Expression of Caveolin-3 in Skeletal, Cardiac, and SmoothMuscle CellsCAVEOLIN-3 IS A COMPONENT OF THE SARCOLEMMA AND CO-FRACTIONATES WITH DYSTROPHIN ANDDYSTROPHIN-ASSOCIATED GLYCOPROTEINS*

(Received for publication, February 28, 1996, and in revised form, April 5, 1996)

Kenneth S. Song‡, Philipp E. Scherer‡§, ZhaoLan Tang‡, Takashi Okamoto¶i, Shengwen Li‡**,Mark Chafel‡, Caryn Chu‡‡, D. Stave Kohtz‡‡, and Michael P. Lisanti‡§§

From the ‡Whitehead Institute for Biomedical Research, Cambridge, Massachusetts 02142-1479, ¶Shriners Hospitals forCrippled Children, Massachusetts General Hospital, Department of Anesthesia, Harvard Medical School, Boston,Massachusetts 02114, and ‡‡The Mount Sinai School of Medicine, Department of Pathology, New York, New York 10029

Caveolae are microdomains of the plasma membranethat have been implicated in signal transduction. Caveo-lin, a 21–24-kDa integral membrane protein, is a princi-pal component of the caveolae membrane. Recently, weand others have identified a family of caveolin-relatedproteins; caveolin has been retermed caveolin-1. Caveo-lin-3 is most closely related to caveolin-1, but caveolin-3mRNA is expressed only in muscle tissue types. Here, weexamine (i) the expression of caveolin-3 protein in mus-cle tissue types and (ii) its localization within skeletalmuscle fibers by immunofluorescence microscopy andsubcellular fractionation. For this purpose, we gener-ated a novel monoclonal antibody (mAb) probe that rec-ognizes the unique N-terminal region of caveolin-3, butnot other members of the caveolin gene family. A surveyof tissues and muscle cell types by Western blot analysisreveals that the caveolin-3 protein is selectively ex-pressed only in heart and skeletal muscle tissues, car-diac myocytes, and smooth muscle cells. Immunolocal-ization of caveolin-3 in skeletal muscle fibersdemonstrates that caveolin-3 is localized to the sarco-lemma (muscle cell plasma membrane) and coincideswith the distribution of another muscle-specific plasmamembrane marker protein, dystrophin. In addition,caveolin-3 protein expression is dramatically inducedduring the differentiation of C2C12 skeletal myoblastsin culture. Using differentiated C2C12 skeletal myo-blasts as a model system, we observe that caveolin-3co-fractionates with cytoplasmic signaling molecules(G-proteins and Src-like kinases) and members of thedystrophin complex (dystrophin, a-sarcoglycan, andb-dystroglycan), but is clearly separated from the bulkof cellular proteins. Caveolin-3 co-immunoprecipitateswith antibodies directed against dystrophin, suggestingthat they are physically associated as a discrete com-

plex. These results are consistent with previous immu-noelectron microscopic studies demonstrating that dys-trophin is localized to plasma membrane caveolae insmooth muscle cells.

Caveolae are small bulb-shaped invaginations located at ornear the cell surface (1–5). They represent a microdomain orsubcompartment of the plasma membrane (3–5). Recent bio-chemical and morphological studies have implicated caveolaein a subset of transmembrane signaling events, including G-protein-coupled signaling (reviewed in Refs. 4 and 5).Caveolin, a 21–24-kDa integral membrane protein, is an

important structural and regulatory component of caveolaemembranes (6–11). Using both Triton-based methods (12–17)and detergent-free methods (18, 19), caveolin co-purifies with anumber of lipid-modified cytoplasmic signaling molecules, in-cluding G-proteins (a and bg subunits), protein kinase Ca,Src-family tyrosine kinases, and Ras proteins. Based on theseobservations, we have proposed the “caveolae signaling hypoth-esis” which states that compartmentalization of certain cyto-plasmic signaling molecules within caveolae membranes couldallow rapid and efficient coupling of activated receptors to morethan one effector system (4, 12).Caveolin may act as a scaffolding protein within caveolae

membranes (20). Caveolin forms high molecular mass homo-oligomers (;14–16 monomers per oligomer) (20–22), and thesecaveolin homo-oligomers have the capacity to bind cholesterol(22, 23) and self-associate into larger structures that resemblecaveolae (20). Within these structures, individual caveolin ho-mo-oligomers (;4–6 nm) apparently pack side by side to form25–50-nm structures that are the same size as caveolae (20).This would provide a “caveolin platform or scaffold” for therecruitment of caveolin-interacting proteins, such as Ga sub-units and Ras proteins, to caveolin-rich areas of the plasmamembrane (20). In accordance with this idea, recombinantcaveolin is sufficient to recruit nonlipid modified forms of Ga

subunits and Ras onto membranes both in vitro and in vivo (19,22). These structural properties of caveolin are also consistentwith a role for caveolin in orchestrating the formation of caveo-lae microdomains.Additional evidence suggests that expression of caveolin is

sufficient to form caveolae membranes. Caveolin mRNA andprotein expression levels are highest in cell types that containnumerous caveolae, i.e. adipocytes, endothelial cells, smoothmuscle cells, and fibroblasts (9, 13). Caveolin expression levelsdirectly correlate with the morphological appearance of caveo-lae: (i) caveolin and caveolae are both induced ;10–25-fold

