Cardiovascular Research

Cardiovascular Research 2007 75(1):109-117; doi:10.1016/j.cardiores.2007.03.017

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Google Scholar Articles by Swift, F. Articles by Sejersted, O. M. Search for Related Content PubMed PubMed Citation Articles by Swift, F. Articles by Sejersted, O. M. Social Bookmarking          What's this?

Copyright © 2007, European Society of Cardiology

The Na⁺/K⁺-ATPase ₂-isoform regulates cardiac contractility in rat cardiomyocytes

Fredrik Swift^a,b,*, Nils Tovsrud^a,b, Ulla H. Enger^a,b, Ivar Sjaastad^a,b,c and Ole M. Sejersted^a,b

^aInstitute for Experimental Medical Research, Ullevaal University Hospital, University of Oslo, Oslo, Norway
^bCenter for Heart Failure Research, Faculty of Medicine, University of Oslo, Oslo, Norway
^cDepartment of Cardiology, Ullevaal University Hospital, Oslo, Norway

^* Corresponding author. Institute for Experimental Medical Research, Ullevål University Hospital, 4th floor Department of Surgery, Kirkeveien 166, N-0407 Oslo, Norway. Tel.: +47 23016800; fax: +47 23016799. fredrik.swift{at}medisin.uio.no

Received 25 August 2006; revised 20 February 2007; accepted 14 March 2007

	Abstract

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
Objective The presence of both ₁- and ₂-isoforms of the Na⁺/K⁺-ATPase (NKA) in cardiomyocytes indicates different functions. We hypothesized that preferential localization of the ₂-isoform to the t-tubules, locally controlling the Na⁺/Ca²⁺-exchanger (NCX), underlies a specific role in Ca²⁺ handling.

Methods We studied NKA isoform distribution in isolated cardiomyocytes from Wistar rats using immunocytochemistry. NKA pump and NCX currents (I_pump and I_NCX) were measured in control and detubulated cardiomyocytes. Intracellular Na⁺ concentration [Na⁺]_i was assessed with the fluorescent dye SBFI.

Results The ₂-isoform abundance was higher in the t-tubules than in the surface sarcolemma. We established that 0.3 µM ouabain specifically blocked the ₂-isoform in isolated rat cardiomyocytes. This low concentration blocked 10.7±0.6% of I_pump in control, but only 6.0±0.5% in detubulated cardiomyocytes. Moreover, measured and calculated ₁-specific and ₂-specific I_pump in control (547±29 pA and 66 pA, respectively) and in detubulated cells (495±30 pA and 31 pA, respectively) showed that 53% of the ₂-isoform, but only 9.5% of the ₁-isoform, were localized to the t-tubules. Despite the small abundance of the ₂-isoform (11% of total NKA), selective inhibition of this isoform induced a 40% increase in contractility in field stimulated cardiomyocytes, but no increase in global [Na⁺]_i. However, inhibition of the ₂-isoform increased I_NCX indicating local subsarcolemmal accumulation of Na⁺ near NCX.

Conclusions The ₂-isoform of the NKA is functionally coupled to the NCX and can regulate Ca²⁺ handling without changing global [Na⁺]_i.

KEYWORDS Contractile function; e–c coupling; Ions; Ion pumps; Na/K-pump

	1. Introduction

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
The Na⁺/K⁺-ATPase (NKA) controls the concentration gradients for Na⁺ and K⁺ necessary to generate and maintain the membrane potential and to regulate cardiac contractility. Control of intracellular Na⁺ concentration ([Na⁺]_i) is of particular importance since it regulates the Na⁺/Ca²⁺-exchanger (NCX), a key player in the excitation–contraction (EC) coupling.

The active NKA contains a catalytic -subunit (110 kDa) and a small β-subunit (55 kDa). In adult rat cardiomyocytes, two different -subunits, ₁ and ₂, and one β-subunit, β₁, are expressed [1]. Although it has been shown that the ₁-isoform influences Ca²⁺ handling through the NCX [2], it has been proposed that the ₂-isoform plays a different and more important role in Ca²⁺ handling than the ₁-isoform [3–5]. Considering the higher abundance of NCX in the t-tubules than in the surface sarcolemma [6,7], and a slow diffusion of Na⁺ in the cytosol [8], one could expect to find the ₂-isoform preferentially located in the t-tubules close to the NCX. Since it has been proposed that diffusion of solutes is slow in the subsarcolemmal region ("fuzzy space") [9,10], the two membrane proteins would be exposed to the same local "pool" of [Na⁺]_i. However, studies differ in their conclusions as regards the subcellular localization of NKA isoforms. Sweadner et al. [11] reported uniform distribution of ₁ and ₂ in the surface membrane, but little in the t-tubules. McDonough et al. [12] showed uniform distribution of ₂-isoform, whereas the ₁-isoform was homogeneously distributed in t-tubules, but sparse in the surface sarcolemma. Other studies have shown that the ₂-isoform is preferentially localized in the surface membrane, and not in the vicinity of the junctional structures (e.g. not in t-tubules) [13,14]. However, in heterozygous mice lacking one copy of the ₂-isoform, contractility was increased [5], suggesting a role for the ₂-isoform in Ca²⁺ handling. Moreover, contractility was decreased in a heterozygous ₁^+/– mouse [5,15], with compensatory upregulated ₂-isoform, but was increased when the ₂-isoform was blocked by ouabain [15].

The main aim of this study was to determine whether the NKA ₂-isoform is important for Ca²⁺ handling in rat cardiomyocytes. We hypothesized that the NKA ₂-isoform is preferentially located in the t-tubules, interacting tightly with NCX through local [Na⁺]_i in the fuzzy space. We first describe the localization of NKA ₁- and ₂-isoforms shown by immunocytochemical labeling of isolated cardiomyocytes, then confirm the results in electrophysiological experiments in control and detubulated cardiomyocytes. For this, we used a concentration of ouabain that specifically blocked the ₂-isoform. We also show that blocking only the ₂-isoform increases contractility through a local, but not global, increase of [Na⁺]_i.

