This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kupriyanov, V. V. Articles by Deslauriers, R. Search for Related Content PubMed PubMed Citation Articles by Kupriyanov, V. V. Articles by Deslauriers, R.

(Circulation Research. 1995;76:839-851.)
© 1995 American Heart Association, Inc.

Articles

Pathways of Rb⁺ Influx and Their Relation to Intracellular [Na⁺] in the Perfused Rat Heart

A ⁸⁷Rb and ²³Na NMR Study

Valerie V. Kupriyanov, Laura C. Stewart, Bo Xiang, June Kwak, Roxanne Deslauriers

From the Institute for Biodiagnostics, National Research Council, Winnipeg, Canada. National Research Council publication NRC 34759.

Correspondence to Dr Valerie V. Kupriyanov, Institute for Biodiagnostics, National Research Council of Canada, 435 Ellice Ave, Winnipeg, Manitoba, Canada R3B 1Y6. E-mail kupriyanov @ibd.lan.nrc.ca.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract The aims of this study were to characterize the routes of influx of the K⁺ congener, Rb⁺, into cardiac cells in the perfused rat heart and to evaluate their links to the intracellular Na⁺ concentration ([Na⁺]_i) using ⁸⁷Rb and ²³Na nuclear magnetic resonance (NMR) spectroscopy. The rate constant for Rb⁺ equilibration in the extracellular space was 8.5 times higher than that for the intracellular space. The sensitivity of the rate of Rb⁺ accumulation in the intracellular space of the perfused rat heart to the inhibitors of the K⁺ and Na⁺ transport systems has been analyzed. The Rb⁺ influx rates were measured in both beating and arrested hearts: both procaine (5 mmol/L) and lidocaine (1 mmol/L) halved the Rb⁺ influx rate. In procaine-arrested hearts, the Na⁺,K⁺-ATPase inhibitor ouabain (0.6 mmol/L) decreased Rb⁺ influx by 76±24% relative to that observed in untreated but arrested hearts. Rb⁺ uptake was insensitive to the K⁺ channel blocker 4-aminopyridine (1 mmol/L). The inhibitor of Na⁺/K⁺/2 Cl^- cotransport bumetanide (30 µmol/L) decreased Rb⁺ uptake only slightly (by 9±8%). Rb⁺ uptake was dependent on [Na⁺]_i: it increased by 58±34% when [Na⁺]_i was increased with the Na⁺ ionophore monensin (1 µmol/L) and decreased by 48±9% when [Na⁺]_i was decreased by the Na⁺ channel blockers procaine and lidocaine. Dimethylamiloride (15 to 20 µmol/L), an inhibitor of the Na⁺/H⁺ exchanger, slightly reduced [Na⁺]_i and Rb⁺ entry into the cardiomyocytes (by 15±5%). ³¹P NMR spectroscopy was used to monitor the energetic state and intracellular pH (pH_i) in a parallel series of hearts. Treatment of the hearts with lidocaine, 4-aminopyridine, dimethylamiloride, or bumetanide for 15 to 20 minutes at the same concentrations as used for the Rb⁺ and Na⁺ experiments did not markedly affect the levels of the phosphate metabolites or pH_i. These data show that under normal physiological conditions, Rb⁺ influx occurs mainly through Na⁺,K⁺-ATPase; the contribution of the Na⁺/K⁺/2 Cl^- cotransporter and K⁺ channels to Rb⁺ influx is small. The correlation between Rb⁺ influx and [Na⁺]_i during infusion of drugs that affect [Na⁺]_i indicates that, in rat hearts at 37°C, Rb⁺ influx can serve as a measure of Na⁺ influx. We estimate that, at normothermia, at least 50% of the Na⁺ entry into beating cardiac cells is provided by the Na⁺ channels, with only minor contributions (<15%) from the Na⁺/K⁺/2 Cl^- cotransporter and the Na⁺/H⁺ exchanger.

Key Words: Rb⁺ uptake • intracellular Na⁺ • rat heart • Na⁺,K⁺-ATPase • Na⁺ and K⁺ transport inhibitors

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Under steady state conditions, K⁺ and Na⁺ fluxes across the sarcolemma of cardiomyocytes have been assessed by tracer techniques that use radioactive isotopes of these ions, ⁴²K and ²²Na.^{1 2 3} For measurements of K⁺ fluxes, Rb⁺ and its radioactive isotope, ⁸⁶Rb⁺, have been used^{4 5 6} because rubidium is a well-established congener for potassium.^{7 8 9 10} The rates of influx and efflux of Rb⁺ have been measured in living tissues and isolated cells by the radioisotope technique, atomic absorption spectrometry, and nuclear magnetic resonance (NMR) spectroscopy.^{4 5 6 7 11 12} The relatively high natural abundance (28%), low biological abundance, and high NMR sensitivity of ⁸⁷Rb make it a good tracer for K⁺ influx and efflux studies by NMR.^{11 12} Rb⁺ uptake data can be used to approximate K⁺ influx if there are no significant differences in ion selectivity of the different K⁺-transporting systems. In addition to the Na⁺ pump, K⁺ and Rb⁺ are transported through the Na⁺/K⁺/2 Cl^- cotransporter, which has been discovered and characterized in isolated chicken, rabbit, and neonatal rat cardiomyocytes.^{13 14 15} However, whether this cotransporter is operative and its relative contribution to K⁺/Rb⁺ uptake in the adult rat heart remain unclear. K⁺ channels may contribute to the influx of Rb⁺ because of its inwardly directed electrochemical gradient at the beginning of Rb⁺ uptake experiments in cardiomyocytes.

It is known that the kinetic properties of K⁺ and Rb⁺ with respect to Na⁺,K⁺-ATPase are similar for cardiac and skeletal muscles, nerves, red cells, and other tissues.^{8 9 10 16 17 18} Direct comparison of ⁸⁶Rb and ⁴²K uptake in human erythrocytes⁷ revealed identical uptake rates. Similarly, ⁸⁶Rb and ⁴²K fluxes in the cardiomyocytes are comparable (12 to 20 µmol · min^-1 · g protein^-1; see References 1, 3, 5, and 6^{1 3 5 6} ). This implies that at least the ratios of the maximal rates to the Michaelis constant (V_max/K_m) for the carriers responsible for K⁺ influx (Na⁺,K⁺-ATPase, Na⁺/K⁺/2 Cl^- cotransporter, K⁺ channels) are also identical in these cells (because [K⁺]>>[Rb⁺] in radioisotope tracer studies). In addition, Rb⁺ quantitatively replaces K⁺ in electrophysiological determinations of Na⁺,K⁺-ATPase activity in cardiac tissue,^{9 10} and the equilibrium transmembrane Rb⁺ gradient is similar to that of K⁺.^{11 12} The latter implies similar ion selectivity of the systems transferring these ions inside (mainly Na⁺,K⁺-ATPase + Na⁺/K⁺/2 Cl^- cotransporter + K⁺ channels) and outside (K⁺ channels + Na⁺/K⁺/2 Cl^- cotransporter) the cells. Less is known about the ion specificity of the Na⁺/K⁺/2 Cl^- symporter in cardiac muscle. However, data available for other mammalian cells show that Rb⁺ is quantitatively similar to K⁺ in its interactions with the Na⁺/K⁺/2 Cl^- cotransporter (for review, see References 19 through 21^{19 20 21} ). Therefore, it is likely that this symporter has analogous ion selectivity in cardiomyocytes, similar to that of the Na⁺,K⁺-ATPase from various mammalian cells (see above). This appears to be true for cardiac K⁺ channels as well.⁴

Under steady state conditions, the fraction of Rb⁺ influx mediated by Na⁺,K⁺-ATPase is tightly coupled to Na⁺ entry and therefore should reflect the Na⁺ influx rate rather than Na⁺/K⁺ pump activity per se, if the contribution of Na⁺/Ca²⁺ exchange to Na⁺ efflux is small.¹⁰ The latter seems to be true in the normal beating heart, in which the average membrane potential is highly negative and Na⁺ entry, rather than efflux, occurs through this exchanger. Furthermore, when the Na⁺ pump is moderately inhibited (decrease in maximal rate), the concomitant increase in [Na⁺]_i is sufficient to compensate for the reduced number of active pumps and maintain Na⁺ efflux and its coupled K⁺ influx through the pump constant.¹⁰ This is a consequence of the relative independence of passive Na⁺ influx from [Na⁺]_i, because the concentration required for reversal of Na⁺ flux is at least 100 times higher than the steady state [Na⁺]_i under normal conditions. However, the arguments presented above are theoretical and require experimental support.

