Adenovirus-Mediated Gene Transfer Reconstitutes Depressed Sarcoplasmic Reticulum Ca2+-ATPase Levels and Shortens Prolonged Cardiac Myocyte Ca2+ Transients
Background Decreased expression of the sarcoplasmic reticulum (SR) Ca2+-ATPase of the cardiac myocyte (SERCA2) and abnormal Ca2+ regulation have been independently linked to human heart failure. This study was designed to determine whether expression of a SERCA2 transgene could reconstitute depressed cardiac myocyte SERCA2 levels, augment SR Ca2+ uptake, and shorten prolonged excitation-contraction (EC)–associated Ca2+ transients in neonatal rat cardiac myocytes (NM).
Methods and Results Cultured NM were treated with phorbol-12-myristate-13-acetate (PMA), a compound that decreases endogenous SERCA2 expression and results in prolongation of EC-associated Ca2+ transients. PMA-treated NM had a 75% reduction in SERCA2 mRNA and a 40% reduction in SERCA2 protein levels. SERCA2 adenovirus infection increased SERCA2 mRNA expression to 2.5 times control and reconstituted SERCA2 protein levels in PMA-treated cells. This reconstitution was associated with a 32.4% reduction in the time for decline of the Indo-1 Ca2+ transient to half-maximum levels (t1/2 [Ca2+]i) (P<.05). A 34.5% augmentation of oxalate-facilitated SR Ca2+ uptake was also documented in SERCA2 adenovirus–infected cells (P<.05).
Conclusions Adenovirus-mediated expression of a SERCA2 transgene can reconstitute depressed endogenous SERCA2 levels, shorten prolonged Ca2+ transients, and augment SR Ca2+ uptake. It is conceivable that such an approach might be used in vivo to normalize altered Ca2+ regulation in human heart failure.
Dynamic regulation of the Ca2+ concentration within cardiac myocytes is critical to the normal contractile function of the heart.1 A major system regulating rapid intracellular Ca2+ shifts in cardiac myocytes is the SERCA2 pump.2 SERCA2 is responsible for reuptake of Ca2+ into the SR and is thought to have a major effect on diastolic relaxation. Additionally, by providing the SR with an abundant store of Ca2+ for excitation-coupled Ca2+ release, SERCA2 may significantly affect systolic function. In pressure overload–associated cardiac hypertrophy induced by aortic banding, SERCA2 mRNA levels decrease concomitant with the development of left ventricular dysfunction.3 Studies of tissue from failing human hearts have demonstrated decreased SERCA2 mRNA and protein levels as well as slower Ca2+ uptake into SR vesicles.4 5 Although supportive of an association between decreased SERCA2 levels and cardiac dysfunction, these studies do not yet prove a functional relationship.
In vitro studies have demonstrated a downregulation of SERCA2 mRNA and protein levels in response to treatment of NM with the phorbol ester PMA.6 7 This downregulation appears to be at least partially related to protein kinase C activation and to be mediated by changes in SERCA2 message stability.7 Importantly, this downregulation correlates with a significant alteration in cardiac myocyte Ca2+ handling marked by a prolongation of Indo-1 Ca2+ transients. The ability of SERCA2 overexpression to shorten Ca2+ transients in neonatal cardiac myocytes with normal baseline SERCA2 levels has recently been shown.8 The goals of our study were to determine whether adenovirus-mediated expression of SERCA2 could rescue reduced SERCA2 mRNA and protein levels and whether this intervention could normalize the prolonged Ca2+ transients associated with PMA-induced SERCA2 downregulation.
Recombinant E1A/E1B-deleted SERCA2–encoding adenovirus was constructed using a standard protocol.9 The promoter construct 14CTK, which contains tandem simian virus 40 enhancer sequences 5′ to the thymidine kinase promoter, was used to drive expression of a rat SERCA2 cDNA in this adenovirus construct.10 Viral DNA was isolated and construction of SERCA2 adenovirus recombinants was verified by Southern blot using a full-length rat SERCA2 cDNA probe. A positive isolate (AdSERCA2) was propagated to high titer in 293 cells, purified by double CsCl gradient ultracentrifugation, and column filtered through Sepharose CL4B. Final stocks were titered both spectrophotometrically and by plaque titer. The control (empty) virus AdSR− was constructed similarly and contains no insert between cytomegalovirus promoter and simian virus 40 polyadenylation sequences.
