| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Max Delbrück Center for Molecular Medicine, 13092 Berlin, Germany;
Max Planck Institute for Physiological and Clinical Research, 61231 Bad Nauheim, Germany;
* Institute of Physiology, Humboldt University (Charité), 10117 Berlin, Germany; and
# Franz Volhard Clinic, Humboldt University (Charité), 13125 Berlin, Germany
1Correspondence: Max Delbrück Center for Molecular Medicine (MDC), Robert-Rössle-Str. 10, D-13092 Berlin, Germany. E-mail: haase{at}mdc-berlin.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: L-type calcium channel • ß subunit • protein–protein interaction • cardiomyocytes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5
6
7
8)
. It was originally identified as desmosomal plaque protein of epidermal cells (1)
and as a nuclear phosphoprotein whose expression is repressed in human cell lines of neuroblastoma and other tumors (2
, 3)
. Ahnak is encoded by an intronless gene located on human chromosome 11q12-q13 (9)
. The deduced amino acid sequence of human ahnak predicts a protein of 5643 amino acids that can be divided into three main structural regions: the amino-terminal 251 amino acids, a large central region of 
4300 amino acids with multiple repeated units, most of which are 128 amino acids in length, and the carboxyl-terminal 1002 amino acids. For the central domain, a heptad structure was identified that is marked by the recurrence of proline at every seventh residue. Within these sequences, hydrophobic and hydrophilic residues alternate resembling a ß strand 1.2 µm in length with a thin polyionic rod representing a fibrous protein (2)
. The carboxyl-terminal region of ahnak is characterized by several stretches of sequences with positively charged residues, indicating nuclear localization signals (2)
. Indeed, ahnak was localized to the nuclei in some human cell lines (3)
. It has been demonstrated that the carboxyl terminus of ahnak is responsible for translocation from the nucleus to the plasma membrane (10)
. Although data on molecular structure of ahnak suggested a role in the cytoarchitecture (2)
, ahnak’s definite function remains unclear. Recent reports support the hypothesis that ahnak might play a role in signal transduction: ahnak binds and activates phospholipase C-
1 in the presence of arachidonic acid (11)
and interacts with the EF hand Ca2+ binding protein S100B (12)
.
We first described ahnak expression in mammalian cardiomyocytes and its cyclic AMP-dependent protein kinase (PKA) -mediated in vivo phosphorylation. The association of ahnak with the regulatory ß2 subunit of L-type Ca2+ channel suggested a potential implication in the regulation of channel activity (6)
. Beside the intracellular ß2 subunit, the cardiac L-type Ca2+ channel complex contains the central pore-forming 
1C subunit and auxiliary 2/
subunits (13)
. Cardiac Ca2+ channel activity is not only regulated by the membrane potential and phosphorylation (13
, 14)
, but also by the cytoskeletal actin filament organization, which modulates the Ca2+ channel inward current (15)
.
In the present study, we identify and characterize interaction sites between ahnak, the ß2 subunit of the L-type Ca2+ channel, and F-actin. Confocal imaging of cryosections of human heart localized ahnak to sarcolemmal compartments. We suggest an important role for ahnak in the subsarcolemmal cytoarchitecture and in tethering the actin cytoskeleton of cardiomyocytes with the L-type Ca2+ channel.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, respectively, were generously donated by Dr. Yun Soo Bae (Ewha Womans University, Seoul, Korea). The expression plasmid of the rabbit ß2a, pcDNAß2a was the generous gift of Dr. Franz Hofmann (Technical University, München, Germany). Antibodies against the Ca2+ channel subunits used in this study have been described: 1S (16)
, 1C (17)
, ß2 (6
, 17
, 18)
, ßcom (6)
. The anti-ahnak antibody KIS directed against the internal repeating units was described in ref 6
. The 18 amino acid 1 interaction domain (AID) peptide corresponding to 1C was as in ref 18
. Isopropyl-(-D)-thiogalactopyranoside was from Diagnostic Chemicals (Oxford, CT). Glutathione-Sepharose 4B and BrCN-activated Sepharose 4B were from Pharmacia Biotech (Piscataway, NJ). The enhanced chemiluminescence reaction kit and [-32P]ATP were from Amersham (Arlington Heights, IL). The PKA was prepared as in ref 19
.
