The spatial arrangement of the cell-surface membranes (sarcolemma and transverse tubules) and internal membranes of the sarcoplasmic reticulum relative to the myofibril is critical for effective excitation-contraction (E-C) coupling in cardiac myocytes; however, the molecular determinants of this order remain to be defined. Here, we ascribe molecular and cellular properties to the coiled-coil, tail-anchored sarcolemmal membrane-associated protein (SLMAP) that are consistent with a potential role in organizing the E-C coupling apparatus of the cardiomyocyte. The expression of SLMAP was developmentally regulated and its localization was distinctly apparent at the level of the membranes involved in regulating the E-C coupling mechanism. Several SLMAP isoforms were expressed in the cardiac myocyte with unique COOH-terminal membrane anchors that could target this molecule to distinct subcellular membranes. Protein interaction analysis indicated that SLMAPs could self assemble and bind myosin in cardiac muscle. The cardiac-specific expression of SLMAP isoforms that can be targeted to distinct subcellular membranes, self assemble, and interact with the myofibril suggests a potential role for this molecule in the structural arrangement of the E-C coupling apparatus.
- sarcoplasmic reticulum
- transmembrane domain
- sarcolemmal membrane-associated protein
the cardiomyocyte has evolved a specialized subcellular architecture that is necessary for the regulation of signal transduction underlying the process of excitation-contraction (E-C) coupling (1, 3, 4, 7, 21). The sarcoplasmic reticulum (SR), derived from the endoplasmic reticulum (ER), constitutes an abundant internal membrane system comprised of two distinct membrane domains: the longitudinal (nonjunctional) SR and the terminal cisternae (junctional) SR (7, 27). Situated in close apposition to the terminal cisternae SR is a system of sarcolemmal invaginations, known as transverse (T)-tubules. Localized entry of calcium ions through the L-type calcium channel situated at the T-tubules is induced on depolarization of the sarcolemma and triggers the release of calcium from the SR-localized calcium release channels into the cytosol (3, 4, 6, 20). This increase in intracellular calcium levels activates the closely situated contractile units composed of the myofilaments to facilitate myocyte shortening. The subsequent removal of calcium by the activities of the SERCA pumps of the longitudinal SR and the Na+/Ca2+ exchanger present at the T-tubules and sarcolemma results in relaxation (12, 19, 22). Thus the proper spatial distribution of these membrane systems relative to the myofilaments is pivotal for the coordination of the E-C mechanism (1, 3, 4, 7).
The precise mechanisms that organize the membrane architecture of the cardiac myocyte so that the appropriate targeting of the E-C coupling apparatus can be achieved and maintained on a beat-to-beat basis remain poorly understood. Thus the identification of the molecular components necessary for the organization and stabilization of the cardiac membrane architecture would not only advance the understanding of cardiomyocyte function but also offer insights into heart disease where membrane disorganization has been correlated with dysfunction (1, 8, 9). Recent studies indicate that a family of tail-anchored membrane proteins designated the junctophilins (JPs) contributes to the formation of the dyadic couplings between the SR/ER and the T-tubules in cardiomyocytes (13, 25, 26). A COOH-terminal hydrophobic segment targets the JPs to the ER/SR membrane, whereas the cytoplasmic domain interacts with lipid moieties of the plasma membrane (25). Although the JPs appear critical for membrane organization, studies also suggest that other as yet unidentified molecular entities are also necessary for the proper arrangement of the E-C coupling mechanism. In previous studies, we defined the primary structure of a family of coiled-coil, tail-anchored sarcolemmal membrane-associated proteins termed the SLMAPs that were highly expressed in the myocardium (11, 28, 29). The genomic analysis of the SLMAP gene revealed that several SLMAP isoforms were generated by alternative splicing mechanisms with unique COOH-terminal hydrophobic domains. The SLMAPs share an overall structural similarity to other tail-anchored membrane proteins such as the syntaxins, which serve roles in membrane docking and fusion events (16). Here, we provide evidence that SLMAPs can localize at the level of the SL, T-tubules, and SR and be targeted to distinct cellular membranes by their unique COOH-terminal tails. We also provide evidence that SLMAPs can self assemble and interact with cardiac myosin. These properties imply a role for these molecules in the structural arrangement of the E-C coupling mechanism of the cardiomyocyte.
