Caveolin-3 Overexpression Attenuates Cardiac Hypertrophy via Inhibition of T-type Ca2+ Current Modulated by Protein Kinase Cα in Cardiomyocytes*

Background: Ventricular remodeling altered caveolin-3 expression, and Ca2+ signaling is associated with cardiac hypertrophy. Results: Cardiomyocyte-specific caveolin-3 overexpression prevented cardiac hypertrophy by inhibiting the T-type Ca2+ current and hyperactivation of calcineurin-dependent nuclear factor of activated T-cell signaling. Conclusion: Caveolin-3 expression is essential for protective Ca2+ signaling in pathological cardiac hypertrophy. Significance: Caveolin-3 overexpression in heart may be used as a therapeutic strategy for treatment of many cardiovascular diseases. Pathological cardiac hypertrophy is characterized by subcellular remodeling of the ventricular myocyte with a reduction in the scaffolding protein caveolin-3 (Cav-3), altered Ca2+ cycling, increased protein kinase C expression, and hyperactivation of calcineurin/nuclear factor of activated T cell (NFAT) signaling. However, the precise role of Cav-3 in the regulation of local Ca2+ signaling in pathological cardiac hypertrophy is unclear. We used cardiac-specific Cav-3-overexpressing mice and in vivo and in vitro cardiac hypertrophy models to determine the essential requirement for Cav-3 expression in protection against pharmacologically and pressure overload-induced cardiac hypertrophy. Transverse aortic constriction and angiotensin-II (Ang-II) infusion in wild type (WT) mice resulted in cardiac hypertrophy characterized by significant reduction in fractional shortening, ejection fraction, and a reduced expression of Cav-3. In addition, association of PKCα and angiotensin-II receptor, type 1, with Cav-3 was disrupted in the hypertrophic ventricular myocytes. Whole cell patch clamp analysis demonstrated increased expression of T-type Ca2+ current (ICa, T) in hypertrophic ventricular myocytes. In contrast, the Cav-3-overexpressing mice demonstrated protection from transverse aortic constriction or Ang-II-induced pathological hypertrophy with inhibition of ICa, T and intact Cav-3-associated macromolecular signaling complexes. siRNA-mediated knockdown of Cav-3 in the neonatal cardiomyocytes resulted in enhanced Ang-II stimulation of ICa, T mediated by PKCα, which caused nuclear translocation of NFAT. Overexpression of Cav-3 in neonatal myocytes prevented a PKCα-mediated increase in ICa, T and nuclear translocation of NFAT. In conclusion, we show that stable Cav-3 expression is essential for protecting the signaling mechanisms in pharmacologically and pressure overload-induced cardiac hypertrophy.


Pathological cardiac hypertrophy is characterized by subcellular remodeling of the ventricular myocyte with a reduction in the scaffolding protein caveolin-3 (Cav-3), altered Ca 2؉ cycling, increased protein kinase C expression, and hyperactivation of calcineurin/nuclear factor of activated T cell (NFAT) signaling.
However, the precise role of Cav-3 in the regulation of local Ca 2؉ signaling in pathological cardiac hypertrophy is unclear. We used cardiac-specific Cav-3-overexpressing mice and in vivo and in vitro cardiac hypertrophy models to determine the essential requirement for Cav-3 expression in protection against pharmacologically and pressure overload-induced cardiac hypertrophy. Transverse aortic constriction and angiotensin-II (Ang-II) infusion in wild type (WT) mice resulted in cardiac hypertrophy characterized by significant reduction in fractional shortening, ejection fraction, and a reduced expression of Cav-3.

In addition, association of PKC␣ and angiotensin-II receptor, type 1, with Cav-3 was disrupted in the hypertrophic ventricular myocytes. Whole cell patch clamp analysis demonstrated increased expression of T-type Ca 2؉ current (I Ca, T ) in hypertrophic ventricular myocytes. In contrast, the Cav-3-overexpressing mice demonstrated protection from transverse aortic constriction or Ang-II-induced pathological hypertrophy with inhibition of I Ca, T and intact Cav-3-associated macromolecular signaling complexes. siRNA-mediated knockdown of Cav-3 in the neonatal cardiomyocytes resulted in enhanced Ang-II stim-ulation of I Ca, T mediated by PKC␣, which caused nuclear translocation of NFAT. Overexpression of Cav-3 in neonatal myocytes prevented a PKC␣-mediated increase in I Ca, T and nuclear translocation of NFAT. In conclusion, we show that stable Cav-3 expression is essential for protecting the signaling mechanisms in pharmacologically and pressure overload-induced cardiac hypertrophy.
Cardiac hypertrophy is a major predictor of many cardiovascular diseases, including arrhythmias, sudden death, and heart failure. Cardiac hypertrophy is an adaptive response of the heart during stress to preserve contractility and cardiac function. However, continued cardiac stress through either pressure or volume overload or neurohormonal stress leads to pathological hypertrophy and heart failure (1,2), during which an alteration in cardiac myocyte Ca 2ϩ handling is commonly observed (3)(4)(5). It is well established that an increase in cytosolic Ca 2ϩ is responsible for activating calcineurin (Cn) 2 and nuclear factor of activated T-cell (NFAT) signaling leading to the expression of genes involved in pathological cardiac hypertrophy. With the progression of cardiac hypertrophy, a structural remodeling of the ventricular myocytes results in T-tubule disruption at advanced heart failure (6). With myocyte remodeling during cardiac hypertrophy and heart failure (7,8), it is likely that the micro-architecture of the sarcolemma and T-tubules, which are major determinants of the local control of Ca 2ϩ in the heart, is altered. * This work was supported, in whole or in part, by National Institutes of Health Caveolae are specialized microdomains in the sarcolemmal membrane of ventricular myocytes that serve to integrate sympathetic and parasympathetic inputs to the heart to precisely regulate cardiac function. Caveolae contain a variety of signaling proteins such as G-protein-coupled receptors, kinases, phosphatases, and ion channels, including the voltage-gated L-type and the T-type Ca 2ϩ channels and other calcium cycling proteins (9 -11). Caveolin-3 (Cav-3) is a muscle-specific scaffolding protein integral to caveolae in the cardiomyocyte and plays a significant role in the physiology of the heart (12). Reduced expression of Cav-3 and caveolae in cardiomyocytes is reported in cardiac diseases, including myocardial infarction and heart failure (13). In contrast, we have shown that overexpression of Cav-3 prevents ischemic injury (14) and cardiac hypertrophy (15). However, the precise role of Cav-3 in the regulation of local Ca 2ϩ signaling and regulation of pathophysiology in cardiac hypertrophy is unclear.
In this study, we determined whether a loss of Cav-3 expression during pressure overload and angiotensin-II (Ang-II) treatment contributes to altered Ca 2ϩ -induced Cn-NFAT signaling and the development of pathological cardiac hypertrophy. We demonstrated that a loss of Cav-3 expression after pressure overload or Ang-II treatment results in the disruption of caveola-associated macromolecular signaling proteins, increased stimulation of T-type Ca 2ϩ channels (TTCC) current (I Ca, T ) mediated by PKC␣, and the activation of Cn-NFAT signaling in cardiomyocytes. Additionally, Cav-3 overexpression in cardiomyocytes inhibits basal and Ang-II-stimulated I Ca, T that is modulated by PKC␣ and the activation of Cn-NFAT signaling. Using mice with cardiac-specific overexpression Cav-3 (Cav-3 OE) (14), generated using the ␣-myosin heavy chain promoter system, we demonstrated that development of pressure overload-induced pathological cardiac hypertrophy in vivo is prevented.