* This work was supported in part by National Institutes of HealthFIRST Award GM-50443 (to M. P. L.), a grant from the Elsa U. PardeeFoundation (to M. P. L.), and a grant from the W. M. Keck Foundationto the Whitehead Fellows program (to M. P. L.). The costs of publicationof this article were defrayed in part by the payment of page charges.This article must therefore be hereby marked “advertisement” in ac-cordance with 18 U.S.C. Section 1734 solely to indicate this fact.§ Funded by a Swiss National Science Foundation Fellowship.i Recipient of Fellowships from the Byotai-Taisha Foundation and

the Mochida Memorial Foundation.** Recipient of National Institutes of Health Postdoctoral Fellowship

CA-71326.§§ To whom correspondence should be addressed: Whitehead Insti-

tute for Biomedical Research, 9 Cambridge Center, Cambridge, MA02142-1479. Tel.: 617-258-5225; Fax: 617-258-9872; E-mail: lisanti@wi.mit.edu.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 271, No. 25, Issue of June 21, pp. 15160–15165, 1996© 1996 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

15160

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from

during the differentiation of 3T3-L1 fibroblasts to the adipoctyeform (24–26), and (ii) caveolin levels are dramatically reducedand caveolae are morphologically absent in NIH 3T3 cellstransformed by various activated oncogenes (v-abl, activatedras, and others) (27). Furthermore, recombinant expression ofcaveolin in caveolin-negative cell lines results in the correcttargeting of caveolin to caveolae-enriched membrane fractions(28) and drives the de novo formation of caveolae (29). Theseresults indicate that caveolin represents an important struc-tural protein for directing the formation of caveolaemembranes.However, there are certain cell lines which morphologically

contain caveolae, but fail to express caveolin (12). This findinghas suggested that other caveolin-related proteins may existthat are immunologically distinct from caveolin. In support ofthis notion, two novel caveolin-related proteins have recentlybeen identified and cloned. These proteins, termed caveolin-2and caveolin-3, are the products of separate caveolin genes (26,30, 31). Thus, caveolin (retermed caveolin-1) is the first mem-ber of a multigene family (26).Caveolin-3 is most closely related to caveolin-1 based on

protein sequence homology; caveolin-1 and caveolin-3 are;65% identical and ;85% similar. (See Tang et al. (30) for analignment.) However, caveolin-3 mRNA is expressed predomi-nantly in muscle tissue types (skeletal muscle, diaphragm, andheart) (30). Identification of a muscle-specific member of thecaveolin gene family may have implications for understandingthe role of caveolin in different muscle cell types (smooth,cardiac, and skeletal), as previous morphological studies havedemonstrated that caveolae are abundant in these cells. Thisappears relevant to the pathogenesis of Duchenne’s musculardystrophy. More specifically, dystrophin has been localized toplasma membrane caveolae in smooth muscle cells using im-munoelectron microscopy techniques (32), and skeletal musclecaveolae undergo characteristic changes in their size and dis-tribution in patients with Duchenne’s muscular dystrophy, butnot in other forms of neuronally based muscular dystrophiesexamined (33). This indicates that muscle cell caveolae mayplay an important role in muscle membrane biology.Here, we (i) characterize the protein expression of caveolin-3

in skeletal, cardiac, and smooth muscle cells and (ii) report itsco-localization, co-fractionation, and co-immunoprecipitationwith dystrophin, a known muscle-specific caveolar markerprotein.

EXPERIMENTAL PROCEDURES

Materials—The cDNAs for caveolin-1, caveolin-2, and caveolin-3were as we described previously (12, 26, 30). Antibodies and theirsources were as follows: anti-caveolin-1 IgG (mAb1 2297; gift of Dr. JohnR. Glenney, Transduction Labs); anti-myc epitope IgG (mAb 9E10;Santa Cruz Biotech); anti-dystrophin (mAb DYS2 (Dy8/6C5); Novocas-tra Laboratories); anti-dystrophin (rabbit polyclonal; gift of Drs. LouisKunkel and Elizabeth McNally, Children’s Hospital, Boston, MA); anti-b-dystroglycan (mAb 43DAG1/8D5; Novocastra Laboratories); anti-a-sarcoglycan (Adhalin; mAb Ad1/20A6; Novocastra Laboratories); anti-Gi,2a (Dupont NEN); anti-Gb (Transduction Labs); anti-Src (OncogeneSciences); anti-Lyn (Santa Cruz Biotech); anti-glutathione S-transfer-ase (Santa Cruz Biotech). A rabbit polyclonal antibody directed againstthe C-terminal 44 amino acids of caveolin-1 (residues 135–178) was ascharacterized previously (28). This antibody specifically recognizes botha- and b-isoforms of caveolin-1, but does not recognize caveolin-2 orcaveolin-3. A variety of other reagents were purchased commercially:fetal bovine serum (JRH Biosciences); prestained protein markers (LifeTechnologies, Inc.); Slow-Fade anti-fade reagent (Molecular Probes,Eugene, OR).Hybridoma Production—A monoclonal antibody to caveolin-3 was

