Progesterone modulates SERCA2a expression and function in rabbit cardiomyocytes

Abstract

We recently showed that progesterone treatment abolished arrhythmias and sudden cardiac death in a transgenic rabbit model of long QT syndrome type 2 (LQT2). Moreover, levels of cardiac sarco(endo)plasmic reticulum Ca²⁺-ATPase type 2a (SERCA2a) were upregulated in LQT2 heart extracts. We hypothesized that progesterone treatment upregulated SERCA2a expression, thereby reducing Ca²⁺-dependent arrhythmias in LQT2 rabbits. We therefore investigated the effect of progesterone on SERCA2a regulation in isolated cardiomyocytes. Cardiomyocytes from neonatal (3- to 5-day-old) rabbits were isolated, cultured, and treated with progesterone and other pharmacological agents. Immunoblotting was performed on total cell lysates and sarcoplasmic reticulum-enriched membrane fractions for protein abundance, and mRNA transcripts were quantified using real-time PCR. The effect of progesterone on baseline Ca²⁺ transients and Ca²⁺ clearance was determined using digital imaging. Progesterone treatment increased the total pool of SERCA2a protein by slowing its degradation. Using various pharmacological inhibitors of degradation pathways, we showed that progesterone-associated degradation of SERCA2a involves ubiquitination, and progesterone significantly decreases the levels of ubiquitin-tagged SERCA2a polypeptides. Our digital imaging data revealed that progesterone significantly shortened the decay and duration of Ca²⁺ transients. Progesterone treatment increases protein levels and activity of SERCA2a. Progesterone stabilizes SERCA2a, in part, by decreasing the ubiquitination level of SERCA2a polypeptides.

inherited long QT (LQT) syndrome (LQTS) is characterized by a prolonged QT interval, polymorphic ventricular tachycardia, and sudden cardiac death (SCD) (22, 24). Together, LQT1 and LQT2 account for 55–65% of LQTS cases. LQT1 is associated with various point mutations in the slow delayed rectifier K⁺ channel (KCNQ1), which diminishes slow delayed rectifier K⁺ currents (I_Ks), thereby slowing repolarization and leading to prolongation of cardiac action potential duration (APD). LQT2 is associated with a point mutation in the α-subunit of the rapid delayed rectifier K⁺ channel [human ether-à-go-go (hERG)], which compromises I_Kr, the current responsible for early termination of action potential, leading to APD prolongation. LQT2 patients are less susceptible to SCD during pregnancy than postpartum (15, 28). Moreover, in drug-induced LQTS, the risk of polymorphic ventricular tachycardia is higher during menses and the follicular phase, when plasma levels of estradiol are high, than during the luteal phase, when progesterone levels are high (25). These studies suggest that sex hormone may influence arrhythmogenesis in LQTS. Indeed, our recent in vivo study (20) utilizing ovariectomized prepubertal transgenic LQT2 female rabbits showed, for the first time, the direct link between sex hormones and the incidence of arrhythmias and SCD. The study demonstrated that progesterone treatment is protective, preventing SCD in LQT2.

The mechanisms by which sex hormones exert their electrophysiological effect at the cardiomyocyte level remain elusive. Wu et al. (31) documented that treatment of primary neonate rat cardiomyocytes with large amounts of progesterone (5 μM) inhibits hERG trafficking, thereby decreasing hERG channel expression levels on the membrane and hERG current densities. Their results indicate that the effect of progesterone is not mediated by progesterone nuclear receptors but, rather, is nongenomic. Progesterone treatment under basal conditions shortened cardiac APD in guinea pig cardiomyocytes, an effect attributable to enhancement of I_Ks and slowed L-type Ca²⁺ currents. This effect of progesterone is mediated by nitric oxide released by endothelial cells and has been suggested to be nongenomic (19). Sex hormones also regulate the function of other ion channel proteins. For example, Na⁺/Ca²⁺ exchanger (NCX) pump current increase in the presence of 17β-estradiol in isolated cardiomyocytes from female rabbits is blunted by an estrogen receptor antagonist, implying a genomic effect of estradiol. This effect is specific to cardiomyocytes isolated from the base of female hearts, but not from the apex or from male hearts (5).

The sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA2a) pump is a core Ca²⁺-cycling protein that plays a critical role in excitation-contraction coupling in the heart. Downregulation of SERCA2a at the level of the mRNA transcript (8, 27, 29), protein (16, 18, 29), or activity (8, 27, 32) has been well documented in experimental models of arrhythmia-associated heart failure (HF) and SCD. In this regard, there is considerable interest in SERCA2a gene therapy in HF. Recently, it has been shown that overexpression of SERCA2a restores the Ca²⁺ transient and improves contraction and relaxation in failing human cardiomyocytes (7), decreases the incidence of ventricular arrhythmias in HF (26), and reduces the incidence of aftercontractions and afterdepolarizations in rabbit cardiomyocytes (6).

It has been suggested that posttranslational modifications of SERCA2a regulate its expression levels and function. For example, oxidative stress posttranslational modification impairs SERCA2a pump function in HF (17). Similarly, the pump function of SERCA2a is compromised by glutathiolation (1). More recently, it has been shown that SERCA2a expression and function are compromised in HF, in part by imbalance between SUMOylation and ubiquitination (14). Kho et al. (14) demonstrate that SUMOylation of SERCA2a (addition of SUMO1 at lysine residues within the ATP-binding domain in SERCA2a) is a prerequisite for stability and pump activity. They were able to ameliorate heart dysfunction induced by pressure overload in SERCA2a or SUMO1 transgenic animals. However, heart function was not restored in SUMO1 transgenic animals in which SERCA2a expression was knocked down.

Our recent study pointed out an association between the antiarrhythmic effect of progesterone and upregulation of SERCA2a levels in LQT2 rabbit hearts (20). However, progesterone's regulation of SERCA2a expression at the cellular and molecular levels in cardiomyocytes from large animals is not well understood. Therefore, the present study is mechanistically designed to elucidate the effect of progesterone on SERCA2a regulation in rabbit cardiomyocytes. We used neonate rabbit cardiomyocytes (NRbCM) because the endogenous hormone effect is minimal and progesterone treatment in culture can be easily controlled.

MATERIALS AND METHODS

NRbCM culture.

All animal experiments and procedures were approved by the Rhode Island Hospital Institutional Animal Care and Use Committee. Cardiomyocytes were isolated from the hearts of neonatal (3- to 5-day-old) rabbits (New Zealand White) of either sex by enzymatic digestion using collagenase I (0.3 mg/ml), as described previously (21). Briefly, hearts were excised, mounted on a Langendorff system, and perfused with a Ca²⁺-free solution followed by collagenase I. Ventricles were separated and minced, and cell suspensions were prepared. Cells were seeded at a density of 1 × 10⁶ cells/mm² on laminin-coated dishes and grown in DMEM supplemented with 7% FBS and 5-bromo-2′-deoxyuridine for 4 days. Cardiomyocyte cultures were ≥80–90% pure as evaluated by α-actinin staining. On day 4 of culture, the medium was changed to DMEM supplemented with 7% charcoal-activated FBS (to strip the hormones and ligands) and treated with progesterone and other drugs at the indicated concentrations and time points.

Treatment with pharmacological agents.

To evaluate the involvement of calpains, proteasomes, and autophagy in the degradation of SERCA2a, we pretreated the cardiomyocytes in the presence of specific inhibitors for each degradation pathway. To inhibit the calpain degradation pathway, we used 20 μM E-64d; to inhibit proteasomes, we used 25 μM clasto-lactacystin-β-lactone (Lcy) or 5 μM MG132. These concentrations effectively blocked proteasomal activity, as assessed by accumulation of ubiquitinated proteins detected in Western blots, similar to findings reported elsewhere. To inhibit macroautophagy, we used 10 mM 3-methyladenine (3-MA).

Preparation of cell lysate for Western blotting and immunoprecipitation.

