracture in ischaemic myocardium. Since rigor contracture activates myosin ATPase it is likely to exacerbate ATP depletion and thereby limit vital cell functions. This positive feedback is consistent with the abrupt depletion of ATP observed in individual cardiomyocytes undergoing deenergization contracture. 1998 Academic Press K W: Ischaemia; Rigor; Myosin ATPase; ADP; Acidosis; Rat cardiomyocytes; Free energy of ATP hydrolysis; Troponin I.
Introduction In cardiac ischaemia blood flow through the myocardium is insufficient to meet the oxygen demand of the tissue. Hypoxia restricts aerobic regeneration of ATP, while poor wash-out of catabolites may exacerbate tissue dysfunction. Shortly after onset of ischaemia contractility in the affected region wanes then ceases, whereupon a contracture develops largely independent of calcium. Eventually myocytes in the poorly-irrigated tissue sustain irreversible injury and an infarct forms. The role of this contracture in the progression to
tissue death has been studied extensively but a clear understanding of the mechanism involved has not yet emerged. Diastolic pressure rises in ischaemic myocardium despite little calcium-activated muscle contraction (Holubarsch et al., 1982), while shortening of individual cardiomyocytes under anoxia precedes any significant rise in cytosolic free calcium (Allshire et al., 1987). In vitro, actin and myosin interact independent of calcium when ATP levels are low (10–100 l) by forming nucleotide-free rigor cross bridges that activate myosin ATPase and crossbridge cycling (Bremel and Weber, 1972), whereas onset of contracture in intact myocardium and car-
Please address all correspondence to: Dr Ashley Allshire, Department of Pharmacology and Therapeutics, University College Cork, University Hospital, Cork, Ireland.
0022–2828/98/071349+10 $30.00/0
1998 Academic Press
1350
M. T. Stapleton and A. P. Allshire
diomyocytes occurs at about 1–2 m ATP (Altschuld et al., 1981; Nichols and Lederer, 1990; Elliott et al., 1992). This may reflect distinct behaviour of myosin ATPase and the cross-bridge cycle in actomyosin solution compared to that in the highly-ordered microenvironment of the myofilaments in situ. In addition, the calcium-independent contracture in ischaemic heart may be modulated by prevailing metabolic conditions. Ischaemia induces a switch from aerobic to anaerobic metabolism, depletion of tissue phosphocreatine and then ATP, and accumulation in the cytosol of catabolites including lactate, ADP, AMP, magnesium ion, inorganic phosphate and protons. Furthermore the in vitro behaviour of actomyosin suggests that rigor activation of myosin ATPase should exacerbate ATP depletion, though as far as we are aware this has not been demonstrated in heart muscle. However this would most readily explain the observation (Haworth et al., 1981; Bowers et al., 1992; Allue et al., 1996) that cardiomyocyte shortening coincides precisely with rapid depletion of cytosolic ATP. The aim of the present study was to identify factors likely to influence rigor in ischaemic myocardium. Cardiomyocytes were permeabilized with digitonin in order to expose the intact myofilament array at 37°C to conditions reflecting the intracellular ionic and nucleotide milieu. This allowed rigor indicated by cell shortening independent of calcium to be measured in parallel with myosin ATPase activity in terms of inorganic phosphate released from MgATP in the presence of inhibitors of the main non-myosin ATPases. At low MgATP myosin ATPase activity correlated with rigor development. Of the metabolites that accumulate in ischaemia, only ADP and protons affected rigor directly.
Materials and Methods Myocytes Ventricular myocytes were isolated from male Wistar rats (200–300 g) by Langendorff perfusion with collagenase (Worthington Biochemical Corporation, NJ, USA), as described previously (Allshire et al., 1987). The preparation contained 59±8% rod-shaped cells (n=25) and was maintained at 37°C in medium 199 (ICN/Flow, UK) under 5% CO2 in air, and used within 4 h of isolation. Cell permeabilization Cardiomyocytes harvested from 1.5-ml portions of the culture suspension by gentle centrifugation
(20 g×60 s) were resuspended in an equal volume of intracellular type medium (ICM) comprising (in m): 164 KCl, 3.5 EGTA, 1.02 MgCl2 and 30 3-[Nmorpholino]propane-sulfonic acid (MOPS), adjusted to pH 7.10 with KOH, at 37°C. After 3 min at 37°C, the cells were cooled to 2°C over 5 min to prevent rigor development at permeabilization as endogenous ATP levels fell. During this period cells were triturated gently at intervals to prevent settling, then harvested and resuspended in 1.5 ml of icecold ICM supplemented with digitonin (2 lmol/mg protein) for 5 min before being washed in four changes of an equal volume of detergent-free ICM. Finally a sample was removed for measurement of rigor frequency and cell permeabilization, immediately before the microfuge tube containing the cell pellet was transferred to a water bath at 37°C. Twenty seconds later the assay was begun by resuspending the cells in ATP-containing solution pre-equilibrated at 37°C, then 250-ll samples of the suspension were taken at 10 and 30 s for measurement of myosin ATPase activity and a 20-ll sample at 40 s for measurement of rigor frequency.
Solutions Composition of solutions containing Ca2+, Mg2+ and multiple ligands were calculated using the computer program of Brooks and Storey (1992). Intracellular-type media were prepared from a basic stock comprising (in m): 30 MOPS, 3.5 EGTA and 1 dithiothreitol (DTT), to which were added ATP (sodium salt), MgCl2, and KCl to give defined MgATP concentrations and total ionic strength of 180 m after pH had been adjusted to 7.10 at 37°C with KOH. Free magnesium concentration was set at 0.5 m and free calcium at 3 n. A pH of 7.10 was chosen to match that in normoxic myocardium. When other metabolites were included (inorganic phosphate, lactate, magnesium, ADP, AMP), or when pH was lowered, buffer composition was adjusted in order to maintain MgATP, free Mg2+ and Ca2+, and ionic strength. Chemicals were obtained from BDH Ltd, Poole, UK, and Sigma Chemical Co., St Louis, MO, USA.
Rigor Samples of cell suspensions (20 ll) were fixed and stained by addition to an equal volume of glutaraldehyde/trypan blue dye solution (each 1% w/ v), then scored for rigor (shortening to length:width Ζ3; Armstrong and Ganote, 1991) and dye uptake
1351
Rigor and Myosin ATPase in Cardiomyocytes
ATPase activity was measured as inorganic phosphate (Pi) release from ATP, using the malachite green assay (Chan et al., 1986). Major non-myosin ATPases were inhibited by exposing permeabilized cells to a cocktail containing 50 l each of ouabain (Na+,K+-ATPase), oligomycin (mitochondrial F0F1ATPase) and cyclopiazonic acid (SR Ca2+-ATPase) in ICM with 1 m DTT for 10 min at 2°C before transfer to ATP-containing solution supplemented with inhibitors at 37°C. Correction was made for spontaneous ATP hydrolysis, always <10% of that in the presence of cells. We verified that residual ATP hydrolysis represented myosin ATPase activity using a peptide mimetic of the troponin-I switch sequence (Tn-I137–148). In experiments where cells had not been pre-incubated with the inhibitor cocktail, Tn-I137–148 (114 l) attenuated rigor and ATPase activity by 96±8% (n=4) and 88±5% (n= 7), respectively, at 56 l MgATP, while for both parameters half-maximal inhibition occurred at 22 l peptide (Stapleton et al., 1995). However at 1 m MgATP a proportion of the ATPase activity was refractory to Tn-I137–148, indicating a growing contribution by ATPase(s) other than myosin to Pi generation as ATP concentration approached millimolar.
Protein Protein content of cell suspensions was measured by the Bio-Rad assay, using bovine serum albumin standards.
75
75
50
50
*
25
0
5.5
5.0
†
4.5
4.0 3.5 p MgATP
25
3.0
2.5
Percent cells in rigor ( )
Myosin ATPase
100
100 ATPase activity (pmol Pi/ g protein/min) ( )
within 1 h, over which time these parameters remained stable. Typically 100–200 cells were scored per sample. As the proportion of rod-shaped cells varied between individual preparations, rigor frequency was normalized.
0
Figure 1 Parallel development of rigor and myosin ATPase activity in permeabilized cardiomyocytes at 37°C. Rigor was defined as cell length:width ratio <3 and myosin ATPase activity was measured as inorganic phosphate release in the presence of non-myosin ATPase inhibitors (see Materials and Methods for details). Datapoints are means±.. (n=4–10); ∗ P<0.05, † P<0.01.
Results Development of rigor At physiologically relevant temperature, pH, osmotic strength and ionic milieu permeabilized rat cardiomyocytes underwent rigor contracture independent of calcium at millimolar to micromolar MgATP. However, to facilitate detection of altered rigor behaviour, we adopted a standard exposure to MgATP of 40 s that caused all cells to shorten over a narrow MgATP range (about 20–60 l) but fewer at very low or high concentrations (6% at 3 l; 8% at 1.8 m; Fig. 1). Myosin ATPase activity measured in parallel followed the general pattern of rigor development, namely a peak activity at 30 l MgATP (Fig. 1). Yet while ATPase activity at pMgATP 4.0 was similar to that at 4.75 rigor was more extreme in the latter, and at pMgATP 3.0 rigor was low but ATP hydrolysis high. This underlying trend of increasing hydrolysis with MgATP concentration that was independent of rigor reflects residual activity of nonmyosin ATPases (see Materials and Methods).
Statistical analysis Values are expressed as means ±.. Data sets were compared by Wilcoxon–Mann–Whitney and unpaired t-tests using the Astute statistical package (DDU Software, University of Leeds, UK), taking significance as PΖ0.05. For clarity, some error bars in the figures are shown in one direction only.
