Frequency dependent force generation correlates with sarcoplasmic calcium ATPase activity in human myocardium

Objective: In congestive heart failure both a decreased function of the sarcoplasmic Ca²⁺-ATPase and a negative force-frequency relationship have been shown. This study aimed to investigate a possible relationship between frequency potentiation, sarcoplasmic Ca²⁺-ATPase activity, and SERCA2 protein expression in human myocardium. Methods: Frequency potentiation was studied in electrically stimulated, isometric, left ventricular papillary muscle strip preparations (37°C, 0.5–3.0 Hz) from terminally failing (NYHA IV; n=5, dilated cardiomyopathy) and nonfailing (donor hearts, n=5) human myocardium. In the identical samples the Ca²⁺-ATPase activity (NADH coupled assay) and the protein expression of sarcoplasmic Ca²⁺-ATPase (SERCA2), phospholamban, and calsequestrin (western blot) were determined. The frequency dependent change in the force of contraction and V_max of the Ca²⁺-ATPase activity and the protein expression of SERCA2 were correlated with each other. Results: In terminally failing myocardium the force-frequency relationship was negative (2.0 Hz vs. 0.5 Hz: –0.2±0.1 ΔmN) contrasting a positive rate dependent potentiation of force in nonfailing tissue (2.0 Hz vs. 0.5 Hz: +0.8±0.2 ΔmN; p<0.01). In failing myocardium the corresponding maximal sarcoplasmic Ca²⁺-ATPase activity (V_max) was reduced significantly compared to nonfailing myocardium (174±24 vs. 296±31 nmol ATP/mg·min, p<0.01). The protein expression of SERCA2, phospholamban, and calsequestrin remained unchanged in failing myocardium. The maximal Ca²⁺-ATPase activity significantly correlated with the frequency dependent change in force of contraction (2 Hz vs. 0.5 Hz: r=0.88, p=0.001; 3 Hz vs. 0.5 Hz: r=0.84, p=0.004). No correlation between protein expression of SERCA2 and Ca²⁺-ATPase activity or change in force of contraction was observed. Conclusion: Due to a significant correlation between sarcoplasmic Ca²⁺-ATPase activity and frequency potentiation, the negative rate dependent force potentiation in human heart failure could be at least in part be attributed to decreased function of the sarcoplasmic Ca²⁺-ATPase. Received: 8 January 1998, Accepted: 2 June 1998     

The force-frequency relationship is dependent on Ca2+-influx via L-type- and SR-Ca2+-channels in human heart

The present study investigated the influence of Bay K 8644 and nifedipine (Nif) on the force-frequency relationship and on tetanic tension and force of contraction of failing human myocardium (PAP, n = 12). In addition, ryanodine (Rya) was studied on the force-frequency relationship. Bay K 8644 (0.1 μM) increased, but Nif (0.01 μM) reduced isometric force of contraction significantly. However, both, Bay K 8644 (2 Hz vs. 0.5 Hz: Control: −31.6 ± 7.8 %; +Bay K 8644: +103 ± 30 % (% basal); p < 0.005) as well as Nif (2 Hz vs. 0.5 Hz: Control: −8.8 ± 9.7 %; +Nif: +90.9 ± 31.5 %) (% basal); p < 0.05), were able to restore a positive FFR in PAP. By measurement of tetanic tension and posttetanic potentiation in the presence of the 1,4-dihydropyridines, we support the hypothesis of the existence and functional relevance of a dihydropyridin-ryanodine receptor junctional complex. In skinned fiber preparations, Bay K 8644 showed no effect on Ca²⁺-sensitivity or caffeine induced Ca²⁺-release. Rya (10 μM) decreased force of contraction in PAP and was effective in restoring a postive FFR (2 Hz vs. 0.5 Hz: Control: −7.3 ± 5.1 %; +Rya: +98.0 ± 31.9 % (% basal); p < 0.05). Thus, the altered FFR and Ca²⁺-homeostasis in failing human myocardium may result from changes in sarcolemmal Ca²⁺-influx and/or from altered SR-Ca²⁺-load. Received: 27 August 1998, Returned for 1. revision: 24 September 1998, 1. Revision received: 23 October 1998, Returned for 2. revision: 20 November 1998, 2. Revision received: 18 December 1998, Accepted: 22 December 1998     

Calcium-sensitivity of the SR calcium release channel in failing and nonfailing human myocardium