Protein was extracted and processed for Western blotting as described previously (21). Briefly, the medium was removed and cells were washed with ice-cold PBS before addition of lysis buffer [1 Complete Mini Protease Inhibitor tablet (Roche), 50 mM PMSF, and 1× phosphatase inhibitor (Bio-Rad) in 10 ml of RIPA buffer (Invitrogen)]. Protein lysates were prepared by centrifugation at 14,000 g. The protein concentration was determined using a Bio-Rad protein assay. The lysates were processed immediately or stored at −80°C. Western blot experiments on total protein extracts or crude membrane extracts of NRbCM were performed using the primary antibodies anti-SERCA2a (catalog no. MA3-919, Thermo Scientific), anti-phospholamban (PLB; catalog no. MA3-922, Thermo Scientific), anti-phosphorylated (Ser¹⁶) PLB (Thermo Scientific), anti-phosphorylated (Thr¹⁷) PLB (Santa Cruz Biotechnology), and anti-NCX (catalog no. MA3-926, Thermo Scientific) and corresponding horseradish peroxidase-conjugated secondary antibodies. After Western blot transfer, the PVDF membranes were stained with 1% Ponceau S red stain prepared in acetic acid to confirm transfer and equal loading. Developed films and Ponceau S-stained blots were scanned using an HP ScanJet 8300 scanner with resolution set to 600 pixels/inch. The scans were saved as TIF files and imported into the ImageJ software package, with samples normalized to the corresponding GAPDH levels. Protein standards were run to ensure that the quantified protein bands were within the linear range for analysis. The relative band intensities of phosphorylated (Ser¹⁶ and Thr¹⁷) PLB were densitometrically quantified and normalized to total PLB levels. For immunoprecipitation assays, cell lysates (100 μg/ml) were diluted with 500 μl of lysis buffer supplemented with the above-mentioned protease inhibitors. Then the lysate was precleared by incubation with 75 μl of protein A-Sepharose beads for 15 min, and the beads were removed. The precleared lysates were incubated with 10 μg of SERCA2a antibody or with isotype-matching control antibody (IgG) for 4 h at 4°C. The immunocomplexes were captured by addition of 75 μl of protein A-Sepharose beads, and the mixture was incubated overnight at 4°C. The beads were pelleted by centrifugation, washed three times in cold lysis buffer, and then resuspended in 2× SDS loading buffer. The samples were boiled for 10 min, and proteins in the resulting supernatants were separated on a 10% SDS gel and analyzed for ubiquitin by Western blotting.

Sarcoplasmic reticulum membrane enrichment.

Sarcoplasmic reticulum (SR) membranes were prepared from cultured NRbCM by ultracentrifugation, as described elsewhere (12) with slight modifications. Briefly, cells (from a 10-cm dish) were washed three times with cold PBS (pH 7.4) and collected in 1 ml of buffer 1 (20 mM PIPES, pH 7.4, 300 mM sucrose, 1 mM PMSF, 1× protease cocktail, and 1.5× phosphatase inhibitor) in Eppendorf tubes with resin (sample grinder kit, GE). Cells were centrifuged at 1,800 rpm at 4°C for 6 min. The supernatant was removed, and the cells were resuspended in 400 μl of buffer 1. Cells were disrupted mechanically using a homogenizer (30–35 strokes) on ice and centrifuged at 500 g at 4°C for 25 min. The supernatant was collected in ultracentrifuge-compatible 1.5-ml Eppendorf tubes and centrifuged at 25,000 g at 4°C (20,000 rpm; Beckmann Coulter) for 1 h. The pellet represented the SR membrane fraction (the supernatant represented the cytosolic protein). The pellet was resuspended in 100–200 μl of buffer 2 (20 mM PIPES, pH 7.4, 30 mM sucrose, 600 mM KCl, 1 mM PMSF, 1× protease cocktail, and 1.5× phosphatase inhibitor). The resuspended fractions (including the cytosolic fraction) were frozen in liquid nitrogen overnight and then transferred to −70°C until further use. Protein was quantitated using the Bradford assay.

Quantitative RT-PCR.

