TY - JOUR
AU - Fiolet,, J.W.T
AB - Abstract Objective: Cytosolic sodium ([Na+]i) is increased in heart failure (HF). We hypothesize that up-regulation of Na+/H+-exchanger (NHE) in heart failure is causal to the increase of [Na+]i and underlies disturbance of cytosolic calcium ([Ca2+]i) handling. Methods: Heart failure was induced in rabbits by combined volume and pressure overload. Age-matched animals served as control. [Na+]i, cytosolic calcium [Ca2+]i and cytosolic pH (pHi) were measured in isolated left ventricular midmural myocytes with SBFI, indo-1 and SNARF. SR calcium content was measured as the response of [Ca2+]i to rapid cooling (RC). Calcium after-transients were elicited by cessation of rapid stimulation (3 Hz) in the presence of 100 nmol/l noradrenalin. NHE and Na+/K+-ATPase activity were inhibited with 10 μmol/l cariporide and 100 μmol/l ouabain, respectively. Results: At all stimulation rates (0–3 Hz) [Na+]i and diastolic [Ca2+]i were significantly higher in HF than in control. With increasing frequency [Na+]i and diastolic [Ca2+]i progressively increased in HF and control, and the calcium transient amplitude (measured as total calcium released from SR) decreased in HF and increased in control. In HF (at 2 Hz), SR calcium content was reduced by 40% and the calcium gradient across the SR membrane by 60%. Fractional systolic SR calcium release was 90% in HF and 60% in control. In HF the rate of pHi recovery following acid loading was much faster at all pHi and NHE dependent sodium influx was almost twice as high as in control. In HF cariporide (10 μmol/l, 5 min) reduced [Na+]i and end diastolic [Ca2+]i to almost control values, and reversed the relation between calcium transient amplitude and stimulation rate from negative to positive. It increased SR calcium content and SR membrane gradient and decreased fractional systolic SR depletion to 60%. Cariporide greatly reduced the susceptibility to develop calcium after-transients. In control animals, cariporide had only minor effects on all these parameters. Increase of [Na+]i with ouabain in control myocytes induced abnormal calcium handling as found in HF. Conclusions: In HF up-regulation of NHE activity is causal to increased [Na+]i and secondarily to disturbed diastolic, systolic and SR calcium handling. Specific inhibition of NHE partly normalized [Na+]i, end diastolic [Ca2+]i, and SR calcium handling and reduced the incidence of calcium after-transients. Chronic treatment with specific NHE inhibitors may provide a useful future therapeutic option in treatment of developing hypertrophy and heart failure. Calcium (cellular), Heart failure, Myocytes, Na/H-exchanger Time for primary review 22 days. 1 Introduction Despite major efforts and growing knowledge on mechanisms underlying development of heart failure (HF), the particular processes involved in the transition from compensated hypertrophy to HF and the genesis of associated arrhythmias are still not entirely understood. Nevertheless, abundant evidence indicates that disturbed calcium handling in HF is instrumental to contractile dysfunction and contributes to the genesis of ventricular arrhythmias. Disturbed calcium handling is evidenced by increased end diastolic and decreased systolic calcium ([Ca2+]i) [1,2], up-regulation of the Na+/Ca2+-exchanger (NCX) (reviewed in Ref. [3]), down regulation of sarcoplasmic reticulum Ca2+-ATPase (SERCA2) [4], increased open probability of the SR calcium release channels (RyR) and a negative force frequency relationship [5]. So far, it has not been established whether disturbance of calcium handling is an adaptive process secondary to other stimuli to cope with altered contractile demands. Intracellular sodium ([Na+]i) is increased in HF [6,7]. From the perspective of contractile adaptation, an increase of [Na+]i would favor positive inotropic purposes through a NCX mediated increase of end diastolic [Ca2+]i[6]. An increase in workload and mechanical stretch is a major stimulus for cardiac hypertrophy and for activation of an autocrine/paracrine cascade involving the release of angiotension-II and endothelin [7], which cause increased NHE activity [8]. Mechanical stretch is associated with a rise in [Na+]i, which can be prevented by inhibition of the NHE [9]. It also has been reported that NHE mRNA and protein levels are increased in transgenic mice over-expressing beta-adrenergic receptors that develop hypertrophy [10] and of mRNA during development of HF in rabbit [11]. Increased [Na+]i may be related to increased activity and/or up-regulation of NHE. The increased end diastolic [Ca2+]i secondary to enhanced [Na+]i positively affects the open probability of RyR [12]. Increased open probability of RyR may lead to spontaneous calcium release from SR, initiating calcium after-transients and delayed after-depolarizations (DADs), easily inducible in human and experimental failing hearts [13]. NHE may be involved in hypertrophic remodeling in and development of HF related to cellular alkalinization, which together with increased [Ca2+]i affects protein synthesis. An increase of pHi as small as 0.1 units accelerates protein synthesis by 40% [14], while elevated [Ca2+]i activates mitogen activated protein kinases in cardiac cells [15]. Inhibition of NHE reduces cellular pH [16], attenuates activation of mitogen-activated proteinkinase by mechanical stretch [17] and favorably affects development of hypertrophy [10,18,19]. In transgenic mice, in which NHE mRNA and protein is up-regulated, inhibition of NHE prevented hypertrophy [10]. Inhibition of NHE also attenuated the degree of reperfusion damage and apoptosis following ischemia [20,21]. Clinically, short-term (7 days) treatment of early post infarction patients with cariporide was evaluated in the Guardian study [22] and found to reduce mortality rate. This study aims to explore whether and to what extent NHE mediated increase of [Na+]i contributes to disturbed calcium handling and whether specific inhibition of NHE with cariporide can normalize cellular calcium handling in myocytes isolated from hearts of rabbits with pressure and volume overload induced HF. 2 Methods Animal care and handling conformed to the Guide for Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and the study was approved by the local ethical committee. 