Correlation of Ischemia-Induced Extracellular and Intracellular Ion Changes to Cell-to-Cell Electrical Uncoupling in Isolated Blood-Perfused Rabbit Hearts
Background The relationships between the metabolic, ionic, and electrical changes of acute ischemia have not been determined precisely because they have been studied under different experimental conditions. We used ion-selective electrodes, nuclear magnetic resonance spectroscopy, and the four-electrode method to perform four series of experiments in the isolated blood-perfused rabbit heart loaded with 5F-BAPTA during 30 to 35 minutes of no-flow ischemia. We sought to determine the relationship between changes in phosphocreatine (PCr), adenosine triphosphate (ATP), intracellular calcium ([Ca2+]i), intracellular pH (pHi), extracellular potassium ([K+]e), extracellular pH (pHe), and whole-tissue resistance (rt).
Methods and Results In the first 8 minutes of ischemia, [K+]e rose from 4.9 to 10.8 mmol/L, PCr fell by 90%, ATP decreased by 25%, and pHi and pHe decreased by 0.5 U, while [Ca2+]i and rt changed only slightly. Between 8 and 23 minutes, [K+]e changed only slightly; pHi, pHe, and ATP continued to fall, and [Ca2+]i rose. rt did not increase until >20 minutes of ischemia, when pHi was <6.0 and [Ca2+]i had increased more than threefold. The increase in rt, indicating electrical uncoupling, coincided with the third phase of the [K+]e change.
Conclusions Our study suggests that cellular uncoupling occurs only after a significant rise in [Ca2+]i and fall in pHi and that these ionic and electrical changes can be identified by the change in [K+]e. Our study underscores the importance of using a common model while attempting to formulate an integrated picture of the ionic, metabolic, and electrical events that occur during acute ischemia.
Acute myocardial ischemia is characterized by interrelated metabolic ionic and neurohumoral events that alter membrane properties of cardiac cells, causing electrophysiological changes that often lead to life-threatening ventricular arrhythmias.1 These include but are not limited to the depletion of PCr and ATP, a fall in pHi and pHe, a rise in [Ca2+]i and [K+]e, and cell-to-cell electrical uncoupling. The various changes have been studied individually in a variety of experimental models using different species and different methods. Thus, it is unclear as to which events occur concurrently and which occur sequentially. To address these issues, we measured the changes in [Ca2+]i, pHi, PCr, ATP, [K+]e, pHe, and rt in the same model under identical experimental conditions.
We performed four separate sets of experiments. In each, juvenile Dutch belted rabbits (HRP, Inc; Denver, Pa) weighing ≈500 g were heparinized and anesthetized with thiamylal sodium (30 mg/kg IV). The hearts were rapidly excised and perfused retrogradely through the aorta from a reservoir 90 cm above the heart.2 The nonrecirculating perfusate contained insulin (1 U/L), heparin (400 U/L), albumin (2 g/L), dextran (MW, 70 000:40 g/L) and Tyrode's solution (mmol/L: Na+ 149, K+ 4.5, Mg2+ 0.49, Ca2+ 1.8, HCO3− 25, HPO4− 0.4, and glucose 20). The perfusate was gassed with 95% O2-5% CO2 and its temperature maintained at 37°C. All hearts then were loaded for 20 minutes with 5F-BAPTA by addition of 225 mL of 5 μmol/L 5F-BAPTA-AM to the perfusate (Molecular Probes).3 Washed bovine erythrocytes then were added to the perfusate to achieve a hematocrit of 25%.4 The hearts were perfused with the blood perfusate for 30 minutes before inducing no-flow ischemia by cross-clamping inflow to the aorta. Right ventricular pacing at 3 Hz was carried out throughout the entire protocol using stimuli of 2-ms duration and twice diastolic threshold strength.