generated by multiple immunizations of Balb/c female mice with thesynthetic peptide TEEHTDLEARIIKDIHCKEIDL. This peptide corre-sponds to amino acids 3–24 of the rat caveolin-3 protein sequence. Miceshowing the highest titer of anti-caveolin-3 immunoreactivity wereused to create fusions with myeloma cells using standard protocols (34).Positive hybridomas were cloned twice by limiting dilution and injectedinto mice to produce ascites fluid. IgGs were purified by affinity chro-matography on protein A-Sepharose. These antibodies were producedin collaboration with Drs. Roberto Campos-Gonzalez and John R.Glenney, Jr. (Transduction Laboratories, Lexington, KY).Transient Expression of Caveolin Genes in COS-7 Cells—Constructs

encoding C-terminally myc-epitope tagged forms of caveolin-1, caveo-lin-2 or caveolin-3, were as described previously (26, 28, 30). Theseconstructs (;5–10 mg) were transiently transfected into COS-7-cells bythe DEAE-dextran method (35). Forty-eight hours post-transfection,cells were scraped into lysis buffer (20 mM Tris, pH 8.0, 150 mM NaCl,1% Triton X-100). Recombinant expression was analyzed by SDS-PAGE(15% acrylamide) followed by Western blotting. Epitope-tagged formsof caveolin-1, caveolin-2, and caveolin-3 were detected using themonoclonal antibody, 9E10, that recognizes the myc-epitope(EQKLISEEDLN).Tissue Western—Approximately 200 mg of various mouse tissues

were lysed in immunoprecipitation buffer and homogenized on ice witha Polytron tissue grinder, as described elsewhere (25). Equal amounts(100 mg protein) were loaded on an SDS-PAGE gel (12% acrylamide).After transfer to nitrocellulose, the blot was probed with antibodiesdirected against caveolin-1 and caveolin-3.Immunofluorescence—Mouse skeletal muscle (psoas) was excised

and incubated for 30 min in 0.1% Triton X-100 in EGTA-Ringer’ssolution (100 mM NaCl, 2 mM KCl, 2 mM MgCl2, 6 mM potassiumphosphate, 1 mM EGTA, 0.1% glucose). The tissue was further incu-bated in this solution with the addition of 1 mM ATP. The musclesamples were then fixed in 4% paraformaldehyde/PBS, rinsed in PBS,and cryoprotected in 0.6 M sucrose, PBS prior to freezing them in anisopentane-liquid nitrogen bath. Frozen sections (5 mm thick) wereextracted in ice-cold acetone for 2 min and rinsed in PBS prior toantibody staining. To prevent nonspecific binding, the sections werepreincubated in 3% BSA in PBS for 1 h at 37 °C. Mouse monoclonalcaveolin-3 and rabbit polyclonal dystrophin antibodies were then ap-plied to the sections and incubated in a similar fashion. Sections werethen washed with three changes of PBS, 0.1% Triton X-100. Binding ofthe primary antibodies was detected by incubating the sections for 1 hat 37 °C, with fluorescein-conjugated goat anti-mouse IgG and lissa-mine rhodamine-conjugated donkey anti-rabbit IgG. Sections were fi-nally washed as above and mounted in 1 mg/ml p-phenylenediamine in90% glycerol. All samples were examined and photographed using aZeiss Axioskop microscope with 403 and 633 objectives.Cell Culture—C2C12–3 cells (36) were derived from a single colony of

C2C12 cells (37) cultured at clonal density and display a more stablephenotype than the parental cell line. C2C12-3 myoblasts were culturedas described elsewhere (36). Briefly, proliferating C2C12-3 cells werecultured in high mitogen medium (Dulbecco’s modified Eagle’s mediumcontaining 15% fetal bovine serum and 1% chicken embryo extract) andinduced to differentiate at confluence in low mitogen medium (Dulbec-co’s modified Eagle’s medium containing 3% horse serum). Overt dif-ferentiation was indicated by the assembly of multinucleated syncytia,which commenced 36–48 h after the cells were switched to low mitogenmedia.Cell Fractionation—Differentiated C2C12 cells grown to confluence

in 150-mm dishes were used to prepare caveolin-enriched membranefractions, essentially as we have described previously for other cell lines(12, 13, 25, 28, 38, 39). However, two specific modifications were intro-duced to allow the purification of caveolin-rich domains without the useof detergent (19). Triton X-100 was replaced with sodium carbonatebuffer and a sonication step was introduced to more finely disruptcellular membranes (19). After two washes with ice-cold PBS, differen-tiated C2C12 cells (two confluent 150-mm dishes) were scraped into 2ml of 500 mM sodium carbonate, pH 11.0. Homogenization was carriedout sequentially in the following order using: (i) a loose-fitting Douncehomogenizer (10 strokes); (ii) a Polytron tissue grinder (three 10-sbursts; Kinematica GmbH, Brinkmann Instruments, Westbury, NY);and (iii) a sonicator (three 20-s bursts; Branson Sonifier 250, BransonUltrasonic Corp., Danbury, CT). The homogenate was then adjusted to45% sucrose by addition of 2 ml of 90% sucrose prepared in MBS (25 mM

Mes, pH 6.5, 0.15 M NaCl) and placed at the bottom of an ultracentri-fuge tube. A 5–35% discontinuous sucrose gradient was formed above (4ml of 5% sucrose/4 ml of 35% sucrose; both in MBS containing 250 mM

sodium carbonate) and centrifuged at 39,000 rpm for 16–20 h in an

1 The abbreviations used are: mAb, monoclonal antibody; PAGE,polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline;Mes, 4-morpholineethanesulfonic acid.