Altered Ca²⁺ metabolism of the sarcoplasmic reticulum results in changes of the contractile behavior in failing human myocardium. The ryanodine-sensitive Ca²⁺ release channel of the sarcoplasmic reticulum plays a key role in the intracellular Ca²⁺ handling in cardiac myocytes. Recently, we showed that the density of ³H-ryanodine binding sites which correspond to the SR Ca²⁺ release channel in human myocardial homogenates is unchanged in failing human myocardium. However, the sensitivity of the channel towards Ca²⁺, which acts as the trigger signal of channel activation and thereby initiates contraction, has not yet been investigated in failing and nonfailing myocardium. Methods: Homogenates (100 μg protein) from hearts with dilated (DCM, n = 10) or ischemic (ICM, n = 9) cardiomyopathy were incubated with a saturating concentration of ³H-ryanodine (12 nM) in the presence of different Ca²⁺ concentrations ranging from 1 nM to 10 mM. For comparison, myocardium of 8 nonfailing hearts which could not be transplanted for technical reasons was investigated. Nonspecific binding was determined in the presence of a high concentration (10 μM) of unlabeled ryanodine. Results: ³H-ryanodine binding to the Ca²⁺ release channel showed a bell-shaped pattern with an increase in specific binding at submicromolar Ca²⁺ concentrations and a decrease at higher Ca²⁺ concentrations than 0.5 mM, whereas nonspecific binding was not influenced by different Ca²⁺ concentrations. In nonfailing myocardium, maximal ³H-ryanodine binding (Bmax) was 85.2 ± 3.1 fmol/mg protein and half-maximal binding was reached at a free Ca²⁺ concentration of 0.25 (0.22 – 0.30)μM (EC₅₀). Neither EC₅₀ values nor maximal specific ³H-ryanodine binding differed between nonfailing and failing myocardium of both etiologies. EC₅₀ values were 0.24 (0.23 – 0.26)μM (DCM, n = 10) or 0.28 (0.25 – 0.31)μM (ICM, n = 9), respectively. Caffeine (2 mM) and the ATP-analogon AMP-PCP (1 mM) led to a shift towards lower Ca²⁺ concentrations consistent with an activation of the channel by these compounds, whereas Mg²⁺ (0.7 mM) shifted the Ca²⁺-dependence of ³H-ryanodine binding towards higher Ca²⁺ concentrations indicating inhibition of channel opening. After activation of the Ca²⁺ release channel by caffeine or AMP-PCP as well as after the inhibition with MG²⁺ EC₅₀ values were the same in failing and nonfailing myocardium. Conclusion: Caffeine and AMP-PCP sensitize, whereas Mg²⁺ desensitizes the myocardial Ca²⁺ release channel to Ca²⁺. The determination of Ca²⁺-dependent ³H-ryanodine binding to the human myocardial Ca²⁺ release channel is a useful tool to investigate its open probability. Furthermore, the Ca²⁺-sensitivity and the pharmacological behavior of the human SR Ca²⁺ release channel are similar in failing and nonfailing myocardium. Received: 7 October 1997, Returned for revision: 9 December 1997, 1.Revision received: 1 December 1998, Accepted: 5 January 1999     

Frequency-dependent changes in calcium cycling and contractile activation in SERCA2a transgenic mice

Objective: This study was undertaken to investigate the mechanism of altered contractility in hearts from transgenic mice overexpressing the sarcoplasmic reticulum (SR) Ca²⁺ ATPase (SERCA2a). In particular, we sought to determine whether the reported increase in contractility is freqnency-dependent, as might be expected if attributable to changes in SR Ca²⁺ loading. Methods: Intracellular [Ca²⁺] and contractile force were measured at room temperature (22 °C) simultaneously in fura-2-loaded isometrically-contracting trabeculae dissected from the hearts of FVB/N control (n=6) or SERCA2a transgenic (n=6) mice. Results: SERCA transgenics exhibit a positive force-frequency relationship, but this was flat in age- and strain-matched controls. SERCA transgenics exhibit a sizable increase in calcium transient amplitude relative to controls, with a concomitant increase in force generation at higher frequencies of stimulation. Amplitudes of Ca²⁺ transients (transgenics: 1.56 ± 0.09 μmol/l, controls: 1.21 ± 0.14) and twitches (transgenics: 21.71 ± 0.91 mN/mm², controls: 13.74 ± 1.67) were significantly different at 2.0 Hz stimulation (P < 0.05). Conclusion: An increase in SERCA expression increases the ability of the sarcoplasmic reticulum to store calcium, such that more calcium is available to be released during each heartbeat at higher stimulation rates. Received: 4 June 1999, Returned for revision: 1 July 1999, Revision received: 4 October 1999, Accepted: 14 October 1999     

Influence of mild hypothermia on myocardial contractility and circulatory function

Myocardial contractility depends on temperature. We investigated the influence of mild hypothermia (37–31°) on isometric twitch force, sarcoplasmic reticulum (SR) Ca²⁺-content and intracellular Ca²⁺-transients in ventricular muscle strips from human and porcine myocardium, and on in vivo hemodynamic parameters in pigs. In vitro experiments: muscle strips from 5 nonfailing human and 8 pig hearts. Electrical stimulation (1 Hz), simultaneous recording of isometric force and rapid cooling contractures (RCCs) as an indicator of SR Ca²⁺-content, or intracellular Ca²⁺-transients (aequorin method). In vivo experiments: 8 pigs were monitored with Millar-Tip (left ventricular) and Swan-Ganz catheter (pulmonary artery). Hemodynamics parameters were assessed at baseline conditions (37°), and after stepwise cooling on cardiopulmonary bypass to 35, 33, and 31°C. Hypothermia increase isometric twitch force significantly by 91 ± 16% in human and by 50 ± 9% in pig myocardium (31 vs. 37°C; p < 0.05, respectively). RCCs or aequorin light emission did not change significantly. In anesthetized pigs, mild hypothermia resulted in an increase in hemodynamic paramters of myocardial contractility. While heart rate decreased from 111 ± 3 to 73 ± 1 min⁻¹, cardiac output increased from 2.4 ± 0.1 to 3.1 ± 0.3 l/min, and stroke volume increased from 21 ± 1 to 41 ± 3 ml. +dP/dt_max increased by 25 ± 8% (37 vs. 31°C; p < 0.05 for all values). Systemic and pulmonary vascular resistance did not change significantly during cooling. Mild hypothermia exerts significant positive inotropic effects in human and porcine myocardium without increasing intracellular Ca²⁺-transients or SR Ca²⁺-content. These effects translate into improved hemodynamics parameters of left ventricular function. Received: 26 June 2000, Returned for revision: 20 July 2000, Revision received: 11 October 2000, Accepted: 17 October 2000     