Cell Culture and Treatment
Primary cultures of rat ventricular neonatal myocytes were prepared as previously described.6 10 Cells were plated at a density of 25 000/cm2 onto 10-cm tissue culture dishes precoated with 1% (wt/vol) gelatin or onto glass coverslips coated with fibronectin. Cells were allowed to adhere to the dishes (or coverslips) for 24 hours. The medium was then changed to serum-free medium. For SERCA2 mRNA expression, SR Ca2+ uptake, and Indo-1 Ca2+ transient studies on NM not treated with PMA, cells were infected with AdSERCA2 or AdSR− at a multiplicity of infection (MOI) of ≈5:1 or sham infected. At 48 hours later, cells were either scrapped for RNA preparation, prepared for 45Ca2+ uptake studies, or used for Indo-1 Ca2+ transients. For SERCA2 downregulation and reconstitution studies, cells were treated with PMA (Sigma Chemical) at a final concentration of 200 ng PMA/mL (300 nmol/L).6 Control plates were treated with dimethylsulfoxide vehicle. At 24 hours after PMA treatment, the cells were infected with either AdSERCA2 or AdSR− (MOI 5:1) or sham infected. At 4 hours after infection, the medium was exchanged with fresh serum-free medium in the dishes and the cells were treated again with PMA or vehicle. At 48 hours later (72 hours after PMA), plated cells were scrapped for Western and Northern blot analysis, and cells on coverslips were used for measurement of Indo-1 Ca2+ transients. After measurement of Ca2+ transients, adenovirus transfection of cells used for Indo-1 studies was documented by standard immunostaining with a polyclonal antibody against Ad hexon protein (Chemicon Inc).
Northern and Western Blot Analyses
Isolation of total RNA from NM was performed as described by Chomszynski and Sacchi.11 Standard protocols were used for electrophoresis, Northern transfer, and hybridization.12 SERCA2 transgene expression was characterized using a 1.8-kb EcoRI restriction fragment corresponding to the 5′-end of the rat SERCA2 cDNA. For Western analysis, a rabbit polyclonal anti-SERCA2 antibody (SERCA2 Ab) was used, and cellular homogenates were prepared as previously described.6 Immunoblotting was performed with a 1:500 dilution of SERCA2 Ab, and an enhanced chemiluminescence kit (Amersham) was used for signal detection. Quantification was done using video images of the blots and Macintosh software for image analysis (IMAGE, NIH; same for Northern analysis).
Indo-1 Ca2+ Transients
Myocytes were incubated with Indo-1/AM (10 mmol/L) for 15 to 30 minutes at 37°C as previously described.13 Fluorescence measurements were performed at room temperature or 37°C using methods and apparatus (Photon Technologies) previously described.13 Cells were stimulated to contract using platinum electrodes in 1 ml Tyrode’s solution in a Biophysica coverslip chamber. The characteristics of Ca2+ transients or [Ca2+]i of cells were represented as the duration of the time for intracellular Ca2+ to decrease from maximum to half-maximum (t1/2 [Ca2+]i).
SR Ca2+ Uptake Determination
Oxalate-facilitated SR Ca2+ uptake measurements were performed in cellular homogenates by a protocol derived from the method of Pagani and Solaro14 using 45Ca2+. Radioactivity was determined by liquid scintillation spectroscopy and protein concentration assayed with a Bradford reagent. Ca2+ uptake was calculated from the slope of the linear regression analysis relating 45Ca2+ uptake per milligram of total protein to reaction time. Linearity of the uptake was confirmed up to 20 minutes.
Data and Statistical Analyses
Data were statistically analyzed with either an ANOVA/Newman-Keuls (Indo-1 data) or Student’s unpaired t test. A value of P≤.05 was considered statistically significant. Functional measurements of Ca2+ handling were blinded at the time of data acquisition. A minimum of three independent experiments were performed for all components of the study. For the Indo-1 studies, n refers to number of coverslips studied per condition, with 15 to 20 cells per coverslip evaluated.
Adenovirus-Mediated Expression of SERCA2
To document transcription of the SERCA2 transgene encoded in AdSERCA2, Northern blot analysis of total RNA from infected NM was performed. The SERCA2 transgene in AdSERCA2 is truncated by 200 bp in the 5′ transcribed untranslated region. Northern blots probed with a 1.8-kb SERCA2 cDNA fragment demonstrated expression of the expected truncated SERCA2 mRNA in all AdSERCA2-infected NM (Fig 1A⇓). To document the ability of the SERCA2 adenovirus to reconstitute downregulated SERCA2 mRNA levels, NM were treated with PMA and infected with AdSERCA2. As shown in Fig 1B⇓, PMA treatment resulted in a reduction in SERCA2 mRNA levels to 25% of controls. Infection with AdSERCA2 resulted in reconstitution and augmentation of SERCA2 mRNA expression 2.5 times that of untreated controls.