Effect of ischaemic metabolites on rigor and myosin ATPase
Acidosis and lactate Reduction of pH from 7.1 to 6.8 and 6.2 progressively and strongly suppressed rigor in the
1352
M. T. Stapleton and A. P. Allshire
Percent cells in rigor
100 †
80
†
60 40
* †
20 0
† 5.5
5.0
†
4.5 4.0 p MgATP
* 3.5
3.0
Figure 2 Inhibition of rigor in cardiomyocytes on decreasing pH from 7.1 to 6.2. Data points are means±.. (n=4–8); ∗ P<0.05, † P<0.01 v control. (Χ) pH 7.1; (Φ) pH 6.8; (Α) pH 6.2.
MgATP range 0.03–1 m (pMgATP 4.5–3.0), while effects at lower MgATP were minimal (Fig. 2). However at pMgATP 4.0 where mean rigor frequency was reduced from 66% at pH 7.1 to 1% at pH 6.2 (P<0.01) ATPase activity remained unchanged [56±19 (n=6) v 42±13 pmol Pi/lmol protein/ min (n=3) at pH 7.1 and 6.2, respectively; P= 0.40]. By contrast, lactate had no effect on either rigor [Fig. 3(b)] or ATPase activity [measured at pMgATP 4.0: 56±19 (n=6) v 80±25 pmol Pi/ lmol protein/min (n=4) in absence and presence of 30 m lactate, respectively; P=0.19].
ATP hydrolysis products Hydrolysis of MgATP releases magnesium ion, and free magnesium reduces myofilament calcium sensitivity (Fabiato and Fabiato, 1975). However increasing free magnesium concentration from 0.5 to 3.0 m had no effect on rigor other than to enhance it somewhat at pMgATP 4.0 [100 l; Fig. 3(c)], nor on ATPase activity [56±19 (n=6) v 80±23 pmol Pi/lmol protein/min (n=4) at 0.5 and 3 m free Mg2+, respectively; P=0.17]. Net hydrolysis of creatine phosphate and ATP causes a substantial increase in cytosolic inorganic phosphate (Pi). However no effect on rigor contracture of permeabilized cells could be discerned at either 15 m (not shown) or 30 m Pi [Fig. 3(d)]. The effect of MgADP on rigor was studied at pMgATP 4.0 and 3.0 (0.1 and 1.0 m MgATP), concentrations at which rigor was close to halfmaximal (66±9%) and minimal (13±9%), respectively. At both MgATP concentrations increasing MgADP promoted rigor strongly once ATP: ADP fell below a threshold of about 1.3 (Fig. 4).
At 1 m MgATP MgADP increased myosin ATPase activity parallel to its effect on rigor (Fig. 5). However inhibition of adenylate kinase (AK) with diadenosine pentaphosphate (AP5A; 10 or 100 l) did not increase rigor further, suggesting that the enzyme did not greatly affect the ATP:ADP ratio within the myofibrillar array (Fig. 4). On this basis, neither could we discern an effect of myofibrillar adenylate kinase on rigor at 0.1 m MgATP and either 25 or 50 l MgADP in the presence of inorganic phosphate (30 m), phosphocreatine (12 m) or at pH 6.2 (data not shown). Although endogenous AK did not appear to alter the nucleotide environment of the myofilaments under the conditions used here, in principle the AK reaction might counter falling ATP:ADP and produce AMP in the process. Therefore we tested the effect of AMP on myofibril response to MgATP. Neither rigor nor ATPase activity were affected by 0.12 m AMP (0.004 m MgAMP) at 0.1 m MgATP or by 1.2 m AMP (0.04 m MgAMP) at 1 m MgATP (data not shown). However at 0.1 m MgATP 3 m AMP stimulated both rigor (from 66±9 to 100±0%, P<0.01) and ATPase activity (from 86±2 to 115±20 pmol/lg protein/min, P<0.05). It may be that under the latter conditions high AMP drove ADP production by reversing the AK reaction at the expense of ATP, and thereby depressed ATP:ADP at the myofilaments. Since increasing ADP concentration potentiated rigor while increasing Pi did not, we tested whether rigor was regulated by DGATP . Varying DGATP between −26 and −51 kJ/mol had essentially no effect on rigor frequency (r=0.03; Table 1). Thus, in the presence of sufficient MgATP to support crossbridge cycling, the primary regulation of rigor was allosteric (by MgADP) rather than thermodynamic. For example, addition of Pi in the presence of MgADP, which would reduce DGATP , tended to inhibit rather than promote rigor at progressively higher ADP concentrations (Table 1).
Combined ADP and acidosis In ischaemic myocardium lactate accumulation and net ATP hydrolysis will contribute to development of acidosis. As protons and MgADP separately had opposing effects on rigor, we tested their net effect in combination (Fig. 6). Rigor was essentially complete at both 0.1 and 1 m MgATP (pMgATP 4 and 3, respectively) in the presence of equimolar MgADP. However, at both MgATP concentrations, reducing pH by 0.3 unit more than halved the extent of rigor, while at pH 6.2 the extent of rigor was no more than 10% of that at pH 7.1. This
1353
Rigor and Myosin ATPase in Cardiomyocytes 100
(a)
75
Percent cells in rigor
Percent cells in rigor
100
50
(b)
75
50
25
25
0 30 mM 0
4.5
5.5
0
2.5
75
50
25
4.5 p MgATP
5.5
3.5
100
(c)
*
Percent cells in rigor
Percent cells in rigor
100
3.5 p MgATP
(d)
75 0 30 mM 50
25 0.5 mM 3 mM
0
4.5
4.0
3.5 p MgATP
3.0
0
5.5
4.5
3.5 p MgATP
2.5
Figure 3 Rigor frequency in isolated cardiomyocytes after 40 s at 37°C, control (a) is largely unaffected by addition of 30 m lactate (b), 3 m free Mg2+ (c) or 30 m Pi (d). Data points are means±.. (n=4–8); ∗ P<0.01 v control.
pattern was not affected by the further addition of 30 m each Pi and lactate (data not shown).
ATPase activity, while rigor was inhibited by acidosis. No other metabolites that accumulate in ischaemic myocardium had any significant effect.
Discussion Rigor and myosin ATPase activity In this study, permeabilized cardiomyocytes were used to develop a model of ischaemic contracture in which both rigor (calcium-independent cell shortening) and myosin ATPase (Pi release) could be measured in parallel. It provides the first direct evidence of which we are aware that rigor is accompanied by increased myosin ATPase activity in intact arrays of cardiac myofilaments, under physiologically-relevant conditions of temperature, pH, ionic strength and milieu (other than very low free calcium). ADP enhanced both rigor and myosin
Actin binds the myosin head strongly in the absence of nucleotide. This rigor complex is formed transiently during the normal cross-bridge cycle; however, during the intermediate state of lowest free energy, rigor complexes accumulate when ATP levels fall below the threshold required to dissociate them. In dilute solutions of actomyosin, rigor complexes “turn on” adjacent actin monomers not combined with myosin in the absence of calcium. However, it does not automatically follow that the
1354
M. T. Stapleton and A. P. Allshire
80 60
*
40 20 0
0
100 Percent cells in rigor
Table 1 Rigor frequency in permeabilized cardiomyocytes at 37°C as a function of free energy of ATP hydrolysis, DGATP , at 1 m MgATP. In the absence of added ADP or Pi values were arbitrarily set at 1% of total ATP and 0.025 m, respectively, in order to estimate DGATP . Rigor frequency is compared without and with added Pi at each ADP concentration, where ∗ P<0.05; † P<0.01; .., not significant.
(a)
20
40 60 MgADP ( M)
80
100
(b)
80 60 40 20 0
0.00
0.25
0.50 0.75 1.00 MgADP (mM)
1.25
† 150
75
100
50
*
25
50
0 0.00
0.25
0.50 0.75 MgADP (mM)
1.00
Percent cells in rigor ( )
ATPase activity (pmol Pi/ g protein/min) ( )
100
†
Pi (m)
−DGATP (kJ/mol)
0.015 0.015 2.0 2.0 3.0 3.0 4.0 4.0
0.025 5.0 0.025 3.0 0.025 2.0 0.025 1.5
51 37 38 26 37 26 36 26
0 1.25
Figure 5 Promotion of rigor and myosin ATPase activity by MgADP at 1 m MgATP. Datapoints are means±.. (n=4–10); ∗ P<0.05, † P<0.01 v absence of MgADP.
same happens within a cardiomyocyte where actin and myosin are sterically constrained in the lattice array of the myofilament. Demonstration of a direct association between rigor and myosin ATPase activity in the intact myofilament is hampered by background ATP hydrolysis. Here, we equated myosin ATPase activity with total inorganic phosphate released from ATP in the presence of inhibitors of the other major
% cells in rigor mean±.. (n) 13±9 23±4 20±17 8±9 45±9 18±10 98±2 55±15
(8) (4) (4) (4) (4) (4) (4) (4)
P ∗ .. † †
100
1.50
Figure 4 Promotion of rigor by MgADP at 0.1 (a) and 1 m (b) MgATP, and the effect of adenylate kinase inhibition with diadenosine pentaphosphate (AP5A; 10 l). Data points are means±.. (n=4); ∗ P<0.05, v control. (Χ) Control; (Β) +AP5A.
200
ADP (m)
80 Percent cells in rigor
Percent cells in rigor
100
60
40
20
0
7.1
6.8
6.5
6.2
pH
Figure 6 Acidosis impairs the enhancement of rigor by MgADP at both 1 (Φ) and 0.1 m (∆) MgATP in the presence of equimolar MgADP. Bars are means±.. (n= 4) and effects are significant at pH Ζ6.8 v pH 7.1 (P<0.01).