Materials and Methods
Transverse Aortic Constriction (TAC) induced Pressure Overload Hypertrophy-TAC was performed in 12-16-week-old male mice to induce pressure overload as described earlier (16). Briefly, the mice were anesthetized with 2% isofluorane inhalation, and insertion was made to expose the aorta. A 27-gauge needle was placed on top of the aorta and ligated using 7-0 silk sutures, following which the needle was removed to produce refined stenosis of the vessel. The muscle cavity and skin were sutured, and the wound was closed with wound clip. Mice of the same genetic background received a sham operation in which a silk suture band was placed around the aorta but not ligated and was subsequently removed.
Ang-II Infusion Induced Cardiac Hypertrophy-Ang-II or saline was infused for 28 days using mini osmotic pumps (model 2002, ALZET Osmotic Pumps, Cupertino, CA). Osmotic pumps primed at constant rate of 0.5 g/h, filled with 5 mg/ml Ang-II (Sigma) or isotonic saline, were inserted subcutaneously above the scapula under sterile conditions in anesthetized mice. For in vitro Ang-II-induced cardiac hypertrophy, NMVM were isolated from 1-to 2-day-old pups and grown in culture treated with Ang-II (10mol/liter) for 48 h.
Echocardiography Analysis-Noninvasive transthoracic echocardiography was performed using Visual Sonics Vevo 770 ultrasonograph. ECG was monitored continuously in anesthetized mice (1.5% isoflurane) maintained on a heated platform. After 4 weeks of saline or Ang-II infusion and sham or TAC surgery in mice, left ventricular wall thickness, chamber dimensions, and contractility were evaluated. The pressure gradients across the aortic constriction were measured to ensure similar pressure overload in the TAC mice.
Transmission Electron Microscopy-Rapidly excised mouse hearts were initially perfused with Tyrode's solution (10 ml) in a Langendorff perfusion system followed by fixative (2.5% glutaraldehyde, 2.0% paraformaldehyde) in 0.1 mol/liter cacodylate buffer for 30 min. The left ventricle was dissected out, cut into 2 ϫ 2-mm blocks, immersed in the same fixative, and left overnight at 4°C. The samples were rinsed in the same buffer, postfixed in 1% osmium tetroxide, dehydrated in a graded ethanol series, rinsed in propylene oxide, and embedded in Epon 812 substitute. After resin polymerization, the samples were then sliced into 70-nm sections with a Leica EM UC6 ultramicrotome and placed on 200 mesh transmission electron microscopy grids. The samples were post-stained in 8% uranyl acetate in 50% EtOH and Reynold's lead citrate, viewed on a Philips CM120 transmission electron microscope, and documented with a SIS MegaView III digital camera. A relative number of caveolae distributed in the myocyte sarcolemmal membranes was estimated by obtaining about 250 images from three preparations of WT or TAC samples. A threshold size for individual caveolae was set between 40 and 100 nm. The number of caveolae was counted as per unit length (m) of myocyte sarcolemmal membranes using ImageJ software from a series of random EM micrographs. To confirm caveola vesicles from other regions, immunogold labeling using anti-Cav-3 antibody was performed. Data were analyzed by plotting frequency histograms of the number of caveolae per m of sarcolemma for each observation.
Isolation of Mouse Ventricular Myocytes-Neonatal or adult mouse ventricular myocytes were enzymatically isolated as described previously (11). Rod-shaped myocytes with clear striations were randomly selected for electrophysiology studies. The neonatal myocytes were transfected by the electroporation method (11) by a Nucleofector device (Lonza, USA) using Ingenio electroporation reagent (catalog no. MIR 50115) from Mirus BioSciences, and cells were used for experiments 72-96 h after transfection.
Quantitative Real Time PCR Analysis-MIQE guidelines were followed in designing qPCR experiments. Total RNA isolated from SHAM, TAC, saline, and Ang-II treated mouse left ventricles using the GenElute Mammalian Total RNA Miniprep kit (Sigma). RNA quantity and quality were determined with UV spectrophotometry. First strand cDNA synthesis was performed with 1 g of total RNA using iScript reverse transcription supermix for RT-qPCR (Bio-Rad). The levels of cDNA were analyzed by quantitative real time PCR using TaqMan gene expression master mix (Applied Biosystems). Probes and primers were designed for multiplex analysis (Integrated DNA Technologies). Primers and probes designed for analysis of the genes of interest are provided in Table 1. RT-qPCR was performed on CFX96 TM real time systems (Bio Rad). For quantification of mRNA levels, the normalized cycle values were obtained by the subtraction of corresponding GAPDH (⌬C T ), and data are presented as fold change (for TAC or Ang-II treatment) with respect to expression in SHAM or saline-treated samples (⌬⌬C T ).
Preparation of Caveolin-enriched Fractions-Caveolin-enriched membrane fractions from mouse ventricular myocytes from WT or Cav-3 OE following TAC or sham treatment were prepared by using a previously described method (18). Briefly, freshly isolated adult mouse myocytes (10 ϫ 10 6 cells) were suspended in 2 ml of ice-cold 0.5 mol/liter sodium carbonate (pH 11.0) and homogenized sequentially by using a loose-fitting Dounce homogenizer (10 strokes), a Polytron tissue grinder (three 10-s bursts; Kinematica, Brinkmann Instruments, Westbury, NY), and a sonicator (three 20-s bursts; Branson Sonifier 250, Branson Ultrasonic, Danbury, CT). The homogenate was adjusted to 45% sucrose in MBS (25 mmol/liter Mes (pH 6.5), 0.15 mol/liter NaCl) and placed at the bottom of an ultracentrifuge tube. A 5-35% discontinuous sucrose gradient (in MBS containing 250 mmol/liter sodium carbonate) was formed and centrifuged at 39,000 rpm for 16 -20 h in an SW41 rotor (Beckman Instruments, Palo Alto, CA). From the bottom of each gradient, 1-ml gradient fractions were collected to yield a total of 12 fractions. Protein concentrations determined by the Lowry assay (Bio-Rad) confirmed that total protein distribution was weighted toward heavier sucrose density gradient fractions (F7 through F11). A lightscattering band confined to fractions 4 -6 typically corresponds to caveola-enriched fractions. Proteins from different fractions were precipitated using 0.1% w/v deoxycholic acid in 100% w/v trichloroacetic acid and then each fraction sample was solubilized into an equal volume (40 l) of sample buffer for SDS-PAGE and Western blot analysis.
Co-immunoprecipitation and Western Blot Analysis-Isolated adult mouse myocytes (ϳ2 mg of protein) were rinsed with ice-cold 25 mmol/liter Tris-HCl (pH 7.4), 150 mmol/liter NaCl (TBS), and lysed in ice-cold solubilization buffer containing 25 mmol/liter Tris-HCl (pH 7.4), 150 mmol/liter NaCl, 60 mmol/liter n-octyl D-glucoside, 1% Triton X-100, 2 mmol/liter phenylmethylsulfonyl fluoride, 5 g/ml aprotinin, 5 g/ml benzamidine, 5 g/ml leupeptin, and 5 mol/liter pepstatin A. The lysate was centrifuged at 10,000 ϫ g for 10 min to remove insol- uble debris, and the soluble supernatant was precleared by using protein G Dynabeads (Invitrogen), followed by incubation for 4 h at 4°C with anti-Cav-3 (2 g) antibodies or control IgG in a total of 450 l. 50 l of a 1:1 slurry of protein G Dynabeads was added to the sample and further incubated for 1 h at 4°C. Beads were washed four times with solubilization buffer on a magnetic stand, and bound proteins were eluted with SDS-PAGE sample buffer by boiling for 5 min. Immune complexes were analyzed by SDS-PAGE (4 -15% gradient gels, Bio-Rad) and Western blot by probing with antibodies to Cav-3, PKC␣, angiotensin receptor type 1, NOS-3, and ␤ 1 AR. Electrophysiology-Electrophysiological experiments were carried out using the whole cell patch clamp technique using Axopatch 200B amplifier (Axon Instruments, Foster City, CA) with pClamp version 10.2 software. The patch pipettes were pulled from thin-walled borosilicate glass capillaries (World Precision Instruments, Inc., Sarasota, FL) on Sutter P-87 micropipette puller (Sutter Instrument Co.) and polished using microforge MF900 (Narishige). All the experiments were carried out at room temperature with pipette resistance of 1.5-2.5 megohms. Recordings were made from the freshly isolated healthy rod-shaped ventricular myocytes. The bath solution to measure T-type Ca 2ϩ channel currents from adult cardiomyocytes consisted of (in mmol/liter) 140 TEA-Cl, 1 MgCl 2 , 1.8 CaCl 2 , 10 glucose, and 10 HEPES (pH 7.4), tetradotoxin 20 mol/liter. For neonatal cardiomyocytes, bath buffer consisted of (in mmol/liter) 145 TEA-Cl, 5 CaCl 2 , 1 MgCl 2 , 5 CsCl, 1,4aminopyridine, 0.01 tetradotoxin, 10 HEPES, 5 D-glucose (pH 7.4), adjusted with TEA-OH. The internal pipette solution consisted of (in mmol/liter) 114 CsCl, 10 EGTA, 10 HEPES, 5 MgATP (pH 7.2) adjusted using CsOH. T-and L-type calcium currents were measured using a dual pulse protocol as described earlier by us (11). Myocytes were held at a holding potential of Ϫ90 mV, and a 10-mV step depolarization was applied up to ϩ60 mV for 200 ms (I Ca, total ), followed by a brief holding potential of Ϫ50 mV, and a further 10-mV step depolarization was applied up to ϩ70 mV for 200 ms (I Ca, L ). First pulse represents the total current (I Ca, total ), and the second pulse represents the L-type current (I Ca, L ). T-type calcium current (I Ca, T ) was obtained by subtracting I Ca, L traces from I Ca, total and indicated as I Ca, difference (I Ca, diff ). When this I Ca, T is absent (absence of peak I Ca at Ϫ30 mV) in the cells, the two I-V curves generated from holding potentials of Ϫ90 and Ϫ50 mV will overlap but may also exhibit a I Ca, diff at, or membrane potentials positive to Ϫ10 mV. A small difference in the currents at potentials positive to Ϫ10 mV was not due to the presence of I Ca, T but could be due to partial voltage-dependent inactivation of I Ca, L recorded with a holding potential of Ϫ50 mV. The current traces were corrected for linear capacitance and leak using ϪP/4 subtraction. The data were filtered at 5 kHz and digitized at 50 kHz and were analyzed using Microcal Origin software (Origin Lab Corp., Northampton, MA). The data were analyzed using OriginPro.9.0.0 (OriginLab Corp.).
Statistics-Statistical significance was analyzed with Student's paired t test. Average data are reported as mean Ϯ S.E.