	2. Methods

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
2.1 Animal care
The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1996).

2.2 Cardiomyocyte preparation
Male Wistar rats (Møllegaard Breeding and Research Center, Skensved, Denmark) weighing 300 g were anaesthetized with 68% N₂O, 29% O₂ and 2–3% isofluran (Abbott Scandinavia, Solna, Sweden), and ventilated on a respirator (Zoovent, Triumph Technical Services, Milton Keynes, UK). Single cardiomyocytes were isolated enzymatically as previously described [16].

2.3 Western blot analysis
Protein homogenates (15 µg for anti-₁ and anti-₂ McHERED, 25 µg for anti-₂ Upstate) were electrophoresed on 6% SDS-PAGE gels, transferred to polyvinylidene difluoride membranes and blocked over night at 4 °C in 5% nonfat dry milk. Primary antibodies used were: anti-₁ (cat#05-369 Upstate, 1:2500), anti-₂ McHERED (courtesy of Alicia McDonough, 1:3000) or anti-₂ (cat#07-674 Upstate, 1:1000). Anti-mouse or anti-rabbit IgG conjugated to horseradish peroxidase (Amersham, cat#NA 931 and NA 934, 1:5000) were used as secondary antibodies. The blots were developed using Enhanced Chemiluminescence (ECL+, Amersham). Signals were quantified using ImageQuant software (Amersham).

2.4 Immunocytochemistry
Isolated cardiomyocytes were fixed in PBS buffer (mM: NaCl 137, Na₂HPO₄ 8, KCl 2.7, KH₂PO₄ 1.5, pH 7.40) containing 4% paraformaldehyde (cat# 04018, Polysciences Inc.) for 15 min, then quenched in PBS+100 mM glycine for 10 min. After 10 min permeabilization in PBS+0.03% Triton X-100, cardiomyocytes were blocked for 2 h in NaCl 150 mM, Na₃ citrate 17.5 mM, 5% goat serum, 3% BSA, 0.02% NaN₃. Cardiomyocytes were incubated over night with primary antibody (₁, ₂ McHERED 1:100; ₂ Upstate 1:200) in NaCl 150 mM, Na₃ citrate 17.5 mM, 2% goat serum, 1% BSA, 0.02% NaN₃, then for 2 h with a secondary antibody (Alexa-488 conjugated IgG, 1:200, Molecular Probes). Cardiomyocytes were scanned on a Zeiss LSM510 confocal microscope (excitation 488 nm Argon laser, emission collected at 505–550 nm). The images were deconvolved using Huygens Essential software (SVI, The Netherlands) before analysis.

2.5 NKA current (I_pump) and NCX current (I_NCX) measurements
We used protocols modified from [17] for whole-cell voltage clamping of cardiomyocytes (Axopatch 200B amplifier, Axon Instruments, Foster City, USA) to –50 mV with electrodes (1.3±0.1 M) filled with a solution containing (mM): NaOH 40, sodium phosphocreatine 5, HEPES 10, TEA-Cl 20, L-aspartic acid 42, EGTA 42, CaCl₂ 29.7, MgATP 10, pH 7.20. The final [Na⁺] was 50 mM. The free [Ca²⁺] was calculated to be 300 nM, using WinMAX C 2.01 software (Chris Patton, Stanford University, CA, USA). Cardiomyocytes were superfused (37 °C) with solution I containing (mM): NaCl 145, MgCl₂ 2, CsCl 2, EGTA 0.1, D-glucose 5.5, Hepes 5, BaCl₂ 2, nicardipine 0.001, pH 7.40 for at least 4 min until the background current was stable. The NKA and NCX were activated by applying solution I containing 5.4 mM KCl and 2 mM Ca²⁺, respectively. The peak outward shift in the membrane current was taken as I_pump or I_NCX. Ouabain was perfused for at least 2 min before measuring I_pump or I_NCX in the presence of ouabain. Voltage ramp protocols were applied before and during activation of the I_pump. Cardiomyocytes were depolarized to +70 mV for 50 ms, then hyperpolarized to –120 mV (dV/dt=380 mV s^–1), and back to –50 mV. The difference between the recorded current at baseline and during activation of NKA defined the I_pump–voltage relationship.

2.6 Detubulation
Detubulation was induced by exposure to formamide as described by Kawai et al. [18] using solution II (mM): NaCl 140, D-glucose 5.5, KCl 5.4, HEPES 5, CaCl₂ 1, MgCl₂ 0.5, NaH₂PO₄ 0.4 (pH 7.40). Normal and detubulated cells were labeled with di-8-ANEPPS to quantify visible membrane fluorescence in the optical sections, and to assess the extent of detubulation both as previously described [19] and by plotting capacitance as a function of cell volume. Membrane capacitance was calculated from the capacitive current induced by a 10 mV depolarizing step. Cell volume was evaluated using v=(lwd)/4 [20], where l was the cell length, and w the width. The cell depth (d) was assumed to be 1/3 of the cell width [20].

2.7 Contractions and Ca²⁺ transients
Laminin plated cardiomyocytes were field stimulated at 1 Hz in solution II containing 1.8 mM Ca²⁺ (37 °C). Contractility (fractional shortening divided by time to peak, FS/TTP) was measured from contractions recorded by a video edge detector (Crescent Electronics, Sandy, UT USA). Ca²⁺ transients were measured in cardiomyocytes loaded with Fluo-4-AM by longitudinal line scans on a Zeiss LSM510 confocal microscope. Fluo-4 was excited at 488 nm and emission was measured at 505–550 nm.

