G-protein-coupled Receptor Signaling Components Localize in Both Sarcolemmal and Intracellular Caveolin-3-associated Microdomains in Adult Cardiac Myocytes*

This study tests the hypothesis that G-protein-coupled receptor (GPCR) signaling components involved in the regulation of adenylyl cyclase (AC) localize with caveolin (Cav), a protein marker for caveolae, in both cell-surface and intracellular membrane regions. Using sucrose density fractionation of adult cardiac myocytes, we detected Cav-3 in both buoyant membrane fractions (BF) and heavy/non-buoyant fractions (HF); (cid:1) 2 -adrener- gic receptors (AR) in BF; and AC5/6, (cid:1) 1 -AR, M 4 -musca-rinic acetylcholine receptors (mAChR), (cid:2) -opioid receptors, and G (cid:3) s in both BF and HF. In contrast, M 2 -mAChR, G (cid:3) i3 , and G (cid:3) i2 were found only in HF. Immunofluores- cence microscopy showed co-localization of Cav-3 with AC5/6, G (cid:3) s , (cid:1) 2 -AR, and (cid:2) -opioid receptors in both sar- colemmal

The caveolin/lipid raft signaling hypothesis proposes that compartmentation of signaling molecules in caveolae provides a mechanism for temporal and spatial signal transduction and cross-talk among signaling pathways (15). Despite substantial data supporting this notion (16,17), little is known regarding the expression of caveolae in the plasma membrane versus non-plasma membrane locations. One cell type in which GPCR compartmentation has been studied is the cardiac myocyte (CM), but virtually all previous studies related to caveolae have involved the use of neonatal CM (9,18). Adult CM differ from neonatal CM in numerous ways, including being more extensively differentiated, multinucleated, and larger in volume and possessing a more developed T-tubule network. In this study, we hypothesized that Cav-3 co-localizes with GPCR signaling components in both sarcolemmal and intracellular membranes and used adult CM to determine whether (a) Cav distributes differently in adult CM versus other cardiac cell types; (b) GPCR (e.g. ␤-adrenergic, muscarinic, and -opioid receptors) and key post-receptor signaling components, including heterotrimeric G-proteins and adenylyl cyclase, co-localize with Cav-3 in intracellular domains (i.e. T-tubules) in addition to the sarcolemma; and (c) Cav-3 scaffolds GPCR signaling components capable of stimulating and inhibiting cAMP synthesis.
CM Preparation-Adult male Sprague-Dawley rats (250 -300 g) were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg), and hearts were excised and retrograde-perfused with medium containing collagenase II (Worthington) as described previously (19). Animals were heparinized (1000 -2000 units intraperitoneal) 5 min prior to administration of anesthesia. Hearts were removed and placed in ice-cold cardioplegic solution (112 mM NaCl, 5.4 mM KCl, 1 mM MgCl 2 , 9 mM NaH 2 PO 4 , and 11.1 mM D-glucose supplemented with 10 mM HEPES, 30 mM taurine, 2 mM DL-carnitine, and 2 mM creatine, pH 7.4). The hearts were retrograde-perfused on a Langendorff apparatus at a rate of 5 ml/min for 5 min at 37°C, followed by perfusion with medium containing collagenase II (250 units/mg) for 20 min. Following perfusion, both ventricles were isolated, minced in collagenase II-containing medium for 10 -15 min, washed several times, and re-acclimated to 1.2 mM Ca 2ϩ over 25 min to produce calcium-tolerant CM. To remove all non-myocytes, myocytes were plated in 4% fetal bovine serum on laminin (2 g/cm 2 )-coated plates for 1 h, followed by serum-free medium (1% bovine serum albumin); CM were incubated at 37°C in 5% CO 2 for 24 h prior to experiments.
