Substrate uptake and metabolism are preserved in hypertrophic caveolin-3 knockout hearts

Ayanna S. Augustus, Jonathan Buchanan, Sankar Addya, Giuseppe Rengo, Richard G. Pestell, Paolo Fortina, Walter J. Koch, Andre Bensadoun, E. Dale Abel, Michael P. Lisanti


Caveolin-3 (Cav3), the primary protein component of caveolae in muscle cells, regulates numerous signaling pathways including insulin receptor signaling and facilitates free fatty acid (FA) uptake by interacting with several FA transport proteins. We previously reported that Cav3 knockout mice (Cav3KO) develop cardiac hypertrophy with diminished contractile function; however, the effects of Cav3 gene ablation on cardiac substrate utilization are unknown. The present study revealed that the uptake and oxidation of FAs and glucose were normal in hypertrophic Cav3KO hearts. Real-time PCR analysis revealed normal expression of lipid metabolism genes including FA translocase (CD36) and carnitine palmitoyl transferase-1 in Cav3KO hearts. Interestingly, myocardial cAMP content was significantly increased by 42%; however, this had no effect on PKA activity in Cav3KO hearts. Microarray expression analysis revealed a marked increase in the expression of genes involved in receptor trafficking to the plasma membrane, including Rab4a and the expression of WD repeat/FYVE domain containing proteins. We observed a fourfold increase in the expression of cellular retinol binding protein-III and a 3.5-fold increase in 17β-hydroxysteroid dehydrogenase type 11, a member of the short-chain dehydrogenase/reductase family involved in the biosynthesis and inactivation of steroid hormones. In summary, a loss of Cav3 in the heart leads to cardiac hypertrophy with normal substrate utilization. Moreover, a loss of Cav3 mRNA altered the expression of several genes not previously linked to cardiac growth and function. Thus we have identified a number of new target genes associated with the pathogenesis of cardiac hypertrophy.

  • cardiomyopathy
  • caveolae
  • fatty acids
  • isolated perfused hearts
  • adenosine 3′,5′-cyclic monophosphate

first described nearly sixty years ago, caveolae are 50- to 100-nm flask-shaped plasma membrane invaginations enriched in cholesterol, sphingolipids, and the primary protein component, caveolin. Caveolins and caveolae are involved in a number of diverse cellular processes including signal transduction, vesicular transport, and cholesterol metabolism. The caveolin gene family consists of three gene products including caveolin-1 and -2, which are coexpressed in several tissues including adipocytes, fibroblast, smooth muscle, and endothelial cells. Caveolin-3 (Cav3) is expressed in skeletal, smooth, and cardiac muscle and is involved in several myocyte-specific functions. During development, Cav3 expression increases threefold beginning at postnatal day 2 until day 90 in the rat heart (41). Cav3 associates with the T tubule system during development (40) and localizes to the sarcolemma in mature muscle fibers (43). Furthermore, Cav3-associated caveolae compartmentalize a number of signaling molecules including α- and β-adrenergic receptors (ARs), PKC isoforms and Src family tyrosine kinases (16, 42), and insulin receptor (23). Muscle-cell Cav3 also associates with several members of the dystrophin-glycoprotein complex, abnormalities of which are linked to several human diseases including Duchenne muscular dystrophy (43).

Given that caveolins act as a scaffold for a number of proteins involved in cardiac substrate utilization, including insulin receptor (23) and fatty acid (FA) transport protein CD36 (49), we set out to characterize the metabolic consequences of Cav3 gene ablation on cardiac energy substrate uptake and metabolism. Several groups have demonstrated diminished insulin responsiveness and glucose uptake in muscle cells lacking Cav3. Fecchi et al. (15) recently reported that the loss of Cav3 expression in conditionally immortalized skeletal muscle cells significantly reduced insulin-stimulated glucose uptake by blocking the activation of key downstream proteins required for the effective transduction of insulin signaling. This resulted in diminished movement of glucose transport protein-4 (GLUT4) containing vesicles to the plasma membrane in Cav3 knockout (KO) skeletal muscle myotubes (15). Our group, along with others, has shown that Cav3KO mice developed insulin resistance in skeletal muscle with significantly reduced insulin receptor protein content (9, 50). Therefore, we wanted to determine whether a loss of Cav3 in the heart would have a similar effect on cardiac glucose utilization.

