J. Biochem. 112, 235-242 (1992)
Structural Characterization of the Dihydropyridine Receptor-Linked Calcium Channel from Porcine Heart1 Akihiko Kuniyasu,* Eozo Oka,* Tomoko Ide-Yamada,* Yasumaru Hatanaka,* Teruo Abe,** Hitoshi Nakayama,*4 and Yuichi Kanaoka 0
Received for publication, March 6, 1992
Ca2+-cbannel was purified 230-fold from digitonin extracts of the porcine cardiac sarcolemmal membranes by means of a four-step procedure. Two antibodies, a site-directed antibody against the sequence 1691-1707 of the rabbit cardiac a\ subunit (anti-CCP5) and a monoclonal antibody directed to rabbit skeletal muscle a2d subunit-complex (MCC-1), effectively immunoprecipitated the '"I-labeled cardiac Ca2+-channel complex in 0.2% digitonin. SDS-PAGE analysis of the immunoprecipitates under reducing conditions revealed that the cardiac channel is mainly composed of two large polypeptides of 190 and 150 kDa, and five smaller polypeptides of 60, 55, 35, 30, and 25 kDa. An additional polypeptide of either 79 or 55 kDa is crosslinked with the 190 kDa component to form 250270 kDa (~270kDa) to the extent of 15-20% through disulfide bond(s). The 190 kDa component (al) is responsible for photoaffinity labeling with [3H]diazipine, since minor photolabeled ~270kDa was converged to the major labeled 190 kDa component when electrophoresed under reducing conditions. The 150 kDa component (a2) was derived by reduction of disulfide bonds from another 190 kDa component of glycopolypeptide which was separated from the channel complex in 1% Triton X-100 and capable of binding to WGA-Sepharose. The four smaller components of 60,35,30, and 25 kDa were not covalentiy associated with the large components through disulfide bonds, whereas the 55 kDa polypeptide was suggested to be a mixture of two kinds of peptides with respect to the disulfide bond: one was crosslinked with al through disulfide linkage and the other was not covalentiy associated with any other component. None of the polypeptides in the purified preparation or the sarcolemmal membranes which were phosphorylated by the cAMPdependent protein kinase, was immunoprecipitated by either of the antibodies.
Voltage-dependent Ca2+-channels in excitable cells play essential roles in many cellular functions. Four types of Cal+-channels at least, namely L, T, N, and P, have been distinguished on the basis of their electrophysiological and pharmacological properties (for a review, see Ref. 1). In the L-type CaI+-channels, cardiac channels show higher affinity (reviewed in Ref. 2) than those from skeletal muscles, brains, or smooth muscles toward 1,4-dihydropyTidines (DHPs), one of the typical ligands for the L-type Ca2+channels, suggesting that cardiac DHP receptor of the Ca1+-channel is different from those of other tissues.
Indeed, this has been proved by recent results of cDNA cloning in rabbit heart (3), skeletal muscle (4), smooth muscle (5, 6), and brain (7). It is believed that L-type Ca2+-channels are composed of several different polypeptides including the DHP receptor, like the well-characterized skeletal muscle channel which is composed of al, a2, p, y, and tfl-3 (8, 9). It was reported that immunologically related +-channel. Samples were treated either with 20 mM NEM (non-reducing conditions, odd-numbered lanes) or with 20 mM DTT (reducing conditions, even-numbered lanes) before being separated by SDS-PAGE using a 5-15% gradient gel and stained with silver: cardiac sarcolemmal membranes (3 n%, lanes 1 and 2), the first heparin-agarose column pool (1.5 ftg, lanes 3 and 4), and the second heparin-agarose column pool (0.5//g, lanes 5 and 6). The skeletal muscle Ca*+-channel (0.3 //g) which was purified according to the method described (32) was analyzed on the same gel for comparison.
