Ca+ loading during reperfusion after myocardial ischemia is linked to reduced cardiac function. Like ischemic preconditioning (IPC), a volatile anesthetic given briefly before ischemia can reduce reperfusion injury. We determined whether IPC and sevoflurane preconditioning (SPC) before ischemia equivalently improve mechanical and metabolic function, reduce cytosolic Ca2+ loading, and improve myocardial Ca2+ responsiveness. Four groups of guinea pig isolated hearts were perfused: no ischemia, no treatment before 30-min global ischemia and 60-min reperfusion (control), IPC (two 2-min occlusions) before ischemia, and SPC (3.5 vol%, two 2-min exposures) before ischemia. Intracellular Ca2+ concentration ([Ca2+]i) was measured at the left ventricular (LV) free wall with the fluorescent probe indo 1. Ca2+ responsiveness was assessed by changing extracellular [Ca2+]. In control hearts, initial reperfusion increased diastolic [Ca2+] and diastolic LV pressure (LVP), and the maximal and minimal derivatives of LVP (dLVP/dt max and dLVP/dt min, respectively), O2consumption, and cardiac efficiency (CE). Throughout reperfusion, IPC and SPC similarly reduced ischemic contracture, ventricular fibrillation, and enzyme release, attenuated rises in systolic and diastolic [Ca2+], improved contractile and relaxation indexes, O2 consumption, and CE, and reduced infarct size. Diastolic [Ca2+] at 50% dLVP/dt min was right shifted by 32–53 ± 8 nM after 30-min reperfusion for all groups. Phasic [Ca2+] at 50% dLVP/dt max was not altered in control but was left shifted by −235 ± 40 nM [Ca2+] after IPC and by −135 ± 20 nM [Ca2+] after SPC. Both SPC and IPC similarly reduce Ca2+ loading, while augmenting contractile responsiveness to Ca2+, improving postischemia cardiac function and attenuating permanent damage.
- contractile function
cardiac mechanical dysfunction and cell death typically occur with reperfusion (RP) after prolonged ischemia. This has been shown during simulated ischemia in isolated myocyte models, in intact isolated hearts, and in whole animal models of several species. The duration and/or the magnitude of ischemia determine whether the cellular injury is reversible (stunned) or irreversible (infarcted) (17). The mechanism of RP is multifactorial but the following factors stand out as being causative: 1) cytosolic Na+ overload (38, 42, 56) and Ca2+ overload (1,56) resulting in myofibrillar hypercontracture, cytoskeletal damage, and cell disruption during RP, 2) mitochondrial Ca2+ overload (4, 37) that may lead to inefficient ATP synthesis or utilization as a result of reduced nicotinamide adenine dinucleotide (NADH) accumulation (56), 3) reduced maximal Ca2+activated force and/or sensitivity (56) of myofilaments to Ca2+, and 4) impaired myocardial vascular perfusion (41). Varadarajan et al. (56) reported that 30-min global ischemia causes dissociation between cardiac function and intracellular Ca2+concentration ([Ca2+]i) during RP.
Hearts subjected to brief periods of ischemia can withstand subsequent periods of prolonged ischemia with better functional recovery. Ischemic preconditioning (IPC), first described in 1986 (39), is most often assessed by observations of reduced infarct size, attenuated mechanical dysfunction, or limited ultrastructural abnormality on RP after prolonged ischemia. This endogenous cardioprotective mechanism has been shown to occur in many species including humans (13, 15, 30, 41, 44, 46). The degree of mechanical and metabolic cardioprotection provided by myocardial preconditioning and the mechanisms that mediate its effects have been the focus of intense investigation.
Anesthetics are also preconditioning agents. Kersten et al. (26) reported that isoflurane mimics IPC, as assessed by reduced infarct size in dogs, and suggested that the protective effect involves ATP-sensitive K+ (KATP) channel opening because KATP channel antagonism by glibenclamide attenuated the reduction in infarct size. We reported (40) that anesthetic preconditioning with sevoflurane (SPC) mimics IPC by improving vascular, mechanical, and metabolic function and induced endothelial nitric oxide release in isolated guinea pig hearts; these effects also were blocked by glibenclamide.
We hypothesized that altered cellular ion homeostasis leads to Ca2+ overload as a primary cause of RP myocardial injury and that two different preconditioning approaches to reduce RP injury can similarly attenuate myocardial Ca2+ overload and damage and improve function during RP.
MATERIALS AND METHODS
Langendorff Heart Preparation
The investigation conformed to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, Publication No. 85-23, Revised 1996). Prior approval was obtained from the Medical College of Wisconsin animal studies committee. Our methods have been described in detail previously (40, 41, 50, 52,56). Ketamine (30 mg) and heparin (1,000 units) were injected intraperitoneally into 60 albino English short-haired guinea pigs (250–300 g) 15 min before the animals were decapitated when unresponsive to noxious stimulation. After thoracotomy, the inferior and superior venae cavae were cut and the aorta was cannulated distal to the aortic valve. Each heart was immediately perfused via the aortic root with a cold oxygenated modified Krebs-Ringer (KR) solution (equilibrated with 97% O2-3% CO2) and rapidly excised. All hearts were perfused at an aortic root perfusion pressure of 55 mmHg. The KR perfusate (pH 7.39 ± 0.01, Po 2 560 ± 10 mmHg) was in-line filtered (5-μm pore size) and had the calculated composition of (nonionized, in mM) 137 Na+, 5 K+, 1.2 Mg2+, 2.5 Ca2+, 134 Cl−, 15.5 HCO , 1.2 H2PO , 11.5 glucose, 2 pyruvate, 16 mannitol, 0.05 EDTA, and 0.1 probenecid, with 5 units of insulin. Perfusate and bath temperatures were maintained at 37.2 ± 0.1°C with the use of a thermostatically controlled water circulator.