Caveolin-3 and the Dystrophin Complex 15161

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from

SW41 rotor (Beckman Instruments, Palo Alto, CA). A light-scatteringband confined to the 5–35% sucrose interface was observed that con-tained caveolin-3, but excluded most other cellular proteins.Immunoblotting of Gradient Fractions—From the top of each gradi-

ent, 1-ml gradient fractions were collected to yield a total of 13 fractions.As shown previously, caveolin-1 migrates mainly in fractions 5 and 6 ofthese sucrose density gradients (13, 19, 25, 28, 38, 39). Gradient frac-tions were separated by SDS-PAGE (15% acrylamide) and transferredto nitrocellulose. After transfer, nitrocellulose sheets were stained withPonceau S to visualize protein bands and subjected to immunoblotting.For immunoblotting, incubation conditions were as described by themanufacturer (Amersham Corp.), except we supplemented our blockingsolution with both 1% BSA and 1% non-fat dry milk (Carnation).Immunoprecipitation—Immunoprecipitations were carried out using

protein A-Sepharose CL-4B (Pharmacia Biotech Inc.) as described pre-viously (40). Briefly, differentiated C2C12 cells were lysed in a buffercontaining 10 mM Tris, pH 8.0, 0.15 M NaCl, 60 mM octyl glucoside andsubjected to immunoprecipitation with specific rabbit polyclonal IgGdirected against dystrophin, glutathione S-transferase (an irrelevantrabbit polyclonal IgG), or protein A-Sepharose alone. After extensivewashing, samples were separated by SDS-PAGE (15% acrylamide) andtransferred to nitrocellulose. Blots were then probed with a mAb di-rected against caveolin-3.Cardiac Myocytes and Smooth Muscle Cells—Rat cardiac myocytes

were the generous gift of Drs. Douglas Sawyer, Thomas W. Smith, andRalph A. Kelley (Brigham and Women’s Hospital, Division of Cardiol-ogy, Harvard Medical School, Boston, MA) (41). Rat aortic smoothmuscle cells were the generous gift of Dr. Lee Graves (University ofNorth Carolina, Chapel Hill, NC) and were isolated and characterizedas described previously (42).

RESULTS

Characterization of a mAb Probe Specific for Caveolin-3—Caveolin-1, -2, and -3 are distinct gene products with differentmolecular masses, all in the range of ;18–24 kDa. Currently,there are no available antibody probes that selectively recog-nize caveolin-3. Comparison of the protein sequences of caveo-lin-1, caveolin-2, and caveolin-3 reveals that these proteinsdiffer most significantly within their extreme N termini. (SeeTang et al. (30) for an alignment.) Thus, a peptide derived fromthe unique N-terminal sequence of caveolin-3 was used to gen-erate a caveolin-3-specific monoclonal antibody probe.Fig. 1 (right panel) demonstrates the specificity of this novel

mAb probe; it selectively recognizes caveolin-3, but does notrecognize caveolin-1 or caveolin-2. In addition, mAb 2297 orig-inally generated against caveolin-1 recognizes only caveolin-1

(Fig. 1, middle panel). Thus, these two selective mAb probescan be used to study the function and differential expression ofdistinct caveolin gene family members. Antibodies for caveo-lin-2 are not yet available; however, caveolin-1 and caveolin-2have the same tissue distribution as revealed by Northernanalysis (26).Muscle-specific Expression of the Caveolin-3 Protein—Fig. 2

(upper panel) shows the tissue distribution of the caveolin-3protein. Caveolin-3 expression is detected only in muscle tissuetypes, i.e. heart and skeletal muscle. The tissue distribution ofcaveolin-1 is shown for comparison (Fig. 2, lower panel). Instriking contrast, caveolin-1 is most abundant in adipose tis-sue; little or no caveolin-1 protein is expressed in muscle tis-sues. The low levels of caveolin-1 expression detected in skel-etal muscle tissue derives from endothelial cells which line theblood vessels, but not skeletal muscle cells themselves (30).Thus, it appears that tissue-specific expression of caveolin-3may be important for the functioning of muscle cells. Thismuscle-specific expression of caveolin-3 protein is in accord-ance with previous Northern analysis indicating that caveo-lin-3 mRNA is expressed only in muscle tissue types (30, 31).Immunolocalization of Caveolin-3 to the Sarcolemma of Skel-

etal Muscle Fibers—Fig. 3 shows the localization of caveolin-3in skeletal muscle fibers. Caveolin-3 immunostaining is con-fined to the plasma membrane (i.e. the sarcolemma) of thesemuscle cells. To confirm the nature of this localization, weperformed double-labeling studies with antibodies to anothermuscle-specific plasma membrane protein, dystrophin. Our re-sults indicate that caveolin-3 and dystrophin co-localize to thesarcolemma (Fig. 3A). In addition, closer inspection of tangen-

