Inhibition of PKCα and rhoA Translocation in Differentiated Smooth Muscle by a Caveolin Scaffolding Domain Peptide

Experimental Cell Research 258, 72– 81 (2000) doi:10.1006/excr.2000.4891, available online at http://www.idealibrary.com on

Inhibition of PKC␣ and rhoA Translocation in Differentiated Smooth Muscle by a Caveolin Scaffolding Domain Peptide Michael J. Taggart,* ,† ,1 Paul Leavis,* Olivier Feron,‡ and Kathleen G. Morgan* ,§ *Boston Biomedical Research Institute, Watertown, Massachusetts 02472; §Cardiovascular Division, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts 02215; ‡Department of Medicine, Universite´ Catholique de Louvain, B-1200 Brussels, Belgium; and †University Department of Medicine, Manchester Royal Infirmary, Manchester, M13 9WL, Great Britain

coordination of many signal transducing events. For example, stimulus-dependent relocalizations of both protein kinase C␣ (PKC␣) and rhoA from the cytosol to the plasma membrane [1] have been suggested to be key factors in agonist-induced Ca 2⫹ sensitization of myofilament contractility [2– 4]. Unfortunately, it is poorly understood how such signal integration is regulated in smooth muscle. Caveolae are invaginations 50 – 80 nm wide that appear in periodic register along the smooth muscle cell membrane interspersed by regions of dense bodies anchoring the cytoskeleton [5, 6]. Following their initial discovery in nonmuscle cells [7, 8] the exact function of caveolae remained uncertain, although they were suggested to be involved in many cellular processes, including cholesterol transport, clathrin-independent endocytocis, and Ca 2⫹ signaling [9 –11]. In recent years, the discovery of a family of integral caveolar proteins— caveolins— has renewed interest in the possible function(s) of caveolae [12–14]. Three main mammalian caveolin isoforms exist: caveolin-1 and -2 are distributed in a wide variety of tissues, whereas caveolin-3 appears to be restricted mainly to muscle. Caveolin expression is critical for caveolar formation [15] and cell-free in vitro studies have illustrated that caveolins interact with a variety of signal transducing molecules, including G ␣ subunits, mitogen-activated protein kinase (MAPK), src tyrosine kinases, endothelial nitric oxide synthase (eNOS), PKC␣, and rhoA [12–14, 16]. A short N-terminal cytoplasmic region of caveolin-1 (residues 82–101) appears essential for these interactions and has thus been termed the scaffolding domain [17]. Analogous regions exist in caveolin-2 and -3. In reconstituted biochemical experiments, the caveolin-1 scaffolding domain regulates the activity of many signaling molecules, including PKC [18]. Furthermore, agonist stimulation has resulted in a redistribution of receptors for contractile agonists (e.g., muscarinic, bradykinin, angiotensin) to caveolin-containing subcellular fractions [19 –21]. It has been proposed, therefore, that caveolins, and by implication caveolae, may act to coordinate the interaction of receptors and a variety of

Receptor-coupled contraction of smooth muscle involves recruitment to the plasma membrane of downstream effector molecules PKC␣ and rhoA but the mechanism of this signal integration is unclear. Caveolins, the principal structural proteins of caveolar plasma membrane invaginations, have been implicated in the organization and regulation of many signal transducing molecules. Thus, using laser scanning confocal immunofluorescent microscopy, we tested the hypothesis that caveolin is involved in smooth muscle signaling by investigating caveolin isoform expression and localization, together with the effect of a peptide inhibitor of caveolin function, in intact differentiated smooth muscle cells. All three main caveolin isoforms were identified in uterine, stomach, and ileal smooth muscles and assumed a predominantly plasma membranous localization in myometrial cells. Cytoplasmic introduction of a peptide corresponding to the caveolin-1 scaffolding domain—an essential region for caveolin interaction with signaling molecules—significantly inhibited agonist-induced translocation of both PKC␣ and rhoA. Translocation was unimpaired by a scrambled peptide and was unaltered in shamtreated cells. The membranous localization of caveolins, and direct inhibition of receptor-coupled PKC␣ and rhoA translocation by the caveolin-1 scaffolding domain, supports the concept that caveolins can regulate the integration of extracellular contractile stimuli and downstream intracellular effectors in smooth muscle. © 2000 Academic Press Key Words: caveolin; PKC; rhoA; smooth muscle.

INTRODUCTION

Efficient receptor-coupled activation of smooth muscle, from the plasma membrane to myofilaments deep in the cell interior, requires the precise and discrete 1 To whom correspondence and reprint requests should be addressed at University Department of Medicine, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, Great Britain. E-mail: [email protected]. Fax: ⫹44 161 274 8433.