Generation of expression constructs
Human ahnak cDNAs were prepared from two plasmids (z37 and z7) (2)
bearing the NH2 terminus and the carboxyl terminus, respectively. The amino-terminal cDNA fragment encoding amino acids (aa) 1–257 was constructed by digesting z37 with EcoRI and BsaBI and ligated with EcoRI/SmaI-cleaved pGEX-4T1 (Amersham Pharmacia Biotech) to yield GST-ahnak-N. The carboxyl terminus encoding aa 4646–5643 was split into two fragments: ahnak-C1 encoding aa 4646–5288 and ahnak-C2 encoding aa 5262–5643. To obtain ahnak-C1, the z7 plasmid was digested with NspV, blunted with Klenow fragment of E. coli DNA polymerase I, and cleaved with XhoI. This fragment was inserted into SmaI and XhoI restriction sites of pGEX-4T1. After removal of the 1.3-kb NcoI-XhoI fragment the ends were blunted and ligated. The ahnak-C2 fragment was constructed by digesting z7 with BspMI, blunted with Klenow fragment and cleaved with XhoI. The resulting 1.4-kb fragment was ligated with SmaI-XhoI sites of pGEX-4T1. The 2.1-kilobase SalI-NotI fragment of pcDNA3 carrying the complete protein-coding region of the rabbit ß2a subunit (20)
was inserted into the SalI-NotI site of pGEX-4T1 vector to yield GST-ß2a. The amino-terminal (aa 1–200; ß2a-N) and carboxyl-terminal (aa 195–606; ß2a-C) regions of ß2a were generated by PCR using the following primers ß2a-N forward: 5'-acgtgaattcggtcgacccacgcgtccgcc-3' and reverse: 5'-gatgcggccgcgacgttacactgtttgcact-3'; ß2a-C forward: 5'-acgtgaattcagtgcaaacagtgtaacgtc-3' and reverse: 5'-gatgcggccgctggcggatgtaaacatccct-3'. EcoRI-NotI restriction sites were incorporated into primers (indicated by italic letters) to facilitate subsequent subcloning to EcoRI-NotI sites of the expression vector pGEX-4T1. All constructs were checked by restriction site mapping and sequencing.
Purification of GST fusion proteins
A single colony of BL21-CodonPlus(DE3)-RIL (Stratagene, San Diego, CA) transformed with the plasmid of interest was grown overnight in LB medium containing 100 µg/mL ampicillin; 50–100 mL of the overnight culture was added to 1 L of LB medium containing 100 µg/mL ampicillin and grown for 3 h. Bacteria were induced with 0.1 mM IPTG for 3–4 h at 37°C. The cells were collected by centrifugation (8100 rpm for 10 min at 4°C on Beckman JLA10.500 rotor) and resuspended in 20 mL STE buffer (10 mM Tris-HCl buffer, pH 8.0, 150 mM NaCl, 1 mM EDTA) containing 0.1 mM PMSF, 1 µM pepstatin A, 0.1 mM benzamidine, 1 mM iodoacetamide, 1% TritonX-100, 2 mM dithiothreitol, and sonicated twice for 45 s (power level 70%, cycle 70) in a SonoPlus HD70 (Bachofer Laboratoriumsgeräte, Reutlingen, Germany). The sonicated material was centrifuged (10,000 rpm for 10 min at 4°C on Beckman JA-20 rotor). The supernatant was incubated with glutathione-Sepharose beads 10 min at room temperature on a rotating wheel. The beads were washed extensively with STE buffer; the GST fusion proteins were eluted with 20 mM glutathione in 50 mM NaHCO3 buffer, pH 9.0, and dialyzed against 30 mM HEPES-Tris buffer, pH 7.4. The identities of the purified recombinant proteins were verified by determination of the amino acid composition according to ref 21
.
Phosphorylation of ß2a fusion proteins
The ß2a fusion proteins, full-length GST-ß2a subunit, GST-ß2a-N, or GST-ß2a-C (200 µg each) were rebound to glutathione-Sepharose beads (50 µL packed gel), equilibrated with basal buffer containing 40 mM HEPES-Tris buffer, pH 7.4, 10 mM MgCl2, 1 mM EDTA, and, if not stated otherwise, 10 µM [-32P]ATP and 0.5 µM PKA. Reactions proceeded for 5 min at 30°C. Phosphorylated proteins were eluted with 20 mM glutathione in 100 mM NaHCO3 buffer, pH 9.0, and diluted with blocking solution for overlay assays.
Overlay binding assay
GST-ahnak fusion proteins (120 pmol) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 10% acrylamide) and blotted onto nitrocellulose membranes. The blots were incubated for 1 h in blocking solution containing 5% bovine serum albumin (BSA), 0.5% nonfat dry milk in phosphate-buffered saline (PBS), pH 7.5. Nitrocellulose filters were overlaid with equimolar concentrations of phosphorylated ß2a fusion proteins (30 nM, each) in blocking solution for 3 h at room temperature. The membranes were washed 1 h with 5% BSA in PBS, air dried, and exposed to a Fuji imaging system (type BAS-III) for 15–25 h.