Cav-3 and Caveola Expression Is Altered in Ventricular Myocytes in TAC or Ang-II Infusion-induced Cardiac
Hypertrophy-Previous studies have indicated that the level of Cav-3 expression and the density of caveolae can change in various models of cardiac disease, including cardiac hypertrophy and heart failure (13,20). We investigated for changes in the expression of Cav-3 in mouse models of pressure overloadinduced cardiac hypertrophy. 12-16-Week-old C57BL6 mice were chronically treated with Ang-II via continuous infusion (via mini osmotic pumps; see under "Materials and Methods") or TAC surgery. Cardiac function was measured by echocardiography before TAC, sham, or Ang-II or saline treatment and then after 4 weeks of treatment. Four weeks of TAC or Ang-II treatment resulted in the development of pathological cardiac hypertrophy as evidenced by significant changes in HW/BW, and reduced fractional shortening and ejection fraction in the WT mice compared with sham or saline-treated mice ( Fig. 1, A-C). The RNA isolated from left ventricular myocytes showed a significant increase in atrial natriuretic peptide, B-type natriuretic peptide expression, and a significant reduction in the expression for Cav-3 and SERCA2a levels in TAC-and Ang-IItreated mice compared with sham or saline-treated mice (Fig. 2, A and B). The mRNA levels for Cav-1 were unchanged between the groups. We then estimated the expression of Cav-3 protein by semi-quantitative Western blot analysis in mouse ventricular myocytes. Cav-3 expression levels (Fig. 1, D and E) were significantly reduced (ϳ50%) after TAC-or Ang-II-induced cardiac hypertrophy when compared with control hearts (sham or saline infusion). We then performed transmission electron microscopy analysis on the left ventricle tissue sections after 4 weeks of TAC or sham treatment. As shown in the representative electron micrograph (Fig. 1F), after 4 weeks of TAC, the number of caveolae was reduced significantly (64%) in the left ventricular myocytes compared with the sham mice (Fig. 1G).
I Ca, T Is Up-regulated in Left Ventricular Myocytes in Cardiac Hypertrophy-Previous studies have shown that I Ca, T is expressed only during cardiac development and is undetectable in adult ventricular myocytes (21,22). However, I Ca, T was shown to be re-expressed in ventricular myocytes in diseased hearts, including pressure overload-induced cardiac hypertrophy (11,23,24) in cardiomyopathic hamster (25), and in postinfarction remodeled rat left ventricle (26). We measured the expression levels for the TTCC subunit isoforms, Ca v 3.1 and Ca v 3.2, in the ventricles by qPCR analysis. We noticed an increased mRNA expression for Ca v 3.1 and Ca v 3.2 subunits in the left ventricles from TAC-or Ang-II-treated mice compared with vehicle-or sham-treated mice, respectively (Fig. 2, A and  B). The mRNA levels for Ca v 3.2 appeared to be significantly higher (p Ͻ 0.05) in the TAC ventricle compared with sham ( Fig. 2A). In contrast, the mRNA levels for the Ca v 1.2 subunit of the LTCC did not change after TAC or Ang-II treatment compared with controls. We then investigated whether the I Ca, T was detectable in the adult left ventricular myocytes after TAC or Ang-II infusion. I Ca, T and L-type Ca 2ϩ channel current (I Ca, L ) were measured using the whole cell patch clamp technique by applying a dual pulse protocol described previously by