ATPases in cardiomyocytes, i.e. the mitochondrial F0F1-ATPase, the Na+,K+-ATPase and the sarcoplasmic reticulum Ca2+-ATPase. That mimetic TnI137–148 almost completely inhibited both rigor and ATPase hydrolysis, and with the same potency, indicates that this assumption is valid. In addition, rigor and residual ATPase activity both showed bell-shaped dependence on MgATP over the same concentration range (Fig. 1), and ADP enhanced both rigor and ATP hydrolysis (Fig. 5), whereas the opposite effect would be expected of an ATP
Rigor and Myosin ATPase in Cardiomyocytes
hydrolysis product on the basis of mass action. Overall, we observed that net ATPase activity clearly attributable to myosin matched the incidence of rigor, both showing a maximum over the range 20–50 l MgATP, which is consistent with earlier reports of rigor in cardiomyocytes and myocardium (Fabiato and Fabiato, 1975; Haworth et al., 1981; Nichols and Lederer, 1990). Furthermore, the biphasic incidence of rigor in the intact myofilament array as a function of MgATP matches qualitatively the behaviour of skeletal actomyosin in solution (Bremel and Weber, 1972), with the ascending phase of the curve probably reflecting substrate affinity of myosin ATPase and the descending phase corresponding to progressive saturation.
Effect of ischaemic metabolites While the effects of ischaemic metabolites on calcium activated force in heart muscle are well characterized, effects on rigor force development are less extensively documented (Silverman et al., 1994; Smith and Steele, 1994; Ventura-Clapier and Veksler, 1994). To our knowledge, the present study is the first to address the impact of ischaemic metabolites on combined rigor and myosin ATPase activity in cardiac myofilaments. Lactic acid dissociation and H+ release from ATP hydrolysis during myocardial ischaemia causes intracellular pH to fall rapidly by about 1 unit. This acidosis inhibits calcium-induced contraction and calcium influx (Altschuld et al., 1981), which effects may lead to conservation of ATP and adenine nucleotide (Bak and Ingwall, 1994). Here, acidosis progressively and powerfully inhibited rigor in permeabilized cardiomyocytes (Fig. 3), as it does in chemically skinned rat ventricular fibres (VenturaClapier and Veksler, 1994) and trabeculae (Smith and Steele, 1994). That acidosis decreases both the number of cross-bridges, and particularly the force per cross-bridge, could reflect a direct effect on myosin ATPase (Ventura-Clapier and Veksler, 1994). However, we found that myosin ATPase activity associated with rigor in permeabilized cardiomyocytes was similar at pH 7.2 and 6.2, while the effect of acidosis on myofibrillar ATPase activity in skinned rat trabeculae undergoing calcium-induced contraction was also reported to be small (Ebus et al., 1994). Hence, acidosis considerably reduces the energy efficiency not only of calciumdependent but also of rigor cross-bridge cycling in the cardiac myofilament. This argues against the suggestion of Ventura-Clapier and Veksler (1994) that rigor might provide sustained tension at low
1355
energy cost in ischaemic myocardium. Finally, the lack of effect of lactate on either rigor or myosin ATPase in permeabilized cardiomyocytes is consistent with the conclusion that injury associated with its accumulation in ischaemic myocardium is actually due to acidosis (Neely and Grotyohann, 1984; Vander Heide et al., 1996a). During hypoxia, cytosolic free magnesium rises to 2.8 m in rat ventricular myocytes (Silverman et al., 1994), commensurate with the depletion of its main intracellular ligand ATP. A comparable rise to 2.3 m in intact rat heart (Headrick and Willis, 1991) coincides with onset of contractile failure and reaches a plateau as rigor contracture develops. However, despite its negative inotropic effect on calcium-activated contraction, increasing free magnesium had only marginal effects on rigor and associated myosin ATPase in permeabilized cardiomyocytes. Therefore, the reduction in infarct size reported in animal studies following magnesium administration at the time of occlusion (Christensen et al., 1995) or reperfusion (Herzog et al., 1995) is unlikely to be due to attenuation of rigor. Intracellular Pi increases rapidly in ischaemic myocardium (Koretsune and Marban, 1990), which together with developing acidosis contributes to contractile failure (Kentish, 1986; Lee and Allen 1991). It is well established that Pi inhibits myofibrillar Ca2+-activated ATPase activity (Cooke and Pate, 1985; Parkhouse, 1991; Ebus et al., 1994) and depresses Ca2+-dependent tension development (Cooke and Pate, 1985; Kentish, 1986; Ebus et al., 1994). However, we could discern no direct effect of Pi at up to 30 m on rigor in permeabilized cardiomyocytes [Fig. 3(d)], consistent with observations by others in fibres (Smith and Steele, 1994; Ventura-Clapier and Veksler, 1994). Instead, we observed an interaction between the ATP hydrolysis products, in that Pi attenuated the promotion of rigor by ADP. Thus, in contrast to active tension, rigor is not inhibited directly by the inorganic phosphate that accumulates during ischaemia, possibly because release of MgADP becomes the main rate-limiting step in contracture development (Ventura-Clapier and Veksler, 1994). Rather, data such as those in Table 1 indicate that Pi inhibits rigor indirectly. We also found that increasing MgADP concentration stimulated myosin ATPase activity in parallel with rigor in cardiomyocytes (Fig. 5). This accords with reports that ADP enhances both development of rigor tension in myocardium (Miller and Smith, 1985) and ventricular fibres (Ventura-Clapier and Veksler, 1994), as well as calcium-independent ATPase activity in skeletal fibres (Shimizu et al., 1992).
1356
M. T. Stapleton and A. P. Allshire
MgADP appears to hinder competitively the dissociation of rigor complexes by MgATP, with the result that the positive co-operativity provided by rigor complex activation of actin-myosin interaction overrides possible product inhibition of myosin ATPase in the cardiac myofilament. The Second Law of Thermodynamics demands that free energy of ATP hydrolysis, DGATP , rather than absolute concentration of adenine nucleotide is the cardinal determinant of cross-bridge cycling (Kammermeier et al., 1982). DGATP in the aerobic beating heart is estimated at −60 kJ/mol (Kammermeier et al., 1982), falling to about −30 kJ/ mol after 30 min global ischaemia in rat heart (Fiolet et al., 1985), which is close to the energy requirement for completion of the cross-bridge cycle (Sleep and Smith, 1981). Nevertheless, over the DGATP range −26 to −51 kJ/mol in the bulk phase we saw no correlation with rigor. Yet it was unlikely that a substantial gradient of DGATP existed in the system, since phosphocreatine was absent and AK had little effect on the ATP:ADP ratio at the myofilaments (see below). By contrast, rigor was sensitive to changes in ADP:Pi ratio at constant DGATP . For example, under conditions of 1 m MgATP and DGATP of −26 kJ/mol, rigor frequency ranged from 8±9% (n=4, with 3 m ADP and 2 m Pi) through 18±10% (n=4, with 2 m ADP and 3 m Pi) to 55±15% (n=4, with 4 m ADP and 1.5 m Pi). At higher DGATP (−37 kJ/mol), a similar pattern was observed. Hence, ADP rather than DGATP was the primary determinant of rigor, but Pi antagonized this effect increasingly as ADP concentration was raised. For example, rigor frequency observed at 1 m MgATP and 4 m MgADP rose from 55±15% in the presence of 1.5 m Pi to 98±2% (n=4, P<0.01) in its absence. Thus, overall, while inorganic phosphate exerted little or no direct effect on rigor it strongly moderated the promotion of rigor by ADP. In principle, ATP hydrolysis to ADP in the vicinity of myofilaments might be countered by the adenylate kinase reaction that produces AMP, and myocardial levels of AMP have been reported to rise during ischaemia (Geisbuhler et al., 1984; Neely and Grotyohann, 1984). However, we found that physiological concentrations of AMP had no effect on either rigor or myosin ATPase activity; only when AMP concentration was raised to 30-fold that of ATP was rigor enhanced. Since in ischaemic heart the molar excess of AMP over ATP appears to be 10-fold (Geisbuhler et al., 1984) or less (Neely and Grotyohann, 1984), it is unlikely that AMP is an important regulator of rigor. The extent to which adenylate kinase (AK) might
modulate the adenine nucleotide environment in cardiac myofibrils remains obscure. Evidence from skeletal muscle suggests that AK plays a role at both the mitochondrial and myofibrillar poles of the phosphocreatine shuttle (Bessman and Carpenter, 1985), with phosphoryl transfer flux through AK increasing during contractile activity (Zeleznikar et al., 1990). In skinned cardiac muscle exposed to creatine phosphate, rigor tension is unaffected by diadenosine pentaphosphate (Ventura-Clapier and Vassort, 1985; Ventura-Clapier and Veksler, 1994), but under these conditions a role for AK is likely to be masked by ADP phosphorylation by CK. When creatine phosphate is depleted, as in ischaemic myocardium, it remains conceivable that AK might postpone rigor by attenuating the fall in ATP:ADP ratio at the myofilaments. However, data in Figure 4 suggest that intrinsic adenylate kinase activity in permeabilized cardiomyocytes caused little or no deviation of myofibrillar ATP/ADP from that in the bulk phase, since inhibition of the enzyme did not affect rigor. On this basis, it appears unlikely that myofibrillar AK plays any significant role in modulation of ischaemic contracture, at least under near steady-state conditions.