MATERIALS AND METHODS
Timed pregnant mice were killed by lethal injection of Somital. Embryos (9–15 days postcoitum) were removed, rinsed briefly in PBS, and then immediately fixed in 4% paraformaldehyde in 100 mM phosphate buffer, pH 7.4. For cryoprotection, embryos were incubated in 20% sucrose, followed by incubation in 15% sucrose in PBS. Fixed embryos were embedded in Tissue-Tek O.C.T. compound, frozen, and stored at −80°C. Cryosections (6–10 μm) were collected on gelatin-coated microscope slides and stored at −80°C. Older embryos (18 days postcoitum) and adult tissues were isolated, rinsed in PBS, embedded in Tissue-Tek OCT compound, and then frozen. Cryosections were mounted onto slides and fixed overnight in 4% paraformaldehyde. Immunohistochemistry was performed by first prewarming slides to 37°C. Paraformaldehyde-fixed sections were washed in PBS and then treated with 50 mM ammonium chloride in PBS for 5 min to reduce nonspecific staining of blood proteins. Sections were blocked in PBS containing 10% goat serum, 1% Triton X-100, and 10% bovine serum albumin for 20 min at room temperature before being exposed to primary antibodies (1 h at room temperature). After several washes in PBS, sections were incubated in PBS containing 5% goat serum, 1% Triton X-100, and the appropriate fluorochrome-linked secondary antibodies for 1 h at room temperature. After several washes in PBS, sections were mounted with antifading solution (Molecular Probes).
The following antibodies were used in the immunocytochemistry/histochemistry studies: monoclonal anti-Myc 9E10 monoclonal antibody (Dr. J. Bell, University of Ottawa); anti-α-actinin monocolonal antibody (Sigma, clone BM-75.2); monoclonal anti-caveolin-3 antibody, anti-ryanodine receptor (Transduction Laboratories). Anti-SLMAP(C) rabbit antisera was raised against the COOH 370 amino acids of SLAP, as previously described (29). The secondary antibodies used in these studies included Alexa-Fluor 488-conjugated anti-rabbit immunoglobulins and Alexa Fluor 594-conjugated anti-mouse immunoglobulins (Molecular Probes).
Generation of glutathione S-transferase fusion proteins.
Rabbit SLMAP1 cDNA was PCR amplified from full-length SLMAP3 cDNA using the forward primer (GEXSL15′: cgGAATTCaaagcagtgacgacac) and reverse primer (GEX3′: cgGAATTCgtgtacggactcaagaaa). Full-length SLMAP3M1 cDNA was generated by PCR using the GEXM15′ forward primer (cgGAATTCccagtctgtcctcgatg) and GEX3′ reverse primer (cgGAATTCgtgtacggactcaagaaa). PCR conditions consisted of 1) 3 min at 94°C; 2) 30 cycles of 94°C for 40 s (denaturing), 60°C for 30 s (annealing), 72°C for 2.5 min (extension); followed by 3) one last step at 72°C for 10 min. PCR products were EcoRI digested and ligated into EcoRI site of pGEX-2TK vector (source) to yield pGEX-SLMAP1. Construct orientation was verified by restriction site mapping and sequencing. pGEX-2TK or pGEX-SLMAP1 was used to transform BL21 Codon Plus (Stratagene) competent cells. Transformed cells were plated on LB agar plates supplemented with ampicillin (1,000 μg/ml), chloramphenicol (40 μg/ml), and 2% glucose. Overnight bacterial cultures for each transformant were grown at 37°C until an optical density reading (A600nm) of 0.5–0.6 was reached. Fusion protein synthesis was induced with 0.1 mM IPTG for 4 h at room temperature. Cells were collected by centrifugation for 10 min at 5,000 rpm and resuspended in lysis buffer (PBS, pH 7.4, 1% Triton X-100, 0.8% NaCl, 5% glycerol) supplemented with protease inhibitor cocktail (Sigma). To lyse the cells, 0.1 mg/ml of lysozyme was added and lysates were incubated on ice for 30 min, followed by brief sonication. Sonicated material was centrifuged for 20 min at 20,000 g, and collected supernatants were then incubated with glutathione Sepharose 4B (Amersham Pharmacia) beads with constant rotation at 4°C. Beads were separated from the lysate by brief centrifugation (4,000 rpm, 10 s), and the protein-linked beads were washed several times with lysis buffer. Two additional washes were performed with purification buffer (PBS, pH 7.4, 0.5% NP-40, 0.8% NaCl, 5% glycerol).