2.8 Intracellular Na⁺ measurements
Global intracellular Na⁺ in cardiomyocytes field stimulated at 1 Hz in solution II+1.8 mM Ca²⁺ was determined by emission ratio mode Na⁺-binding benzofuran isophthalate (SBFI) fluorescence as described by Baartscheer et al. [21]. To adjust for variations in the SBFI loading of the cardiomyocytes, the ratio was adjusted to zero for [Na⁺]=0 mM. After the experiment each cardiomyocyte was perfused with solution III containing: monensin 30 µM, gramicidin 2 µg/ml, ouabain 1 mM, Hepes 5 mM, Glucose 5.5 mM, EGTA 2 mM, supplemented with NaCl (0 or 140 mM) and KCl (140 or 0 mM), pH 7.20.

To calculate [Na⁺]_i, the fluorescence signal was calibrated in a separate series of experiments. Cardiomyocytes were perfused with solution III containing 0, 7, 24, 70 or 140 mM NaCl ([NaCl]+[KCl]=140 mM). The calibration data were also adjusted to 0 for [Na⁺]=0 mM. [Na⁺]_i was calculated according to:

where K_d is the intracellular dissociation constant, β is the ratio of fluorescence signals in sodium free and sodium saturated cells measured at 590 nm, R the ratio of F₄₁₀/F₅₉₀, R_min and R_max are minimal and maximal ratios in sodium free and sodium saturated cardiomyocytes. K_d, β, R_min and R_max used to solve Eq. (A) were determined in the calibration experiments.

2.9 Statistics
Values are expressed as means±SE. Comparisons between means were made using Student's t-test (homoscedastic, two-tailed distribution), and differences were considered significant when p<0.05.

	3. Results

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
3.1 Antibody specificity
The anti-₁ antibody labeled a single prominent band at 100 kDa (Fig. 1A), which was not visible in the negative control (Fig. 1B). This indicated a high specificity of the antibody. The anti-₂ McHERED antibody labeled 3 bands at 100 kDa, 75 kDa and 65 kDa (Fig. 1C). The bands at 100 kDa and 65 kDa were not labeled when the primary antibody was saturated with an excess of the peptide sequence (KHEREDSPQSHVL, Eurogentec, Belgium) used to produce the antibody, indicating that the 65 kDa band was a degradation product of the ₂-isoform (Fig. 1D). The nature of the 75 kDa protein is unknown. All 3 bands were invisible in the negative control (Fig. 1E). The anti-₂ Upstate antibody labeled a single prominent band at 97 kDa (Fig. 1F), indicating high specificity to the ₂-isoform.

View larger version (30K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 1 Antibody specificity. Western blots of (A) Anti-1; (B) Anti-1, negative control; (C) Anti-2 McHERED; (D) Anti-2 McHERED, preabsorbed; (E) Anti-2, negative control; (F) Anti-2 Upstate.

 
3.2 The ₂-isoform of NKA were abundant in the t-tubules
Fig. 2A, C and E show deconvolved images of cardiomyocytes labeled with anti-₁, anti-₂ McHERED and anti-₂ Upstate antibodies, respectively. Labeling was stronger in the intercalated discs than in the rest of the cell for all antibodies, but this was most pronounced for the anti-₁ antibody. Moreover, the anti-₁ antibody labeled both transverse and longitudinal t-tubules, whereas labeling by the two anti-₂ antibodies was predominant only in the transverse t-tubules. The same distribution pattern as for the ₂ antibodies was also observed for NCX by Thomas et al. [6]. In Fig. 2B, D and F, the mean fluorescence intensity along a line in 10 distinct regions of the surface membrane (not including the intercalated discs) was compared to the mean fluorescence intensity of 30 distinct regions of the longitudinal t-tubules. The ₁-isoform was significantly less abundant in the t-tubules than in the surface membrane (Fig. 2B). The ₂-isoform, on the other hand, was more abundant in the t-tubules than in the surface membrane as shown with both anti-₂ antibodies (Fig. 2D and F, p<0.05). By comparison, labeling intensity of the t-tubules with di-8-ANEPPS was merely 51% of that of the sarcolemma indicating that less t-tubule membrane was included in the optical section. Adjusting for this, the relative abundance of the ₂-isoform in the t-tubules would be significantly higher than indicated in Fig. 2, but the low accuracy of the method does not justify more exact calculations. For instance, differences in epitope accessibility between t-tubules and surface membrane could exist. To verify that the ₂-isoform was preponderant in the t-tubules, the NKA current attributable to the ₂-isoform was measured before and after detubulation.

View larger version (85K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 2 Immunolocalization using (A) Anti-1, (C) anti-2 McHERED and (E) anti-2 Upstate antibodies. Scale bar=5.4 µm. Mean fluorescence in the t-tubules (white bars) was compared to the mean fluorescence in the surface membrane (black bars) for each antibody used: (B) anti-1, (D) anti-2 McHERED and (F) anti-2 Upstate.

 
3.3 A low concentration of ouabain selectively inhibited the ₂-isoform
In the rat, the -isoforms differ with regard to their affinity for cardiac glycosides such as ouabain [22]. Voltage clamp experiments were performed to determine the concentration of ouabain that selectively blocked the ₂-isoform in isolated rat cardiomyocytes (Fig. 3A). The dose-response curve, fitted with a two site competition, nonlinear regression equation, is presented as percent reduction from baseline I_pump (Fig. 3B). The K_d of the high-affinity component (₂-isoform) was 19.9 nM, whereas that of the low-affinity component (₁-isoform) was 43.4 µM. Thus, a concentration of 0.3 µM ouabain blocked 94% of the ₂-isoform, but less than 1% of the ₁-isoform. Fig. 3C shows the normalized current–voltage (IV) relationship for I_pump,1 (₁-isoform related I_pump) and I_pump,2 (₂-isoform related I_pump). I_pump,1 was taken as the difference between currents recorded during voltage ramps c and d in Fig. 3A. I_pump,2 was taken as the difference between total I_pump (ramp a–ramp b) and I_pump,1. The currents in each experiment were fitted with a sigmoid function [23]. The IV-relationship for I_pump,2 was steeper than I_pump,1 at low potentials. This is in accordance with IV-relationships measured in oocytes [24,25].