The aims of this study were to characterize the routes of influx of the K⁺ congener Rb⁺ into cardiac cells in the perfused rat heart and to evaluate their links to the intracellular Na⁺ concentration ([Na⁺]_i) using ⁸⁷Rb and ²³Na NMR spectroscopy. We analyzed the sensitivity of the rate of Rb⁺ accumulation in the perfused rat heart to inhibitors of the K⁺ and Na⁺ transporting systems and found that, in the steady state at 37°C, this rate may be used as an index of the Na⁺ entry rate, which occurs mostly through Na⁺ channels. The contributions of the Na⁺/H⁺ exchanger and Na⁺/K⁺/2 Cl^-cotransporter are each approximately 10% to 15%.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Solutions
For the ⁸⁷Rb and ³¹P NMR experiments, the hearts were perfused with phosphate-free Krebs-Henseleit (KH) buffer containing (in mmol/L) NaCl 118, NaHCO₃ 25, KCl 4.7, CaCl₂ 1.75 (1.1 mmol/L free Ca²⁺), MgSO₄ 1.2, EDTA 0.5, and pyruvate 5 or glucose 11. The perfusate was equilibrated with 95% O₂/5% CO₂, with the pH maintained at 7.4. The KH buffer used for Rb⁺ loading contained [Rb⁺] 0.94 mmol/L and [K⁺] 3.76 mmol/L instead of 4.7 mmol/L K⁺. The Rb⁺-free flushing solution (solution 1) contained (in mmol/L) NaCl 143, KCl 4.7, MgSO₄ 1.2, and CaCl₂ 1.75.

For the ²³Na NMR experiments, a 5 mmol/L complex of dysprosium with triethylenetetraminehexaacetic acid (TTHA), Na₃Dy(TTHA) · 3NaCl, was added to the KH buffer, and [NaCl] was reduced to 88 mmol/L, maintaining the total [Na⁺] at 144 mmol/L. CaCl₂ was increased to 2.75 mmol/L to compensate for the Ca²⁺ binding effect of EDTA and Dy(TTHA)^3-; free [Ca²⁺] was 1.1 mmol/L. The flushing solution (solution 2) contained (in mmol/L) choline chloride (ChCl) 140, KCl 4.7, CaCl₂ 1.75, MgSO₄ 1.2, and Tris₃Dy(TTHA) · 3(Tris · HCl) 5. Stock solutions of Na₃Dy(TTHA) · 3NaCl (0.1 mol/L) and Tris₃Dy(TTHA) · 3 (Tris · HCl) (0.1 mol/L) were prepared as described previously.²²

Heart Perfusion
Male Sprague-Dawley rats (360 to 460 g, n=102) were obtained from the NRC animal facility. The rats were anesthetized with sodium pentobarbital (120 mg/kg IP), and the hearts were quickly removed and perfused retrogradely via the aorta at a constant pressure of 80 mm Hg or at constant flow of 10 to 12 mL/g wet wt. After placement of a left ventricular apical drain, a water-filled balloon was inserted through the mitral valve and secured in the left ventricle. The balloon was connected to a Statham P23Db pressure transducer and a Gould two-channel physiological recorder, which allowed for monitoring of the heart rate and left ventricular pressure. The end-diastolic pressure (EDP) was set to approximately 5 mm Hg. The pressure-rate product (PRP), the product of developed pressure (systolic minus diastolic pressures) and heart rate, was used as an index of mechanical work. The coronary flow rate was monitored with an ultrasonic blood-flow meter (Transonic Systems Inc), and perfusion pressure was measured continuously through the catheter connecting aortic line and pressure transducer. PRP and coronary flow were measured simultaneously with acquisition of NMR spectra. Typical functional parameters during control perfusion were as follows: left ventricular developed pressure, 90 to 120 mm Hg; EDP, 0 to 5 mm Hg; heart rate, 200 to 300 beats per minute; and coronary flow, 16 to 22 mL/min (10 to 12 mL/g wet wt).

Measurements of Rb⁺ Uptake
For measurements of Rb⁺ accumulation, perfusion was switched to a Rb⁺-containing solution for 10 to 15 minutes (to obtain control values) and then switched back to normal KH buffer for 20 minutes before any other interventions. Then drug infusion was started (2.5 to 5 minutes), and the same loading/washout cycle was repeated (see protocols below). To reduce contributions to the Rb⁺ signal from the bath, either the hearts were perifused with Rb⁺-free flushing solution 1 throughout Rb⁺ loading and washout or the perifusion line was used as a suction line placed at the bottom of the NMR tube to remove fluid around the heart ("dry" heart).

Estimation of Extracellular Compartment
The amount of Rb⁺ in the "extracellular" space (interstitium + vasculature + cannula + aorta + right ventricle + sample tube) was estimated for both dry and submerged hearts. Hearts perfused in the dry mode (n=5) were equilibrated with Rb⁺ in buffer containing 0.94 mmol/L RbCl and 3.76 mmol/L KCl for 30 minutes. The buffer KCl and RbCl concentrations were then increased fivefold (23.5 mmol/L total concentration) over a period of 10 minutes by infusion of a mixture of 0.8 mol/L KCl and 0.2 mol/L RbCl. This infusion was then terminated, and after return of Rb⁺ to its preinfusion level (10 minutes), washout was started. This approach allowed measurement of the rates of Rb⁺ influx and efflux from intracellular and extracellular compartments in the same heart and the contribution of extracellular Rb⁺ to the total NMR signal. To assess the possible effects of increased osmolarity, ionic strength, and [Cl^-], hearts were infused with ChCl (n=3) in place of KCl+RbCl to reach a concentration of 18.8 mmol/L (4.7x4) in the perfusate (see Fig 2, left).

View larger version (12K): [in this window] [in a new window]   Figure 2. Left, Graph showing kinetics of Rb+ equilibration in extracellular space. Hearts (n=5) perfused in a "dry" mode were equilibrated with Rb+ for 30 minutes under standard conditions (0.94 mmol/L RbCl+3.76 mmol/L KCl). The [K+]+[Rb+] in the perfusate was then increased fivefold for 10 minutes and then returned to normal level. In control experiments (n=3), instead of K++Rb+ mixture, an equal amount of choline chloride (ChCl) was infused to bring its concentration to 18.8 mmol/L. The beginning and end of the infusion are indicated by arrows. Right, Graph showing relation between total Rb+ and Na+ contents and perfusion pressure in perfused rat hearts. Hearts (n=3) perfused in a "dry" mode were equilibrated with Rb+ for 30 minutes under standard conditions (0.94 mmol/L RbCl+3.76 mmol/L KCl). The perfusion pressure was then altered by changing the coronary flow, and 87Rb and 23Na NMR spectra were acquired as described in "Materials and Methods" and "Results." Additional 87Rb nuclear magnetic resonance data were also taken from experiments (n=3) in which hearts equilibrated with Rb+ and tested with ChCl were subsequently subjected to changes in perfusion pressure as described in "Materials and Methods" and "Results." The data are expressed as percentages of the values measured for each heart at minimal perfusion pressures (25 to 50 mm Hg). The correlations obtained by the least-squares method are described by the following equations: sodium, y=83.2+0.33x (R2=.82) and rubidium, y=98.5+0.035x (R2=.016).