For the first series of experiments, we determined the changes in [K+]e and pHe with the use of K+- and pH-sensitive electrodes constructed and calibrated as described previously.5 Data were accepted only from electrodes that demonstrated a 57- to 63-mV shift per tenfold change in K+ or pH activity. Two pairs of K+ and pH electrodes were placed in the mid myocardium in the left ventricular wall.
Millivolt readings from the K+- and pH-sensitive electrodes were amplified and filtered (1 kHz) by a high-impedance differential amplifier. The signals were sampled every 30 seconds and processed as described previously.5
pHi, PCr, and ATP were determined in the second series of experiments and [Ca2+]i in the third. In both, NMR spectroscopy with a Nicolet wide-bore NT-360 NMR spectrometer (Nicolet Magnetics Co) was used. For the determinations of PCr, ATP, and pHi, a 20-mm, 31P broad-band NMR probe (Nicolet Magnetics Co) tuned to 146.1 MHz was used. 31P data were time-averaged over 5-minute intervals throughout the protocol. pHi was determined from the chemical shifts of Pi and PCr peaks in the 31P NMR spectrum.6 ATP and PCr were determined from the area under their respective resonance peaks.2
For the measurements of [Ca2+]i, a 20-mm 19F NMR probe (Doty Scientific) tuned to 339.6 MHz was used as described previously.2 3 7 8 [Ca2+]i was determined from the relative areas of the peaks for the calcium-bound form of 5F-BAPTA compared with peak for the free 5F-BAPTA using a kD of 700 mmol/L.3 7 9 These spectra were also time-averaged over 5-minute intervals.2 9 10
In the fourth series of experiments, rt was determined by use of the four-electrode technique described by Oosterom et al,11 Plonsey and Barr,12 and Ellenby et al.13 The outer two Teflon-coated Ag Cl electrodes, 0.006 to 0.007 mm in diameter, were separated by 10 mm. The inner two electrodes were each placed 2.5 mm from the outer electrodes and were separated by 5 mm. Constant-current subthreshold pulses of 20-ms duration were delivered between the outer electrodes at end diastole (WP1A365 stimulator). The voltage drop between the inner two electrodes resulting from the current injection was measured with high-gain differential amplifiers, digitized at 20 KHz and stored on a Macintosh IIX computer with Superscope software (GW Instruments). Measurements were taken each minute throughout the experiment and then were time-averaged over 5-minute intervals for comparison with the NMR measurements. The relative change in rt (Δrt) during ischemia was determined by the relative change in the voltage drop induced by the current pulse Δrt(%)=(ΔVisI−ΔVc/Ic)·100/ΔVc/Ic, where is is ischemia and c is control.
All results are expressed as mean±SEM. Differences between means were evaluated with the use of one-way ANOVA or Student's t test with the Bonferroni correction when appropriate. A value of P≤.05 was considered statistically significant.
The results of the four series of experiments are shown in the Figure.⇓ Changes in [K+]e, pHe, PCr, ATP, pHi, [Ca2+]i, and rt are illustrated; first and second phases of the rise in [K+]e are also shown.14
[K+]e increased from 4.9±0.24 to 10.8±1.3 mmol/L in the first 8 minutes of ischemia. During the second or plateau phase, which lasted from 8 to 23 minutes of ischemia, [K+]e rose to 11.7±2.2 mmol/L. During the third phase, beginning 23 minutes after the onset of ischemia, [K+]e rose to 22±9.2 mmol/L. pHe fell from 7.4±0.1 to 6.98±1.0 during the first 8 minutes of ischemia. Thereafter, pHe fell gradually and eventually plateaued at 6.45±0.6 ≈30 minutes after the onset of no-flow ischemia.
PCr fell to 10% of its original value between 5 and 10 minutes after the onset of acute ischemia and was undetectable after 20 minutes. There was no significant change in the time-averaged value of ATP within the first 5 minutes of ischemia. Thereafter, ATP levels fell steadily, reaching 25% of its control value between 15 and 20 minutes after the onset of ischemia. ATP was undetectable after 25 minutes. Time-averaged pHi did not change significantly during the first 5 minutes of ischemia, but thereafter its fall paralleled the decline in ATP. pHi fell to approximately 6 between 15 and 20 minutes of ischemia.