0014-4827/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

72

CAVEOLINS AND SMOOTH MUSCLE SIGNALING

downstream signal transducing molecules that localize to the cell membrane following cell stimulation. In fibroblasts, both PKC␣ and rhoA have been found to colocalize with caveolae-enriched membrane domains [22, 23]. Caveolin levels also change with disease states such as cancer and hypercholesterolemia, while caveolin gene defects have recently been associated with muscular dystrophy [24 –26]. As impaired smooth muscle signal transduction may contribute to many disease conditions, such as atherosclerocis, hypertension, asthma, and uterine dystocia, it is essential to clarify the role of caveolins in receptor-coupled signaling. Although caveolin mRNA and protein have been described in cultured smooth muscle [21, 27], the identity, localization, and role of caveolin proteins in intact smooth muscle cells retaining a contractile phenotype has not been determined. Therefore, in order to test the hypothesis that caveolin is involved in smooth muscle signaling, we studied the expression, localization, and effect of a peptide inhibitor of caveolin function in intact differentiated smooth muscle cells. We demonstrate that each caveolin isoform assumes a plasma membranous localization in freshly isolated intact smooth muscle cells. Furthermore, cytoplasmic introduction of the caveolin-1 scaffolding domain to intact single smooth muscle cells results in inhibition of agonist-dependent PKC␣ and rhoA translocation. Our results suggest that caveolins are prime candidates for interaction with key signaling molecules recruited to the smooth muscle plasma membrane following Ca 2⫹sensitizing agonist stimulation. EXPERIMENTAL PROCEDURES Isolation of single smooth muscle cells. Uterine horns were dissected from late-term (19 –21 days; term 22 days) pregnant Sprague– Dawley rats killed by cervical dislocation following CO 2 anesthesia by procedures approved by the institutional ACUC. Approximately 40 strips (4 –5 mm in length, 1 mm in width) of longitudinal smooth muscle were carefully freed from the underlying circular muscle and endometrium and placed in ice-cold Hanks’ buffer (composition: 0.44 mM KH 2PO 4, 0.42 mM Na 2HPO 4, 4.17 mM NaHCO 3, 10 mM Hepes (N-[2-hydroxyethylpiperazine-N⬘-2-ethanesulfonic acid), pH 7.4), 137 mM NaCl, 5.4 mM KCl, 5.5 mM dextrose, 4.9 mM MgCl 2, 1 mM CaCl 2). The muscle strips were washed three times in ice-cold Ca 2⫹free Hanks’ and finely dissected into 1-mm 2 sections. Tissues were then incubated for 25 min, with shaking (34°C), in 7 ml of Ca 2⫹- and Mg 2⫹-free Hanks’ supplemented with 0.2% bovine serum albumin fraction V (BSA; Gibco-BRL), 1.5 mg/ml trypsin inhibitor type II-S (Sigma), 1.9 mg/ml collagenase type II (228 u/mg; Worthington), and 1 mg/ml grade II elastase (3.65 u/mg; Boerhinger Manheim). The supernatant was decanted through a woven mesh (500-␮m diameter; Spectrum) and the tissues were shaken for a further 25 min (34°C) in 7 ml of Ca 2⫹- and Mg 2⫹-free Hanks’ containing 0.2% BSA, 1.25 mg/ml trypsin inhibitor, 0.95 mg/ml collagenase, and 1 mg/ml elastase. The supernatant was filtered through the woven mesh and diluted threefold with ice-cold Ca 2⫹-free Mg 2⫹-containing Hanks’. This filtrate, containing isolated smooth muscle cells, was plated over 10 glass coverslips and left at 4°C for 60 min.

73

Chemical loading procedure for isolated cells. Peptides were loaded into intact smooth muscle cells according to the following procedure. Plated cells were incubated for 30 min (4°C) in solution A of the following composition: 120 mM KCl, 2 mM MgCl 2, 5 mM Na 2ATP, 20 mM Tes (N-tris[hydroxymethyl]methyl-2-aminoethanesulfonic acid), pH 7.0), 10 mM EGTA (ethylene glycol bis(␤-aminoethyl ether)-N,N,N⬘,N⬘-tetraacetic acid). Cells were then placed in solution B at 4°C [as solution A except (EGTA) ⫽ 0.1 mM] containing 100 ␮M relevant peptide (from 10 mM stock dissolved in dimethylsulfoxide containing 25% (w/v) pluronic acid (Molecular Probes)) for 2.0 –2.5 h. This was followed by 30 min incubation in peptide-free solution C [as solution B except (MgCl 2) ⫽ 10 mM; 4°C]. CaCl 2 was then sequentially added over a period of 1 h to a final concentration of 1 mM. Sham-treated cells were exposed to an identical procedure in the absence of any exogenous peptide. Cells were allowed to warm to room temperature (22–24°C) for 30 min in Hanks’ solution before performing experimental maneuvers. Stimulation, fixation, and labeling of smooth muscle cells. As previously reported [1], we employed only cells with a flattened, relaxed appearance (length ⬎150 ␮m; width ⬎8 ␮m) with no semblance of membrane ruffling or blebbing. Cells were then either stimulated for 10 min with the muscarinic agonist carbachol (cch, 100 ␮M) or left untreated for the same time period before fixation in 2.5% paraformaldehyde (20 min). Cells were subsequently permeabilized with 0.15% Triton X-100 (10 min) in Ca 2⫹- and Mg 2⫹-free Hanks’ supplemented with 1% BSA, followed by blocking (1 h) with 1% BSA and 2% goat serum (Vector Laboratories). Cells were then incubated (1 h, room temperature) in the absence of BSA and goat serum with the chosen monoclonal or polyclonal antibody for 1 h at room temperature followed by three separate washes (10 min each) in the absence of antibody. Primary antibodies used in this study were anti-caveolin-1 (monoclonal; 1:500), anti-caveolin-2 (monoclonal; 1:200), anti-caveolin-2 (goat polyclonal; Santa Cruz Biotechnology; 1:250), anti-caveolin-3 (monoclonal, 1:4000), anti-PKC␣ (monoclonal; 1:400; Upstate Biotechnology), and anti-rhoA (monoclonal; Santa Cruz Biotechnology; 1:400). Cells were then probed (30 min, room temperature) with the appropriate secondary antibody labeled with rhodamine-X (goat anti-mouse IgG; Molecular Probes) or Oregon Green (donkey anti-goat IgG; Molecular Probes) followed by three separate washes (10 min each) in the absence of antibody. Coverslips were then mounted on microscope slides with FluorSave reagent (Calbiochem) and analyzed within 48 h. Laser scanning confocal microscopy of labeled cells. Slides containing immunolabeled cells were mounted on a laser scanning (Kr/ Ar) confocal microscope (Leica TC4D) and visualized with a 100X oil immersion objective (NA 1.4). Excitation/emission pathways of 568/ 590 nm (rhodamine-X; Molecular Probes) and 488/515–545 nm (fluoresceine isothyocyanate (FITC)/Oregon Green/Yo-Yo-1; Molecular Probes) were utilized. Laser power settings were optimized to ensure no bleed-through of different fluorochromes in multiple labeling experiments. Up to 12 serial z-section images were digitally collected and saved from a single cell, with each z-section image being an average of eight whole field scans. Offline image restoration and quantification was performed in Adobe Photoshop 3.0.5 and NIH Image 1.6.1. Analysis of cellular distributions of caveolins, PKC␣, rhoA. An assessment of the peripheral:cytosolic distribution of each of the immunolabeled proteins of interest was performed as detailed elsewhere [1]. Briefly, line scans (0.5 ␮m in width) of digitized images were performed across the breadth of nonnuclear regions of central cell sections and the corresponding surface plots of these scans analyzed as follows: the peak pixel intensities over the outer 15% of each cell width were averaged and taken as an estimate of peripheral staining, whereas the peak pixel intensities of the remaining 70% of the stained cell were averaged and taken as an estimate of cytosolic staining. In this manner, a peripheral:cytosolic ratio of 1.0 would indicate a homogeneous distribution of labeled protein. Three to four

