Journal Logo


Contribution of Cytosolic Ionic and Energetic Milieu Change to Ischemia- and Reperfusion-Induced Injury in Guinea Pig Heart: Fluorometry and Nuclear Magnetic Resonance Studies

Hotta, Yoshihiro; Fujita, Michiko; Nakagawa, Junichi; Ando, Hiroaki; Takeya, Kazumi; Ishikawa, Naohisa; Sakakibara, Jinsaku*

Author Information
Journal of Cardiovascular Pharmacology: January 1998 - Volume 31 - Issue 1 - p 146-156
  • Free


The mechanism of acute ischemic injury in cardiac muscle has received extensive clinical attention, but further information obtained from living hearts about ionic and energetic states is needed to elucidate this mechanism. Ischemic muscle undergoes metabolic and functional alterations that are reversible if reperfusion occurs early (1). However, a time-dependent transition from reversible to irreversible changes in mechanical function occurs after 20-40 min during normothermic (37°C) ischemia. Although many time-dependent alterations in cellular function occur during ischemia, such as loss of total adenine nucleotides, a decrease in pH, and changes in the content of Na+, Ca2+, K+, and other ions, none of these alterations has been shown to be the major cause of the development of irreversible mechanical failure.

Of the many alterations of cellular function that have been reported, an excessive accumulation of intracellular Ca2+ ([Ca2+]i) in reperfused ischemic cells appears to be a consistent finding. Recent studies revealed an increase in cytosolic free Ca2+ concentration during early ischemia in intact mammalian hearts (2-4). Because excessive Ca2+ uptake is commonly observed in reperfused ischemic myocardium, it has been proposed that this additional Ca2+ may be a major cause of cellular damage (5). However, a direct role of Ca2+ in cellular dysfunction either during ischemia or with reperfusion has not been established.

It has been proposed that the increased uptake of Ca2+ by Na+-Ca2+ exchange in response to increased intracellular Na+ ([Na+]i). The increase in [Na+]i that occurs during ischemia may result from an increased H+-Na+ exchange coupled with decreased activity of Na+,K+-adenosine triphosphatase (ATPase; 6) or Na+ influx through the voltage-gated Na+ channel (7). Thus the hypothesis proposed by Lazdunski et al. (6) and Haigney et al. (7) is that H+ produced by anaerobic metabolism during ischemia is exchanged for extracellular Na+ and that the increased [Na+]i is exchanged for extracellular Ca2+; the excessive [Ca2+]i thereby results in cellular damage. In our study, the role of [Na+]i in ischemia was studied by altering [Na+]i levels with the combined use of asebotoxin-III (ATX-III) and dihydroouabain (DHO; 8-11).

We also investigated isolated heart mitochondria to obtain information about the ion-transport mechanism in mitochondrial membranes that may contribute to the cytosolic ion milieu. The marked increases of mitochondrial matrix fura-2 Ca2+ signals ([Ca2+]m) were reduced by the pretreatment of mitochondria with proton inhibitors of 5-(N-ethyl-N-isopropyl)-amiloride (EIPA) and omeprazole. In our previous study, we observed beneficial effects of these inhibitors on either H+-Na+ or H+-K+ exchanger in ischemia-reperfusion injuries of the Langendorff heart preparation (12) associated with a decrease in the Ca2+ level and a small reduction of ATP content with an increase of Pi, even at the last stage during a long ischemia. Ca2+ overloading on mitochondria, therefore, produced a deterioration in the contractility of muscle, probably associated with the thermodynamic deviation of diminished ATP utilization (13). These findings help to explain the "Ca paradox" at the organelle level of mitochondria.


Heart preparation and examination procedures

Hartley strain guinea pigs of either sex, weighing between 300 and 350 g, were anesthetized with diethyl ether and heparinized (250 IU, i.p.). The heart was rapidly excised, and the aorta was cannulated. The Langendorff heart preparations were perfused with Krebs-Henseleit solution (KH solution, pH 7.4, at 37°C) containing (in mM) NaCl, 115; NaHCO3, 25; KCl, 4.7; CaCl2, 2.0; MgCl2, 1.2; KH2PO4, 1.2; and glucose, 10. The solution was presaturated with a gas mixture containing 95% O2 and 5% CO2, and the heart was perfused either at a constant pressure of 75 cm H2O for nuclear magnetic resonance (NMR) measurement or at the flow rate of 7 ml/min by using a peristaltic pump for fluorometry. A latex balloon was inserted into the left ventricle and inflated enough to measure the left ventricular pressure (LVP) at the end-diastolic pressure of 10 mm Hg. Subsequent to the equilibration period of 30 min required to stabilize mechanical function, the perfused hearts were exposed to 40 min of global ischemia by clamping the perfusion flow line and were then reperfused for 30 min with the control medium of perfusate. Drugs were introduced into the perfusate at 2 min before the start of ischemia. The recovery percentages of LVP in each drug-treatment condition were determined and compared with that of the preischemic level.

Throughout the experiments, all animals were dealt with in accordance with the guidelines for animal experimentation set by the Japanese Association for Laboratory Animal Science.