During the first 5 minutes of ischemia, the time-averaged value for [Ca2+]i increased from a control of 1031±110 to 1305±159 mmol/L. This increase was statistically significant (P<.05). [Ca2+]i then continued to increase, reaching a level close to 3000 mmol/L, a twofold increase over the control. After 25 minutes, only a calcium 5F-BAPTA resonance could be resolved. Thus, a further increase in [Ca2+]i could not be measured, and the plateau shown in the Figure after 25 minutes may be an artifact.
rt increased by 28±7% within the first 5 minutes of ischemia and then did not change until after 23 minutes. It then increased in a linear fashion, reaching levels 100% greater than control within 35 minutes.
Thus, the first (rapidly rising) phase of the increase in [K+]e is characterized by a 90% fall in PCr, a 25% decrease in ATP, and a decrease in pHi and pHe of ≈0.5 U. [Ca2+]i rises slightly but significantly, and there is a slight increase in rt. pHi, pHe, and ATP continue to fall during the plateau phase of the [K+]e change. [Ca2+]i continues to rise during this phase, but there is no significant increase in rt. The increase in rt occurs in association with the third phase in the rise in [K+]e beginning 23 minutes after the onset of ischemia, after ATP and PCr have been depleted. At this time, pHe changes only slightly. The rise in rt occurs only after [Ca2+]i has risen to three times its control levels and pHi has fallen to <6.0.
The purpose of our experiments was to determine the relationships between the metabolic and ionic changes occurring during acute no-flow ischemia and to correlate these changes with the simultaneously occurring changes in rt with the use of a common preparation, the isolated blood-perfused rabbit heart loaded with 5F-BAPTA, and a common protocol, 30 to 35 minutes of no-flow ischemia. Our study is unique for the following reasons: (1) It provides a precise description of the relationship between the changes in [K+]e and pHe and the changes in high-energy phosphates and pHi during acute ischemia. (2) It establishes the usefulness of the changes in [K+]e as a surrogate for the concurrent ionic and electrical events during ischemia. (3) It establishes the changes in pHi and [Ca2+]i that occur before the onset of cell-to-cell electrical uncoupling during acute ischemia.
The relationship between the first phase of the rise in [K+]e and the initial changes in pHe and pHi supports the anion-linked hypothesis of K+ efflux, which postulates that the increased efflux is related to the transmembrane movements of anions generated intracellularly during ischemia.15 However, this does not rule out a contribution of other mechanisms such as an increase in K+ conductance through the ATP-sensitive potassium channel or through voltage-gated potassium channels.16 17 18
Intracellular acidification is thought to result from ATP hydrolysis, glycolysis, lipolysis, and the triglyceride synthesis/degradation cycle.19 The parallel change in ATP and pHi that we observed is consistent with this explanation and with other studies that used NMR spectroscopy.6 20 Our findings contrast with experiments in which pHi was measured by ion-selective microelectrodes in the blood-perfused rabbit papillary muscle.21 22 In the study of Yan and Kle´ber,21 pHi changed less during the first 20 minutes of no-flow ischemia than we have recorded and less than the change in pHe. However, in these experiments extracellular CO2 was clamped at 35 mm Hg. As extracellular PCO2 was increased, the change in pHi became more pronounced and approached the values that we obtained. In our experiments, PCO2 was not clamped and therefore probably rose to levels of 300 mm Hg, as originally described.23 The differences also may relate to the greater accumulation of metabolically derived CO2 in the whole heart than in isolated papillary muscles of ≤2 mm in diameter.