The intracellular Ca2+-homeostasis influences the frequency-dependent force-generation in man

The present study invesitgates the effect of stimulation frequency and external Ca²⁺-concentration on intracellular systolic and diastolic Ca²⁺ as well as on the force-frequency relationship (FFR, 0.5 to 3.0 Hz, 1.0 mmol/l extracellular Ca²⁺) in human myocardium using fura-2 AM loaded electrically stimulated right atrial muscle strips (coronary bypass surgery, n = 15, age: 60.0 ± 1.9 years). The FFR was positive (3.0 vs. 0.5 Hz: 184 ± 43 % of basal value) and linked to an increase in peak systolic (R340/380 _sys, 119 ± 7 %) as well as diastolic Ca²⁺ (R340/380_ED, Δ fura-2 ratio +0.20 ± 0.02). After elevating the extracellular Ca²⁺ concentration from 1.0 to 2.4 mmol/l, force of contraction (FOC) increased from 0.5 up to 1.0 Hz (128 ± 8 %) and declined after further augmentation of stimulation frequency (3.0 Hz: 87 ± 15 %). However, this decrease in FOC was accompanied by an increase in diastolic Ca²⁺ (Δ fura-2 ratio +0.45 ± 0.08), while systolic Ca²⁺ declined at high stimulation frequencies. In conclusion, the frequency-dependent force generation is accompanied by an increase in both systolic and diastolic Ca²⁺ levels. Thus, especially at high stimulation frequencies the Ca²⁺-lowering mechanisms may become crucial and may be responsible for the blunted force-frequency relationship in failing human myocardium. Received: 2 November 1998, Returned for revision: 30 November 1998, Revision received: 3 February 1999, Accepted: 10 February 1999     

Evidence for calcineurin-mediated regulation of SERCA 2a activity in human myocardium

Compromised SERCA 2a activity is a key malfunction leading to the Ca(2+) cycling alterations in failing human myocardium. SERCA 2a activity is regulated by the Ca(2+)/calmodulin-dependent protein kinase (CaM-kinase) but alterations of the CaM-kinase pathway regarding SERCA 2a in heart failure are unresolved. Therefore we investigated the CaM-kinase and phosphatase calcineurin mediated regulation of SERCA 2a in failing and non-failing human myocardium. We studied human myocardial preparations from explanted hearts from non-failing organ donors (NF, n=8) and from patients with terminal heart failure undergoing cardiac transplantation (dilated cardiomyopathy, DCM, n=8). SERCA 2a activity was determined using a NADH-coupled enzyme assay [expressed in nmol ATP/(mg protein x min)] and by(45)Ca(2+) uptake. Protein expression of SERCA 2a, phospholamban, calsequestrin and calcineurin was assessed by Western blotting (expressed as densitometric units/microg protein); phosphorylation of cardiac proteins was detected with specific phospho-antibodies for phospholamban at threonine-17 (PT17) or by incorporation of [gamma -(32)P] (expressed as pmol(32)P/mg). Maximal(45)Ca(2+) uptake (in pmol/mg/min) (NF: 3402+/-174; DCM: 2488+/-189) and maximal SERCA 2a activity were reduced in DCM compared to NF (V(max): NF: 125+/-9; DCM: 98+/-5). The V(max) reduction could be mimicked by calcineurin in vitro in NF (NF(control): 72.1+/-3.7; NF(+calcineurin): 49.8+/-2.9) and restored in DCM by CaM-kinase in vitro (DCM(control): 98+/-5; DCM(+CaM-kinase): 120+/-6). Protein expression of SERCA 2a, phospholamban and calsequestrin remained similar, but calcineurin expression was significantly increased in failing human hearts (NF: 11.6+/-1.5 v DCM: 17.1+/-1.6). Although the capacity of endogenous CaM-kinase to phosphorylate PT17 was significantly higher in DCM (DCM(control): 128+/-36; DCM(+endogenous CaM-kinase): 205+/-20) compared to NF myocardium (NF(control): 273+/-37; NF(+endogenous CaM-kinase): 254+/-31), net phosphorylation at threonine-17 phospholamban was significantly lower in DCM (DCM 130+/-11 v NF 170+/-11). A calcineurin-dependent dephosphorylation of phospholamban could be mimicked in vitro by incubation of NF preparations with calcineurin (NF(control) 80.7+/-4.4 v NF(+calcineurin) 30.7+/-4.1, P<0.05). In human myocardium, the V(max) of SERCA 2a and the phosphorylation of phospholamban is modulated by CaM-kinase and calcineurin, at least in vitro. In failing human myocardium, despite increased CaM-kinase activity, calcineurin dephosphorylation leads to decreased net phosphorylation of threonine-17 phospholamban in vivo. Increased calcineurin activity contributes to the impaired V(max) of SERCA 2a in failing human myocardium and the disorder in Ca(2+)-handling in heart failure.     