To estimate the possible effects of changes in vascular + interstitial volume on the Rb⁺ signal, two series of the experiments were performed. First, after ChCl washout (10 minutes), the perfusion pressure was decreased from 85±5 to 32±6.3 mm Hg by decreasing the coronary flow from 17.7±1.2 to 5.8±0.3 mL/min for 10 minutes. The coronary flow and perfusion pressure were subsequently returned to normal. Second, after 30 minutes of Rb⁺ equilibration, the perfusion pressure was increased from 83±6 to 123±12 mm Hg, then decreased to 48±3 mm Hg, and returned to the initial level (83 mm Hg) by increasing and decreasing coronary flow, respectively, in 3 hearts perfused in the dry mode (see Fig 2, right). During each step (6 minutes), two ²³Na (no shift reagent) and two ⁸⁷Rb NMR spectra were acquired in the following sequence: Na (16 seconds), Rb (2 minutes), Rb (2 minutes), and Na (16 seconds). The probe was manually switched from one frequency to the other without retuning because the Rb and Na channels are independent. The total Na⁺ signal is a marker of extracellular space, which is proportional to the extracellular volume because in the normal heart, 95% of the signal is of extracellular origin.²³ A similar approach has been used for evaluating the stability of the extracellular volume during cycles of Rb⁺ loading and washout. Five hearts were subjected to two consecutive cycles of Rb⁺ loading and washout (30 minutes each), and in 3 hearts, three ²³Na NMR spectra were interleaved with ⁸⁷Rb NMR spectra: before the first (0 minutes) and second (30 minutes) cycles and at the end of the second cycle (60 minutes).

In addition, switching the hearts equilibrated with Rb⁺ from the dry to the submerged (with flushing from outside with Rb⁺-free solution) mode did not markedly affect the observed Rb⁺ level (<5%), demonstrating the equivalence of the two techniques.

Measurements of Intracellular Na⁺
For the shift reagent–aided ²³Na NMR experiments, the hearts were superfused with sodium-free solution to minimize the signal arising from the bath.²³ The Dy(TTHA)^3--containing buffer and flushing solution did not significantly affect the measured functional parameters of the heart, which remained stable for at least 2 hours.

Experimental Protocols
All hearts were perfused with basic KH buffer for 15 to 20 minutes before the spectral acquisition was begun and then perfused according to the following protocols.

Protocol 1 (General)

View this table: [in this window] [in a new window]   Table 5.

Drugs 1 and 2 are 1 mmol/L 4-aminopyridine (4-AP), 1 mmol/L lidocaine (Lido), 30 µmol/L bumetanide (Bum), 15 to 20 µmol/L dimethylamiloride (DMA), and 1 µmol/L monensin (Mon), which were infused 2.5 to 4 minutes before Rb⁺ loading and washed out simultaneously with Rb⁺ washout. 4-AP, Lido, and DMA were infused as aqueous solutions (pH 7.4) and Bum and Mon as DMSO or ethanol solutions. The concentrations of the organic solvents in the perfusate did not exceed 0.2% (vol/vol). The reliability of repetitive measurements of Rb⁺ influx was tested in the separate control experiments (n=5) in which the rates of Rb⁺ uptake and total Na⁺ content were measured and compared during two consecutive cycles of Rb⁺ loading and washout (see above).

Protocol 2 (Procaine, Ouabain)

View this table: [in this window] [in a new window]   Table 6.

Procaine (5 mmol/L) and ouabain (0.6 mmol/L) were added 5 minutes and 2 minutes, respectively, before Rb⁺ loading and washed out simultaneously with the last Rb⁺ washout. This protocol was used in ⁸⁷Rb, ²³Na, and ³¹P NMR experiments.

Protocol 3 (Lido-Arrested Hearts)

View this table: [in this window] [in a new window]   Table 7.

Lido (1 mmol/L) and Lido in combination with DMA (15 µmol/L) or Bum (30 µmol/L) were infused 2.5 minutes before Rb⁺ loading and washed out simultaneously with Rb⁺ washout. To minimize the possible effects of temporal metabolic instability as well as the effects of previous drug infusion, the sequence of DMA and Bum infusion was varied.

Protocol 4 (Effects of 4-AP, DMA, Bum, and Lido on Energy Metabolism and pH_i)

View this table: [in this window] [in a new window]   Table 8.

Normally the intracellular inorganic phosphate (P_i) peak is used to measure intracellular pH (pH_i), but this peak is frequently undetectable in pyruvate-perfused hearts. The hearts were therefore perfused briefly (15 minutes) with 4 mmol/L 2-deoxyglucose (DG). After washout of DG, a small but clearly detectable peak of intracellular 2-DG-6-phosphate was visible and was used as a pH_i probe. Later, three cycles of drug infusion and washout were repeated. The sequence of drug infusion varied in each experiment to minimize possible effects of any temporal metabolic instability and previous drug infusion. The infusions were performed as described in protocol 1.

NMR Spectroscopy
All NMR experiments were performed on a Bruker AM-360 spectrometer. A 20-mm broadband probe (Morris Instruments or Bruker) was used for data acquisition. The ²³Na signal from the heart and the surrounding bath was used for shimming. ⁸⁷Rb NMR spectra were acquired at 117.8 MHz with 1.0- to 2.5-minute time resolution (90° pulse, 60 µs; recycle time, 14 ms) with a Morris Instruments broadband probe. The sweep width was 18 kHz, and 512 data points were collected. A capillary containing 1 mol/L RbCl+5 mol/L KI (10 to 15 µmol Rb) was used as a concentration reference.¹¹ To minimize the contribution of bath Rb⁺ to the observed NMR signal, the hearts were flushed from outside with Rb⁺-free flushing solution 1 at flow rates about three times greater than the coronary flow. The ratio of coronary and flushing flows was kept constant throughout the Rb⁺ loading and washout cycle. Representative spectra are shown in Fig 1. Some experiments (n31) were carried out in dry hearts. After the shimming procedure, the perfusate was aspirated from the bottom of the NMR tube, maintaining a minimal amount of fluid around the heart. This did not lead to detectable broadening of the already broad ⁸⁷Rb⁺ resonance (400 Hz), changes in mechanical function, or changes in the Rb⁺ influx or efflux relative to hearts that were immersed and flushed from outside with Rb⁺-free solution.

View larger version (22K): [in this window] [in a new window]   Figure 1. Left, Representative 87Rb nuclear magnetic resonance spectra of perfused rat heart during Rb+ load and washout. Each spectrum represents sum of 7800 acquisitions collected during 2 minutes. Line broadening, 150 Hz. In this experiment, heart was fully immersed in the perfusate (bath). To reduce the signal from Rb+ in the bath, the heart was perfused from outside at a flow about three times that of the coronary flow. Other details are described in "Materials and Methods." Right, Graph showing kinetics of Rb+ accumulation and washout in perfused rat hearts. Means±SD for five hearts are given. Rb+ content was calculated as follows: the ratio of heart peak area to the reference peak area was multiplied by the Rb+ content in the reference and divided by dry heart weight. The first two points of the kinetic curve (0 to 2 and 2 to 4 minutes) were omitted because they correspond to equilibration of Rb+ in the extracellular compartments. Extrapolation of the initial linear part of the kinetic curve to "zero" time gives the approximate amount of the Rb+ in the extracellular space and in the bath surrounding heart (3 µmol/g dry wt). Rb+ loading and washout was done as described in "Materials and Methods."