We used 5F-BAPTA and 19F NMR spectroscopy to determine [Ca2+]i. This method has been used by other laboratories8 as well as our own.2 10 It is important to stress that the hearts were loaded with 5F-BAPTA before the introduction of the erythrocytes. Thus, the calcium present within the erythrocytes did not contribute to the calcium signal obtained before or during acute ischemia. The level of resting [Ca2+]i of >1000 nmol/L that we observed is higher than the time-averaged values reported in the ferret8 or in the rat.10 Some of this difference is due to the use of a lower KD for 5F-BAPTA binding to calcium. Given the propensity of acetoxymethylester-loaded indicators to distribute into intracellular organelles, we considered the possibility that this high basal [Ca2+]i value might be due to 5F-BAPTA loading into intracellular organelles (such as the sarcoplasmic reticulum). We tested this hypothesis by using a newly developed fast exchange indicator, TF-BAPTA, which gives separate resonances for different calcium compartments.24 In preliminary experiments, rabbit hearts loaded with TF-BAPTA-AM showed three fluorine resonances, a calcium-insensitive fluorine resonance, and two calcium-sensitive resonances. This suggests that the TF-BAPTA was distributed into two compartments: one compartment with an ionized calcium level in the 500 nmol/L range, which we believe represents cytosolic calcium, and a second compartment with an ionized calcium level >100 μmol/L, which we believe represents calcium in intracellular organelles. In the preliminary experiments with TF-BAPTA, 20 minutes of global ischemia resulted in a doubling of cytosolic calcium, whereas there was no significant change in the putative intracellular organelle resonance.
We used the four-electrode method to determine changes in rt induced by ischemia.11 12 13 The early rise in rt that we observed is similar to that reported by Cascio et al4 and Kle´ber et al25 and most likely reflects collapse of the microvasculature and diminution of the interstitial space. rt did not change for 20 to 25 minutes. It then increased in concert with the third phase of the potassium rise. This is similar to the results obtained in the papillary muscle4 25 and corresponds to an increase in internal longitudinal resistance, indicating cell-to-cell uncoupling.
Changes in [Ca2+]i and [pH]i are considered of primary importance in regulation of cell-to-cell electrical coupling.26 27 28 We observed a significant rise in [Ca2+]i and a significant fall in pHi before the rise in rt. This result suggests that [Ca2+]i and pHi may need to change to a threshold level before cell-to-cell uncoupling occurs. This level appears to be a fall in pHi to approximately 6.0 and at least a doubling of [Ca2+]i. A synergistic effect of calcium and protons on gap junctional conductance has been described previously.28 29 Our results are consistent with this observation and suggest a pH-dependent effect of [Ca2+]i on cell-to-cell coupling. It also appears that the third phase of the rise in [K+]e occurs in concert with and identifies the achievement of these intracellular changes.
Our study is the first to record metabolic, ionic, and electrical events in the same experimental model under the same experimental conditions and indicates that changes in [K+]e can be used as a surrogate for the concurrent ionic and electrical events. The fact that our data both confirm and contradict findings made in different experimental models underscores the importance of the use of a common and consistent model while attempting to formulate a single picture of the ionic, metabolic, and electrical events that occur during acute ischemia.
This research was supported in part by a Howard Hughes Medical Institute Research Training Fellowship for Medical Students; the Program Project Grant on Sudden Cardiac Death, 5-PO1-HL-27430; and a Merit Award, 5-R-37-HL-38885, from the NHLBI, Bethesda, Md. The authors gratefully acknowledge Prof Andre´ Kle´ber of Bern, Switzerland, for suggestions made during the planning stages of our study and Anne Middleton for secretarial support.
Selected Abbreviations and Acronyms
|[K+]e||=||extracellular K+ concentration|
|[Ca2+]i||=||intracellular Ca2+ concentration|
|NMR||=||nuclear magnetic resonance|
|rt||=||whole tissue resistance|
↵1 Members of the Experimental Cardiology Group include W. Cascio, C. Engel, L. Gettes, T. Griggs, T. Johnson, D. Martin, G. Meissner, T. Nichols, R. Rosenberg, W. Sun, and H. Yang.
- Received February 5, 1996.
- Revision received April 22, 1996.