NMR Measurements

Intracellular sodium [Na+]imeasurements by the inversionrecovery (IR) method and data analysis. The hearts were placed in the 20-mm probe of a GSX400 FT NMR spectrometer (Bo = 9.4T; JEOL, Tokyo, Japan) as described previously (11). The heart for 23Na NMR was paced at the driving rate of 3- to 4-Hz stimulation at the right ventricle via the electrode of a catheter filled with 3 M KCl-agar. The perfusate was quickly and continuously removed by the suction pump through thin polyethylene tubing from the bottom of an NMR glass sample tube. [Na+]i was selectively estimated at a frequency of 105.74 MHz by the IR method by using the change (VT) in interval between 180° and 90° (180°-VT-90°). The VT value was calculated from T1 of extracellular Na ([Na+]o; T1 of the perfusate, 57-65 ms; n = 5); the intensity of the Na resonance of the perfusate was nil at 39-45 ms (0.69 × T1 of [Na+]o). Sequential 64 double pulses of 180° and 90° flip angles were reported at an interval of 0.4 s (total acquisition time, 1 min). Capillaries filled with a mixture of 100 mM NaCl and 10 mM Dy (PPPi) placed within the NMR tube served as the external standard, and the relative intensities of each peak were used for quantitative analysis. Datum Station ALICE software (JEOL DATUM; Tokyo) was used to determine the area under each peak with a personal computer.

31P-NMR measurement. The perfused hearts were placed in a 31P-selective NMR probe and paced at the driving rate of 3- to 4-Hz stimulation via a 3 M KCl-agar electrode in the same manner as the 23Na-NMR experiment previously described. The perfusate was removed to the level of 4 cm from the bottom of the glass tube to adjust the shim-balance by using the water proton signal, which was easily detected even with the probe tuned for 31P. 31P-NMR spectra were recorded at 161.8 MHz with 45° flip-angle pulses at 2-s intervals. Spectra were acquired for 4 min and averaged from 90 transient samples.

The intracellular pH (pHi) was calculated from the chemical shift between the phosphocreatine (PCr) and inorganic phosphate (Pi) resonances by the following equation: Equation (1) where δo is the chemical shift of Pi from PCr expressed as parts per million (ppm; 14). PCr, Pi, and β-ATP were quantified by comparison with a capillary tube of standard methylenediphosphonic acid (MDP, 0.25 M) fixed inside the NMR tube. Phosphate peaks expressed as percentages of control values were determined by measuring the area under each resonance peak, as was done for the 23Na-NMR.


Cytosolicion fluorometry in Langendorff hearts. Cytosolic Na+, Ca2+ and H+ fluorometry were performed with a fluorometer (CAF100, or CAM230 with the biological fiberscope, FB30; Japan Spectroscopic Co., Tokyo, Japan) as described previously (15,16). The changes in the contents of each ion during ischemia and reperfusion in guinea-pig Langendorff hearts preloaded with a specific fluorescent indicator for each ion were measured, with a simultaneous recording of the mechanical performance (LVP). The hearts were field-stimulated at the driving rate of 3-4 Hz by using a pair of platinum-plate electrodes. After the loading of each indicator (added 0.025% cremophore EL), the heart was perfused in a nonrecirculating mode for 10 min to remove excess indicator. Subsequent to the equilibration period of 20 min required to stabilize mechanical performance, the changes in ion contents and LVP during ischemia and reperfusion were measured. The ion-fluorescence of the hearts that were incubated with fura-2 AM (2 μM, for Ca2+, 30 min as loading time) and SBFI AM (6 μM, for Na+ measurements, 90 min loading time) were measured at 500 nm as the ratio of strengths of fluorescence excited at 340 and 380 nm (R340/380). BCECF AM (2 μM, for pH measurements, 30 min loading time) was measured at 530 nm for the ratio of strengths of fluorescence excited at 450 and 500 nm (R450/500).

Intramitochondria calcium and pH measurements with fura-2 and BCECF. Heart mitochondria were isolated from guineapigs as previously described (17) and suspended in 70 mM sucrose, 210 mM mannitol, containing 20 mM MOPS-KOH (pH 7.4). Mitochondria were incubated at 24°C in a medium of various proportions of 70 mM sucrose, 210 mM mannitol, 1 mM MgCl2, 1 mM KH2PO4, 10 mM succinate, 5 mM pyruvate, 5 mM malate, and 20 mM MOPS adjusted to pH 7.4 with KOH and containing 10 μM fura-2 AM or BCECF AM, and 0.025% cremophor EL. A 0.5-ml aliquot of mitochondria was settled for 30 min on a glass coverslip on the chamber bottom (18), which was treated with poly-L-lysine to promote mitochondria adhesion. After dye loading, the mitochondria on the coverslip were mounted on the stage of an inverted microscope (CAM230; Japan Spectroscopic Co., Tokyo, Japan) and washed in dye-free solution for 10 min and continuously circulated with 10 ml medium solution. The mitochondrial pH measurement by BCECF was calculated by the method of James-Kracke (19). The solution temperature in the mitochondria chamber was 24°C. The final mitochondrial protein concentration, determined by the method of Lowry et al. (20) was adjusted to 30-35 mg/ml in the suspension medium.


Dihydroouabain was prepared by semisynthesis from ouabain (Merck, Darmstadt, Germany). Fura-2 AM and BCECF AM were obtained from Dojindo Laboratories (Kumamoto, Japan). SBFI AM, 5-(N-ethyl-N-isopropyl) amiloride hydrochloride, and cremophore EL were obtain from Molecular Probes (Eugene, OR, U.S.A.). Omeprazole (OMP) was kindly donated by Fujisawa Co. (Tokyo, Japan).

Statistical analysis

Data are expressed as the mean ± SEM. in Figs. 3 and 5. Estimation of significance was performed by Student's unpaired t test. ATP, Pi, and PCr content (during ischemia-reperfusion) were calculated on linear coordinates by using a personal computer. Probability values <5% were considered significant.