2.1 The rabbit model of pressure and volume overload induced heart failure and isolation of left ventricular myocytes HF was induced in rabbits (New Zealand White, SPF, 3–3.5 kg) by combined volume and pressure overload as described previously [13,23]. Volume overload was produced by rupture of the aortic valve until pulse pressure increased by 100%. Three weeks after the first operation, pressure overload was applied by suprarenal abdominal aortic stenosis of approximately 50%. Twelve weeks after the second operation hearts were isolated and heart failure was assessed based on relative heart and lung weight, left ventricular and diastolic pressure, the presence of ascites and cell dimensions as described previously [23,24]. Age matched untreated animals served as controls. Table 1 summarizes the parameters relevant to the development of heart failure in this model. Table 1 Animal characteristics . Control . HF . Bodyweight (kg) 4.4±0.26 4.3±0.09 Relative heart weight (ww/kg 10−2) 2.4±0.09 5.2±0.23* Relative lung weight (ww/kg 10−2) 2.5±0.09 4.6±0.3* LVEDP (mmHg) 3±0.4 18±1.8* Ascites (n/n) 0/12 10/20* Cell width (μm) 28±1.9 36±0.7* Cell length (μm) 144±2.2 199±4.0* . Control . HF . Bodyweight (kg) 4.4±0.26 4.3±0.09 Relative heart weight (ww/kg 10−2) 2.4±0.09 5.2±0.23* Relative lung weight (ww/kg 10−2) 2.5±0.09 4.6±0.3* LVEDP (mmHg) 3±0.4 18±1.8* Ascites (n/n) 0/12 10/20* Cell width (μm) 28±1.9 36±0.7* Cell length (μm) 144±2.2 199±4.0* ww=wet weight, LVEDP=left ventricular end diastolic pressure. Number of animals: control n=12, HF n=20. Data expressed as mean±S.E.M. * P<0.01 versus control. Open in new tab Table 1 Animal characteristics . Control . HF . Bodyweight (kg) 4.4±0.26 4.3±0.09 Relative heart weight (ww/kg 10−2) 2.4±0.09 5.2±0.23* Relative lung weight (ww/kg 10−2) 2.5±0.09 4.6±0.3* LVEDP (mmHg) 3±0.4 18±1.8* Ascites (n/n) 0/12 10/20* Cell width (μm) 28±1.9 36±0.7* Cell length (μm) 144±2.2 199±4.0* . Control . HF . Bodyweight (kg) 4.4±0.26 4.3±0.09 Relative heart weight (ww/kg 10−2) 2.4±0.09 5.2±0.23* Relative lung weight (ww/kg 10−2) 2.5±0.09 4.6±0.3* LVEDP (mmHg) 3±0.4 18±1.8* Ascites (n/n) 0/12 10/20* Cell width (μm) 28±1.9 36±0.7* Cell length (μm) 144±2.2 199±4.0* ww=wet weight, LVEDP=left ventricular end diastolic pressure. Number of animals: control n=12, HF n=20. Data expressed as mean±S.E.M. * P<0.01 versus control. Open in new tab Midmural left ventricular myocytes were isolated as described previously [25]. Myocytes were stored at room temperature in separate vials, each containing about 105 myocytes in 5 ml solution containing (mmol/l): [Na+] 156, [K+] 4.7, [Ca2+] 2.6, [Mg2+] 2.0, [Cl−] 150.6, [HCO3−] 4.3, [HPO42−] 1.4, [HEPES] 17, [Glucose] 11 supplied with 1% fatty acid free albumin (pH 7.3). 2.2 Measurement of cytosolic [Na+]i, [Ca2+]i, pHi, SR calcium content and ΔGATP [Na+]i, [Ca2+]i and pHi were measured in parallel experiments with SBFI, indo-1 and SNARF. Myocytes were loaded during 120 min with 10 μmol/l SBFI-AM and 0.01% pluronic, or during 30 min with 5 μmol/l indo-1/AM, or during 10 min with 10 μmol/l SNARF-AM, washed twice with fresh HEPES solution (without albumin) and kept for another 15 min to ensure complete de-esterification. Myocytes were attached to a polylysine (0.1 g/l) treated cover slip on the stage of a fluorescence microscope (Nikon Diaphot). A cell chamber (height 0.4 mm, diameter 10 mm, volume 30 μl), having two needles at opposite sides for superfusion purposes and two parallel platinum electrodes at 8 mm distance for field stimulation (40 V/cm bipolar square pulses of 0.5 ms duration), was tightly pressed onto the cover slip. The microscope stage and the perfusion chamber were temperature controlled at 37 °C. The contents of the chamber could be replaced within 100 ms. The measuring window was adjusted to the cellular surface of one quiescent rod-shaped myocyte with a rectangular diaphragm. Before collecting data, each myocyte was conditioned by 2 min stimulation at 2 Hz. Stimulation rate dependence (0–3 Hz) of steady state [Na+]i and [Ca2+]i was measured after 2 min of conditioning at each frequency. For SBFI, indo-1 and SNARF excitation wavelengths were 340, 340 and 515 nm, respectively. Fluorescence was recorded in dual wavelength emission mode at 410/590, 410/516 and 580/640 nm, respectively. Signals were digitized at 1 kHz, corrected for background signals recorded in probe-free myocytes and stored for later analysis [26,27]. SR calcium content was measured as the response of [Ca2+]i to rapid cooling (within 200 ms) with ice-cold HEPES solution of the same composition. Calcium after-transients (3 myocytes per rabbit) were elicited by cessation of stimulation following 10 s of rapid pacing (3 Hz) in the presence of 100 nmol/l noradrenaline. Free energy of ATP hydrolysis, the phosphorylation potential (ΔGATP), was calculated from the creatinephosphate, creatine and inorganic phosphate content measured in field stimulated (2 Hz) cell suspensions containing about 3 mg of myocyte as described previously [25]. 2.3 Calculation of [Na+]i, cytosolic free [Ca2+]i, cytosolic bound [Ca2+]i, SR calcium content and fractional SR calcium release SBFI fluorescence was measured in dual wavelength emission mode and [Na+]i was calculated using previously obtained values for Rmax, Rmin, β and kd; dual wavelength emission mode provides a more sensitive and sodium specific measurement of [Na+]i than the dual excitation technique [26]. Because [Na+]i does not change on a beat to beat basis calculated [Na+]i data were averaged over the entire cardiac cycle. Indo-1 signals were corrected for background signals and calculations were performed using previously established values for Rmax, Rmin, β and kd (see Fig. 1 and Ref. [27]). In order to calculate cytosolic free [Ca2+]i, a second correction was applied on fluorescence data, which allows for mitochondrially compartmentalized indo-1 signals [27] and stimulation rate dependence of mitochondrial calcium [28]. Fig. 1 Open in new tabDownload slide Representative examples of calcium transients and response to rapid