FIG. 1. Characterization of a mAb probe specific for caveo-lin-3. Caveolin-1, -2, and -3 are distinct gene products with differentmolecular masses, all in the range of ;18–24 kDa. C-terminally myc-tagged forms of caveolin-1, caveolin-2, and caveolin-3 were transientlyexpressed in COS-7 cells. Lysates were generated and used to deter-mine the specificity of caveolin antibody probes by immunoblotting. Asa control for equal loading, immunoblotting was first performed withmAb 9E10 that recognizes the myc-epitope; this antibody reveals allthree myc-tagged caveolin gene products (left panel). Note that mAb2297 recognizes only caveolin-1 (middle panel), while a mAb generatedagainst an N-terminal peptide that is unique to caveolin-3 recognizesonly caveolin-3 (right panel). FIG. 2. Western blot analysis of the tissue distribution of

caveolin-3. Extracts of mouse tissues were prepared as described un-der “Experimental Procedures.” After SDS-PAGE and transfer to nitro-cellulose, blots were probed with anti-caveolin IgG. Upper panel, im-munoblot analysis with a mAb directed against caveolin-3. Note thatcaveolin-3 is selectively expressed in heart and skeletal muscle tissues.Lower panel, immunoblot analysis with both caveolin-1 (mAb 2297) andcaveolin-3 specific antibody probes. The distribution of caveolin-1 isshown for comparison. Caveolin-1 is most abundantly expressed in fat,pancreas, and lung tissues. Note that little or no caveolin-1 is expressedin heart or skeletal muscle tissues. The low levels of caveolin-1 expres-sion detected in skeletal muscle tissue derive from endothelial cellswhich line the blood vessels, but not skeletal muscle cells themselves(30).

Caveolin-3 and the Dystrophin Complex15162

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from

tial sections revealed punctate immunostaining for both caveo-lin-3 and dystrophin that closely coincided, suggesting thatthey co-localize to the same microdomains of the plasma mem-brane (Fig. 3B). In support of these findings, recent immuno-electron microscopy studies have directly localized dystrophinto caveolae in smooth muscle cells (32).Up-regulation of Caveolin-3 Expression during the Differen-

tiation of C2C12 Skeletal Muscle Myoblasts: Co-fractionationwith Dystrophin and Dystrophin-associated Glycoproteins—Cultured C2C12 cells offer a convenient system to study skel-

etal myoblast differentiation. These cells can be induced todifferentiate from myoblasts into myotubes bearing an embry-onic phenotype in low mitogen medium over a period of 2 days(36, 37). Fig. 4 shows that caveolin-3 protein was undetectablein precursor myoblasts and strongly induced during myoblastdifferentiation. In contrast, no caveolin-1 expression was de-tected in either precursor myoblasts or differentiated myotubes(not shown). These results are consistent with the selectiveexpression of caveolin-3 in skeletal muscle and other muscletissues (Figs. 2 and 3) and suggest that caveolin-3 may functionin muscle from the earliest stages of its development.We next used differentiated C2C12 cells to examine the

subcellular distribution of caveolin-3 and its potential associa-tion with other proteins. To separate C2C12 membranes en-riched in caveolin-3 from the bulk of cellular membranes andcytosolic proteins, an established equilibrium sucrose densitygradient system was employed (12, 13, 25, 26, 28, 30, 38, 39). Inthis detergent-free fractionation scheme (19), immunoblottingwith anti-caveolin-3 IgG can be used to track the position ofcaveolae-derived membranes within these bottom-loaded su-crose gradients. Fig. 5 illustrates that in this fractionationscheme ;90–95% of caveolin-3 (fractions 5 and 6) is separatedfrom the bulk of cellular proteins. In addition, caveolin-3 co-fractionated with known caveolin-1-associated proteins, suchas G protein subunits and Src-family tyrosine kinases (Fig. 5,lower panels). This is in accordance with the idea that inmuscle tissues caveolin-3 might subsume the functional role ofcaveolin-1 (30).As previous immunoelectron microscopy studies have dem-

onstrated that dystrophin is selectively localized to plasmamembrane caveolae in smooth muscle cells (32), we also exam-ined the distribution of dystrophin and dystrophin-associatedglycoproteins (a-sarcoglycan/b-dystroglycan) in these sucrosedensity gradients. Fig. 6 shows that dystrophin and these dys-trophin-associated glycoproteins also co-fractionate with caveo-lin-3. This is consistent with results demonstrating the co-localization of caveolin-3 and dystrophin in intact skeletalmuscle fibers by fluorescence microscopy (Fig. 3).To further examine whether caveolin-3 and dystrophin are

physically associated as a discrete complex, co-immunoprecipi-tation experiments were performed. Lysates from differenti-ated C2C12 were immunoprecipitated with anti-dystrophinIgG, an irrelevant IgG or protein A-Sepharose beads alone.

FIG. 3. Immunolocalization of caveolin-3 in skeletal muscletissue. Frozen sections of mouse skeletal muscle tissue were preparedand doubly immunostained with a mouse mAb directed against caveo-lin-3 and a rabbit polyclonal antibody to dystrophin. Bound primaryantibodies were visualized by incubation with distinctly tagged fluores-cent secondary antibodies; see “Experimental Procedures.” A, trans-verse section, low magnification, bar 5 50 mm; B, tangential section,higher magnification, bar5 15 mm. Note that caveolin-3 and dystrophinco-localize to the sarcolemma (the plasma membrane of the musclefiber).