FIG. 3
FIG. 3:
Changes in left ventricular pressure (LVP; mm Hg) (a) and cytosolic Ca2+ signal (ratio) (b) in a preischemic (control) heart (A), after 1 min of ischemia (B), after 40 min of ischemia (C) and at 30 min after reperfusion (D) (for details, see Results section). Pretreatment without drugs (C), with the combined use of asebotoxin-III (0.5 μM) and dihydroouabain (50 μM; A + D), and 5-(N-ethyl-N-isopropyl)-amiloride (EIPA; 5 μM; E). *p < 0.05; **p < 0.01, significantly different from the control (drug-free) value. Data are the mean values of five preparations; vertical lines represent SEM.
FIG. 5
FIG. 5:
Changes of intracellular Na+ (Nai, mM) concentration estimated by 23Nai-NMR signals during ischemia/reperfusion in the hearts with various treatments without drugs (C), the combined use (A + D) of asebotoxin-III (0.5 μM) and dihydroouabain (50 μM), and 5-(N-ethyl-N-isopropyl)-amiloride (EIPA; 5 μM; E), as described in Materials and Methods. 23Nai-NMR was measured in preischemic (control) hearts (A), 14-15 min after the start of ischemia (B), 39-40 min after the start of ischemia (C), and 29-30 min after the start of reperfusion (D). *p < 0.05; **p < 0.01; significantly different from the control (drug-free) value. Vertical lines represent SEM (n = 5).


Changes in fura-2 Ca2+ signals during ischemia and reperfusion

Simultaneous recordings of fluorescence from fura-2-loaded heart preparations and LVP during ischemia-reperfusion are shown in Fig. 1a and b. During the 40-min ischemia, the fura-2 Ca2+ signals changed triphasically; the level and amplitude of transient Ca2+ (TCa) increased when the LVP decreased exponentially during the first 3-5 min of ischemia (phase I, B) and were maintained for 20-30 min at a high steady level (phase II), followed by a significant elevation of TCa (A: diastolic Ca2+ ratio, 0.49 ± 0.04 → C: TCa ratio, 0.72 ± 0.04; p < 0.01; n = 5) level in the final stage with no twitch (phase III, C). On reperfusion with new KH solution, the increased diastolic Ca2+ level of TCa (C) returned rapidly to control level (A) with the recovery of LVP.

FIG. 1
FIG. 1:
Simultaneous recordings of the changes in left ventricular pressure (LVP) and fura-2 Ca2+ signals during ischemia/reperfusion of a guineapig Langendorff heart preparation. a: During ischemia for 40 min, fura-2 Ca2+ signals changed triphasically (phases I, II, and III), LVP was restored during the reperfusion, with a rapid return to the control level, as described in the Results section. b: Bottom trace in magnified time scale shows the alteration of contraction strength and transient Ca2+ (TCa) in a preischemic (control) heart (A), after 1 min of ischemia (B), after 40 min of ischemia (C), and at 30 min after reperfusion (D).

Different effects of the combined use of asebotoxin-III and dihydroouabain, and H+-Na+ inhibitor on the triphasic changes of TCa

Figure 2a and b show a typical change of contractile (LVP) and Ca2+ signals in drug-free and drug-treated hearts exposed to ischemia-reperfusion. The values obtained are presented in Table 1 and the changes of LVP (mm Hg) and Ca2+ signals (ratio) are summarized in Fig. 3a and b. The combined use of ATX-III (0.5 μM) and DHO (50 μM), which caused an increase in Nai, resulted in an elevation of Cai level in phase III [156.2 ± 6.5% (drug-free) vs. 171.2 ± 11.2%; p < 0.01; n = 5; Figs. 3b,C, and 2a), and induced a strong contracture with a loss of the signal of twitch after reperfusion (Figs. 2a and 3a,D). The inhibitor of H+-Na+ exchange, EIPA, 5μM produced a significant restorative effect on LVP associated with a decrease in the Ca2+ level in phase III (156.2 ± 6.5% vs. 137.2 ± 6.2%; p < 0.05; n = 5; Figs. 3b,C, and 2b). Just before the start of ischemia, the combined use of ATX-III and DHO caused an rapid increase in the LVP with an elevation of the diastolic Ca2+ level, whereas EIPA caused slight decreases in the LVP and the Ca2+ signal.

FIG. 2
FIG. 2:
Effects of positive inotropic agents (a) and H+-Na+ inhibitor (b) on the triphasic changes of Ca2+ signals. a: The combined use of asebotoxin-III (ATX-III, 0.5 μM) and dihydroouabain (DHO, 50 μM) caused an increase in the Ca2+ level in phase III and induced a strong contracture with a loss of twitch in left ventricular pressure (LVP) after reperfusion. b: Top trace, simultaneous recordings of the changes of LVP and Ca2+ signals induced by ischemia/reperfusion in a heart without drugs. 5-(N-ethyl-N-isopropyl)-amiloride (EIPA; 5 μM) had a significant restorative effect on LVP, associated with a decrease in Ca2+ level in phase III (bottom trace). The underlines before ischemia were drug-treatment time.
Left ventricular pressure and cytosolic electrolytes and pH changes during ischemia/reperfusion of the Langendorff hearts under drug action

Changes in SBFI-Nai signals and 23Nai-NMR signals during ischemia and reperfusion

Increases in the SBFI-Nai and 23Nai-NMR signals, as shown in Fig. 4a and b, were observed during early ischemia (1-5 min, phase I), and the signals remained at constant levels from the middle to the end of the ischemia, at which time the Ca2+ signal level increased (phase II and III). Both Nai signals were returned to the control level of preischemia by reperfusion. The rapid increase in SBFI-Nai induced by the combined use of DHO and ATX-III was observed with loss of twitch after reperfusion (Fig. 4a).