cooling recorded in a control and a HF myocyte before and after 5 min of cariporide treatment. Fig. 1 Open in new tabDownload slide Representative examples of calcium transients and response to rapid cooling recorded in a control and a HF myocyte before and after 5 min of cariporide treatment. Fig. 1 shows representative cytosolic free [Ca2+]i data of calcium transients and rapid cooling response recorded in a control and a HF myocyte. From cytosolic free [Ca2+]i and literature data on cytosolic calcium buffer capacity, total cytosolic bound [Ca2+]i was calculated as described previously [27]. It should be stressed, that consequent to calcium buffering characteristics, the relationship between free [Ca2+]i and bound [Ca2+]i is non-linear and that the actual value of a change of bound [Ca2+]i associated with a particular change in free [Ca2+]i strongly depends on the magnitude of diastolic free [Ca2+]i. For example: with higher diastolic [Ca2+]i in HF than in control and the same calcium transient amplitude expressed as free [Ca2+]i, the amplitude expressed as bound [Ca2+]i is smaller in HF than in control. In general, at a particular calcium transient amplitude measured as free [Ca2+]i, the associated change expressed as bound [Ca2+]i progressively decreases with increasing diastolic [Ca2+]i. SR calcium content was calculated from the difference between diastolic bound [Ca2+]i and the rapid cooling induced increase of cytosolic bound [Ca2+]i and a fractional SR volume of 10% [29]. Fractional SR calcium release was calculated as the ratio between the calcium transient amplitude expressed as cytosolic bound [Ca2+]i and SR calcium content. 2.4 Statistics Data are expressed as mean±S.E.M. Values for myocytes of individual hearts were averaged. Two-way ANOVA (with a post-hoc test according to Student's–Newman–Keuls) or Student's t-test was used to test for statistical significance where appropriate at a level of significance of P<0.05. 3 Results Fig. 2 shows the stimulation rate dependence of cytosolic sodium ([Na+]i) in myocytes of HF and control rabbits. At all frequencies [Na+]i was significantly higher in HF than in control (P<0.01). Also the increase of [Na+]i with stimulation rate (0–3 Hz) was significantly larger in HF than in control (P<0.01). At thermodynamic equilibrium of the Na+/K+-ATPase an upper limit is set to the trans-sarcolemmal electrochemical potential difference of sodium equal to one third of the phosphorylation potential (ΔGATP) and, consequently, a lower limit to [Na+]i. At 2 Hz the phosphorylation potential (ΔGATP) was 54.2±1.8 kJ/mol in HF and 55.8±1.1 kJ/mol in control. The concentrations of cytosolic sodium corresponding to equilibrium, are 4.8 and 3.9 mmol/l, respectively. The deviation of measured [Na+]i from equilibrium was significantly larger in HF than in control (P<0.01). Fig. 2 Open in new tabDownload slide Stimulation rate dependence of [Na+]i, Data (mean±S.E.M.) for eight HF animals (closed symbols) and six control animals (open symbols). Data of three myocytes per animal were pooled. Equilibrium concentrations of cytosolic sodium at 2 Hz were 3.9 and 4.8 mmol/l, respectively, calculated from equilibrium with the phosphorylation potential (ΔGATP) of 55.8±1.1 kJ/mol in control (n=27 cell suspensions, nine rabbits) and 54.2±1.8 kJ/mol in HF (n=21 cell suspensions, seven rabbits). The magnitude of the arrows indicate the deviation from equilibrium of measured [Na+]i. * P<0.01 versus control at the same stimulation frequency, †P<0.05 versus HF. Fig. 2 Open in new tabDownload slide Stimulation rate dependence of [Na+]i, Data (mean±S.E.M.) for eight HF animals (closed symbols) and six control animals (open symbols). Data of three myocytes per animal were pooled. Equilibrium concentrations of cytosolic sodium at 2 Hz were 3.9 and 4.8 mmol/l, respectively, calculated from equilibrium with the phosphorylation potential (ΔGATP) of 55.8±1.1 kJ/mol in control (n=27 cell suspensions, nine rabbits) and 54.2±1.8 kJ/mol in HF (n=21 cell suspensions, seven rabbits). The magnitude of the arrows indicate the deviation from equilibrium of measured [Na+]i. * P<0.01 versus control at the same stimulation frequency, †P<0.05 versus HF. Deviation from equilibrium reflects non-zero sodium influx. Apart from Na-channel related sodium influx during the action potential a quantitatively important contribution to sodium influx may be due to Na+/H+-exchanger (NHE) activity. Fig. 3 shows the effect of 10 μmol/l cariporide, a specific inhibitor of NHE, on [Na+]i (top panel), end diastolic [Ca2+]i (middle panel) and calcium transient amplitude (bottom panel). In HF myocytes, 5 min of cariporide treatment reduced [Na+]i, much more than in control myocytes, although the difference with control values remained significant. In a few experiments the HEPES buffered solutions was replaced by HCO3-buffered (25 mmol/l) solution. This caused a small increase of [Na+]i, but the effect of cariporide was the same as in HEPES buffered solution (data not shown). Fig. 3 Open in new tabDownload slide The effect of cariporide on [Na+]i (upper panel), end diastolic [Ca2+]i (middle panel) and calcium transient amplitude (lower panel) in HF (closed symbols) and control (open symbols) rabbits. Stimulation rate was 2 Hz. Concentration of cariporide was 10 μmol/l. Data for [Na+]i, (mean±S.E.M.) for six HF and five control rabbits. Data for [Ca2+]i (mean±S.E.M.) for six HF and control rabbits. Data of three myocytes per animal were pooled. * P<0.01 versus control at time zero, †P<0.01 versus HF at time zero, ‡P<0.01 versus control at 5 min. Fig. 3 Open in new tabDownload slide The effect of cariporide on [Na+]i (upper panel), end diastolic [Ca2+]i (middle panel) and calcium transient amplitude (lower panel) in HF (closed symbols) and control (open symbols) rabbits. Stimulation rate was 2 Hz. Concentration of cariporide was 10 μmol/l. Data for [Na+]i, (mean±S.E.M.) for six HF and five control rabbits. Data for [Ca2+]i (mean±S.E.M.) for six HF and control rabbits. Data of three myocytes per animal were pooled. * P<0.01 versus control at time zero, †P<0.01 versus HF at time zero, ‡P<0.01 versus control at 5 min. Maintenance of end