Relevance to ischaemia The finding that in the presence of MgADP the myofibrillar array in cardiomyocytes begins to enter rigor at MgATP concentrations in the low millimolar range, is consistent with direct measurements of cytosolic ATP in individual cells undergoing deenergization contracture (Allue et al., 1996). Since rigor development stimulates myosin ATPase activity, a positive feed-back loop is established that could drive the abrupt depletion of residual cytosolic ATP observed in individual cardiomyocytes that coincides with de-energization contracture (Bowers et al., 1992; Allue et al., 1996). Following deenergization-induced contracture in intact cardiomyocytes diastolic calcium then rises, priming the cells for reoxygenation injury in the form of a powerful calcium- and ATP-activated hypercontracture (Stern et al., 1985; Allshire et al., 1987). Of the metabolic changes in ischaemic myocardium, increased ADP is likely to be the primary determinant of onset of rigor. Data in Figure 6 suggest that rigor becomes significant at millimolar ATP as ADP enters the high micromolar range. In hypoxic or ischaemic myocardium ADP levels rise to high micromolar (Kammermeier et al., 1982) or millimolar (Neely and Grotyohann, 1984;; Geisbuhler et al., 1984; Vander Heide et al., 1996b),
Rigor and Myosin ATPase in Cardiomyocytes
according to the method of measurement, while tissue ATP:ADP ratio is substantially decreased (Neely and Grotyohann, 1984; Bak and Ingwall, 1994). However, ADP stimulation of rigor is antagonized by inorganic phosphate and prevented under acidotic conditions. Since these factors may be less prominent in superfused cardiomyocytes subjected to metabolic inhibition with cyanide and 2-deoxyglucose, the estimate of a 1–2 m ATP threshold for onset of rigor (Allue et al., 1996) could actually represent an upper limit in the case of the ischaemic myocardium. Subsequent progression of rigor contracture will then be determined by the dynamic balance between development of acidosis and changes in concentration of ADP and inorganic phosphate. In conclusion, we show that at physiological temperature rigor development in cardiomyocytes entails an activation of myosin ATPase. This accords with the hypothesis that rigor dramatically accelerates ATP depletion in de-energized cells (Allshire and Cobbold, 1989), and demonstrates the likely mechanism underlying the abrupt loss of cytosolic ATP in individual cardiomyocytes undergoing contracture (Bowers et al., 1992). Consequently, rigor may represent a key target for interventions aimed at enhancing survival of the ischaemic myocardium.
Acknowledgements This study was supported by the Irish Heart Foundation and undertaken within the ambit of EU Biomed II Reinforced Concerted Action PL-951254.
References A AP, C PH, 1989. Cytosolic Ca2+ in hypoxic cardiomyocytes. In: Piper HM, Isenberg I (eds). Isolated Adult Cardiomyocytes. Boca Raton: CRC Press, 287–308. A AP, P HM, C KSR, C PH, 1987. Cytosolic free Ca2+ in single rat heart cells during anoxia and reoxygenation. Biochem J 244: 381–385. A I, G O, D E, U N, C P, 1996. Evidence for rapid consumption of millimolar concentrations of cytoplasmic ATP during rigor-contracture of metabolically compromised single cardiomyocytes. Biochem J 319: 463–469. A RA, H JR, B GP, 1981. Response of isolated rat heart cells to hypoxia, reoxygenation, and acidosis. Circ Res 49: 307–316. A SC, G CE, 1991. Effects of 2,3-butanedione monoxime (BDM) on contracture and injury
1357
of isolated rat myocytes following metabolic inhibition and ischemia. J Mol Cell Cardiol 23: 1001–1014. B MI, I JS, 1994. Acidosis during ischaemia promotes adenosine triphosphate resynthesis in postischaemic rat heart. J Clin Invest 93: 40–49. B SP, C CL, 1985. Creatine phosphate shuttle. Annu Rev Biochem 54: 831–862. B KC, A AP, C PH, 1992. Bioluminescent measurement in single cardiomyocytes of sudden cytosolic ATP depletion coincident with rigor. J Mol Cell Cardiol 24: 213–218. B RD, W AM, 1972. Cooperation within actin filament in vertebrate skeletal muscle. Nature 238: 97–101. B SPJ, S KB, 1992. Bound and determined: a computer program for making buffers of defined ion concentrations. Anal Biochem 201: 119–126. C KM, D D, J KD, 1986. A direct colorimetric assay for Ca2+-stimulated ATPase activity. Anal Biochem 157: 375–380. C CM, R MA, S EL, G NE, 1995. Magnesium sulfate reduces myocardial infarct size when administered prior to but not after coronary reperfusion in a canine model. Circulation 92: 2617–2621. C R, P E, 1985. The effect of ADP and phosphate on the contraction of muscle fibres. Biophys J 48: 789–798. E JP, S GJM, E G, 1994. Influence of phosphate and pH on myofibrillar ATPase activity and force in skinned cardiac trabeculae from rat. J Physiol 476: 501–516. E AC, S GL, E DA, A DG, 1992. Metabolic changes during ischaemia and their role in contractile failure in isolated ferret hearts. J Physiol 454: 467–490. F A, F F, 1975. Effects of magnesium on contractile activation of skinned cardiac cells. J Physiol 249: 497–517. F SAE, B A, S CA, C R, W HF, 1985. The change of the free energy of ATP hydrolysis during global ischaemia and anoxia in the rat heart. J Mol Cell Cardiol 16: 1023–1036. G T, A RA, T RW, A AZ, L K, B G, 1984. Adenine nucleotide metabolism and compartmentalization in isolated adult rat heart cells. Circ Res 54: 536–546. H RA, H DR, B HA, 1981. Contracture in isolated adult rat heart cells: role of Ca2+, ATP and compartmentation. Circ Res 49: 1119–1128. H JP, W RJ, 1991. Cytosolic free magnesium in stimulated, hypoxic and underperfused rat heart. J Mol Cell Cardiol 23: 991–999. H WR, S ML, MM KS, E LR, G MJ, V RA, S VL, 1995. Timing of magnesium therapy affects experimental infarct size. Circulation 92: 2622–2626. H C, A NR, G R, M LA, 1982. Heat production during hypoxic contracture of rat myocardium. Circ Res 51: 777–786. K H, S P, J¨ E, 1982. Free energy change of ATP hydrolysis: a causal factor of early hypoxic failure of the myocardium. J Mol Cell Cardiol 14: 267–277. K JC, 1986. The effects of inorganic phosphate
1358
M. T. Stapleton and A. P. Allshire
and creatine phosphate on force production in skinned muscles from rat ventricle. J Physiol 370: 585–604. K Y, M E, 1990. Mechanism of ischemic contraction in ferret hearts: relative roles of [Ca2+]i elevation and ATP depletion. Am J Physiol 258: H9– H16. L JA, A DG, 1991. Mechanisms of acute ischaemic contractile failure of the heart. J Clin Invest 88: 361– 367. M DJ, S GL, 1985. The contractile behaviour of EGTA and detergent-treated heart muscle. J Musc Res Cell Motil 6: 541–567. N JR, G LW, 1984. Role of glycolytic products in damage to ischemic myocardium: dissociation of adenosine triphosphate levels and recovery of function of reperfused ischemic hearts. Circ Res 55: 816–824. N CG, L WJ, 1990. The role of ATP in energydeprivation contractures in unloaded rat ventricular myocytes. Can J Physiol Pharmacol 68: 183–194. P WS, 1991. The effects of ATP, inorganic phosphate, protons, and lactate on isolated myofibrillar ATPase activity. Can J Physiol Pharmacol 70: 1175– 1181. S H, F T, I S, 1992. Regulation of tension development by MgADP and Pi without Ca2+. Biophys J 61: 1087–1098. S HS, D L D, H RC, M H, S SJ, H RG, L EG, S MD, 1994. Regulation of intracellular free Mg2+ and contraction in single adult mammalian cardiac myocytes. Am J Physiol 266: C222–C233.
S JA, S SJ, 1981. Actomyosin ATPase and muscle contraction. Curr Top Bioenerg 11: 239–286. S GL, S DS, 1994. Effects of pH and inorganic phosphate on rigor tension in chemically skinned rat ventricular trabeculae. J Physiol 478: 505–512. S M, D P, C P, A A, 1995. Troponin-I137–148 peptide abolishes rigor in cardiomyocytes. Circulation 92: I–189 (Abstract). S MD, C AM, C MC, P DJ, L EG, 1985. Direct observation of the ‘oxygen paradox’ in single rat ventricular myocytes. Circ Res 56: 899–903. V H RS, D JA, J RB, R KA, S C, 1996a. Reducing lactate accumulation does not attenuate lethal ischemic injury in isolated perfused rat hearts. Am J Physiol 270: H38–H44. V H RS, H ML, R KA, J RB, 1996b. Effect of reversible ischemia on the activity of the mitochondrial ATPase: relationship to ischemic preconditioning. J Mol Cell Cardiol 28: 103–112. V-C R, V G, 1985. Role of myofibrillar creatine kinase in the relaxation of rigor tension in skinned cardiac muscle. Pflu¨gers Arch 404: 157–161. V-C R, V V, 1994. Myocardial ischemic contracture:metabolites affect rigor tension development and stiffness. Circ Res 47: 920–929. Z RJ, H RA, G RM, W TF, D SM, B EA, G ND, 1990. Evidence for compartmentalized adenylate kinase catalysis serving a high energy phosphoryl transfer function in rat skeletal muscle. J Biol Chem 265: 300–311.