Total RNA was extracted from left ventricles of LQT2 female rabbits treated with progesterone (20) and NRbCM using the RNeasy Mini kit (Qiagen, Valencia, CA), and cDNA was synthesized using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems). Gene expression was detected by quantitative PCR using TaqMan probes and 2× Gene Expression Master Mix (Applied Biosystems) on the 7900HT Fast Real-Time PCR system (Applied Biosystems). To detect SERCA2a gene expression, we used 5′-/56-FAM/ATCACACCG/ZEN/CTGAATGTGACCCAGT/3IABkFQ/-3′ (probe), 5′-TGTCACTGCACTTCCTCATCCTCT-3′ (primer 1), and 5′-ACTCCAGTATTGCAGGTTCCAGGT-3′ (primer 2). For signal recognition particle 14 kDa (SRP14) gene expression, acting as internal control, we used 5′-/56-FAM/TCAAAGCCCTCCACAGAACCTTTCC/3IABkFQ/-3′ (probe), 5′-CCCATCAGTAGCTCTTAACAGAC-3′ (primer 1), and 5′-AGACTCTTCCAGAAATGCCG-3′ (primer 2). All oligomers were obtained from Integrated DNA Technologies. The changes in gene expression of SERCA2a were derived from comparison of the average of the triplicate reactions of the progesterone-treated group with the average of the control group. Relative expression of SERCA2a was normalized to the level of SRP14 in the same cDNA.

Digital imaging of Ca²⁺ transients.

Ca²⁺ transient decay was determined using digital imaging. Briefly, on day 6 of culture, NRbCM were loaded with 2 μM fluo 3-AM indicator dye for 20 min in a temperature-controlled chamber (Dual Automatic Temperature Controller TC-344B, Warner Instruments) maintained at 35 ± 1°C. The cultured cells were stimulated using a platinum electrode (Myopacer EP field stimulator, IonOptix) with 10–20 V/cm electric field strength, 4-ms duration, and 1-Hz frequency. Fluorescence images were acquired on a Nikon Eclipse Ti-S microscope connected to a xenon arc lamp (Cairn Technology, Cairn, UK) through a μStep filter (IonOptix) and a high-speed electron-multiplying charge-coupled device camera (Evolve128, Photometrics) at 979 frames/s (2 × 2 binning with 2.4 × 2.4-μm² pixel resolution). Ca²⁺ transient duration was measured as the interval between activation time, which is determined from the maximum time point of the first derivative signal (dF/dt) and the point at which 75% amplitude is recovered (see Fig. 5A, black line). The Ca²⁺ decay time constant (τ) was measured by fitting an exponential decay function between the minimum time point of dF/dt (which represents the start of Ca²⁺ decay) and immediately before the next beat activation time point (which represents full recovery; see Fig. 5A, magenta line). All Ca²⁺ measurements are from multiple cells.

Fig. 5.

Progesterone shortens Ca²⁺ transient (CaT) duration and decay time constant (τ) in NRbCM. NRbCM were cultured on optical dishes, and τ and CaT duration were determined using digital imaging. A–C: CaT duration was measured as the interval between activation time, which is determined from the maximum time point of the 1st derivative signal (dF/dt) and the point at which 75% amplitude is recovered (horizontal black line in A) and plotted as a bar graph (B). τ was measured by fitting an exponential decay function between the minimum time point of dF/dt (which represents the start of Ca²⁺ decay) and immediately before the next beat activation time point (which represents full recovery; magenta line in A) and plotted as a bar graph (C). D: representative traces of Ca²⁺ transients from NRbCM culture with and without (CTR) progesterone. Values are means ± SE from 3 different isolations.

Statistical analysis.

Values are means ± SE of the number (n) of independent experiments. Data were analyzed by Student's t-test or one-way ANOVA followed by Tukey's test to determine statistical significance; P < 0.05 was deemed statistically significant.

RESULTS

Progesterone increases SERCA2a protein expression levels.

In a recent study (20) we showed significantly higher levels of SERCA2a protein in transgenic LQT2 rabbits treated with progesterone than in untreated rabbits. We further verified the effect of progesterone on SERCA2a using the NRbCM culture system, in which the concentration and treatment duration can be better regulated than in an in vivo system. To determine whether cultured NRbCM might serve as a model system in which to study the effects of progesterone, we first needed to determine whether progesterone treatment would have an effect on Ca²⁺-handling proteins in these cells similar to that seen in our in vivo transgenic rabbit model. Therefore, NRbCM were cultured and stimulated with 1 nM progesterone. We investigated the effects of progesterone treatment for 0, 24, 48, and 72 h on the levels of SERCA2a polypeptides in the protein lysates from cultured NRbCM. Our immunoblot data indicated an ∼100-kDa SERCA2a band (Fig. 1A), as we previously found for these cells (21). At 24 h poststimulation, SERCA2a protein levels were increased (∼1.2-fold by densitometric analysis) in response to progesterone. This increase reached its maximal level (1.5-fold) at 48 h; 72 h later, there was little difference between stimulated and nonstimulated NRbCM (Fig. 1).