diastolic [Ca2+]i depends on the Na+/Ca2+-exchanger (NCX), the forward mode driving force of which greatly depends on [Na+]i. With increasing [Na+]i the driving force of NCX decreases and end diastolic [Ca2+]i increases. End diastolic [Ca2+]i was significantly higher in HF than in control. Cariporide greatly reduced the difference, but not entirely to control values; after 5 min the difference between HF and control was still significant. Diastolic [Ca2+]i affects SR calcium handling. The lower panel shows the effect of cariporide on the calcium transient amplitude. The amplitude was significantly smaller in HF than in control. Five min treatment with cariporide caused a small but non-significant decrease of amplitude in control (P=0.19), but in HF it caused a significantly larger decrease of the calcium transient amplitude. Fig. 4 evaluates the contribution of NHE to steady state sodium influx in myocytes stimulated at 2 Hz. Immediately following inhibition of the Na+/K+-ATPase with 100 μmol/l ouabain (open symbols), [Na+]i linearly increased during 10 min. The rate of increase was significantly faster in HF than in control, 2.2 versus 1.5 mmol/l/min, respectively (P<0.05). The presence of cariporide, reduced the rate of increase of [Na+]i in HF and control to 1.0 and 0.8 mmol/l/min, respectively (NS). This indicates that NHE dependent sodium influx is larger in HF than in control. The rates of sodium influx measured with combined inhibition of NHE and Na+/K+-ATPase can be attributed at least for a large part to Na-channel related sodium influx. This is supported by the observation that [Na+]i remained constant in quiescent myocytes (data not shown). A realistic calculated estimate of Na-channel related sodium influx is about 6 μmol/l/beat, which corresponds to about 0.75 mmol/l/min at 2 Hz (μmol/l/beat=F*C*dVm/Volcell, with Faraday constant F=105, cell capacitance C=150 pF, action potential amplitude dVm=0.1 V and cell volume Volcell=2.5×10−11 l). Fig. 4 Open in new tabDownload slide Time course of increase of [Na+]i (upper panels) and diastolic [Ca2+]i (lower panels) after inhibition of Na+/K+-ATPase with 100 μmol/l ouabain in the absence (open circles) or presence (closed circles) of 10 μmol/l cariporide (pre-incubated during 5 min). The stimulation rate was 2 Hz. Data (mean±S.E.M.) for seven HF rabbits (right panels) and five control rabbits (left panels). Data of three myocytes per animal were pooled. Fig. 4 Open in new tabDownload slide Time course of increase of [Na+]i (upper panels) and diastolic [Ca2+]i (lower panels) after inhibition of Na+/K+-ATPase with 100 μmol/l ouabain in the absence (open circles) or presence (closed circles) of 10 μmol/l cariporide (pre-incubated during 5 min). The stimulation rate was 2 Hz. Data (mean±S.E.M.) for seven HF rabbits (right panels) and five control rabbits (left panels). Data of three myocytes per animal were pooled. The lower panels of Fig. 4 show the corresponding changes of end diastolic [Ca2+]i. In parallel with changes of [Na+]i, end diastolic [Ca2+]i increased more in HF than in control after inhibition of the Na+/K+-ATPase. With combined inhibition of NHE and Na+/K+-ATPase the rate of increase of end diastolic [Ca2+]i in HF was reduced to almost control. Fig. 5 directly demonstrates that the proton flux through NHE (JNHE), measured from the rate of recovery of pHi following acid loading, is substantially larger in HF than in control. Fig. 5 Open in new tabDownload slide Rate of proton flux through NHE (JNHE) calculated from the rate of recovery of pHi following acid loading with an ammonium pre-pulse in non-stimulated control (open symbols) and HF (closed symbols) myocytes. The insert shows a representative example of raw data. Data for pHi recovery rates (mean±S.E.M.) six myocytes from three HF and three control rabbits. Fig. 5 Open in new tabDownload slide Rate of proton flux through NHE (JNHE) calculated from the rate of recovery of pHi following acid loading with an ammonium pre-pulse in non-stimulated control (open symbols) and HF (closed symbols) myocytes. The insert shows a representative example of raw data. Data for pHi recovery rates (mean±S.E.M.) six myocytes from three HF and three control rabbits. Fig. 6 shows the effect of inhibition of NHE on SR calcium content (upper panel), the magnitude of the calcium gradient across the SR membrane during end diastole (middle panel) and the maximal fractional calcium release during systole (lower panel). SR calcium content and the ratio of SR calcium content and end-diastolic [Ca2+]i (the SR calcium gradient) were significantly smaller in HF than in control (P<0.001). Maximal systolic fractional release was significantly higher in HF (P<0.001). Cariporide did not affect either of these parameters in normal myocytes. In HF cariporide increased SR calcium content (P<0.001), increased the diastolic SR calcium gradient (P<0.001) and reduced maximal systolic fractional release (P<0.01). Fig. 6 Open in new tabDownload slide SR calcium content (upper panel), calcium gradient across the SR membrane (middle panel) and fractional depletion of SR during systole (bottom panel). SR calcium content was measured by rapid cooling. SR gradient as the ratio of SR calcium content and diastolic [Ca2+]i and fractional SR depletion as the ratio between the systolic and diastolic SR calcium content. The stimulation rate was 2 Hz. Data (mean±S.E.M.) for 12 HF rabbits and seven control animals. Data of three myocytes per animal were pooled. * P<0.001 versus control, †P<0.001 versus HF, ‡P<0.01 versus control+cariporide. Fig. 6 Open in new tabDownload slide SR calcium content (upper panel), calcium gradient across the SR membrane (middle panel) and fractional depletion of SR during systole (bottom panel). SR calcium content was measured by rapid cooling. SR gradient as the ratio of SR calcium content and diastolic [Ca2+]i and fractional SR depletion as the ratio between the systolic and diastolic SR calcium content. The stimulation rate was 2 Hz. Data (mean±S.E.M.) for 12 HF rabbits and seven control animals. Data of three myocytes per animal were pooled. * P<0.001 versus control, †P<0.001 versus HF, ‡P<0.01 versus control+cariporide. Fig. 