View larger version (27K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 3 (A) Representative current recording during activation of Ipump in the absence and presence of 0.3 µM ouabain. Recorded currents during voltage ramps were truncated and are labeled a–d. (B) The percent reduction of Ipump from baseline was plotted for each concentration of ouabain used. Currents from 3–7 cardiomyocytes were measured for each concentration. (C) IV-relationship for Ipump,1 and Ipump,2 normalized to 1 at 50 mV. The figure shows mean values and SE of current values read from the fitted curves of each recorded IV-relationship (n=10). *p<0.05.

 
3.4 The I_pump,2 was functionally located in the t-tubules
Detubulated cardiomyocytes were used to discriminate between I_pump,2 in the t-tubules and the surface membrane. Fig. 4A and B shows di-8-ANEPPS stained control and detubulated cardiomyocytes, respectively. Almost no t-tubules were observed in the formamide treated cardiomyocytes. Fig. 4C shows the surface-to-volume ratio which was 22% lower in detubulated than in control cardiomyocytes (7.3±0.5 pF/pL vs. 9.3±0.5 pF/pL, p<0.05). When control cardiomyocytes were exposed to the low concentration of ouabain, I_pump density was reduced by 10.7±0.6% (2.62±0.13 pA/pF vs. 2.33±0.12 pA/pF, p<0.01), indicating that only a minor portion of the I_pump was carried by the NKA ₂-isoform. The reduction in I_pump density induced by the low concentration of ouabain was even less after detubulation (6.0±0.5%, 2.95±0.12 pA/pF vs. 2.78±0.11 pA/pF, p<0.01), indicating a higher density of the ₂-isoform in the t-tubules than in the surface membrane. T-tubular I_pump,1 and I_pump,2 were calculated and are shown in Table 1: 53% of the ₂-isoform current, but only 9.5% of the ₁-isoform current were located in the t-tubular membrane. Thus, the ₂-isoform was functionally preponderant in the t-tubules.

View larger version (36K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 4 Confocal images of (A) control and (B) detubulated cardiomyocytes stained with di-8-ANEPPS. (C) Cell capacitance vs. cell volume for control cells () and detubulated cells (). Mean values are presented by square symbols, and linear regressions were calculated for control (solid, =175, β=2.1) and detubulated (dotted, =134, β=1.7) cardiomyocytes.

 

View this table: [in this window] [in a new window]   Table 1 Relative amounts of 1 and 2 present in the t-tubules in rat cardiomyocytes

 
3.5 The ₂-isoform regulates contractility
Despite the rather low amount of the ₂-isoform in the cardiomyocyte, the high abundance in the t-tubules could indicate a role in Ca²⁺ handling. The low concentration of ouabain gradually increased FS/TTP to 40% higher after 8–9 min than at baseline (p<0.05) (Fig. 5). Also the maximum shortening and relaxation velocities (CV and RV) increased by 40% (baseline: CV 0.27±0.02 µm/ms, RV 0.29±0.03 µm/ms; ouabain: CV 0.39±0.02 µm/ms, RV 0.41±0.02 µm/ms, both p<0.05). Moreover, ouabain increased both the diastolic and systolic [Ca²⁺]_i and the amplitude of the transient by 70%. Also the rate of decline of the Ca²⁺ transient as assessed by the time from peak transient to 90% recovery was increased (350±28 ms vs. 269±28 ms). The question is whether the positive inotropic effect of ouabain was due to a global or local rise in [Na⁺]_i.

View larger version (26K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 5 Contractions (A) and longitudinal line scans (B) were recorded in separate experiments in field stimulated single cardiomyocytes (1Hz) before (black) and after 8–9 min application of 0.3 µM ouabain (grey). (C) Mean values of FS/TTP, n=9. *p<0.05. (D) Representative tracings of averaged transients.

 
3.6 Exposure to 0.3 µM ouabain did not increase global [Na⁺]_i
Fig. 6A shows a typical recording of the F₄₁₀/F₅₉₀ ratio. Mean results from 12 experiments are presented in Fig. 6B, showing no detectable increase in [Na⁺]_i after 9 min of exposure to 0.3 µM ouabain. However, subsequent addition of 1 mM ouabain increased [Na⁺]_i within 1–2 min (p<0.05). The experiment was aborted before maximal response due to Ca²⁺ overload of the cell, and the measured F₄₁₀/F₅₉₀ at 1 mM ouabain only served as a positive control. Mean data from 9 calibration experiments (4 hearts) are shown in Fig. 6C. The data were fitted with a one site saturation, nonlinear regression equation (Sigmaplot, Systat Software Inc., California, USA) with a correlation coefficient of 0.85±0.01), giving R_max=0.092±0.006 and K_d=20.7±4.6 mM. These values were comparable to those determined by Baartscheer et al. [21]. With R_min adjusted to 0, and β=0.95±0.04, Eq. (A) gave a [Na⁺]_i of 12.7±0.4 mM at baseline and 12.5±0.5 mM during exposure to 0.3 µM ouabain (NS). The smallest detectable change in [Na⁺]_i was calculated to be 1.7 mM (power 0.80, alpha<0.05). Thus, increased contractility during 0.3 µM ouabain exposure could not be explained by a substantial global rise in [Na⁺]_i.

View larger version (15K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 6 (A) Representative tracing of SBFI F410/F590 ratio recorded before and during exposure to 0.3 µM and 1 mM ouabain. Values of [Na+]i were calculated from the calibration curve. (B) Mean results from 12 experiments showed no increase in [Na+]i after 9 min of exposure to 0.3 µM ouabain, but [Na+]i increased after exposure to 1 mM ouabain. (C) SBFI calibration curve. The solid line represents the fitted data, and the dotted line represents the maximal value of the fitted curve.