Analytical ultracentrifugation
Sedimentation equilibrium studies of ahnak-C2 and ß2a or its truncates were performed in an XL-A type analytical ultracentrifuge (Beckman, Palo Alto, CA) equipped with UV absorbance optics. Sedimentation equilibrium was reached after 2 h of overspeed at 18,000 rpm, followed by an equilibrium speed of 14,000 rpm for 30 h at 10°C. Depending on the loading concentration, the radial absorbance in each compartment was recorded at three different wave lengths between 240 and 290 nm using the molar absorbance coefficients. Molecular mass determinations used the global fit of three radial distributions described by Ar = ArmeMF with F = [(1-
) 
2 (r2-rm2)]/2RT using the program Polymole (22)
. In these equations is the solvent density, is the partial specific volume, is the angular velocity, R is the gas constant, and T is the absolute temperature. Ar means the radial absorbance and Arm represents the corresponding value at meniscus position. Determination of molecular mass and analyzing absorbance profiles at three different wave lengths allowed us to estimate the total concentration of the components. Dissociation constants and stoichiometry for the reacting components were derived from the sum of exponentials (radial concentration distribution curves) considering molecular mass, loading concentration, and extinction coefficients of the reactants as described in detail by us (22)
. This technique was successfully applied for analysis of heterologous associations between different proteins (22
23
24)
and to study the interaction between proteins and nucleic acids (25
26
27
28)
. In the case of more than one binding site on the receptor molecule, the statistic binding mode according to Wyman and Gill (29)
was considered.
Preparation of tissue lysates, Ca2+ channels, and actin
Lysates from heart and skeletal muscle were prepared using SDS sample buffer as described (17)
. Cardiac Ca2+ channels were solubilized from pig sarcolemma membranes with 1% digitonin in the presence of a protease inhibitor mixture as in ref 30
. Rabbit skeletal muscle microsomal membranes (according to ref 16
) were solubilized and purified on wheat germ lectin Sepharose (30)
using buffer A consisting of 25 mM Tris-HCl (pH 7.4), 0.1% CHAPS, 50 mM NaCl, 0.1 mM PMSF, 1 µM pepstatin A, 0.1 mM benzamidine, 1 mM iodoacetamide. Actin from rabbit skeletal muscle (Sigma, Deisenhofen, Germany) was purified as described in ref 31
.
Affinity bead assays
Purified GST-ahnak-C2 fusion proteins (860 µg) were covalently coupled to 0.3 g BrCN-activated Sepharose 4B according to the manufacturer‘s protocol. The affinity beads were used in pull-down experiments with Ca2+ channels or actin. For control experiments, the equimolar concentration of unfused GST protein was coupled under the same conditions. Solubilized cardiac sarcolemma (1 mL) were diluted with 9 mL of buffer A and incubated with either ahnak-C2 affinity beads (100 µL swollen gel) or GST control beads (100 µL swollen gel) for 15 h at 4°C on a rotating wheel. WGA-purified skeletal muscle Ca2+ channels (500 µL) were incubated with affinity beads or control beads under the same experimental conditions. Tissue lysates (1 mg protein) were diluted (1:50) with buffer A to a final SDS concentration of 0.1% and incubated with the beads as above. Actin binding was performed using 15 µg purified G-actin in 200 µL G-actin buffer (31)
complemented with 800 µL of buffer A for 1 h at 4°C. After binding, the beads were washed three times with buffer A, resuspended in 70 µL 2x SDS-sample buffer, and incubated at 37°C for 30 min. Proteins released from the beads were separated on a 8% SDS-polyacrylamide gel, blotted to nitrocellulose, and analyzed by immunodetection according to standard procedures. The following sequence-directed antibodies were used: anti-1C (0.5 µg IgG/mL), anti-ß2 (0.5 µg IgG/mL), and anti-ßcom (1.2 µg IgG/mL). The 1S subunit was detected by a monoclonal antibody (1:100,000 dilution). Immunoreactive protein bands were visualized by the ECL reaction kit.
F-actin cosedimentation assays
F-actin cosedimentation assays were done essentially as described by the manufacturer (Cytoskeleton, Denver, CO). Recombinant ahnak protein preparations were incubated with 40 µg freshly polymerized actin (F-actin from striated muscle) for 30 min at room temperature. After incubation, the protein plus F-actin solution was subjected to centrifugation at 150,000 g for 1.5 h (Airfuge, Beckman) to pellet F-actin and protein bound to F-actin. After solubilization of the pellet fraction in a volume equal to the initial incubation volume, equal volumes of the pellet and supernatant fractions were analyzed by SDS-PAGE and Coomassie blue staining.