²³Na NMR spectra in the presence of shift reagent were acquired at 95.2 MHz with 2- or 4-minute resolution (90° pulse, 36 µs; recycle time, 0.3 seconds) with a Bruker broadband probe. The sweep width was 4 kHz, and 4096 data points were collected. Representative spectra (showing the upfield region only) are shown in Fig 6. The extracellular Na⁺ peak was minimized by flushing around the heart with Na⁺-free flushing solution 2 as described above and reported previously.^{22 23} Sodium NMR spectra of the total Na⁺ were the sum of 64 acquisitions accumulated by applying similar parameters and using a Morris Instruments broadband probe.

View larger version (13K): [in this window] [in a new window]   Figure 6. Changes in the 23Na spectra of perfused rat hearts caused by procaine, ouabain, and monensin. The experiments with procaine and procaine + ouabain were performed according to protocol 2, and those with monensin according to basic protocol 1 (see "Materials and Methods" and legends to Figs 3 and 4). Spectra represent the changes in intracellular Na+ peak intensity induced by 10 to 15 minutes of exposure to 5 mmol/L procaine, 0.6 mmol/L ouabain in the presence of procaine, or 1 µmol/L monensin alone. The spectra represent sum of 480 acquisitions collected during 4 minutes. Gaussian multiplication was used (line broadening=-30 Hz, GB parameter=0.15) to improve resolution of the intracellular and extracellular Na+ peaks (0.5 and 2.5 ppm, respectively). The chemical shifts are given relative to that of the Na+ peak (not shown) measured before perfusion with the shift reagent.

³¹P NMR spectra were acquired at 145.8 MHz using a Bruker broadband probe. ³¹P NMR spectra were acquired with a 4-minute time resolution (60° pulse, 24 µs; recycle time, 2.3 seconds). The sweep width was 10 kHz, and 4096 data points were collected.

NMR Data Analysis
⁸⁷Rb NMR data were processed by exponential multiplication with line broadening (LB) of 150 to 200 Hz to increase sensitivity. After automatic baseline correction, the spectra were integrated and the ratio of the integral of "heart" Rb⁺ to that of the reference was multiplied by the amount of Rb⁺ in the reference and divided by the heart weight to calculate the total Rb⁺ content of the heart. The total Rb⁺ content is equal to the sum of the Rb⁺ in the intracellular and extracellular spaces (interstitium + vasculature + cannula and aorta + right ventricle + sample tube, respectively).

²³Na NMR data obtained in the presence of the shift reagent were processed by gaussian multiplication (typically LB, -30 Hz; GB parameter, 0.15 to 0.25) to improve resolution of the extracellular and intracellular Na⁺ resonances. The relative intensities of the intracellular Na⁺ signals were used to estimate changes in intracellular Na⁺ content assuming no change in line width or Na⁺ visibility throughout the experimental protocols. ²³Na NMR data obtained in the absence of the shift reagent were exponentially multiplied (LB, 10 Hz) before Fourier transformation. After automatic baseline correction, the spectra were integrated by the Bruker integration subroutine.

³¹P NMR data were processed by exponential multiplication for sensitivity enhancement (LB, 20 Hz). The relative contents of phosphorus-containing compounds were calculated from either peak intensities or their integrals. ³¹P NMR resonance areas were calculated with the NMR1 program (NMRi) on a SUN 3 data station. Total NMR visible phosphorus was calculated as the sum of the integrals of the peaks of phosphocreatine (PCr), P_i, ATP (++ß), NAD⁺, and phosphomonoesters (PME) and referred to the initial level (see Table 2). The integral of the PCr peak was corrected for saturation with a saturation factor of 1.2.²² The integral of the PME peak was not corrected for saturation, because in our experiments its contribution to the total NMR visible phosphorus integral did not exceed 10%.

View this table: [in this window] [in a new window]   Table 2. Effects of Procaine and Ouabain on Functional and Metabolic Parameters of Perfused Rat Hearts During Rb+ Transport Experiments

Estimation of pH_i
pH_i was calculated from the chemical shifts of the P_i and/or 2-DP-6-phosphate resonances relative to PCr²⁴ with calibration curves obtained by titrating P_i and glucose-6-phosphate dissolved in a solution of high ionic strength, mimicking the intracellular milieu described previously.²² The calibration curve fitted the Henderson-Hasselbach equation with the following parameters: A=3.30 ppm, B=5.75 ppm, and pK₂=6.85.

Total Rb⁺ Content
The Rb⁺ accumulation rate was also measured in a separate series of hearts (n=5) by atomic emission spectrometry (AES). The hearts were loaded with Rb⁺ for 10 minutes according to the protocol for the ⁸⁷Rb NMR experiments (control), freeze-clamped, lyophilized, and digested in HNO₃. The resulting solutions were analyzed by a Jobin-Yvon JY-38 ICP atomic emission spectrometer.

Kinetic Analysis
The rates of Rb⁺ accumulation and washout were calculated from the slopes of the "initial" linear portions of accumulation (4 to 10 minutes) and washout (14 to 20 minutes) curves. The first points were excluded from the analysis to reduce interference from the kinetics of equilibration of the extracellular spaces. The observed first-order rate constants for Rb⁺ influx (k_in*) and efflux (k_eff) were calculated by dividing the respective rates (R_in and R_eff) by the equilibrium intracellular Rb⁺ content (Rb⁺_i(eq)). For the system with constant influx rate, k_in*=k_eff,

(1)

Rb⁺_i, intracellular Rb⁺ content=[Rb⁺]_i V_i,

(2)

(3)

where V is the volume of the respective compartment (in mL/g dry wt) and subscripts e and i denote extracellular and intracellular compartments. At equilibrium,

(4)

Therefore,

(5)

At the same time, k_in*=k_in(V_e/V_i)([Rb⁺]_e/[Rb⁺]_i(eq)) and the true influx rate constant

(6)

Taking V_i=2.5 and V_e=1.5 mL/g dry wt and [Rb⁺]_i(eq)/[Rb⁺]_e=20 to 25, we obtain k_in of 34 to 42 k_in*. However, the "initial" rates are not truly initial because at 4 to 10 minutes' perfusion with Rb⁺ when the [Rb⁺]_i level is no longer zero, the efflux rate is significant compared with the influx rate, resulting in an underestimation of the influx rate. Combining Equations 1, 3, and 4, we have the expression

(7)

which shows that the degree to which Rb⁺ influx rate is underestimated is directly proportional to the degree of Rb⁺ loading. For the changes in Rb⁺ loading of 0% to 40% that are observed in our experiments, the average error will be about 20%. To address this problem experimentally, full kinetic curves corrected for extracellular Rb⁺ were analyzed by use of the equation for monoexponential kinetics derived from Equations 1 through 4:

(8)

The analysis confirmed that the measured initial rates and k_in* are approximately 20% smaller than the true constants.