1500 "D
ss
1000 -
8 5 8
I I2 H-
f 2
500 •
Slice Number Fig. 2. Photoaffinity labeling of cardiac DHP receptor. Purified cardiac Cal+-channel preparations which were prelabeled with [*H]diazipine were photolabeled as described in 'MATERIALS AND METHODS." Irradiated samples (15 ng protein) were treated with either 20 mM NEM (non-reducing conditions, • ) or 20 mM DTT (reducing conditions, O) and electrophoresed on a 5-15% gradient polyacrylamide gel. After SDS-PAGE, gels were stained with Coomasaie Brilliant Blue. Individual gel lanes were cut into 3 mm slices which were solubilized with 30% H,O! and counted for radioactivity. Photolabeled 270- and 190-kDa bands under non-reducing conditions (filled arrows) and 190 kDa band under reducing conditions (open arrow) are indicated. A£ standards are also shown (myoain, 200 kDa; /3-galactosidase, 116 kDa; phosphorylase b, 97 kDa; bovine serum albumin, 66 kDa; ovalbumin, 45 kDa).
DHP receptor deduced from the cDNA sequence (3). The antibody immunoprecipitated the [sH]PN200-110-binding component of the porcine cardiac Ca2+-channel in a dosedependent manner, up to 70% of the applied radioactivity (Fig. 3). The immunoprecipitation was sequence specific, as it was blocked to the control level in the presence of antigen peptide CCP5. The other antibody is a monoclonal antibody (MCC-1) specific for ar2cfsubunit complex of Ca2+-channels
10
10'
10'
Ig G Log [ pg ]
Fig. 3. Immunoprecipitation activity of anti-CCP6 antibody against solnblllzed cardiac DHP receptors prelabeled with [*H]PN200-110. With antibody beads which were prepared by addition of various amounts of purified IgG (4-200 /ig) to protein ASepharose (25 //I gel), solubilired and ['H]PN200-110-boundcardiac DHP receptors (3,000 dpm) were incubated in the absence (•) or presence (O) of peptide CCP5 (10 ^M). After incubating for 2 h at 4'C and washing with TBS-0.2% digitonin, the radioactivity of the beads was determined. As a control, non-immune IgG (A) was similarly treated and measured.
from bovine heart as well as skeletal muscle and brain, as reported previously (12). (1) Immunoblot analysis: The purified cardiac Ca2+-channel was first investigated by immunoblot analysis. As shown in Fig. 4A, anti-CCP5 antibody recognized polypeptides of 190 and ~270 kDa in approximately 4:1 ratio under the non-reducing conditions (lane 1). Two polypeptides were specifically reacted with the anti-CCP5, since the immunoblot was blocked with peptide CCP5. After reduction of disulfide bonds, the immunoblotted band of ~ 270 kDa diminished while the 190 kDa and a new band of 66 kDa were immunoblotted (lane 2). The 66 kDa band immunoreacted nonspecifically, as it was observed even in the presence of the competitive peptide CCP5 (lane 4). The results demonstrate that the polypeptides 190 kDa and minor ~ 270 kDa, which is converged to 190 kDa by reduction, possess sequences homologous to those of rabbit DHP receptor of the cardiac CaJ+-channel. Therefore, the 190 kDa polypeptide is considered to be the DHP receptor of porcine cardiac CaI+-channel. Similar experiments with anti-CCP5 were performed for Ca1+-channels from rabbit skeletal muscle, since the cardiac sequence CCP5 is almost identical to the corresponding region of the skeletal muscle channel (3). Figure 4B shows that the size-unchanged band of 170 kDa is specifically immunoreacted with the antibody under non-reducing and reducing conditions. Apparently the major immunoreactive 190 kDa peptide of cardiac DHP receptor is of higher molecular muss than the corresponding skeletal muscle counterpart. The cardiac 190 kDa polypeptide can be termed al subunit by analogy with the skeletal muscle 170 kDa. The purified cardiac preparation was also analyzed by immunoblotting using the monoclonal antibody MCC-1. Figure 4A (lane 5) shows that MCC-1 reacted with 190 kDa band under non-reducing conditions. After reductive treatment of the purified cardiac preparation, MCC-1 could not react with any polypeptides (lane 6) as expected from the J. Biochem.