LV pressure (LVP) was measured isovolumetrically with a transducer connected to a thin, saline-filled latex balloon inserted into the LV through the mitral valve from a cut in the left atrium. Balloon volume was initially adjusted to a diastolic LVP of 0 mmHg so that any subsequent increase in diastolic LVP reflected an increase in LV wall stiffness or diastolic contracture. A pair of bipolar electrodes was placed in the right atrial appendage and in the right ventricular free wall to monitor spontaneous heart rate and atrioventricular conduction time. Coronary flow (CF; aortic inflow) was measured at constant temperature and constant perfusion pressure (55 mmHg) by an ultrasonic flowmeter (model T106X, Transonic Systems; Ithaca, NY) placed directly into the aortic inflow line.
Coronary inflow and coronary venous (v) Na+, K+, Ca2+, Po 2, pH, and Pco 2 were measured off-line with an intermittently self-calibrating analyzer system (model ABL-505, Radiometer; Copenhagen, Denmark). Coronary sinus effluent was collected by the placement of a small catheter into the right ventricle through the pulmonary artery after both venae cavae were ligated. Coronary sinus venous Po 2 tension was also measured continuously on-line with an O2 Clark type electrode (model 203B, Instech; Plymouth Meeting, PA). Percent O2 extraction (%O2E) was calculated as 100 × (PaO2 − PvO2)/PaO2, myocardial O2consumption (MV˙o 2) as CF/g × (PaO2 − PvO2) × 24 μl O2/ml at 760 mmHg; and cardiac efficiency (CE) as systolic-diastolic LVP × heart rate/MV˙o 2. Effluent was spot collected during RP and frozen for later analysis of creatine kinase (sensitivity >10 U/l; Creatine Kinase Flex Reagent Cartridge, Dade Behring Dimension; Newark, DE). If ventricular fibrillation (VF) occurred, a 0.25-ml bolus of lidocaine (250 μg) was administered immediately via the aortic cannula. The data were collected from hearts naturally in, or converted to, sinus rhythm.
Infarct size was determined by the 2, 3,5-triphenyltetrazolium chloride (TTC) staining method. At the end of 70 min, RP hearts were cut into four horizontal sections and TTC stained. TTC stains the noninfarcted myocardium a brick red color, which indicates the presence of a formazan precipitate that results from the reduction of TTC by dehydrogenase enzymes present in viable tissue. After storage overnight in 10% formaldehyde, infarcted and noninfarcted tissues of whole hearts were carefully separated and weighed. Infarct size was expressed as a percentage of the whole heart.
Measurement of Cytosolic and Noncytosolic Free Ca2+ in Intact Hearts
Loading fluorescent probe indo 1 and recording Ca2+ transients.
Experiments were carried out in a light-blocking Faraday cage. The heart was partially immobilized by hanging it from the aortic cannula, the pulmonary artery catheter, and the LV balloon catheter. The heart was immersed continuously in the bath at 37°C. The distal end of a trifurcated fiber silica fiber-optic cable (optical surface area 3.85 mm2) was placed gently against the LV epicardial surface through a hole in the bath. A rubber O-ring was placed over the fiber-optic tip to seal the hole, and netting was applied around the heart for optimal contact with the fiber-optic tip. This maneuver did not affect LVP. Background autofluorescence was determined for each heart after initial perfusion and equilibration at 37°C.
We (50, 52, 56) have published loading and washout methods, as well as calibration, and recording techniques for indo 1. Indo 1-acetoxymethyl ester (AM) (Sigma; St. Louis, MO) was freshly dissolved in 1 ml of dimethyl sulfoxide containing 16% (wt/vol) Pluronic F-127 (Sigma) and diluted to 165 ml with modified KR solution. Each heart was then loaded with indo 1-AM for 20–30 min with the recirculated KR solution at a final indo 1-AM concentration of 6 μM (0.4 mM dimethyl sulfoxide). Loading was stopped when the fluorescence (F) intensity at 385 nm increased by ∼10-fold. Residual interstitial indo 1-AM was washed out by perfusing the heart with standard perfusate for at least another 20 min. Probenecid (100 μM) was present in the perfusate to retard cell leakage of indo 1. We (52) reported that loading and washout of indo 1 reduces LVP ∼25%; this effect is due to the vehicle and to intracellular Ca2+buffering by indo 1 per se.
Fluorescence emissions at 385 and 456 nm (F385 and F456) were recorded using a modified luminescence spectrophotometer (SLM Aminco-Bowman II, Spectronic Instruments; Urbana, IL) as described previously (50, 52, 56). The LV region of the heart was excited with light from a xenon arc lamp, and the light was filtered through a 360-nm monochromator with a bandwidth of 16 nm. The beam was focused onto the in-going fibers of the optic bundle. The arc lamp shutter was opened only for 2.5-s recording intervals to prevent photobleaching. Emission fluorescence was collected by fibers of the remaining two limbs of the cable and was filtered by square interference filters (Corion; Franklin, MA) at 385 nm (390 ± 5 nm) and 456 nm (460 ± 5 nm).