Caveolin-3 Overexpression Attenuates Cardiac Hypertrophy
us (11). As shown in Fig. 3, a re-expression of I Ca, T (Ϫ1.6 Ϯ 0.4 pA/pF) in ventricular myocytes after 4 weeks of TAC compared with negligible current (Ϫ0.02 Ϯ 0.05 pA/pF) in ventricular myocytes from sham-treated mice (Fig. 3, B and C) was observed. In the sham myocytes (Fig. 3B), there was no detectable I Ca, T at Ϫ30 mV. Similarly, Ang-II infusion resulted in re-expression of I Ca, T in ventricular myocytes (Ϫ0.84 Ϯ 0.11 pA/pF) compared with saline-treated animals (Ϫ0.19 Ϯ 0.34 pA/pF) (Fig. 3C). To confirm the expression of I Ca, T in the WT hypertrophied (TAC) cardiomyocytes, we first measured I Ca, T and then perfused cells with an I Ca, T inhibitor Ni 2ϩ (300 M), which completely abolished the I Ca, T (Fig. 3E) but did not significantly impact the I Ca, L (data not shown). The inhibition of I Ca, T by Ni 2ϩ confirmed that the Ca 2ϩ current elicited at Ϫ30 mV is indeed I Ca, T . These data confirm that I Ca, T is re-expressed in the ventricular myocytes during TAC-or Ang-IIinduced cardiac hypertrophy. I Ca, L was not significantly different in the ventricular myocytes after TAC or Ang-II treatment compared with controls (Fig. 3D).
Cardiac-specific Cav-3 Overexpression Attenuates Cardiac Hypertrophy-Recently, we have demonstrated that the cardiacspecific overexpression of Cav-3 resulted in attenuation of

. Reduced Cav-3 and caveola expression levels in ventricular myocytes in TAC and Ang-II infusion-induced cardiac hypertrophy.
A, average heart weight to body weight (HW/BW) ratio was significantly increased in control WT mice after TAC or Ang-II infusion compared with sham or saline infusion, respectively. B and C, echocardiography in WT mice reveals a significant decrease in percentage of fractional shortening and ejection fraction in TAC or Ang-II infusion compared with sham or saline infusion respectively. #, p Ͻ 0.05; *, p Ͻ 0.005, n ϭ 8. D, representative Western blot analysis shows reduced Cav-3 protein expression in ventricular myocyte lysates from WT mice after 4 weeks of TAC or Ang-II infusion compared with sham or saline infusion, respectively. E, mean Cav-3 expression levels normalized to ␤-actin levels in TAC-or Ang-II-treated WT mice compared with sham-or saline-treated mice. *, p Ͻ 0.005, n ϭ 6.  SEPTEMBER 4, 2015 • VOLUME 290 • NUMBER 36

JOURNAL OF BIOLOGICAL CHEMISTRY 22089
TAC-induced cardiac hypertrophy via enhanced natriuretic peptide expression (15). Here, we investigated whether Cav-3 OE mice have attenuation of cardiac hypertrophy after Ang-II infusion. Male 12-16-week-old Cav-3 OE and WT mice were subjected to Ang-II or saline infusion or TAC or sham surgery for 4 weeks. As shown in Table 2, echocardiography revealed that WT mice had decreased ejection fraction and percentage fractional shortening after 4 weeks of TAC or continuous Ang-II infusion (Table 2), whereas Cav-3 OE mice subjected to TAC had no change in either measure of cardiac function compared with sham-treated mice. WT mice showed an increase in cardiac hypertrophy in response to TAC or Ang-II infusion with increased ventricular wall thickness and an increase in HW/BW ratio, but TAC or Ang-II infusion in Cav-3 OE mice did not show significant differences in these measures compared with sham or saline treatment, respectively ( Table 2). The above data confirm that Cav-3 OE mice are protected from Ang-II-induced cardiac hypertrophy. The data with TAC studies are in agreement with and confirm our previously published results (15).

Caveolin-3 Overexpression Inhibits I Ca, T in Cardiac
Hypertrophy-A recent study suggested that a re-expression of the Ca v 3.2 TTCC current is responsible for the induction of pathological cardiac hypertrophy via calcineurin/NFAT hypertrophic signaling (27). We have shown that Cav-3 overexpression inhibits Ca v 3.2 (␣ 1H ) channel current but not the Ca v 3.1 (␣ 1G ) current in mouse neonatal cardiomyocytes (11). Therefore, we hypothesized that ventricular myocytes from Cav-3 OE mice will inhibit re-expression of I Ca, T , specifically the I Cav3.2 , and attenuate pressure overload-induced pathological cardiac hypertrophy. We investigated the role of Cav-3 on I Ca, T inhibition in pathological hypertrophy using the Cav-3 OE or littermate WT control mice subjected to TAC or Ang-II infusion for 4 weeks. I Ca, T and I Ca, L were measured in adult ventricular myocytes (AVMs) from mice subjected to different treatment groups. Cell capacitance measured during voltage clamp measurement showed TAC or Ang-II infusion caused 27 and 34% increase, respectively, in the AVM size in the WT mice compared with saline-treated animals (Fig. 4E). Cell capacitance of the AVMs from Cav-3 OE mice was 50% greater than AVMs from WT saline-treated mice. TAC or Ang-II infusion did not significantly alter the cell capacitance in AVMs from Cav-3 OE mice (Fig. 4F). Peak I Ca, T , measured at Ϫ30 mV normalized to cell capacitance and expressed as pA/pF (Fig. 3C), was significantly increased in the AVMs from WT mice after either TAC or Ang-II infusion. I Ca, T expression was negligible in salineand sham-treated WT AVMs (Fig. 3C). As shown in Fig. 4, the peak I Ca, T was completely inhibited in the AVMs from Cav-3 OE mice after TAC (Fig. 4, A and B) or Ang-II infusion (Fig. 4C), suggesting that cardiac myocyte-specific overexpression of Cav-3 inhibits the TAC-or Ang-II-induced increase in I Ca, T during pathological hypertrophy. The peak I Ca, L density elicited at 0 mV was not different in the AVMs from mice with all treatment groups (Fig. 4D). In addition, the activation and inactivation of I Ca, L were not different in the AVMs from mice with all treatment groups (data not shown). The I Ca, L data also confirmed our earlier demonstration that Cav-3 overexpression does not alter peak I Ca, L density in neonatal mouse ventricular myocytes (11).
Changes in Expression of Key Signaling Proteins in Ventricular Myocytes during Cardiac Hypertrophy-To examine whether the increase in I Ca, T expression in hypertrophic myocytes is associated with changes to the protein level for TTCC isoforms Ca v 3.1 and Ca v 3.2, and other key signaling proteins involved in cardiac hypertrophy, we performed semi-quantitative Western blot analysis on ventricular lysates prepared from WT and Cav-3 OE mice following TAC or sham treatments. As shown in Fig. 5, the expression levels for the Ca v 3.2 and PKC␣ proteins were significantly increased in WT TAC myocytes compared with sham mice. However, the expression levels of Ca v 3.2 and PKC␣ were normalized in the Cav-3 OE hearts following TAC and were not different compared with WT and Cav-3 OE sham hearts. The expression level of Ca v 3.1 and PKC␤1 was unchanged in all groups. Also, the Cav-3 overexpression in the hearts (Cav-3 OE mice) did not impact the expression profiles of any of the above proteins. However, the