74

TAGGART ET AL.

line scans were performed per serial z section and the data meaned. Furthermore, three central confocal z-section images per cell were so analyzed and the data averaged. Preparation of peptides. The caveolin peptides used in this study consisted of the following sequences: (a) scaffolding domain peptide of caveolin-1 (DGIWKASFTTFTVTKYWFYR) termed cav-1 and (b) a scrambled version of the scaffolding domain peptide (WGIDKAFFTTSTVTYKWFRY) termed cav-X. Each was synthesized on an Applied Biosystems Model 431A Peptide Synthesizer using Fmoc (9fluorenylmethoxycarbonyl) as the ␣-amino protecting group. Coupling of amino acids to the nascent peptide was carried out in N-methylpyrrolidone (NMP) using HOBT/HBTU (N-hydroxybenzotriazole/2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate). Fmoc was removed by treatment of the peptide with 15% piperidine in NMP. ␤-Alanine was coupled to the N-terminus of each peptide followed by addition of a 5⫻ excess of fluorescein isothiocyanate (FITC) in N-methylmorpholine/dimethylformamide. FITC was coupled to the free N-terminus of the ␤-alanine overnight at room temperature. Cleavage of the peptide from the resin and deprotection of amino acid side chains was carried out by reaction with 95% trifluoroacetic acid (3 h, room temperature) in water containing phenol, ethandithiol, and thioanisole as scavengers. Peptide was removed from the resin by filtration and precipitated in diethyl ether. Peptides were purified by reversed-phase HPLC and their molecular weights confirmed by mass spectrometry. SDS–PAGE and Western blotting. Intact smooth muscle strips were homogenized in 6 vol of ice-cold buffer (10% (v/v) glycerol, 5 mM EGTA, 140 mM NaCl, 0.5% Nonidet P-40, 0.02% BSA, 5.5 ␮M leupeptin, 5.5 ␮M pepstatin, 20 IU aprotonin, 1.2 mM EDTA (ethylenediaminetetraacetic acid), DTT, glycerophosphate, 50 mM Tris ([Tris(hydroxymethyl)aminomethane]; pH 7.4) and centrifuged (10,000g, 2 min). The protein concentration of the supernatant was determined using Bio-Rad assay reagents before 1:1 dilution with SDS–PAGE sample buffer (to final concentrations of 125 mM Tris, 4% SDS (sodium dodecyl sulfate), 20% glycerol, 0.0005% bromophenol blue, 5% ␤-mercaptoethanol). Protein samples were electrophoresed on 8 or 4 –20% linear gradient polyacrylamide gels (Bio-Rad) and electrophoretically transferred to Immobilon-P membranes (Millipore). Blocked membranes (5% dried milk and 0.05% Tween in phosphate-buffered saline (PBS); 1 h) were incubated overnight (4°C; PBS–Tween with 1% dried milk) with primary antibody. Washed membranes were then incubated with the appropriate horseradish peroxidase-conjugated secondary antibody (Calbiochem; 1:2000) for 2 h at room temperature and membrane-bound antibody was visualized using enhanced chemiluminescence (Pierce). Protein molecular weights were established by densitometric scanning (Bio-Rad GS-700) of Western blots using known prestained molecular weight standards (Bio-Rad Kaleidoscope standards). Electron microscopy of uterine smooth muscle. Longitudinal uterine muscle sections (3– 4 mm 2) were dissected from late-term pregnant rats as described above and fixed with 2% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4), postfixed with 1% osmium tetroxide in cacodylate buffer, and en bloc stained with 2% aqueous uranyl acetate. Fixed tissues were then dehydrated in graded alcohols and embedded in LX112 resin. Thin sections were cut with an Ultracut E ultramicrotome, placed on Formvar- and carbon-coated grids, and analyzed in a JEOL 100CX electron microscope.

RESULTS

Caveolae and Caveolin Isoform Expression Transmission electron microscopy of ultrathin transverse longitudinal uterine smooth muscle sections illustrated the regular appearance of caveolaer invaginations of the plasma membrane (denoted by filled

FIG. 1. Smooth muscle caveolae and caveolin expression. (A) Transmission electron microscopy of ultrathin sections of fixed uterine longitudinal smooth muscle illustrating rows (, Œ) of plasma membranous caveolae. An adjacent noncaveolar region of intense electron-dense staining is labeled (ƒ, ‚). The scale bar equals 0.3 ␮m. (B) Immunoblot analysis using monoclonal antibodies to caveolin-1, -2, and -3 of smooth muscle cell lysates boiled in SDS–PAGE sample buffer and run on 8% polyacrylamide gels illustrating the high specificity of each antibody. (C) Immunoblot analysis of uterine smooth muscle cell lysates treated in SDS–PAGE sample buffer without boiling and run on 4 –20% polyacrylamide gels revealing high-molecular-weight oligomeric complexes formed by each caveolin isoform in vivo.

triangles, Fig. 1A) adjacent to areas of intense electrondense staining (denoted by open triangles, Fig. 1A). Caveolae typically appeared singularly or in rows and were of an approximate width of 75 nm. As illustrated in Fig. 1B, Western blotting with the isoform-specific anti-caveolin monoclonal antibodies indicates that all three caveolin isoforms are expressed in homogenates of rat uterine longitudinal smooth muscle as well as ileal and stomach smooth muscle. Indeed, each isoform has been detected in every smooth muscle tested thus

CAVEOLINS AND SMOOTH MUSCLE SIGNALING

75

FIG. 2. Localization of caveolin-1 in smooth muscle cells. Longitudinal uterine smooth muscle cells were freshly isolated, fixed in paraformaldehyde, and permeabilized as described under Experimental Procedures. Cells were probed with anti-caveolin-1 monoclonal antibody followed by rhodamine-X-labeled anti-mouse secondary antibody. Cells were also incubated with the selective nucleic acid stain YoYo-1 and analyzed by laser scanning confocal microscopy. (A) Twelve confocal serial z sections (0.6 ␮m apart) of a single cell illustrating caveolin-1 (red) and nuclear (green) staining. (B) Enlarged representations of three central cell sections [(iv)–(vi) in A] demonstrating the predominantly plasma membranous staining of caveolin-1. Areas of intense staining at the cell extremities and regions closely apposing the nucleus are marked by asterisks. (C) Surface plot representations of the cell sections in B.

far, including vascular and nonvascular tissues (e.g., portal vein and stomach), phasic and tonic tissues (e.g., ileum and aorta), and different species (e.g., rat and ferret; data not shown). Preparation of uterine smooth samples in SDS–PAGE sample buffer without boiling revealed that caveolins-1, -2, and -3 all formed highmolecular-weight oligomeric complexes (Fig. 1C). Caveolin-1 and -2 formed oligomers of similar molecular weight up to approximately 600 kDa, whereas caveolin-3 oligomeric complexes ⬎200 kDa were not readily observed.