FIG. 4
FIG. 4:
Simultaneous recordings of the changes in left ventricular pressure (LVP) and SBFI-Nai signals (a) or 23Nai-NMR signals (b) during ischemia/reperfusion. Increases in SBFI-Nai and 23Nai-NMR signals were observed early during ischemia (1-5 min; phase I) and remained at a constant high level to the end of ischemia (15-40 min; phases II and III; [(a) top and second traces, and (b) top]. a: The ischemia (15 min)-induced SBFI-Nai elevation was returned to the preischemic level by reperfusion, increased by monensine (MON, 10 μM), and then reduced by tetrodotoxin (TTX, 30 μM; (top). The rapid increase in SBFI-Nai by the combined use of DHO and asebotoxin-III (ATX-III) induced a strong contracture with a loss of twitch after reperfusion (third trace). b: The changes of 23Nai-NMR between the 5-(N-ethyl-N-isopropyl)-amiloride (EIPA; 5 μM) and EIPA-free hearts during ischemia/reperfusion were similar to those of the SBFI-Nai signals (bottom).

The change of intracellular Na+ concentration (Table 1) estimated by 23 Na-NMR signals (Fig. 4b) during ischemia and reperfusion in a heart pretreated with these drugs (C, drug-free; A + D, ATX-III and DHO; and E, EIPA) is shown in Fig. 5. The Na+ concentration in the preparation with combination of ATX-III and DHO was significantly higher than that of the control (drug-free) preparation at the early stage of ischemia (150.0 ± 4.4% vs. 177.7 ± 5.1%; p < 0.05; n = 5; Fig. 5B) and also at the end of the long reperfusion period (40 min; 118.3 ± 4.4% vs. 170.0 ± 7.8%; p < 0.01; Fig. 5D). The cytosolic Na+ concentration difference between EIPA-free and EIPA-treated hearts during ischemia and reperfusion did not change significantly (150.0 ± 4.4% vs. 127.0 ± 5.9%, Fig. 5B; 131.7 ± 4.4% vs. 125.3 ± 3.7 mM;Fig. 5C; 118.3 ± 4.4% vs. 111.3 ± 1.5%; Fig. 5D; n = 5).

Changes of pHi and high-energy phosphate contents determined by measuring 31P-NMR spectra during ischemia and reperfusion

31P-NMR spectra from representative hearts in each of two groups (presence and absence of EIPA,) are shown in Fig. 6a. The PCr contents of ischemic heart preparations diminished rapidly with a concomitant increase in Pi production and loss of contractility but with only a small reduction of ATP content. There was an ATP peak even at the end stage of 40-min ischemia (ATP, 24.6 ± 1.1% vs. 41.2 ± 4.3%; p < 0.01; ATP/Pi, 0.12 ± 0.01 vs. 0.23 ± 0.04; p < 0.05; n = 5; Fig. 7) and a more complete recovery of high-energy phosphate contents (ATP, 41.1 ± 3.1% vs. 55.4 ± 2.0%; p < 0.01; PCr, 60.1 ± 5.3% vs. 80.5 ± 7.8%; p < 0.05; Pi, 119.6 ± 3.8% vs. 106.3 ± 5.3%; p < 0.05; PCr/Pi, 0.39 ± 0.04 vs. 0.59 ± 0.06; p < 0.05; ATP/Pi, 0.18 ± 0.01 vs. 0.28 ± 0.02; p < 0.01; n = 5; Fig. 7 and Table 1) after reperfusion. The drop in pHi, estimated from the Pi-PCr chemical shift of 31P-NMR (Fig. 6b) suggested an augmented H+-Na+ exchange during ischemia. EIPA (5 μM) had a marked restorative effect on the LVP during reperfusion (32 ± 3.3% vs. 58.7 ± 6.0%; p < 0.01; n = 5; Fig. 7) associated with an increase in intracellular proton concentration during ischemia (pH 6.0 ± 0.01 vs. 5.9 ± 0.04; p < 0.05; n = 5; Fig. 6b and Table 1).

FIG. 6
FIG. 6:
Changes of 31P-NMR spectra (a) and pHi (b) determined during ischemia/reperfusion. a: 31P-NMR spectra collected over a 4-min period from representative control (left) and 5-(N-ethyl-N-isopropyl)-amiloride (EIPA; 5 μM) (right) hearts obtained during control perfusion (top), at the end of ischemia (37-40 min; middle), and at the end of reperfusion (27-30 min; bottom). The PCr contents of ischemic heart preparations diminished rapidly with a concomitant increase in Pi production and loss of contractility, but with a small reduction of ATP content. In a heart pretreated with EIPA, a peak of β-ATP (↓) was obtained at the end stage of a long ischemia and reperfusion. b: Plots of pHi during ischemia/reperfusion in the presence and absence of EIPA. The decrease in pHi during ischemia in the presence of EIPA ([-•-]) was greater than that in the control ([-○-]; pH 6.1 ± 0.01 → 5.9 ± 0.04; *p < 0.05). Vertical lines represent SEM (n = 5).
FIG. 7
FIG. 7:
Relation of contractile function (LVP) and high-energy phosphates (ATP, Pi, PCr, PCr/Pi, and ATP/Pi) estimated by 31P-NMR spectra in the presence and absence of 5-(N-ethyl-N-isopropyl)-amiloride (EIPA) during ischemia/reperfusion. At the end stage of ischemia of a heart pretreated with EIPA, the level of ATP was partially restored and showed significantly greater recovery during reperfusion than did the control (drug-free). *p < 0.05; **p < 0.01; significantly different from the control (drug-free) value. Vertical lines represent SEM (n = 5).

Mitochondrial Ca2+ measurement

The fura-2 Ca2+ signal of the mitochondria ([Ca2+]m) immersed in extremely high Ca2+ concentration (30 mM) of the perfusate underwent only a steady, gradual, but slight elevation of 110% (ratio 0 → 0.11; n = 5) during the perfusion period of 30 min. The displacement with perfusate of normal matrix Ca2+ concentration (100 nM) produced a rapid and intense increase up to 524.2 ± 20% (ratio 0.11 → 0.58 ± 0.02; n = 5) relative to that at displacement with perfusate as 100% (Fig. 8a; 1). Pretreatment of mitochondria with either 10 μM EIPA or 0.1 mM OMP depressed by about half the elevation that occurred in the drug-free perfusate [EIPA ratio, 0.11 → 0.28 ± 0.03 (254.9 ± 24.0%); p < 0.01; n = 4; OMP ratio, 0.12 ± 0.01 → 0.30 ± 0.03 (260 ± 26.0%); p < 0.01; n = 4].