Statistical Analysis
The data are presented as mean±SD. Groups were compared by one-factor ANOVA at a significance level of 95%. When differences between the groups were found, Student's two-tailed t test was applied to determine P values. Comparisons within the groups were done with a paired two-tailed Student's t test.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Kinetics of Rb⁺ Influx and Efflux in the Beating Rat Heart
Fig 1 (left) shows a series of ⁸⁷Rb NMR spectra from a perfused rat heart acquired with a 2-minute time resolution during Rb⁺ loading and washout. The increase in the intensity of the broad Rb⁺ signal during the 10-minute loading and the decrease during the 20-minute washout are clearly seen. Fig 1 (right) shows the kinetics of Rb⁺ accumulation and washout (n=5 hearts). Rb⁺ accumulation was nearly linear between 4 and 10 minutes, and its rate of uptake determined from the slope was 2.25±0.72 µmol · min^-1 · g dry wt^-1 for pyruvate-perfused hearts and 1.69±0.39 µmol · min^-1 · g dry wt^-1 for glucose-perfused hearts. These values were not significantly different from that determined by AES (2.14±0.28, n=5 pyruvate-perfused hearts), implying that the visibility factor for intracellular Rb⁺ relative to the reference used was close to 1 under our experimental conditions. The observed net Rb⁺ accumulation is a consequence of the superposition of four sequential processes: (1) physical supply of Rb⁺ to the coronary vessels by coronary flow; (2) substitution of 20% of the perfusate K⁺ with Rb⁺ in the extracellular space; (3) passive transport of Rb⁺ through the vascular walls to the interstitium; and (4) active and passive Rb⁺ transport into the cardiac cells. Rb⁺ supply with the perfusate flowing at normal coronary rates (56.6±7.2 mL · min^-1 · g dry wt^-1) was not limiting (56.6x0.94=53 µmol · min^-1 · g dry wt^-1), and the fraction of extracted Rb⁺ was only 4% of its supply (2.24/53=0.04). Moreover, a decrease in the coronary flow by a factor of 4 (to 13.4±2.2 mL · min^-1 · g dry wt^-1) decreased the Rb⁺ influx rate only marginally (1.39±0.25 versus 1.69±0.39 µmol · min^-1 · g dry wt^-1 in glucose-perfused hearts, P=NS, n=5), probably as a result of an increase in Rb⁺ extraction to 18% (2.24/13.4x0.94=0.18), showing no limitation of Rb⁺ supply to the vascular space in this range of flows. At these flow rates, the hearts still remained well oxygenated, based on the observation that lactate production did not increase above the values seen during control perfusion (<4 µmol · min^-1 · g dry wt^-1). The process of Rb⁺ equilibration in the vascular space is fast, since vascular volume is approximately 1% of the coronary flow rate values (18 mL/min).

To estimate the kinetics of Rb⁺ equilibration in the extracellular compartment, the experiment shown in Fig 2 was performed. Hearts perfused in the dry mode (n=5) were equilibrated with Rb⁺ for 30 minutes under standard conditions, and the concentrations of K⁺ and Rb⁺ were then increased fivefold for 10 minutes, from 4.7 to 23.5 mmol/L, with the molar fraction of Rb⁺ maintained constant (20%). This resulted in a rapid increase in the observed Rb⁺ level (by 6.5 µmol), which decreased rapidly (by 5.5 µmol) on termination of the infusion (Fig 2 and Table 1). To assess the possible effects of increased osmolarity, ionic strength, and [Cl^-], an equal amount of ChCl was infused in place of the K⁺+Rb⁺ mixture (n=3) for 10 minutes. The perfusate [ChCl] reached 18.8 mmol/L, with no significant change in the Rb⁺ level. Thus, the observed Rb⁺ increase is mainly due to the increase in [Rb⁺] in the extracellular compartment by a factor of 4. Therefore, the amount of extracellular Rb⁺ under normokalemic conditions can be estimated as (Rb⁺)/4=1.45±0.30 µmol, that is, 19±2.7% of the equilibrium intracellular Rb⁺ content or 16% of its total content. This implies that the extracellular water volume is 1.54 mL (1.45 µmol/0.94 µmol/mL=1.54 mL).

View this table: [in this window] [in a new window]   Table 1. Kinetic Parameters for Rb+ Equilibration in Extracellular and Intracellular Compartments of Perfused Rat Heart

The initial rates of Rb⁺ equilibration in the extracellular space were six times faster and the respective rate constants were 8.5 times higher than those for net Rb⁺ loading and washout measured in the same hearts (Table 1). This indicates that at 4 and 6 minutes of Rb⁺ loading, the [Rb⁺] in the interstitium is about 86% and 95% of its nominal concentration, respectively, and consequently does not significantly restrict the rate of Rb⁺ entry across the sarcolemma. In the system with a constant influx rate, the constants k_in* and k_eff should be identical; however, the former is smaller than the latter. It is explained by the fact that k_in* calculated from the initial slopes is 20% lower than the true constant (see "Kinetic Analysis" in "Materials and Methods"), whereas k_eff does not differ from the true constant, because the efflux rate measured at [Rb⁺]_e equals 0. Therefore, we calculated the arithmetic mean of k_in* and k_eff (k_ave) to obtain the constants of equilibration for intracellular and extracellular compartments (Table 1).

An increase in K⁺+Rb⁺ concentrations resulted in cardiac arrest, whereas ChCl did not affect function. However, both high KCl+RbCl and ChCl reversibly increased the perfusion pressure by 45% (from 69±9 to 100±4 and from 77±6 to 110±13 mm Hg, respectively), which could increase extracellular space and background Rb⁺ signal, resulting in overestimation of the extracellular volume. To investigate the effects of perfusion pressure on the total Rb⁺ signal, coronary flow was decreased from 17.7±1.2 to 5.8±0.3 mL/min for 10 minutes after ChCl washout (10 minutes). This led to a decrease in perfusion pressure from 85±5 to 32±6.3 mm Hg, after which the coronary flow and perfusion pressure were returned to normal. In an additional experimental group, 3 hearts equilibrated with Rb⁺ for 30 minutes were subjected to increases and decreases in perfusion pressure while both ⁸⁷Rb and ²³Na NMR spectra were acquired. It is seen from Fig 2 (right) that the total Rb content observed in both experimental groups (n=6) does not correlate with perfusion pressure (R²=.016), whereas the total Na⁺ content depends linearly on this parameter (R²=.82). The variable part of the total Na⁺ content (one third of total) may reflect changes in the extracellular space (vascular+interstitial) associated with perfusion pressure alterations, whereas the constant part, obtained by extrapolation to zero perfusion pressure, reflects the bath + right ventricle volumes (about two thirds of total). This observation is consistent with the absence of any detectable changes in the equilibrium Rb⁺ level in response to any alterations in perfusion pressure. In fact, the background Rb⁺ signal (extracellular Rb⁺) was estimated to be 16% of the total (see above) and the variable part to be 5% (one third) of the total, which is beyond the sensitivity of the method. This finding is also consistent with a volume of the vascular + interstitial space of 0.5 mL for a 1.5-g heart (one third of total 1.54 mL) because the extracellular space at perfusion pressures of 60 to 80 mm Hg is equal to 0.35 mL/g blotted weight.²⁵ However, under the conditions of kinetic experiments, where Rb⁺ loading during 10 minutes is about 45% of the equilibrium level [100-100xexp -(0.06x10)], the background signal is about 30% and its variable part is about 10% of the total signal.

To test the stability of background Rb⁺ signal, hearts (n=5) were subjected to two consecutive cycles of Rb⁺ loading and washout (30+30 minutes), and the total Na⁺ content was measured at the beginning (0 minutes), in the middle (30 minutes), and at the end (60 minutes) of the protocol. The total Na⁺ integrals were 46.0±1.92, 46.8±1.25, and 47.0±2.6 arbitrary units, respectively, demonstrating stability of the extracellular volume during the Rb⁺ kinetic experiments. The rates of Rb⁺ influx measured during the first and second loading cycles were identical, 1.90±0.30 µmol · min^-1 · g dry wt^-1. This implies stability of the Rb⁺ transport systems and the absence of detectable inhibition of Rb⁺ influx by any residual Rb⁺ remaining inside the cells after the first loading and washout cycle. According to our estimates, this residual Rb⁺ did not exceed 10% of its equilibrium level.

Thus, in accordance with previous observations,¹¹ Rb⁺ uptake by perfused rat hearts measured with ⁸⁷Rb NMR spectroscopy reflects Rb⁺ accumulation in cardiac cells rather than its distribution in extracellular compartments.