 
3.7 Exposure to 0.3 µM ouabain increased I_NCX
To test if the increased contractility could be due to local accumulations of Na⁺ in proximity to NKA and NCX, I_NCX and I_pump were measured simultaneously in the absence and in the presence of 0.3 µM ouabain (Fig. 7A). At steady state I_pump(step 4 in Fig. 7), I_NCX was reduced by 78±3% in control and by 69±3% in cardiomyocytes exposed to 0.3 µM ouabain, compared to I_NCX at baseline (control: 161±12 pA at baseline, 36±6 pA during steady state I_pump; 0.3 µM ouabain: 201±7 pA at baseline, 63±6 pA during steady state I_pump). Assuming 1) a linear relationship between [Na⁺]_i and I_NCX at –50 mV, 2) baseline I_NCX was measured at 50 mM [Na⁺]_i which was used in the pipette, and 3) [Na⁺]_e, [Ca²⁺]_i and [Ca²⁺]_e were fixed at 145 mM, 300 nM and 2 mM respectively, we calculated the local [Na⁺]_i at steady state I_pump. An I_NCX of 36 pA would correspond to 11.2 mM local [Na⁺]_i and 63 pA would correspond to 15.7 mM local [Na⁺]_i. Thus, ₂-isoform inhibition by ouabain seems to cause an increase in local [Na⁺]_i of 4–5 mM. Moreover, the relationship between I_NCX and [Na⁺]_i has previously been established in experiments with giant membrane patches from guinea pig ventricular cells [26]. According to that relation, the observed difference between I_NCX measured during steady state I_pump in control and during exposure to 0.3 µM ouabain could fit with a local increase in [Na⁺]_i of 3 mM, which is in reasonable agreement with the present data.

View larger version (23K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 7 Simultaneous Ipump and INCX recordings. (A) Representative current recording during activation of NCX (addition of 2 mM Ca2+) and NKA (addition of 5.4 mM K+). (B) Peak INCX was measured every 2 min and normalized to the INCX measured before activation of Ipump. Experiments were done in the absence (black bars) and presence (white bars) of 0.3 µM ouabain. *p<0.05.

 

	4. Discussion

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
In this study, we show by immunocytochemical and electrophysiological methods that the ₂-isoform of the NKA is abundant in the t-tubules of rat cardiomyocytes whereas the ₁-isoform is abundant in the surface membrane. Selective inhibition of the ₂-isoform by a low concentration of ouabain increased contractility by 40%, but did not induce an increase in global [Na⁺]_i detectable by SBFI. However, a local accumulation of [Na⁺]_i was detectable by I_NCX measurements. This suggests that the ₂-isoform is efficiently coupled to contractility through a subsarcolemmal pool of Na⁺, e.g. in the fuzzy space [9,10].

We determined immunocytochemically the subcellular distribution of the two -isoforms of the NKA. Only the ₂-McHERED antibody showed one unspecific band in Western blots, but gave similar immunocytochemical labeling as the ₂ Upstate antibody. We therefore assume that the epitope giving the unspecific band in the Western blot was not accessible in isolated cells, and thus all the antibodies had a high specificity for its target protein in isolated cells. Localization of the ₂-subunit was predominant in the t-tubules and the ₁-isoform was more abundant in the surface sarcolemma than in the t-tubules, especially in the intercalated discs. Similar distribution patterns have been observed by others in guinea pig, rat and rabbit cardiomyocytes [27–29]. However, in another study in the rat [12], the authors observed ₁-isoforms enriched in the t-tubules, and uniformly distributed ₂-isoforms in the t-tubules and surface sarcolemma.

Detubulation of cardiomyocytes was used to separate I_pump in the t-tubules from the I_pump in the surface membrane. Detubulation has been used in several studies and has been extensively validated [6,7,18,30]. Cell capacitance was 30% lower in the detubulated cells than in control cells, but the cell volume was not changed. Exposure to formamide could affect NKA properties and distribution in the cell membrane. The detubulation experiments should therefore be interpreted with care. However, no effects on cytoskeleton or protein properties were found in formamide treated atrial cells lacking t-tubules [30]. Furthermore, the distribution of NKA ₂-isoforms shown in the electrophysiological experiments (53% of the ₂-isoform in the t-tubules) corresponds with the distribution pattern shown by immunocytochemistry. As shown in Table 1, only 9.5% of the I_pump,1 originates from the t-tubules. Even though the labeling of the ₁-isoform was more intense in the surface membrane than in the t-tubules, the difference in intensity could not account for the difference in I_pump,1. An explanation for this could be that the intercalated discs were intensely labeled with the ₁-antibody, indicating that a large portion of the ₁-isoforms is located in the intercalated discs. However, we were not able to discern pump activity in the intercalated discs from that in the rest of the surface membrane. Intense labeling of the intercalated discs has been observed with several antibodies to membrane proteins [6], including other antibodies for NKA [12], and could partly be due to excessive membrane folding in this region [31].

In the rat, ouabain affinity differs between the various NKA isoforms. The dissociation constants obtained in our study were in the same order of magnitude as the values from the literature [13,28,32,33]. The consequences of inhibition of the NKA by ouabain are increased Ca²⁺ load of the sarcoplasmic reticulum (SR) and larger Ca²⁺ transients that will cause increased contractility [34]. It has also been proposed that ouabain exposure will increase the Ca²⁺ sensitivity of the ryanodine receptors (RyR) [35,36]. This mechanism could act in concert with the effect of ouabain on the NKA to increase the size of the Ca²⁺ transient. However, as pointed out by Eisner et al. [37], increasing the open probability of RyR will only transiently increase cytosolic Ca²⁺, and thence cell shortening. When steady state is achieved, the SR Ca²⁺ load will be lowered, due to increased leak through RyR. Thus, the Ca²⁺ transient would return to control levels. In our experiments, the cell shortening increased gradually and reached a sustained steady state after 8–9 min. Thus, effects of ouabain on RyR are unlikely to explain the increased contraction observed in our experiments. Moreover, a recent study showed that the inotropic effect of ouabain requires intact NCX function [38].