FA oxidation accounts for 60–70% of oxygen consumption in the normal adult heart with the balance provided by glucose and lactate (46, 47). Physiological hypertrophy, a beneficial adaptive response to bouts of exercise training, increases myocardial reliance on FA as an energy source and increases the expression of FA metabolism genes (29). However, during pathological hypertrophy (caused by pressure overload due to hypertension or valvular disease), the heart becomes more dependent on glucose as an energy source (47). Pathological cardiac hypertrophy is a compensatory adaptation to an increase in the workload of the heart, marked by reduced left ventricular cardiac function. Initially, this adaptive response normalizes myocardial wall stress and oxygen consumption, maintaining cardiac output and ATP levels (1). However, under continuous stress, this adaptive response leads to decompensated cardiac hypertrophy and heart failure (34). The metabolic phenotype associated with pathological cardiac hypertrophy includes reduced expression of genes involved in FA uptake and oxidation, increased expression of genes involved in glucose metabolism, and accelerated rates of basal glucose uptake and glycolysis with decreased glucose oxidation.

We previously demonstrated that the loss of Cav3 expression (Cav3KO) in cardiomyocytes resulted in a 20% increase in left ventricular wall thickness and a 20% decrease in fractional shortening (FS) in hearts from 4-mo-old Cav3KO mice (50). Histological analysis revealed progressive interstitial and perivascular fibrosis and increased myocyte cell size and number likely due to the hyperactivation of the −p42/44 MAPK (ERK1/2) pathway, a known regulator of myocyte cell growth (8). Recent studies have shown that the transgenic overexpression of Ras, a GTPase that activates the ERK1/2 pathway, produces hypertrophic cardiomyopathy characterized by pathological ventricular remodeling and premature death (34). Interestingly, many of the genes in the electron transport chain, FA metabolism, and Krebs cycle were significantly decreased in this mouse model, suggesting diminished FA metabolism. Our studies describe for the first time the metabolic consequences of Cav3 gene ablation on myocardial substrate metabolism.



Mice were housed and maintained in a barrier facility at Thomas Jefferson University. All experiments were performed on 6 h-fasted, 5- to 6-mo-old male mice. Studies were approved by the Institutional Care and Use Committees at Thomas Jefferson University and the University of Utah. The generation of Cav3KO mice on the C57Bl/J6 background was as previously described (21).

Transaortic echocardiography.

Transthoracic two-dimensional echocardiography in mice was performed using a 12-MHz transducer (VisualSONICS VeVo 770 imaging system) as described previously (14). M-mode echocardiography was carried out in the parasternal short axis in Cav3KO and wild-type mice at 5 to 6 mo of age to assess left ventricular anterior wall (AW) and posterior wall (PW) thickness, FS, ejection fraction, and heart rate.

Lipoprotein lipase activity and mass.

Postheparin plasma lipoprotein lipase (LPL) activity was analyzed as described previously (48). We injected fasting male mice intravenously with 10 units of heparin and collected blood 10 min later. Plasma samples were assayed in triplicate for LPL activity. LPL activity in homogenized tissues was measured as described by Hocquette et al. (27). Murine LPL protein was measured by enzyme-linked immunosorbent assay as described previously by van Vlijmen et al. (48).

In vivo kinetics studies.

Intralipid emulsion labeled with [3H]triolein was prepared as previously described (2). [14C]palmitate (PerkinElmer Life Sciences) was complexed to 6% FA-free bovine serum albumin (Sigma) as described previously (10). Labeled FA and triglyceride (TG)-labeled Intralipid emulsion were injected simultaneously into 6 h-fasted Cav3KO and wild-type mice to assess FA delivery to the heart via albumin-bound and TG-derived FA. Cav3 is not expressed in the liver; thus, a loss of Cav3 should not impact hepatic lipid metabolism. The uptake of lipid moieties by the liver was also measured as a control. A bolus of 400,000 counts/min (cpm) of [14C]palmitate with 1 × 106 cpm of labeled [3H]TG containing particles at a final concentration of 1 mg TG/mouse was injected. Eighty microliters of blood were collected from the retroorbital plexus 30 s and 1, 2, and 5 min later. Ten microliters of plasma from each time point were used to determine plasma decay. At the completion of the study, mice were anesthetized, the heart was isolated and perfused with 10 ml of 1× PBS through the left ventricle. Tissues were collected and homogenized in 5 ml of 1× PBS, 1-ml aliquots were added to 3.5 ml of scintillation fluid, and radioactivity was counted to determine tissue radioactivity.

In vivo assessment of cardiac glucose metabolism.

Basal glucose uptake was measured in hearts following an intravenous administration of 3 μCi of 2-deoxy-d-[1-14C]glucose (NEN Life Science Products). Sixty minutes later, hearts were perfused with 1× PBS, tissues were excised, and radioactive counts were measured.

Substrate metabolism in isolated working mouse hearts.