FIG. 8
FIG. 8:
Recordings of effects of the elevation and the reduction of Na+, K+, and Ca2+ concentrations in perfusate on the intramitochondrial Ca2+ (a) and pH signals (b) from the fura-2- or BCECF-loaded mitochondria preparation. a: (1), The fura-2 Ca signal gradually increased with high Ca2+ (30 mM), and perfusion with the low-Ca2+-content solution (100 nM) perfused rapidly increased the intramitochondrial Ca2+ level. (2), The perfusate acidification in a low-Ca2+-content solution (100 nM) produced a more rapid Ca2+ signal increase than that with changes in Ca2+-content perfusate (a (1)). 5-(N-ethyl-N-isopropyl)-amiloride (EIPA; 10 μM) reduced the increase of intramitochondrial Ca2+ signals by the displacement of perfusate [a (1) and (2)]. b: The BCECF-pH signal was rapidly increased by the addition of 10 mM Na+ and 110 mM K+ into Na+-, K+-, and Ca2+-free perfusate, but not by 100 nM Ca2+ perfusate. With the addition of Ca2+ ≤30 mM, the pH signal rapidly increased, and the perfusion with a low-Ca2+-content solution decreased the intramitochondrial pH. EIPA (10 μM) reduced the degree of the elevation (A) and the depression (B) of pH signals. After low-Ca2+ perfusion, the change of fluorescent signal used to determine the maximal or minimal contents of ions was smaller (a and b) than that of the normal condition (b, C).

Changing the pH from 7.4 to 6.5 in a low-Ca2+ content perfusate (100 nM) elevated [Ca2+]m to the maximal extent, whereas an increase to pH 8.0 had no effect (Fig. 8a; 2). The appreciable [Ca2+]m elevation induced by acidification (ratio, 0.01 → 0.76 ± 0.03; n = 5) was reduced by a prior infusion of the proton-antiport inhibitors 10 μM EIPA [ratio, 0.05 ± 0.02; p < 0.01; n = 4; → 0.78 ± 0.05 (with drug-free perfusate] or 0.1 mM OMP [ratio, 0.01 → 0.03 ± 0.02; p < 0.01; n = 3; → 0.75 ± 0.06 (with drug-free perfusate)] and was greater and more rapid than that by perfusion with low-Ca2+ perfusate (Fig. 8a; 1)

The administration of Na+ at 10 mEq or K+ at 110 mEq in perfusate of physiologic cytosolic conditions to mitochondria rapidly increased the BCECF-pH signal ([pH]m; Na+, ΔpH 0.24 ± 0.03; K+, ΔpH 0.34 ± 0.03). These elevations (ΔpH) were depressed by 64.6 ± 9.2% (as 100% drug-free ΔpH · ΔpH 0.12 ± 0.02; p < 0.05; n = 5) or by 46.3 ± 11.6% (ΔpH 0.11 ± 0.02; p < 0.05; n = 4) in the presence of 10 μM EIPA or 0.1 mM OMP. In contrast, the cytosolic Ca2+ concentration of 100 nM had almost no effect on the [pH]m, and further elevation up to Ca2+ 30 mM showed a similar rapid increase in pHm (104.8 ± 0.5%; Fig. 8b). Subsequently, the [pH]m decreased gradually by perfusion with normoxic low Ca2+ perfusion (100 nM) in Ca2+-loaded mitochondria preparations (97.1 ± 0.8%; p < 0.05; n = 4).


In our study, we measured the changes in intracellular Ca2+ transient signals (TCa) and the LVP simultaneously by using a fura-2 Ca2+ fluorescent probe during ischemia and reperfusion of guinea-pig hearts. The fura-2 Ca2+ signal increased markedly during ischemia, and reperfusion reversed this increase, as shown in Fig. 1. The basal level and amplitude of TCa increased when the LVP was depressed exponentially at an early stage (phase I) of ischemia, and the increase was sustained for 30 min (phase II) followed by a marked further increase during the last 10 min (phase III) of 40-min ischemia.

The combined use of ATX-III and DHO, which increase Nai(8-11) increased the Ca2+ level in diastole without any change in the amplitude of TCa at the last stage of ischemia and induced strong contracture after reperfusion (Fig. 2a). From our observation of both SBFI-Nai and 23Nai-NMR signals, it became clear that [Na+]i increased rapidly during phase I and remained at a constant level in phase III (Fig. 4a and b, Fig. 5, and Table 1). With regard to the fura-2 Ca2+ increase during the last stage of phase III, the cytosolic free Ca2+ at this phase would be increased by Na+i-Ca2+o exchange. The ischemia-induced fura-2 Ca2+ level increase for the diastolic pressure was returned rapidly to the control level by first the reperfusion and then showed gradual Ca2+ elevation with a similar contracture and a loss of twitch of LVP.

Many of the H+-Na+-exchange inhibitors, amiloride derivatives have been demonstrated to protect the heart from aspects of ischemic insult (21-28). In our study, EIPA reduced the Ca2+ level in phase III as shown in Fig. 2b, resulting in an 86.6 ± 6.0% (n = 5) restorative effect on LVP (drug-free; 58.9 ± 4.8%, n = 5) associated with an increase in intracellular H+, thereby stopping the increase of [Na+]i(Figs. 4 and 5) to promote Na+i-Ca2+o exchange. Weiss et al. (22) and Murphy et al. (23) also showed that during hypoxia, the cellular Ca2+ in amiloride-treated hearts was lower than that of drug-free controls.