Membrane Fractionation-CM were fractionated using both detergent-free and detergent-containing (1% Triton X-100) methods (20,21). Buffer containing 10 mM KH 2 PO 4 , 5 mM MgCl 2 , 5 mM EDTA, and 1 mM EGTA was used to extract the contractile myofibrils as described previously (22). CM from a 15-cm plate were washed twice with ice-cold phosphate-buffered saline (PBS) and scraped in 3 ml of either 500 mM sodium carbonate, pH 11.0, to extract peripheral membrane proteins or TNE buffer (25 mM Tris-HCl, 150 mM NaCl, and 5 mM EDTA) containing 1% Triton X-100. For detergent-free extraction, cells were homogenized by three 10-s bursts of a tissue grinder and then sonicated by three cycles of 20-s bursts of sonication and 1 min of incubation on ice. Approximately 2 ml of homogenate were mixed with 2 ml of 90% sucrose in MES-buffered saline (25 mM MES and 150 mM NaCl, pH 6.5) to form 45% sucrose and loaded at the bottom of an ultracentrifuge tube. A discontinuous sucrose gradient was generated by layering 4 ml of 35% sucrose prepared in MES-buffered saline and 250 mM Na 2 CO 3 , followed by 4 ml of 5% sucrose also in MES-buffered saline/Na 2 CO 3 . Gradients were centrifuged at 280,000 ϫ g using a Beckman SW 41Ti rotor for 16 -20 h at 4°C. For subcellular fractionation using Triton X-100, 2 ml of homogenate were mixed with 2 ml of 90% sucrose in TNE buffer. A discontinuous sucrose gradient was generated by layering 4 ml of 30% sucrose in TNE buffer, followed by 4 ml of 5% sucrose in TNE buffer and centrifugation at 190,000 ϫ g using the SW 41Ti rotor for 16 -20 h at 4°C. Samples were removed in 1-ml aliquots to form 12 fractions.
Immunoblot Analysis-Proteins in fractions and cell lysates were separated by SDS-PAGE using 10 or 12% acrylamide precast gels (Invitrogen) and transferred to polyvinylidene difluoride membranes (Millipore Corp.) by electroelution. Membranes were blocked with 20 mM PBS and 1% Tween containing 1.5% nonfat dry milk and incubated overnight with primary antibody at 4°C. Primary antibodies were visualized using horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology, Inc.) and ECL reagent (Amersham Biosciences). All displayed bands migrated at the appropriate size as determined by comparison with molecular weight standards (Santa Cruz Biotechnology, Inc.). The amount of protein per fraction was determined using a dye binding protein assay (Bio-Rad).
Immunofluorescence Microscopy of Adult Heart and CM-Adult rat heart ventricles and CM were prepared for immunofluorescence microscopy as described (23,24). CM were plated on laminin (2 g/cm 2 )precoated glass coverslips and grown for 24 h. Rat ventricles were freshly harvested, frozen, and then mounted on a cryostat (Ϫ23°C) to cut 10-m semithin sections. Semithin sections and cells were fixed with 2% paraformaldehyde in PBS for 10 min at room temperature; incubated with 100 mM glycine, pH 7.4, for 10 min to quench aldehyde groups; permeabilized in 0.1% buffered Triton X-100 for 10 min; blocked with 1% bovine serum albumin, PBS, and 0.05% Tween for 20 min; and then incubated with primary antibodies (1:100) in 1% bovine serum albumin, PBS, and 0.05% Tween for 24 -48 h at 4°C. Excess antibody was removed by incubation with PBS and 0.1% Tween for 15 min, and samples were incubated with fluorescein isothiocyanate-or Alexa-conjugated secondary antibody (1:250) for 1 h. To remove excess secondary antibody, semithin sections and cells were washed six times at 5-min intervals with PBS and 0.1% Tween and incubated for 20 min with the nuclear stain 4Ј,6-diamidino-2-phenylindole (1:5000) diluted in PBS. Sections and cells were washed for 10 min with PBS and mounted in gelvatol for microscopic imaging.
Deconvolution Image Analysis-Deconvolution images were obtained as described (25,26) and captured with a DeltaVision deconvolution microscope system (Applied Precision, LLC, Issaquah, WA). The system includes a Photometrics CCD camera mounted on a Nikon TE-200 inverted epifluorescence microscope. Between 30 and 80 optical sections spaced by ϳ0.1-0.3 m were generally taken. Exposure times were set such that the camera response was in the linear range for each fluorophore. Lenses included ϫ100 (numerical aperture of 1.4), ϫ60 (numerical aperture of 1.4), and ϫ40 (numerical aperture of 1.3). The data sets were deconvolved and analyzed using SoftWorx software (Applied Precision, LLC) on a Silicon Graphics Octane workstation. Image analysis was performed with the Data Inspector program in SoftWorx. Maximal projection volume views or single optical sections are shown as indicated.