Fig. 1.

Progesterone (PROG) increases sarco(endo)plasmic reticulum Ca²⁺-ATPase type 2a (SERCA2a) protein expression levels in neonatal rabbit cardiomyocytes (NRbCM). NRbCM were stimulated with progesterone on day 4 of culture for 24, 48, and 72 h. Cells were lysed and processed for immunoblotting and densitometric scanning of SERCA2a polypeptides. A: representative Western blot for SERCA2a expression in response to progesterone. Level of GAPDH served as a reference for protein loading. B: SERCA2a protein levels (fold change over control). Values are means ± SE from 3 different isolations. *P < 0.05. NS, not significant.

Progesterone does not affect SERCA2a mRNA transcript levels.

We next wanted to investigate the mechanism underlying progesterone-induced SERCA2a upregulation. A possible reason for increased SERCA2a expression is that progesterone binds to nuclear progesterone receptors, which, upon binding to possible DNA regulatory elements, may activate SERCA2a gene expression. Total RNA was isolated from the hearts of female LQT2 rabbits treated with progesterone (20) and from cultured NRbCM stimulated with progesterone and processed for quantitative RT-PCR to determine SERCA2a mRNA transcript levels. We found no change in SERCA2a mRNA levels upon progesterone treatment (Fig. 2), suggesting that the increase in SERCA2a protein expression may involve SERCA2a stabilization or a delay in SERCA2a protein degradation by progesterone.

Fig. 2.

Progesterone has no effect on SERCA2a mRNA transcript levels. A: total RNA was extracted from hearts of progesterone-treated female rabbits with long QT syndrome type 2 (LQT2) and processed for quantitative RT-PCR. Cycle threshold (2^ΔΔC_T) method was used to present data as fold change in SERCA2a transcript level in hearts of progesterone-treated animals normalized to the signal recognition particle 14 kDa (SRP14) gene (endogenous control) and relative to the SRP14-normalized SERCA2a transcript level in hearts of control rabbits (CTRL), which was arbitrarily set at 1.0. Values are means ± SE from 3 hearts for each condition. B: NRbCM were stimulated for 48 h with 0.1, 1, and 10 nM progesterone. Total RNA was extracted, and SERCA2a mRNA transcript was determined using quantitative RT-PCR. Data are shown as fold change in SRP14-normalized SERCA2a transcript level in progesterone-treated cells over the SERCA2a transcript level in vehicle-treated cells. Values are means ± SE from 3 different isolations.

Progesterone delays the degradation of SERCA2a protein.

Since progesterone has no effect on SERCA2a mRNA transcript levels, we speculated that it instead affects SERCA2a protein stability. To determine whether progesterone contributes to the protein degradation of SERCA2a, we conducted experiments with cycloheximide (CHX), which is known to block de novo protein synthesis (see materials and methods). Cultured NRbCM were stimulated with progesterone for 20 h, treated with 20 ng/ml CHX, and harvested after 0 and 24 h. Untreated samples and samples treated with progesterone (with and without CHX) were run on different gels (Fig. 3A). We compared the effect of progesterone treatment on SERCA2a protein levels by immunoblotting and densitometry after CHX-mediated inhibition of translation. At 24 h after CHX treatment, only 10% of SERCA2a was degraded in progesterone-treated NRbCM compared with 38% in vehicle-treated NRbCM (P < 0.05; Fig. 3B). This suggests that progesterone inhibited the degradation of SERCA2a polypeptides. GAPDH, an internal control, was measured by Western blotting and densitometry, indicating no significant difference in degradation kinetics between progesterone- and vehicle-treated groups at 0 and 24 h after CHX treatment.