Immunolocalization of Caveolin Isoforms Laser scanning confocal immunofluorescent microscopy of fixed uterine smooth muscle cells labeled with anti-caveolin-1 antibody illustrated the pattern of staining throughout the cell depth, as indicated from the 12 serial z sections of a single cell in Fig. 2A. Viewing consecutive sections through the cell depth illustrates that, in the central cell sections (e.g., [Fig. 2A (iv)–(vi)], caveolin staining is located predominantly close to the plasma membrane. This is more

76

TAGGART ET AL.

FIG. 3. Caveolin-2 localization in single smooth muscle cells. Freshly isolated cells were probed with anti-caveolin-2 polyclonal antibody followed by Oregon Green-labeled anti-goat secondary antibody and examined by laser scanning confocal microscopy. (A) Three confocal serial z sections (0.55 ␮m apart) demonstrating the plasma membranous localization of caveolin-2. Asterisks denote areas of intense staining. (B) Surface plot representations of caveolin-2 staining of cell sections in A.

clearly represented in the magnified representations of central cell sections and the corresponding surface plots (Figs. 2B and 2C). Colabeling of cells with the nucleic acid dye YoYo-1 (Figs. 2A and 2B) showed little evidence of nuclear staining by the anti-caveolin-1 antibody. Single cells were also separately immunolabeled with anti-caveolin-2 or anti-caveolin-3 antibodies. Analysis of central cell sections, including the resultant surface plots, illustrated that caveolin-2 (Fig. 3) and -3 (Fig. 4) were also both predominantly peripherally localized, with little intracellular staining evident. For all three isoforms of caveolin there was noticeably strong staining at the cell extremities and regions closely apposed to the nucleus (asterisks in Figs. 2– 4).

distribution of rhoA in sham-treated chemically loaded single cells at rest was relatively uniform (peripheral: cytosolic ratio of 1.04 ⫾ 0.02; n ⫽ 12; see Fig. 5B) whereas, following carbachol stimulation, a significant relocalization of the protein close to the plasma membrane was observed (peripheral:cytosolic ratio of 2.42 ⫾ 0.23; n ⫽ 11). The PKC␣ and rhoA distributions in these sham-treated cells, at rest and following stimulation, were not significantly different from those observed in intact cells which had not been exposed to the chemical loading procedure [1]. This indicates that cellular integrity and receptor coupling had not been compromised by the experimental intervention of reversible permeabilization.

Receptor-Coupled Redistribution of PKC␣ and rhoA

Effects of Caveolin Peptides on Receptor-Coupled Translocation of PKC␣ and rhoA

The cellular distributions of PKC␣ and rhoA, at rest and following muscarinic stimulation, were studied in isolated immunolabeled uterine cells which had been exposed to the chemical loading procedure (see Experimental Procedures). Line-scan analysis of central z sections of cells immunolabeled with anti-PKC␣, and sham-treated with the chemical loading procedure, indicated that the protein distribution was fairly homogenous in resting cells (e.g., Fig. 5A). The peripheral: cytosolic ratio was 1.12 ⫾ 0.05 (n ⫽ 10). Following stimulation with carbachol, however, a significant redistribution of PKC␣ closer to the plasma membrane became evident such that the peripheral:cytosolic ratio was 2.01 ⫾ 0.08 (n ⫽ 6; P ⬍ 0.01). Similarly, the

Utilizing the fluorescent tags attached to each of the cav-1 and cav-X peptides, the cellular localization of each compound following chemical loading was monitored. Both cav-1 and cav-X peptides were successfully introduced in to the isolated cells and assumed homogeneous distributions throughout the cell cytoplasm following the chemical loading procedure (Fig. 5). The peripheral:cytoplasmic ratios were 1.03 ⫾ 0.01 (n ⫽ 10) for cav-1 peptide and 0.98 ⫾ 0.02 (n ⫽ 15) for cav-X peptide, respectively. Cytoplasmic introduction of the cav-1 peptide was found to significantly inhibit the redistribution of either PKC␣ or rhoA from the cytosol to the cell periphery upon carbachol stimulation (Fig. 5). For example, in cav-1 peptide-containing

CAVEOLINS AND SMOOTH MUSCLE SIGNALING

77

FIG. 4. Caveolin-3 distribution in single smooth muscle cells. Isolated smooth muscle cells were probed with anti-caveolin-3 monoclonal antibody and subsequently with rhodamine-X-labeled anti-mouse secondary antibody and inspected by laser scanning confocal microscopy. (A) Three confocal serial z sections (0.6 ␮m apart) showing the plasma membranous localization of caveolin-3. Areas of intense staining are denoted by asterisks. (B) Surface plot representations of caveolin-3 staining of cell sections in A.

cells stimulated with carbachol the peripheral:cytosolic distribution of PKC␣ was 1.06 ⫾ 0.01 (n ⫽ 5) and that of rhoA was 1.06 ⫾ 0.03 (n ⫽ 5). These ratios were not significantly different from those in resting cells shamtreated with the chemical loading procedure (P ⬎ 0.1). Alternatively, the cav-X peptide did not significantly alter the muscarinic receptor-coupled redistribution of PKC␣ from that observed in sham-treated cells (Fig. 5). Under these conditions, the peripheral:cytosolic ratio of PKC␣ was 1.84 ⫾ 0.06 (n ⫽ 10). DISCUSSION