Myocardial ischemia and reperfusion reduce the levels of cyclic ADP-ribose in rat myocardium

Cyclic ADP-ribose (cADPR) is a novel Ca²⁺-mobilizing second messenger in mammalian cells including cardiomyocytes. It is unknown whether myocardial ischemia and reperfusion affect the metabolism of cADPR in the myocardium. The present study therefore examined the effects of myocardial ischemia and reperfusion on the concentrations of myocardial cADPR using high-performance liquid chromatography. Basal levels of cADPR in rat myocardium were 5.3 ± 1.8 nmol· mg⁻¹ protein. Myocardial ischemia for 30 min significantly decreased cADPR concentrations to 2.1 ± 0.4 nmol·mg⁻¹ protein. During reperfusion, cADPR was maintained at ischemic levels. The activity of ADP-ribosyl cyclase was expressed as the conversion rate of nicotinamide guanine dinucleotide (NGD⁺) to cyclic GDP-ribose. Myocardial ischemia and reperfusion did not alter the activity of ADP-ribosyl cyclase. However, cADPR hydrolase activity, as measured by the conversion rate of cADPR to ADP-ribose, was significantly elevated by ischemia and reperfusion. To determine the mechanism resulting in the enhancement of cADPR hydrolase activity, we examined the effects of changes in ADP, ATP, pH, and PO₂ on the conversion rate of cADPR to ADPR. Alterations of ADP, ATP, or pH in myocardial tissue had no effect on the degradation of cADPR, whereas a decrease in tissue PO₂ markedly increased the hydrolysis of cADPR. These results suggest that myocardial ischemia and reperfusion decrease cADPR in the myocardium by increasing its hydrolysis. Tissue hypoxia may be one of the important mechanisms to activate cADPR hydrolase. Received: 24 July 2001/Returned for 1. revision: 20 August 2001/1. Revision received: 25 October 2001/Returned for 2. revision: 20 November 2001/2. Revision received: 12 December 2001/Accepted: 2 January 2002     

Altered force-frequency relation in hypertrophic obstructive cardiomyopathy

The present study was designed to test the hypothesis that in hypertrophied myocardium of patients with hypertrophic obstructive cardiomyopathy (HOCM) a reduced contractile reserve provided by frequency dependent potentiation of force of contraction contributes to the myocardial dysfunction. Myectomy was performed in 8 HOCM patients with normal systolic left ventricular function at rest. Nonfailing myocardium from the hearts of three multiorgan donors was investigated for comparison. In thin myocardial strips we measured the inotropic effects of different stimulation frequencies (0,5–3.0Hz) at different extracellular Ca²⁺ concentrations (1.8–16.2 mmol/l). At 1.8 mmol/l extracellular Ca²⁺ concentration, increasing stimulation rates had no positive inotropic effect in HOCM myocardium, whereas in nonfailing myocardium force of contraction increased up to 3 Hz. Increasing extracellular Ca²⁺ concentrations induced a positive force-frequency relation in HOCM with a maximum at 5.4 mmol/l Ca²⁺. A further increase to 16.2 mmol/l Ca²⁺ resulted in a negative force-frequency relation in these specimens. The time to peak tension and the time to relaxation decreased at increasing stimulation frequencies at all Ca²⁺ concentrations investigated. In conclusion, in hypertrophied myocardium of HOCM patients increasing stimulation frequencies failed to have a positive inotropic effect at physiological extracellular Ca²⁺ concentrations. The induction of a positive force-frequency relation by higher Ca²⁺ concentrations suggests that an abnormal cellular Ca²⁺ handling may play an important pathophysiological role. Received: 27 December 1996, Returned for revision: 3 March 1997, Revision received: 1 September 1998, Accepted: 30 September 1998     

Differences in time course of myocardial mRNA expression in non-infarcted myocardium after myocardial infarction

In non-infarcted myocardium after myocardial infarction, the change of cardiac phenotypic modulation of contractile protein, extracellular matrix and intracellular Ca²⁺ transport protein, such as sarcoplasmic reticulum Ca²⁺(SR-Ca²⁺)-ATPase, Na⁺-Ca²⁺ exchanger, have a important role during cardiac remodeling. However, the time course in this gene expression in the adjacent and remote left ventricular, or right ventricular myocardium after myocardial infarction has not been well examined. The purpose of this study was to examine the left ventricular function and regional cardiac gene expression after myocardial infarction. Myocardial infarction was produced in Wistar rats by the ligation of the left anterior descending coronary artery. After 3 weeks, 2 months and 4 months from myocardial infarction, we performed Doppler echocardiography and measured the systolic and diastolic function. Then, we analyzed the contractile protein, extracellular matrix and intracellular Ca ²⁺ transport protein mRNAs of cardiac tissues in the adjacent and the remote noninfarcted myocardium, and right ventricular myocardium by Northern blot hybridization. Fractional shortening of infarcted heart progressively decreased. Peak early diastolic filling wave (E wave) velocity increased, and the deceleration rate of the E wave velocity was more rapid in myocardial infarction areas. Atrial filling wave (A wave) velocity decreased, resulting in a marked increase in the ration of E wave to A wave velocity. Expression of myocardial α-skeletal actin, β-MHC and ANP mRNA, or collagen I and III mRNA were higher at 3 weeks after myocardial infarction. SR Ca²⁺-ATPase mRNA in the adjacent non-infarcted myocardium was decreased at 2 months, and that in remote myocardium was decreased at 4 months after infarction. Na⁺-Ca²⁺ exchanger mRNA levels were increased at 3 weeks, but was decreased at 2 months in the adjacent non-infarcted myocardium and at 4 months in the remote myocardium. These findings suggest that the compensation for myocardial infarction by myocardial gene expression in non-infarcted myocardium may occur at an early phase after myocardial infarction, and myocardial dysfunction may begin from adjacent to remote non-infarcted myocardium during progressive cardiac remodeling. Received: 9 August 1999, Returned for revision: 16 September 1999, Revision received: 5 January 2000, Accepted: 26 January 2000     