Electron Microscopy-Cells were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer, gently scraped, and pelleted. The resuspended pellet was fixed for 1 h at room temperature with 8% paraformaldehyde and then overnight with 4% paraformaldehyde at 40°C. Cells were washed three times with 0.1 M phosphate buffer, embedded in 10% gelatin, and cryoprotected overnight in 2.3 M sucrose at 40°C. Cryosections were cut on a Leica UltraCut E ultramicrotome; 80-nm ultrathin sections were mounted on glow-discharged nickel grids and stored on 2% gelatin until labeled. Sections were blocked with 1% goat serum and 1% bovine serum albumin and incubated with primary antibodies for 2 h to overnight, followed by 5-or 10-nm gold-labeled goat anti-rabbit or goat anti-mouse IgG (Amersham Biosciences). Sections were then absorption-stained with uranyl acetate and embedded in 0.2% methyl cellulose. For AC localization, cells were treated with an adenoviral construct containing either LacZ (control) or AC6 for 24 h prior to fixation (18).
Measurement of AC Activity-AC activity was measured in Cav-3 immunoprecipitates using a modification of a previously described method (10). A 15-cm plate of adult CM was homogenized on ice in lysis buffer, precleared with protein A-agarose for 1 h, and incubated with primary antibody for 1 h at 4°C. Antibody conjugates were immunoprecipitated with protein G-agarose for 1 h at 4°C and centrifuged at 13,000 ϫ g for 5 min. Agarose pellets were washed once with lysis buffer and subsequently with wash buffers A and B and then resuspended in 30 mM Na-HEPES, 5 mM MgCl 2 , and 2 mM dithiothreitol, pH 7.5. Protein (30 l of immunoprecipitate) was added to tubes containing 30 mM Na-HEPES, pH 7.5, 100 mM NaCl, 1 mM EGTA, 10 mM MgCl 2 , 1 mM isobutylmethylxanthine (a cyclic nucleotide phosphodiesterase inhibitor), 1 mM ATP, 10 mM phosphocreatine, 5 M GTP, 60 units/ml creatine phosphokinase, and 0.1% bovine serum albumin. After 5 min, 0.1 M naloxone (an opioid receptor antagonist) or vehicle was added, followed 5 min later by addition of 1 M [D-Ala 2 ,N-MePhe 4 ,Gly 5 -ol]enkephalin (DAMGO; a selective -OR agonist) or vehicle. After 5 min, 10 M forskolin was added, and samples were incubated for an additional 15 min. The reaction was stopped by boiling for 5 min, and cAMP content was assayed as described previously (27).

Caveolin Distributes Differently in Adult CM Compared with
Other Cardiovascular Cells-The distribution of Cav was investigated in human coronary artery smooth muscle cells (hCASMC) and adult rat cardiac fibroblasts (ACF) and CM following sucrose density fractionation. We detected the majority of Cav-1 in BF 4 and 5 from ACF and hCASMC (Fig. 1A). In adult CM, Cav-3 was detected in both BF 4 and 5 and HF/ non-BF 9 -12 (Fig. 1A). Following equal protein loading of sucrose density fractions of adult CM (Fig. 1B), BF were enriched in Cav-3; HF had a lower proportion of Cav-3 (Fig. 1A, lower panels). BF contained Ͻ5% of total cellular protein (Fig.  1B). Adult CM fractions in Fig. 1A (from equal volume-loaded gels) that were immunoblotted for the T-tubule markers vinculin and DHPR (a marker for voltage-sensitive calcium chan-nels) showed a distribution pattern similar to that with Cav-3 (Fig. 1C). RyR (a marker of the sarcoplasmic reticulum (SR)) and ␤-adaptin (a marker of clathrin-coated pits) were detected only in HF (Fig. 1C). To test whether the Cav-3 distribution pattern observed upon Na 2 CO 3 /sucrose density fractionation of adult CM was unique to this method of cellular disruption and fractionation, we fractionated adult CM in alternative lysis buffers, one containing 1% Triton X-100 and the other containing a high salt buffer to extract contractile myofibrils. Immunoblotting of equal volume-loaded fractions from the Triton X-100 fractionation revealed a broad distribution of Cav-3 (Fig.  1D). Following extraction of the contractile myofibrils using high salt buffer, we detected buoyant and heavy pools of Cav-3, albeit with a substantially greater amount of Cav-3 in the heavier fractions (Fig. 1D). Overall, these results show that Cav-3 was present in both BF and HF following sucrose density fractionation, indicating a different cellular distribution compared with Cav-1 in two other cardiovascular cell types, hCASMC and ACF.