Although our results show that the ₂-isoforms only constitute 11% of the total NKA, the inhibition of ₂ induced a large positive inotropic effect (40%). The percentage of the ₂-isoform is similar to that suggested by Lingrel et al. [4]. The role of the ₂-isoform has also been investigated in mice expressing ouabain-insensitive ₂-isoforms [39]. Here, ouabain failed to induce the positive inotropic effect observed in control mice. This indicates that the ₂-isoform is efficiently linked to contractility, probably through the NCX. We have previously shown that NCX is present both in the t-tubules and in the surface sarcolemma, but the density is higher in the t-tubules [6]. The predominant localization of NCX in the t-tubules has also been shown in other studies [7,40]. The interaction between the ₂-isoform and NCX probably occurs through a local subsarcolemmal pool of Na⁺, since inhibition of the ₂-isoform did not induce a global increase in [Na⁺]_i, but did increase I_NCX. Increasing the concentration of ouabain to concentrations also inhibiting the ₁-isoform often led to spontaneous Ca²⁺ waves (data not shown), indicating that inhibition of the ₁-isoform induced Ca²⁺ overload associated with increase in global [Na⁺]_i. It seems clear that the ₂-isoform plays an important role in the fine tuning of contractility, whereas the ₁-isoform is more important in the maintenance of Na⁺ and K⁺ homeostasis of the whole-cell, and thereby also regulate contractility. In addition to increasing the SR Ca²⁺ load, reduced Na⁺/K⁺-ATPase activity could increase contractility through a more efficient triggering of Ca²⁺ release. Lines et al. [41] recently showed that NCX could play an important role in the triggering of SR Ca²⁺ release. This requires slow diffusion of Na⁺ in the fuzzy space so that Na⁺ can accumulate and reverse the NCX. Our data support this prediction. Since the ₂-isoform seems to control [Na⁺]_i in the fuzzy space, a regulatory role of the triggering of SR Ca²⁺ release can also be predicted for the ₂-isoform. Interestingly, the IV-relationship was steeper for the ₂-isoform (Fig. 3C) indicating a more dynamic role for the ₂-isoform than for the ₁-isoform during the action potential. The ₂-isoform activity will be very low during the resting potential, avoiding depletion of Na⁺ in the fuzzy space during diastole. This will facilitate the early and rapid reversal of the NCX early during the action potential as predicted by Lines et al. [41]. However, during the action potential, the ₂-isoform will rapidly be activated, lowering local [Na⁺]_i and favoring forward mode NCX.

Downregulation of the ₂-isoform during heart failure [1] could impair the tight control of the excitation–contraction coupling. Such downregulation would increase [Na⁺]_i in the subsarcolemmal pool. This would impair the possible contribution of rapid reversal of the NCX to trigger Ca²⁺ release. This could contribute to more dyssynchronous Ca²⁺ release as observed in models of heart failure [42]. Also, reduced Ca²⁺ extrusion through the NCX would increase SR Ca²⁺ load. In heart failure, SR Ca²⁺ load is most often decreased [43–45]. This suggests that the downregulation of the ₂-isoform is a compensatory mechanism to limit the decrease in SR Ca²⁺ load. The use of low doses of cardiac glycosides to improve contractile properties might enhance this compensatory mechanism. Further studies are required to determine the causes and consequences of ₂-isoform downregulation in heart failure. Extrapolations to other species are not straightforward since rat cardiomyocytes are special due to their extensive t-tubular network and since their -isoforms have different affinities for ouabain. However, a recent study showed that the human ₂-isoform exhibit more rapid ouabain association and dissociation rates than the ₁-isoform, and also different Na⁺ affinity and voltage dependence [25].

We conclude that the different -isoforms of the NKA are differentially distributed in the sarcolemma in normal rat cardiomyocytes. The ₁-isoform is located throughout the whole sarcolemma, especially in the intercalated discs but to a lesser extent in the t-tubules. This suggests a "housekeeping" role for the ₁-isoform, particularly important in the maintenance of ionic gradients across the membrane. The ₂-subunit seems to be preferentially located to the t-tubules and is functionally linked to the NCX. This provides evidence for a role for the ₂-isoform in the tight regulation of cardiac contractility.

Time for primary review 23 days

	Acknowledgements

 
We thank Tor Skomedal and Jon Arne Kro Birkeland for helpful discussions, Clive Orchard and Fabien Brette for help with the detubulation technique. Animal care was carried out by the Section for Comparative Medicine. This study was supported by grants from the Norwegian Council for Cardiovascular Diseases, the Norwegian Research Council and Anders Jahre's Fund for Promotion of Science.