. I Ca, T is increased in the left ventricular myocytes from TAC-or Ang-II-infused hypertrophic mice.
A, representative calcium current traces were measured using whole cell patch clamp technique in left ventricular myocytes from TAC or sham mice using a dual pulse voltage protocol (inset). I Ca, T is referred to as the difference between current recorded between step depolarization at holding potential Ϫ90 and Ϫ50 mV. B, mean current to voltage response of L-type (E) and T-type (OE) current recorded from ventricular myocytes after 4 weeks of TAC or sham in WT mice. C, mean peak current densities of I Ca, T at Ϫ30 mV were significantly increased in ventricular myocytes from WT mice after 4 weeks of TAC or Ang-II infusion compared with sham or saline infusion, respectively. D, mean peak I Ca, L density at 0 mV was unchanged in the ventricular myocytes from WT mice after TAC or sham treatment and Ang-II infusion or saline treatment. p Ͻ 0.001, n ϭ 9 cells from 5 animals in each group. E, representative I Ca, T traces from WT TAC myocytes perfused with 300 M Ni 2ϩ (left) and mean peak I Ca, T at Ϫ30 mV (right). n ϭ 4 from three animals. Data represent means Ϯ S.E.   SEPTEMBER 4, 2015 • VOLUME 290 • NUMBER 36 expression levels for AT-R were significantly reduced in WT and Cav-3 OE hearts following TAC in comparison with shamtreated hearts. These above data indicate that increased expression of the Ca v 3.2 and PKC␣ in cardiac hypertrophy could contribute to an increase in I Ca, T and altered Ca 2ϩ signaling.

Cav-3 Overexpression Prevents Disruption of Caveola-associated Macromolecular
Signaling Complex-We investigated the impact of reduced Cav-3 and caveola expression on caveolalocalized signaling proteins in pathological hypertrophy. Myocyte lysates from mice after 4 weeks of TAC or sham treatment were co-immunoprecipitated (co-IP) using anti-Cav-3 or control IgG antibody. The co-IP samples were analyzed by Western blot by probing with specific antibodies to proteins that are known to associate with Cav-3 such as NOS3, PKC␣, and AT1 receptors. Representative Western blots (Fig. 6A) show that NOS3, PKC␣, and AT1 receptors co-precipitated with Cav-3 from WT or Cav-3 OE myocytes following sham treatment. In contrast, PKC␣ and AT1 receptor did not co-IP with anti-Cav-3 antibody from TAC myocyte lysates. In contrast, NOS3, a Cav-3 associated protein, was found to co-IP with sham and TAC myocyte lysates. ␤ 1 AR, which does not associate with Cav-3, did not co-IP with anti-Cav-3 antibody from sham or TAC myocyte lysates. We then tested whether the overexpression of Cav-3 in the ventricular myocytes from Cav-3 OE mice

. Cardiac-specific Cav-3 OE inhibits I Ca, T in ventricular myocytes from TAC-or Ang-II infusion-induced hypertrophic mice.
A and B, mean current-voltage response of I Ca, L (E) and I Ca, T (OE) measured from left ventricular myocytes from Cav-3 OE mice after 4 weeks of sham or TAC treatment, respectively. C, mean peak I Ca, T density (measured at Ϫ30 mV). D, mean I Ca, L density (measured at 0 mV) in ventricular myocytes from Cav-3 OE mice after 4 weeks of sham or TAC, and saline or Ang-II infusion were not significantly different. E, whole cell membrane capacitance of ventricular myocytes from WT mice subjected to TAC or Ang-II infusion was significantly increased compared with sham or saline infusion, respectively. F, whole cell membrane capacitance of ventricular myocytes from Cav-3 OE mice subjected to TAC or Ang-II infusion was not significantly different compared with sham or saline infusion, respectively. Data represent means Ϯ S.E. *, p Ͻ 0.05, n ϭ 8 -10 cells from five mice each group. N.S., not significant. FIGURE 5. Changes in expression of key signaling protein in ventricular myocytes during cardiac hypertrophy. Left ventricular myocyte lysates from either WT or Cav-3 OE mice subjected to TAC or sham surgery were separated by SDS-PAGE and Western blot analysis by probing with specific antibodies to Ca v 3.1, Ca v 3.2, PKC␣, PKC␤1, NFATc3, AT1 receptor, and GAPDH. Representative immunoblots indicated for respective proteins and GAPDH signals as loading control are shown on the left. The bar plots on the right show semi-quantitative densitometry analysis for indicated protein expression normalized to GAPDH signals. The expression levels for Ca v 3.2 and PKC␣ were significantly increased in WT TAC hearts compared with WT sham hearts. The AT1 receptor levels were significantly reduced in WT and Cav-3 OE TAC hearts compared with respective sham hearts. (Note that same immunoblot membrane was used to probe for Ca v 3.2 and AT1-R and also for PKC␤ and NFATc3). Data represents mean Ϯ S.E. n ϭ 4 experiments, *, p Ͻ 0.05.
prevented TAC-induced myocyte remodeling and disruption of caveola-localized signaling proteins. As shown in Fig. 6A, PKC␣, AT1 receptor, and NOS3 co-IPed with anti-Cav-3 antibody from Cav-3 OE mice subjected to TAC or sham myocytes. To further confirm these above results, we also performed sucrose density membrane fractionation on the WT and Cav-3 OE subjected to TAC or sham surgery and isolated caveolaenriched membrane fractions. As described under "Materials and Methods," equal volumes of the sucrose density gradient membrane fractions were loaded onto SDS-polyacrylamide gels and analyzed by Western blot by probing with antibodies to Cav-3, PKC␣, and AT1 receptor. Representative Western blot analysis on the density gradient membrane fractions from WT sham (left panel) or WT TAC (middle panel) Cav-3 OE TACtreated hearts is shown in Fig. 6B. For the WT sham hearts, the highest enrichment for the Cav-3 signal, indicating enrichment for caveolae, was noticed in the lower density fractions 4 -6, as has been reported in many previous studies (18, 28 -30). Identical enrichment and distribution for PKC␣ and AT1 receptor was also detected in the same caveola-enriched fractions (frac-tion 4 -6). The corresponding optical density for the distribution of Cav-3, PKC␣, and AT1 receptor is shown in Fig. 6C (left  panel). In contrast, the Cav-3, PKC␣, and AT1-receptor distribution was shifted to a higher density gradient fraction (fraction 7 and above; middle panels, Fig. 6, B and C) from WT hearts subjected to TAC suggesting that reduced expression of Cav-3 may have resulted in disruption of caveolae and altered distribution of these signaling proteins. In contrast, the distribution pattern for Cav-3, PKC␣, and the AT1 receptor from Cav-3 OE hearts subjected to TAC was similar (Fig. 6, B and C, right panels, fractions 4 -6) to that of WT sham hearts (control). The above experiment was repeated to confirm reproducibility of the results. These results suggest that Cav-3 overexpression is associated with reduced TAC-induced myocyte remodeling and decreased disruption of the Cav-3-associated macromolecular signaling complexes.