- Accepted May 1, 1996.
- Copyright © 1996 by American Heart Association
Gettes LS, Cascio WE. Effect of acute ischemia on cardiac electrophysiology. In: Fozzard HA, ed. The Heart and Cardiovascular System: Scientific Foundations. 2nd ed. New York, NY: Raven Press, Ltd; 1992.
Steenbergen C, Murphy E, Levy LA, London RE. Elevation in cytosolic free calcium concentration early in myocardial ischemia in perfused rat heart. Circ Res.. 1987;60:700-707.
Levy LA, Murphy E, London RE. Synthesis and characterization of 19F NMR chelators for measurement of cytosolic free calcium. Am J Physiol.. 1987;252:C441-C449.
Cascio WE, Yan G-X, Kle´ber AG. Passive electrical properties, mechanical activity, and extracellular potassium in arterially perfused and ischemic rabbit ventricular muscle. Circ Res.. 1990;66:1461-1473.
Johnson TA, Engle CL, Kusy RP, Knisley SB, Graebner CA, Gettes LS. Fabrication, evaluation, and use of extracellular K+ and H+ ion-selective electrodes. Am J Physiol.. 1990;258:H1224-H1231.
Jacobus WE, Pores GJ, Nunnaly RL, Hollis DP. Tight coupling of intracellular pH and ventricular performance. J Moll Cell Cardiol. 1978;10(suppl1):39.
Smith GA, Hesketh RT, Metcalfe JC, Feeney J, Morris PG. Intracellular calcium measurements by 19F NMR of fluorine labeled chelators. Proc Natl Acad Sci U S A.. 1983;80:7178-7182.
Marban E, Kitakaze M, Koretsune Y, Yue DT, Chacko VP, Pike MM. Quantification of [Ca2+]i in perfused hearts: critical evaluation of the 5F-BAPTA and nuclear magnetic resonance method as applied to the study of ischemia and reperfusion. Circ Res.. 1990;66:1255-1267.
Marban E, Kitakaze M, Kusuoka H, Porterfield JK, Yue DT, Chacko VP. Intracellular free calcium concentration measured with 19F NMR spectroscopy in intact ferret hearts. Proc Natl Acad Sci U S A.. 1987;84:6006-6009.
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;66:135-146.
Hill JL, Gettes LS. Effect of acute coronary artery occlusion on local myocardial extracellular K+ activity in swine. Circulation.. 1980;61:768-778.
Weiss WN, Lamp WT, Shine KT. Cellular K+ loss and anion efflux during myocardial ischemia and metabolic inhibition. Am J Physiol.. 1989;256:H1165-H1175.
Yan GX, Kle´ber AG. Changes in extracellular and intracellular pH in ischemic rabbit papillary muscle. Circ Res.. 1992;71:460-470.
VanHeel B, Leybaert L, DeHemptinne A, Leusen I. Simulated ischemia and intracellular pH in isolated ventricular muscle. Am J Physiol.. 1989;257:365-376.
Case RB, Felix A, Castellana FS. Rate of rise of myocardial PCO2 during early myocardial ischemia in the dog. Circ Res.. 1979;45:324-330.
Murphy E, Steenbergen C, Levy LA, Gabel S, London RE. Measurement of cytosolic free calcium in perfused rat heart using TF-BAPTA. Am J Physiol. 1994;266(Cell Physiol 35):C1323-C1329.
Kle´ber AG, Riegger CB, Janse MJ. Electrical uncoupling and increase of extracellular resistance after induction of ischemia in isolated, arterially perfused rabbit papillary muscle. Circ Res.. 1987;66:271-279.
Burt J. Block of intercellular communication: interaction of intracellular H+ and Ca2+. Am J Physiol.. 1987;253:C607-C612.
White RL, Doeller JE, Verselis VK, Wittenberg BA. Gap junctional conductance between pairs of ventricular myocytes is modulated synergistically by H+ and Ca++. J Gen Physiol.. 1990;95:1061-1075.