Modulation of Rigor and Myosin ATPase Activity in Rat Cardiomyocytes Mary T. Stapleton and Ashley P. Allshire Department of Pharmacology and Therapeutics, University College Cork, University Hospital, Cork, Ireland (Received 26 January 1998, accepted in revised form 31 March 1998) M. T. S A. P. A. Modulation of Rigor and Myosin ATPase Activity in Rat Cardiomyocytes. Journal of Molecular and Cellular Cardiology (1998) 30, 1349–1358. Ischaemic myocardium undergoes calciumindependent contracture at millimolar tissue ATP, though in actomyosin solutions ATP must be reduced to micromolar before rigor complexes form. This contracture is associated with myosin ATPase activity that may contribute to tissue de-energization. Here we used isolated rat cardiomyocytes permeabilized with digitonin to analyse in parallel how rigor and myosin ATPase activity are modulated by metabolic conditions that develop during ischaemia. At pH 7.1 and 37°C rigor and myosin ATPase showed co-ordinated bell-shaped dependence on ATP concentration over 3–1000 l. Rigor, but not myosin ATPase, was inhibited by acidosis (pH 6.2), indicating reduced efficiency of cross-bridge cycling, while both parameters were stimulated by ADP (Ζ1 m) and unaffected by inorganic phosphate (Pi, 30 m), AMP, Mg2+, lactate or inhibition of adenylate kinase with diadenosine pentaphosphate. Combined acidosis and high ADP inhibited rigor, while Pi attenuated the enhancement of rigor by ADP. Thus, rigor complex formation activates myosin ATPase in the intact myofilament array, modulated by ADP, Pi and acidosis in the ranges that occur in ischaemia. There was no evidence that adenylate kinase might attenuate falling ATP/ADP ratio at the myofilaments. In combination these effects are sufficient to resolve the apparent discrepancy between ATP concentrations triggering rigor in actomyosin and onset of contracture in ischaemic myocardium. Since rigor contracture activates myosin ATPase it is likely to exacerbate ATP depletion and thereby limit vital cell functions. This positive feedback is consistent with the abrupt depletion of ATP observed in individual cardiomyocytes undergoing deenergization contracture. 1998 Academic Press K W: Ischaemia; Rigor; Myosin ATPase; ADP; Acidosis; Rat cardiomyocytes; Free energy of ATP hydrolysis; Troponin I.
Introduction In cardiac ischaemia blood flow through the myocardium is insufficient to meet the oxygen demand of the tissue. Hypoxia restricts aerobic regeneration of ATP, while poor wash-out of catabolites may exacerbate tissue dysfunction. Shortly after onset of ischaemia contractility in the affected region wanes then ceases, whereupon a contracture develops largely independent of calcium. Eventually myocytes in the poorly-irrigated tissue sustain irreversible injury and an infarct forms. The role of this contracture in the progression to
tissue death has been studied extensively but a clear understanding of the mechanism involved has not yet emerged. Diastolic pressure rises in ischaemic myocardium despite little calcium-activated muscle contraction (Holubarsch et al., 1982), while shortening of individual cardiomyocytes under anoxia precedes any significant rise in cytosolic free calcium (Allshire et al., 1987). In vitro, actin and myosin interact independent of calcium when ATP levels are low (10–100 l) by forming nucleotide-free rigor cross bridges that activate myosin ATPase and crossbridge cycling (Bremel and Weber, 1972), whereas onset of contracture in intact myocardium and car-
Please address all correspondence to: Dr Ashley Allshire, Department of Pharmacology and Therapeutics, University College Cork, University Hospital, Cork, Ireland.
0022–2828/98/071349+10 $30.00/0
1998 Academic Press
1350
M. T. Stapleton and A. P. Allshire
diomyocytes occurs at about 1–2 m ATP (Altschuld et al., 1981; Nichols and Lederer, 1990; Elliott et al., 1992). This may reflect distinct behaviour of myosin ATPase and the cross-bridge cycle in actomyosin solution compared to that in the highly-ordered microenvironment of the myofilaments in situ. In addition, the calcium-independent contracture in ischaemic heart may be modulated by prevailing metabolic conditions. Ischaemia induces a switch from aerobic to anaerobic metabolism, depletion of tissue phosphocreatine and then ATP, and accumulation in the cytosol of catabolites including lactate, ADP, AMP, magnesium ion, inorganic phosphate and protons. Furthermore the in vitro behaviour of actomyosin suggests that rigor activation of myosin ATPase should exacerbate ATP depletion, though as far as we are aware this has not been demonstrated in heart muscle. However this would most readily explain the observation (Haworth et al., 1981; Bowers et al., 1992; Allue et al., 1996) that cardiomyocyte shortening coincides precisely with rapid depletion of cytosolic ATP. The aim of the present study was to identify factors likely to influence rigor in ischaemic myocardium. Cardiomyocytes were permeabilized with digitonin in order to expose the intact myofilament array at 37°C to conditions reflecting the intracellular ionic and nucleotide milieu. This allowed rigor indicated by cell shortening independent of calcium to be measured in parallel with myosin ATPase activity in terms of inorganic phosphate released from MgATP in the presence of inhibitors of the main non-myosin ATPases. At low MgATP myosin ATPase activity correlated with rigor development. Of the metabolites that accumulate in ischaemia, only ADP and protons affected rigor directly.
Materials and Methods Myocytes Ventricular myocytes were isolated from male Wistar rats (200–300 g) by Langendorff perfusion with collagenase (Worthington Biochemical Corporation, NJ, USA), as described previously (Allshire et al., 1987). The preparation contained 59±8% rod-shaped cells (n=25) and was maintained at 37°C in medium 199 (ICN/Flow, UK) under 5% CO2 in air, and used within 4 h of isolation. Cell permeabilization Cardiomyocytes harvested from 1.5-ml portions of the culture suspension by gentle centrifugation
(20 g×60 s) were resuspended in an equal volume of intracellular type medium (ICM) comprising (in m): 164 KCl, 3.5 EGTA, 1.02 MgCl2 and 30 3-[Nmorpholino]propane-sulfonic acid (MOPS), adjusted to pH 7.10 with KOH, at 37°C. After 3 min at 37°C, the cells were cooled to 2°C over 5 min to prevent rigor development at permeabilization as endogenous ATP levels fell. During this period cells were triturated gently at intervals to prevent settling, then harvested and resuspended in 1.5 ml of icecold ICM supplemented with digitonin (2 lmol/mg protein) for 5 min before being washed in four changes of an equal volume of detergent-free ICM. Finally a sample was removed for measurement of rigor frequency and cell permeabilization, immediately before the microfuge tube containing the cell pellet was transferred to a water bath at 37°C. Twenty seconds later the assay was begun by resuspending the cells in ATP-containing solution pre-equilibrated at 37°C, then 250-ll samples of the suspension were taken at 10 and 30 s for measurement of myosin ATPase activity and a 20-ll sample at 40 s for measurement of rigor frequency.
Solutions Composition of solutions containing Ca2+, Mg2+ and multiple ligands were calculated using the computer program of Brooks and Storey (1992). Intracellular-type media were prepared from a basic stock comprising (in m): 30 MOPS, 3.5 EGTA and 1 dithiothreitol (DTT), to which were added ATP (sodium salt), MgCl2, and KCl to give defined MgATP concentrations and total ionic strength of 180 m after pH had been adjusted to 7.10 at 37°C with KOH. Free magnesium concentration was set at 0.5 m and free calcium at 3 n. A pH of 7.10 was chosen to match that in normoxic myocardium. When other metabolites were included (inorganic phosphate, lactate, magnesium, ADP, AMP), or when pH was lowered, buffer composition was adjusted in order to maintain MgATP, free Mg2+ and Ca2+, and ionic strength. Chemicals were obtained from BDH Ltd, Poole, UK, and Sigma Chemical Co., St Louis, MO, USA.
Rigor Samples of cell suspensions (20 ll) were fixed and stained by addition to an equal volume of glutaraldehyde/trypan blue dye solution (each 1% w/ v), then scored for rigor (shortening to length:width Ζ3; Armstrong and Ganote, 1991) and dye uptake
1351
Rigor and Myosin ATPase in Cardiomyocytes
ATPase activity was measured as inorganic phosphate (Pi) release from ATP, using the malachite green assay (Chan et al., 1986). Major non-myosin ATPases were inhibited by exposing permeabilized cells to a cocktail containing 50 l each of ouabain (Na+,K+-ATPase), oligomycin (mitochondrial F0F1ATPase) and cyclopiazonic acid (SR Ca2+-ATPase) in ICM with 1 m DTT for 10 min at 2°C before transfer to ATP-containing solution supplemented with inhibitors at 37°C. Correction was made for spontaneous ATP hydrolysis, always <10% of that in the presence of cells. We verified that residual ATP hydrolysis represented myosin ATPase activity using a peptide mimetic of the troponin-I switch sequence (Tn-I137–148). In experiments where cells had not been pre-incubated with the inhibitor cocktail, Tn-I137–148 (114 l) attenuated rigor and ATPase activity by 96±8% (n=4) and 88±5% (n= 7), respectively, at 56 l MgATP, while for both parameters half-maximal inhibition occurred at 22 l peptide (Stapleton et al., 1995). However at 1 m MgATP a proportion of the ATPase activity was refractory to Tn-I137–148, indicating a growing contribution by ATPase(s) other than myosin to Pi generation as ATP concentration approached millimolar.
Protein Protein content of cell suspensions was measured by the Bio-Rad assay, using bovine serum albumin standards.
75
75
50
50
*
25
0
5.5
5.0
†
4.5
4.0 3.5 p MgATP
25
3.0
2.5
Percent cells in rigor ( )
Myosin ATPase
100
100 ATPase activity (pmol Pi/ g protein/min) ( )
within 1 h, over which time these parameters remained stable. Typically 100–200 cells were scored per sample. As the proportion of rod-shaped cells varied between individual preparations, rigor frequency was normalized.
0
Figure 1 Parallel development of rigor and myosin ATPase activity in permeabilized cardiomyocytes at 37°C. Rigor was defined as cell length:width ratio <3 and myosin ATPase activity was measured as inorganic phosphate release in the presence of non-myosin ATPase inhibitors (see Materials and Methods for details). Datapoints are means±.. (n=4–10); ∗ P<0.05, † P<0.01.