7 shows the stimulation rate dependence of the calcium transient amplitude in the presence and absence of cariporide expressed as cytosolic bound [Ca2+]i. With increasing stimulation rate, the amplitude progressively increased in control and decreased in HF. In HF, cariporide restored a positive relationship between stimulation rate and amplitude. Cariporide hardly affected the calcium transient amplitude in control. Fig. 7 Open in new tabDownload slide Stimulation rate dependence of the calcium transient amplitude in the absence (circles) or presence (triangles) of 10 μmol/l cariporide. Amplitudes are expressed as total calcium released from SR into the cytosol, calculated from the transient amplitudes of free cytosolic [Ca2+]i and calcium buffer characteristics (see Section 2). Measurements were started after 5 min of pre-incubation with cariporide. Data (mean±S.E.M.) for seven HF rabbits (closed symbols) and five control animals (open symbols). Data of at least three myocytes per animal were pooled. * P<0.01 versus control, †P<0.01 versus HF. Fig. 7 Open in new tabDownload slide Stimulation rate dependence of the calcium transient amplitude in the absence (circles) or presence (triangles) of 10 μmol/l cariporide. Amplitudes are expressed as total calcium released from SR into the cytosol, calculated from the transient amplitudes of free cytosolic [Ca2+]i and calcium buffer characteristics (see Section 2). Measurements were started after 5 min of pre-incubation with cariporide. Data (mean±S.E.M.) for seven HF rabbits (closed symbols) and five control animals (open symbols). Data of at least three myocytes per animal were pooled. * P<0.01 versus control, †P<0.01 versus HF. Fig. 8 shows a representative example of the susceptibility of HF myocytes to develop calcium after-transients upon cessation of rapid pacing in the presence of noradrenalin (100 nmol/l), before (upper panel) during (middle panel) and after washout of cariporide (bottom panel). Cariporide reversibly suppressed development of the after-transient. After-transients were never observed in control myocytes in the same conditions. Table 2 summarizes the effect of cariporide on the incidence of calcium after-transients in HF myocytes. Without cariporide, in all HF myocytes (18 myocytes, nine rabbits) calcium after-transients were inducible. In 15 of these myocytes, only a single after-transient occurred (amplitude 16.1±2.0% relative to the last stimulated beat) and in three myocytes a train of more than one after-transient occurred (relative amplitude 79±2.6%). Cariporide treatment significantly reduced the incidence to a single after-transient in 2 of 18 myocytes (relative amplitude 6.1±0.6%); these two myocytes exhibited a train of after-transients before cariporide treatment. Fig. 8 Open in new tabDownload slide The effect of cariporide on calcium after-transients in a HF myocyte. Before cariporide (upper panel), after 5 min cariporide treatment (middle panel) and after washout (bottom panel). Calcium after-transients were elicited by cessation of stimulation following 10 s of rapid pacing (3 Hz) in the presence of 100 nmol/l noradrenaline. Fig. 8 Open in new tabDownload slide The effect of cariporide on calcium after-transients in a HF myocyte. Before cariporide (upper panel), after 5 min cariporide treatment (middle panel) and after washout (bottom panel). Calcium after-transients were elicited by cessation of stimulation following 10 s of rapid pacing (3 Hz) in the presence of 100 nmol/l noradrenaline. Table 2 Calcium after-transients . − Cariporide . + Cariporide . Incidence (n/n) 18/18 2/18* Relative amplitude (%) 30 9 end diastolic [Ca2+]i (nmol/l) 150±7.5 101±3.8* . − Cariporide . + Cariporide . Incidence (n/n) 18/18 2/18* Relative amplitude (%) 30 9 end diastolic [Ca2+]i (nmol/l) 150±7.5 101±3.8* Data expressed as mean±S.E.M. * P<0.05 versus control. Open in new tab Table 2 Calcium after-transients . − Cariporide . + Cariporide . Incidence (n/n) 18/18 2/18* Relative amplitude (%) 30 9 end diastolic [Ca2+]i (nmol/l) 150±7.5 101±3.8* . − Cariporide . + Cariporide . Incidence (n/n) 18/18 2/18* Relative amplitude (%) 30 9 end diastolic [Ca2+]i (nmol/l) 150±7.5 101±3.8* Data expressed as mean±S.E.M. * P<0.05 versus control. Open in new tab With ouabain [Na+]i increased in control myocytes (Fig. 4). Fig. 9 shows an example of calcium transients and the response of [Ca2+]i to rapid cooling in a control myocyte before and after 5 min ouabain treatment. In four control myocytes ouabain (mean±S.E.M., n=4): increased diastolic [Ca2+]i by 34±8.6%, decreased calcium transient amplitude by 28±11.1%, decreased SR calcium content by 46±11.6% and increased fractional SR calcium release from 63 to 90%. Fig. 9 Open in new tabDownload slide A representative example of calcium transients and response of [Ca2+]i to rapid cooling in a control myocyte after 5 min of inhibition of Na+/K+-ATPase with 100 μmol/l ouabain. The upper panels show data expressed as cytosolic free [Ca2+]i and the lower panels show the corresponding total cytosolic bound [Ca2+]i, from which SR calcium content and fractional SR release can be calculated (see Section 2). Fig. 9 Open in new tabDownload slide A representative example of calcium transients and response of [Ca2+]i to rapid cooling in a control myocyte after 5 min of inhibition of Na+/K+-ATPase with 100 μmol/l ouabain. The upper panels show data expressed as cytosolic free [Ca2+]i and the lower panels show the corresponding total cytosolic bound [Ca2+]i, from which SR calcium content and fractional SR release can be calculated (see Section 2). 