FIG. 4. Induction of caveolin-3 protein during the differentia-tion of C2C12 skeletal myoblasts. Cell lysates were prepared fromproliferating (P) or differentiated (D) C2C12 skeletal myoblasts andsubjected to immunoblot analysis with a caveolin-3-specific mAb probe.Note that caveolin-3 expression is selectively induced during myoblastdifferentiation. No expression of caveolin-1 was detected using the samecell extracts (not shown).

Caveolin-3 and the Dystrophin Complex 15163

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Immunoprecipitates were then subjected to immunoblot anal-ysis with a mAb directed against caveolin-3. Fig. 7 shows thatcaveolin-3 specifically co-immunoprecipitates with dystrophin;little or no caveolin-3 was found associated with an irrelevantIgG or beads alone.Expression of Caveolin-3 in Cardiac Myocytes and Smooth

Muscle Cells—Primary cultures of cardiac myocytes andsmooth muscle cells were used to further examine the muscle-specific expression of caveolin-3. Fig. 8 shows the differentialexpression of caveolin-1 and caveolin-3 in these cell types.While only caveolin-3 is expressed in cardiac myocytes (leftpanel), both caveolin-1 and caveolin-3 are co-expressed insmooth muscle cells (right panel). Thus, it appears that striated

muscle cell types (cardiac myocytes and skeletal muscle fibers)express predominantly caveolin-3.

DISCUSSION

Duchenne and Becker muscular dystrophies result from mu-tations in the dystrophin gene (43). Dystrophin is a high mo-lecular mass cytoskeletal protein that co-purifies with a com-plex of dystrophin-associated proteins which are thought toanchor dystrophin to the cytoplasmic face of the muscle cellplasma membrane (44). Several dystrophin-associated proteinshave been identified and cloned (reviewed in Ref. 45). Theseinclude: (i) the dystroglycan complex (156- and 43-kDa glyco-proteins, termed a- and b-dystroglycan); (ii) the sarcoglycan

FIG. 6.Co-fractionation of caveolin-3 with dystrophin and dys-trophin-associated glycoproteins. Gradient fractions depicted inFig. 5 were subjected to immunoblot analysis with specific antibodiesdirected against caveolin-3, dystrophin, and dystrophin-associated gly-coproteins (a-sarcoglycan and b-dystroglycan). Note that both dystro-phin and dystrophin-associated glycoproteins co-fractionate withcaveolin-3.

FIG. 5. Subcellular fractionation of differentiated C2C12 skel-etal myoblasts that endogenously express caveolin-3. Differenti-ated C2C12 cells were subjected to subcellular fractionation afterhomogenization in a buffer containing sodium carbonate (see “Experi-mental Procedures”). Note that in this fractionation scheme: (i) caveo-lin-3 (fractions 5 and 6) is separated from most cellular proteins (frac-tions 9–12) and (ii) caveolin-3 co-fractionates with Gbg and c-Src(fractions 5 and 6). Other G protein subunits (Gi,2a) and Src-familymembers (Lyn) also co-fractionated with caveolin-3 in this detergent-free system (not shown).

FIG. 7.Co-immunoprecipitation of caveolin-3 with dystrophin.Differentiated C2C12 cells were lysed and subjected to immunoprecipi-tation with specific rabbit polyclonal IgG directed against dystrophin,glutathione S-transferase (GST) (an irrelevant rabbit polyclonal IgG),or protein A-Sepharose alone. After extensive washing, samples wereseparated by SDS-PAGE (15% acrylamide) and transferred to nitrocel-lulose. Blots were then probed with a mAb directed against caveolin-3.Note that that caveolin-3 specifically co-immunoprecipitates with dys-trophin; little or no caveolin-3 was found associated with an irrelevantIgG (GST) or beads alone.

FIG. 8. Expression of caveolin-3 in cardiac myocytes andsmooth muscle cells. Lysates from primary cultures of rat cardiacmyocytes (left panel) and aortic smooth muscle cells (right panel) wereprepared and subjected to immunoblot analysis with mAbs to eithercaveolin-1 (2297) or caveolin-3. Note that only caveolin-3 was detectedin cardiac myocytes, while smooth muscle cells co-express caveolin-1and caveolin-3. Smooth muscle cells are known to express caveolin-1(14).

Caveolin-3 and the Dystrophin Complex15164

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from

complex (50, 43, and 35 kDa, termed a-, b-, and g-sarcoglycan);and (iii) an unknown ;20–25-kDa nonglycosylated integralmembrane protein. Mutations in a-, b-, and g-sarcoglycan havebeen detected in either autosomal recessive muscular dystro-phy, limb-girdle muscular dystrophy, or severe childhood au-tosomal recessive muscular dystrophy (45). To date, no geneticlesions have been described in a- or b-dystroglycan. The func-tion of the dystrophin complex remains unknown, althoughsome have suggested that the dystrophin complex may formpart of a stretch-activated calcium channel that is defective inDuchenne muscular dystrophy.Two related morphological observations seemingly implicate

muscle cell caveolae in the pathogenesis of Duchenne musculardystrophy: (i) dystrophin has been localized to plasma mem-brane caveolae in smooth muscle cells using immunoelectronmicroscopy techniques (32) and (ii) another electron microscopystudy demonstrates that skeletal muscle caveolae undergocharacteristic changes in size and their distribution in patientswith Duchenne muscular dystrophy, but not in other forms ofneuronally based muscular dystrophies examined (33). How-ever, little or no caveolin-1 is expressed in cardiac myocytesand skeletal muscle fibers (30). As such it has been difficult toassess the potential role of caveolin proteins and caveolae inmuscle cell membrane biology, as no caveolin probes exist forskeletal and cardiac muscle. However, recent studies indicatethat skeletal and cardiac tissues express mRNA encoding amuscle-specific caveolin gene, termed caveolin-3 (30).Here, we have generated and characterized a mAb probe that

recognizes the caveolin-3 protein, but not other known membersof the caveolin gene family. Undoubtedly, this novel probe willgreatly facilitate the study of caveolae in striated muscle cells.Immunolocalization of caveolin-3 in skeletal muscle fibers re-veals that caveolin-3 is localized to the sarcolemma (muscle cellplasma membrane) and this immunostaining coincides with thedistribution of another muscle-specific plasma membranemarker protein, dystrophin. In addition, detergent-free subcellu-lar fractionation studies indicate that caveolin-3 co-fractionateswith members of the dystrophin complex (dystrophin, a-sarcogly-can, and b-dystroglycan) and cytoplasmic signaling molecules(G-proteins and Src-like kinases), but is clearly separated fromthe bulk of cellular proteins. This is consistent with previousreports demonstrating that in non-muscle cells caveolin-1 co-purifies with components of the membrane cytoskeleton andcytoplasmic signaling molecules (12–19).Several independent lines of evidence suggest that caveo-

lin-3 may represent the “unknown 20–25-kDa integral mem-brane component of the dystrophin complex” mentioned above.Caveolin-3 is a 20–25-kDa integral membrane protein (i) thatis selectively expressed in muscle tissue types, (ii) co-localizeswith dystrophin to the sarcolemma of muscle fibers, (iii) co-fractionates with dystrophin and dystrophin-associated pro-teins, and (iv) co-immunoprecipitates with dystrophin. In thisregard, we are currently searching for mutations within thehuman caveolin-3 gene in patients with unknown causes ofmuscular dystrophy.

Acknowledgments—We thank Dr. Harvey F. Lodish for his enthusi-asm and encouragement; Dr. John R. Glenney for monoclonal antibod-ies (2297 and 2234) directed against caveolin; Drs. John R. Glenney andRoberto Campos-Gonzalez for mAb production; Drs. Justin Fallon,

Elizabeth McNally, and Louis Kunkel for helpful and stimulating dis-cussions; and Marcia Glatt and other members of the Whitehead pur-chasing department for their dedicated service.

REFERENCES

1. Severs, N. J. (1988) J. Cell Sci. 90, 341–3482. Simionescu, N., Simionescu, M. & Palade, G. E. (1975) J. Cell Biol. 64,

586–6073. Anderson, R. G. W. (1993) Curr. Opin. Cell Biol. 5, 647–6524. Lisanti, M. P., Scherer, P., Tang, Z.-L. & Sargiacomo, M. (1994) Trends Cell

Biol. 4, 231–2355. Lisanti, M. P., Scherer, P. E., Tang, Z.-L., Kubler, E., Koleske, A. J. &

Sargiacomo, M. S. (1995) Semin. Dev. Biol. 6, 47–586. Rothberg, K. G., Heuser, J. E., Donzell, W. C., Ying, Y., Glenney, J. R. &

Anderson, R. G. W. (1992) Cell 68, 673–6827. Glenney, J. R. (1989) J. Biol. Chem. 264, 20163–201668. Glenney, J. R. & Soppet, D. (1992) Proc. Natl. Acad. Sci. U. S. A. 89,

10517–105219. Glenney, J. R., Jr. (1992) FEBS Lett. 314, 45–4810. Dupree, P., Parton, R. G., Raposo, G., Kurzchalia, T. V. & Simons, K. (1993)

EMBO J. 12, 1597–160511. Tang, Z.-L., Scherer, P. E. & Lisanti, M. P. (1994) Gene (Amst.) 147, 299–30012. Sargiacomo, M., Sudol, M., Tang, Z. & Lisanti, M. P. (1993) J. Cell Biol. 122,

789–80713. Lisanti, M. P., Scherer, P. E., Vidugiriene, J., Tang, Z.-L., Hermanoski-

Vosatka, A., Tu, Y.-H., Cook, R. F. & Sargiacomo, M. (1994) J. Cell Biol.126, 111–126

14. Chang, W. J., Ying, Y., Rothberg, K., Hooper, N., Turner, A., Gambliel, H., DeGunzburg, J., Mumby, S., Gilman, A. & Anderson, R. G. W. (1994) J. CellBiol. 126, 127–138

15. Robbins, S. M., Quintrell, N. A. & Bishop, M. J. (1995) Mol. Cell. Biol. 15,3507–3515

16. Schnitzer, J. E., Liu, J. & Oh, P. (1995) J. Biol. Chem. 270, 14399–1440417. Smart, E. J., Ying, Y.-S. & Anderson, R. G. W. (1995) J. Cell Biol. 131, 929–93818. Smart, E. J., Ying, Y., Mineo, C. & Anderson, R. G. W. (1995) Proc. Natl. Acad.