Results Development of rigor At physiologically relevant temperature, pH, osmotic strength and ionic milieu permeabilized rat cardiomyocytes underwent rigor contracture independent of calcium at millimolar to micromolar MgATP. However, to facilitate detection of altered rigor behaviour, we adopted a standard exposure to MgATP of 40 s that caused all cells to shorten over a narrow MgATP range (about 20–60 l) but fewer at very low or high concentrations (6% at 3 l; 8% at 1.8 m; Fig. 1). Myosin ATPase activity measured in parallel followed the general pattern of rigor development, namely a peak activity at 30 l MgATP (Fig. 1). Yet while ATPase activity at pMgATP 4.0 was similar to that at 4.75 rigor was more extreme in the latter, and at pMgATP 3.0 rigor was low but ATP hydrolysis high. This underlying trend of increasing hydrolysis with MgATP concentration that was independent of rigor reflects residual activity of nonmyosin ATPases (see Materials and Methods).
Statistical analysis Values are expressed as means ±.. Data sets were compared by Wilcoxon–Mann–Whitney and unpaired t-tests using the Astute statistical package (DDU Software, University of Leeds, UK), taking significance as PΖ0.05. For clarity, some error bars in the figures are shown in one direction only.
Effect of ischaemic metabolites on rigor and myosin ATPase
Acidosis and lactate Reduction of pH from 7.1 to 6.8 and 6.2 progressively and strongly suppressed rigor in the
1352
M. T. Stapleton and A. P. Allshire
Percent cells in rigor
100 †
80
†
60 40
* †
20 0
† 5.5
5.0
†
4.5 4.0 p MgATP
* 3.5
3.0
Figure 2 Inhibition of rigor in cardiomyocytes on decreasing pH from 7.1 to 6.2. Data points are means±.. (n=4–8); ∗ P<0.05, † P<0.01 v control. (Χ) pH 7.1; (Φ) pH 6.8; (Α) pH 6.2.
MgATP range 0.03–1 m (pMgATP 4.5–3.0), while effects at lower MgATP were minimal (Fig. 2). However at pMgATP 4.0 where mean rigor frequency was reduced from 66% at pH 7.1 to 1% at pH 6.2 (P<0.01) ATPase activity remained unchanged [56±19 (n=6) v 42±13 pmol Pi/lmol protein/ min (n=3) at pH 7.1 and 6.2, respectively; P= 0.40]. By contrast, lactate had no effect on either rigor [Fig. 3(b)] or ATPase activity [measured at pMgATP 4.0: 56±19 (n=6) v 80±25 pmol Pi/ lmol protein/min (n=4) in absence and presence of 30 m lactate, respectively; P=0.19].
ATP hydrolysis products Hydrolysis of MgATP releases magnesium ion, and free magnesium reduces myofilament calcium sensitivity (Fabiato and Fabiato, 1975). However increasing free magnesium concentration from 0.5 to 3.0 m had no effect on rigor other than to enhance it somewhat at pMgATP 4.0 [100 l; Fig. 3(c)], nor on ATPase activity [56±19 (n=6) v 80±23 pmol Pi/lmol protein/min (n=4) at 0.5 and 3 m free Mg2+, respectively; P=0.17]. Net hydrolysis of creatine phosphate and ATP causes a substantial increase in cytosolic inorganic phosphate (Pi). However no effect on rigor contracture of permeabilized cells could be discerned at either 15 m (not shown) or 30 m Pi [Fig. 3(d)]. The effect of MgADP on rigor was studied at pMgATP 4.0 and 3.0 (0.1 and 1.0 m MgATP), concentrations at which rigor was close to halfmaximal (66±9%) and minimal (13±9%), respectively. At both MgATP concentrations increasing MgADP promoted rigor strongly once ATP: ADP fell below a threshold of about 1.3 (Fig. 4).
At 1 m MgATP MgADP increased myosin ATPase activity parallel to its effect on rigor (Fig. 5). However inhibition of adenylate kinase (AK) with diadenosine pentaphosphate (AP5A; 10 or 100 l) did not increase rigor further, suggesting that the enzyme did not greatly affect the ATP:ADP ratio within the myofibrillar array (Fig. 4). On this basis, neither could we discern an effect of myofibrillar adenylate kinase on rigor at 0.1 m MgATP and either 25 or 50 l MgADP in the presence of inorganic phosphate (30 m), phosphocreatine (12 m) or at pH 6.2 (data not shown). Although endogenous AK did not appear to alter the nucleotide environment of the myofilaments under the conditions used here, in principle the AK reaction might counter falling ATP:ADP and produce AMP in the process. Therefore we tested the effect of AMP on myofibril response to MgATP. Neither rigor nor ATPase activity were affected by 0.12 m AMP (0.004 m MgAMP) at 0.1 m MgATP or by 1.2 m AMP (0.04 m MgAMP) at 1 m MgATP (data not shown). However at 0.1 m MgATP 3 m AMP stimulated both rigor (from 66±9 to 100±0%, P<0.01) and ATPase activity (from 86±2 to 115±20 pmol/lg protein/min, P<0.05). It may be that under the latter conditions high AMP drove ADP production by reversing the AK reaction at the expense of ATP, and thereby depressed ATP:ADP at the myofilaments. Since increasing ADP concentration potentiated rigor while increasing Pi did not, we tested whether rigor was regulated by DGATP . Varying DGATP between −26 and −51 kJ/mol had essentially no effect on rigor frequency (r=0.03; Table 1). Thus, in the presence of sufficient MgATP to support crossbridge cycling, the primary regulation of rigor was allosteric (by MgADP) rather than thermodynamic. For example, addition of Pi in the presence of MgADP, which would reduce DGATP , tended to inhibit rather than promote rigor at progressively higher ADP concentrations (Table 1).
Combined ADP and acidosis In ischaemic myocardium lactate accumulation and net ATP hydrolysis will contribute to development of acidosis. As protons and MgADP separately had opposing effects on rigor, we tested their net effect in combination (Fig. 6). Rigor was essentially complete at both 0.1 and 1 m MgATP (pMgATP 4 and 3, respectively) in the presence of equimolar MgADP. However, at both MgATP concentrations, reducing pH by 0.3 unit more than halved the extent of rigor, while at pH 6.2 the extent of rigor was no more than 10% of that at pH 7.1. This
1353
Rigor and Myosin ATPase in Cardiomyocytes 100
(a)
75
Percent cells in rigor
Percent cells in rigor
100
50
(b)
75
50
25
25
0 30 mM 0
4.5
5.5
0
2.5
75
50
25
4.5 p MgATP
5.5
3.5
100
(c)
*
Percent cells in rigor
Percent cells in rigor
100
3.5 p MgATP
(d)
75 0 30 mM 50
25 0.5 mM 3 mM
0
4.5
4.0
3.5 p MgATP
3.0
0
5.5
4.5
3.5 p MgATP
2.5
Figure 3 Rigor frequency in isolated cardiomyocytes after 40 s at 37°C, control (a) is largely unaffected by addition of 30 m lactate (b), 3 m free Mg2+ (c) or 30 m Pi (d). Data points are means±.. (n=4–8); ∗ P<0.01 v control.
pattern was not affected by the further addition of 30 m each Pi and lactate (data not shown).
ATPase activity, while rigor was inhibited by acidosis. No other metabolites that accumulate in ischaemic myocardium had any significant effect.
Discussion Rigor and myosin ATPase activity In this study, permeabilized cardiomyocytes were used to develop a model of ischaemic contracture in which both rigor (calcium-independent cell shortening) and myosin ATPase (Pi release) could be measured in parallel. It provides the first direct evidence of which we are aware that rigor is accompanied by increased myosin ATPase activity in intact arrays of cardiac myofilaments, under physiologically-relevant conditions of temperature, pH, ionic strength and milieu (other than very low free calcium). ADP enhanced both rigor and myosin
Actin binds the myosin head strongly in the absence of nucleotide. This rigor complex is formed transiently during the normal cross-bridge cycle; however, during the intermediate state of lowest free energy, rigor complexes accumulate when ATP levels fall below the threshold required to dissociate them. In dilute solutions of actomyosin, rigor complexes “turn on” adjacent actin monomers not combined with myosin in the absence of calcium. However, it does not automatically follow that the
1354
M. T. Stapleton and A. P. Allshire
80 60
*
40 20 0
0
100 Percent cells in rigor
Table 1 Rigor frequency in permeabilized cardiomyocytes at 37°C as a function of free energy of ATP hydrolysis, DGATP , at 1 m MgATP. In the absence of added ADP or Pi values were arbitrarily set at 1% of total ATP and 0.025 m, respectively, in order to estimate DGATP . Rigor frequency is compared without and with added Pi at each ADP concentration, where ∗ P<0.05; † P<0.01; .., not significant.
(a)
20
40 60 MgADP ( M)
80
100
(b)
80 60 40 20 0
0.00
0.25
0.50 0.75 1.00 MgADP (mM)
1.25
† 150
75
100
50
*
25
50
0 0.00
0.25
0.50 0.75 MgADP (mM)
1.00
Percent cells in rigor ( )
ATPase activity (pmol Pi/ g protein/min) ( )
100
†
Pi (m)
−DGATP (kJ/mol)
0.015 0.015 2.0 2.0 3.0 3.0 4.0 4.0
0.025 5.0 0.025 3.0 0.025 2.0 0.025 1.5
51 37 38 26 37 26 36 26
0 1.25
Figure 5 Promotion of rigor and myosin ATPase activity by MgADP at 1 m MgATP. Datapoints are means±.. (n=4–10); ∗ P<0.05, † P<0.01 v absence of MgADP.
same happens within a cardiomyocyte where actin and myosin are sterically constrained in the lattice array of the myofilament. Demonstration of a direct association between rigor and myosin ATPase activity in the intact myofilament is hampered by background ATP hydrolysis. Here, we equated myosin ATPase activity with total inorganic phosphate released from ATP in the presence of inhibitors of the other major
% cells in rigor mean±.. (n) 13±9 23±4 20±17 8±9 45±9 18±10 98±2 55±15
(8) (4) (4) (4) (4) (4) (4) (4)
P ∗ .. † †
100
1.50
Figure 4 Promotion of rigor by MgADP at 0.1 (a) and 1 m (b) MgATP, and the effect of adenylate kinase inhibition with diadenosine pentaphosphate (AP5A; 10 l). Data points are means±.. (n=4); ∗ P<0.05, v control. (Χ) Control; (Β) +AP5A.