GPCR Signaling Components Distribute Non-uniformly with Cav-3 in Adult CM-Using immunoblot analyses, we investigated co-localization of G s and G i protein-coupled receptor signaling components with Cav-3 in adult CM. AC5/6 and ␤ 1 -and ␤ 2 -AR were detected in BF, whereas a portion of ␤ 1 -AR was also detected in HF ( Fig. 2A). G␣ s was detected as two bands (which represent the short and long splice variants of G␣ s ) in BF and HF ( Fig. 2A). Because the ␤-AR and mAChR pathways act antagonistically in the regulation of cardiac rate and force of contraction, we investigated whether M 2 -and M 4 -mAChR localize to BF. We found that M 2 -mAChR were excluded from BF, as were G␣ i3 and G␣ i2 , G-proteins through which both M 2 -and M 4 -mAChR signal, whereas we detected M 4 -mAChR in both BF and HF (Fig. 2, B and C). We also detected -OR, another G i -coupled receptor, in adult CM by PCR (data not shown) and immunoblotting of both BF and HF (Fig. 2B).
As an alternative means to test for interaction of the signaling components with Cav-3, we assessed BF and HF following immunoprecipitation with an anti-Cav-3 antibody (Fig. 2D). AC5/6, ␤ 1 -and ␤ 2 -AR, G␣ s , M 4 -mAChR, -OR, and Cav-3 were all detected in Cav-3 immunoprecipitates from both BF and HF. By contrast, M 2 -mAChR and G␣ i3 (Fig. 2D) were detected only in Cav-3 immunoprecipitates from HF, whereas G␣ i2 and ␤-adaptin were not detected in immunoprecipitates from either BF or HF. Thus, the results with the Cav-3 immunoprecipitates from BF and HF confirm the immunoblot findings with sucrose density fractions and demonstrate that GPCR signaling components vary in their cellular distribution with Cav-3.

FIG. 4. Immunofluorescence and deconvolution analysis of the co-localization of Cav-3 with -OR and M 4 -mAChR in adult CM.
Cells were stained with antibodies for Cav-3, -OR, and M 4 -mAChR. Images were deconvolved and are shown as single-stained or overlaid to show co-localization. As a negative control, incubation with secondary antibodies (2ab CTNL) revealed minimal background staining (lower panel).

FIG. 2. Expression and localization of ␤-adrenergic, muscarinic, and opioid receptor signaling components upon sucrose density fractionations of Na 2 CO 3 extracts and Cav-3 immunoprecipitates of BF and HF generated from Na 2 CO 3 extracts from adult CM.
A, shown is the AC5/6, ␤ 2 -and ␤ 1 -AR, and G␣ s localization upon sucrose density fractionation of Na 2 CO 3 extracts. B, fractions 4 and 5 (BF) and 9 -12 (HF) were pooled and compared with whole cell lysates (WCL) from adult CM for expression of mAChR subtypes (M 2 and M 4 ), -OR, G␣ i3 , and Cav-3 by immunoblot analysis. Approximately 4 g of protein from each fraction were loaded into each lane. C, shown is the G␣ i2 and Cav-3 localization upon sucrose density fractionation of Na 2 CO 3 extracts (equal volume). D, BF and HF generated from adult CM sucrose density fractionations were pH-neutralized with HCl and immunoprecipitated with anti-Cav-3 antibody, and immunoprecipitates were then probed with antibodies for ␤-adaptin, AC5/6, ␤ 2 -and ␤ 1 -AR, -OR, M 2 -and M 4 -mAChR, G␣ s , G␣ i3 , G␣ i2 , and Cav-3. TX, Triton X-100.