All hearts were prepared and perfused in the working mode, using protocols that have been previously described (7). In brief, the working heart buffer was Krebs-Henseleit buffer containing (in mM) 118.5 NaCl, 25 NaHCO3, 4.7 KCl, 1.2 MgSO4, 1.2 KH2PO4, 2.5 CaCl2, 0.5 EDTA, and 5 glucose, gassed with 95% O2-5% CO2 and supplemented with 0.4 mM palmitate bound to 3% BSA in the absence of insulin. Glycolytic flux was determined by measuring the amount of 3H2O released from the metabolism of exogenous [5-3H]glucose (specific activity, 177 Gbq/mol). Glucose oxidation was determined by trapping and measuring 14CO2 released by the metabolism of [U-14C]glucose (specific activity, 96 Mbq/mol). Palmitate oxidation was determined in separate perfused hearts by measuring the amount of 3H2O released from [9,10-3H]palmitate (specific activity, 42 Gbq/mol). Cardiac hydraulic work, [J·min−1·g−1 WHW] = CO (ml/min) × DevP (mmHg) × 1.33 × 10−4/g WHW, where WHW is wet heart weight, CO is the cardiac output, and DevP is the developed pressure, was determined.

Determination of cAMP level and PKA activity.

An Amersham cAMP [3H] assay kit was used to measure cAMP concentration in the heart. Snap-frozen tissue was homogenized on ice in 4 mM EDTA containing buffer, followed by heating for several minutes to coagulate the protein. After centrifugation, cAMP in the supernatant was assayed. PKA activity was measured with a cAMP-dependent protein kinase (PKA) assay system (Promega, Madison, WI) according to manufacturer's recommendations.

Real-time PCR.

Total RNA was isolated from hearts using TRIzol reagent (Invitrogen, Carlsbad, CA). Quantitative real-time PCR was performed with SYBR green PCR core reagents (Applied Biosystems, Foster City, CA). The incorporation of the SYBR green dye into the PCR products was monitored in real time with a 7900 HT Sequence detection system (Applied Biosystems). Several primer pairs were designed using OligoPerfect designer (Invitrogen). Samples were normalized against cylophilin. Available primer sequences used are as follows: cyclophilin, sense 5′-AGCACTGGAGAGAAAGGATTTGG-3′ and anti-sense 5′-TCTTCTTGCTGGTCTTGCCATT-3′; GLUT1, sense 5′-TCGTAACGAGGAGAACCG-3′ and anti-sense 5′-GGCCGTGTTGACGATA-3′; GLUT4, sense 5′-AGAGTCTAAAGCGCCT-3′ and anti-sense 5′-CCGAGACCAACGTGAA-3′; pyruvate dehydrogenase kinase 4, sense 5′-GACCGCTTAGTGAACAC-3′ and anti-sense 5′-GTAACGGGGTCCACTG-3′; CD36, sense 5′-ATTGGTCAAGCCAGCT-3′ and anti-sense 5′-TGTAGGCTCATCCACTAC-3′; FATP6, sense 5′-ACACCTACGAGGACGTGGAC-3′ and anti-sense 5′-CAGTGGGTTCACAGGTGTTG-3′; and FATP1, sense 5′-GACGTTGACATCCGTAAAG-3′ and anti-sense 5′-CAGTAACAGTCCGCCT-3′; acyl-CoA oxidase (ACO), sense 5′-CCCAGATTGTGAGTGTGTGG-3′ and anti-sense 5′-AGCCACCATGATTGAAGTCC-3′; carnitine palmitoyltransferase 1 (CPT-1), sense 5′- TACAACCCTGACGATG-3′ and anti-sense 5′-GGTACGAGTTCTCGAT-3′; peroxisome proliferator-activated receptor-α (PPARα), sense 5′-GGCTGAGCGTAGGTAAT-3′ and anti-sense 5′-CGCAACGTAGAGTGCTG-3′; and PPARδ, sense 5′-CCTACAACGAGATCAG-3′ and anti-sense 5′-CGTCTTTGTTGACGATG-3′. Primers for aldolase, retinal binding protein-7 (cellular), hydroxysteroid (17-β) dehydrogenase 11, and Rab4a were synthesized and purchased from SuperArray Bioscience (Frederick, MD).

Microarray analysis.