To document translation of the AdSERCA2-encoded SERCA2 mRNA and to investigate the ability of AdSERCA2 to reconstitute downregulated endogenous SERCA2 protein levels, Western blot analysis of total cellular protein from PMA-treated NM infected with either AdSERCA2 or AdSR− was performed. As shown in Fig 1C⇑, PMA-treated cells demonstrated a 40% reduction in SERCA2 protein levels that was not affected by infection with AdSR−. Infection with AdSERCA2 resulted in reconstitution of SERCA2 protein levels (1.1 times control) in the PMA-treated cells within 48 hours of infection.
Effects of Modulated SERCA2 Expression on Ca2+ Transients and SR Ca2+ Uptake
To investigate the effects of adenovirus-mediated SERCA2 expression on Ca2+ handling in NM, Indo-1 fluorescent Ca2+ transients were measured. In AdSERCA2-infected NM, the time for decline of the Indo-1 Ca2+ transient to half-maximum was reduced by 32.6% relative to controls (P<.001), documenting the ability of SERCA2 gene transfer to alter baseline Ca2+ handling (Fig 2A⇓). As reported, PMA treatment of NM results in a prolongation of the t1/2 [Ca2+]i.7 8 At 72 hours after initiation of PMA treatment, NM infected with AdSERCA2 demonstrated an average t1/2 [Ca2+]i 32.4% briefer than that of uninfected PMA-treated cells (P<.05). The t1/2 [Ca2+]i of PMA-treated NM infected with AdSR− was not different than that of the uninfected PMA-treated cells (Fig 2B⇓). Immunostaining using an anti-hexon antibody demonstrated near 100% Ad infection of the myocytes used for these studies (data not shown).
A number of factors can potentially affect the duration of the Indo-1 Ca2+ transient. To more directly investigate the contribution of adenovirus-mediated SERCA2 expression on SR function, oxalate-facilitated SR 45Ca2+ uptake was measured in NM at 48 hours after infection with AdSERCA2 or AdSR−. 45Ca2+ uptake (nmol · mg−1 · min−1) revealed significantly augmented SR Ca2+ sequestration in AdSERCA2-infected NM (27.7±0.5, n=9) relative to uninfected controls (21.9±0.7, n=8, P<.05) and cells infected with AdSR− (21.9±2.6, n=8, P<.05; Table⇓).
This study demonstrates adenovirus-mediated expression of a SERCA2 transgene in cardiac myocytes, resulting in shortening of Indo-1 Ca2+ transients and specific augmentation of SR Ca2+ uptake. Furthermore, we show for the first time that adenovirus-mediated SERCA2 expression can reconstitute depressed SERCA2 protein levels and shorten prolonged Indo-1 Ca2+ transients by decreasing t1/2 [Ca2+]i. This effect is of particular interest considering that the pathophysiology of some forms of human heart failure may involve reduced SERCA2 levels with consequent abnormalities in dynamic Ca2+ handling. The SERCA2 downregulation that occurs in response to phorbol ester is mediated predominantly posttranscriptionally.7 Although the mechanism of SERCA2 reduction in human heart failure has not been elucidated, the possibility that it too is mediated posttranscriptionally has important implications pertaining to strategies to ameliorate heart failure by transcriptional overexpression of a SERCA2 transgene. We show here that constitutive overexpression of a SERCA2 transgene can overcome posttranscriptional downregulation and reconstitute SERCA2 levels.
The mechanism of the effects of AdSERCA2 on Indo-1 Ca2+ transients is presumably a transgenemediated increase in the total number of functioning SERCA2 pumps in the SR. Although Indo-1 Ca2+ transients can be affected by a number of factors, the oxalate-facilitated Ca2+ uptake data are highly supportive that the shortening of t1/2 [Ca2+]i is at least partially due to an augmentation in SERCA2 function. Although it has been shown that an increase in [Ca2+]i can result in a decrease in the duration of Indo-1 Ca2+ transients,15 basal and peak Indo-1 ratios in this study demonstrate no significant differences in control versus AdSERCA2-treated cells. Therefore, it is unlikely that the shortening of t1/2 [Ca2+]i is a result of alterations of peak [Ca2+]i independent of augmentation of SR Ca2+ uptake. The present study demonstrates a strong correlation between induced SERCA2 expression and dynamic intracellular Ca2+ regulation as tested by two separate and complementary methods of analysis: oxalate-facilitated Ca2+ uptake and Indo-1 fluorescent Ca2+ transients. Although this does not prove a correlation between SERCA2 expression and in vivo contractile function, it is additional evidence suggesting an important relationship.