Sci. U. S. A. 92, 10104–1010819. Song, K. S., Li, S., Okamoto, T., Quilliam, L., Sargiacomo, M. & Lisanti, M. P.

(1996) J. Biol. Chem. 271, 9690–969720. Sargiacomo, M., Scherer, P. E., Tang, Z.-L., Kubler, E., Song, K. S., Sanders,

M. C. & Lisanti, M. P. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 9407–941121. Monier, S., Parton, R. G., Vogel, F., Behlke, J., Henske, A. & Kurzchalia, T.

(1995) Mol. Biol. Cell 6, 911–92722. Li, S., Song, K. S. & Lisanti, M. P. (1996) J. Biol. Chem. 271, 568–57323. Murata, M., Peranen, J., Schreiner, R., Weiland, F., Kurzchalia, T. & Simons,

K. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 10339–1034324. Fan, J. Y., Carpentier, J.-L., van Obberghen, E., Grunfeld, C., Gorden, P. &

Orci, L. (1983) J. Cell Sci. 61, 219–23025. Scherer, P. E., Lisanti, M. P., Baldini, G., Sargiacomo, M., Corley-Mastick, C.

& Lodish, H. F. (1994) J. Cell Biol. 127, 1233–124326. Scherer, P. E., Okamoto, T., Chun, M., Lodish, H. F. & Lisanti, M. P. (1996)

Proc. Natl. Acad. Sci. U. S. A. 93, 131–13527. Koleske, A. J., Baltimore, D. & Lisanti, M. P. (1995) Proc. Natl. Acad. Sci.

U. S. A. 92, 1381–138528. Scherer, P. E., Tang, Z., Chun, M., Sargiacomo, M., Lodish, H. F. & Lisanti, M.

P. (1995) J. Biol. Chem. 270, 16395–1640129. Fra, A. M., Williamson, E., Simons, K. & Parton, R. G. (1995) Proc. Natl. Acad.

Sci. U. S. A. 92, 8655–865930. Tang, Z.-L., Scherer, P. E., Okamoto, T., Song, K., Chu, C., Kohtz, D. S.,

Nishimoto, I., Lodish, H. F. & Lisanti, M. P. (1996) J. Biol. Chem. 271,2255–2261

31. Way, M. & Parton, R. (1995) FEBS Lett. 376, 108–11232. North, A. J., Galazkiewicz, B., Byers, T. J., Glenney, J. R. & Small, J. V. (1993)

J. Cell Biol. 120, 1159–116733. Bonilla, E., Fishbeck, K. & Schotland, D. (1981) Am. J. Pathol. 104, 167–17334. Harlow, E. & Lane, D. (1988) Antibiotics: A Laboratory Manual, Cold Spring

Harbor Laboratory, Cold Spring Harbor, NY35. Seed, B. & Aruffo, A. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 3365–336936. Cole, F., Fasy, T. M., Rao, S., de Peralta, M. & Kohtz, D. S. (1993) J. Biol.

Chem. 268, 1580–158537. Blau, H., Chiu, C.-P. & Webster, C. (1983) Cell 32, 1171–118038. Li, S., Okamoto, T., Chun, M., Sargiacomo, M., Casanova, J. E., Hansen, S. H.,

Nishimoto, I. & Lisanti, M. P. (1995) J. Biol. Chem. 270, 15693–1570139. Lisanti, M. P., Tang, Z.-T., Scherer, P. & Sargiacomo, M. (1995) Methods

Enzymol. 250, 655–66840. Lisanti, M. P., Tang, Z.-L. & Sargiacomo, M. (1993) J. Cell Biol. 123, 595–60441. Springhorn, J. P., Ellingsen, O., Berger, H. J., Kelly, R. A. & Smith, T. W.

(1992) J. Biol. Chem. 267, 14360–1436542. Graves, L. M., Bornfeld, K., Argast, G., Krebs, E., Kong, X., Lin, T. &

Lawrence, J. C. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 7222–722643. Hoffman, E. P. & Kunkel, L. M. (1989) Neuron 2, 1019–102944. Ervasti, J. M. & Campbell, K. P. (1991) Cell 66, 1121–113145. Worton, R. (1995) Science 270, 755–756

Caveolin-3 and the Dystrophin Complex 15165

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Mark Chafel, Caryn Chu, D. Stave Kohtz and Michael P. LisantiKenneth S. Song, Philipp E. Scherer, ZhaoLan Tang, Takashi Okamoto, Shengwen Li,

GLYCOPROTEINSCO-FRACTIONATES WITH DYSTROPHIN AND DYSTROPHIN-ASSOCIATED

CAVEOLIN-3 IS A COMPONENT OF THE SARCOLEMMA AND Expression of Caveolin-3 in Skeletal, Cardiac, and Smooth Muscle Cells:

doi: 10.1074/jbc.271.25.151601996, 271:15160-15165.J. Biol. Chem. 

  http://www.jbc.org/content/271/25/15160Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/271/25/15160.full.html#ref-list-1

This article cites 44 references, 31 of which can be accessed free at

by guest on July 6, 2020http://w

ww

.jbc.org/D

ownloaded from