RNA isolated from Cav3KO and wild-type hearts (n = 3 per group) was treated with DNase (Invitrogen), and RNA was further purified using the RNeasy purification kit (Qiagen). RNA samples were not pooled. RNA was quantified on a Nanodrop ND-100 spectrophotometer, followed by RNA quality assessment by analysis on an Agilent 2100 bioanalyser (Agilent, Palo Alto, CA). Two micograms of total RNA from two groups (wild-type and Cav3KO) and three replicate in each group were used for Affymetrix one-cycle target labeling method as recommended by the manufacturer (Affymetrix, Santa Clara, CA). Each of six Affymetrix GeneChip for Mouse Genome 430 2.0 was hybridized for 16 h with biotin-labeled fragmented cRNA (10 μg) in 200 μl of hybridization cocktail according to Affymetrix protocol. Arrays were washed and stained using GeneChip Fluidic Station 450, and hybridization signals were amplified using antibody amplification with goat IgG (Sigma-Aldrich) and anti-streptavidin biotinylated antibody. Chips were scanned on an Affymetrix GeneChip Scanner 3000 using GeneChip Operating Software version 3.0. Normalization was done using Robust Multichip Average (RMA) and per gene normalized to control samples (wild-type) with GeneSpring GX v7.3.1 software (Agilent). Volcano plot (data not shown) was used to identify differentially expressed genes using the parametric testing assuming variances are equal and no multiple testing correction. These data are located on the National Center for Biotechnology Information Gene Expression Omibus Web site ( The accession numbers are as follows (Cav3KO samples, GSM2753233, GSM2753232, and GSM2753231; and wild-type samples, GSM2753215, GSM2753216, and GSM2753217). Independent validation of select differentially expressed genes was performed using real-time PCR.

Heart lipids.

Heart lipids were extracted as described by Folch et al. (17). Heart TG, cholesterol, and FA levels were measured enzymatically.

Analysis of tissue glycogen levels.

Tissue glycogen levels were determined according to previously described methods (30). Briefly, hearts were treated with 1 N NaOH at 80°C. Following digestion, protein precipitation was achieved by adding an equal volume of 2 N TCA. Glycogen was precipitated from the supernatant with ethanol (2:1 vol/vol). The precipitate was washed with 80% ethanol and solubilized in water. An aliquot of the glycogen precipitate was digested to glucose with 1 mg/ml amyloglucosidase (Sigma) in 0.5 M Na acetate buffer, pH 4.5, and the amount of glucose was determined using a glucose assay. Glycogen levels were determined as the difference between amyloglucosidase-treated and -untreated samples.

Immunohistochemical analysis.

Mice were euthanized and transcardially perfused with 4% paraformaldehyde in phosphate-buffered saline under physiological pressure. Fixed hearts were embedded in paraffin. Sections (5 μm) were immunostained with antibodies directed against CD45 (BD Pharmingen, San Diego, CA). Briefly, paraffin sections were dewaxed in xylene for 20 min, rehydrated in alcohol, and washed in PBS. All sections were boiled in antigen retrieval solution for 15 min. After three washes, slides were incubated for 30 min with 3% H2O2 to block endogenous peroxidase activity, incubated with 10% normal goat serum for 1 h, and then incubated with the corresponding antibody overnight at 4°C. Sections were incubated with biotinylated IgG (1:500) for 30 min and stained using the immunoperoxidase technique (Dako, Carpinteria, CA). Samples were developed using diaminobenzidine and hydrogen peroxide and then counterstained with hematoxylin, dehydrated, and coverslips were mounted.


Data are presented as means ± SE and analyzed using the Student's t-test.


Hearts from 5- to 6-mo-old Cav3KO mice are hypertrophic.

We previously reported cardiac hypertrophy and dysfunction in Cav3KO hearts from 2- and 4-mo-old mice (50). The degree of cardiac dysfunction and ventricular hypertrophy observed in 5- to 6-mo-old Cav3KO mice is similar to that observed in 4-mo-old mice (Table 1). Echocardiography revealed a 22% decrease in left ventricular FS. In addition, AW and PW diameter during diastole are significantly increased in Cav3KO hearts, indicative of left ventricular hypertrophy.

View this table:
Table 1.

Echocardiographic data for wild-type and Cav3KO mice at 5 to 6 months of age

Loss of Cav3 does not alter cardiac energy substrate uptake.

We assessed the impact of Cav3 gene ablation on cardiac substrate uptake. The uptake of 14C-palmitate complexed to BSA (Fig. 1A) and 3H-triolen from an Intralipid emulsion (Fig. 1B) was normal in hearts from 5- to 6-mo-old mice lacking Cav3 (n = 10) compared with wild-type (n = 10). Plasma clearance of 14C-palmitate and 3H-triolen was normal in Cav3KO mice compared with wild-type mice (data not shown). The uptake of 3H-deoxyglucose by Cav3KO hearts was comparable with wild-type hearts (Fig. 1C). We next examined the protein and activity levels of LPL, the rate-limiting enzyme in the delivery of lipoprotein TG-derived FA to the heart (22). The loss of Cav3 in cardiac myocytes did not alter LPL activity or total LPL protein levels (Fig. 2, A and B).