Glutathione S-transferase pull-down assays.
H9c2 cells were lysed in RIPA buffer supplemented with protease inhibitor cocktail (Sigma), and unbroken cells were removed by centrifugation. The H9c2 cell lysate was precleared with glutathione Sepharose 4B (Amersham Pharmacia) beads linked to glutathione S-transferase (GST; 30-min rotation at 4°C). Six milligrams of precleared cell lysate were incubated overnight at 4°C with equivalent amounts of purified GST or GST-SLMAP1 bound to glutathione Sepharose 4B beads. After being bound, the beads were extensively washed with 1× RIPA, 0.5× RIPA or PBS, and then resuspended in 2× sample buffer. Captured proteins were eluted from the beads by boiling and separated on 7.5% SDS-PAGE gels.
Microsomes were isolated from adult rabbit ventricles by homogenization in buffer consisting of 50 mM Tris·HCl, pH 7.8, 20 mM sodium tetraphosphate, and 1 mM EDTA supplemented with protease inhibitor cocktail (Sigma). The homogenate was cleared of debris by centrifugation at 10,000 g for 10 min at 4°C and then filtered through cheesecloth. KCl was added to a final concentration of 0.6 M, and the cleared homogenate was further centrifuged at 100,000 g for 45 min at 4°C. The pellet was retained and solubilized overnight with constant rotation in buffer consisting of 12.5 mM MOPS, pH 7.2, 0.8% CHAPS, 0.5% NP-40, 450 mM NaCl, 150 mM sucrose, and 25 μM EGTA. Solubilized membranes were then subjected to high-speed centrifugation (190,000 g, 1 h). The supernatant was retained as the solubilized crude microsome fraction.
Silver staining, digestion, and mass spectrometry.
SDS-polyacrylamide gels were fixed in 50% ethanol, 5% acetic acid solution for 30 min at room temperature. Gels were then washed for 10 min in 50% ethanol followed by two 10-min washes in double-distilled water and incubated in silver-staining sensitizer solution (0.02% sodium thiosulphate) for 5 min and then washed twice in double-distilled water before staining in 0.1% silver nitrate for 30 min. Gels were briefly rinsed in double-distilled water and then exposed to developer solution (0.04% formalin in 2% sodium carbonate). This step was carefully monitored, and the reaction was stopped by the addition of 5% acetic acid solution on visualization of bands. Gels were stored in 1% acetic acid solution. Gel slices were destained (15 mM potassium ferricyanide, 50 mM sodium thiosulphate) and incubated at room temperature with shaking. Gel pieces were then washed (3 × 10 min) in deionized water, and acetonitrile was added to shrink the gel pieces. Gel pieces were rehydrated with buffer consisting of 0.01 μg/μl of trypsin in 50 mM ammonium bicarbonate and digested overnight at 37°C. The buffer containing the digested peptides was transferred into a mass spectrometer analysis plate and analyzed on the Q-tof ultima instrument (Micro Mass, courtesy of Dr. J. Kelly, National Research Council of Canada). Samples were analyzed using a nano-LC-MS/MS method with runs of 45 min/samples and a gradient of acetonitrile ranging from 5 to 55%.
Cell culture and transfections.
COS7 African green monkey kidney cells and H9c2 cells (CRL-1446, ATCC) were maintained at 37°C in DMEM supplemented with 10% heat-inactivated FBS and antibiotics. Transient transfection experiments were performed using the Fugene (Roche Biochemicals) transfection reagent according to the manufacturer’s specifications. 6Myc-tagged SLMAP1 expression constructs encoding transmembrane domain 1 (TM1), TM domain 2 (TM2), or the construct lacking both TM (ΔTM) domains were generated.
Cells grown on sterile glass coverslips were fixed in either 4% paraformaldehyde in phosphate buffer for 15 min at room temperature or in ice-cold methanol for 5 min. After either fixation method, coverslips were mounted onto glass slides and cells were incubated with the relevant antibody(s) diluted in 0.01 M PBScontaining 0.3% Triton X-100 (PBS-T) for 2 h at room temperature. After several washes in PBS, cells were incubated in the appropriate fluorochrome-conjugated secondary antibody(s) for 45 min at 37°C. Cells were washed extensively in PBS and then mounted with antifade media (Molecular Probes).