Cav-3 Knockdown Results in Increased Ang-II Stimulation of I Ca, T Mediated by PKC␣ in Neonatal Mouse Ventricular
Myocytes-To further investigate the impact of reduced Cav-3 expression and mechanism of increased I Ca, T in Ang-II-medi-FIGURE 6. Cardiac-specific Cav-3 OE in mice prevents disruption of caveola-associated macromolecular signaling complexes following TAC-induced cardiac hypertrophy. Left ventricular myocyte lysates from either WT or Cav-3 OE mice subjected to TAC or sham were used for immunoprecipitation with anti Cav-3 antibody. Immunoprecipitates were separated by Western blot analysis and probed with antibodies to NOS3, ␤ 1 AR, PKC␣, AT1-R, and Cav-3. Representative immunoblots show NOS3, PKC␣, and AT1-R but not the ␤ 1 -adrenergic receptor (␤1-AR) co-immunoprecipitated with anti-Cav-3 antibody from WT sham lysates, whereas control IgG does not immunoprecipitate the proteins (A). However, PKC␣ and AT1-R did not co-IP with anti-Cav-3 from WT TAC myocyte lysates, whereas NOS3 co-immunoprecipitated with Cav-3. In contrast to WT TAC, the NOS3, AT1-R, and PKC␣ co-immunoprecipitated with anti-Cav-3 from ventricular myocyte lysates from Cav-3 OE mice subjected to TAC. B, representative Western blot analysis performed on caveola-enriched membrane fractions prepared using ventricular myocytes from WT sham, WT TAC, and Cav-3 OE TAC hearts. Precipitated proteins from gradient membrane fractions analyzed Western blot by probing with antibodies to PKC␣, AT1-R, and Cav-3. C, respective plots show relative distribution for PKC␣ (f), AT1-R (•), and Cav-3 (OE) and protein recovery in each of the gradient fractions as indicated (E). Results are representative of data from two separate experiments. SEPTEMBER 4, 2015 • VOLUME 290 • NUMBER 36 ated cardiac hypertrophy, we used cultured NMVMs, which are known to endogenously express the I Ca, T (11,31,32). Cultured NMVMs were treated with Ang-II (10 mol/liter) or vehicle for 48 h, and I Ca, T was measured. As shown in Fig. 7, A and B, Ang-II treatment caused a significant increase (38%) in the peak I Ca, T (Ϫ8.7 Ϯ 0.8 pA/pF) compared with control (Ϫ6.3 Ϯ 1.1 pA/pF) NMVMs. In separate experiments using NMVMs, we performed siRNA-mediated knockdown of Cav-3 using specific siRNA oligonucleotides to Cav-3 or overexpression of Cav-3 using cDNA of Cav-3 as described previously (11). siRNA-mediated knockdown of Cav-3 or overexpression of Cav-3 was confirmed by Western blot analysis (Fig. 7, C and D). As shown in Fig. 7, A and B, siRNA-mediated knockdown of Cav-3 further enhanced (112%) Ang-II stimulation of peak I Ca, T (Ϫ18.6 Ϯ 7 pA/pF) compared with scrambled Cav-3 siRNA (Ϫ7 Ϯ 0.8 pA/PF) or vehicle-treated NMVMs (Ϫ3.9 Ϯ 0.8 pA/pF; 372%). In contrast, Cav-3 overexpression inhibited the basal peak I Ca, T and abolished the Ang-II stimulation of peak I Ca, T (Ϫ2 Ϯ 0.7 pA/pF). Previous studies have demonstrated that PKC␣ activates and regulates the Ca v 3.2 channel current (33,34). Increased PKC␣ expression and signaling were reported in cardiac hypertrophy and heart failure (35,36). Chronic activation of the renin angiotensin system is known to induce cardiac hypertrophy, and Ang-II stimulation of cardiomyocytes causes increase in I Ca, T in a PKC-dependent fash-ion (37,38). We rationalized that with reduced expression of Cav-3, PKC␣ may couple to the Ca v 3.2 channels resulting in an enhanced regulation of the I Ca, T in the myocytes. To test this, we performed knockdown of the PKC␣ using specific shRNA. The NMVMs were co-transfected with eGFP and either Cav-3 siRNA ϩ shRNA PKC␣, scrambled Cav-3 siRNA ϩ shRNA PKC␣, scrambled Cav-3 siRNA ϩ empty vector, Cav-3 ϩ shRNA PKC␣, or Cav-3 ϩ empty vector. Knockdown of the PKC␣ or Cav-3 or overexpression of Cav-3 was confirmed by semi-quantitative Western blot analysis as shown in Fig. 8, C and D. The transfected myocytes were treated with vehicle and Ang-II (10 mol/liter) for 48 h, and I Ca, T was measured in single cells expressing GFP. The knockdown of PKC␣ completely abolished the Ang-II stimulation of peak I Ca, T (Ϫ1 Ϯ 0.37 pA/pF) compared with Ang-II stimulation of peak I Ca, T (Ϫ21.8 Ϯ 4.6 pA/pF) (Fig. 8, A and B). Interestingly, knockdown of PKC␣ did not impact the peak I Ca, T in the vehicle-treated control NMVMs (4.6 Ϯ 1 pA/pF) compared with vehicletreated NMVMs transfected with GFP alone (Ϫ6.3 Ϯ 1.1 pA/pF) or scrambled Cav-3 siRNA (Ϫ6.3 Ϯ 3.7 pA/pF), indicating that PKC␣ did not regulate the basal I Ca, T currents in the NMVMs. These data clearly suggest that the Ang-II stimulation of the I Ca, T is specifically mediated by PKC␣ in the NMVMs. Moreover, in NMVMs co-transfected with shRNA to PKC␣ and Cav-3 cDNA, the basal (vehicle-treated) and Ang-II-stimulated I Ca, T

FIGURE 7. siRNA-mediated knockdown of Cav-3 expression increases Ang-II stimulation of I Ca, T in neonatal mouse ventricular myocytes.
NMVMs were transfected with either eGFP alone, Cav-3 siRNA ϩ eGFP, scrambled Cav-3 siRNAϩ eGFP, and Cav-3 cDNA ϩ eGFP. I Ca, T was measured using a dual pulse protocol as described under "Materials and Methods." A, representative peak I Ca, T traces (at Ϫ30 mV) recorded from NMVMs that were transfected as indicated and treated with vehicle (control) or Ang-II (10 mol/liter for 48 h). B, mean peak I Ca, T in NMVMs. Ang-II treatment caused significant increase in the I Ca, T (*, p Ͻ 0.05). Cav-3 siRNA caused a further robust increase in the Ang-II stimulation of peak I Ca, T (#, p Ͻ 0.005). Scrambled Cav-3 siRNA, used as control, also significantly increased I Ca, T compared with vehicle (#, p Ͻ 0.005). Basal and Ang-II stimulation of I Ca, T was significantly inhibited in NMVMs transfected with Cav-3 cDNA compared with control treatment (*, p Ͻ 0.05). Data represent means Ϯ S.E.; n ϭ 5-7 cells from three separate transfections. C, representative Western blots show protein expression for Cav-3 and GAPDH in NMVMs. D, semi-quantitative densitometry analysis for Cav-3 expression normalized to GAPDH signals. siRNA-mediated knockdown of Cav-3 caused a significant reduction in the expression of Cav-3 in the NMVMs compared with control scrambled siRNA-transfected cells (p Ͻ 0.001). The NMVMs transfected with Cav-3 cDNA showed significantly higher Cav-3 expression compared with eGFP (p Ͻ 0.005). Data represent means Ϯ S.E., n ϭ 6.