Fig. 1.

Cardiac uptake of 14C-palmitate-BSA (A), 3H-triolein from Intralipid emulsion (B), and 2-deoxy-d-[1-14C]glucose (C) in hearts from 5- to 6-mo-old caveolin-3 knockout (Cav3KO; white bars; n = 10) was comparable with wild-type (black bars; n = 10) male mice. Data are expressed as a percentage of wild-type ±SE.

Fig. 2.

Hearts from 5- to 6-mo-old wild-type (black bars; n = 5) and Cav3KO (white bars; n = 5) male mice were homogenized and assayed for lipoprotein lipase (LPL) activity (A) and LPL mass (B). Values are expressed as means ± SE.

Gene expression and substrate metabolism are normal in hypertrophic Cav3KO hearts.

To determine the effects of Cav3 gene ablation on cardiac substrate metabolism, glycolysis, glucose oxidation, and palmitate oxidation rates were measured. Substrate utilization in hearts from Cav3KO mice was comparable with wild-type mice: glycolysis [1,781 ± 516 nmoles·min−1·g−1 dry weight (dwt) vs. 2,294 ± 382 nmoles·min−1·g−1 dwt], glucose oxidation (685 ± 211 nmoles·min−1·g−1 dwt vs. 434 ± 97 nmoles·min−1·g−1 dwt), and palmitate oxidation (384 ± 34 nmoles·min−1·g−1 dwt vs. 343 ± 28.5 nmoles·min−1·g−1 dwt; Fig. 3, AC).

Fig. 3.

Myocardial substrate utilization was measured in 5- to 6-mo-old male wild-type (black bars; n = 5) and Cav3KO (white bars; n = 5) mice. Loss of Cav3 did not significantly alter glycolysis (A), glucose oxidation (B), and myocardial palmitate oxidation rates (C). Values are expressed as means ± SE.

Cardiac hypertrophy is associated with the decreased expression of genes involved in lipid uptake and utilization and increased expression of genes involved in glucose metabolism (1). Interestingly, we did not observe a significant difference in the expression of genes involved in FA uptake, peroxisomal, or mitochondrial β-oxidation (CD36, ACO, CPT-1, PPARα, and PPARδ; Fig. 4). Conversely, we found a small but significant increase in the mRNA levels of heart-specific FATP6 (P = 0.05) in mice Cav3KO mice. The loss of Cav3 gene expression did not affect GLUT1 and GLUT4 mRNA levels.

Fig. 4.

Quantitative RT-PCR analysis of genes involved in fatty acid metabolism and glucose uptake. The expression of fatty acid translocase (CD36), fatty acid transport protein (FATP)1, FATP6, acyl-CoA oxidase (ACO), carnitine palmitoyltransferase 1 (CPT-1), peroxisome proliferator-activated receptor (PPAR)α, PPARδ, glucose transport protein (GLUT)1, and GLUT4 in hearts from 5- to 6-mo-old male Cav3KO mice (white bars; n = 9) were comparable with wild-type mice (black bars; n = 8). Gene expression was normalized to cyclophilin. Values are expressed as means ± SE. *P < 0.05. dwt, dry weight.

cAMP levels are increased in Cav3KO heart.

Caveolins are known to negatively regulate adenylyl cyclase (44), which generates the second messenger cAMP. cAMP content was significantly elevated in hearts lacking Cav3 (111.2 ± 9.6 pmoles/mg) compared with wild-type (78.4 ± 7.4 pmoles/mg; P < 0.05; Fig. 5A). Surprisingly, increased cAMP content did not result in a change in PKA activity in Cav3KO hearts (Fig. 5B).

Fig. 5.

cAMP levels (A) were significantly elevated in hearts from Cav3KO (white bars; n = 5) compared with wild-type (black bars; n = 6) mice. However, PKA activity (B) in Cav3KO hearts (n = 5) was comparable with wild-type (n = 5) hearts. Values are expressed as means ± SE. *P < 0.05.

Targeted disruption of Cav3 alters myocardial lipid content.

Cardiac TG, free FA, cholesterol, and glycogen levels were measured in 6 h-fasted, 5- to 6-mo-old wild-type and male mice. Interestingly, cardiac TG content is significantly increased (3.262 ± 0.25 mg/g vs. 2.031 ± 0.145 mg/g; P < 0.05), and cholesterol content was significantly reduced (0.467 ± 0.144 mg/g vs. 0.886 ± 0.064 mg/g; P < 0.05) in Cav3KO hearts (Fig. 6, A and C). We also observed a trend toward reduced cardiac free FA levels in Cav3KO mice (0.623 ± 0.083 μmoles/g vs. 0.356 ± 0.045 μmoles/g; P = 0.08; Fig. 6B). However, a loss of Cav3 gene expression in cardiac myocytes did not affect myocardial glycogen content (Fig. 6D).