Immunofluorescence Microscopy of Adult Heart Shows Colocalization of Cav-3 and G␣ s Protein in Both Sarcolemmal and Intracellular Regions-Immunofluorescence microscopy re-vealed that the cellular distribution of Cav-3 in adult heart was similar to that in adult CM (Figs. 3 and 4), with Cav-3 in a punctate pattern in the sarcolemma and in striations that ran transversely across the cell interior (Fig. 6, upper panel). In tissue sections, G␣ s co-localized with Cav-3 in intracellular regions, most predominantly in intercalated discs, which demarcate junctions between adjacent myocytes (Fig. 6, lower  panel, arrow).

Immunofluorescence Microscopy Shows Co-localization of Cav-3 with T-tubule (DHPR and Vinculin) and Sarcoplasmic Reticulum (RyR) Markers in Both Adult CM and
Heart-Immunostaining of adult CM revealed that Cav-3 co-localized with two different T-tubule markers (vinculin and DHPR) in a transverse pattern and along the sarcolemma (Fig. 7A). Tissue sections of heart assessed with antibodies for Cav-3 and DHPR revealed a pattern similar to that found in isolated CM, with Cav-3 and DHPR co-localizing in intracellular regions running transversely across the cell, along the sarcolemma, and in intercalated discs (Fig. 7B, right panel, arrow). Staining for Cav-3 and RyR (an SR marker) revealed co-localization predominantly in intracellular regions running transversely across the cell in both isolated CM and heart (Fig. 8A)  intercalated discs in sections of heart (Fig. 8B, right panel,  arrows). We detected minimal co-localization of Cav-3 and ␣-actinin, a Z-disc marker (data not shown). As an alternative means to test for Cav-3/RyR interaction, we assessed lysates from CM immunoprecipitated with anti-Cav-3 and anti-RyR antibodies (Fig. 8C) and observed Cav-3 and RyR in both Cav-3 and RyR immunoprecipitates, thus confirming results from immunofluorescence microscopy showing co-localization of Cav-3 and the SR marker. Thus, Cav-3 localizes not only along the sarcolemma, but also in T-tubule/SR regions in both adult CM and heart.
Immunoelectron Microscopy Detects Cav-3 in Sarcolemmal Invaginations and Intracellular Domains in Adult CM-We used immunogold labeling as an additional means to assess the distribution of Cav-3 in adult CM. Immunoelectron microscopy demonstrated abundant sarcolemmal caveolae, present as invaginations that labeled with antibodies directed against Cav-3 (Fig. 9, A and B). We also detected Cav-3 in membranes flanked by Z-discs, which correspond to the T-tubule network within the cell interior (Fig. 9C). These immunoelectron microscopic data confirmed and extended results from immunofluorescence microscopy demonstrating Cav-3 in both sarcolemmal and intracellular regions and are consistent with the detection of Cav-3 in HF following sucrose density fractionation (Fig. 1).
Immunoelectron Microscopy Shows Co-localization of Cav-3 with AC6 and G␣ s -Using immunogold labeling, we detected Cav-3 (10-nm gold) with AC6 (5-nm gold) in invaginations of the sarcolemma (Fig. 9D) and in intracellular membranes (Fig.  9E) located between adjacent Z-discs. We also detected Cav-3 (10-nm gold) with G␣ s (5-nm gold) in vesicles near myofibrils (Fig. 9, F and G). These immunoelectron microscopic findings are consistent with the results of Cav-3 immunoprecipitation studies showing multiprotein interaction between Cav-3 and AC6 and G␣ s in adult CM (Fig. 2C).

AC6 Overexpression Enhances GPCR-stimulated cAMP Production in
Adult CM-To confirm that the overexpressed AC6 that we analyzed microscopically (Figs. 3 and 9) was functional, we assessed cAMP production in adult CM. Overexpression of AC6 increased the levels of cAMP produced in response to isoproterenol (a ␤-AR agonist) without an increase in agonist potency (Fig. 10A). Thus, the overexpressed AC6 that co-localized with Cav-3 in adult CM (as shown by both immunofluorescence and immunoelectron microscopy) was enzymatically active. Moreover, by assaying forskolin-stimulated cAMP accumulation, we obtained functional evidence for the presence of -OR in adult CM: the -OR agonist DAMGO significantly reduced forskolin-stimulated cAMP production (p Ͻ 0.05, n ϭ 5) (Fig. 10B).