Effect of Drugs Specific to K⁺ and Na⁺ Transport on Rb⁺ Uptake
To estimate contributions of different Rb⁺ transporters to Rb⁺ uptake, we measured changes in Rb⁺ uptake in response to specific inhibitors (ouabain, Bum,^{13 14 15} 4-AP^{4 26} ) of these systems. The sensitivity of Rb⁺ uptake to the Na⁺,K⁺-ATPase inhibitor ouabain was measured in procaine-arrested hearts (protocol 2) to minimize the indirect effects of ouabain on energetics, ionic homeostasis, mechanical function, and coronary flow associated with the high doses required for complete inhibition of the Na⁺ pump (0.6 mmol/L). Ouabain frequently induced arrhythmias, severe contracture, and a dramatic decrease in coronary flow (not shown) in beating hearts. Procaine (5 mmol/L) induced cardiac arrest and decreased the Rb⁺ uptake rate by half (48±9% of initial); ouabain further reduced Rb⁺ influx to 26±24% of that with procaine or to 13% of the initial value (Fig 3). Other K⁺ transporters contribute only slightly to Rb⁺ uptake, since Rb⁺ influx was slightly inhibited by Bum (30 µmol/L, 9±8%) and was insensitive to 1 mmol/L 4-AP (Fig 3).

View larger version (15K): [in this window] [in a new window]   Figure 3. Bar graph showing effects of procaine (Proc), ouabain (Ouab), bumetanide (Bum), and 4-aminopyridine (4-AP) on the Rb+ uptake rate in perfused rat hearts. The experiments with Proc and Proc+Ouab were performed according to protocol 2, in which the Rb+ loading and washout cycle was repeated three times in each heart in the following sequence: control ProcProc+Ouab (see "Materials and Methods"). The effects of 4-AP and Bum were tested according to basic protocol 1, in which the control cycle of Rb+ loading and washout was followed by the cycle in the presence of drug. The drugs were infused as described in "Materials and Methods." Ouab was added directly to the perfusate to a concentration of 0.6 mmol/L. Data are presented as mean±SD.

In addition to direct inhibition of the Na⁺ pump and K⁺ transporters, various drugs were used to alter intracellular Na⁺ (infusion of procaine/Lido, Mon, DMA) to determine the correlation between [Na⁺]_i and Rb⁺ influx. Because the Rb⁺ uptake is determined primarily by Na⁺,K⁺-ATPase, it should depend on [Na⁺]_i and agents capable of changing [Na⁺]_i. The sodium-specific ionophore Mon (1 µmol/L) stimulated the Rb⁺ uptake rate 1.6-fold, and the Na⁺ channel blocker Lido (1 mmol/L)^{4 27} reduced the uptake by the same amount (48±18%) as did procaine. DMA (15 to 20 µmol/L), the inhibitor of Na⁺/H⁺ exchange,^{28 29} decreased Rb⁺ influx slightly, by 15±5% (Fig 4).

View larger version (14K): [in this window] [in a new window]   Figure 4. Bar graph showing effects of lidocaine (Lido), dimethylamiloride (DMA), and monensin (Mon) on the Rb+ influx rate in perfused rat hearts. The experiments were performed according to basic protocol 1, in which the control cycle of Rb+ loading and washout was followed by the cycle in the presence of drug. The drugs were infused as described in "Materials and Methods." Data are presented as mean±SD.

The observed effects of DMA and Bum on Rb⁺ uptake were small; therefore, the sensitivity of the assay was increased by inhibiting Na⁺ entry through the Na⁺ channels with Lido (protocol 3). This reduced Rb⁺ uptake by half. DMA further decreased the Rb⁺ uptake rate (from 1.1±0.30 to 0.76±0.17 µmol · min^-1 · g dry wt^-1, a reduction of 16% of control; P=.005; Fig 5), which is equal to the DMA-induced reduction seen in beating hearts (15±5%, Fig 4). Bum also decreased the rate of Rb⁺ uptake from 1.10±0.30 to 0.62±0.22 µmol · min^-1 · g dry wt^-1 (by 22% of control, P=.028, Fig 5), which was a greater change than that observed in beating hearts (9±8%, Fig 4).

View larger version (12K): [in this window] [in a new window]   Figure 5. Bar graph showing inhibition of the Rb+ uptake rate in perfused rat hearts with dimethylamiloride (DMA) and bumetanide (Bum) in the presence of lidocaine (Lido). The data for Rb transport were pooled for two experimental protocols, 1 and 3: (1) controlLidoLido/DMA and (2) LidoLido/drug 1Lido/drug 2, where drug 1 and drug 2 are either DMA or Bum. The drugs were infused as described in "Materials and Methods." Means±SD are presented. Data are expressed as a percentage of values obtained in the presence of lidocaine alone (1.10±0.30 µmol · min-1 · g dry wt-1).

Thus, in beating hearts, the Na⁺,K⁺-ATPase provides more than 80% of the Rb⁺ uptake; the contributions of aminopyridine-sensitive K⁺ channels and Na⁺/K⁺/2 Cl^- cotransport are minimal. Accordingly, Rb⁺ uptake was increased by the agents increasing [Na⁺]_i (Mon) and decreased by the agents decreasing [Na⁺]_i (procaine, Lido).

Effect of Drugs Specific to K⁺ and Na⁺ Transport on Intracellular Na⁺ Levels
All the above inhibitors were tested for their effects on [Na⁺]_i, estimated by ²³Na NMR spectroscopy in the presence of the shift reagent (see "Materials and Methods"). Fig 6 shows representative ²³Na NMR spectra in which the small upfield peak (+0.5 ppm) corresponds to intracellular Na⁺. Procaine reduced the Na⁺_i content by one third of initial content, and subsequent ouabain perfusion (protocol 2) increased [Na⁺]_i twofold relative to the levels observed in the presence of procaine (Figs 6 and 7). In beating hearts (protocol 1), Mon also doubled [Na⁺]_i, Bum and DMA slightly decreased Na⁺_i to 90±3 (P<.02) and 73±11% (P<.03) of control, respectively, and 4-AP had no detectable effect on Na⁺_i (91±13% of control, P=NS) (Figs 6 and 7). Thus, as expected, the Na⁺ (and Ca²⁺) channel inhibitor procaine decreased intracellular Na⁺, and the Na⁺-specific ionophore Mon elevated intracellular Na⁺. The changes in [Na⁺]_i paralleled the changes in the Rb⁺ uptake rate (except in the case of ouabain), as shown in Fig 8.

View larger version (15K): [in this window] [in a new window]   Figure 7. Bar graph showing effect of drugs that alter Rb+ and Na+ transport on intracellular Na+ content in perfused rat hearts. The experiments with procaine (Proc) and Proc+ouabain (Ouab) were performed according to protocol 2 under constant pressure perfusion. For details see "Materials and Methods" and the legend to Table 1. The experiments with monensin (Mon, 1 µmol/L), 4-aminopyridine (4-AP, 1 mmol/L), dimethylamiloride (DMA, 15 µmol/L), and bumetanide (Bum, 30 µmol/L) were carried out according to protocol 1 under both constant-pressure (80 mm Hg) and constant-flow (16 to 22 mL/min) perfusion. The shift reagent, 5 mmol/L Dy(TTHA)3-, was present in the perfusate, and 23Na spectra were constantly acquired throughout each of the protocols. The Rb+ loading and washout procedure was used in experiments with Proc, Proc+Ouab, 4-AP, and Mon, whereas the experiments with Bum and DMA were done without Rb+. The periods of drug perfusion (12 to 15 minutes) and washout (20 minutes) were the same in both sets of experiments. Means±SD are given for three hearts in each group.

View larger version (15K): [in this window] [in a new window]   Figure 8. Graph showing correlation between the Rb+ influx rate and the intracellular Na+ content in perfused rat hearts. Data are taken from Figs 3, 4, and 7 and expressed as percentages of control (CON) values. Details of the experimental protocols are given in "Materials and Methods" and in the legends to the respective figures. 4-AP indicates 4-aminopyridine; Bum, bumetanide; DMA, dimethylamiloride; Mon, monensin; Proc, procaine; and Ouab, ouabain.