	References

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 

1. Semb S.O., Lunde P.K., Holt E., Tønnessen T., Christensen G., Sejersted O.M. Reduced myocardial Na⁺/K⁺-pump capacity in congestive heart failure following myocardial infarction in rats. J Mol Cell Cardiol (1998) 30:1311–1328.[CrossRef][ISI][Medline]
2. Dostanic I., Schultz J.J., Lorenz J.N., Lingrel J.B. The ₁ isoform of Na,K-ATPase regulates cardiac contractility and functionally interacts and co-localizes with the Na/Ca exchanger in heart. J Biol Chem (2004) 279:54053–54061.[Abstract/Free Full Text]
3. Yamamoto T., Su Z., Moseley A.E., Kadono T., Zhang J., Cougnon M., et al. Relative abundance of ₂ Na⁺ pump isoform influences Na⁺–Ca²⁺ exchanger currents and Ca²⁺ transients in mouse ventricular myocytes. J Mol Cell Cardiol (2005) 39:113–120.[CrossRef][ISI][Medline]
4. Lingrel J., Moseley A., Dostanic I., Cougnon M., He S., James P., et al. Functional roles of the isoforms of the Na,K-ATPase. Ann N Y Acad Sci (2003) 986:354–359.[Abstract/Free Full Text]
5. James P.F., Grupp I.L., Grupp G., Woo A.L., Askew G.R., Croyle M.L., et al. Identification of a specific role for the Na⁺/K⁺-ATPase ₂-isoform as a regulator of calcium in the heart. Mol Cell (1999) 3:555–563.[CrossRef][ISI][Medline]
6. Thomas M.J., Sjaastad I., Andersen K., Helm P.J., Wasserstrom J.A., Sejersted O.M., et al. Localization and function of the Na⁺/Ca²⁺-exchanger in normal and detubulated rat cardiomyocytes. J Mol Cell Cardiol (2003) 35:1325–1337.[CrossRef][ISI][Medline]
7. Despa S., Brette F., Orchard C.H., Bers D.M. Na⁺/Ca²⁺-exchange and Na⁺/K⁺-ATPase function are equally concentrated in transverse tubules of rat ventricular myocytes. Biophys J (2003) 85:3388–3396.[Abstract/Free Full Text]
8. Despa S., Kockskamper J., Blatter L.A., Bers D.M. Na/K pump-induced [Na]_i gradients in rat ventricular myocytes measured with two-photon microscopy. Biophys J (2004) 87:1360–1368.[Abstract/Free Full Text]
9. Lederer W.J., Niggli E., Hadley R.W. Sodium–calcium exchange in excitable cells: fuzzy space. Science (1990) 248:283.[Free Full Text]
10. Leblanc N., Hume J.R. Sodium current-induced release of calcium from cardiac sarcoplasmic reticulum. Science (1990) 248:372–376.[Abstract/Free Full Text]
11. Sweadner K.J., Herrera V.L., Amato S., Moellmann A., Gibbons D.K., Repke K.R. Immunologic identification of Na⁺/K⁺-ATPase isoforms in myocardium. Isoform change in deoxycorticosterone acetate–salt hypertension. Circ Res (1994) 74:669–678.[Abstract/Free Full Text]
12. McDonough A.A., Zhang Y., Shin V., Frank J.S. Subcellular distribution of sodium pump isoform subunits in mammalian cardiac myocytes. Am J Physiol (1996) 270:C1221–C1227.[ISI][Medline]
13. Noël F., Wibo M., Godfraind T. Distribution of ₁ and ₂ Na⁺,K⁺-ATPase isoforms between the junctional (t-tubular) and non-junctional sarcolemmal domains of rat ventricle. Biochem Pharmacol (1991) 41:313–315.[CrossRef][ISI][Medline]
14. Slezak J., Schulze W., Stefankova Z., Okruhlicova L., Danihel L., Wallukat G. Localization of ₁,₂,₃-subunit isoforms of Na,K-ATPase in cultured neonatal and adult rat myocardium: the immunofluorescence and immunocytochemical study. Mol Cell Biochem (1996) 163–164:39–45.
15. Moseley A.E., Cougnon M.H., Grupp I.L., El S.J., Lingrel J.B. Attenuation of cardiac contractility in Na,K-ATPase ₁ isoform-deficient hearts under reduced calcium conditions. J Mol Cell Cardiol (2004) 37:913–919.[CrossRef][ISI][Medline]
16. Qvigstad E., Brattelid T., Sjaastad I., Andressen K.W., Krobert K.A., Birkeland J.A., et al. Appearance of a ventricular 5-HT4 receptor-mediated inotropic response to serotonin in heart failure. Cardiovasc Res (2005) 65:869–878.[Abstract/Free Full Text]
17. Fujioka Y., Matsuoka S., Ban T., Noma A. Interaction of the Na⁺/K⁺-pump and Na⁺/Ca²⁺-exchange via [Na⁺]_i in a restricted space of guinea-pig ventricular cells. J Physiol (1998) 509:457–470.[Abstract/Free Full Text]
18. Kawai M., Hussain M., Orchard C.H. Excitation–contraction coupling in rat ventricular myocytes after formamide-induced detubulation. Am J Physiol (1999) 277:H603–H609.[ISI][Medline]
19. Swift F., Strømme T.A., Amundsen B., Sejersted O.M., Sjaastad I. Slow diffusion of K⁺ in the t-tubules of rat cardiomyocytes. J Appl Physiol (2006) 101:1170–1176.[Abstract/Free Full Text]
20. Boyett M.R., Frampton J.E., Kirby M.S. The length, width and volume of isolated rat and ferret ventricular myocytes during twitch contractions and changes in osmotic strength. Exp Physiol (1991) 76:259–270.[Abstract]
21. Baartscheer A., Schumacher C.A., Fiolet J.W. Small changes of cytosolic sodium in rat ventricular myocytes measured with SBFI in emission ratio mode. J Mol Cell Cardiol (1997) 29:3375–3383.[CrossRef][ISI][Medline]
22. Blanco G., Mercer R.W. Isozymes of the Na⁺/K⁺-ATPase: heterogeneity in structure, diversity in function. Am J Physiol (1998) 275:R633–R650.
23. Gadsby D.C., Nakao M. Steady-state current–voltage relationship of the Na/K pump in guinea pig ventricular myocytes. J Gen Physiol (1989) 94:511–537.[Abstract/Free Full Text]