We have assessed the expression and localization of caveolins in differentiated contractile smooth muscle and find that all three mammalian caveolin isoforms are abundantly expressed. Laser scanning confocal microscopy of immunolabeled single smooth muscle cells indicated that each caveolin isoform was predominantly localized at, or very close to, the plasma membrane. Introduction of the caveolin-1 scaffolding domain peptide to isolated smooth muscle cells resulted in inhibition of the agonist-induced translocation to the plasma membrane of both PKC␣ and rhoA. Uterine smooth muscle preparations contained nu-

merous caveolae appearing periodically in the smooth muscle plasma membrane when viewed by transmission electron microscopy. Caveolae were approximately 70 –90 nm wide, in agreement with previously published morphological data [28]. All three caveolin isoforms were present in longitudinal uterine smooth muscle (thus precluding any contamination from endometrial cells) and, indeed, in each smooth muscle tested from different species in support of our previous data [29]. It is likely, therefore, that all three isoforms are ubiquitously expressed in mammalian smooth muscle. Each caveolin isoform formed high-molecularweight oligomeric complexes in smooth muscle similar to those previously reported in various other cell types [15, 30 –32]. There was often increased intensity of caveolin staining in areas of the plasma membrane close to the nucleus and cell extremities. Caveolae often, although not exclusively, are closely apposed to the peripheral sarcoplasmic reticulum [5, 6, 28] which forms a continuous tubular network with the nuclear envelope [33]. A close geometrical association of peripheral sarcoplasmic reticulum and caveolin-containing surface membrane may indicate a prominent role for caveolae in Ca 2⫹ homeostasis [34 –39].

78

TAGGART ET AL.

We have reported that PKC␣ and rhoA—suggested to be important modulators of smooth muscle contractility— exhibited a stimulus-dependent translocation to the plasma membrane of smooth muscle cells where an interaction with caveolins may take place [1]. Indeed, caveolin-1 in particular has been found to interact with a wide variety of signal transducing molecules (reviewed in [13]), including PKC␣ and rhoA [18, 16]. We introduced the caveolin-1 scaffolding domain peptide into isolated single smooth muscle cells by modification of a previous chemical loading procedure. The peptide assumed a mainly homogeneous distribution, as predicted from the primary sequence and the distributions of PKC␣ and rhoA in resting cells containing the peptide remained unaltered (as evidenced by peripheral:cytosolic ratios close to 1.0). The successful introduction of biologically active peptides into differentiated smooth muscle cells by reversible permeabilization obviates the need for primary or long-term culture, procedures that often result in cellular attainment of a noncontractile phenotype [40] and possibly altered signaling mechanisms. Cellular introduction of the caveolin-1 peptide resulted in a significant inhibition of carbachol-induced relocalization of both PKC␣ and rhoA to the plasma membrane. The chemical loading procedure per se did not interfere with the receptor coupling of the smooth muscle cells: shamtreated cells, and cells chemically loaded with the scaffolding domain scrambled peptide cav-X, exhibited agonist-dependent translocations of PKC␣ and rhoA which were unaltered from those observed in intact cells not exposed to the chemical loading procedure [1]. These results suggest that caveolin-1 directly interacts, via the N-terminal proximal scaffolding domain, with both PKC␣ and rhoA in intact smooth muscle. Consistent with these findings, use of the caveolin-1 scaffolding domain to select peptide ligands from bacteriophage display libraries has identified linear amino acid sequences that are characteristic of many signaling proteins suggested to interact with caveolins [41]. PKC␣ ( 522WAYGVLLY 528) and rhoA ( 34YVPTVFENY 42) both contain such sequences in the catalytic and effector domains, respectively. Indeed, in vitro biochemical assays have indicated that the caveolin-1 scaffolding domain negatively regulates the activity of PKC [18,

FIG. 5. Inhibition of cch-stimulated translocation of PKC␣ and rhoA. Isolated cells were fixed either at rest or following muscarinic stimulation by carbachol (cch, 100 ␮M). Cells were then probed with either anti-PKC␣ or anti-rhoA monoclonal antibodies followed by rhodamine-X-labeled anti-mouse secondary antibodies. FITC-labeled

cav-1 scaffolding domain peptide and FITC-labeled cav-X scrambled peptide were chemically loaded in to the cells as described under Experimental Procedures. (A) PKC␣ distribution at rest and following cch stimulation. (B) rhoA localization at rest and following cch activation. (C) Cav-X peptide (green) assumes a homogenous distribution in the cell and does not prevent the cch-stimulated localization of PKC␣ (red) to the plasma membrane. (D and E) Cav-1 scaffolding domain peptide (green) assumes a homogenous distribution in the cell and inhibits the cch-stimulated redistribution of PKC␣ (D, red) and rhoA (E, red).

79

CAVEOLINS AND SMOOTH MUSCLE SIGNALING

42]. Recent in situ measurements of PKC␣ activity have indicated that membranous distribution of the protein is associated with increased enzymatic activity [43], and thus localization of PKC␣ to caveolae, and binding to caveolins, may be part of a mechanism for coordinating the active state of the enzyme. Also of note, however, is the report that an antibody recognizing possibly multiple isoforms of activated PKC stained both membranous and filamentous structures in stimulated smooth muscle cells [44]. Peptides derived from RACK-1 (receptor for activated C kinase) have also been shown to inhibit PKC translocation to the plasma membrane [45]. However, RACK-1 is found substantially in noncaveolar cell fractions [22, 46], perhaps indicating the existence of more than one protein scaffolding system capable of recruiting PKC␣ to the plasma membrane. A direct association of rhoA with caveolin-1 and/or caveolar fractions has also been reported [16, 23, 47]. This is compatible with the suggestion that the scaffolding domain of caveolin-1 could interact directly with rhoA following cell stimulation. As smooth muscle caveolins reside predominantly at the plasma membrane, then C-terminal prenylation of rhoA following stimulation, and consequent membranous localization, may serve to increase association between the two protein molecules. Since membranous localization of rhoA is essential for receptor-coupled smooth muscle Ca 2⫹ sensitization [3], and the caveolin scaffolding domain has been found to inhibit the activity of many lipid-modified signaling molecules [13, 14], it is possible that caveolins directly regulate the activity of rhoA. It is particularly noteworthy that the sites of ADP ribosylation (N 41) and monoglucosylation (T 37) in rhoA, phenomena that result in an inhibition of agonist-induced rhoA translocation and/or Ca 2⫹ sensitization [4, 48, 49], are both contained within the putative caveolin-binding region mentioned above. Presently, there are conflicting data as to whether treatment with ADPribosylating agents affects rhoA interaction with caveolin/caveolae [16, 23]. Rho-associated kinase (ROK) has been identified as a downstream effector mediating the receptor-coupled smooth muscle Ca 2⫹-sensitizing actions of rhoA [50 – 52]. The ROK␣ isoform also translocates to the cell periphery following muscarinic stimulation in isolated smooth muscle cells [1] and contains putative caveolin scaffolding domain-interacting sequences in its catalytic domain separate from suggested regions of rhoA– ROK interaction ( 151WVVQLFCAF 159 and 164YLYMVMEY 171). However, it remains unknown if there is a direct regulatory interaction of caveolins with ROK. Muscarinic type 3 receptors are present in uterine smooth muscle and are thought to mediate the contractile actions of carbachol stimulation [53]. It has been shown in nonmuscle cells that m3 receptor activation