4 Discussion We demonstrated that [Na+]i is increased in HF at all stimulation frequencies compared to control. The deviation of [Na+]i from thermodynamic equilibrium was larger in HF than in control due to enhanced NHE activity. Elevation of [Na+]i in HF was associated with (1) increased end diastolic [Ca2+]i, (2) decreased SR calcium content, (3) decreased SR–membrane calcium gradient, (4) increased fractional systolic calcium release from SR, and (5) a high incidence of calcium after-transients compared to normal myocytes. The relation between the calcium transient amplitude and stimulation frequency was positive in control and negative in HF. Cariporide, a widely recognized highly specific inhibitor of NHE, partly normalized all these parameters to control values. In control myocytes increase of [Na+]i after application of ouabain induced HF like characteristics with respect to cytosolic and SR calcium handling. We used a well characterized model of combined volume and pressure overload induced HF in rabbits as previously described [13–23]. Heart failure reproducibly (Table 1) develops in this model within 2–3 months with all features of failure (hypertrophy, dilatation, dyspnoe, decreased diastolic function and arrhythmogenesis (ventricular premature beats and tachycardia) [24,29]. Elevated [Na+]i has been reported in hypertrophy [30–33]. Data on [Na+]i in heart failure are limited. Very recently Despa et al. [34], using the same animal model, reported increased [Na+]i in HF, which in non-stimulated myocytes was mainly TTX sensitive and only little NHE dependent. In quiescent myocytes metabolic rate and associated NHE fluxes are considerably less than in stimulated conditions, but the TTX sensitivity observed at rest is quite intriguing. Contrast to their findings, in stimulated conditions we could not demonstrate a large difference in Na-channel related sodium influx between HF and control. The substantially increased cariporide sensitive sodium influx in HF rather indicates up-regulated NHE dependent sodium influx (Figs. 3 and 4). In support of this we demonstrate a substantially increased recovery rate of pHi in non-stimulated HF myocytes following acid loading (Fig. 5) [35], which corroborates literature data on increased mRNA and/or protein expression of NHE in hypertrophy [10] and HF [11] and increased NHE activity (but unchanged protein expression) in human HF [36]. Interestingly, attenuation of post-infarction hypertrophy in rats by chronic administration of cariporide suggests a role for enhanced NHE activity also in the remodelling process [19]. Other causes for increased [Na+]i in HF: (1) Reduced sodium efflux by depressed Na+/K+-ATPase activity. Although literature data indicate reduced expression of Na+/K+-ATPase in HF (for review, see Ref. [37]) and hypertrophy in Ref. [38], Despa et al. [34] did not find altered Na+/K+-pump activity. From steady state considerations it follows that Na+/K+-ATPase activity in HF, even if down-regulated molecularly, is capable to balance increased sodium influx. In order to achieve steady state, increased [Na+]i could kinetically compensate for molecular down-regulation of Na+/K+-ATPase. In this respect it is of interest that [Na+]i in HF did not entirely return to control values after 5 min of cariporide treatment (Fig. 3). (2) Increased Na+/Ca2+-exchanger (NCX) mediated influx. Whether or not NCX is up-regulated in HF (for review, see Ref. [3]), and regardless of the fact that its forward driving force is reduced by elevated [Na+]i, steady state conditions require that NCX calcium efflux exactly balances L-type Ca-channel related influx during the cardiac cycle. Therefore, with unchanged L-type Ca-channel current the NCX carried sodium influx must be the same in HF and control. (3) Because our experimental HEPES-buffered solutions contained some HCO3−, we cannot entirely exclude some contribution of the Na+–HCO3− co-transporter to sodium influx. However, the effect of cariporide was not influenced even at high [HCO3−]. End diastolic [Ca2+]i is coupled to [Na+]i via NCX. With increasing [Na+]i, ΔGexch diminishes and [Ca2+]i increases. The opposite occurs with reduction of [Na+]i with cariporide in HF. Steady state diastolic SR calcium depends on diastolic [Ca2+]i and the efficiency and driving force of SERCA. This driving force is the free energy of ATP hydrolysis (ΔGATP). At 2 HZ stimulation rate, ΔGATP in HF was 1.6 kJ/mol less than in normal cells (Fig. 1). Assuming similar efficiency of SERCA in HF and normal cells, it follows from thermodynamic calculations that the magnitude of the calcium gradient across the SR membrane would be reduced by 25% in HF, which in combination with a 60% higher diastolic [Ca2+]i (Fig. 3) would cause a 10% increase of SR calcium content in HF relative to control. In fact, we measured a 40% reduction of SR calcium content and a 60% reduction of the gradient in HF (Fig. 6). Therefore, the overall efficiency of the SR calcium uptake process is decreased in HF, either due to down-regulated SERCA, which indeed has been amply documented [4] and/or increased leak via calcium release channels (RyR) [39,40]. The latter mechanism agrees with our observation that the fractional systolic SR calcium depletion was substantially higher in HF (Fig. 6). Cariporide partly restored SR calcium content and gradient, but also reduced fractional systolic SR calcium depletion to control values. This strongly suggests that the reduction of diastolic [Ca2+]i is instrumental to this. Indeed it has been found that open probability of RyR and the incidence of spontaneous calcium sparks is rather sensitive to [Ca2+]i[12,41]. Our observation that ouabain induced increase of [Na+]i and [Ca2+]i in control myocytes also enhanced fractional SR calcium release (Fig. 9) provides strong support for this contention. In 2 Hz stimulated HF cells cariporide caused a decrease of the calcium transient amplitude, when expressed as cytosolic free [Ca2+]i (Fig. 3), but when expressed as cytosolic bound [Ca2+]i the amplitude hardly changed (Fig. 7). This entirely follows from the cytosolic calcium buffering characteristics, which establish strong non-linearity of the relationship between cytosolic free [Ca2+]i and bound [Ca2+]i as a function of diastolic [Ca2+]i (see Section 2 and Ref. [27]). Therefore, to appreciate calcium transient amplitude in terms of SR function and contractility it is more appropriate to express the amplitude as the change of bound cytosolic calcium (see Section 2 and Ref. [27]). It may certainly be argued that calcium buffering characteristics could differ in HF and control myocytes. However, no data are presently available on this. A common feature of myocardium with high [Na+]i is a negative force frequency relationship [30]. Correspondingly, we found a negative relationship between the calcium transient amplitude and stimulation rate in HF and a positive relationship in control myocytes (Fig. 7). Cariporide partly normalized the relationship between the calcium transient amplitude and stimulation rate. In the absence of cariporide we found that SR calcium content was similar in normal and HF myocytes at low frequencies, increased with stimulation rate in normal myocytes, but decreased in HF and that fractional SR calcium release remained 30% higher in HF than in normal cells over the entire range of stimulation rates (submitted); these observations explain the differences in calcium transient amplitudes between HF and control as a function of stimulation rate in the absence of cariporide. Presently no data are available on the stimulation rate dependence of SR calcium content and fractional release in the presence of cariporide. Therefore, an explanation for the observed phenomenon remains speculative. The incidence of spontaneous calcium release from SR in HF myocytes correlated with the incidence of ventricular arrhythmias in vivo [5]. Open probability of RyR progressively increases with increasing diastolic [Ca2+]i[12,40,41]. This may well underlie increased propensity to development of calcium after-transients in conditions of SR calcium overload after rapid pacing in the presence of noradrenalin. In this study we demonstrate that inhibition of NHE with cariporide reversibly decreased [Ca2+]i and development of after transients (Fig. 7 and Table 2). This antiarrhythmic effect could well be attributed to decrease of the RyR open probability with decreasing diastolic [Ca2+]i. We conclude that NHE activity is up-regulated in HF and causes increased [Na+]i, which causes NCX mediated elevation of diastolic [Ca2+]i and secondarily increased calcium leak through RyR. This effect on RyR is responsible for reduced SR calcium content, reduced diastolic SR membrane calcium gradient, increased fractional systolic SR depletion and increased propensity to develop calcium after-transients. All of these parameters characteristic for HF are partly restored by inhibition of NHE. The observations described in this study provide a promising perspective for potentially beneficial clinical applicability of NHE inhibition in the setting of heart failure, both from a perspective of contractile dysfunction and of arrhythmogenesis. Acknowledgements We acknowledge Hoechst, Frankfurt, Germany for kindly supplying cariporide. This study was supported by the Nederlandse Hart Stichting (Dutch Heart Foundation, NHS, Grant No. 96039). References [1] Gwathmey J.K. Slawsky M.T. Hajjar R.J. Briggs G.M. Morgan J.P. Role of intracellular calcium handling in force–interval relationships of human ventricular myocardium J Clin Invest 1990 85 1599 1613 Google Scholar Crossref Search ADS PubMed WorldCat [2] Harding S.E. Jones S.M. O'Gara P. et al. Isolated ventricular myocytes from failing and non-failing human heart: the relation of age and clinical status of patients to isoproterenol response J Mol Cell Cardiol 1992 24 549 564 Google Scholar Crossref Search ADS PubMed WorldCat [3] Sipido K.R. Volders P.G.A. Vos M.A. Verdonck F. Altered Na/Ca exchange activity in cardiac hypertrophy and heart failure: a new target for therapy Cardiovasc Res 2002 53 782 805 Google Scholar Crossref Search ADS PubMed WorldCat [4] Hasenfuss G. Alterations of calcium-regulatory proteins in heart failure Cardiovasc Res 1998 72 681 685 OpenURL Placeholder Text WorldCat [5] Baartscheer A. Fiolet J.W.T. Schumacher C.A. Belterman C.N.W. Vermeulen J.T. SR Ca content and Ca aftertransients in pressure and volume overload induced heart failure [abstract] Biophys J 2001 80 608a OpenURL Placeholder Text WorldCat [6] Baartscheer A. Schumacher C.A. Belterman C.N.W. Fiolet J.W.T. The driving force of the Na+/Ca2+-exchanger and the increase on diastolic calcium in a rabbit model of pressure and volume overload induced heart failure Eur J Heart Fail 2001 3 S61 S62 OpenURL Placeholder Text WorldCat [7] Matsui H. Barry W.H. Livsey C. Spitzer K.W. Angiotensin II stimulates sodium–hydrogen exchange in adult rabbit ventricular myocytes Cardiovasc Res 1995 29 215 221 Google Scholar Crossref Search ADS PubMed WorldCat [8] Cingolani H.E. Alvarez B.V. Ennis I.L. Camilión de Hurtado M.C. Stretch induced alkalinization of feline papillary muscle Circ Res 1998 83 775 779 Google Scholar Crossref Search ADS PubMed WorldCat [9] Alvarez B.V. Pérez N.G. Ennis I.L. Camilión de Hurtado M.C. Cingolani H.E. Mechanisms underlying the increase in force and Ca2+ transient that follow stretch of cardiac muscle: a possible explanation of the anrep effect Circ Res 1999 85 716 722 Google Scholar Crossref Search ADS PubMed WorldCat [10] Engelhardt S. Hein L. Keller U. Klämbt K. Lohse M.J. Inhibition of Na+–H+ exchange prevents hypertrophy, fibrosis, and heart failure in β1-adrenergic receptor transgenic mice Circ Res 2002 90 814 819 Google Scholar Crossref Search ADS PubMed WorldCat [11] Takewaki S. Kuro-o M. Hiroi Y. et al. Activation of Na(+)–H(+) antiporter (NHE-1) gene expression during growth, hypertrophy and proliferation of the rabbit cardiovascular system J Moll Cell Cardiol 1995 25 729 742 Google Scholar Crossref Search ADS WorldCat [12] Zahradníková A. Zahradní I. Analysis of calcium-induced calcium release in cardiac sarcoplasmic reticulum vesicles using models derived from single-channel data Biochim Biophys Acta 1999 1418 268 284 Google Scholar Crossref Search ADS PubMed WorldCat [13] Vermeulen J.T. McGuire M.A. Opthof T. et al. Triggered activity and automaticity in ventricular trabeculae of failing human and rabbits hearts Cardiovasc Res 1994 28 1547 1554 Google Scholar Crossref Search ADS PubMed WorldCat [14] Fuller S.J. Gaitanaki C.J. Sugden P.H. Effects of catecholamines on protein synthesis in cardiac myocytes and perfused heart isolated from adult rats Biochem J 1990 266 727 736 Google Scholar Crossref Search ADS PubMed WorldCat [15] Eguchi S. Matsumoto T. Motley E.D. Utsunomiya H. Inagami T. Identification of an essential signalling cascade for mitogen-activated protein kinase activation by angiotension