In our study, the pretreatment of guinea-pig hearts with EIPA also had a significant preservation effect on tissue levels of ATP and PCr measured at the end stage of ischemia or after reperfusion (Figs. 6 and 7 and Table 1). In experiments using rat hearts, Pike et al. (27) reported that ATP was well preserved and PCr resynthesis was higher in EIPA-pretreated hearts than in control hearts during reperfusion. Therefore the interrelation between cytosolic Ca2+ and ATP depletion and lethal myocyte injury, and other factors that vary in parallel with ATP depletion could also be involved. These findings lead to the conclusion that an increase in cytosolic Ca2+ is an important mechanism of myocardial ischemic injury.

The pHi estimated by 31P-NMR declined during ischemia in our study. The pHi at the end stage of ischemia in EIPA-treated hearts (5.9) was slightly (0.02 units) lower than that in control hearts (6.1; Figs. 6 and 7 and Table 1). Other investigators (27-29) did not clearly detect the progressive pHi decline during the period preceding ischemia by 31 P-NMR by using rat or ferret hearts with preischemic treatment with amiloride derivatives. In cardiac ischemia, protons in the cytosol are produced mainly from anoxic glycolytic ATP turnover and CO2 retention and buffered by Pi which is produced by PCr breakdown to maintain ATP levels or other intracellular protein residues (30,31). However, Koike et al. (32) demonstrated in neonatal rabbit hearts that the pHi decline (5.24) was 1.0 unit lower than that in control hearts (6.24) after the administration of 5-(N,N-dimethyl) amiloride to St. Thomas Hospital's cardioplegic solution before the onset of 45-min ischemia. It is suggested that H+-Na+ exchange plays an important role not only during reperfusion but also during ischemia for the development of postischemic cardiac dysfunction, probably by first inducing a [Na+]i elevation and secondary Ca2+ overload.

As shown in Fig. 1, rapid acute global ischemia in perfusion stopped the guinea-pig Langendorff hearts, rapidly elevating fura-2 Ca2+ transient signals, and the reperfusion after ischemia caused the Ca2+ signals to decrease, without any further increase. The results of these experiments are in agreement with studies concerning Indo-1 Ca2+ transient signals in rabbit hearts (2) and aequolin Ca2+ transient signals in ferret hearts (3). The cause of intracellular Ca2+ increases during ischemia or reperfusion or both are still controversial, although more and more evidence favors the view that the cytosolic Ca2+ increases during ischemia and even more so during reperfusion (33).

In connection with our investigation of Ca2+ systems that are important to the behavior of mitochondria during reperfusion, we examined the ion-transport mechanisms of mitochondrial membranes by using a superfusion technique on mitochondria attached to the polylysinecoated coverglasses and fluorometric analysis of proton and calcium ions. The intramitochondrial pH ([pH]m) was increased by the superfusion of physiologic cytosolic concentrations of 10 mM Na+ or 110 mM K+ or both, providing evidence for the existence of H+-Na+ or H+-K+ exchange in mitochondrial membranes (Fig. 8b). The [pH]m was decreased by EIPA and by OMP, suggesting the existence of monovalent cation-proton exchange mechanisms in mitochondrial membranes, as is likely in the sarcolemma (see Results).

Although a normal cytosolic Ca2+ concentration (100 nM) did not affect the [pH]m, an extremely high concentration of Ca2+ (>10 mM) produced an increase in [pH]m(Fig. 8b). The intramitochondrial Ca2+ signal ([Ca2+]m) increased steadily and linearly but remained at only 10% excess for 1 h with 30 mM Ca2+ (Fig. 8a; 1), as also observed by Vasington and Murphy (34). However, [Ca2+]m increased rapidly, reaching a high maximum level during the displacement with the perfusate of a solution containing the physiologic cytosolic level of Ca2+ (∼100 nM), as shown in Fig. 8a; 1). The pHi during the period of ischemia in nonperfused Langendorff hearts decreased progressively to ∼6.0 at the time point of 40 min of ischemia (Fig. 6b). In the superperfused mitochondrial preparations, we observed the appreciable increase of [Ca2+]m induced by a decrease in pH from 7.4 to 6.5 in a low Ca2+-content perfusate. The perfusate acidification produced a more rapid [Ca2+]m elevation (Fig. 8a; 2) than did changes in Ca2+ concentration in the perfusate (Fig. 8a; 1), but alkalization of the perfusate to 8.0 did not increase [Ca2+]m. It is thus anticipated that in pathologic conditions, the cytosolic acidic pH (protons increases) connects with Ca2+ uptake into mitochondria.

This marked [Ca2+]m elevation with either the change in the Ca2+ concentration or acidification of the perfusate was reduced by pretreatments of mitochondria with the proton-pump inhibitors EIPA and OMP, which had beneficial effects on LVP in the ischemia-reperfusion injury of Langendorff hearts (restorative ratio by EIPA, 86.6 ± 6.0%; OMP, 80.1 ± 4.4%, vs. drug-free, 58.9 ± 4.8%; n = 5; 12). These findings led to the conclusion that mitochondrial Ca2+ overloading during reperfusion plays an essential role in the homeostasis of Ca2+ for the maintenance of cell functions in the heart, as reported by Crompton (35).