Cav-3 Immunoprecipitates Multiprotein Complexes Capable of Producing and Inhibiting cAMP-To demonstrate that Cav-3 interacts with and organizes components that mediate stimulation and inhibition of cAMP formation, we assessed AC activity in Cav-3 immunoprecipitates of adult CM lysates. Forskolin increased cAMP production (Fig. 10C); this stimulation was significantly (p Ͻ 0.01) inhibited by DAMGO, a response that was inhibited by the opioid receptor antagonist naloxone. Immunoblot analysis detected -OR and G␣ i3 in the Cav-3 immunoprecipitates used in the AC activity experiments (data not shown). These data show functional evidence for the existence of -OR and support the conclusion that Cav-3 interacts with multiprotein complexes that can both stimulate and inhibit cAMP production in adult CM.
FIG. 6. Immunofluorescence and deconvolution analysis of the co-localization of Cav-3 and G␣ s in adult heart. Semithin sections (5 m) of adult heart were co-stained with antibodies for Cav-3 and G␣ s . Cav-3 co-localized with G␣ s in transverse striations within the cell and in intercalated discs between two cardiac myocytes. Images were deconvolved and are shown as single-stained or overlaid to show colocalization. Scale bar ϭ 10 m.

FIG. 7. Immunofluorescence and deconvolution analysis of the co-localization of Cav-3 and the T-tubule markers DHPR and vinculin in adult CM and heart. A, cells stained with antibodies for
Cav-3, DHPR, and vinculin revealed co-localization of Cav-3 and Ttubule markers in the sarcolemma and in a transverse intracellular pattern. B, semithin sections (5 m) of heart stained with antibodies for Cav-3 and DHPR displayed co-localization along the sarcolemma and in transverse intracellular striations and intercalated discs (arrow) between adjacent CM. Images were deconvolved and are shown as singlestained or overlaid to show co-localization. Scale bar ϭ 30 m in B.

DISCUSSION
Caveolae-localized signaling microdomains have been proposed as sites that concentrate GPCR, heterotrimeric G-proteins, and G-protein-regulated effector molecules so as to facilitate coordinated, precise, and rapid regulation of cell function (5, 8 -14). Cav-rich domains thus may serve as spatial organizers of GPCR signaling, although not all data have supported this conclusion (13,28). Using multiple experimental approaches (subcellular fractionation, immunoprecipitation, immunofluorescence, immunoelectron microscopy, and functional assays of cAMP formation), we have provided new evidence that extends the notion of Cav-rich domains as organizers of GPCR signaling components and in support of the novel conclusion that Cav-3 organizes GPCR signaling components in both sarcolemmal and intracellular regions (e.g. T-tubules in cardiac myocytes) in adult heart.
Most previous studies on compartmentation of GPCR signaling components in CM have utilized embryonic and neonatal cells (9,18,27,29). However, it is preferable to study adult CM, which are more akin to the in vivo setting than are neonatal CM and other non-striated cardiovascular cells (30). Because of development-related changes in ion channels and contractile proteins, it can be difficult to extrapolate results from the neonatal to the adult heart. The present results regarding Cav-3 distribution in adult CM differ from findings reported for neonatal CM and certain non-cardiac cell types in which Cav distributed only to BF (9,18,27,29,31). Our demonstration that M 2 -mAChR were excluded from BF and expressed in intracellular sites confirms and extends previous results for adult CM (32) and contrasts with findings obtained with neonatal CM in which M 2 -mAChR were detected in both BF and HF (9). Such differences imply developmental changes related to caveolar compartmentation of the ␤-AR and mAChR signaling cascades, which perhaps contribute to differences during development in response to physiologic stimuli. Compartmentation of GPCR signaling components in adult CM also contrasts with findings obtained with adult rat aortic smooth muscle cells in which ␤ 1 -and ␤ 2 -AR were found only in HF, implying that localization of particular GPCR to Cav-rich fractions is cell type-dependent (2,10).