Cell isolation and labeling.
Ventricular myocytes were isolated from adult male Wistar rats and fixed using methods previously described (23). The fixed cells were then adhered to acid-washed coverslips using poly-l-lysine (Sigma), rinsed with 3 × 10-min washes in PBS (in mM: 137 NaCl, 8 NaH2PO4, 2.7 KCl, 1.5 K H2PO4, pH 7.4), permeabilized in a PBS solution containing 0.1% Triton X-100 for 10 min, and given a final 3 × 10-min rinse in PBS. The cells were then incubated simultaneously with the polyclonal anti-SLMAP (1:100) and a monoclonal antibody, either anti-caveolin-3 (1 μg/ml, BD Transduction Laboratories) or anti-ryanodine receptor (1 μg/ml, Affinity Bioreagents) overnight in a humidified environment at room temperature. The antibodies were diluted in antibody buffer (in mM: 75 NaCl, 18 Na3citrate, 2% goat serum, 1% BSA, 0.05% Triton X-100, and 0.02% NaN3). After 3 × 10-min rinses in antibody wash solution (in mM: 75 NaCl, 18 Na3citrate, 0.05% Triton X-100), the cells were incubated with highly cross-adsorbed anti-rabbit and anti-mouse secondary antibodies that were labeled with fluorescein isothiocyanate (FITC) or Texas red (Jackson Immunobiologicals) for 2 h at room temperature and then given a final 3 × 10-min rinse in antibody wash solution. Some of the coverslips were used for control experiments and were incubated only with the antibody buffer overnight and then labeled with the fluorescently tagged secondary antibodies. All of the coverslips were mounted onto frosted slides in a solution composed of 90% glycerol, 10% 10× PBS, 2.5% triethylenediamine, and 0.02% NaN3. Fluorescent microspheres (0.2-μm diameter, Molecular Probes) labeled with FITC and Texas red were added to the mounting medium to act as fiduciary markers and permit accurate alignment of the three-dimensional data sets.
Image acquisition and deconvolution.
A series of two-dimensional images were acquired through the cells at 0.25-μm spacing using a Nikon Diaphot 200 microscope and a Planapo 60/1.4 objective. This configuration produced voxel dimensions of 100 × 100 × 250 nm and satisfied the Nyquist criteria for sampling and prevented aliasing. The image detector was a thermoelectrically cooled CCD camera with a Site S1502AB chip with a peak quantum efficiency of 80%. Other details of the microscope can be found in Scriven et al. (23). The point spread function of the microscope was measured using fluospheres of the appropriate color (0.1-μm diameter, Molecular Probes). The images were prepared, deconvolved, and analyzed as previously reported (5).
Samples were visualized using Axiophot (Carl Zeiss) microscope equipped with a 3CCD color video camera. An Olympus IX70 laser-scanning inverted microscope with a ×63 oil immersion objective was also used for observation. Images were acquired using the Bio-Rad MRC 1024 confocal. Acquired images were digitally processed using Northern Eclipse (Version 5.0, Empix Imaging) acquisition software as well as the Confocal Assistant 40 software. Images were further processed using Adobe Photoshop 5.0 (Adobe Systems).
SLMAP expression in the developing myocardium and adult cardiomyocytes.