Caveolin-3 Overexpression Attenuates Cardiac Hypertrophy
were significantly reduced (vehicle Ϫ0.55 Ϯ 1 pA/pF versus Ang-II Ϫ1.1 Ϯ 0.5 pA/pF). These data suggest that Cav-3 overexpression inhibits both the basal and Ang-II stimulation of the I Ca, T that is mediated by the PKC␣. These above measurements were performed using a dual pulse protocol as described under "Materials and Methods," which allowed us to also record the I Ca, L density in the NMVMs under various treatment conditions. Importantly, the Ang-II treatment did not alter peak I Ca, L density compared with vehicle (control)-treated NMVMs. In addition, the peak I Ca, L density was unchanged following siRNA-mediated knockdown of Cav-3 or shRNA-mediated knockdown of PKC␣ or with Cav-3 overexpression in any of the treatment groups (data not shown). Taken together, our results strongly suggest that knockdown of Cav-3 results in enhanced Ang-II stimulation of the I Ca, T , whereas overexpression of Cav-3 abrogates both basal and PKC␣ mediated Ang-II stimulation of the I Ca, T .
Cav-3 Overexpression Attenuates the Activation of Calcineurin/ NFAT Signaling in Neonatal Ventricular Myocytes-The activation of the Ca 2ϩ -dependent calcineurin-NFAT signaling pathway is involved in Ang-II-induced pathological cardiac hypertrophy (39,40). We hypothesized that the enhanced I Ca, T due to reduced Cav-3 expression will activate the Ca 2ϩ -dependent calcineurin/ NFAT signaling following Ang-II stimulation, and this effect can be reversed with stable Cav-3 expression. We tested our hypothesis in the cultured NMVMs by transfecting with either Cav-3 siRNA or scrambled control Cav-3 siRNA or cDNA to Cav-3 (Cav-3 overexpression) and co-transfected with m-Cherry. After 24 h in culture, the NMVMs were infected with NFATc3-GFP adenovirus (gift from Dr. Steve Houser, Temple University). Cells were then treated with Ang-II or vehicle in the presence of 4 mmol/liter extracellular Ca 2ϩ . 24 h later, NMVMs were stained with the nuclear stain DAPI and imaged under a confocal microscope to determine cytoplasm versus nuclear localization of the NFATc3-GFP signal. As shown in representative images (Fig. 9A), in the vehicle-treated control NMVMs, almost the entire NFATc3-GFP signal was observed in the cytoplasm. In contrast, Ang-II treatment caused significant nuclear translocation of NFATc3-GFP in 75% (p Ͻ 0.05; Fig. 9B) of the cells. Similarly, Ang-II treatment in NMVMs caused NFATc3-GFP translocation into the nucleus in 87.5% of NMVMs (p Ͻ 0.05; Fig. 9B), where Cav-3 was knocked down by specific siRNA. However, in NMVMs overexpressing Cav-3, the nuclear translocation of NFATc3-GFP was almost completely inhibited (only 5% of cells showed nuclear localization of NFATc3-GFP) upon treatment with Ang-II, which was similar to the cells treated with vehicle. Interestingly, Cav-3 knockdown alone did not cause nuclear translocation of the NFATc3-GFP. These data show that Cav-3 overexpression prevents nuclear translocation of the NFATc3-GFP upon Ang-II stimulation. Taken together, these data suggest that Cav-3 overexpression prevents local increase in the Ca 2ϩ levels via inhibition of Ang-II stimulation of I Ca, T , and it thereby prevents activation of the calcineurin/NFAT signaling.

Discussion
This study investigated whether and how a loss of Cav-3 in pressure overload-induced cardiac hypertrophy impacts myocyte Ca 2ϩ signaling and leads to pathological cardiac hypertro-  SEPTEMBER 4, 2015 • VOLUME 290 • NUMBER 36 phy. The results presented here highlight several novel and important findings. We show reduced Cav-3 expression and abundance of caveolae and a simultaneous increase in the Ca v 3.2 protein and I Ca, T in the ventricular myocytes in cardiac hypertrophy. Reduced Cav-3 expression resulted in dissociation of the AT1 receptor and PKC␣ from Cav-3 in the hypertrophic ventricular myocytes. In NMVMs, siRNA-mediated knockdown of Cav-3 results in increased Ang-II stimulation of I Ca, T mediated by PKC␣ and caused calcineurin-dependent NFAT translocation into the nucleus. In contrast, Cav-3 overexpression inhibited the PKC␣-mediated Ang-II stimulation of the I Ca, T and prevented NFATc3 translocation into the nucleus. In addition, mice with cardiac-specific Cav-3 overex-pression had reduced expression of I Ca, T and prevented disruption of the Cav-3-associated macromolecular signaling complexes after exposure to cardiac hypertrophic stimuli. Taken together, our data demonstrate that Cav-3 overexpression protects against pressure overload-induced cardiac hypertrophy via inhibition of I Ca, T and suppression of the Ca 2ϩ -dependent hypertrophic calcineurin-NFAT signaling pathway.