Miyata et al. (36) also directly measured mitochondrial free Ca2+ ([Ca2+]m) by Mn2+ quenching of the cytosolic Ca2+ ([Ca2+]c) of indo-1-loaded rat myocytes during hypoxia and reoxygenation. They observed that during anoxia, [Ca2+]m and [Ca2+]c increased slowly and in parallel and that at reoxygenation, the rapid decrease in [Ca2+]c was blunted, and [Ca2+]m showed an immediate increase in these cells. In addition, Allen et al. (37) reported that in mitochondria isolated from rat hearts, the fura-2 Ca2+ signal ([Ca2+]m) remained steady during hypoxia and increased sharply on reoxygenation. These findings further support the view that reoxygenation after anoxia in mitochondria is associated with an impairment of the electron-transfer system in state 4 respiration (38) and oxidative phosphorylation in particular (24).

A key feature of the transition from reversible to irreversible cell injury is mitochondrial dysfunction, which may open nonspecific pores in the mitochondrial inner membrane and allow the collapse of ΔμH+ and mixing of low-molecular weight (<1,500 Da) compounds of the cytosol and matrix (35,39,40). Pore opening can be induced in vitro by exposure of isolated mitochondria to high Ca2+ and Pi concentrations (41). Therefore pathologic pore formation requires a higher Ca2+ concentration (25 μM matrix-free Ca2+) than that at which physiologic dehydrogenase control occurs (1-2 μM). Mitochondrial Ca2+ transport will protect the cytosol against hypercalcemia under a limited range of pathologic conditions.

In our study, we observed that the changes in fura-2 Ca2+ ratio in guinea-pig Langendorff hearts could be recorded with sufficient accuracy during ischemia and reperfusion. The intramitochondrial fura-2 Ca2+ signals were increased rapidly by Ca2+ concentration change and acidification of the perfusate similar to reperfusion after global ischemia in the Langendorff heart preparations. These results provide basic information about the ion transport mechanism (Ca2+-H+ symport, Na+, K+, Ca2+, and proton antiport) of mitochondrial membranes, thereby adding to our understanding of the mechanisms of some types of positive inotropy and ischemia-reperfusion injury of the heart. The use of fluorescent indicators in in vivo preparations can be particularly useful in testing emerging concepts of the physiologic relevance of mitochondrial ion transport-coupled electrochemical proton gradients under pathologic conditions. Further studies of the mitochondria are needed to clarify the possible mechanisms of coupling between Ca2+/H+ transport and additional production (free radicals and heat-shock proteins, etc.) during reperfusion.

Acknowledgment: We thank Mr. M. Naruse (Aichi Medical University) for his skillful assistance with the NMR measurements. This work was supported in part by a grant-in-aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan.