alters myosin filament arrangement by both PKC- and rhoA-dependent pathways acting independently [54]. This may be in agreement with the suggestion that receptor-coupled activation of PKC and rhoA in smooth muscle results in separate pathways of Ca 2⫹ sensitization of contractility converging downstream by inhibiting myosin light chain phosphatase activity [55]. Smooth muscle caveolar accumulation of contractile receptors [20, 21], together with the caveolin-1 scaffolding domain results presented herein, suggests that caveolins may serve as upstream regulators of these signaling pathways linking receptor occupancy with myofilament activation. Although membranous localizations of PKC and rhoA in smooth muscle are required for the receptor-coupled Ca 2⫹-sensitizing actions of these molecules [2– 4, 56], future experimentation will clarify the contractile effects of caveolin peptide-dependent inhibition of PKC/rhoA translocation. Collectively, these data suggest that caveolins may interact with, and regulate the activity of, PKC␣ and rhoA. It remains to be established if the analogous scaffolding domains of caveolin-2 and -3 act in a similar manner to that of caveolin-1 in smooth muscle. PKC␣ undergoes a stimulus-dependent translocation to caveolin-3-containing subcellular fractions of cardiac muscle cells [22]. Furthermore, while the caveolin-2 scaffolding domain appears to lack the in vitro regulatory activity associated with caveolin-1, the caveolin-3 scaffolding domain exhibits interactions similar to those of caveolin-1 [42]. In summary, we have shown that all three mammalian caveolin isoforms are expressed in differentiated smooth muscle and exhibit a predominantly plasma membranous localization. The scaffolding domain of caveolin-1 inhibits the receptor-coupled recruitment of PKC␣ and rhoA to the cell periphery. It is therefore suggested that caveolae, as a result of the regulatory actions of caveolin(s), may be specialized plasmalemmal regions involved in the integration of extracellular signals and intracellular effectors in intact smooth muscle. This work was supported by the Wellcome Trust (to M.J.T.) and by NIH Grant HL 42293 (to K.G.M.). The assistance with confocal microscopy of Audrey Hutcheon and Donald Pottle (Schepens Eye Research Institute, Boston, MA) is greatly appreciated. We are grateful to Drs. Susan Hagen and Dan Brown of The Morphology Core of the Harvard Digestive Diseases Center (Grant DK-34854) for performing electron microscopic experiments.

REFERENCES 1.

Taggart, M. J., Lee, Y-H., and Morgan, K. G. (1999). Cellular redistribution of PKC␣, rhoA and ROK␣ following smooth muscle agonist stimulation. Exp. Cell Res. 251, 92–101.

2.

Lee, Y-H., Kim, I., Laporte, R., Walsh, M. P., and Morgan, K. G. (1999). Isozyme-specific inhibitors of PKC translocation: Effects

80

TAGGART ET AL. on contractility of single permeabilized vascular muscle cells. J. Physiol. 517, 709 –720.

3.

Gong, M. C., Fujihara, H., Somlyo, A. V., and Somlyo, A. P. (1997). Translocation of rhoA associated with Ca2⫹ sensitization of smooth muscle. J. Biol. Chem. 272, 10704 –10709.

4.

Fujihara, K., Walker, L. A., Gong, M. C., Lemichez, E., Boquet, P., Somlyo, A. V., and Somlyo, A. P. (1997). Inhibition of rhoA translocation and calcium sensitization by in vivo ADP-ribosylation with the chimeric toxin DC3B. Mol. Biol. Cell. 8, 2437– 2447.

21.

22.

23.

5.

Gabella, G. (1984). Structural apparatus for force transmission in smooth muscles. Physiol. Rev. 64, 455– 477.

6.

Somlyo, A. P., and Somlyo, A. P. (1992). Smooth muscle structure and function. In “The Heart and Cardiovascular System,” 2nd ed., pp. 1295–1324.

7.

Palade, G. E. (1953). Fine structure of blood capillaries. J. Appl. Physics. 24, 1424.

8.

Yamade, E. (1955). The fine structure of the gall bladder epithelium of the mouse. J. Biophys. Biochem. Cytol. 1, 445– 458.

9.

Chang, W. J., Rothberg, K. G., Kamen, B. A., and Anderson, R. G. (1992). Lowering the cholesterol content of MA104 cells inhibits receptor-mediated transport of folate. J. Cell Biol. 118, 63– 69.

10.

Lamaze, C., and Schmid, S. L. (1995). The emergence of clathrin-independent pinocytic pathways. Curr. Opin. Cell Biol. 7, 573–580.

11.

Popescu, L., Diculescu, I., Zelck, U., and Ionescu, N. (1974). Ultrastructural distribution of calcium in smooth muscle cells of guinea pig taenia coli. Cell Tissue Res. 154, 357–358.

12.

Parton, R. G. (1995). Caveolae and caveolins. Curr. Opin. Cell Biol. 8, 542–548.

28.

13.

Okamoto, T., Schlegel, A., Scherer, P. E., and Lisanti, M. P. (1998). Caveolins, a family of scaffolding proteins for organizing “preassembled signaling complexes” at the plasma membrane. J. Biol. Chem. 273, 5419 –5422.