Our previous studies indicate that there is robust SLMAP mRNA and protein expression in adult cardiac muscle (28, 29), but nothing is known about its expression in the developing myocardium. At 9 days postcoitum, the primitive heart exists as a tubular structure with presumptive atrial and ventricular chambers. SLMAP-antibody labeling of formalin-fixed sagital cryosections from whole mouse embryos revealed that SLMAP was expressed in both the atria and ventricles at this early developmental stage; however, a more robust staining was observed in the ventricular myocardium (Fig. 1A). At 13 days postcoitum, the developing heart is divided by septa into a four-chambered heart and SLMAP was found to be uniformly expressed in the atrial and ventricular myocardium, as well as the interventricular and interatrial septum (Fig. 1B, a, c, d). No immunostaining was observed in sections incubated in preimmune rabbit serum (Fig. 1Bc) or in the absence of the primary antibody (not shown). Confocal images of ventricular myocytes stained for anti-SLMAP at 13 days postcoitum revealed diffuse SLMAP localization, and a distinct distribution of SLMAPs relative to the myofibrils was not observed at this stage of development (Fig. 2, a-c). In older embryos (18 days postcoitum), a clear striated pattern of SLMAP localization was apparent and mostly distinct from that of the Z-line marker α-actinin (Fig. 2, d-f). Immunostaining and confocal microscopy were used to further examine the subcellular localization of SLMAP in isolated adult rat ventricular myocytes. A clear cross-striated pattern of SLMAP distribution was observed in adult cardiac cells (Fig. 2, g-i). Anti-SLMAP labeling of the Z-line was apparent by confocal microscopy (Fig. 2i); however, substantial SLMAP staining was also noted at reticular structures perpendicular to the Z-line staining. Further analysis of SLMAP localization in single adult cardiomyocytes using confocal and deconvolution techniques (Figs. 3 and 4) indicated that SLMAP-specific staining extended toward the periphery of the cell as well as the invaginations of the plasma membrane (Fig. 3). Thus we examined whether SLMAP codistributed with markers of external membrane systems and the internal membranes such as the SR. Caveolin-3 proteins are expressed in cardiac, skeletal, and smooth muscle and reside at the vesicular invaginations of the plasma membrane known as caveolae (24). Caveolin-3 antibodies clearly labeled a subcompartment of the surface cardiac membrane as well as plasma membrane invaginations (Figs. 3b and 4B), consistent with the distribution of caveolin-3 in the T tubular membrane system in cardiac cells (17). SLMAP was seen to colocalize with caveolin-3 at the cell-surface membrane invaginations and along the Z-lines (white voxels) as resolved by deconvolution microscopy (Figs. 4B and 3c). Regional differences in the SLMAP-caveolin-3 colocalization were apparent as caveolin-3 along the Z-lines was more heavily colocalized with SLMAP (white voxels) than the cell surface (Figs. 3c and 4B).
The ryanodine receptors (RyRs) or the calcium release channels are essential components of the E-C coupling machinery and are localized at the terminal cisternae SR membranes. By immunofluorescent microscopy, RyRs are typically detected as a series of regularly spaced doublets along the Z-disc in the cardiomyocytes (Figs. 3e and 4C in red) (8). Dual staining with SLMAP and RyR2-specific antibodies revealed substantial colocalization of these molecules along the Z-disc as indicated by white voxels (Figs. 3, d-f, and 4C). Thus Fig. 4 shows that the SLMAP is distributed in multiple locations but it is prominent at the cell surface around the nuclear envelope and along the Z-lines. At the cell surface and in the nuclear envelope 32 ± 6% of the voxels containing signal specific for caveolin-3 also contained SLMAP (n = 3; Fig. 4B). Although at the Z-lines, 36 ± 7% (n = 4) of the voxels containing the RyR also contained SLMAP (Fig. 4C). Each of the images also shows prominent staining of the M-line by the SLMAP antibody.
Unique COOH-terminal anchors in SLMAP target distinct cellular membranes.
Our previous studies indicate that three distinct mRNAs encoding SLMAP polypeptides of 35, 60, and ∼80 kDa are expressed in cardiac muscle, and biochemical fractionation experiments revealed that these polypeptides are enriched in surface membrane and SR fractions (17). Our previous studies also reveal that alternative splicing generates SLMAPs with unique COOH-terminal TM domains TM1 or TM2 that are equally expressed in the myocardium (18). Whether these unique COOH-terminal membrane anchors can target SLMAPs to distinct subcellular membranes is not known. We therefore examined whether the unique COOH-terminal TM domains in SLMAPs can target these polypeptides to distinct subcellular organelles. SLMAP constructs encoding either TM1 or TM2 sequences were fused to the 6Myc epitope tag and transfected into COS7 cells and examined by immunohistochemistry. Transient expression of SLMAP sequences encompassing TM1 generated a pronounced perinuclear distribution, which also extended throughout the cytoplasm in a diffuse reticular pattern consistent with the distribution of the ER (Fig. 5A). Immunostaining with anti-calnexin confirmed the identity of the SLMAP-localized compartment as ER (data not shown).