Ahnak antibodies
The recombinant ahnak fragments GST-ahnak-N and GST-ahnak-C2 were used for immunization of New Zealand White rabbits. The antibody-containing serum fractions were affinity purified on the respective columns of immobilized GST-ahnak-N or GST-ahnak-C2 fusion proteins. The resulting affinity-purified antibody fractions were depleted of anti-GST antibodies by chromatography on immobilized GST. For immunoblot analysis of ahnak, lysates (40 µg of protein) from frozen tissue specimens of human heart were separated on a 6.5% SDS-polyacrylamide gel and transferred to nitrocellulose for 2 h at 300 mA. The transfers were subsequently incubated with the affinity-purified ahnak antibodies at a concentration of 0.5–1 µg IgG/mL for 90 min at room temperature and the peroxidase-coupled anti-rabbit IgG (BioGenes, Berlin, Germany). Immunoreactive protein bands were visualized by the ECL reaction kit.
Immunohistochemistry
Left ventricular myocardium from a human donor heart not used for transplantation was frozen in liquid nitrogen and stored in -80°C until further use. Longitudinal and transversal sections 5 µm thick were cut using a cryostat (Leica CM3000) and fixed in 4% formalin for 10 min. Sections were blocked with 0.1% BSA for 15 min and incubated with the affinity-purified, region-specific ahnak antibodies raised against ahnak-N or ahnak-C2 (5 µg IgG/mL) at room temperature for 3 h. For the detection system, a biotinylated donkey anti-rabbit IgG (Dianova, Hamburg, Germany) in a concentration of 1:100 was applied for 1.5 h, followed by Cy2-conjugated streptavidin (Rockland, Gilbertsville, PA) at a dilution of 1:100 for 1 h. For double labeling experiments, vinculin (clone hVIN-1, Sigma, 1:40) or collagen (COL-94 Sigma, 1:100) were stained overnight, followed by an incubation step with rhodamine-conjugated goat anti-mouse IgG (Jackson ImmunoResearch, West Grove, PA; 1:50, 2 h). Nuclei were stained with toto 3 (Molecular Probes) at a dilution of 1:200 for 15 min. Between all staining steps, slides were rinsed in PBS, pH 7.4, three times for 3 min. Omission of primary antibodies served as negative controls. Tissue sections were mounted with Mowiol (Hoechst, Somerville, NJ) and protected with cover glasses.
Confocal Microscopy
Tissue sections were investigated using laser scanning confocal microscopy (Leica TCS SP). A series of confocal sections (0.5 µm interval) was taken through the specimen for consecutive 3-dimensional reconstruction at a Silicon Graphics Octane workstation using multichannel image processing software ‘Imaris’ and ‘Selima’ (Bitplane).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, suggesting a strong association. To search for interaction sites on both the ahnak and ß2a subunit, we tested a series of GST fusion proteins in overlay binding assays. Purified GST fusion proteins comprising amino acids (aa) 1–257 (ahnak-N), 3740–3882 (ahnak-R1), 3817–4412 (ahnak-R4), 4646–5288 (ahnak-C1), and 5262–5643 (ahnak-C2) of ahnak (Fig. 1
A) were blotted on nitrocellulose (Fig. 1B
) and incubated with recombinant [32P]-labeled ß2a subunit. The binding was analyzed by subsequent autoradiography. As demonstrated in Fig. 1C
, the ß2a subunit bound strongly to the most carboxyl-terminal region of ahnak (ahnak-C2). A weak interaction was detected between ahnak-C1 and [32P]-ß2a. No binding was observed for equivalent amounts of the GST fusion proteins ahnak-N, ahnak-R1, and ahnak-R4 or for the control GST protein alone.
|
Interaction of ahnak-C2 with ß2a truncation mutants
To identify regions in the ß2a protein responsible for ahnak-C2 interaction, we generated two GST fusion proteins that expressed the amino- and carboxyl-terminal portions of the ß2a subunit ß2a-N (aa 1–200) and ß2a-C (aa 195–606), respectively (Fig. 2
A). Overlay experiments revealed that truncation mutants ß2a-N and ß2a-C bound to ahnak-C2 (Fig. 2B
). The interaction between ahnak-C2 and ß2a-N appeared weaker than between ahnak-C2 and ß2a-C. In both cases, no interaction was detected with the control GST protein indicating specific interaction between ahnak-C2 and the ß2a subunit.
|
To investigate whether the ß interaction domain (BID) located in the second conserved domain of ß subunits (Fig. 2A
) is involved in ahnak-C2 interaction, we probed ahnak-C2/ß2a interaction in the presence of the synthetic 18 amino acid peptide resembling the cardiac AID to saturate the BID. Inclusion of 10 µM AID peptide did not prevent ahnak-C2 binding to both full-length ß2a and ß2a-C (data not shown), indicating that the BID region of ß2a does not participate in ahnak-C2 binding.
To quantify ahnak-C2/ß2a interaction, sedimentation equilibrium experiments were carried out and analyzed with respect to the dissociation constants (Kds) or amounts of complexes and free reactants. A prerequisite for such analysis is knowledge of the molecular masses of the reactants. The values obtained for recombinant ß2a and ahnak-C2 were 95 kDa and 65 kDa, respectively, indicating that both proteins are monomers in solution. Different mixtures consisting of 0.13 µM ß2a and variable amounts of ahnak-C2 were centrifuged to the sedimentation equilibrium and analyzed with respect to the complex formation. As shown in Fig. 3
A, the amount of complexes increased with growing recombinant ahnak-C2/ß2a ratio. An 1:1 complex was presumably formed up to the ahnak-C2/ß2a ratio of 2. When increasing the ratio to values of >2, more than one ahnak-C2 molecule was bound to the ß2a (Fig. 3A
), indicating a second binding site into the ß2a protein. The Kd value for the high-affinity binding site amounted to 53 ± 6 nM (n=9); the apparent Kd of the low-affinity binding site was approximated to be 200–300 nM. Subsequently, we analyzed truncated ß2a proteins with respect to their ahnak-C2 affinity. The ß2a-C appeared as a monomeric protein in solution with a molecular mass of 73 kDa. It bound ahnak-C2 with high affinity (Kd=55±11 nM, n=11) (Fig. 3B
), comparable to the high-affinity binding site of the full-length ß2a protein. In contrast, the amino-terminal fragment of ß2a (ß2a-N) formed a dimeric protein with a molecular mass of 97 kDa. The affinity was calculated with a Kd value of 328 ± 24 nM (n=7), (Fig. 3C
). Corresponding to the dimeric state, the truncation mutant ß2a-N bound more than one molecule of ahnak-C2.
|
Interaction of ahnak-C2 with ß subunits of native Ca2+ channels
To study whether the carboxyl-terminal ahnak portion mediates interaction to ß subunits of native Ca2+ channels, we performed pull-down assays with Ca2+ channel preparations from cardiac and skeletal muscle and ahnak-C2 that had been covalently coupled onto cyanogen bromide-activated Sepharose 4B (ahnak-C2 affinity beads). Bound proteins were subsequently released from the affinity beads and analyzed on immunoblots. Pig cardiac sarcolemma used as the starting material contained high levels of immunoreactive ß2 subunits (Fig. 4
, top, lane 1). Incubation of ahnak-C2 affinity beads with solubilized sarcolemmal membranes resulted in remarkable ß2 binding. The binding was specific, since GST control beads did not interact with ß2 subunits (Fig. 4
, top, lanes 2, 3).
|
We next addressed the question of whether the interaction between ahnak-C2 and ß2 subunit was ß isoform specific. A skeletal muscle Ca2+ channel preparation was used containing the ß1a isoform. As in the ß2 studies, we examined whether ahnak-C2 affinity beads would recover the ß1a isoform from the Ca2+ channel complex. Incubation with ahnak-C2 affinity beads, but not with the GST-control beads, resulted in the pull-down of a 50/55 kDa ß-immunoreactive double band that coincided with the ß-stained bands of the skeletal muscle Ca2+ channel preparation used as the starting material (Fig. 4
, lanes 4–6), indicating specific precipitation of the ß1a isoform. To determine whether the ß subunits remain associated with their cognate 1 subunits during this pull-down assay, the nitrocellulose transfers were stripped and reprobed with anti-1C and anti-1S antibodies. Figure 4
(bottom) illustrates that 1C and 1S indeed copurified with ß2 and ß1a, respectively.
Subcellular localization of ahnak in human cardiomyocytes
To gain insight into the subcellular organization of ahnak in human cardiomyocytes, we developed region-specific ahnak antibodies using the ahnak fragments ahnak-N and ahnak-C2. Antibody fractions were affinity purified and their specificity was examined and compared with that of the KIS antibody, a well-characterized peptide antibody against ahnak’s repeating units (3
4
5
6)
. On Western blots of total human cardiac proteins, the ahnak-N antibody and the ahnak-C2 antibody both reacted with a single protein of 
700 kDa (Fig. 5
), consistent with prominent KIS immunostaining and the predicted molecular mass of ahnak. Besides the 700 kDa full-length ahnak protein, the KIS antibody showed minor immunoreactivity with bands migrating at 500–700 kDa. These diffusely stained bands are probably proteolytic fragments derived from ahnak, since inclusion of a 100-fold molar excess of the antigenic peptide prevented immunostaining by KIS (data not shown). The results demonstrate that the newly developed, region-specific ahnak antibodies are useful tools for subcellular localization studies.
|
Figure 6
shows the labeling pattern of ahnak-C2 in representative examples of longitudinal (Fig. 6A
) and cross sections (Fig. 6B, C
) of normal human myocardium. Confocal imaging of cryosections revealed ahnak labeling along the entire surface sarcolemma (Fig. 6A, C
, asterisks), in the T-tubular system (Fig. 6
, small arrows), and at the level of intercalated disks (Fig. 6A
, big arrow). The fine T-tubular network appears as intracellular striations in longitudinal sections (Fig. 6A
) and as radially oriented stripes inside the myocytes in cross sections (Fig. 6B
), especially visible at high magnification (Fig. 6C
). Ahnak-C2 labeling was observed around the cardiomyocytes and vessels (Fig. 6B
, arrowheads).
|
Previous studies have demonstrated that vinculin, a cell-matrix focal adhesion protein, and collagen IV, a member of basement membrane proteins, serve as markers for T-tubular structures in human heart (32)
. The topology of ahnak-C2 was therefore further studied by laser scanning confocal microscopy in double labeling experiments. Longitudinal sections of human myocardium clearly show T-tubular structures stained by ahnak-C2 (Fig. 7
, arrows). The intercalated disks and the lateral sarcolemma are marked continuously by ahnak-C2 but not by vinculin. Vinculin is associated with T-tubules (Fig. 7C
, arrowheads), with the sarcolemma at the intercalated disks, and with costameres at the lateral sarcolemma. Colocalization of ahnak-C2 (Fig. 7A
, green) and vinculin (Fig. 7A
, red) was observed at the costameres and at the intercalated disks. T-tubules are stained for vinculin, located at the cytoplasmic side of the T-tubular membrane, and ahnak-C2, which is located in the sarcolemma flanking vinculin in double bands. A longitudinal section of human myocardium stained for ahnak-C2 and collagen IV demonstrate that both proteins are localized at the T-tubular system (Fig. 7D-F
). Collagen is continuously distributed at the outside of the sarcolemma and of the T-tubules whereas ahnak-C2 is present in the sarcolemma membrane, in the T-tubular system, and at the intercalated disk level. Intercalated disks are not stained by collagen IV (Fig. 7F
). High magnification of human myocardium in longitudinal section (Fig. 7G-I
) stained for ahnak-C2 and collagen IV clearly shows that collagen IV is peripherally located to ahnak-C2. The oblique line in the middle is due to the lateral membranes of two myocytes; T-tubules marked for both proteins emerge from this lateral membrane.
|
Similar to ahnak-C2 staining, ahnak-N labeling was observed at the lateral sarcolemma, at T-tubular structures, and at the intercalated disk level in sections of normal human myocardium (data not shown).
Ahnak-C2 binds to G-actin and F-actin in vitro
To identify potential novel interaction partners for the carboxyl-terminal ahnak portion in the striated muscle, we used ahnak-C2 affinity beads and GST control beads in pull-down assays with muscle protein preparations solubilized from heart and skeletal muscle of different species. Associated proteins were released from the beads, separated by SDS-PAGE, and stained with Coomassie brilliant blue. Figure 8
A illustrates a typical experiment using a rat cardiac preparation. Obviously, the 200 kDa myosin heavy chain band (MyHC) was retained by ahnak-C2 affinity beads and GST control beads. Thus, MyHC binding was considered to be nonspecific. Among the proteins with higher affinity to ahnak-C2 beads than to control beads, there was a prominent 40 kDa protein (Fig. 8A
, lane 3) that was subsequently identified as actin by immunoblotting (data not shown). The interaction between ahnak-C2 and actin was further investigated using ahnak-C2 affinity beads and purified muscle actin under conditions that preserve the globular actin form (G-actin). Indeed, specific binding of G-actin to immobilized ahnak-C2 could be detected which was nearly complete within 10 min at 4°C (Fig. 8B
). Moreover, direct binding of ahnak-C2 to actin was studied in an ELISA-based assay. We found reversible actin binding to wells coated with ahnak-C2 but not to those coated with GST or BSA (data not shown).
|
To test whether the carboxyl-terminal ahnak fragment bound directly to filamentous actin (F-actin), cosedimentation assays were performed using F-actin and ahnak-C2 (Fig. 8C
). In the absence of F-actin, ahnak-C2 was located solely in the soluble fraction (lane 1). However, in the presence of F-actin, 90% of ahnak-C2 was located in the pellet fraction (lane 4). -Actinin was used as positive control for F-actin sedimentation assays and BSA as negative control. As shown in Fig. 8C
, the known actin binding protein -actinin was recovered to 90% in the pellet fraction (lane 6) whereas BSA remained solely in the soluble fraction (lane 7), as determined by Coomassie blue staining of gels.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Binding assays and equilibrium sedimentation revealed specific, high-affinity ß2a binding for the most carboxyl-terminal ahnak region, designated as ahnak-C2 (aa 5262–5643) throughout this study. We found that 1 mol of ß2a subunit was able to bind 2 mol of ahnak-C2 with differing affinities in the submicromolar range. We dissected the interaction sites using ß2a truncation mutants and localized the high-affinity site (Kd50 nM) to the carboxyl-terminal (aa 195–606) ß2a portion; the binding site of lower affinity (Kd300 nM) resides in the amino-terminal (aa 1–200) portion of the ß2a subunit. All known ß subunits harbor two conserved domains that are flanked and interconnected by variable regions (33)
. The BID located in the second conserved domain (Fig. 2A
) binds to the AID within the I-II linker region of the 1 subunit (34
, 35)
. It has been suggested that this tight interaction with a Kd of 5 nM (36)
plays an important structural (37)
and modulatory role (18)
. We excluded the involvement of the BID in the interaction between ß2 subunit and ahnak-C2 by two independent approaches. First, we mimicked conditions that exist in native Ca2+ channels by inclusion of the 18 amino acid cardiac AID peptide in ß2a/ahnak-C2 binding assays. The cardiac AID peptide is able to bind strongly to ß subunits (18)
. Inclusion of the AID peptide did not prevent ß2a binding to ahnak-C2, indicating that the ß2a subunit exhibits independent interaction sites to ahnak-C2 and the 1C subunit. Thus, we propose a role for the cytoplasmic ß subunit as a tether between the carboxyl-terminal ahnak region and the channel-forming 1C subunit (Fig. 9
). Consistent with this peptide approach, we found that ahnak-C2 interacts with ß subunits of native Ca2+ channels, since ahnak-C2 affinity beads recovered cardiac ß2 subunits and skeletal muscle ß1a subunits together with the corresponding 1C and 1S proteins. These results underline the significance of ß subunits in tethering the 1 subunit of L-type Ca2+ channels to ahnak and suggest that 1) that the conserved domains of the ß subunits are critical for binding and 2) this interaction may be widespread within the Ca2+ channel family.
|
Confocal microscopy of human left ventricular tissue revealed localization of ahnak at the sarcolemma, including the T-tubular system and the intercalated disks of cardiomyocytes. This labeling pattern was obtained using two region-specific ahnak antibodies raised against either amino-terminal or carboxyl-terminal epitopes of the 700 kDa ahnak protein. In a previous study, we observed a similar plasma membrane location of ahnak in rat cardiomyocytes using a sequence-directed antibody against the internal repeating units (6)
. Thus, immunofluorescence data strongly suggest the association of the full-length ahnak protein with cardiac sarcolemma. This notion seems important, since the proposed 1.2 µm rod-like structure of ahnak (2)
could theoretically span the distance between different cellular compartments.
Plasma membrane location of ahnak was also observed in normal human skin (5)
and mouse keratinocytes (4)
. However, ahnak was predominantly cytoplasmic in epithelial cells (1
, 4
, 5)
and it was preferentially nuclear in nonepithelial cells (2
, 3)
. This puzzling subcellular distribution can be explained by ahnak’s potential for translocation. For instance, in epithelial cells, an increase in extracellular Ca2+ or stimulation of PK-C can induce the translocation of ahnak from cytosol to the plasma membrane (8)
. The region of ahnak that regulates its localization was shown to be contained within the carboxyl terminus of the protein (10)
. Shtivelman’s group (38)
identified very recently the formation of cell–cell contacts as a trigger for plasma membrane shuttling of ahnak from its nuclear localization in two epithelial cell lines. Nuclear exclusion of ahnak is mediated through a nuclear export signal that involves the phosphorylation of serine residue 5335 by PK-B. The authors speculate that the plasma membrane-anchored ahnak may be involved in the growth arrest of normal epithelial cells (38)
. Cardiomyocytes are terminally differentiated cells that withdraw from the cell cycle during early postnatal development. Considering that ahnak’s function may be determined by subcellular colocalization, we propose that ahnak targeted to the sarcolemma membrane of cardiomyocytes constitutes a membrane support protein. We observed a uniform distribution of ahnak at the plasma membrane. At the outer side of the lateral plasma membrane and the T-tubules, ahnak was surrounded by collagen IV, a basement membrane protein known to contribute to the outer scaffold of cardiomyocytes. Double labeling experiments with vinculin revealed an intimate association of ahnak with the plasma membrane. Vinculin, a member of the cell matrix focal adhesion protein, was found to colocalize with ahnak at the costameres at the lateral sarcolemma. Vinculin, which belongs to the inner scaffold of T-tubular structures, flanked ahnak at the cytoplasmic side. From this location, ahnak could constitute a plasma membrane protein. However, structure prediction models do not reveal the existence of transmembrane domains. Rather, the recurrent central motifs resemble a ß strand and are thought to form a polyionic rod with hydrophobic residues facing inward and charged residues facing outward (2)
. Membrane attachment may be achieved by interaction with other plasma membrane support proteins or the lipid bilayer.