Although both F385 and F456 declined over time, the F385/F456 ratio remained stable during the course of these studies. The time-related effect of changes in indo 1 fluorescence units and contractility was examined at 2.1 mM extracellular ionized [Ca2+] over 3 h in nonischemic hearts (time control, n = 9). Systolic and diastolic F385, respectively, decreased from 7.2 ± 0.4 and 4.8 ± 0.4 initially to 3.2 ± 0.3 and 1.9 ± 0.2 units after 3 h; the units for F456 nm decreased from 4.8 ± 0.4 and 3.9 ± 0.4 to 1.6 ± 0.2 and 1.4 ± 0.2 units after 3 h. However, the F385/F456 ratio, initially 1.6 ± 0.3 and 1.1 ± 0.3, was unchanged at 1.9 ± 0.1 and 1.3 ± 0.3, indicating that no change in effective [Ca2+] occurred over that time. Developed LVP after indo 1 loading, initially 68 ± 4 mmHg and then 76 ± 4 mmHg after 3 h, also was not significantly altered over time. In companion studies, tissue autofluorescence was measured throughout each experiment in the absence of ischemia (autofluorescence control, n = 8) and before, during, and after ischemia (ischemia control, n = 8). The mean F385 and F456 background values obtained were then subtracted from the corresponding F385 and F456 values from the indo-1-treated hearts at the same time point and for the same experimental condition.
Calculation of cytosolic and noncytosolic [Ca2+] from F385/F456 transients.
The Ca2+ transient obtained from the fluorescence ratio F385/F456 is nonlinearly proportional to [Ca2+]. The calibration curves were derived according to the protocols by Brandes et al. (6, 7) with the use of modifications of a standard equation for fluorescent indicators (22). Total (tot) [Ca2+]i([Ca2+]tot) was calculated from the F385,tot/F456,tot ratio (Rtot), Rmax = Sr/bH (for >100 μM Ca2+), Rmin = Rmax · S385/S456 (for 0 Ca2+), S385 = I385/I385 (at min/max Ca2+) = 0.05, S456 = I456/I456 (at min/max Ca2+) = 2.40, and the dissociation constant (K d), according to the equation Equation 1where S is the ratio of light intensities (I) at the same wavelength at min/max Ca2+, Sr = (1 − S456)/(1 − S385) = −1.48, andbH is the average slope (b) of F385,tot as a function of F456,tot = −0.25.
We calculated the K d of indo 1 by using homogenized guinea pig heart protein as 249 ± 8 nM (4). Rmax was calculated as 5.986 and Rmin as 0.059.
Noncytosolic (primarily mitochondrial) [Ca2+] ([Ca2+]mito) was calculated similarly as Equation 2where Rmito was calculated as the ratio of the noncytosolic fluorescence, F385,mito and F456,mito, respectively, Noncytosolic fluorescence was measured at the end of each experiment (185 min) in control (n = 12), IPC (n = 12), and SPC (n = 12) groups after perfusing hearts with 100 μM MnCl2 for 10 min to quench fluorescence derived from the cytosolic (cyto) compartment (8). F385,mitoand F456,mito were calculated at each time point by multiplying the residual mitochondrial fluorescence fractions (f385 and f456) by total end-diastolic fluorescence so that Equation 3 Similar to Eqs. 1 and 2 , cytosolic [Ca2+]cyto was calculated as Equation 4where Rcyto was derived from the ratio of the cytosolic fluorescence, F385,cyto and F456,cyto, respectively, calculated at each time point by effectively subtracting [Ca2+]mito from [Ca2+]tot and multiplying the remainder by total end-diastolic fluorescence (as in Eq. 3 ) so that Equation 5 Estimated mitochondrial compartment Ca2+ (not shown) is similar to actual mitochondrial Ca2+ measured by MnCl2 quenching of indo 1 at the beginning of the experiment (56). Nonstimulated endothelium does not contribute significantly to [Ca2+]tot(7, 52).
Data Presentation and Interpretation
Cytosolic-systolic and diastolic-Ca2+.
F385, F456, and F385/F456 Ca2+ transient signals, LVP, and the derivative of developed LVP over time (dLVP/dt) were displayed simultaneously on a computer screen and stored digitally using proprietary software on an IBM OS/2 system. After correcting for tissue autofluorescence over time, with or without ischemia and RP, and quenching cytosolic compartment Ca2+ to reveal noncytosolic compartment Ca2+, we calibrated the signals to nanomolar [Ca2+1) before and after ischemia in one representative experiment from each group. For each group, contractile indexes as a function of [Ca2+] were plotted (Figs. 7 and 8) during increases in CaCl2.2+ transient data, were later analyzed together (Excel, Microsoft; Redmond, WA). Several characteristics of cyto[Ca2+] were analyzed: peak systolic (sys), peak diastolic (dia), and systolic-diastolic (sys-dia) [Ca2+], i.e., released or phasic [Ca2+]; mitochondrial (noncytosolic) [Ca2+] is not phasic. The following characteristics of LVP were analyzed: sys, dia, and sys-diaLVP, and the maximal and minimal derivatives of LVP, respectively (dLVP/dt max and dLVP/dt min). Cyclic LVP [Ca2+] coordinates (250 data points over 2.5 s) were plotted (Fig.
Potential limitation of Ca2+analysis.
It is not possible to separate cytosolic compartment Ca2+from other intracellular Ca2+ compartments, e.g., mitochondrial or nuclear, or Ca2+ from other cells, e.g., endothelial, vascular smooth muscle, or nerve when using Ca2+ fluorescent dyes in intact hearts. However, the flux of Ca2+ through mitochondrial and nuclear compartments is likely very slow so it is improbable that these compartments contribute much to the total rapid phasic signal derived from the myocyte cytosol. We estimated average noncytosolic Ca2+ by quenching cytosolic Ca2+ with MnCl2 at the end of each experiment. We assume that fluorescence determined after cytosolic Ca2+ quenching arises predominantly from mitochondrial Ca2+, but this could be an overestimate.
It is possible that an immediate increase in Ca2+ on RP could arise in part from washed-out Ca2+ bound indicator dye leaking out of damaged cells during ischemia. To test this, we carried out preliminary experiments with very low-flow normoxic ischemia or normal flow hypoxia (Po 2, 35 mmHg) for 30 min to provide flow during ischemia and so to continuously wash out any interstitial indicator. In an additional protocol, we subjected hearts to 30-min global ischemia and then perfused them with a hypoxic solution (10 ml/min) for 2 min, followed by an additional 2 min of global ischemia before final RP with normoxic KR solution. These experiments confirmed that Ca2+ increases rapidly over twofold on initial RP after 30-min ischemia, and this effect is not due to indicator washout.
There were four primary indo 1-treated groups; each experiment lasted 185 min after a 30-min period of equilibration. The untreated time control (nonischemic) group was not subjected to ischemia or transient preconditioning. The three 30-min index ischemia groups were control (Con), IPC, and SPC. Each group underwent 85 min of perfusion, followed by 30-min ischemia and 70-min RP. The same time protocols were used for both IPC and SPC. The IPC group was subjected to two 2-min periods of transient no-flow global ischemia separated by 5 and 6 min of RP and followed by 30-min global ischemia and 70-min RP. The SPC group was subject to two 2-min periods of perfusion with sevoflurane separated by 5 and 6 min of RP.
Sevoflurane (3.5 vol%) was bubbled into the perfusate using an agent-specific vaporizer placed in the O2-CO2gas mixture line. Samples of coronary perfusate were collected from a port in the aortic cannula to measure sevoflurane concentration by gas chromatography (9, 10, 34). Inflow sevoflurane concentration was 0.64 ± 0.02 mmol/l, which is equivalent to 3.34 ± 0.22% atmospheres and a minimal alveolar concentration of ∼1.5 ± 0.4%. Sevoflurane was not detectable in the effluent during the initial equilibration period, the ischemic period, or the RP period.
Initial control measurements were obtained at the end of fluorescence dye or vehicle washout. Recordings were obtained every 5 min at the nominal 2.1 mM ionized extracellular Ca2+ concentration ([Ca2+]o). To assess differences in association between cyto[Ca2+] and contractility before and after ischemia, [Ca2+]o was increased incrementally from 0.4 to 4.9 mM over 18 min by infusing a concentrated CaCl2 solution into the perfusate initially lacking CaCl2. During these increases in [Ca2+]o, contractility and F385/F456 measurements were made at 1-min intervals until LVP reached a stable maximum. Each heart underwent a change in [Ca2+]o twice, once 30 min before ischemia and once again 30 min after ischemia (RP), so that each heart served as its own control. In the nonischemic time control (n = 9) studies, [Ca2+]o was again changed 1 h after the first change. Preliminary results of Vadarajan et al. (56,57) show that increasing [Ca2+]o over 20 min does not by itself increase mitochondrial compartment Ca2+, nor does IPC or SPC reduce the rise in mitochondrial compartment Ca2+ after RP (2, 58).
All data were expressed as means ± SE. Within-group data for a given variable were compared with a preischemia control period (at 70 min) by Duncan's comparison of means tests whenever univariate analysis of variance (ANOVA) for repeated measures was significant (SuperANOVA version 1.11 software for Macintosh, Abacus Concepts; Berkeley, CA). Within-group data were analyzed over time using univariate ANOVA for repeated measures. If F values (P < 0.05) were significant, post hoc comparisons of means compared with the preischemia control period (at 70 min) (Student's t-test with Duncan's adjustment for multiplicity) were used to differentiate treatment groups. The incidence of VF versus sinus rhythm was determined by χ2analysis and differences in VF duration were determined by unpairedt-tests. Contractile and relaxation indexes as a function of myoplasmic [Ca2+] over the range of [Ca2+]o were determined by nonlinear regression with slope comparisons by parallelism tests using the Boltzmann equation (Prism Software; GraphPad). Differences among means were considered statistically significant when P ≤ 0.05 (*time value vs. 70-min Con, #IPC or SPC vs. Con, and §IPC vs. SPC). Some statistical notations that are not given in the figures are given in the text.
Variables unchanged at 70-min RP from 20 min before ischemia, respectively, and averaged for all groups, were heart rate, 250 ± 3 to 252 ± 3 beats/min, and atrioventricular conduction time, 75 ± 2 to 73 ± 2 ms. Venous pH (pHv) before ischemia, for Con, IPC, and SPC groups, respectively, 7.16 ± 0.03, 7.11 ± 0.04, and 7.11 ± 0.05, and at 70-min RP, 7.09 ± 0.05, 7.10 ± 0.07, and 7.08 ± 0.05, were not different. CF before ischemia, 8.4 ± 0.4, 8.6 ± 0.3, and 8.5 ± 0.4 ml · min−1 · g−1, was not different in Con, IPC, and SPC groups, respectively; CF at 70-min RP was lower in each group (P < 0.05) but higher in the IPC group, 7.4 ± 0.5, and SPC group, 7.6 ± 0.4 ml · min−1 · g−1, than in the Con group, 5.0 ± 0.3 ml · min−1 · g−1. %O2E before ischemia, 75 ± 3, 77 ± 3, and 77 ± 3%, was not different in Con, IPC, and SPC groups, respectively; %O2E at 70-min RP was lower in the Con group, 63 ± 3% (P < 0.05), than in IPC, 79 ± 35%, and SPC, 75 ± 3%, groups.
Figure 1 displays representative cardiac loops (coordinates) of cyclic LVP as a function of cyclic cyto[Ca2+] before ischemia and at 2 and 60 min RP (Fig. 1, A and B, respectively). At 2-min RP, each loop was shifted upward and rightward, but more so in Con than IPC and SPC groups; at 60-min RP, these differences persisted but were much smaller. Figures 2 A to 6 B display the associated temporal changes in cyto[Ca2+] and cardiac function at 30 time points beginning 20 min before 30 min of ischemia and lasting up to 60-min RP. Figure2 A shows that sys[Ca2+] decreased slowly over 10 min during ischemia and then increased abruptly during initial RP in each group. However, the rise in sys[Ca2+] throughout RP was less after IPC and SPC and returned to control levels after 5-min RP; preischemia control levels were attained only after 30 min in the Con group. Figure 2 B shows that sysLVP decreased slightly during SPC and toward 0 mmHg during the brief IPC pulses and during prolonged ischemia. On initial RP, sysLVP returned to the control level after IPC but not after SPC; however, sysLVP remained higher after IPC and SPC than after Con during 30-min RP.
Figure 3 A shows that dia[Ca2+] rose incrementally during ischemia and rose abruptly and markedly during initial RP in each group; dia[Ca2+] remained similarly elevated throughout RP in each group. Figure 3 B shows that diaLVP fell briefly during IPC and during 20 min of ischemia in each group but then increased during the last 10 min; however, the time of onset of LV diastolic contracture was earlier and its magnitude greater in Con than in IPC and SPC groups, as also shown in Fig. 2 B. DiaLVP on RP increased much less after IPC and SPC compared with Con but remained above control levels in each group throughout RP. Figure4 A shows that phasic sys-dia[Ca2+] decreased over the initial 10 min of ischemia and then approached a minimum in each group. Sys-dia[Ca2+] rose abruptly during initial RP in each group but more so in Con than in IPC and SPC groups. Figure4 B shows that sys-diaLVP decreased slightly during SPC and to 0 mmHg during IPC and ischemia. Sys-diaLVP remained depressed compared with before ischemia in each group throughout RP, but was improved more after IPC compared with after SPC during the first 5 min of RP, and more after IPC and SPC than in the Con group throughout RP.
Figure 5, A and B, shows that dLVP/dt max and dLVP/dt min decreased moderately during SPC and approached zero during IPC and ischemia; on RP these values remained depressed throughout RP but were improved more so in the IPC and SPC groups. Myocardial O2 consumption, MV˙o 2, (Fig.6 A) and CE (Fig.6 B) were indeterminate during zero flow IPC and ischemia. MV˙o 2 was equivalent to that before ischemia after IPC and SPC throughout RP but remained much reduced in the Con group. MV˙o 2 was 26% higher after RP in both IPC and SPC groups than in the Con group. CE returned to preischemia control levels after 5-min RP in IPC and SPC groups but remained depressed by ∼50% in the Con group. For data not displayed, CEmax obtained over the range of [Ca2+]o was not depressed in IPC and SPC groups after RP but was 24% lower in the Con group. The curve of CE as a function of sys-dia[Ca2+] was right-shifted by 160 ± 25 nM in the Con group, but was not shifted in IPC or SPC groups.
Figures 7 A to 8 B display indexes of contractility and relaxation as a function of cyto[Ca2+]. Figure7 A displays dLVP/dt max (contractility) as a function of sys[Ca2+] during the induced change in ionized [Ca2+]o from 0.40 ± 0.03 to 4.90 ± 0.24 mM conducted 30 min before ischemia and 30 min after RP. Postischemia curves were corrected for a small left shift (∼30 nM) in a second curve obtained 90 min after the first curve in the nonischemia Con group, as we reported earlier (56). Contractility increased as a function of sys[Ca2+] in each group; however, after ischemia, peak dLVP/dt max was decreased by 53 ± 3%, 34 ± 4%, and 37 ± 3% in Con, IPC, and SPC groups, respectively. Figure 7 B displays the same data after all peak values for dLVP/dt max were normalized to 100%. There was no significant difference between the pre- and postischemia Con curves, but in the IPC and SPC groups, the curves at 50% dLVP/dt (ED50) were left shifted by −190 ± 30 nM [Ca2+] and −130 ± 24 nM [Ca2+], respectively. For data not displayed, dLVP/dt max was also plotted as a function of sys-dia[Ca2+]. From before to after ischemia the curves at ED50 were left shifted by −235 ± 40 nM [Ca2+] and −135 ± 20 nM [Ca2+], respectively, for IPC and SPC groups; ED50 curves were not different for the Con group.
Figure 8 A displays dLVP/dt min as a function of dia[Ca2+]. The rate of relaxation increased maximally as dia[Ca2+] increased by ∼50 nM in each group before and after ischemia and RP; however, after ischemia, peak dLVP/dt min was decreased by 49 ± 2, 23 ± 3, and 24 ± 3% in Con, IPC, and SPC groups, respectively. Figure 8 B displays the same data after all peak values for dLVP/dt min were normalized to 100%. There were no significant differences in the shape of pre- and postischemia curves, but the curves at ED50 were similarly right shifted by 36 ± 10 nM [Ca2+] and 53 ± 8 and 32 ± 6 nM [Ca2+], for Con, IPC, and SPC groups, respectively.
Figure 9 A displays the average number of ventricular fibrillations per heart; 2.8 for Con, 1.0 for IPC, 0.8 for SPC groups; this was lower (P < 0.05) in IPC and SPC groups. VF was the only notable dysrhythmia observed on RP. The percentage of hearts that fibrillated during RP (not shown) was Con 100%, IPC 75%, and SPC 64%. Figure 9 B displays infarct size in Con, IPC, and SPC groups. Infarct size was less in IPC, 13.8 ± 1.6%, and SPC, 14.5 ± 1.1%, groups (P < 0.05) than in the Con group, 51.3 ± 1.1%. Infarcted tissue was confined to the middle one-third of the myocardium in all ischemic groups. Creatine kinase (not displayed) for Con, IPC, and SPC groups, respectively, was 190 ± 61, 15 ± 9, and 12 ± 9 U/l at 1-min RP; 93 ± 62, 26 ± 12, and 0 U/l at 5-min RP; 9 ± 7, 1 ± 1, and 0 U/l at 10-min RP; 19 ± 16, 0, and 0 U/l at 15-min RP; and 6 ± 3, 0, and 0 at 30-min RP. Creatine kinase was not measurable at 30- and 70-min RP in any group.
This study first demonstrates in intact hearts that two brief exposures to ischemia or sevoflurane before sustained ischemia similarly lead to reduced cytosolic Ca2+loading on initial RP; this was associated with improved perfusion, contractility, relaxation, and metabolic function, and reduced creatine kinase release and infarct size. On RP, both IPC and SPC improved the loop relationship of LVP as a function of cyto[Ca2+] during each cardiac cycle. More importantly, these improvements after IPC and SPC were accompanied by augmented contractile responsiveness to [Ca2+] during later RP compared with controls. Reduced diastolic contracture during late ischemia after IPC and SPC could contribute to improved CF during RP via reduced microvascular compression. Moreover, CE returned to preischemia values on RP only after IPC and SPC. Because only viable cells consume O2 and produce work, restored CE may indicate improved functioning of remaining viable cells over time during RP. IPC and SPC also reduced the occurrence of VF on RP. This work confirms and extends our previous results (40, 41) on the vascular protective effects of preconditioning.
Postischemic RP injury likely results in large part from cytosolic Ca2+ overloading. If the rise in [Ca2+] concentration is prolonged, a cascade of events is initiated, which ultimately results in lethal injury (39). We (56) recently detailed the time course of change in contractility and relaxation with [Na+], cytosolic, and mitochondrial [Ca2+] and [NADH] during ischemia RP injury in the intact heart. From that study, it was apparent that contractile performance becomes dissociated from cytosolic Ca2+, particularly during early RP, in that the higher Ca2+ levels were associated with reduced contractile force and impaired relaxation. The study supported the notion that altered Na+/H and Na+/Ca+ exchange are important factors in ischemia RP injury because cytosolic Ca2+ loading was temporally associated with Na+loading during early RP. The study suggested also that continued elevations in [Na+] and mitochondrial [Ca2+] during RP largely contribute to continued contractile dysfunction (56).
The present study shows that both IPC and SPC improved maximal contractility and relaxation at high [Ca2+]o(Figs. 7 and 8). IPC may reduce systolic Ca2+ loading and improve function in part by improving sarcoplasmic reticular (SR) Ca2+ pump and release activities (25, 58). A surprising finding in the present study was that both IPC and SPC shifted the contractile response curves (sysLVP and dLVP/dt max/ sys[Ca2+] or /sys-dia[Ca2+]) to the left (lower ED50), during the increase in [Ca2+]o. There was no shift in the absence of preconditioning. On the other hand, ischemia, with or without IPC or SPC, shifted each relaxation response curve (dLVP/dt min/dia[Ca2+]) to the right (higher ED50). Thus there was a greater contractile and relaxant effect after IPC and SPC as a function of phasic [Ca2+] but this was achieved at a higher dia[Ca2+].
It is not clear how IPC and SPC reduce the Ca2+ loading that leads to improved function and reduced infarct size. Severe or prolonged Ca2+ overload leads to cell necrosis (29,37, 39). It is thought that index ischemia reduces Ca2+-ATPase activity so that SR uptake is reduced, and in a feedback manner, the rise in [Ca2+] increases the opening probability for SR Ca2+ release (59). IPC has been shown to reduce the rate constant of SR Ca2+ release (59) but it is unknown if this occurs with anesthetic preconditioning. IPC promotes translocation and activation of specific protein kinase C (PKC) isozymes (33, 48, 49), tyrosine kinases (3, 35), and Ca2+/calmodulin-dependent PK II (25). Because Ca2+/calmodulin-dependent PK II and protein kinase A phosphorylate the Ca2+ pump inhibitory protein phospholamban (36, 53), this may increase SR Ca2+ uptake and so may reduce Ca2+-induced SR Ca2+ release.
The effect of cardiac stunning on myofilament responsiveness to Ca2+ is controversial (11, 16, 19, 29). The rate and magnitude of tension development in cardiac muscle are regulated by the rate and magnitude of myoplasmic Ca2+availability (upstream mechanism), by the binding affinity of myoplasmic Ca2+ to troponin C (central mechanism), and by processes occurring after Ca2+ binding to troponin C. The latter implies altered myofilament contractile response to a given number of occupied Ca2+ binding sites on troponin C (downstream mechanism) (5). Changes in cross-bridge cycling rate and interaction of the troponin-tropomyosin complex with actin are two examples (5). Our results in the intact heart are based on these classic descriptions and findings obtained from force/Ca2+ curves in skinned or isolated muscle preparations. Our studies support the notion that sustained ischemia, in the absence of preconditioning (Con), decreases the number of functioning contractile elements (11, 16, 18,29) but does not alter central or downstream mechanisms on later RP (16). Both IPC and SPC effectively reduced the sys[Ca2+] (Fig. 7 B) required for a given level of contractility. Also contractile elements in remaining viable cells on RP appear to function more efficiently at a given level of cytosolic Ca2+ after either IPC or SPC. This novel result suggests an enhanced central and/or downstream effect of Ca2+ on contractility on RP after preconditioning.
The mechanism underlying improved contractile responses at given sys[Ca2+] or phasic sys-dia[Ca2+] after IPC and SPC is not clear. The leftward shift in the contractility/Ca2+ curve occurs despite the higher diastolic Ca2+ loading conditions on RP. This may reflect a myofilament adaptation to reduced diastolic contracture and improved relaxation rate after IPC and SPC for several reasons. First, because troponin I inhibits the Ca2+ troponin C interaction at low dia[Ca2+], the relative increase in dia[Ca2+] on RP may result in less troponin I inhibition, thus allowing an increased or prolonged Ca2+ troponin C interaction (5). However, each group exhibited a similarly elevated dia[Ca2+]. Second, IPC may reduce accumulation of cell [H+] during ischemia (18,32) and lower cell [H+] is associated with enhanced Ca2+ troponin C interaction (5). However, pHv was similar among the groups on RP and cell [H+] is rapidly normalized after RP (18,32). Third, IPC, and perhaps SPC, promotes translocation and activation of specific PKC isozymes (33) and downstream KATP channel opening (48). Fourth, certain PKC isozymes have been shown to also phosphorylate myofilament regulatory proteins (24). This could alter Ca2+sensitivity by modifying the association of Ca2+ with troponin C and/or the cross-bridge cycling rate (47).
A fifth possibility for augmented Ca2+ responsiveness after IPC is that brief IPC may induce a brief and small release of reactive O2 species (ROS) during IPC and cause reduced ROS formation during initial RP and so improve contractile responsiveness to Ca2+. Indeed, IPC was shown to reduce the oxidant burst at RP, and this effect was blocked by a PKC inhibitor and a KATP channel inhibitor (55). However, it is unknown if ROS formation is stimulated during SPC. ROS can oxidize sulfhydryl groups in ryanodine receptors and reduce SR ryanodine Ca2+ binding and Ca2+ release (60). Rat trabeculae treated with antioxidants before 20 min of ischemia and RP exhibited a greater contractile twitch response with a smaller Ca2+ transient compared with the control group; from this it was suggested that antioxidants have a myofilament Ca2+ sensitizing effect (43).
It is thought that ischemic preconditioning is mediated primarily via sarcolemmal G protein-linked receptors such as adenosine (A1, A3), bradykinin, and opioid receptors (21) coupled to PKC. Activation of PKC by IPC has been shown to activate kinase pathways in rabbits (33) that in turn promote phosphorylation of sarcolemmal and mitochondrial KATP channels (3, 20, 21, 31, 48, 49). Biochemical analysis indicates that IPC helps to conserve ATP during the subsequent ischemic period (39). Low ATP levels during ischemia opens KATP channels to allow K+ efflux at all depolarized voltages beyond the K+ equilibrium potential. This promotes earlier Ca2+ current repolarization and inactivation to cause a modest negative inotropic effect. KATP channel agonists enhance functional recovery of postischemic reperfused myocardium in vivo, and this effect can be blocked by KATPchannel antagonists such as glibenclamide (20). We (40) reported that vascular and myocardial protective effects of IPC were eliminated by glibenclamide. These studies indicate that endogenous KATP channel activation is involved, at least in part, in the mechanical and vascular protective effects of IPC. Yet it is not clear how KATP channel opening mediates IPC or SPC.
It is not yet known how volatile anesthetics protect the myocardium from RP damage. We reported that anesthetics improve function and metabolism on RP and reduce dysrhythmia development when given 10 min before, during, and 10 min after global ischemia (9, 10) or hypoxia (34) in isolated guinea pig hearts. Anesthetics also protect against dysrhythmias and improve mechanical, metabolic, and vascular endothelial function when given during low-flow perfusion for 1 day at 3°C (51). Isoflurane given for 45 min before and during 15 min of coronary occlusion in dogs enhances recovery of regional myocardial contractile function after 5-h RP (26). This effect was partially blocked by glibenclamide, which suggests a role for isoflurane to enhance KATP channel activation during ischemia and RP.
Preconditioning by the use of isoflurane, enflurane, and halothane was first shown to reduce infarct size in rabbit hearts (14). Isoflurane given for 15 min beginning 30 min before 30-min global ischemia in rabbit hearts partially mimicked ischemic IPC by reducing infarct size (12, 23). In dogs, isoflurane given alone or during four 5-min occlusions before 60 min of regional myocardial ischemia reduced infarct size (28). Because the protective effect was reversed by glibenclamide, this again supported a role for KATP channel opening. Anesthetics may activate A1 receptors or increase the sensitivity of adenosine A1 receptors to attenuated adenosine release (27). In turn, anesthetic-induced increases in PKC activity may underlie reduced infarct size (23) and improved contractility (54) after ischemia. In isolated human atrial trabecular muscle, isoflurane, but not halothane, has been shown to protect against simulated ischemia (45).
Using a protocol identical to the present study, we reported that SPC is as effective as IPC on improving basal and nitric oxide-mediated CF as well as cardiac rhythm, perfusion, mechanical, and metabolic function (40). Moreover, the protective effects of SPC and IPC were antagonized by glibenclamide, suggesting a common final mechanism via activation of KATP channels. If KATP channel antagonism also reverses the reduced Ca2+ loading effects of IPC and SPC, this would suggest that cardioprotection is afforded, at least in part, by reduced Ca2+ loading when KATP channels are open.
Overall, our study shows that SPC and IPC similarly reduce cytosolic Ca2+ loading and improve mechanical and metabolic function and contractile responsiveness to Ca2+ on RP. It is unclear if SPC and IPC produce myocardial protective effects via similar signaling pathways, but both interventions clearly lead to protection via a common final mechanism that reduces cytosolic Ca2+loading. Volatile anesthetics are often selected for patients with coronary artery disease at risk for ischemia and infarction during cardiac and noncardiac surgery. Temporary ischemia is often induced during cardiac surgery and angioplasty. Prior administration of a volatile anesthetic may be a more practical, but just as efficacious, method to protect the heart. Further research will determine if improved cardiac function after IPC and SPC results from improved Ca2+ homeostasis, or if improved Ca2+homeostasis is a consequence of effects of IPC and SPC on intracellular signaling pathways that lead to KATP channel opening.
Our studies in the guinea pig may not be easily compared with those of other species, especially the rat. Ischemia in rats causes a marked contracture beginning early during ischemia that is accompanied by a marked increase in dia[Ca2+] (D. F. Stowe, S. Fujita, and N. Novalija, laboratory observations); on early RP diastolic LVP remains elevated and dia[Ca2+] does not increase further. It is possible, but unlikely, that a portion of the 150-nM increase in dia[Ca2+] we found in each group on initial RP was derived from washout of indo 1-bound Ca2+ rather than from cytosolic Ca2+. Subepicardial LV tissue at the location of the fiber-optic probe was not infarcted but likely stunned. Sys-dia[Ca2+], because it is phasic, must represent cyto[Ca2+]. It is not possible to separate mitochondrial compartment Ca2+ from other compartments, e.g., nuclear, or Ca2+ from other cells, i.e., endothelial, vascular, or nerve. The flux of Ca2+ through each of these compartments is likely very slow. These compartments likely contribute little to the total rapid phasic signal of the myoplasm. We estimated the average noncytosolic Ca2+ from MnCl2 quenching at the end of each experiment. It is possible that Mn2+ might leak into this compartment over time. We assumed that the residual fluorescence recorded after quenching cytosolic fluorescence arises predominantly from mitochondrial Ca2+, but this may be an overestimate. Our preconditioning protocol of two 2-min occlusions with 5-min RP in between, though quite effective, is different from that used in other species by some investigators, e.g., 2–4 occlusions of 5 min each.
The authors thank Dr. Amadou Camara, Dr. Qun Chen, Dr. Ming-Tao Jiang, Jim Heisner, Samhita Shahane Rhodes, and Anita Tredeau for valuable contributions to this study.
This research was supported in part by National Heart, Lung, and Blood Institute Grants RO1-HL-58691 and RO1-5T32 GM-08377 and by a grant from Abbott Laboratories.
Portions of this work have appeared in abstract form: An JZ, Varadarajan SG, Smart SC, and Stowe DE. Anesth Analg 88: SCA47, 1999; Novalija E, An JZ, and Stowe DF. FASEB J 12: A712, 1998; and An JZ, Varadarajan SG, Smart SC, and Stow DF.Anesthesiology 89: A631, 1998.
Address for reprint requests and other correspondence: D. F. Stowe, Box M4280, 8701 Watertown Plank Rd., Medical College of Wisconsin, Milwaukee Regional Medical Center, Milwaukee, WI 53226.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society