29.

14.

Shaul, P. W., and Anderson, R. G. W. (1998). Role of plasmalemmal caveolae in signal transduction. Am. J. Physiol. 275, L843–L851.

15.

Lipardi, C., Mora, R., Colomer, V., Paladino, S., Nitsch, L., Rodriguez-Boulan, E., and Zurzolo, C. (1998). Caveolae transfection results in caveolae formation but not apical sorting of glycosylphosphatidylinositol (GPI)-anchored proteins in epithelial cells. J. Cell Biol. 140, 617– 626.

16.

17.

18.

19.

20.

24.

25.

26.

27.

30.

31.

Gingras, D., Gauthier, F., Lamy, S., Desrosiers, R. R., and Be´liveau, R. (1998). Localization of RhoA GTPase to endothelial caveolae-enriched membrane domains. Biochem. Biophys. Res. Commun. 247, 888 – 893.

32.

Li, S., Couet, J., and Lisanti, M. P. (1996). Src tyrosine kinases, Galpha subunits, and H-Ras share a common membrane-anchored scaffolding protein, caveolin. Caveolin binding negatively regulates the auto-activation of Src tyrosine kinases. J. Biol. Chem. 271, 29182–29190.

33.

Couet, J., Sargiacomo, M., and Lisanti, M. P. (1997). Interaction of a receptor tyrosine kinase, EGF-R, with caveolins. Caveolin binding negatively regulates tyrosine and serine/threonine kinase activities. J. Biol. Chem. 272, 30429 –30438. Feron, O., Smith, T. W., Michel, T., and Kelly, R. A. (1997). Dynamic targeting of the agonist-stimulated m2 muscarinic acetylcholine receptor to caveolae in cardiac myocytes. J. Biol. Chem. 272, 17744 –17748. Ishizaka, N., Griendling, K. K., Lassegue, B., and Alexander, R. W. (1998). Angiotensin II type 1 receptor: relationship with

34.

35. 36.

caveolae and caveolin after initial agonist stimulation. Hypertension 32, 459 – 466. de Weerd, W. F. C., and Leeb-Lundberg, L. M. F. (1997). Bradykinin sequesters B2 bradykinin receptors and the receptorcoupled G ␣ subunits G ␣q and G ␣i in caveolae in DDT1 MF-2 smooth muscle cells. J. Biol. Chem. 272, 17858 –17866. Mineo, C., Ying, Y-S., Chapline, C., Jaken, S., and Anderson, R. G. W. (1998). Targeting of protein kinase C alpha to caveolae. J. Cell Biol. 141, 601– 610. Michaely, P. A., Mineo, C., Ying, Y., and Anderson, R. G. (1999). Polarized distribution of endogenous Rac1 and RhoA at the cell surface. J. Biol. Chem. 274, 21430 –21436. Feron, O., Dessy, C., Moniotte, S., Desager, J., and Balligand, J. (1999). Hypercholesterolemia decreases nitric oxide production by promoting the interaction of caveolin and endothelial nitric oxide synthase. J. Clin. Invest. 103, 897–905. Racine, C., Belanger, M., Hirabayashi, H., Boucher, M., Chakir, J., and Couet, J. (1999). Reduction of caveolin 1 gene expression in lung carcinoma cell lines. Biochem. Biophys. Res. Commun. 255, 580 –586. Minetti, C., Sotgia, F., Bruno, C., Scartezzini, P., Broda, P., Bado, M., Masetti, E., Mazzocco, M., Egeo, A., Donati, M. A., Volonte, D., Galbiati, F., Cordone, G., Bricarelli, F. D., Lisanti, M. P., and Zara, F. (1998). Mutations in the caveolin-3 gene cause autosomal dominant limb-girdle muscular dystrophy. Nat. Genet. 18, 365–368. Peterson, T. E., Kleppe, L. S., Caplice, N. M., Pan, S., Mueske, C. S., and Simari, R. D. (1999). The regulation of caveolin expression and localization by serum and heparin in vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 265, 722–727. Broderick, R., and Broderick, K. A. (1990). “Uterine Function: Molecular and Cellular Aspects,” 1st ed., pp. 1–70, Plenum, New York. Taggart, M. J., Leavis, P., Feron, O., and Morgan, K. G. (1998). Distribution and function of caveolins in differentiated smooth muscle. Mol. Biol. Cell. 9, 98a. Monier, S., Parton, R. G., Vogel, F., Behlke, J., Henske, A., and Kurzchalia, T. V. (1995). VIP21-caveolin, a membrane protein constituent of the caveolar coat, oligomerizes in vivo and in vitro. Mol. Biol. Cell. 6, 911–927. Li, S., Galbiati, F., Volonte, D., Sargiacomo, M., Engelman, J. A., Das, K., Scherer, P. E., and Lisanti, M. P. (1998). Mutational analysis of caveolin-induced vesicle formation. Expression of caveolin-1 recruits caveolin-2 to caveolae membranes. FEBS Lett. 434, 127–134. Das, K., Lewis, R. Y., Scherer, P. E., and Lisanti, M. P. (1999). The membrane-spanning domains of caveolins-1 and -2 mediate the formation of caveolin hetero-oligomers. Implications for the assembly of caveolae membranes in vivo. J. Biol. Chem. 274, 18721–18728. Nixon, G. F., Migneri, G. A., and Somlyo, A. V. (1994). Immunogold localization of inositol 1,4,5-trisphosphate receptors and characterization of ultrastructural features of the sarcoplasmic reticulum in phasic and tonic smooth muscle. J. Muscle Res. Cell Motil. 15, 682–700. Taggart, M. J., and Wray, S. (1997). Agonist mobilization of sarcoplasmic reticular calcium in smooth muscle: Functional coupling to the Na/Ca exchanger? Cell Calcium 22, 333–341. Fujimoto, T. (1993). Calcium pump of the plasma membrane is localized in caveolae. J. Cell Biol. 120, 1147–1157. Moore, E. D. W., Etter, E. F., Phillipson, K. D., Carrington, W. A., Fogarty, K. E., Lifshitz, L. M., and Fay, F. S. (1993). Coupling of the Na⫹/Ca2⫹ exchanger, Na⫹/K⫹ pump and