Effects of Drugs Specific to K⁺ and Na⁺ Transport on Energy Metabolism and Mechanical Function
³¹P NMR spectroscopy was used to assess effects of the inhibitors on energetic state and intracellular pH. Procaine-induced cardiac arrest did not completely eliminate the side effects of ouabain: during ouabain infusion there was a decrease in PCr (by 45±29%), ATP (by 29±13%), pH_i (by 0.16, P=NS), and total NMR-visible phosphorus (12±6%) as well as a decline in coronary flow (by 60%) and increase in EDP (to 25±7 mm Hg) (Table 2). Changes in PCr, EDP, and coronary flow were reversible upon procaine and ouabain washout (Table 2). These changes in energetic state per se should not markedly affect Rb⁺ uptake, since previous studies have shown that much more significant changes in energetic state induced by 2-DG + insulin treatment (PCr and ATP were 20% and 30% of initial, respectively) and similar changes in intracellular pH (0.16) did not change the Rb⁺ uptake rate.²² Similarly, if a decrease in total NMR-visible phosphorus is used as an index of cell lysis, then only 16% of the observed decrease in the rate of Rb⁺ uptake can be attributed to nonviable cells.

The effects of other drugs (4-AP, Mon, DMA, Bum, Lido) were much less than those of ouabain in terms of their influence on energy metabolism. Table 3 summarizes the immediate effects of these agents on mechanical function observed in the presence and absence of shift reagent. 4-AP (1 mmol/L) increased PRP by 42% as a consequence of an increase in systolic and developed pressure. The increase in PRP was greater in the presence of Mon (1 µmol/L, by 120±52%) and was predominantly due to an increase in the systolic and developed pressure (67%). DMA (15 to 20 µmol/L) and Bum (30 µmol/L, data not shown) did not markedly affect PRP. Lido (1 mmol/L) induced cardiac arrest, and PRP recovery after a 20-minute washout period was nearly complete (87±8% of initial). DMA and Bum applied to Lido-arrested hearts (protocol 3) further reduced functional recovery to 63±8% (P=.001 versus Lido) and 76±16% (P=.028 versus Lido), respectively. This irreversibility probably reflects some cumulative side effects of these drugs, which are well documented for amiloride derivatives.^{29 30} Note, however, that these effects can be overestimated to some extent because the experimental protocol consisted of three repetitive steps of drug infusion and washout. Therefore, some temporal instability and effects of previous drug infusion could add to the observed decrease in PRP.

View this table: [in this window] [in a new window]   Table 3. Effects of 4-Aminopyridine, Monensin, and Dimethylamiloride on Functional Parameters of Perfused Rat Hearts

The metabolic effects of these drugs were assessed by ³¹P NMR according to protocol 4, and the results are presented in Table 4. Because of low levels of P_i characteristic for perfusion with pyruvate, hearts were loaded with minimal amounts of DG-6P (an intracellular pH probe) by short-term perfusion with 4 mmol/L 2-DG. Later, a drug was infused (15 minutes), then washed out (20 minutes), and the procedure was repeated with other drugs (see "Materials and Methods" and legend to Table 4). None of the measured parameters (PCr, ATP, and pH_i) changed significantly during perfusion with 4-AP, DMA, or Bum.

View this table: [in this window] [in a new window]   Table 4. Effects of 4-Aminopyridine, Dimethylamiloride, Bumetanide, and Lidocaine on Metabolic Parameters of Perfused Rat Hearts

Thus, the effects of the drugs on Rb⁺ uptake and [Na⁺]_i described above were not due to changes in the energetic state or intracellular pH of cardiomyocytes and therefore can be ascribed to their direct interactions with K⁺ and Na⁺ transporting systems.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Analysis of Data Provided by ²³Na and ⁸⁷Rb NMR Spectroscopy
NMR spectra of the rat heart reflect changes in the amounts of phosphorus compounds or intracellular Na⁺ and Rb⁺ in the whole heart, which has intrinsic macroscopic and microscopic heterogeneity. The observable NMR signals arise mostly from the left and right ventricles (mass ratio, 4:1), with only minor contributions from the atria or from cells other than cardiomyocytes (endothelial and smooth muscle cells). Therefore, the observable intracellular Na⁺ signal and the intracellular fraction of the Rb⁺ signal reflect the ionic state of the cardiomyocytes. Furthermore, in cardiomyocytes, the contribution from intramitochondrial sodium is small because the intramitochondrial water space is 10% of the total intracellular space³¹ and [Na⁺] in the mitochondrial matrix is lower than that in the cytoplasm.³²

The contribution of the Rb⁺ in the extracellular space to the observed Rb⁺ signal is constant between 4 and 10 minutes of Rb⁺ loading because of its fast equilibration with Rb⁺, as we have demonstrated experimentally in this study (see Fig 2 and the "Results"). However, this conclusion is correct if the extracellular space is also constant. The rates of Rb⁺ accumulation were measured under conditions when background Rb⁺ signal arising from the extracellular space was relatively high, about one third of the maximal value achieved at the end of the kinetic experiments (10 to 15 minutes). However, the potentially variable part of the background arising from the true extracellular space (vascular + interstitial) is only one third of the total background, ie, 10% of the total Rb⁺ signal, as was demonstrated in the ²³Na NMR experiments shown in Fig 2 (right). In addition, the extracellular space was stable during at least two cycles of Rb⁺ loading and washout (60 minutes), and the Rb⁺ influx rates were identical. In constant-flow experiments, drugs could alter extracellular space by changing the perfusion pressure, thereby causing some "drift" in the Rb background level, which could contribute to the observed Rb⁺ influx rates. However, the maximal effect could not exceed 10% even at 50% inhibition of Rb⁺ uptake. In addition, the drugs used in the constant-flow experiments (Lido, Bum, DMA, 4-AP) did not markedly affect the perfusion pressure. Ouabain inhibition was tested under constant-pressure conditions, which precludes significant changes in the extracellular volume. Therefore, the Rb⁺ accumulation rate corresponds to its influx rate into the cardiomyocytes.

The interpretation presented above is correct if some assumptions are valid. First, changes in the intensities of the Na⁺_i and Rb⁺ resonances reflect changes in their concentrations, assuming no change in the visibility factors or intracellular volumes. These assumptions are reasonable under our experimental conditions, since there were no dramatic alterations in the intracellular environment (pH or Mg²⁺). Swelling or shrinking of the cardiomyocytes in response to increased (Mon) or decreased (procaine or Lido) [Na⁺]_i could attenuate the observed changes in the Na⁺_i peak but could not eliminate them completely.

As we indicated in the "Results" section, the absolute magnitudes of the measured Rb⁺ influx rates and k_in* are systematically lower than the unidirectional influx rates and k_eff by 20%. However, this error cannot significantly affect the relative changes in Rb⁺ flux, ie, degrees of inhibition or stimulation of Rb⁺ uptake. The rate constant for Rb⁺ efflux is 0.070 to 0.077 min^-1 and that for Rb⁺ uptake (k_in) is 1.36 to 1.92 min^-1. The equilibrium constant for the transmembrane Rb⁺ distribution is k_in/k_eff=17.7 to 24.9, which is in good agreement with the expected equilibrium ratio [Rb⁺]_i/[Rb⁺]_e of 20 to 25.

Contributions of K⁺ Transporters to Rb⁺ Uptake
We tested various potential pathways of Rb⁺ entry into cardiomyocytes, including Na⁺,K⁺-ATPase, K⁺ channels, and the Na⁺/K⁺/2 Cl^- cotransporter, using specific inhibitors (ouabain, 4-AP, and Bum). Despite complications in interpretation of our experimental data due to the metabolic side effects produced by high concentrations of ouabain (0.6 mmol/L), which necessitated the use of arrested hearts, our data suggest that the Na⁺ pump is responsible for 75% to 87% of the total Rb⁺ uptake. This conclusion is supported by measurements of the Rb⁺ efflux induced by ouabain (0.3 to 0.45 mmol/L) from beating hearts (n=3) equilibrated with Rb⁺. The rate of the efflux (1.24±0.24 µmol · min^-1 · g dry wt^-1), reflecting Na⁺,K⁺-ATPase activity, was 83% of that of Rb⁺ influx (1.48±0.26) in the same hearts (V. Kupriyanov, unpublished observation). These estimations are consistent with the data available in the literature.¹ The contribution of the aminopyridine-sensitive K⁺ channels is minimal under our experimental conditions, since 4-AP had no detectable effect on Rb⁺ influx. However, the contributions of other types of K⁺ channels remain unknown, except ATP-sensitive channels, which should be inactive in metabolically normal hearts (high [ATP]/[ADP] ratio).

The contribution of the Na⁺/K⁺/2 Cl^- symporter can be roughly estimated as <22% because inhibition with Bum may affect Rb⁺ entry by two mechanisms: direct inhibition of K⁺ (Rb⁺) influx via the symporter and inhibition of Na⁺,K⁺-ATPase by reduced intracellular [Na⁺] (see below) due to a decrease in Na⁺ entry through the symporter. This contribution to K⁺ influx can be even less because Rb⁺ influx via the cotransporter may be in part provided by Rb⁺/K⁺ exchange because of reversibility of this system. In other words, unidirectional Rb⁺ influx via the cotransporter should reflect unidirectional K⁺ influx and therefore exceed the net K⁺ influx (because net influx equals unidirectional influx minus unidirectional efflux) mediated by the cotransporter.

We estimate K⁺ influx [J(K⁺)] into cardiomyocytes to be five times that of Rb⁺ uptake {J(K⁺)=J(Rb⁺)[([K⁺]_e+ [Rb⁺]_e)/[Rb⁺]_e]}, ie, about 11 (13.8 with 20% correction) µmol · min^-1 · g dry wt^-1, which is lower than the rates of K⁺ uptake calculated by Allis et al¹¹ from ⁸⁷Rb NMR experiments (18.8 µmol · min^-1 · g dry wt^-1). However, taking into account that these authors used 40% higher K⁺ and Rb⁺ concentrations (total is 6.65 mmol/L), we find that normalized K⁺ fluxes (influx rate constants) are similar (2.0 versus 1.9 min^-1). It is interesting that K⁺ influx rates (in µmol · min^-1 · g protein^-1) determined with the radioisotopes (⁴²K and ⁸⁶Rb) in isolated guinea pig cardiomyocytes (12.4),⁶ cultured neonatal rat heart cells (15.2),¹ and chick embryo ventricular cells (20)³ are in the same range, which again confirms that transporting properties of Rb⁺ are quantitatively similar to those of K⁺. Assuming that 80% of this flux is provided by Na⁺,K⁺-ATPase and taking the enzyme content to be between 1 and 5 nmol/g dry wt^{33 34} and a K⁺/ATP ratio of 2, we can estimate that its turnover number in the beating rat heart is in the range of 14 to 70 s^-1. These values are well below the turnover number (130 to 200 s^-1) determined for isolated Na⁺,K⁺-ATPase¹⁸ in isolated sarcolemmal vesicles,³⁵ isolated cardiomyocytes in the presence of high Mon concentration,⁶ and in the samples of cardiac tissue.³⁴ This is quite reasonable because in situ the enzyme is not saturated with K⁺ and Na⁺.

Relations Between Rb⁺ and Na⁺ Fluxes
The experiments with effectors specific to different pathways of Na⁺ entry (procaine, Lido, Mon, and DMA) showed a close relation between Rb⁺ uptake and Na⁺ influx (see Fig 8). A decrease in Na⁺ entry via Na⁺ channels (procaine or Lido) or blocking Na⁺ influx via the Na⁺/H⁺ exchanger (DMA) resulted in a decrease in Rb⁺ uptake. An increase in Na⁺ entry by means of a Na⁺ ionophore (Mon) stimulated Rb⁺ uptake similarly to that observed in isolated cardiomyocytes.⁶ These findings are in agreement with previous conclusions that Na⁺,K⁺-ATPase is the major pathway of Rb⁺ and K⁺ influx into cardiomyocytes^{1 10} and that this enzyme is a major pathway of Na⁺ extrusion¹⁰ under normal conditions. In the beating heart, the time-averaged transsarcolemmal membrane potential is highly negative (-50 mV), which precludes net Na⁺ extrusion through the Na⁺/Ca²⁺ exchanger under normal conditions.^{9 36} Under these conditions, Na⁺ influx should be equal to Na⁺ efflux via the Na⁺,K⁺-ATPase and should therefore be proportional to K⁺ entry [3J(Na⁺)=2J(K⁺)] and Rb⁺ uptake [J(K⁺)5J(Rb⁺)]. If these assumptions are valid, Na⁺ influx can be estimated as 15 µmol · min^-1 · g dry wt^-1 in the beating heart and half that value in procaine- or Lido-arrested hearts. Assuming that [Na⁺]_i is 25 µmol/g dry wt,^{1 6 37} we can calculate an intracellular Na⁺ turnover number of about 0.6 min^-1 (0.01 s^-1), which is in good agreement with data available in the literature (0.3 to 0.6 min^-1).¹ The rest of the Na⁺ influx (26%) can be attributed to other Na⁺ cotransporters such as Na⁺/HCO₃^-38 and Na⁺/Cl^-14 and some nonidentified permeability.

To better understand the relations between Rb⁺ (K⁺) fluxes and fluxes of Na⁺ and its intracellular concentration (activity), it is useful to refer to the recent review by Eisner and Smith.¹⁰ Briefly, Na⁺ pump flux is a hyperbolic function of [Na⁺]_i, and passive Na⁺ entry is independent of [Na⁺]_i. The latter approximation seems to be reasonable, in view of our findings that the contribution of transporters that are significantly dependent on [Na⁺]_i (Na⁺/H⁺ and Na⁺/K⁺/2 Cl^-) to the total Na⁺ influx does not exceed 25%. Under these conditions, moderate inhibition of the Na⁺ pump by ouabain should not markedly affect Na⁺ efflux/K⁺ influx, provided that V_max for the Na⁺ pump is above the rate of Na⁺ entry, ie, in the steady state. A reduction in the number of active pumps is compensated by stimulation of the remaining pumps resulting from the increased [Na⁺]_i. At high ouabain concentrations, when V_max is lower than Na⁺ influx and the system is no longer in steady state, Na⁺ efflux/K⁺ influx decreases and inhibition is manifest by further elevation of [Na⁺]_i. This explains why the ouabain concentrations required to inhibit Rb⁺ uptake are much higher (0.6 mmol/L) than the concentrations required to induce a positive inotropic effect in the rat hearts (70 µmol/L).³⁹ This theoretical consideration was qualitatively confirmed by titration experiments with ouabain, which showed that in hearts equilibrated with Rb⁺, ouabain concentrations <0.1 mmol/L did not affect the Rb⁺ level, whereas 0.3 to 0.45 mmol/L induced easily detectable Rb⁺ efflux (V. Kupriyanov, unpublished observation).

Under steady state conditions, the rate of Rb⁺/K⁺ uptake does not reflect changes in the number or activity (V_max) of the Na⁺ pumps. Instead, it depends on and reflects changes in the Na⁺ entry rate. However, in combination with measurements of [Na⁺]_i, the rate of Rb