Ectopic expression of TM2 sequences targeted the 6Myc-tagged SLMAP fusion protein to distinct reticular formations as well as tubular-like projections throughout the cell that were not observed in cells expressing TM1 (Fig. 5A). Extraction of the cells with Triton X-100 before fixation resulted in lack of myc staining in paraformaldehyde-fixed cells expressing 6Myc-SLMAP constructs, indicating that 6Myc-SLMAP is targeted to subcellular membrane structures (data not shown). A characteristic feature of all cardiac SLMAP isoforms is the presence of a coiled-coil leucine-rich motif in the COOH-terminal region encoded by exons 20 and 21. When the leucine-rich coiled-coil region was deleted from SLMAP, it still targeted the reticular structures (Fig. 5Ac), but when the TM domain was deleted, SLMAP was seen to be diffusely distributed throughout the cytoplasm (Fig. 5Ad). To further examine whether the expression of the two distinct COOH-terminal TM anchors mediates differential SLMAP-membrane associations, cotransfection experiments were performed using 1) SLMAP cDNA encompassing TM1 fused to 6Myc (6Myc-SLMAP-TM1) and 2) SLMAP cDNA encompassing TM2 fused to green fluorescent protein (GFP-SLMAP-TM2). The subcellular distribution of the 6Myc-SLMAP-TM1 fusion protein as well as the GFP-SLMAP-TM2 fusion proteins was monitored by visualizing GFP immunofluorescence relative to myc antibody labeling in cells coexpressing the two SLMAP fusion proteins (Fig. 5B). Whereas some overlap in the GFP-myc immunostaining was apparent in cells coexpressing TM1 and TM2, the localizations were predominantly distinct (Fig. 5B). Inclusion of the GFP tag did not appear to affect membrane targeting as cells that were cotransfected with GFP-SLMAP and 6Myc-SLMAP fusion proteins encompassing the same membrane anchor (TM2) showed completely overlapping distribution profiles (Fig. 5B). Thus it appears that the function of the two different membrane anchors in SLMAP is to target this polypeptide to distinct subcellular membrane compartments.
SLMAP can homodimerize in vivo and bind cardiac myosin.
To gain insight into the molecular interactions of cardiac SLMAP, GST-SLMAP fusion proteins were immobilized on glutathione-Sepharose beads and incubated with detergent extracts of rabbit ventricles H9c2 cardiomyoblasts. After being washed extensively to remove nonspecific and weakly bound proteins, the specifically bound proteins were eluted and analyzed by SDS-PAGE and visualized by silver staining (Fig. 6). Two polypeptides of apparent molecular mass ∼35 and ∼70 kDa were found to consistently bind to GST-SLMAP1 from cardiac muscle and cell extracts (Fig. 6). In addition, a polypeptide of ∼200 kDa was also eluted from the immobilized GST-SLMAP1 incubated with the cardiac fractions. Each of these polypeptides specifically interacted with the GST-SLMAP fusion proteins, as none of these were bound by GST alone or in other control experiments where GST or GST-SLMAP recombinant proteins were incubated with lysis buffer alone. The ∼200-, ∼70-, and ∼35-kDa proteins were excised from the silver-stained gels, trypsin digested, and further analyzed for amino acid composition by mass spectrometry. Protein database comparisons of the tryptic peptide sequences corresponding to the 70- and 35-kDa polypeptides matched accession numbers 790240, 790238, and 20870858, each representing SLMAP sequences. The peptide sequences acquired from the mass spectrometric analysis of the 200-kDa protein identified with the α- and β-subunits of myosin heavy chain of the myocardium.
We further analyzed the localization of SLMAP to see whether it resides at the level of the M-line where myosin is located in cardiomyocytes (Fig. 7). Images acquired from three different rat ventricular myocytes stained with anti-SLMAP demonstrate that SLMAP is distributed at the Z-lines (single-headed arrows) and along the M-lines (double-headed arrows) in cardiomyocytes.