Caveolin-3 Overexpression Attenuates Cardiac Hypertrophy
Previous work (15) and current investigations demonstrate that caveola and Cav-3 expression is essential to cardiac protection (anti-hypertrophic signaling). The reduced Cav-3 expression in the cardiomyocytes during cardiac hypertrophy (Fig. 1) is consistent with previously published results (13,20). Besides a variety of signaling proteins, Cav-3 associates and localizes the Ca v 3.2 channels, AT1 receptor, and the PKC isoforms into caveolae and provides local regulation of Ca 2ϩ signaling in the cardiomyocytes (9). During pathological remodeling of myocardium, the structural integrity of myocytes is altered, resulting in changes in the distribution of the ion channels and associated signaling proteins, which causes a loss of protein-protein interaction (41). A reduction in Cav-3 expression and reduced abundance of caveolae in cardiomyocytes in cardiac hypertrophy could lead to altered subcellular localization and changes in composition of caveola-associated macromolecular signaling proteins. Previous studies have demonstrated that caveolar localization of key signaling proteins, including soluble guanylyl cyclase and cGMP-dependent protein kinase, is disrupted during pressure (30) or volume overload (29)-induced cardiac hypertrophy. The latter study also showed that caveolar localization protected soluble guanylyl cyclase against oxidation. It was shown that Cav-3 knockdown prevented the redistribution of 5-HT 2A receptors into caveolar domains (20). Similarly, our data show that Cav-3, AT-1 receptor, and PKC␣ were associated and formed a macromolecular signaling complex in normal cardiomyocytes, which was disrupted by a loss of Cav-3 and caveola expression in the hypertrophic ventricular myocytes (Fig. 6). The loss of Cav-3 expression and combined with an up-regulation of the PKC␣ and dissociation of PKC␣ from Cav-3 augmented enhanced coupling of PKC␣ with the Ca v 3.2 channels resulting in increased Ang-II stimulation of the I Ca, T . In contrast, the overexpression of Cav-3 reversed these effects. Therefore, we propose that caveolae provide a safety mechanism against activation of hypertrophic signaling. Re-expression of fetal I Ca, T in the ventricular myocytes during pathological hypertrophy is well established (22,23,42). Studies have demonstrated the expression of the Ca v 3.2 (␣ 1H ) channel current responsible for the development of cardiac hypertrophy (27,44) and the expression of the Ca v 3.1 (␣ 1G ) channels is attributed to anti-hypertrophic effect and cardioprotective function (45). It was reported that in pathological hypertrophy the Ca 2ϩ influx via the re-expressed Ca v 3.2 channel initiates the binding of calcineurin to the C terminus of Ca v 3.2 leading to activation of NFAT (46). Furthermore, treatment with TTCC blockers could prevent the development of cardiac hypertrophy and heart failure (27,47,48). We have recently demonstrated that both the cardiac TTCC isoforms, Ca v 3.1 and Ca v 3.2 subunits, are associated with Cav-3. However, Cav-3 specifically inhibits the Ca v 3.2 current but not the Ca v 3.1 cur- FIGURE 9. Overexpression of Cav-3 inhibits Ang-II-induced NFATc3-GFP translocation to nucleus. Representative images of NMVMs showing localization NFATc3-GFP following vehicle (control) or Ang-II treatment. Freshly isolated NMVMs were transfected with Cav-3 siRNA or Cav-3 cDNA plasmid, grown in culture for 12 h, and then infected with NFATc3-GFP adenovirus. 24 h after infection, the cells were treated with vehicle (control) or Ang-II (10mol/liter), and the culture media were supplemented with 4 mmol/liter Ca 2ϩ . GFP-tagged NFAT-C3-infected NMVMs were co-stained with nuclear stain DAPI (blue) and m-Cherry as transfection control. 24 h of Ang-II treatment caused a nuclear translocation of NFATc3-GFP compared with control. siRNA-mediated Cav-3 knockdown caused nuclear translocation of NFATc3-GFP in nearly all of NMVMs. In contrast, the NMVMs overexpressing Cav-3 nuclear translocation of NFATc3-GFP was completely inhibited. Scale bar, 50 m. Data are representative of four separate experiments. N.S., not significant.

Caveolin-3 Overexpression Attenuates Cardiac Hypertrophy
rents (11). These above reports, including ours, suggest a likely scenario of Ca v 3.1 and Ca v 3.2 channels activating different signaling pathways within the same caveola via specific coupling mechanisms with different signaling proteins. In this study, we show an increase in the mRNA level for Ca v 3.1 (␣ 1G ) and Ca v 3.2 (␣ 1H ) mRNA (Fig. 2) and protein (Fig. 5) and an increase in the I Ca, T in cardiac hypertrophy . We could not specifically measure the contribution of I Cav3.2 versus I Cav3.1 in the cardiomyocytes during hypertrophy due to nonavailability of specific inhibitors for these TTCC isoforms. Interestingly, we did not observe any changes to the I Ca, L density in cardiomyocytes in cardiac hypertrophy in the WT or the Cav-3 OE mice (Figs. 3D and 4D). Some studies reported a reduction or no change or an increase in the I Ca, L density in hypertrophy (49,50), whereas other reports suggested a role for the LTCC current in the pathological hypertrophy (51,52). A recent report indicates that caveolalocalized LTCC can activate the calcineurin/NFAT-mediated hypertrophic signaling in cardiomyocytes (53). Subsequently, it was shown that Ca 2ϩ influx through LTCCs primarily activates the Cn-NFAT signaling, and Ca 2ϩ entry through transient receptor potential (TRP) channels also participated in this process (54). An earlier report indicated that TRP channels as necessary mediators of pathological cardiac hypertrophy through a calcineurin-NFAT signaling pathway (55). Although the TRP3 channel has been shown to localize to caveolae in the arterial smooth muscle cells (17), it is not known whether the TRP channels are associated with caveolar signaling proteins in the ventricular myocytes. We did not examine the role of Cav-3 in FIGURE 10. Proposed model of Cav-3-mediated cardiac protection during cardiac hypertrophy. A, during pressure overload-induced cardiac hypertrophy, a reduced expression of Cav-3 and caveola leads to disruption of caveola-localized and Cav-3-associated signaling proteins, including the AT1-R and PKC␣. As a result, an increased Ang-II stimulation and PKC␣-mediated activation of the I Ca, T lead to an increase in the local intracellular Ca 2ϩ levels. This enhanced Ca 2ϩ then activates the calmodulin-sensitive calcineurin, which then dephosphorylates NFATc3 triggering a hypertrophic response. However, in model B Cav-3 overexpression inhibits I Ca, T and prevents up-regulation of local Ca 2ϩ levels and prevents activation of downstream calcineurin/NFATc3 signaling. Cav-3 overexpression also causes caveola formation, which may stabilize the Cav-3-associated macromolecular signaling proteins and therefore protects against pressure overload-induced cardiac hypertrophy. regulation TRP channel currents. Our data clearly suggest that cardiomyocyte caveolae localize essential signals that regulate the Ca 2ϩ influx-mediated hypertrophic signaling. Likely differences between our observations of unchanged I Ca, L density with other studies could be due to differences in the models of pathological cardiac hypertrophy (early stage) versus heart failure (43). Future studies should investigate a clear role for the Ca v 1.2 channels, including the expression of auxiliary subunits and isoforms during the development of pathological cardiac hypertrophy and heart failure. Nevertheless, this study clearly establishes the essential protective function of Cav-3 and caveolae in the regulation of Ca 2ϩ -dependent signaling mechanisms in pathological hypertrophy.

Conclusion
We demonstrate that the loss of Cav-3 and caveola expression in ventricular myocytes in cardiac hypertrophy impacts the Cav-3-mediated compartmentalized regulation of local signaling. A loss of Cav-3 inhibition of the I Ca, T , specifically the I Ca v3.2 , results in increased local intracellular Ca 2ϩ levels that activate calmodulin-dependent calcineurin, which then dephosphorylates the NFAT and triggers hypertrophic responses (Fig. 10A). In contrast, Cav-3 overexpression in the ventricular myocytes prevents pressure overload-induced cardiac hypertrophy via at least two possible mechanisms. Overexpression of Cav-3 directly inhibits I Ca, T and prevents an increase in microdomain Ca 2ϩ levels. Next. Cav-3 stabilizes the caveola-localized macromolecular signaling complexes and prevents increased coupling of PKC␣ with the Ca v 3.2 channels (Fig. 10B). We conclude that Cav-3 overexpression in ventricular myocytes is essential for promoting the protective signaling during pressure overloadinduced cardiac hypertrophy and thus could be used as therapeutic strategy for treatment of such disease.