Fig. 6.

Loss of Cav3 (white bars; n = 7) increased cardiac triglyceride (A) and free FA (FFA; B) levels while decreasing total cholesterol (C) content (wild-type; black bars; n = 6). Glycogen levels (D) in hearts from Cav3KO mice (n = 7) were comparable with those of wild-type hearts (n = 7). Values are expressed as means ± SE. *P < 0.05; **P < 0.01.

Loss of Cav3 in the heart alters the expression of a number of genes involved in endocytosis, retinol transport, lipid metabolism, and the innate immune response.

A gene expression profile was generated from total RNA isolated from 5- to 6-mo-old, 6 h-fasted wild-type and Cav3KO male hearts using Affymetrix GeneChip for Mouse Genome 430 2.0. We were surprised to find a relatively small number of genes were differentially expressed in hearts lacking Cav3 (see supplementary data; all supplemental data can be found with the online version of this article). Table 2 lists the genes of known function that were significantly changed in hearts lacking Cav3. GeneSpring software identified an increased expression of several genes encoding chemokine ligands including chemokine ligand 14, chemokine ligand 6, and chemokine ligand 9. In addition, genes involved in retinoid and steroid metabolism were upregulated in hearts lacking Cav3, cellular retinol-binding protein 7 (4.0-fold), and hydroxysteroid (17-β) dehydrogenases 11 (3.6-fold). We also observed a significant increase in the expression of the small GTPase Rab4a (2.5-fold). Primarily expressed in the liver, aldolase B, an isoenzyme of fructose 1,6-bisphosphate aldolase, was markedly reduced in hearts lacking Cav3. Microarray gene expression analysis for several genes was confirmed by real-time PCR (Fig. 7).

Fig. 7.

Microarray data were confirmed by real-time PCR. Hearts from 5- to 6-mo-old Cav3KO male mice (white bars; n = 5) had significantly increased gene expression for cellular retinol binding protein (Rbp)7, 17β-hydroxysteroid dehydrogenase type 11 (HSD17β11), and Rab4a vs. wild-type (black bars; n = 5). The expression of aldolase B was markedly reduced in Cav3KO hearts. Values are expressed as means ± SE. *P < 0.05.

View this table:
Table 2.

List of genes of known function differentially expressed in Cav3KO hearts

CD45+ positive cells are present in the fibrotic regions in hypertrophic Cav3KO hearts.

Excessive chemokine expression and activation has been observed in several models of cardiac hypertrophy, ischemia, and failing hearts (5, 11, 25). We previously demonstrated the presence of cellular infiltrates in fibrotic lesions present in Cav3KO hearts. Using an immunohistochemical approach, we observed an infiltration of CD45+ cells in perivascular fibrotic regions present in Cav3KO hearts (Fig. 8) compared with wild-type hearts, which do develop fibrosis. Low grade inflammation and leucocyte infiltration in the myocardium have been linked to cardiac dysfunction through the activation and production of reactive oxygen species and cytokines.

Fig. 8.

Immunohistochemical analysis revealed infiltration of CD45+ cells in fibrotic lesions in Cav3KO hearts. Paraffin-embedded Cav3KO (A and C) and wild-type (B and D) mouse hearts from 5- to 6-mo-old mice were subjected to immunostaining. Interestingly, CD45, a leukocyte marker, was only observed in perivascular fibrotic lesions in Cav3KO hearts (arrows). Interstitial and perivascular fibrosis and leukocyte staining were not observed in wild-type hearts (B and D).


The purpose of this study was to assess the impact of Cav3 gene ablation on cardiac energy substrate uptake and metabolism. We previously demonstrated that loss of Cav3 gene expression in cardiomyocytes results in increased left ventricular wall thickness and decreased cardiac function in 4-mo-old Cav3KO hearts (50). We also confirmed the presence of cardiac hypertrophy and diminished contractile function by echocardiography in hearts from 5- to 6-mo-old Cav3KO mice (Table 1). Echocardiography revealed a 22% decrease in FS and an increase in left ventricular wall thickness in Cav3KO hearts. These results demonstrate that the degree of hypertrophy observed in Cav3KO mice is similar to that observed in hearts from 4-mo-old Cav3KO mice.