Fig. 3.

Progesterone delays degradation of SERCA2a protein. A: representative Western blot for SERCA2a polypeptides, with GAPDH as an internal control. NRbCM were stimulated with progesterone for 20 h and then treated with cycloheximide (CHX, 20 ng/ml), and protein lysates were prepared 0 and 24 h after CHX treatment. B: immunoblots were analyzed densitometrically and plotted as a line graph representing total SERCA2a and GAPDH remaining (% of initial) 0 and 24 h after CHX treatment. Values are means ± SE from 3 different isolations. *P < 0.05.

Progesterone-mediated SERCA2a protein degradation involves ubiquitination.

If progesterone does not simulate de novo synthesis of SERCA2a polypeptides, it may regulate its degradation. Since we have shown that progesterone delays the degradation of SERCA2a protein, we sought to determine the degradation pathway that is affected by progesterone treatment. Ischemia-reperfusion injury is associated with calpain activation and SERCA2a degradation, and pharmacological inhibition of calpain activity is cardioprotective in ischemia-reperfusion injury (10). In cardiomyocytes, cellular proteins are degraded by two major pathways: autophagy and the ubiquitin-proteasome system. Autophagy is the process by which the degraded protein or organelles are sequestered in an autophagosome that fuses with the lysosome for complete degradation. The proteasome is the site for degradation of ubiquitin-tagged proteins (23). First, we set out to establish the pathway by which endogenous SERCA2a is degraded. Pharmacological inhibitors of autophagy, 3-MA, and the ubiquitin (MG132 and Lcy) and calpain (E-64D) pathways were added to NRbCM culture on day 4 and allowed to incubate for 48 h before harvest. Our Western blot data revealed no difference in SERCA2a expression levels between cells treated with vehicle, 3-MA, or E-64D (data not shown). However, SERCA2a protein levels increased in association with MG132 and Lcy treatment. We noticed a similar increase in SERCA2a levels in cells treated with progesterone, MG132, and Lcy. Compared with progesterone alone, cotreatment with Lcy or MG132 did not result in a significant further increase in SERCA2a levels in NRbCM (Fig. 4, A and B). These findings suggest that the progesterone-associated modulation of SERCA2a protein involves the ubiquitin pathway. It is highly likely that progesterone interferes with the ubiquitination of SERCA2a and, thereby, its degradation by the proteasome pathway.

Fig. 4.

Progesterone-mediated SERCA2a protein degradation involves ubiquitination. NRbCM were treated with clasto-lactacystin-β-lactone (Lcy, 25 μM) and MG132 (5 μM) for 48 h to block the proteasome complex, and SERCA2a levels were determined by immunoblotting and densitometric analysis. A: representative Western blot for SERCA2a expression levels in response to progesterone and blockers of the proteasome complex [Lcy (25 μM) and MG132 (5 μM)]. B: SERCA2a polypeptide levels expressed as fold change over control (GAPDH). Values are means ± SE from 3 different isolations. *P < 0.05. C: representative immunoblot for ubiquitin (top) and SERCA2a and GAPDH (bottom) from 3 independent experiments. Cells were stimulated with progesterone for 48 h, and cell lysates were immunoprecipitated with IgG (isotype-matching control antibody) and anti-SERCA2a. Immune complexes were then separated on 10% SDS gels and subjected to Western blot analysis with antibody directed against total ubiquitin. Lysates without immunoprecipitation (bottom) were probed with anti-SERCA2a and anti-GAPDH antibodies. IP, immunoprecipitation; WB, Western blotting; Ub-SERCA2a, ubiquitinylated SERCA2a polypeptide.

To corroborate this finding, we examined whether progesterone treatment had an effect on the association of SERCA2a with ubiquitin. We therefore performed immunoprecipitation to test this association and how it is affected by progesterone. We treated NRbCM with progesterone for 48 h and immunoprecipitated SERCA2a polypeptides using SERCA2a monoclonal antibody followed by Western blotting with ubiquitin antibody. We included IgG (isotype-matching control antibody) to demonstrate the specificity of our assay. We demonstrated that progesterone treatment significantly decreased the amount of ubiquitinylated SERCA2a polypeptides (Fig. 4C). These findings indicate that progesterone stabilizes SERCA2a, in part, by decreasing the ubiquitination level of SERCA2a polypeptides.