Recombinant adenovirus infection is a highly efficient method of in vitro transfection of cardiac myocytes.16 Although generally not as efficient for this purpose in vivo, we recently demonstrated the efficacy of this approach to induce functional changes and angiogenesis in an animal model of myocardial ischemia.9 In that study, a secreted protein was encoded by the recombinant adenovirus. Conceivably, greater transfection efficiency than was required for an effect by a secreted protein would be needed to demonstrate a functional effect in vivo for an intracellular protein such as SERCA2. However, as gene transfer technology advances, it can be envisioned that in vivo modulation of the expression of proteins such as SERCA2 might have a significant impact on the treatment of human heart failure. In fact, intracellular proteins such as SERCA2 may be the most practical targets for cardiovascular gene therapy, given that secreted proteins such as growth factors can be given directly, avoiding the potential safety concerns of in vivo gene transfer.
In summary, this study demonstrates the feasibility of adenovirus-induced gene transfer to mediate SERCA2 transgene expression, reconstitute decreased SERCA2 levels in cardiac myocytes, increase SR Ca2+ uptake, and shorten prolonged intracellular Ca2+ transients. These findings constitute further evidence of a direct functional relationship between SERCA2 levels and dynamic regulation of Ca2+ in cardiac myocytes. In vivo studies with this adenovirus construct might add further to understanding the relationship between SERCA2 levels and cardiac contractile function.
Selected Abbreviations and Acronyms
|[Ca2+]i||=||intracellular Ca2+ concentration|
|NM||=||neonatal rat cardiac myocytes|
|SERCA2||=||sarcoplasmic reticulum Ca2+-ATPase of the cardiac myocyte|
This work was supported by National Institutes of Health grant HL-52946 (Dr Dillmann) and postdoctoral fellowship grant F32-HL-08711 (Dr Giordano).
- Received November 4, 1996.
- Revision received May 19, 1997.
- Accepted May 20, 1997.
- Copyright © 1997 by American Heart Association
Feldman AM, Weinberg EO, Ray PE, Lorell BH. Selective changes in cardiac gene expression during compensated hypertrophy and the transition to cardiac decompensation in rats with chronic aortic banding. Circ Res. 1994;73:184-192.
Meyer M, Schillinger W, Pieske B, Holubarsch C, Heilmann C, Posival H, Kuwajina G, Mikoshiba K, Just H, Hasenfuss G. Alterations of sarcoplasmic reticulum proteins in failing human dilated cardiomyopathy. Circulation. 1995;92:778-784.
Qi M, Bassani WM, Bers DM, Samarel A. Phorbol 12-myristate 13-acetate alters SR Ca2+-ATPase gene expression in cultured neonatal rat heart cells. Am J Physiol. 1996;271:H1031-H1039.
Hajjar RJ, Kang JX, Gwathmey JK, Rozenzweig A. Physiological effects of adenoviral gene transfer of sarcoplasmic reticulum calcium ATPase in isolated rat myocytes. Circulation. 1997;95:423-429.
Rohrer DK, Hartong R, Dillmann WH. Influence of thyroid hormone and retinoic acid on slow sarcoplasmic reticulum Ca2+ ATPase and myosin heavy chain, a gene expression in cardiac myocytes. J Biol Chem. 1990;266:8638-8646.
Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning, 2nd ed. New York, NY: 1989.
McDonough P, Salvatore LS, Glembotski CC. Involvement of cytoplasmic calcium and protein kinases in the regulation of atrial natriuretic factor secretion by contraction rate and endothelin. J Biol Chem. 1994;269:9466-9472.
Pagani ED, Solaro RJ. Methods for measuring functional properties of sarcoplasmic reticulum and myofibrils in small samples of myocardium. Methods Pharmacol. 1984;5:2542-2547.
Kohmoto O, Ikenouchi H, Hirata Y, Momomura S, Serizawa T, Barry WH. Variable effects of endothelin-1 on [Ca2+]i transients, pHi, and contraction in ventricular myocytes. Am J Physiol. 1993;265:H793-H800.
Kirshenbaum LA, MacLellan WR, Mazur W, French BA, Schneider MD. Highly efficient gene transfer into adult ventricular myocytes by recombinant adenovirus. J Clin Invest. 1993;92:381-387.