II in cultured rat vascular smooth muscle cells J Biol Chem 1996 271 14169 14175 Google Scholar Crossref Search ADS PubMed WorldCat [16] Camilion M.C. Pérez N.G. Cingolani H.E. An electrogenic sodium-bicarbonate cotransport in the regulation of myocardial intracellular pH J Mol Cell Cardiol 1995 27 231 242 Google Scholar Crossref Search ADS PubMed WorldCat [17] Yamazaki T. Komuro I. Kudoh S. et al. Role of ion channels and exchanger in mechanical Stretch-induced cardiomyocyte hypertrophy Circ Res 1998 82 430 437 Google Scholar Crossref Search ADS PubMed WorldCat [18] Spitznagel H. Chung O. Qin-Gui X. et al. Cardioprotective effects of the Na+/H+-exchange inhibitor cariporide in infarct-induced heart failure Cardiovasc Res 2000 46 102 110 Google Scholar Crossref Search ADS PubMed WorldCat [19] Yoshida H. Karmazyn M. Na+/H+-exchanger inhibition attenuates hypertrophy and heart failure in 1-week postinfarction rat myocardium Am J Physiol 2000 278 H300 H304 OpenURL Placeholder Text WorldCat [20] Chakrabarti S. Hoque A.N.E. Karmazyn M. A rapid induced ischemia-induced apoptosis in isolated rat hearts and its attenuation by the sodium–hydrgen exchange inhibitor HOE 642 (cariporide) J Mol Cell Cardiol 1997 11 3169 3174 Google Scholar Crossref Search ADS WorldCat [21] Karmazyn M. Amiloride enhances postischemic ventricular recovery: role of the Na–H exchange Am J Physiol 1988 255 H608 615 Google Scholar PubMed OpenURL Placeholder Text WorldCat [22] Theroux P. Chaitman B.R. Danchin N. et al. Inhibition of the sodium–hydrogen exchanger with cariporide to prevent myocardial infarction in high-risk ischemic situations Circulation 2000 102 3032 3038 Google Scholar Crossref Search ADS PubMed WorldCat [23] Bril A. Forest M. Gout B. Ischemia and reperfusion-induced arrhythmias in rabbits with chronic heart failure Am J Physiol 1991 261 H301 H307 Google Scholar PubMed OpenURL Placeholder Text WorldCat [24] Opthof T. Coronel R. Rademaker H.M.E. et al. Changes in sinus function in a rabbit model of heart failure with ventricular arrhythmias and sudden death Circulation 2000 101 2975 2980 Google Scholar Crossref Search ADS PubMed WorldCat [25] ter Welle H.F. Baartscheer A. Fiolet J.W.T. Schumacher C.A. The cytoplasmic free energy of ATP hydrolysis in isolated rod-shaped rat myocytes J Mol Cell Cardiol 1988 20 435 441 Google Scholar Crossref Search ADS PubMed WorldCat [26] Baartscheer A. Schumacher C.A. Fiolet J.W.T. Small changes of cytosolic sodium in rat ventricular myocytes measured with SBFI in emission ratio mode J Mol Cell Cardiol 1997 29 3375 3383 Google Scholar Crossref Search ADS PubMed WorldCat [27] Baartscheer A. Schumacher C.A. Fiolet J.W.T. SR calcium depletion following reversal of the Na+/Ca2+-exchanger in rat ventricular myocytes J Mol Cell Cardiol 2000 32 1025 1037 Google Scholar Crossref Search ADS PubMed WorldCat [28] Miyata H. Silverman H.S. Sollot S.J. et al. Measurement of mitochondrial free Ca2+ concentration in living single rat cardiac myocytes Am J Physiol 1991 261 H1123 H1134 Google Scholar PubMed OpenURL Placeholder Text WorldCat [29] Pogwizd S.M. Schlotthauer K. Weilong Y. Bers D.M. Arrhythmogenesis and contractile dysfunction in heart failure Circ Res 2001 88 1159 1167 Google Scholar Crossref Search ADS PubMed WorldCat [30] Gray R.P. McIntyre P. Sheridan D.S. Fry C.H. Intracellular sodium and contractile function in hypertrophied human and guinea-pig myocardium Eur J Physiol 2001 442 117 123 Google Scholar Crossref Search ADS WorldCat [31] Jelicks L.A. Siri F.M. Effects of hypertrophy and heart failure on [Na+]i in pressure-overload guinea-pig heart Am J Hypertens 1995 8 934 943 Google Scholar Crossref Search ADS PubMed WorldCat [32] Mészáros J. Khananshvili D. Hart G. Mechanisms underlying delayed after-depolarizations in hypertrophied left ventricular myocytes of rats Am J Physiol 2001 281 H903 H914 OpenURL Placeholder Text WorldCat [33] Verdonck F. Volders P.G.A. Vos M.A. Cardiac hypertrophy is associated with an increase in subsarcolemmal Na+ Biophys J 2001 80 59A OpenURL Placeholder Text WorldCat [34] Despa S. Mohammed A.I. Christopher R.W. Pogwizd S.M. Bers D.M. Intracellular Na+ concentration is elevated in heart failure but Na/K pump function is unchanged Circulation 2002 105 2543 2548 Google Scholar Crossref Search ADS PubMed WorldCat [35] von Borren M.M.G.J. Baartscheer A. Ravesloot J.H. Altered pHi regulation in rabbit failing hearts [abstract] J Physiol 2001 533 suppl 31P OpenURL Placeholder Text WorldCat [36] Yokoyama H. Gunasegaram S. Harding S.E. Avkiran M. Sarcolemmal Na+/H+ exchanger activity and expression in human ventricular myocardium J Am Coll Cardiol 2000 36 534 540 Google Scholar Crossref Search ADS PubMed WorldCat [37] Schmidt T.A. Kjeldsen P. Human myocardial Na,K-ATPase quantification, regulation and relation to Ca Cardiovasc Res 1998 37 335 345 Google Scholar Crossref Search ADS PubMed WorldCat [38] Charlemagne D. Orlowski J. Oliviero P. et al. Alterations of Na,K-ATPase subunit mRNA and protein levels in hypertrophied rat heart J Biol Chem 1994 269 1151 1547 OpenURL Placeholder Text WorldCat [39] Ono K. Yano M. Ohkusa T. et al. Altered interaction of FKBP12.6 with ryanodine receptor as a cause of abnormal Ca2+ release in heart failure Cardiovasc Res 2000 48 323 331 Google Scholar Crossref Search ADS PubMed WorldCat [40] Litwin S. Zang D. SR Ca2+ content in myocytes from infarcted hearts [abstract] Biophys J 2001 80 599a OpenURL Placeholder Text WorldCat [41] Jafri M.S. Sobie E.A. Lederer W. Modeling the effects of sarcoplasmic reticulum lumenal and subspace calcium on spontaneous spark rate in cardiac myocytes [abstract] Biophys J 2002 82 1 281 OpenURL Placeholder Text WorldCat Copyright © 2003, European Society of Cardiology
TI - Increased Na+/H+-exchange activity is the cause of increased [Na+]i and underlies disturbed calcium handling in the rabbit pressure and volume overload heart failure model
JO - Cardiovascular Research
DO - 10.1016/S0008-6363(02)00809-X
DA - 2003-03-15
UR - https://www.deepdyve.com/lp/oxford-university-press/increased-na-h-exchange-activity-is-the-cause-of-increased-na-i-and-KJ0FKUqr6q
SP - 1015
EP - 1024
VL - 57
IS - 4
DP - DeepDyve
ER -