The enhanced contractility in phospholamban deficient mouse hearts is not associated with alterations in Ca2+-sensitivity or myosin ATPase-activity of the contractile proteins

Work performing heart preparations from hypercontractile, phospholamban deficient mouse hearts showed no change in parameters of contraction or relaxation in response to isoproterenol stimulation. Thus, the aim of the present study was to investigate whether or not changes at the level of the contractile apparatus occur in addition to the altered expression of Ca²⁺-regulating proteins observed in these mouse models, e. g., phospholamban, ryanodine receptors. Triton-X skinned fiber preparations from phospholamban deficient (n = 9) and wild-type (n = 10) mice were used and the Ca²⁺-activated force as well as the myosin ATPase-activity were simultaneously measured. The tension dependent ATPase-activity was unchanged in phospholamban deficient animals when compared to controls. The SERCA 2a-inhibitor cyclopiazonic acid did not affect myosin ATPase-activity in this system. The Ca²⁺-sensitivity of Ca²⁺-activated force and myosin ATPase were unchanged as well. Comparison of the concentrations needed to achieve half maximal activation of the myosin ATPase-activity and force demonstrated that the Ca²⁺-sensitivity of the myosin ATPase was higher compared to the Ca²⁺-sentivity of tension development. This holds true for phospholamban deficient mice (EC₅₀ ATPase: 0.9 ± 0.2 μmol/l; tension: 1.7 ± 0.3 μmol/l; p < 0.001) and wild-type controls (1.1 ± 0.01 μmol/l; 2.2 ± 0.4 μmol/l; p < 0.01). The myosin ATPase-activity and force were correlated to each other in both, phospholamban deficient mice and controls and did not change at submaximal Ca²⁺ concentrations. The ATPase/force-ratio, as a parameter of tension cost, was similar in either phospholamban deficient mice or controls. Thus, the present study provides evidence that at the level of the contractile proteins regulation of Ca²⁺-activated force and energy demand of force development are not altered in phospholamban deficient mice with enhanced myocardial performance. At the level of the regulation of crossbridge interaction, no adaptive or compensatory mechanisms have been initiated by ablation of phospholamban. Received: 22 March 1999, Returned for revision: 26 April 1999, Revision received: 1 July 1999, Accepted: 12 July 1999     

Endocardial versus epicardial differences of sarcoplasmic reticulum Ca2+-ATPase gene expression in the canine failing myocardium

It is unknown whether the transmural heterogeneity of sarcoplasmic reticulum (SR) Ca²⁺-ATPase gene expression is present within the left ventricular (LV) wall. Moreover, the changes of transmural distribution have not been examined in the failing hearts. We thus quantified steady-state mRNA abundance of SR Ca²⁺ regulatory proteins by Northern blot analysis in both subendocardial and subepicardial LV layers from normal and rapid pacing-induced heart failure (HF) dog hearts. For normal LV, Ca²⁺-ATPase mRNA abundance (normalized to glyceraldehyde-3-phosphate dehydrogenase [GAPDH] mRNA) was significantly reduced in the subendocardium, whereas calsequestrin mRNA abundance was comparable between the two layers. For HF LV, Ca²⁺-ATPase mRNA abundance in the subendocardium was also reduced compared to the subepicardium. However, the endocardium to epicardium ratio was comparable between control and HF (0.62 ± 0.08 vs. 0.65 ± 0.07; p = NS). Therefore, the transmural gradient of this gene was constant in both control and HF. Even though the data on the transmural heterogeneity of protein level is not available, the subendocardium contained significantly less Ca²⁺-ATPase mRNA, which might contribute, at least in part, to the transmural gradients of biochemical and mechanical function. Received: 8 February 1999, Returned for revision: 8 March 1999, Revision received: 1 April 1999, Accepted: 23 April 1999     

Cyclic GMP attenuates cyclic AMP-stimulate inotropy and oxygen consumption in control and hypertrophic hearts

We tested the hypothesis that increasing myocardial cyclic GMP would attenuate cyclic AMP induced positive inotropy and O₂ consumption, in part, through changes in cyclic AMP and that renal hypertension-induced cardiac hypertrophy (HYP) would alter this relationship. Anesthetized, open chest rabbits (N = 48) were divided into four groups of control (CON) and HYP animals which received vehicle (VEH), isoproterenol 10⁻⁶M (ISO), 3-morpholinosyndnonimine 10⁻⁴M, (SIN-1), or a combination of ISO+SIN-1. Coronary blood flow (micro-spheres) and O₂ extraction (microspectrophotometry) were used to determine O₂ consumption in both subepicardium (EPI) and subendocardium (ENDO). Left ventricular change in wall thickness (%) was increased significantly by ISO in both CON (16 ± 4 to 31 ± 6) and HYP (17 ± 2 to 24 ±3). Percent change in wall thickness was similar in the CON, SIN-1, and ISO+SIN-1 groups. Myocardial O₂ consumption (ml O₂/min/100 g) was increased by ISO in CON (10.3 ± 1.0 to 13.6 ± 2.0 EPI; 10.9 ± 1.0 17.1 ±1.7 ENDO) and HYP (8.2 ± 1.4 to 12.3 ± 2.2 EPI; 6.6 ± 1.4 to 14.8 ± 1.8 ENDO). Oxygen consumption was unaffected by SIN-1 in CON and HYP animals. ISO+SIN-1 caused attenuated ISO-induced increases in O₂ consumption in CON in EPI and ENDO, and in EPI in HYP. Cyclic GMP (pmol/g) was unchanged by ISO in CON and HYP, and increased by SIN-1 in CON (8.1 ± 1.3 to 19.2 ± 2.3 EPI) and HYP (9.1 ± 1.5 to 12.8 ± 2.0 EPI). Cyclic GMP remained elevated with ISO+SIN-1 in both groups. Cyclic AMP (pmol/g) was increased significantly by ISO in CON (496 ± 43 to 725 ± 106 EPI; 534 ± 44 to 756 ± 148 ENDO) and insignificantly in HYP (435 ± 50 to 566 ± 35 EPI; 497 ± 51 to 583 ± 47 ENDO). Cyclic AMP levels were unaffected by SIN-1 in either group. Isoproterenol induced increases in cyclic AMP were blunted by ISO+SIN-1 in CON (496 ± 43 to 537 ± 59 EPI) and not affected in HYP. The current study demonstrated attenuation of cyclic AMP mediated increased inotropy and O₂ consumption by increasing cyclic GMP, which appeared, in part, related to cyclic GMP-induced reduction in cyclic AMP. This effect of cyclic GMP on cyclic AMP was not observed in myocardial hypertrophy. Received: 4 January 1999, Returned for 1. revision: 29 January 1999, 1. Revision received: 30 March 1999, Returned for 2. Revision: 3 May 1999, 2. Revision received: 3 May 1999, Returned for 3. Revision: 12 May 1999, 3. Revision received: 23 June 1999, Returned for final revision: 7 July 1999, Accepted: 22 July 1999     

Sarcoplasmic reticulum Ca 2+ load in human heart failure

Excitation-contraction coupling and intracellular Ca ²⁺ homeostasis are altered in heart failure. We tested the hypothesis that these changes are related to disturbed Ca ²⁺ handling of the sarcoplasmic reticulum (SR). Isolated, electrically stimulated trabeculae were obtained from end-stage failing (NYHA IV) and nonfailing human hearts. Isometric twitch tension, intracellular Ca ²⁺ transients (aequorin method) and SR Ca ²⁺ content (rapid cooling contractures) were assessed under basal conditions (1 Hz, 37 °C) as well as after stepwise increasing rest intervals from 2 – 240 s (post-rest contractions). Protein expression of SERCA2a and phospholamban (Western blot) was assessed in a subset of failing trabeculae. In addition, the effects of SERCA1 overexpression on contractile function of isolated myocytes was tested. On average, post-rest twitch tension continuously increased with increasing rest intervals in nonfailing, but declined with rest intervals longer than 15s in failing myocardium. The rest-dependent contractile changes were accompanied by parallel changes in intracellular Ca ²⁺ transients. Failing trabeculae (n = 40) were grouped (group A: post-rest potentiation (force of contraction > pre-rest twitch force) after 120s rest interval; group B: post-rest decay (force of contraction < pre-rest twitch force) after 120 s rest interval), and post-rest contractile function was related to SERCA2a and PLB expression. While PLB protein expression was not different, SERCA2a protein expression as well as SERCA2a/PLB ratio was significantly higher in group A vs. group B. Transfection of SERCA1 increased shortening amplitude and enhanced relaxation kinetics in failing human myocytes. In conclusion, SR Ca ²⁺ handling is severely altered in human heart failure. Reduced SR Ca ²⁺ release is due to diminished SR Ca ²⁺ content directly related to a depressed expression of SERCA2a protein. Enhancing SERCA function or expression may improve SR Ca ²⁺ handling in failing human myocardium.     

The effects of levosimendan on the left ventricular function and protein phosphorylation in post-isschemic guinea pig hearts

The widely accepted theories for the decreased function in the stunned myocardium relate to Ca²⁺ desensitization and free radical-mediated tissue damage of the myofilaments. The aim of the present study was to examine whether the depressed contractile function and Ca²⁺ responsiveness of the stunned myocardium may be restored by a new Ca²⁺ sensitizer (levosimendan), which has been shown to improve the Ca²⁺ response of the myofilaments. The effects of levosimendan on the left ventricular function and the in vivo protein phosphorylation were examined in both the non-ischemic and the stunned myocardium. Myocardial stunning was induced in Langendorff-perfused guinea pig hearts by suspending the circulation for 8 min, followed by a 20-min reperfusion period. Perfusion of post-ischemic guinea pig hearts with levosimendan (0.03–0.48 μM, 6 min) was associated with dose- and time-dependent increases in both dP/dt_max (contractility) and dP/dt_min (speed of relaxation). When the effectiveness of levosimendan was compared in non-ischemic and post-ischemic hearts, no significant differences were noted in the relative stimulatory effects on contractility and relaxation, at any given time point (time-response curve) or concentration (dose-response curve). Perfusion of the guinea pig hearts with a high (0.3 μM) levosimendan concentration did not reveal any qualitative or quantitative difference in the phosphodiesterase inhibitory potential of the compound (elevation of tissue cyclic AMP levels and characteristics of protein phosphorylation) between the non-ischemic and the post-ischemic myocardium. However, when isoproterenol was adminstered to induce maximal in vivo phosphorylation of cardiac phosphorproteins, an attenuation of the ³²P-incorporation into troponin I was noted in the post-ischemic hearts. The decrease in isoproterenol-induced ³²P-incorporation into troponin I was associated with similar alterations in the tissue level of this protein. We conclude that the Ca²⁺ sensitizer levosimendan exerts dose- and time-dependent positive inotropic and lusitropic effects on the postischemic myocardium, lending support to the hypothesis tha Ca²⁺ desensitization of the myofibrils is involved in myocardial stunning. Received: 20 July 1998, Returned for 1. revision: 27 August 1998, 1. Revision received: 6 January 1999, Returned for 2. revision: 5 February 1999, 2. Revision received: 25 February 1999, Accepted: 3 March 1999     

Electrophysiology of single cardiomyocytes isolated from rabbit pulmonary veins: implication in initiation of focal atrial fibrillation

Pulmonary veins (PVs) are important foci in initiation of paroxysmal atrial fibrillation. However, the mechanisms of the high arrhythmogenic activity of PVs are unclear. This study aimed to isolate single cardiomyocytes from PVs and evaluate their electrophysiological characteristics and arrhythmogenic potential. Cardiomyocytes of rabbit PVs were isolated by retrograde perfusion with digestive enzymes from aorta via left ventricle and left atrium. The action potentials and ionic currents were investigated in isolated single PV cardiomyocytes using the whole-cell clamp technique. Dissociation of PVs yielded single pacemaker cardiomyocytes (76 %) and non-pacemaker cardiomyocytes with a fast response action potential. Both the pacemaker and non-pacemaker cardiomyocytes had similar inward Ca²⁺ currents and transient outward K⁺ currents. However, the pacemaker cardiomyocytes had a smaller inward rectifier K⁺ current (1.50 ± 0.22 versus 4.21 ± 1.15 pA/pF, P < 0.005) and a larger delayed rectifier K⁺ current (0.60 ± 0.05 versus 0.24 ± 0.05 pA/pF, P < 0.005) than non-pacemaker cardiomyocytes. Acetylcholine induced hyperpolarization and inhibited the spontaneous action potential. Isoproterenol (10 nM) accelerated the spontaneous activity and induced early or delayed afterdepolarization, which could be suppressed by nifedipine. The PV cardiomyocytes with early afterdepolarization have a greater prolongation of action potential duration (ΔAPD, + 67 ± 17 versus −109 ± 20 ms, P < 0.0001) and a greater increase of inward Ca²⁺ current (0.90 ± 0.23 versus 0.38 ± 0.08 pA/pF, P < 0.05) after isoproterenol than those without early afterdepolarization. These findings suggest that PV cardiomyocytes have distinct action potentials and ionic current profiles, which may be responsible for the high arrhythmogenic activity of the PVs. Received: 17 April 2001, Returned for revision: 23 May 2001, Revision received: 11 July 2001, Accepted: 17 July 2001     

Altered hetero- and homeometric autoregulation in the terminally failing human heart

OBJECTIVE AND METHODS: To further investigate length-dependent force generation in human heart, nonfailing (donor hearts, NF) and terminally failing (heart transplants, dilated cardiomyopathy, DCM) left ventricular myocardium was studied under various preload (4-40 mN/mm2) or length conditions. In addition, morphological studies (van Giesson Trichrome staining, electron microscopy) were performed. RESULTS: In NF, a biphasic increase in force of contraction (FOC) was observed after elevating the preload (4-40 mN/mm2): there was an immediate fast increase (FOCf,), followed by a slow increase over several minutes (FOCs), which was paralleled by an increase in the systolic fura-2 transient. In DCM, FOCf, FOCs and the systolic fura-2 transient were blunted and diastolic tension was increased at increasing muscle length. Only in NF, a stretched induced increase in diastolic fura-2 ratio was observed. In DCM, no obvious interstitial fibrosis and no difference in basement membrane structure and attachment were observed. CONCLUSIONS: Since FOCf has been attributed to the Frank-Starling mechanism, whereas FOCs represents a length-dependent increase in the intracellular Ca2+-transient, the impaired length-dependent force generation in failing myocardium results from a dysregulation of both myofibrillar Ca2+-sensitivity as well as the intracellular Ca2+-homeostasis. Interstitial fibrosis may have only minor impact on force generation in human end-stage heart failure.     

Effects of different inotropic interventions on myocardial function in the developing rabbit heart

The development of the mammalian heart is characterized by substantial changes in myocardial performance. We studied the ontogeny of myocardial function with and without various inotropic interventions in the developing isolated, antegrade-perfused rabbit heart (2d, 8d, 14d, 28d, n = 96). Myocardial function was related to the protein expression of the sarcolemmal Na⁺-Ca²⁺ exchanger and to the sarcoplasmic Ca²⁺-ATPase. In neonatal hearts an age-dependent increase in maximal developed pressure velocity (dP/dt_max) by 45 % and peak negative pressure velocity (dP/dt_min) by 75 % within days 2 to 8 were observed. In response to inotropic intervention with isoproterenol, ouabain, calcium and the Na⁺-channel modulator BDF 9148, dP/dt_max and dP/dt_min increased in a concentration dependent manner. Significant differences between neonatal, juvenile and adult hearts could be demonstrated in a repeated measurement ANOVA model on the concentration-response curves for BDF 9148 (dP/dt_max and dP/dt_min), ouabain (dP/dt_min) and calcium (dP/dt_min), but not for isoproterenol. At the maximum isoproterenol concentration of 1 μmol/l, the increase in dP/dt_max and dP/dt_min was significantly higher in adult compared to neonatal hearts (t-test, p < 0.01). The significant decline of the Na⁺-Ca²⁺ exchanger protein expression from neonatal (1822 ± 171) to adult hearts (411 ± 96 S.E.M. [units per 20 μg protein], p < 0.01) was related to an increase in myocardial function (dP/dt_max r = 0.63, p < 0.01, dP/dt_min r = 0.62, p < 0.01). Contractility, relaxation and the observed positive inotropic effects were in general significantly lower in neonatal compared to adult hearts. In the individual heart an increase in contractility and relaxation was related to a decrease in Na⁺-Ca²⁺ exchanger expression. Received: 22 May 2000, Returned for 1. revision: 21 June 2000, 1. Revision received: 27 November 2000, Returned for 2. revision: 19 December 2000, 2. Revision received: 2 January 2001, Returned for 3. revision: 17 January 2001, 3. Revision received: 25 May 2001, Accepted: 11 June 2001     

L-type calcium currents in atrial myocytes from patients with persistent and non-persistent atrial fibrillation

Objective. In patients with persistent atrial fibrillation (AF), the atrial myocardium is characterized by a reduced contractile force, by a shortened duration of the action potential and a recently demonstrated reduction of the L-type Ca²⁺ currents. We analyzed potential effects on L-type Ca²⁺ currents of the patients' medication and of the duration of AF. Methods and results. Human atrial myocytes were prepared from the right auricles of patients undergoing open-heart surgery. Three groups of patients were studied: a control group with sinus rhythm (SR, n = 26 patients) and a group with persistent AF (> 3 months duration; n = 10), a group with non-persistent AF (3 patients with SR but with documented episodes of AF in their history). L-type Ca²⁺ currents were measured during depolarizing pulses from a holding potential of −70mV to a test potential of +10mV and are given as mean ±SEM of current densities (currents normalized to the cell capacitance). Ca²⁺ current densities were significantly (p < 0.0001) smaller in cells from patients with persistent AF than in control cells (0.54 ± 0.08 pA/pF vs. 1.96 ± 0.12 pA/pF). No indication was found that these changes were caused by medication with Ca²⁺ channel antagonists, β blockers, or digitalis. Stimulation with the dihydropyridine Bay K 8644 (1 μM) or with isoproterenol (0.1 μM) increased Ca²⁺ currents in control cells 3.5 ± 0.2 and 3.5 ± 0.3-fold. In persistent AF, this increase was significantly larger (6.0 ± 0.5 and 5.2 ± 0.6-fold) but stimulated currents were still significantly lower than in control cells. Patients with non-persistent AF exhibited Ca²⁺ currents well within the control range. Conclusion. A reduction in Ca²⁺ currents, due to a reduction in number as well as a depression of L-type channels, is a characteristic and pathophysiologically important part of the myocardial remodeling during long-lasting atrial fibrillation. It is not present in patients with non-persistent AF and not caused by medication. Received: 27 September 2000, Returned for revision: 9 October 2000, Revision received: 8 November 2000, Accepted: 9 November 2000     

Production and metabolism of ceramide in normal and ischemic-reperfused myocardium of rats

Ceramide has been shown to be a key signaling molecule involved in the apoptotic effect of tumor necrosis factor α (TNF-α) and other cytokines. Given the importance of cytokines such as TNF-α in myocardial ischemia-reperfusion injury, we hypothesize that ceramide is increased during ischemia or reperfusion, and that the activity of enzymes responsible for its production or breakdown should be increased and/or decreased, respectively. Therefore, in the present study, we characterized the enzymatic activities responsible for ceramide production and metabolism in the myocardium of rats, and determined the contribution of these enzymes to altered ceramide levels during myocardial ischemia and reperfusion. The basal ceramide concentration in the myocardium of rats was 34.0 pmol/mg tissue. As determined by the conversion of ¹⁴C-sphingomyelin into ceramide and ¹⁴C-choline phosphate, both neutral (N-) and acidic (A-) SMase were detected in the myocardium, with a conversion rate of 0.09 ± 0.008 and 0.32 ± 0.05 nmol/min per mg protein, respectively. The activity of A-SMase (78 % of total cellular activity) was significantly higher in microsomes than in cytosol, while the activity of N-SMase was similar in both fractions. Ceramidase, a ceramide-metabolizing enzyme, was also detected in the myocardium of rats. It metabolized ceramide into sphingosine at a rate of 9.94 ± 0.42 pmol/min per mg protein. In anesthetized rats, 30 min of ischemia had no apparent effect on ceramide concentrations in the myocardium, while 30 min of ischemia followed by 3 h of reperfusion resulted in a significant increase in ceramide by 48 %. The activities of both N- and A-SMase were reduced by 44 % and 32 %, respectively, in the myocardium subjected to ischemia followed by reperfusion, but unaltered in the ischemic myocardium. It was also found that myocardial ischemia followed by reperfusion produced a marked inhibition of ceramidase (by 29 %). These results demonstrate that the myocardium of rats expresses N- and A-SMase and ceramidase, which contribute to the production and metabolism of ceramide, respectively. Tissue ceramide concentrations increased in reperfused myocardium. These increases in ceramide were not associated with enhanced SMase activity, but rather with reduced ceramidase activity. Received: 4 October 2000, Returned for revision: 23 October 2000, Revision received: 22 November 2000, Accepted: 5 December 2000