CAVEOLINS AND SMOOTH MUSCLE SIGNALING

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

sarcoplasmic reticulum in smooth muscle. Nature 365, 657– 660. Fujimoto, T., Nakade, S., Miyawaki, A., Mikoshiba, K., and Ogawa, K. (1992). Localization of inositol 1,4,5-trisphosphate receptor-like protein in plasmalemmal caveolae. J. Cell Biol. 119, 1507–1513. Gollasch, M., Wellman, G. C., Knot, H. J., Jagger, J. H., Damon, D. H., Bonev, A. D., and Nelson, M. T. (1998). Ontogeny of local sarcoplasmic reticulum Ca2⫹ signals in cerebral arteries: Ca2⫹ sparks as elementary physiological events. Circ. Res. 83, 1104 –1114. Isshiki, M., Ando, J., Korenaga, R., Kogo, H., Fujimoto, T., Fujita, T., and Kamiya, A. (1998). Endothelial Ca2⫹ waves preferentially originate at specific loci in caveolin-rich cell edges. Proc. Natl. Acad. Sci. USA 95, 5009 –5014. Blank, S. R., Thompson, M. M., and Owens, G. K. (1988). Transforming growth factor-beta-induced growth inhibition and cellular hypertrophy in cultured vascular smooth muscle cells. J. Cell Biol. 107, 299 –306. Couet, J., Li, S., Okamoto, T., Ikezu, T., and Lisanti, M. P. (1997). Identification of peptide and protein ligands for the caveolin-scaffolding domain. Implications for the interaction of caveolin with caveolae-associated proteins. J. Biol. Chem. 272, 6525– 6533. Oka, N., Yamamoto, M., Schwencke, C., Kawabe, J., Ebina, T., Ohno, S., Couet, J., Lisanti, M. P., and Ishikawa, Y. (1997). Caveolin interaction with protein kinase C. Isoenzyme-dependent regulation of kinase activity by the caveolin scaffolding domain peptide. J. Biol. Chem. 272, 33416 –33421. Ng, T., Squire, A., Hansra, G., Bornancin, F., Prevostel, C., Hanby, A., Harris, W., Barnes, D., Schmidt, S., Mellor, H., Bastiaens, P. I. H., and Parher, P. J. (1999). Imaging protein kinase C␣ activation in cells. Science 283, 2085–2089. Meininger, G. A., Moore, E. D. W., Schmidt, D. J., Lifshitz, L. M., and Fay, F. S. (1999). Distribution of active protein kinase C in smooth muscle. Biophys. J. 77, 973–984. Yedovitzky, M., Mochly-Rosen, D., Johnson, J. A., Gray, M. O., Ron, D., Abramovitch, E., Cerasi, E., and Nesher, R. (1997). Translocation inhibitors define specificity of protein kinase C isoenzymes in pancreatic beta-cells. J. Biol. Chem. 272, 1417– 1420. Rybin, V. O., Xu, X., and Steinberg, S. F. (1999). Activated protein kinase C isoforms target to cardiomyocyte caveolae: Stimulation of local protein phosphorylation. Circ. Res. 84, 980 –988. Teixeira, A., Chaverot, N., Schroder, C., Strosberg, A. D., Couraud, P. O., and Cazaubon, S. (1999). Requirement of caveo-

Received January 10, 2000 Revised version received March 15, 2000

81

lae microdomains in extracellular signal-regulated kinase and focal adhesion kinase activation induced by endothelin-1 in primary astrocytes. J. Neurochem. 72, 120 –128. 48.

Otto, B., Steusloff, A., Just, I., Aktories, K., and Pfitzer, G. (1996). Role of Rho proteins in carbachol-induced contractions in intact and permeabilised guinea-pig intestinal smooth muscle. J. Physiol. 496, 317–329.

49.

Lucius, C., Arner, A., Steusloff, A., Troschka, M., Hofmann, F., Aktories, K., and Pfitzer, G. Clostridium difficile toxin B inhibits carbachol-induced force and myosin light chain phosphorylation in guinea-pig smooth muscle: role of Rho proteins. J. Physiol. 506.2, 83–93.

50.

Matsui, T., Amano, M., Yamamoto, T., Chihara, K., Nakafuku, M., Ito, M., Nakano, T., Okawa, K., Iwamatsu, A., and Kaibuchi, K. (1995). Rho-associated kinase, a novel serine/threonine kinase, as a putative target for the small GTP binding protein rho. EMBO J. 15, 2208 –2216.

51.

Kimura, K., Ito, M., Amano, M., Chihara, K., Fukata, Y., Nakafuku, M., Yamamori, B., Feng, J., Nakano, T., Okawa, K., Iwamatsu, A., and Kaibuchi, K. (1996). Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science 273, 245–248.

52.

Ueheta, M., Ishizaki, T., Satoh, H., Ono, T., Kawahara, T., Morishita, T., Tamakawa, H., Yamagami, K., Inui, J., Maekawa, M., and Narumiya, S. (1997). Calcium sensitization of smooth muscle mediated by a rho-associated protein kinase in hypertension. Nature 389, 990 –994.

53.

Varol, F. G., Hadjiconstantantinou, M., Zuspan, F. P., and Neff, N. H. (1989). Pharmacological characterization of the muscarinic receptors mediating phosphoinositide hydrolysis in rat myometrium. J. Pharmacol. Exp. Ther. 249, 11–15.

54.

Strassheim, D., May, L. G., Varker, K. A., Puhl, H. L., Phelps, S. H., Porter, R. A., Aronstam, R. S., Noti, J. D., and Williams, C. L. (1999). M3 muscarinic acetylcholine receptors regulate cytoplasmic myosin by a process involving RhoA and requiring conventional protein kinase C isoforms. J. Biol. Chem. 274, 18675–18655.

55.

Jensen, P. E., Gong, M. C., Somlyo, A. V., and Somlyo, A. P. (1996). Separate upstream and convergent downstream pathways of G-protein- and phorbol ester-mediated Ca 2⫹-sensitisation of myosin light chain phosphorylation in smooth muscle. Biochem. J. 318, 469 – 475.

56.

Ru¨egg, J. C. (1999). Smooth muscle: PKC-induced Ca 2⫹-sensitisation by myosin phosphatase inhibition. J. Physiol. 520.1, 3.