1. Neely JR, Grotyoharn LW. Role of glycolytic products in damage to ischemic myocardium: dissociation of adenosine triphosphate levels and recovery of function of reperfused ischemic hearts. Circ Res 1984;55:816-24.
2. Lee H-C, Mohabir R, Smith N, Franz MR, Clusin WT. Effect of ischemia on calcium-dependent fluorescence transients in rabbit hearts containing indo-1: correlation with monophasic action potentials and contraction. Circulation 1988;78:1047-59.
3. Kihara Y, Grossman W, Morgan JE. Direct measurement of changes in intracellular calcium transients during hypoxia, ischemia and reperfusion in intact mammalian heart. Circ Res 1989;65:1029-44.
4. Steenbergen C, Murphy E, Watts JA, London RE. Correlation between cytosolic free calcium, contracture, ATP, and irreversible ischemic injury in perfused rat heart. Circ Res 1990;66:135-46.
5. Pierce GN, Czubryt MP. The contribution of ionic imbalance to ischemia/reperfusion-induced injury. J Mol Cell Cardiol 1995;27:53-63.
6. Lazdunski M, Frelin C, Vigne P. The sodium/hydrogen exchange system in cardiac cells: its biochemical and pharmacological properties and its role in regulating internal concentrations of sodium and internal pH. J Mol Cell Cardiol 1985;17:1029-42.
7. Haigney MCP, Lakatta EG, Stern MD, Silverman HS. Sodium channel blockade reduces hypoxic sodium loading and sodium-dependent calcium loading. Circulation 1994;90:391-9.
8. Takeya K, Hotta Y, Yajima M, Sakakibara J. Frequency-dependent nature of cardiac effect of asebotoxin III. 8th International Congress of Pharmacology, Abstracts, No. 904, Tokyo, 1981.
9. Takeya K, Yajima M, Kobayashi S, Hotta Y, Sakakibara J. Synergistic relation between cardiac glycosides and grayanotoxin I on the positive inotropic effect. Jpn J Pharmacol 1981;31(suppl I):115P.
10. Hotta Y, Takeya K, Kobayashi S, Harada N, Sakakibara J, Shirai N. Relationship between structure, positive inotropic potency and lethal dose of grayanotoxins in guinea pig. Arch Toxicol 1980;44:259-67.
11. Hotta Y, Ando H, Takeya K, Sakakibara J. Direct measurement of increased myocardial cellular 23Na NMR signals in perfused guinea-pig heart induced by dihydroouabain and grayanotoxin-I. Mol Cell Biochem 1994;139:59-70.
12. Hotta Y, Takeya K. Cytosolic electrolytes interchange during ischemia and reperfusion of the heart under drug action. 6th International Symposium, Pharmacological control of calcium and potassium homeostasis, biological, therapeutical and clinical aspects. Florence (Italy) 1994:54.
13. Takeya K, Hotta Y. The role of mitochondria for the contractility regulation of calcium ions in ischemic myocardium. 6th International Symposium, Pharmacological control of calcium and potassium homeostasis, biological, therapeutical and clinical aspects. Florence (Italy) 1994;39.
14. Moon RB, Richards JH. Determination of intracellular pH by 31P nuclear magnetic resonance. J Biol Chem 1973;248:7276-8.
15. Hotta Y, Nishiyama M, Ando H, Kato M, Takeya K. Different modes of the effects of dihydroouabain and isoproterenol on the fura-2 calcium transients and contractile response of the isolated guinea-pig heart. J Aichi Med Univ Assoc 1989;17:911-5.
16. Hotta Y, Ando H, Fujita M, Nakagawa J, Takeya K, Sakakibara J. Different effects of isoproterenol and dihydroouabain on cardiac Ca2+ transients. Eur J Pharmacol 1995;282:121-30.
17. Brierley GP, Jurkowitz MS, Farooqui T, Jung DW. K+/H+ antiport in heart mitochondria. J Biol Chem 1984;259:14672-8.
18. Kohmoto O, Spitzer KW, Movsesian MA, Barry WH. Effects of intracellular acidosis on [Ca2+]i transients, transsarcolemmal Ca2+ fluxes, and contraction in ventricular myocytes. Circ Res 1990;66:622-32.
19. James-Kracke MR. Quick and accurate method to convert BCECF fluorescence to pHi: calibration in three different types of cell preparations. J Cell Physiol 1992;151:596-603.
20. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-75.
21. Tani M, Neely JR. Role of intracellular Na+ in Ca2+ overload and depressed recovery of ventricular function of reperfused ischemic rat hearts. Circ Res 1989;65:1045-56.
22. Weiss RG, Lakatta EG, Gerstenblith G. Effects of amiloride on metabolism and contractility during reoxygenation in perfused rat hearts. Circ Res 1990;66:1012-22.
23. Murphy E, Perlman MA, London RE, Steenbergen C. Amiloride delays the ischemia-induced rise in cytosolic free calcium. Circ Res 1991;68:1250-8.
24. Duan J, Karmazyn M. Protective effects of amiloride on the ischemic reperfused rat heart, relation to mitochondrial function. Eur J Pharmacol 1992;210:149-57.
25. Counillon L, Scholz W, Lang HJ, Pouysségur J. Pharmacological characterization of stable transfected Na+/H+ antiporter isoforms using amiloride analogs and a new inhibitor exhibiting anti-ischemic properties. Mol Pharmacol 1993;44:1041-5.
26. Scholz W, Albus U, Lang HJ, et al. Hoe 694, a new Na+/H+ exchange inhibitor and its effects in cardiac ischemia. Br J Pharmacol 1993;109:562-8.
27. Pike MM, Luo CS, Clark D, et al. NMR measurements of Na+ and cellular energy in ischemic rat heart: role of Na+-H+ exchange. Am J Physiol 1993;265:H2017-26.
28. Hendrikx M, Mubagwa K, Verdonck F, et al. New Na+-H+ exchange inhibitor HOE 694 improves postischemic function and high-energy phosphate resynthesis and reduces Ca2+ overload in isolated perfused rabbit heart. Circulation 1994;89:2787-98.
29. Khandoudi N, Bernard M, Cozzone P, Feuvray D. Intracellular pH and role of Na+/H+ exchange during ischemia and reperfusion of normal and diabetic rat hearts. Cardiovasc Res 1990;24:873-8.
30. Allen DG, Orchard CH. Myocardial contractile function during ischemia and hypoxia. Circ Res 1987;60:153-68.
31. Dennis SC, Coetzee WA, Cragoe EJ, Opie LH. Effects of proton buffering and of amiloride derivatives on reperfusion arrhythmias in isolated rat hearts: possible evidence for an arrhythmogenic role of Na+-H+ exchange. Circ Res 1990;66:1156-9.
32. Koike A, Abe T, Hotta Y, Takeya K, Kodama I, Toyama J. Protective effects of dimethyl amiloride, a potent Na+/H+ exchange inhibitor, against post-ischemic myocardial dysfunction: 31P-NMR measurements of pHi and cellular energy in isolated perfused rabbit hearts. J Thorac Cardiovasc Surg 1996;112:765-77.
33. Opie LH, Coetzee WA. Role of calcium ions in reperfusion arrhythmias: relevance to pharmacologic intervention. Cardiovasc Drugs Ther 1988;2:623-36.
34. Vasington FD, Murphy JV. Ca++ uptake by rat kidney mitochondria and dependence on respiration and phosphorylation. J Biol Chem 1962;237:2670-7.
35. Crompton M. The roll of Ca2+ in the function and dysfunction of heart mitochondria. In: Langer GA, ed. Calcium and the heart. New York: Raven Press, 1990:167-98.
36. Miyata H, Lakatta EG, Stern MD, Silvermann HS. Relation of mitochondrial and cytosolic free calcium to cardiac myocyte recovery after exposure to anoxia. Circ Res 1992;71:605-13.
37. Allen SP, Darley-Usmar VM, McCormack JG, Stone D. Changes in mitochondrial matrix free calcium in perfused rat hearts subjected to hypoxia-reoxygenation. J Mol Cell Cardiol 1993;25:949-58.
38. Nishida T, Inoue T, Kamiike W, Kawashima Y, Tagawa K. Involvement of Ca2+ release and activation of phospholipase A2 in mitochondrial dysfunction during anoxia. J Biochem 1989;106:533-8.
39. Gunter TE, Gunter KK, Sheu S-S, Gavin CE. Mitochondria calcium transport: physiological and pathological relevance. Am J Physiol 1994;267:C313-39.
40. Hansford RG. Physiological role of mitochondrial Ca2+ transport. J Bioenerg Biomembr 1994;26:495-508.
41. Langer GA. Calcium and the heart: exchange at the tissue, cell, and organelle levels. FASEB J 1992;6:893-902.

Guinea-pig heart; Mitochondria; H+-Na+, Na+-Ca2+, H+-K+ exchange; Ca2+ overload; 5-(N-ethyl-N-isopropyl)-amiloride (EIPA); Omeprazole (OMP)

© Lippincott-Raven Publishers