200
ADP (m)
80 Percent cells in rigor
Percent cells in rigor
100
60
40
20
0
7.1
6.8
6.5
6.2
pH
Figure 6 Acidosis impairs the enhancement of rigor by MgADP at both 1 (Φ) and 0.1 m (∆) MgATP in the presence of equimolar MgADP. Bars are means±.. (n= 4) and effects are significant at pH Ζ6.8 v pH 7.1 (P<0.01).
ATPases in cardiomyocytes, i.e. the mitochondrial F0F1-ATPase, the Na+,K+-ATPase and the sarcoplasmic reticulum Ca2+-ATPase. That mimetic TnI137–148 almost completely inhibited both rigor and ATPase hydrolysis, and with the same potency, indicates that this assumption is valid. In addition, rigor and residual ATPase activity both showed bell-shaped dependence on MgATP over the same concentration range (Fig. 1), and ADP enhanced both rigor and ATP hydrolysis (Fig. 5), whereas the opposite effect would be expected of an ATP
Rigor and Myosin ATPase in Cardiomyocytes
hydrolysis product on the basis of mass action. Overall, we observed that net ATPase activity clearly attributable to myosin matched the incidence of rigor, both showing a maximum over the range 20–50 l MgATP, which is consistent with earlier reports of rigor in cardiomyocytes and myocardium (Fabiato and Fabiato, 1975; Haworth et al., 1981; Nichols and Lederer, 1990). Furthermore, the biphasic incidence of rigor in the intact myofilament array as a function of MgATP matches qualitatively the behaviour of skeletal actomyosin in solution (Bremel and Weber, 1972), with the ascending phase of the curve probably reflecting substrate affinity of myosin ATPase and the descending phase corresponding to progressive saturation.
Effect of ischaemic metabolites While the effects of ischaemic metabolites on calcium activated force in heart muscle are well characterized, effects on rigor force development are less extensively documented (Silverman et al., 1994; Smith and Steele, 1994; Ventura-Clapier and Veksler, 1994). To our knowledge, the present study is the first to address the impact of ischaemic metabolites on combined rigor and myosin ATPase activity in cardiac myofilaments. Lactic acid dissociation and H+ release from ATP hydrolysis during myocardial ischaemia causes intracellular pH to fall rapidly by about 1 unit. This acidosis inhibits calcium-induced contraction and calcium influx (Altschuld et al., 1981), which effects may lead to conservation of ATP and adenine nucleotide (Bak and Ingwall, 1994). Here, acidosis progressively and powerfully inhibited rigor in permeabilized cardiomyocytes (Fig. 3), as it does in chemically skinned rat ventricular fibres (VenturaClapier and Veksler, 1994) and trabeculae (Smith and Steele, 1994). That acidosis decreases both the number of cross-bridges, and particularly the force per cross-bridge, could reflect a direct effect on myosin ATPase (Ventura-Clapier and Veksler, 1994). However, we found that myosin ATPase activity associated with rigor in permeabilized cardiomyocytes was similar at pH 7.2 and 6.2, while the effect of acidosis on myofibrillar ATPase activity in skinned rat trabeculae undergoing calcium-induced contraction was also reported to be small (Ebus et al., 1994). Hence, acidosis considerably reduces the energy efficiency not only of calciumdependent but also of rigor cross-bridge cycling in the cardiac myofilament. This argues against the suggestion of Ventura-Clapier and Veksler (1994) that rigor might provide sustained tension at low
1355
energy cost in ischaemic myocardium. Finally, the lack of effect of lactate on either rigor or myosin ATPase in permeabilized cardiomyocytes is consistent with the conclusion that injury associated with its accumulation in ischaemic myocardium is actually due to acidosis (Neely and Grotyohann, 1984; Vander Heide et al., 1996a). During hypoxia, cytosolic free magnesium rises to 2.8 m in rat ventricular myocytes (Silverman et al., 1994), commensurate with the depletion of its main intracellular ligand ATP. A comparable rise to 2.3 m in intact rat heart (Headrick and Willis, 1991) coincides with onset of contractile failure and reaches a plateau as rigor contracture develops. However, despite its negative inotropic effect on calcium-activated contraction, increasing free magnesium had only marginal effects on rigor and associated myosin ATPase in permeabilized cardiomyocytes. Therefore, the reduction in infarct size reported in animal studies following magnesium administration at the time of occlusion (Christensen et al., 1995) or reperfusion (Herzog et al., 1995) is unlikely to be due to attenuation of rigor. Intracellular Pi increases rapidly in ischaemic myocardium (Koretsune and Marban, 1990), which together with developing acidosis contributes to contractile failure (Kentish, 1986; Lee and Allen 1991). It is well established that Pi inhibits myofibrillar Ca2+-activated ATPase activity (Cooke and Pate, 1985; Parkhouse, 1991; Ebus et al., 1994) and depresses Ca2+-dependent tension development (Cooke and Pate, 1985; Kentish, 1986; Ebus et al., 1994). However, we could discern no direct effect of Pi at up to 30 m on rigor in permeabilized cardiomyocytes [Fig. 3(d)], consistent with observations by others in fibres (Smith and Steele, 1994; Ventura-Clapier and Veksler, 1994). Instead, we observed an interaction between the ATP hydrolysis products, in that Pi attenuated the promotion of rigor by ADP. Thus, in contrast to active tension, rigor is not inhibited directly by the inorganic phosphate that accumulates during ischaemia, possibly because release of MgADP becomes the main rate-limiting step in contracture development (Ventura-Clapier and Veksler, 1994). Rather, data such as those in Table 1 indicate that Pi inhibits rigor indirectly. We also found that increasing MgADP concentration stimulated myosin ATPase activity in parallel with rigor in cardiomyocytes (Fig. 5). This accords with reports that ADP enhances both development of rigor tension in myocardium (Miller and Smith, 1985) and ventricular fibres (Ventura-Clapier and Veksler, 1994), as well as calcium-independent ATPase activity in skeletal fibres (Shimizu et al., 1992).
1356
M. T. Stapleton and A. P. Allshire
MgADP appears to hinder competitively the dissociation of rigor complexes by MgATP, with the result that the positive co-operativity provided by rigor complex activation of actin-myosin interaction overrides possible product inhibition of myosin ATPase in the cardiac myofilament. The Second Law of Thermodynamics demands that free energy of ATP hydrolysis, DGATP , rather than absolute concentration of adenine nucleotide is the cardinal determinant of cross-bridge cycling (Kammermeier et al., 1982). DGATP in the aerobic beating heart is estimated at −60 kJ/mol (Kammermeier et al., 1982), falling to about −30 kJ/ mol after 30 min global ischaemia in rat heart (Fiolet et al., 1985), which is close to the energy requirement for completion of the cross-bridge cycle (Sleep and Smith, 1981). Nevertheless, over the DGATP range −26 to −51 kJ/mol in the bulk phase we saw no correlation with rigor. Yet it was unlikely that a substantial gradient of DGATP existed in the system, since phosphocreatine was absent and AK had little effect on the ATP:ADP ratio at the myofilaments (see below). By contrast, rigor was sensitive to changes in ADP:Pi ratio at constant DGATP . For example, under conditions of 1 m MgATP and DGATP of −26 kJ/mol, rigor frequency ranged from 8±9% (n=4, with 3 m ADP and 2 m Pi) through 18±10% (n=4, with 2 m ADP and 3 m Pi) to 55±15% (n=4, with 4 m ADP and 1.5 m Pi). At higher DGATP (−37 kJ/mol), a similar pattern was observed. Hence, ADP rather than DGATP was the primary determinant of rigor, but Pi antagonized this effect increasingly as ADP concentration was raised. For example, rigor frequency observed at 1 m MgATP and 4 m MgADP rose from 55±15% in the presence of 1.5 m Pi to 98±2% (n=4, P<0.01) in its absence. Thus, overall, while inorganic phosphate exerted little or no direct effect on rigor it strongly moderated the promotion of rigor by ADP. In principle, ATP hydrolysis to ADP in the vicinity of myofilaments might be countered by the adenylate kinase reaction that produces AMP, and myocardial levels of AMP have been reported to rise during ischaemia (Geisbuhler et al., 1984; Neely and Grotyohann, 1984). However, we found that physiological concentrations of AMP had no effect on either rigor or myosin ATPase activity; only when AMP concentration was raised to 30-fold that of ATP was rigor enhanced. Since in ischaemic heart the molar excess of AMP over ATP appears to be 10-fold (Geisbuhler et al., 1984) or less (Neely and Grotyohann, 1984), it is unlikely that AMP is an important regulator of rigor. The extent to which adenylate kinase (AK) might
modulate the adenine nucleotide environment in cardiac myofibrils remains obscure. Evidence from skeletal muscle suggests that AK plays a role at both the mitochondrial and myofibrillar poles of the phosphocreatine shuttle (Bessman and Carpenter, 1985), with phosphoryl transfer flux through AK increasing during contractile activity (Zeleznikar et al., 1990). In skinned cardiac muscle exposed to creatine phosphate, rigor tension is unaffected by diadenosine pentaphosphate (Ventura-Clapier and Vassort, 1985; Ventura-Clapier and Veksler, 1994), but under these conditions a role for AK is likely to be masked by ADP phosphorylation by CK. When creatine phosphate is depleted, as in ischaemic myocardium, it remains conceivable that AK might postpone rigor by attenuating the fall in ATP:ADP ratio at the myofilaments. However, data in Figure 4 suggest that intrinsic adenylate kinase activity in permeabilized cardiomyocytes caused little or no deviation of myofibrillar ATP/ADP from that in the bulk phase, since inhibition of the enzyme did not affect rigor. On this basis, it appears unlikely that myofibrillar AK plays any significant role in modulation of ischaemic contracture, at least under near steady-state conditions.