Previous workers have observed intracellular Cav-3 in skeletal myocytes (6,33,34). A possible role for intracellular Cav in adult CM may be as a regulator of calcium homeostasis (35): T-tubules in adult CM are continuous with the sarcolemma and are essential for the influx of calcium via L-type Ca 2ϩ channels or DHPR and regulation of myocyte contractility (36 -39). DHPR are located primarily at the T-tubule/SR junction proximal to where the SR Ca 2ϩ release channels or RyR are found  9. Immunoelectron microscopic localization of Cav-3 with AC6 and G␣ s in the sarcolemma and in intracellular membranes morphologically corresponding to T-tubules in adult CM. Immunogold labeling detected Cav-3 (10-nm gold) in the sarcolemmal membrane (A and B) and in intracellular membranes anchored to Z-discs (Z) (C). Additional immunogold labeling detected Cav-3 (10-nm gold) and AC6 (5-nm gold) in sarcolemmal invaginations (D) and in intracellular membranes near Z-discs (E). Similarly, Cav-3 (10-nm gold) and G␣ s (5-nm gold) were detected in intracellular membranes corresponding to T-tubules (t) (F and G). Scale bars ϭ 0.05 m. (40,41). We have shown co-localization of Cav-3 with DHPR and RyR in adult CM and whole tissue (Figs. 7 and 8) and Cav-3 in regions between adjacent Z-discs morphologically corresponding to T-tubules (Fig. 9), results that complement evidence for Cav-3 in T-tubules in striated myocytes (6,33). In addition, our detection of AC5/6 in intracellular membranes that correspond to T-tubules agrees with results that suggest a T-tubule localization of the AC5/6 protein (42). The biochemical and immunofluorescence data indicate that the inhibition of AC by G i -coupled receptors occurs in Cav-3-rich microdomains in the T-tubule system. Intracellular Cav may provide a scaffold that helps organize GPCR signaling components and proteins that regulate calcium homeostasis at the T-tubule/SR junction. We attempted to assess for differences in AC activity between sarcolemmal and intracellular Cav-rich regions, but because of the lengthy preparation required to isolate fractions and to conduct activity assays, we were unable to detect stable AC enzyme activity (data not shown).
Although caveolae are morphologic entities, virtually all previous work has utilized subcellular fractionation or immunoprecipitation to infer co-localization of GPCR signaling components and caveolins. Plating of adult CM on laminin-coated surfaces allowed us to utilize microscopic techniques (both light and electron) to show Cav-3 in a punctate pattern along the sarcolemma and in intracellular transverse striations. This study thus provides the first microscopic evidence of Cav-3 co-localization with GPCR and GPCR signaling components in both sarcolemma and intracellular membranes (in particular, T-tubule-associated membrane regions) in adult CM (Fig. 9, D-G). The results from immunofluorescence microscopy of adult myocardium showing a Cav-3 distribution similar to that seen in isolated adult CM (Figs. 6, 7B, and 8B) support the use of isolated adult CM as an in vitro model (30).
Opioid receptors have been shown to play an important role in protecting the heart from ischemic injury (43,44) and arrhythmias (45). However, ambiguity exists regarding the receptor subtypes expressed and activated by agonists in the myocardium (46); in particular, the presence of -OR has been disputed (47). Past studies that utilized radioligand binding experiments were performed on membranes prepared from whole hearts, making it difficult to distinguish sarcolemmal from intracellular CM membranes and membranes contributed by other cell types. Our results obtained using four different techniques (PCR, Western blotting, immunofluorescence microscopy, and assay of cAMP generation) provide evidence consistent with the idea that functional -OR are expressed in CM of adult heart.
In conclusion, in this study, we used multiple complementary techniques (subcellular fractionation, immunochemistry, morphology, and functional assays) to document a role for cardiac Cav-3 as an organizer of signaling components that regulate cAMP production for multiple classes of GPCR (i.e. ␤-AR, mAChR, and -OR). The results imply that spatial organization of GPCR signaling components occurs in microdomains in both sarcolemmal and intracellular membrane regions in the heart.