Effect of progesterone on SERCA2a pump function and regulator proteins.

Since progesterone treatment increases SERCA2a expression in cultured cells, we next investigated its effect on Ca²⁺ dynamics. Digital imaging was performed and Ca²⁺ transients were measured in fluo 3-AM dye-loaded cultured cells to determine the Ca²⁺ decay time constant (τ) and the duration of Ca²⁺ transients. We observed a significant shortening of decay of Ca²⁺ transients in response to progesterone treatment: 139 ± 24 ms (vehicle) vs. 97 ± 10 ms (progesterone) (Fig. 5, A and B), which is consistent with increased SERCA2a expression caused by progesterone. We also observed a significant decrease in Ca²⁺ transient duration (Fig. 5, C and D). Since NCX may also play a critical role in regulating Ca²⁺ transients in NRbCM, we compared the expression of NCX with and without progesterone treatment. Since we noted no significant changes in NCX protein levels in response to progesterone (Fig. 6D), it is highly likely that progesterone increases SERCA activity, thereby promoting faster clearance of Ca²⁺ from the cytosol.

Fig. 6.

Progesterone exerts no effect on Na⁺/Ca²⁺ exchanger (NCX) pump or phosphorylation status of phospholamban (PLB). NRbCM were treated with progesterone for 48 h, and the sarcoplasmic reticulum (SR)-enriched membrane fraction was extracted using differential centrifugation and probed for total PLB and phosphorylated (Ser¹⁶ and Thr¹⁷) PLB. A: representative immunoblot for total and phosphorylated (Ser¹⁶ and Thr¹⁷) PLB from 3 different experiments. After transfer of proteins, PVDF membranes were stained with 1% Ponceau S (Pon-S) to confirm equal loading. P, pentameric form of PLB; M, monomeric form of PLB. B and C: phosphorylated (Ser¹⁶ and Thr¹⁷) PLB levels normalized to total PLB levels. Values are means ± SE from 3 different isolations. cont, Control. D: representative Western blot for NCX and relative band intensities determined by densitometry analysis using National Institutes of Health ImageJ software and normalized to GAPDH. Cells were lysed 48 h after progesterone treatment, and total cell lysate was used to determine NCX protein levels. Values are means ± SE from 3 different isolations.

SERCA2a activity is tightly regulated by phosphorylation of PLB at Ser¹⁶ (mediated by PKA) and Thr¹⁷ (mediated by Ca²⁺-calmodulin-dependent protein kinase) (3, 30). When unphosphorylated, PLB inhibits SERCA2a activity; however, phosphorylated PLB releases the inhibition on SERCA2a, increasing its activity (13). Thus we asked whether progesterone affected the phosphorylation status of PLB. Under our experimental conditions, progesterone treatment resulted in no apparent change in the phosphorylation status of PLB or total PLB levels (sum of pentamers and monomers) (Fig. 6).

DISCUSSION

Ca²⁺ handling is impaired in ovariectomized rat heart, as reflected by a decrease in SERCA2a mRNA transcript and SERCA2a protein and activity levels (4). Furthermore, supplementation with estrogen and progesterone restored SERCA2a mRNA levels, protein expression levels, and pump function (4). Recently, our laboratory showed that SERCA2a protein levels were moderately upregulated in progesterone-treated ovariectomized transgenic female LQT2 rabbits (20). Here we attempted to decipher the molecular mechanism of SERCA2a regulation by progesterone at the cellular level. We also observed a moderate 1.5-fold increase in SERCA2a protein levels in NRbCM, which is consistent with our findings and those of other researchers (4). This increase in SERCA2a protein levels was not associated with changes in mRNA levels, suggesting that the effect of progesterone is posttranslational.

Our finding that progesterone acts on SERCA2a via a nongenomic pathway in rabbit cardiomyocytes is supported by the fact that the human SERCA2 gene lacks progesterone-responsive elements (33). However, in contrast with our results, others showed that increased SERCA2a in progesterone-treated ovariectomized rat hearts is associated with mRNA transcript levels (4). This discrepancy can be explained by differences among species in 3′-untranslated regions of the RNA that may affect mRNA stability or by differences in the effects of progesterone-responsive elements in the SERCA2 gene.