The data here show that SLMAPs can reside at the level of the E-C coupling apparatus of the cardiomyocytes as well as perinuclear membrane structures including the ER. Molecular analysis of the SLMAP gene has predicted the presence of a large number of isoforms generated by alternative splicing mechanisms that are expressed in a tissue-specific manner (10, 11, 28, 29). The different locations of SLMAP noted here may reflect the presence of these diverse isoforms in cardiomyocytes. SLMAPs may potentially serve a role in the subcellular architecture of the E-C coupling mechanism by perhaps linking the SR to the myofibril on the one hand and organizing the cell-surface membrane to SR junctions on the other. Immunoconfocal imaging revealed that SLMAPs are expressed early in the developing myocardium and are localized within the SR and cell-surface membranes of the ventricular myocytes. Confocal imaging of SLMAP expression in mature cardiomyocytes indicates that SLMAPs are colocalized with markers of the surface membrane such as caveolin and those of the junctional SR such as the RyR. These immunohistochemical data are consistent with our previous biochemical studies, which indicated that three SLMAP polypeptides of ∼80, 70, and 35 kDa encoded by three transcripts were expressed in cardiac muscle and subfractionated with the cell-surface membranes and the SR (29). SLMAPs carry two distinct COOH-terminal TM domains that are expressed in cardiac muscle, and our data here show that either of these domains can target the SLMAPs to subcellular membrane compartments. Furthermore, SLMAP appeared to be targeted to distinct subcellular membrane compartments depending on the COOH-terminal membrane anchor used. Thus the function of the unique COOH-terminal membrane anchors in SLMAP is to direct this molecule into different subcellular membranes. It is notable that fetal, neonatal, and adult heart tissue was found to express multiple SLMAP transcripts encoding one or other COOH-terminal TM anchors generated by alternative splicing mechanisms (28). Taken together, these observations suggest that alternative COOH-terminal TM domains could target SLMAPs to the SR and the cell-surface membranes and also to other perinuclear structures where SLMAP was detected.
Protein-protein interaction analysis indicated that a cardiac-specific SLMAP variant (SLMAP1) binds with the β- and α-subunit of the myosin heavy chain in cardiac muscle, whereas immunolocalization studies indicated some SLMAP staining at the level of the M-line in adult cardiomyocytes. These data imply that SLMAP can target the SR membrane via the COOH-terminal TM domain and interact with the contractile apparatus via its N-terminal sequences. These findings may be significant as little is known regarding the molecular mechanisms that mediate the interactions between the SR membrane and the myofibrils, which would be critical for the proper alignment of these two organelles. In a recent report, a splice variant of the ankyrin 1 gene (ankyrin 1.5) was proposed to provide a molecular link between the SR membrane and myofibril by binding to a novel sarcomeric protein, termed obscurin in skeletal muscle (1, 2). Recent data suggest that obscurin interacts with sarcomeric myosin to serve a role in its ability to assemble into A bands in muscle (14). Whether the specific SLMAP isoforms may serve to provide molecular links between the membranes and the contractile cytoskeleton in the myocardium remains to be investigated.
Protein-protein interaction analysis also demonstrated that SLMAP can interact with itself in cardiac muscle. In this regard, GST-SLMAP fusion proteins can bind the 70- and 35-kDa cardiac isoforms, suggesting that these molecules can form homodimers in vivo. Our in vitro studies suggest that SLMAPs can form homodimers via their leucine-rich coiled-coil motifs (11). The distribution of SLMAPs in various cardiac membranes, coupled with their ability to self-assemble, suggests that SLMAPs may participate in the regulation of membrane function through homodimer formation. These polypeptides would be ideally situated to organize the specialized membrane junctions involving the cell-surface membranes and the SR. The SLMAPs could conceivably be targeted to these distinct membranes via their unique COOH-terminal anchors and through homodimer formation give rise to junctional couplings. The confocal microscopy of SLMAP distribution relative to other components of the E-C coupling machinery indicates that a pool of SLMAPs colocalized with cell-surface and junctional membranes. Recent studies suggest that the tail-anchored membrane proteins such as JP subtypes are involved in the formation of the junctional membrane structure in excitable cells (14–16). Additional studies have suggested that other molecules such as Mitsugumin 29 are also required for effective junctional couplings. Mitsugumin 29 is a synaptophysin family-related protein that is found to be located in the junction between the plasma membrane and SR of skeletal muscle and appears to contribute to the membrane arrangement and E-C coupling (18). Our data here suggest that the cardiac tail-anchored membrane protein SLMAP is a novel component and perhaps a unique candidate for a role in organizing the E-C coupling apparatus of the cardiomyocyte. In view of the extensive alternative splicing mechanisms that generate a multitude of SLMAP isoforms, the presence of this molecule at distinct subcellular locations suggests diverse roles in cell function (10).
This work was supported by a grant to B. S. Tuana from Heart and Stroke Foundation of Ontario, and R. M. Guzzo was supported by a studentship award from Canadian Institutes of Health Research.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 by the American Physiological Society