A key finding of the present study is the identification of actin binding site(s) within the ahnak protein. In vitro binding of G-actin and cosedimentation with F-actin were observed with recombinant ahnak fragments representing the carboxyl-terminal region, but not with the repeating units of the large internal rod structure (H. Haase et al., unpublished data). A database search gave no homology to one of the known actin binding motifs. Rather, nine potential N-myristoylation sites were identified along with six PK-C phosphorylation sites and seven potential phosphorylation sites for casein kinase II (Prosite, EMBL). In the context of the present study, ahnak N-myristoylation is of interest, since this protein modification promotes membrane association. However, proteolytic cleavage would be a prerequisite to expose the ‘hidden’ myristoylation motifs.
Subcellular localization and interaction studies presented here formed the basis for a model of a sarcolemmal ahnak complex. The key feature of this model is that ahnak provides a link between the subsarcolemmal actin-based cytoskeleton and the ß2 subunit of voltage-dependent, L-type Ca2+ channels (Fig. 9)
. This interaction model resembles in some respects the dystrophin-associated protein complex that plays a fundamental role in maintenance of the structural integrity of muscle. Dystrophin is a 427 kDa protein attached to the F-actin cytoskeletal network through its amino-terminal region, whereas carboxyl-terminal domains mediate binding to a group of transmembrane and subsarcolemmal proteins, referred to as a dystrophin-associated protein complex (39)
. Syntrophin, a member of this complex, binds to skeletal muscle Na+ channels and determines its clustering in neuromuscular junctions (40)
. The ß subunit can be considered a cytoplasmic peripheral adaptor, similar to syntrophin, that couples the membrane-spanning 1 subunit to ahnak at the cytoplasmic side of cardiac sarcolemma. Since the ß subunit has two ahnak binding sites, it could anchor two carboxyl-terminal ahnak regions. Besides this proposed connection, ahnak possesses coiled-coiled regions within the repeating units and a leucine zipper in the carboxyl terminus (7)
.
In addition to this structural role, we believe that the actin–ahnak complex is involved in signaling events. The actin cytoskeleton has been implicated in the regulation of cation channels, including the epithelial Na+ channels (eNaC; 41
) and anion channels such as the cystic fibrosis transmembrane conductance regulator (CFTR; 42
). The actin cytoskeleton has been shown to confer PKA responsiveness to the reconstituted eNaC and to CFTR (41
, 42)
. In cardiac tissue, the proper regulation of Ca2+ channels is critically dependent on components of the cytoskeleton. In fact, both microtubule and actin filament disruption affects L-type Ca2+ channel inward current. Microtubule disassembly in rat cardiomyocytes leads to an increase in L-type Ca2+ channel activity accompanied by a blunted ß-adrenergic responsiveness of the channel (43)
. In contrast, actin filament disruption greatly reduced L-type Ca2+ channel current in myocytes of smooth (44)
and cardiac muscle (15)
. The latter study showed that actin filament stabilization was achieved by a genetic lack of the actin-severing protein gelsolin or that intracellular application of phalloidin leads to an increase in Ca2+ channel activity. We therefore suggest that 1) ahnak is part of a subsarcolemmal stabilizing network and 2) part of an intracellular signaling complex that confers the regulatory role of the cytoskeletal actin filaments to the L-type Ca2+ channels.
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 16, 2001.
Revision received April 1, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 in the presence of arachidonic acid. J. Biol. Chem. 274,13900-139071C and ß subunit isoforms. J. Cell. Biochem. 76,695-703[CrossRef][Medline]
1 subunit (1C-b) by the ß2a subunit: a peptide which inhibits binding of ß to the I-II linker of 1 induces functional uncoupling. Biochem. J. 348,657-665
domains from the human RNA editing enzyme ADAR1. FEBS Lett. 458,27-31[CrossRef][Medline]
1-ß anchoring site in voltage-dependent Ca2+ channels. J. Biol. Chem. 270,12056-20611C subunit are independent functions of the ß subunit. J. Physiol. (London) 517,353-368This article has been cited by other articles:
|
|
 |
|