Pathological cardiac hypertrophy is associated with decreased uptake and metabolism of FA and increased utilization of glucose (1). Surprisingly, our studies show that the loss of Cav3 does not affect myocardial substrate uptake. The uptake of albumin-bound FA and Intralipid TG and glucose under basal conditions was normal in Cav3KO hearts (Fig. 1). LPL activity and protein, a key regulator of TG-derived FA delivery to tissues, was normal in hearts lacking Cav3 (Fig. 2). Furthermore, isolated perfused heart studies revealed normal glycolytic and glucose oxidation rates in 5- to 6-mo-old Cav3KO hearts (Fig. 3). The characteristic uncoupling of glycolysis and glucose oxidation, commonly observed in more severe forms of hypertrophy, was not observed in our model.

Lack of changes in the expression of genes involved in FA and glucose metabolism, as assessed by real-time PCR (Fig. 4) and Western blotting (data not shown), fits with the absence of changes in substrate uptake and utilization. Our findings contrast with previously published reports that demonstrate a marked decrease in FA uptake and oxidation during cardiac hypertrophy (4, 24), including a model of hypterophy induced by the targeted activation of the MAP kinase pathway. Transgenic overexpression of Ras, an upstream regulator of ERK1/2 activation, resulted in the decreased expression of genes involved in the electron transport chain and FA metabolism (34). We previous reported the hyperactivation of ERK1/2 in Cav3KO hearts; however, the resulting increase in myocyte cell number and size did not alter myocardial energy metabolism in our model.

Several recent studies suggest that overt changes in cardiac energy metabolism may depend upon the degree of hypertrophy (13). As compensatory hypertrophy progresses toward decompensated hypertrophy and eventually heart failure, changes in gene expression become more pronounced resulting in a shift away from FA as a fuel source (1). Studies by Miyamoto et al. (35) demonstrated that protein levels for several enzymes involved in FA metabolism and a key transcriptional regulator of metabolic genes, PPARα were unchanged in volume-overloaded hypertrophied rabbit hearts (35). Additional studies have reported normal FA uptake, metabolism, and carnitine palmitoyltransferase-1 activity, the rate-limiting enzyme in mitochondrial β-oxidation (13). Our results indicate that the loss of Cav3 in the heart and the resulting changes in cardiac contractility and MAPK activation may not be harsh enough to trigger the metabolic switch commonly observed in other models of cardiac hypertrophy and heart failure. Additional studies are needed to assess myocardial substrate metabolism in more advanced decompensated hypertrophy.

We previously demonstrated that Cav3 acts as natural inhibitor of adenylyl cyclase (44), which generates the second messenger cAMP. cAMP-mediated signaling regulates a number of diverse pathways in the heart, from calcium ion cycling and cardiac excitation-contraction coupling to regulating protein expression and energy metabolism (36). Myocardial cAMP content was increased by 42% in Cav3KO hearts (Fig. 5A). However, this increase did not alter PKA activity (Fig. 5B), a key mediator of cAMP signaling. Recent studies have shown that the compartmentalization of cAMP/PKA signaling is key for the PKA-mediated regulation of myocyte processes such as muscle contractility (31). The loss of Cav3-associated caveolae likely disrupts the intracellular localization of these proteins. Interestingly, several groups have also shown that cAMP can impact cellular energy production by directing the flux of FA (32) and glucose (18) toward the mitochondria for oxidation. Although it is unclear how cAMP shuttles substrates to the mitochondria, this may provide a potential mechanism by which Cav3KO hearts maintain normal substrate metabolism in the presence of diminished contractile function and increase myocyte cell size. Perhaps the increase energy needed by the heart to maintain suboptimal contractile performance and maintain normal cellular processes is generated by a small but significant increase in substrate flux to the mitochondria.

Despite the normal uptake and metabolism of FA and TG-derived FA in Cav3KO hearts, myocardial TG stores were elevated by 41% (Fig. 6A). Several groups have shown that excess lipid accumulation in the heart leads to cardiac dysfunction either due to the increased production of reactive oxygen species or increased generation of lipid intermediates such as ceramide (39). However, recent studies suggest that lipoprotein production by the heart may be a potential mechanism by which the heart can eliminate the excess lipid. Apolipoprotein B (apoB) is primarily expressed in the liver and is the primary protein component of TG-rich very low-density lipoproteins. Studies by Boren and coworkers (6, 37) have shown that human and murine hearts express apoB mRNA and secrete apoB-containing lipoproteins. Moreover, recent studies have shown that the overexpression of human apoB in a mouse model of lipotoxic cardiomyopathy reduced the cardiomyopathy and premature death observed in this mouse model (52). Although a role for Cav3 in lipid droplet formation and secretion by the heart has not been described, it is possible that Cav3 may regulate apoB processing and secretion in the heart by some unknown mechanism. New studies are needed to explore this possibility.