Relevance to ischaemia The finding that in the presence of MgADP the myofibrillar array in cardiomyocytes begins to enter rigor at MgATP concentrations in the low millimolar range, is consistent with direct measurements of cytosolic ATP in individual cells undergoing deenergization contracture (Allue et al., 1996). Since rigor development stimulates myosin ATPase activity, a positive feed-back loop is established that could drive the abrupt depletion of residual cytosolic ATP observed in individual cardiomyocytes that coincides with de-energization contracture (Bowers et al., 1992; Allue et al., 1996). Following deenergization-induced contracture in intact cardiomyocytes diastolic calcium then rises, priming the cells for reoxygenation injury in the form of a powerful calcium- and ATP-activated hypercontracture (Stern et al., 1985; Allshire et al., 1987). Of the metabolic changes in ischaemic myocardium, increased ADP is likely to be the primary determinant of onset of rigor. Data in Figure 6 suggest that rigor becomes significant at millimolar ATP as ADP enters the high micromolar range. In hypoxic or ischaemic myocardium ADP levels rise to high micromolar (Kammermeier et al., 1982) or millimolar (Neely and Grotyohann, 1984;; Geisbuhler et al., 1984; Vander Heide et al., 1996b),
Rigor and Myosin ATPase in Cardiomyocytes
according to the method of measurement, while tissue ATP:ADP ratio is substantially decreased (Neely and Grotyohann, 1984; Bak and Ingwall, 1994). However, ADP stimulation of rigor is antagonized by inorganic phosphate and prevented under acidotic conditions. Since these factors may be less prominent in superfused cardiomyocytes subjected to metabolic inhibition with cyanide and 2-deoxyglucose, the estimate of a 1–2 m ATP threshold for onset of rigor (Allue et al., 1996) could actually represent an upper limit in the case of the ischaemic myocardium. Subsequent progression of rigor contracture will then be determined by the dynamic balance between development of acidosis and changes in concentration of ADP and inorganic phosphate. In conclusion, we show that at physiological temperature rigor development in cardiomyocytes entails an activation of myosin ATPase. This accords with the hypothesis that rigor dramatically accelerates ATP depletion in de-energized cells (Allshire and Cobbold, 1989), and demonstrates the likely mechanism underlying the abrupt loss of cytosolic ATP in individual cardiomyocytes undergoing contracture (Bowers et al., 1992). Consequently, rigor may represent a key target for interventions aimed at enhancing survival of the ischaemic myocardium.
Acknowledgements This study was supported by the Irish Heart Foundation and undertaken within the ambit of EU Biomed II Reinforced Concerted Action PL-951254.
References A AP, C PH, 1989. Cytosolic Ca2+ in hypoxic cardiomyocytes. In: Piper HM, Isenberg I (eds). Isolated Adult Cardiomyocytes. Boca Raton: CRC Press, 287–308. A AP, P HM, C KSR, C PH, 1987. Cytosolic free Ca2+ in single rat heart cells during anoxia and reoxygenation. Biochem J 244: 381–385. A I, G O, D E, U N, C P, 1996. Evidence for rapid consumption of millimolar concentrations of cytoplasmic ATP during rigor-contracture of metabolically compromised single cardiomyocytes. Biochem J 319: 463–469. A RA, H JR, B GP, 1981. Response of isolated rat heart cells to hypoxia, reoxygenation, and acidosis. Circ Res 49: 307–316. A SC, G CE, 1991. Effects of 2,3-butanedione monoxime (BDM) on contracture and injury
1357
of isolated rat myocytes following metabolic inhibition and ischemia. J Mol Cell Cardiol 23: 1001–1014. B MI, I JS, 1994. Acidosis during ischaemia promotes adenosine triphosphate resynthesis in postischaemic rat heart. J Clin Invest 93: 40–49. B SP, C CL, 1985. Creatine phosphate shuttle. Annu Rev Biochem 54: 831–862. B KC, A AP, C PH, 1992. Bioluminescent measurement in single cardiomyocytes of sudden cytosolic ATP depletion coincident with rigor. J Mol Cell Cardiol 24: 213–218. B RD, W AM, 1972. Cooperation within actin filament in vertebrate skeletal muscle. Nature 238: 97–101. B SPJ, S KB, 1992. Bound and determined: a computer program for making buffers of defined ion concentrations. Anal Biochem 201: 119–126. C KM, D D, J KD, 1986. A direct colorimetric assay for Ca2+-stimulated ATPase activity. Anal Biochem 157: 375–380. C CM, R MA, S EL, G NE, 1995. Magnesium sulfate reduces myocardial infarct size when administered prior to but not after coronary reperfusion in a canine model. Circulation 92: 2617–2621. C R, P E, 1985. The effect of ADP and phosphate on the contraction of muscle fibres. Biophys J 48: 789–798. E JP, S GJM, E G, 1994. Influence of phosphate and pH on myofibrillar ATPase activity and force in skinned cardiac trabeculae from rat. J Physiol 476: 501–516. E AC, S GL, E DA, A DG, 1992. Metabolic changes during ischaemia and their role in contractile failure in isolated ferret hearts. J Physiol 454: 467–490. F A, F F, 1975. Effects of magnesium on contractile activation of skinned cardiac cells. J Physiol 249: 497–517. F SAE, B A, S CA, C R, W HF, 1985. The change of the free energy of ATP hydrolysis during global ischaemia and anoxia in the rat heart. J Mol Cell Cardiol 16: 1023–1036. G T, A RA, T RW, A AZ, L K, B G, 1984. Adenine nucleotide metabolism and compartmentalization in isolated adult rat heart cells. Circ Res 54: 536–546. H RA, H DR, B HA, 1981. Contracture in isolated adult rat heart cells: role of Ca2+, ATP and compartmentation. Circ Res 49: 1119–1128. H JP, W RJ, 1991. Cytosolic free magnesium in stimulated, hypoxic and underperfused rat heart. J Mol Cell Cardiol 23: 991–999. H WR, S ML, MM KS, E LR, G MJ, V RA, S VL, 1995. Timing of magnesium therapy affects experimental infarct size. Circulation 92: 2622–2626. H C, A NR, G R, M LA, 1982. Heat production during hypoxic contracture of rat myocardium. Circ Res 51: 777–786. K H, S P, J¨ E, 1982. Free energy change of ATP hydrolysis: a causal factor of early hypoxic failure of the myocardium. J Mol Cell Cardiol 14: 267–277. K JC, 1986. The effects of inorganic phosphate
1358
M. T. Stapleton and A. P. Allshire
and creatine phosphate on force production in skinned muscles from rat ventricle. J Physiol 370: 585–604. K Y, M E, 1990. Mechanism of ischemic contraction in ferret hearts: relative roles of [Ca2+]i elevation and ATP depletion. Am J Physiol 258: H9– H16. L JA, A DG, 1991. Mechanisms of acute ischaemic contractile failure of the heart. J Clin Invest 88: 361– 367. M DJ, S GL, 1985. The contractile behaviour of EGTA and detergent-treated heart muscle. J Musc Res Cell Motil 6: 541–567. N JR, G LW, 1984. Role of glycolytic products in damage to ischemic myocardium: dissociation of adenosine triphosphate levels and recovery of function of reperfused ischemic hearts. Circ Res 55: 816–824. N CG, L WJ, 1990. The role of ATP in energydeprivation contractures in unloaded rat ventricular myocytes. Can J Physiol Pharmacol 68: 183–194. P WS, 1991. The effects of ATP, inorganic phosphate, protons, and lactate on isolated myofibrillar ATPase activity. Can J Physiol Pharmacol 70: 1175– 1181. S H, F T, I S, 1992. Regulation of tension development by MgADP and Pi without Ca2+. Biophys J 61: 1087–1098. S HS, D L D, H RC, M H, S SJ, H RG, L EG, S MD, 1994. Regulation of intracellular free Mg2+ and contraction in single adult mammalian cardiac myocytes. Am J Physiol 266: C222–C233.
S JA, S SJ, 1981. Actomyosin ATPase and muscle contraction. Curr Top Bioenerg 11: 239–286. S GL, S DS, 1994. Effects of pH and inorganic phosphate on rigor tension in chemically skinned rat ventricular trabeculae. J Physiol 478: 505–512. S M, D P, C P, A A, 1995. Troponin-I137–148 peptide abolishes rigor in cardiomyocytes. Circulation 92: I–189 (Abstract). S MD, C AM, C MC, P DJ, L EG, 1985. Direct observation of the ‘oxygen paradox’ in single rat ventricular myocytes. Circ Res 56: 899–903. V H RS, D JA, J RB, R KA, S C, 1996a. Reducing lactate accumulation does not attenuate lethal ischemic injury in isolated perfused rat hearts. Am J Physiol 270: H38–H44. V H RS, H ML, R KA, J RB, 1996b. Effect of reversible ischemia on the activity of the mitochondrial ATPase: relationship to ischemic preconditioning. J Mol Cell Cardiol 28: 103–112. V-C R, V G, 1985. Role of myofibrillar creatine kinase in the relaxation of rigor tension in skinned cardiac muscle. Pflu¨gers Arch 404: 157–161. V-C R, V V, 1994. Myocardial ischemic contracture:metabolites affect rigor tension development and stiffness. Circ Res 47: 920–929. Z RJ, H RA, G RM, W TF, D SM, B EA, G ND, 1990. Evidence for compartmentalized adenylate kinase catalysis serving a high energy phosphoryl transfer function in rat skeletal muscle. J Biol Chem 265: 300–311.