Our study indicates that the upregulation of SERCA2a in neonatal cardiomyocytes is the result of its delayed degradation, as confirmed in CHX experiments. We next decided to explore the role of other degradation pathways responsible for stabilizing SERCA2a levels in response to progesterone treatment. A study by Kho et al. (14) pointed to the involvement of the ubiquitin-proteasome pathway in SERCA2a degradation and concluded that SUMOylation stabilizes SERCA2a by inhibiting its ubiquitin-mediated proteasomal degradation. Our Western blot and immunoprecipitation data showed that progesterone stabilizes SERCA2a, in part, by decreasing the level of ubiquitin-associated SERCA2a polypeptides independent of SUMOylation. Our results on the ubiquitin-dependent regulation of SERCA2a protein levels are consistent with the results of Kho et al. (14).

It is known that Ca²⁺ transient decay in cardiomyocytes is species-dependent (2) and relies only on NCX and SERCA2a activity. Our results indicate that progesterone exerts no change in NCX protein levels and, importantly, that the shortening of the Ca²⁺ transient decay in NRbCM is most likely due to enhanced SERCA2a activity. It is interesting to note that, despite the observations of others (11), our progesterone data suggest that changes in SERCA2a activity in NRbCM can indeed lead to changes in Ca²⁺ kinetics.

The critical phosphoproteins regulating Ca²⁺ handling and contraction in the heart are ryanodine receptor 2, PLB, and troponin I. PLB controls the Ca²⁺ affinity of the SERCA pump (9). Dephosphorylated PLB is inhibitory, and phosphorylation of PLB restores attenuated SERCA2a pump functions, thereby facilitating Ca²⁺ uptake in the SR. Our understanding of the regulatory role of PLB arises basically from genetically modified rodent models that exhibit fundamental differences in myocardial Ca²⁺ handling compared with humans. Other workers observed no change in total PLB or phosphorylated (Ser¹⁶) PLB protein in ovariectomized rat hearts, which was associated with a significant downregulation of phosphorylated (Thr¹⁷) PLB (4). Hormone supplementation restored the phosphorylation of PLB at Thr¹⁷. We therefore examined whether progesterone also regulates SERCA2a activity through PLB, since regulation often takes place at different levels. We observed no change in PLB protein levels or phosphorylation status, consistent with our previous in vivo findings (20). Thus the progesterone-mediated increase in SERCA2a activity takes place through increased SERCA2a expression and stability and is independent of PLB.

In summary, our findings in cultured NRbCM suggest that progesterone treatment increases SERCA2a protein levels and stabilizes SERCA2a, in part, by decreasing the level of ubiquitin-associated SERCA2a polypeptides and, thereby, its degradation by the proteosome pathway. Moreover, progesterone treatment speeds clearance of Ca²⁺ from the cytosol, very likely by increasing SERCA2a activity, without exerting any effect on the major SERCA2a regulatory protein PLB.

GRANTS

This study was supported in part by National Heart, Lung, and Blood Institute Grants 5R01 HL-093205-04 and 1R01 HL-110791-01A1 (to G. Koren), 5R01 HL-096669 (to B.-R. Choi), and 5T32 HL-094300 (to K. S. Moshal and Z. Zhang).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.S.M., Z.Z., K.R., and G.K. are responsible for conception and design of the research; K.S.M., Z.Z., T.Y.K., L.C., B.P.L., Y.L., and V.R. performed the experiments; K.S.M., Z.Z., T.Y.K., L.C., D.T., and B.-R.C. analyzed the data; K.S.M., K.R., B.P.L., D.T., B.-R.C., and G.K. interpreted the results of the experiments; K.S.M., Z.Z., T.Y.K., L.C., B.P.L., and V.R. prepared the figures; K.S.M. drafted the manuscript; K.S.M., K.R., B.-R.C., and G.K. edited and revised the manuscript; K.R., B.-R.C., and G.K. approved the final version of the manuscript.