To gain a greater understanding of the role of Cav3 in cardiac hypertrophy, we performed microarray gene expression analysis (see supplementary data). We were surprised to find very few genes were differentially expressed in Cav3KO hearts (Table 2). Microarray analysis revealed the expression of genes involved in FA and glucose uptake, and metabolism was normal in Cav3KO hearts. Moreover, the expression of caveolin isoforms-1 and -2 did not change, and we did not observe changes in the downstream genes involved in MAPK signaling (data not shown). However, we observed a 2.4-fold increase in Rab4a and a sixfold increase in the expression of WD repeat/FYVE domain sequence. Rab4a has been implicated in regulating cardiac contractility (38) and β2-AR receptor movement from endosomes to the plasma membrane (12, 45). Proteins containing WD repeats and a FYVE domain include ProF, a protein shown to target vesicular membranes, binds Akt and PKCζ (19, 20). Both Akt and PKCζ are known to regulate GLUT4 trafficking (28). Since a number of receptors have been shown to localize to caveolae, including insulin receptor (23) and β2-AR (42), these data further implicate Cav3 in regulating receptor cycling through secretory pathways (26).

Surprisingly, we observed a 3.5-fold increase in the expression of 17β-hydroxysteroid dehydrogenase type 11 (17β-HSD11), a member of the short-chain dehydrogenase/reductase family primarily expressed in the liver. 17β-HSD11 has been shown to play a role in the biosynthesis and inactivation of steroid hormones by localizing to the endoplasmic reticulum, where it regulates neutral lipid droplet formation in transfected cells (51). Although a role for 17β-HSD11 in the heart has not yet been described, these findings, in combination with the 41% increase in TG and 47% decrease in cholesterol content, lead us to speculate that 17β-HSD11 may play a role in regulating cardiac TG storage and utilization. Recently, the cardiac-specific overexpression of another member of the HSD family, 11β-HSD2, has been shown to produce marked cardiac hypertrophy, with pronounced dilation, increased cardiomyocyte cell size, and interstitial fibrosis (33), a phenotype that is similar to that observed in Cav3KO hearts.

Microarray gene expression analysis revealed a marked increase in the expression of several chemokine ligands, suggesting alternations in innate immunity in the Cav3KO mouse. The primary role of chemokines is the recruitment and activation of specific subpopulations of leucocytes. Chemokines are essential for control of wound healing and infection; however, excessive chemokine activation may lead to an appropriate inflammation, resulting in cell and tissue damage (3). Several studies have shown that chemokine expression is increased in failing (11), pressure-overload hypertrophic (5), and ischemic hearts (25). Moreover, low-grade inflammation with infiltrating leucocytes has been found in failing human hearts (3). We previously demonstrated the presence of cellular infiltrates in myocardial fibrotic lesions of Cav3KO hearts (50). Immunohistochemical analysis of Cav3KO hearts revealed the presence of CD45+ cells (marker of leukocytes) in fibrotic lesions (Fig. 8). Our data support recent findings implicating chemokines in pathological cardiac remodeling and function.

In summary, we have shown that hypertrophic Cav3KO hearts have normal FA and glucose uptake and metabolism. The expression of key genes known to regulate substrate metabolism was unchanged despite decreased left ventricular contractile function. Several studies suggest that the severity of pathological cardiac hypertrophy and heart failure dictates the metabolic switch toward glucose and away from FA as the primary fuel source (13). Our model of cardiac hypertrophy is unique in that the cardiomyopathy does not appear to result in altered cardiac substrate metabolism. The hypertrophy observed in Cav3KO hearts may not be severe enough to produce marked changes in the expression of genes involved in substrate metabolism. Increased cAMP levels observed in Cav3KO hearts may also provide a potential mechanism by which Cav3KO hearts maintain normal substrate metabolism in the presence of diminished contractile function. Several studies suggest that cAMP may enhance the metabolic capacity of Cav3KO hearts by shuttling substrates to the mitochondria for oxidation (18, 32). Additional studies are needed to determine whether substrate utilization swifts as Cav3KO hearts progress from compensatory cardiac hypertrophy toward a decompensated state and heart failure. Microarray analysis revealed significant changes in genes involved in vesicular transport, steroid hormone metabolism, cellular retinol transport, and innate immunity. Further exploration is needed to address the functional link between each individual gene and its role in regulating cardiac growth and metabolism.


This work was supported by grants from the National Institutes of Health, the Muscular Dystrophy Association, and the American Heart Association (all to M. P. Lisanti) and grants from the Pennsylvania Department of Health (SAP4100026302 to P. Fortina) and the Kimmel Cancer Center (Philadelphia, PA; to P. Fortina). A. S. Augustus was supported by postdoctoral training grants 5T32HL001675 and 5T32DK07705. J. Buchanan was supported by training grant T35HL07744.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract