Previous studies in adult myocytes isolated from rat hearts 3 wk after myocardial infarction (MI) demonstrated abnormal contractility and intracellular Ca2+ concentration ([Ca2+]i) homeostasis and decreased sarcoplasmic reticulum Ca2+-ATPase (SERCA2) expression and activity, but sarcoplasmic reticulum Ca2+ leak was unchanged. In the present study, we investigated whether SERCA2 overexpression in MI myocytes would restore contraction and [Ca2+]i transients to normal. Compared with sham-operated hearts, 3-wk MI hearts exhibited significantly higher left ventricular end-diastolic and end-systolic volumes but lower fractional shortening and ejection fraction, as measured by M-mode echocardiography. Seventy-two hours after adenovirus-mediated gene transfer, SERCA2 overexpression in 3-wk MI myocytes did not affect Na+-Ca2+ exchanger expression but restored the depressed SERCA2 levels toward those measured in sham myocytes. In addition, the reduced sarcoplasmic reticulum Ca2+ uptake in MI myocytes was improved to normal levels by SERCA2 overexpression. At extracellular Ca2+ concentration of 5 mM, the subnormal contraction and [Ca2+]i transient amplitudes in MI myocytes (compared with sham myocytes) were restored to normal by SERCA2 overexpression. However, at 0.6 mM extracellular Ca2+ concentration, the supernormal contraction and [Ca2+]i transient amplitudes in MI myocytes (compared with sham myocytes) were exacerbated by SERCA2 overexpression. We conclude that SERCA2 overexpression was only partially effective in ameliorating contraction and [Ca2+]i transient abnormalities in our rat model of ischemic cardiomyopathy. We suggest that other Ca2+ transport pathways, e.g., Na+-Ca2+ exchanger, may also play an important role in contractile and [Ca2+]i homeostatic abnormalities in MI myocytes.
- primary cardiac myocyte culture
- fura 2
- excitation-contraction coupling
- gene transfer
ca2+ transport systems in sarcoplasmic reticulum (SR) play a major role in cardiac excitation-contraction coupling. During an action potential, Ca2+ entry via L-type Ca2+ channels triggers SR Ca2+ release by the ryanodine receptor, resulting in increases in intracellular Ca2+ concentration ([Ca2+]i) and myofilament activation. During diastole, most of the myoplasmic Ca2+ is resequestered in the SR by sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2). The amount of Ca2+ that has entered during systole is extruded by the Na+-Ca2+ exchanger (NCX1) and to a minor extent by sarcolemmal Ca2+-ATPase. It is the coordinated actions of L-type Ca2+ channels, SERCA2, NCX1, and to a lesser extent, sarcolemmal Ca2+-ATPase that allow the myocyte to maintain beat-to-beat Ca2+ balance.
In most animal models of heart failure as well as human heart failure, SR Ca2+ uptake activities and SERCA2 expression were depressed (13, 15). Focusing on the rat model of myocardial infarction (MI), Afzal and Dhalla (1) first showed depressed ATP-dependent Ca2+ uptake in SR membrane fractions isolated from rat left ventricles (LV) at 4, 8, and 16 wk after MI. Specifically, maximal velocity of ATP-dependent Ca2+ uptake by SR membranes was decreased, but affinity for Ca2+ was unchanged. Follow-up studies suggested the reduction in maximal velocity of SR Ca2+ uptake was due to decreases in SERCA2 mRNA and protein levels after an MI (27). Using a fundamentally different approach, we reported SR Ca2+ uptake activity, and SERCA2 protein levels were depressed in intact myocytes isolated from rat hearts 3 wk after MI (32). In addition, SR Ca2+ content in post-MI myocytes was significantly reduced when compared with sham-operated myocytes (35). Decreased SR Ca2+ content not only reduced the maximal amount of SR Ca2+ available for release during a twitch but also lowered the gain (ratio of trigger Ca2+ to Ca2+ released from the SR) of SR Ca2+-release channels (24). Decreased SR Ca2+ content in heart failure may be due to reduced SR Ca2+ uptake by SERCA2, enhanced Ca2+ extrusion by NCX1, or increased SR Ca2+ leak (3). The role of increased SR Ca2+ leak in causing decreased SR Ca2+ content in heart failure is at present controversial (3, 16, 17). In rat hearts studied 3 wk post-MI, we did not find any changes in SR Ca2+ leak compared with sham-operated hearts (32). In addition, in this rat post-MI model, NCX1 activities were also reduced rather than increased (11, 35). These findings, together with similar L-type Ca2+ current densities between sham and post-MI myocytes (30), suggest that reduced SR Ca2+ content was most likely due to decreased SERCA2 amounts and activities in post-MI myocytes. An attractive hypothesis is that reduced SERCA2 levels and activities can account for decreased SR Ca2+ content, resulting in abnormal [Ca2+]i transient (5, 29) and contractile behavior (5, 31) in post-MI rat myocytes. Because it is possible to transiently upregulate SERCA2 in rat myocytes by adenovirus-mediated gene transfer (12), the present study was undertaken to test the hypothesis that overexpression of SERCA2 in post-MI rat myocytes would restore contractile and [Ca2+]i transient abnormalities toward normal.
To induce MI, the left main coronary artery of each anesthetized (ketamine 60 mg/kg ip; xylazine 2 mg/kg ip), intubated, and ventilated (air) male Sprague-Dawley rat (weight ∼300 g) was ligated 3–5 mm distal to its origin from the ascending aorta (5, 23, 28–33, 35, 36). Sham operation was identical except that the coronary artery was not occluded. All surviving rats (∼30% perioperative mortality for MI and ∼0% for sham) received rat chow and water ad libitum and were maintained on a 12:12-h light-dark cycle. Survivors of coronary artery ligation typically had 36 ± 3% of myocardium infarcted as determined histologically (20). In addition, despite no overt signs of congestive heart failure (edema, ascites) in MI rats, infarcted hearts perfused in vitro had 20% lower LV systolic pressure at 1 and 3 wk after infarction (5). Three weeks after MI, after anesthetized with pentobarbital sodium (35 mg/kg body wt ip), hearts from sham-operated and MI rats were excised for myocyte isolation. Our previous studies indicate that myocyte maladaptations post-MI were well established by 3 wk (5, 29, 30–32, 35). The protocols for induction of MI and myocyte isolation were approved by the Institutional Animal Care and Use Committee.
Included in this study are a total of 10 sham-operated and 17 rats that had survived MI. For contraction studies, myocytes isolated from 7 sham and 13 MI rats were used. For [Ca2+]i transient studies, myocytes isolated from three sham and four MI rats were used. For Western blots, myocyte homogenates from six sham and seven MI rat hearts were used. Echocardiographic studies were performed on the first five sham-operated and five MI rats.
Three weeks after sham or MI operation, rats were anesthetized with isoflurane (3–5%) and oxygen (95–97%) via facemask. The left hemithorax was shaved, and a prewarmed ultrasound transmission gel was applied to the precordium. Transthoracic echocardiography was performed using an Agilent Sonos 5500 M2424A Ultrasound System (Andover, MA) equipped with a 12-MHz transducer (model 15-6L). The heart was first imaged in the two-dimensional mode in the parasternal long-axis view identifying the mitral and aortic valves and the apex. Short axis views were recorded at the level of midpapillary muscles. M-mode images obtained at the midpapillary muscle level were used for measurement of wall thickness and chamber dimensions. LV internal dimension at end-diastole (LVIDD) and at end-systole (LVIDS), as well as anterior and posterior wall thicknesses, were measured from M-mode tracings. Fractional shortening was calculated as (LVIDD − LVIDS)/LVIDD. LV end-diastolic (LVEDV) and end-systolic volumes (LVESV) were calculated by the formulae 1.047 × LVIDD3 and 1.047 × LVIDS3, respectively (21). LV stroke volume was defined as LVEDV − LVESV, and ejection fraction was calculated as (LVEDV − LVESV)/LVEDV. LV mass was given by the formula LV mass = 1.04[(anterior wall thickness + LVIDD + posterior wall thickness)3 − (LVIDD)3] (21). All measurements were averaged over three consecutive cardiac cycles. Data were analyzed offline using Access Point 2000 software (Freeland Systems LLC, Westfield, IN).
Construction of recombinant, replication-deficient adenovirus-expressing SERCA2.
The basic protocol is described by He et al. (14). Briefly, the coding sequence of rat heart SERCA2 (a generous gift from Dr. Jonathan Lytton at the University of Calgary) was released from pMT2 by sequential digestion with HindIII and SacI. The SERCA2 sequence (3,634 bp) was first subcloned into pSP73 vector (Promega, Madison, WI) using HindIII and SacI restriction sites and then released from pSP73 by digestion with HindIII and EcoRV. The released SERCA2 sequence was inserted into the shuttle vector pAdTrack-cytomegalovirus (CMV) with the same HindIII and EcoRV restriction sites on the shuttle vector. The SERCA2 sequence in pAdTrack-CMV was authenticated by sequencing. The shuttle vector plasmid was linearized with PmeI, mixed with supercoiled pAdEasy-1, and used to electroporate BJ5183 cells. The recombinants were identified by restriction endonuclease mapping (PacI). Once recombination was confirmed, supercoiled plasmid DNA was transformed into DH10B cells for large-scale amplification. The recombinant construct was linearized with PacI and used to transfect HEK293 cells. Recombinant adenovirus expressing both green fluorescent protein (GFP) and SERCA2 [each under a separate CMV promoter] (Adv-GFP-SERCA2) was harvested, purified by CsCl gradient centrifugation, and stored at −20°C in 5 mM Tris (pH 8.0), 50 mM NaCl, 0.05% bovine serum albumin, and 25% glycerol. The presence of SERCA2 coding sequence in recombinant adenovirus was confirmed by PCR with primers for the CMV promoter and the polyadenylation site of pAdTrack-CMV.
Myocyte isolation and culture.
Cardiac myocytes were isolated from the septum and LV free wall of rat hearts by successive perfusion with collagenase and hyaluronidase (6). Isolated myocytes were seeded on laminin-coated coverslips and cultured in modified serum-free medium 199 [extracellular Ca2+ concentration ([Ca2+]o) = 1.8 mM], as described previously (22, 23, 25, 33, 34). After 2 h, media were changed to remove nonadherent myocytes. The myocytes were incubated for an additional 3–4 h before initiation of pacing (1 Hz, 5-ms pulses of alternating polarity, field strength of 4 V/cm) using a custom-designed amplifier described previously (22, 34). In later experiments, the C-Pacer system (Ionoptix, Milton, MA) was used to pace myocytes in culture (1 Hz, 4-ms pulses of alternating polarity, field strength of 5.3 V/cm). We found both stimulators maintained contractile function of myocytes in culture equally well. Culture media were changed daily. We have previously demonstrated that continuous pacing in culture preserved normal contractile function of rat myocytes for at least 72 h after isolation (22, 23). In addition, we have also shown that MI myocytes maintained their contractile differences from sham myocytes even after 2–3 days of continuous pacing culture (23, 33).
Adenoviral infection of cardiac myocytes.
Two hours after isolation, myocytes seeded in four-well trays (Nuclone) were infected with either Adv-GFP-SERCA2 or adenovirus-expressing GFP alone (Adv-GFP) at a multiplicity of infection of two to five for 3–4 h. Media were then changed, and myocytes were studied after 72 h in continuous pacing culture. Over 95% of myocytes fluoresced green (excitation 478 nm, emission 535 nm) within 12–18 h, indicating successful adenoviral infection and GFP expression. We have previously shown that adenoviral infection did not affect contractility in sham and MI myocytes (23, 33). For brevity, sham myocytes infected with Adv-GFP are referred to as sham-GFP myocytes, and MI myocytes infected with Adv-GFP and Adv-GFP-SERCA2 are referred to as MI-GFP and MI-SERCA2 myocytes, respectively.
Myocyte shortening measurements.
Cell contraction was measured in myocytes incubated in HEPES-buffered (20 mM, pH 7.4) medium 199 (37°C), containing either 0.6, 1.8, or 5.0 mM [Ca2+]o, by using a charged-couple device video camera and edge-detection software (Ionoptix) as described previously (22, 23, 25, 31, 33, 34). For calibration of pixels vs. micrometer, a high-resolution test target (model 22-8635, Ealing Electro-Optics; Natick, MA) was used.
[Ca2+]i transient measurements.
Myocytes were exposed to 0.67 μM of fura 2-AM for 15 min at 37°C. Fura 2 epifluorescence (excitation 360 and 380 ± 10 nm bandpass; emission 510 ± 18 nm bandpass) from paced myocytes (1 Hz, 37°C) was measured with our quantitative fluorescence microscopy system described previously (22, 25, 28, 32–34) using an Olympus DApo Ultraviolet ×40/1.30 numerical aperature oil objective. Background and cellular autofluorescence measured in sham-GFP and MI-GFP myocytes not loaded with fura 2 accounted for <10% of the fura 2 signal (34). Intracellular fura 2 fluorescence was calibrated daily for each batch of myocytes, as previously described (7). [Ca2+]i transient data derived from fura 2 signals were analyzed with custom-written software (Ionoptix) (22, 25, 28, 29, 32–34).
SERCA2, NCX1, and calsequestrin immunoblotting.
Three days after adenovirus infection, cultured myocytes were rinsed three times with ice-cold phosphate buffered saline and then scraped into ice-cold lysis buffer as described previously (22, 23, 25, 33, 34). Cell lysates were snap-frozen with dry ice/ethanol and stored at −80°C. Myocyte lysates (50 μg) in SDS sample buffer were subjected to 7.5% polyacrylamide gel electrophoresis under either nonreducing (10 mM N-ethylmaleimide for NCX1) or reducing (5% β-mercaptoethanol for SERCA2 and calsequestrin) conditions. Fractionated proteins were transferred onto ImmunBlot polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA). NCX1 was detected with a mouse monoclonal antibody (1:1,000 dilution, R3F1, Swant; Bellinzona, Switzerland). SERCA2 was detected with another monoclonal antibody (1:2,500, MA3–919, Affinity Bioreagents; Golden, CO). Sheep anti-mouse antibody (1:2,000, Amersham; Piscataway, NJ) was used as the secondary antibody in both cases. To detect calsequestrin, a rabbit anti-calsequestrin antibody (1:5,000, Swant) and donkey anti-rabbit IgG (1:5,000, Amersham) were used. Immunoreactive proteins were detected with enhanced chemiluminescence-Western blotting system. Densitometric analysis was performed on a personal computer using the public domain National Institutes of Health Image Program (http://rsb.info.nih.gov/nih-image/).
All results are expressed as means ± SE. In experiments in which maximal contraction amplitudes and [Ca2+]i transient dynamics were measured as a function of group (sham-GFP vs. MI-GFP vs. MI-SERCA2) and [Ca2+]o, two-way ANOVA was performed to determine significances of differences. A linear model-fitted standard least squares analysis (JMP version 4.0.5, SAS Institutes; Cary, NC) was used. Significance of differences among the means of proteins (NCX1 and SERCA2) in the three experimental groups was determined by one-way ANOVA. A priori comparisons of means of any two groups were then performed by using F tests as tests of significance. Echocardiographic indexes of cardiac function in sham and MI rats were compared by Student's t-tests. In all analyses, P ≤ 0.05 was taken to be statistically significant.
Effects of MI on in vivo cardiac performance.
Three weeks after MI, there were no differences in body weight between sham and MI rats, although heart rate was significantly lower in MI rats (Table 1). Both LVIDD and LVIDS were significantly elevated in MI hearts, whereas fractional shortening was significantly reduced. MI affected neither anterior wall thickness nor posterior wall thickness. Both LVEDV and LVESV were significantly elevated in MI hearts, and ejection fraction was significantly lower. LV stroke volumes were similar between sham and MI hearts, as were cardiac outputs (sham 170 ± 12 ml/min; MI 152 ± 11 ml/min; P < 0.30) and cardiac indexes (sham 460 ± 36 ml·kg−1·min−1; MI 411 ± 30 ml·kg−1·min−1; P < 0.33). LV mass was similar between sham and MI rats.
Effects of MI and Adv-GFP-SERCA2 infection on NCX1 and SERCA2 levels.
In agreement with our laboratory's previous studies, which showed reduction in SERCA2 protein levels in 3-wk MI rat myocytes (32), SERCA2 levels were significantly lower in MI-GFP myocytes compared with sham-GFP myocytes (Fig. 1; Table 2). As expected, adenovirus-mediated SERCA2 overexpression restored the depressed SERCA2 levels in MI-GFP myocytes to normal. Although NCX1 abundance tended to be lower in MI-GFP myocytes, statistically there were no significant differences in NCX1 protein levels among the three experimental groups.
Effects of MI and SERCA2 overexpression on contractile function in cultured myocytes.
We have previously characterized contractile abnormalities in freshly isolated myocytes from 3-wk MI rat hearts (5, 31, 33). To reiterate, compared with sham myocytes, steady-state contraction amplitudes of MI myocytes were higher at 0.6, similar at 1.8, and lower at 5.0 mM [Ca2+]o. We have also shown that adenoviral infection had no effect on the MI contractile phenotype when examined after 48 (33) or 72 h (23) of continuous pacing culture. Our current studies confirmed our previous observations in that contractile differences between sham and MI myocytes were maintained 72 h after adenoviral infection (Fig. 2; Table 3). Two-way ANOVA showed absence of group (P > 0.06) but strong group × [Ca2+]o (P < 0.02) interaction effects, indicating that the magnitude and/or direction of the effects of [Ca2+]o on cell shortening was different between sham-GFP and MI-GFP myocytes.
Compared with MI-GFP myocytes, SERCA2 overexpression by adenovirus-mediated gene transfer in MI myocytes resulted in increases in steady-state contraction amplitudes at all [Ca2+]o examined (Fig. 2; Table 3). In addition, SERCA2 overexpression in MI myocytes resulted in contraction amplitudes that were even higher than sham-GFP myocytes at all [Ca2+]o. In all three groups of myocytes, increasing [Ca2+]o brought about the expected increased inotropy (Table 3), as confirmed by the highly significant [Ca2+]o effect (P < 0.0001).
Effects of MI and SERCA2 overexpression on [Ca2+]i transients in cultured myocytes.
As a group, diastolic [Ca2+]i levels in MI-GFP myocytes were significantly higher than those in sham-GFP myocytes across the range of [Ca2+]o examined (Fig. 3; Table 4). Overexpressing SERCA2 in MI myocytes decreased the elevated diastolic [Ca2+]i in MI-GFP myocytes at all three [Ca2+]o to levels indistinguishable from those measured in sham-GFP myocytes (Table 4). In all three groups, increasing [Ca2+]o increased diastolic [Ca2+]i (significant [Ca2+]o effect, P < 0.0001).
With respect to systolic [Ca2+]i, at low [Ca2+]o, values for MI-GFP myocytes were higher than those found for sham-GFP myocytes (Fig. 3; Table 4). At 5 mM [Ca2+]o, however, systolic [Ca2+]i was lower in MI-GFP myocytes compared with sham-GFP myocytes. There were no differences in systolic [Ca2+]i between the two groups at 1.8 mM [Ca2+]o. This conclusion is supported by the insignificant group (P > 0.17) but strong group × [Ca2+]o (P < 0.0001) interaction effects. SERCA2 overexpression in MI myocytes increased systolic [Ca2+]i at all [Ca2+]o examined, with the magnitude of increase larger at high [Ca2+]o (Fig. 3; Table 4). In fact, the levels of systolic [Ca2+]i achieved in MI-SERCA2 myocytes were even higher than those observed in sham-GFP myocytes. In all three groups, elevating [Ca2+]o resulted in higher systolic [Ca2+]i levels.
The magnitude of [Ca2+]i transient is reflected by the percent increase in fura 2 fluorescence intensity ratio. Similar to systolic [Ca2+]i, [Ca2+]i transient amplitudes in MI-GFP myocytes were higher at 0.6 mM [Ca2+]o but lower at 5.0 mM [Ca2+]o compared with sham-GFP myocytes (Fig. 3; Table 4). SERCA2 overexpression in MI myocytes increased [Ca2+]i transient amplitudes to levels that were even higher than those observed in sham-GFP myocytes (Table 4). In all three groups of myocytes, increasing [Ca2+]o resulted in the expected increases in [Ca2+]i transient amplitudes.
As a group, the half-time (t1/2) of [Ca2+]i decline was significantly longer in MI-GFP myocytes when compared with sham-GFP myocytes (Fig. 3; Table 4). SERCA2 overexpression in MI myocytes shortened t1/2 of [Ca2+]i decline to values similar to those observed in sham-GFP myocytes (Table 4). When [Ca2+]o was increased, which increased systolic [Ca2+]i in all three groups, t1/2 of [Ca2+]i decline in all three groups of myocytes was significantly lowered (Table 4). This observation is consistent with the findings of Bers and Berlin (2) that the kinetics of [Ca2+]i decline were dependent on peak [Ca2+]i.
Previous invasive hemodynamic studies on our rat MI model with moderate-size infarct (36 ± 3%) indicated that resting heart rate, cardiac output, and arteriovenous oxygen difference were similar between sham and MI rats (20). The normal cardiac output in MI rats was achieved at the expense of significantly elevated LVEDP, indicating a compensated heart failure state (20). Depressed cardiac function post-MI was clearly demonstrated when hearts were excised and perfused in vitro. Under conditions of constant left atrial filling pressure, afterload, and pacing frequency, MI hearts developed significantly lower LV systolic pressure and rate of rise of pressure (dP/dt) (5). In the present experiments using noninvasive echocardiography, we confirmed that our 3-wk rat MI model represented a state of compensated heart failure in that relatively normal cardiac output was maintained at higher LVEDV and LVESV. Contractility, as measured by fractional shortening or ejection fraction, was depressed in MI hearts. There was no significant myocardial hypertrophy, in agreement with other studies that employed a similar rat MI model (4, 19). Lack of increase in LV mass post-MI is also consistent with the mild (∼10%) increase in cell length but no change in cell width in myocytes isolated from MI rat hearts (5).
There are many factors that account for compromised cardiac function post-MI: loss of viable myocardium, altered chamber geometry, mismatched capillary supply, interstitial fibrosis, and myocyte maladaptations. At the cellular level, known excitation-contraction coupling abnormalities post-MI include prolongation of action potential (36), no change in L-type Ca2+ current density (30), decreased Na+-Ca2+ exchange activity (11, 35), decreased SR Ca2+ uptake activity and SERCA2 levels (13, 15, 32), reduced SR Ca2+ content (35), no alterations in SR Ca2+ leak rate (16, 32), shift of myosin heavy chain isoenzyme distribution from fast to slow forms, defective β-adrenergic receptor coupling (29), and altered [Ca2+]i transients (5, 29) and contractile behavior (5, 31). Focusing on SR Ca2+ regulatory systems, virtually all models of heart failure demonstrated decreased SERCA2 protein levels and SR Ca2+ uptake activities (13, 15). This observation, coupled with the recent finding that adenovirus-mediated SERCA2 transfer in neonatal rat myocytes resulted in larger [Ca2+]i transients and increased myocyte shortening (12), suggests that SERCA2 overexpression may be a rational therapeutic strategy in ischemic cardiomyopathy.
The signature contractile phenotype exhibited by our rat myocytes studied 3–9 wk after MI is that, when compared with sham myocytes, contraction amplitudes were higher at 0.6 mM, not different at 1.8 mM, but lower at 5 mM [Ca2+]o (23, 31, 33). This reduced dynamic range of inotropic response to increasing [Ca2+]o post-MI was still manifest in myocytes that had been subjected to 72 h of continued pacing culture and adenovirus infection. SERCA2 overexpression in MI myocytes resulted in increased contraction amplitudes at all three [Ca2+]o to levels that even surpassed those measured in sham myocytes. In particular, SERCA2 overexpression failed to correct the “supernormal” contraction amplitude at 0.6 mM [Ca2+]o in MI myocytes, indicating that adenovirus-mediated SERCA2 transfer did not totally ameliorate the abnormal contractile phenotype in MI myocytes.
Improvements in myocytes contraction amplitudes in MI-SERCA2 myocytes was most likely related to positive changes in [Ca2+]i transients. Compared with sham-GFP myocytes, MI-GFP myocytes had elevated diastolic [Ca2+]i and decreased SR Ca2+ uptake activity, as indicated by prolonged t1/2 of [Ca2+]i decline (32). Systolic [Ca2+]i and [Ca2+]i transient amplitudes were higher at 0.6 mM but lower at 5.0 mM [Ca2+]o in MI-GFP myocytes. These findings in cultured and adenovirus-infected myocytes are consistent with our previous observations on myocytes freshly isolated from sham and 3-wk MI hearts (29, 32). SERCA2 overexpression restored diastolic [Ca2+]i and SR Ca2+ uptake activity to levels observed in sham-GFP myocytes. In human failing myocytes, lowering of diastolic [Ca2+]i by SERCA overexpression was also observed (8). Importantly, the supernormal systolic [Ca2+]i levels and [Ca2+]i transient amplitudes measured at 0.6 mM [Ca2+]o in MI-GFP myocytes were exacerbated by SERCA2 overexpression. This observation provides additional support that SERCA2 overexpression after MI did not totally restore myocyte contractility and [Ca2+]i homeostasis to normal.
Overexpression of SERCA2 by adenovirus-mediated gene transfer has been used to rescue cardiac dysfunction in other models of heart failure. In myocytes isolated from failing human hearts (5 ischemic and 5 dilated cardiomyopathy), adenovirus-mediated SERCA2 transfer resulted in increased SERCA2 expression and SR Ca2+-ATPase activity (8). In addition, when examined at physiological [Ca2+]o, diastolic [Ca2+]i was decreased, systolic [Ca2+]i was increased, and maximal amplitude of contraction was enhanced by SERCA2 overexpression in failing myocytes. These results obtained in failing human myocytes are in agreement with our present study on post-MI rat myocytes. However, del Monte et al. (8) did not examine [Ca2+]i transient dynamics and contractility over a wider range of [Ca2+]o, and therefore potentially deleterious effects of SERCA2 overexpression (as unmasked at low [Ca2+]o levels in the present study) may not be evident. In a rat model of heart failure induced by aortic banding, SERCA2 overexpression was demonstrated to result in improvements in contractility, myocardial energetics, and animal survival (10). Heart failure by aortic banding was likely due to pressure-overload and the pathogenetic mechanisms of heart failure from pressure overload are likely different than those in chronic ischemic cardiomyopathy. A more recent model of transient ischemia (30 min) followed by reperfusion in the rat also benefited by prior SERCA2 overexpression (9). Specifically, decreased regional cardiac wall motion, reduced maximal rate of rise of LV pressure and relaxation, increased frequency of ventricular arrhythmias, and infarct size observed in ischemia/reperfusion rats were all improved by transient overexpression of SERCA2 before the onset of ischemia/reperfusion injury. By contrast, our present study utilized a chronic ischemic model in which SERCA2 gene transfer occurred weeks after the initial ischemic insult, and thus the results of the two studies are not directly comparable. In transgenic mice overexpressing SERCA2, mortality at 24 h after surgically induced MI was unacceptably high (71%) compared with wild-type mice (35%) despite similar infarct sizes, perhaps due to higher frequency of ventricular arrhythmias (4). Disappointingly, despite transient (up to 1 mo) improvement in systolic and diastolic function in the noninfarcted LV inferior wall of transgenic hearts, global LV function (ejection fraction, isovolumic relaxation time) measured at 1-wk, 1-mo, and 3-mo intervals was not different between wild-type and transgenic mice overexpressing SERCA2 that had survived surgically induced MI (4). It should be noted that, although adenovirus-mediated SERCA2 transfer resulted in two- (10) to fourfold (Table 2) increases in SERCA2 protein levels, SERCA2 expression was only increased by 37% in transgenic mice (4). This small increase in SERCA2 levels in transgenic hearts may account for the relatively modest effects on LV contractility post-MI. In any event, the results of published studies suggest that, whereas SERCA2 overexpression may indeed be beneficial in some models of heart failure (8–10), it may not entirely ameliorate cardiac dysfunction in chronic ischemic cardiomyopathic models in rats (present study) and mice (4). Indeed, caution should be exercised in recommending SERCA2 gene therapy for ischemic heart disease.
Of the major Ca2+ transport systems in the cardiac myocyte, arguably SERCA2 and NCX1 are quantitatively the most important. We have previously suggested that depressed NCX1 activity in post-MI rat myocytes accounted for much of the contractile and [Ca2+]i transient abnormalities (23, 25, 33). To reiterate, contractile and [Ca2+]i transient abnormalities in post-MI rat myocytes (5, 29, 31) mimicked those observed in normal rat myocytes in which NCX1 was downregulated by antisense exposure (25). In addition, contractile and [Ca2+]i transient abnormalities in MI myocytes were corrected, both at 0.6 and 5.0 mM [Ca2+]o, by NCX1 overexpression (23, 33). Furthermore, the beneficial effects of high-intensity sprint training on [Ca2+]i homeostasis (28) and contractility (23) in post-MI myocytes, part of which was mediated by restoring the depressed NCX1 activity, was totally abrogated by exposure to NCX1 antisense (23). The results of the present study demonstrate that SERCA2 overexpression in post-MI rat myocytes did not totally correct [Ca2+]i transient and contractile abnormalities and support our hypothesis that decreased NCX1 activity, rather than reduced SERCA2 or its activity, played a more important role in contractile abnormalities in our post-MI rat model. In this context, it is interesting to note that, 5 wk post-MI, transgenic mice overexpressing NCX1 maintained higher LV systolic pressure and maximal dP/dt lower LV end-diastolic pressure than wild-type mice, despite similar infarct sizes (18). In addition, there were no significant ventricular arrhythmias in both wild-type and NCX1-overexpressing mice when examined 2 days post-MI (18).
We have previously demonstrated that overexpression of NCX1 did not affect SERCA2 expression in post-MI rat myocytes (23). In the present study, SERCA2 overexpression did not affect NCX1 expression in MI myocytes. This observation suggests that the effects of SERCA2 overexpression on contractility and [Ca2+]i transients in MI myocytes was unlikely to be mediated by changes in NCX1 levels.
There are limitations to the study. The first is that echocardiographic evaluation was not performed on all of the experimental rats used in this study. In addition, post-MI hearts had significantly lower heart rates compared with sham-operated hearts. These slower heart rates found in the MI rats may potentially lead to a longer diastolic time interval that could be contributing to the greater LVEDV found in these animals. The second limitation is that we did not measure levels of phospholamban, which could affect SERCA2 activity. We note, however, that phospholamban mRNA levels were unchanged in a tachycardia-induced heart failure model in the dog (13). In addition, phospholamban protein levels were similar in failing and nonfailing human myocardium (13) and between sham and MI rat myocytes (28). We also note that adenoviral-mediated SERCA2 transfer in rat cardiac myocytes did not affect phospholamban expression (12). It is thus reasonable to assume that, in the present study, the effect on contractility and [Ca2+]i homeostasis were due to SERCA2 overexpression rather than secondary changes in phospholamban levels. A third limitation is that, in this study, we focused on the role of SERCA2 on contraction abnormalities in post-MI rat myocytes and ignored the other important steps involved in excitation-contraction coupling that are altered in post-MI myocytes (11, 13, 15, 30, 35, 36). The importance of alterations in these other pathways (e.g., prolonged action potential duration, shift in myosin heavy chain isoenzyme distribution, etc.) in causing contractile dysfunction in post-MI myocytes needs to be addressed in future studies.
In summary, we have established an in vitro myocyte culture model system in which phenotypic differences between sham and post-MI myocytes were maintained for at least 72 h after isolation. Overexpression of SERCA2 in post-MI myocytes by adenovirus-mediated gene transfer, although improving contractile dysfunction at high [Ca2+]o, did not correct the supernormal contraction and [Ca2+]i transient amplitudes at low [Ca2+]o. We suggest that decreased SERCA2 activity played an important, but not dominant role, in abnormal contractility in postinfarction rat myocytes.
This study was supported in part by the National Institutes of Health Grants HL-58672 (J. Y. Cheung), DK-46078 (J. Y. Cheung, coinvestigator); American Heart Association Pennsylvania Affiliate Grants-in-Aid 0265426U (X. Zhang) and 0355744U (J. Y. Cheung); American Heart Associate Pennsylvania Affiliate Postdoctoral Fellowship 0425319U (B. A. Ahlers); and by grants from the Geisinger Foundation (J. Y. Cheung and L. I. Rothblum).
We thank Caitlin Custer for assistance in the preparation of the manuscript.
Present address of G. M. Tadros: Division of Cardiovascular Diseases, University of Minnesota, Minneapolis, MN 55455.
↵* B. A. Ahlers and J. Song contributed equally to this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society