24. Horisberger J.D., Kharoubi-Hess S. Functional differences between -subunit isoforms of the rat Na⁺/K⁺-ATPase expressed in Xenopus oocytes. J Physiol (2002) 539:669–680.[Abstract/Free Full Text]
25. Crambert G., Hasler U., Beggah A.T., Yu C., Modyanov N.N., Horisberger J.D., et al. Transport and pharmacological properties of nine different human Na,K-ATPase isozymes. J Biol Chem (2000) 275:1976–1986.[Abstract/Free Full Text]
26. Matsuoka S., Hilgemann D.W. Steady-state and dynamic properties of cardiac sodium–calcium exchange. Ion and voltage dependencies of the transport cycle. J Gen Physiol (1992) 100:963–1001.[Abstract/Free Full Text]
27. Fransen P., Hendrickx J., Brutsaert D.L., Sys S.U. Distribution and role of Na⁺/K⁺ ATPase in endocardial endothelium. Cardiovasc Res (2001) 52:487–499.[Abstract/Free Full Text]
28. Silverman B.Z., Fuller W., Eaton P., Deng J., Moorman J.R., Cheung J.Y., et al. Serine 68 phosphorylation of phospholemman: acute isoform-specific activation of cardiac Na/K ATPase. Cardiovasc Res (2005) 65:93–103.[Abstract/Free Full Text]
29. Harada K., Lin H., Endo Y., Fujishiro N., Sakamoto Y., Inoue M. Subunit composition and role of Na⁺,K⁺-ATPases in ventricular myocytes. J Phys Sci (2006) 56:113–121.[CrossRef]
30. Brette F., Komukai K., Orchard C.H. Validation of formamide as a detubulation agent in isolated rat cardiac cells. Am J Physiol (2002) 283:H1720–H1728.[ISI]
31. Page E. Quantitative ultrastructural analysis in cardiac membrane physiology. Am J Physiol (1978) 235:C147–C158.[ISI][Medline]
32. Ishizuka N., Fielding A.J., Berlin J.R. Na pump current can be separated into ouabain-sensitive and -insensitive components in single rat ventricular myocytes. Jpn J Physiol (1996) 46:215–223.[CrossRef][ISI][Medline]
33. Erdmann E., Werdan K., Brown L. Multiplicity of cardiac glycoside receptors in the heart. Trends Pharmacol Sci (1985) 6:293–295.[CrossRef]
34. Wier W.G., Hess P. Excitation–contraction coupling in cardiac Purkinje fibers. Effects of cardiotonic steroids on the intracellular Ca²⁺ transient, membrane potential, and contraction. J Gen Physiol (1984) 83:395–415.[Abstract/Free Full Text]
35. Sagawa T., Sagawa K., Kelly J.E., Tsushima R.G., Wasserstrom J.A. Activation of cardiac ryanodine receptors by cardiac glycosides. Am J Physiol (2002) 282:H1118–H1126.[ISI]
36. Wasserstrom J.A., Aistrup G.L. Digitalis: new actions for an old drug. Am J Physiol (2005) 289:H1781–H1793.[ISI]
37. Eisner D.A., Diaz M.E., Neill S.C., Trafford A.W. Physiological and pathological modulation of ryanodine receptor function in cardiac muscle. Cell Calcium (2004) 35:583–589.[CrossRef][ISI][Medline]
38. Altamirano J., Li Y., DeSantiago J., Piacentino V. III, Houser S.R., Bers D.M. The inotropic effect of cardioactive glycosides in ventricular myocytes requires Na⁺/Ca²⁺ exchanger function. J Physiol (2006) 575:845–854.[Abstract/Free Full Text]
39. Dostanic I., Lorenz J.N., Schultz J.J., Grupp I.L., Neumann J.C., Wani M.A., et al. The ₂ isoform of Na,K-ATPase mediates ouabain-induced cardiac inotropy in mice. J Biol Chem (2003) 278:53026–53034.[Abstract/Free Full Text]
40. Yang Z., Pascarel C., Steele D.S., Komukai K., Brette F., Orchard C.H. Na⁺/Ca²⁺-exchange activity is localized in the T-tubules of rat ventricular myocytes. Circ Res (2002) 91:315–322.[Abstract/Free Full Text]
41. Lines G.T., Sande J.B., Louch W.E., Mørk H.K., Grøttum P., Sejersted O.M. Contribution of the Na–Ca exchanger to rapid Ca release in cardiomyocytes. Biophys J (2006) 91:779–792.[Abstract/Free Full Text]
42. Louch W.E., Mørk H.K., Sexton J., Strømme T.A., Laake P., Sjaastad I., et al. T-tubule disorganization and reduced synchrony of Ca²⁺ release in murine cardiomyocytes following myocardial infarction. J Physiol (2006) 574:519–533.[Abstract/Free Full Text]
43. Pogwizd S.M., Schlotthauer K., Li L., Yuan W., Bers D.M. Arrhythmogenesis and contractile dysfunction in heart failure: roles of sodium–calcium exchange, inward rectifier potassium current, and residual ß-adrenergic responsiveness. Circ Res (2001) 88:1159–1167.[Abstract/Free Full Text]
44. Hobai I.A., O'Rourke B. Decreased sarcoplasmic reticulum calcium content is responsible for defective excitation–contraction coupling in canine heart failure. Circulation (2001) 103:1577–1584.[Abstract/Free Full Text]
45. Lindner M., Erdmann E., Beuckelmann D.J. Calcium content of the sarcoplasmic reticulum in isolated ventricular myocytes from patients with terminal heart failure. J Mol Cell Cardiol (1998) 30:743–749.[CrossRef][ISI][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				F. Swift, J. A. K. Birkeland, N. Tovsrud, U. H. Enger, J. M. Aronsen, W. E. Louch, I. Sjaastad, and O. M. Sejersted Altered Na+/Ca2+-exchanger activity due to downregulation of Na+/K+-ATPase {alpha}2-isoform in heart failure Cardiovasc Res, April 1, 2008; 78(1): 71 - 78. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Google Scholar Articles by Swift, F. Articles by Sejersted, O. M. Search for Related Content PubMed PubMed Citation Articles by Swift, F. Articles by Sejersted, O. M. Social Bookmarking