Downloaded from https://academic.oup.com/jb/article-abstract/112/2/235/784340 by Tulane University Medical Library user on 23 January 2019
116 — 97 —
X
219
Dihydropyridine Receptor-Linked Calcium Channel
A ) Porcine Cardiac
B) Rabbit Skeletal
Origin—
«-> 200— o
-270 190
-170
* lies'^ 9 7 66 — 45 — Dye—
1 2 3 4 1 2 3 4 5 6 Fig. 4. Immnnoblot analysis of the purified cardiac (A) and skeletal muscle (B) Ca2+-channel preparations. The purified preparations from either porcine hearts (A) or rabbit skeletal muscles (B) were electrophoresed on 4-12% gradient polyacrylamide gels under non-reducing (odd-numbered lanes) or reducing conditions (even-numbered lanes), followed by electrotransfer onto an Immobilon-P membrane, as described in 'MATERIALS AND METHODS." The membrane was immunoblotted with anti-CCP5 antibody (lanes 1 and 2, in A and B), anti-CCP5 antibody plus 10^M peptide CCP5 (lanes 3 and 4, in A and B), or MCC-1 (lanes 5 and 6 in A) and stained with ['"I]protein A to make autoradiograms. Note that the band 66 kDa under reducing conditions (lane 2 in A) was immunoblotted nonspedfically by anti-CCP5 antibody, as it was also observed in the presence of competitive peptide CCP5 (lane 4 in A).
immunoprecipitated in the presence of competitive peptide CCP5 (lane 3). Under reducing conditions, three new bands of 150, 79, and 66 kDa appeared and the 55 kDa band increased, while the ~270 kDa band diminished (lane 2). Immunoprecipitation of the 66 kDa peptide was nonspecific, since it was also observed in the presence of CCP5 peptide (lane 4). Almost identical results were obtained with antibody MCC-1 (Fig. 5B), except that several additional nonspecific bands between 57 and 30 kDa were observed under reducing conditions (Fig. 5B, lanes 2 and 4). Although the 30 kDa band increased greatly after reduction (lane 2), it also increased in the preimmune serum (lane 4). We, therefore, did not take it into account as a reductiongenerated component. By surveying the polypeptides which are commonly observed in the immunoprecipitates with the two antibodies, the polypeptides of -270,190, 60, 55, 35, 30, and 25 kDa under non-reducing conditions and those of 190, 150, 79, 60, 55, 35, 30, and 25 kDa under reducing conditions, are considered as components of the cardiac Ca2+-channel. al and a2 Subunits—In rabbit skeletal muscles, al subunits are associated with a 2 subunits in a digitonin solution, but separate from each other in a Triton X-100 solution (8). It has also been demonstrated that the a2 subunit is heavily glycosylated and binds to lectins, such as WGA (8). We examined the mode of association of al and a2 subunits in porcine cardiac muscles. Figure 5C (lanes 14) shows the polypeptide composition of the immunoprecipitates of the Triton X-100-treated Ca2+-channel by the two antibodies. Anti-CCP5 antibody, which is a l subunit sequence-directed, immunoprecipitated —270 and 190 kDa peptides under non-reducing conditions, but 190, 79, and 55 kDa peptides were immunoprecipitated under reducing conditions (lanes 1 and 2). On the other hand,
Structural Characterization of the Dihydropyridine Receptor-Linked Calcium Channel from Porcine Heart1 Akihiko Kuniyasu,* Eozo Oka,* Tomoko Ide-Yamada,* Yasumaru Hatanaka,* Teruo Abe,** Hitoshi Nakayama,*4 and Yuichi Kanaoka 0
Received for publication, March 6, 1992
Ca2+-cbannel was purified 230-fold from digitonin extracts of the porcine cardiac sarcolemmal membranes by means of a four-step procedure. Two antibodies, a site-directed antibody against the sequence 1691-1707 of the rabbit cardiac a\ subunit (anti-CCP5) and a monoclonal antibody directed to rabbit skeletal muscle a2d subunit-complex (MCC-1), effectively immunoprecipitated the '"I-labeled cardiac Ca2+-channel complex in 0.2% digitonin. SDS-PAGE analysis of the immunoprecipitates under reducing conditions revealed that the cardiac channel is mainly composed of two large polypeptides of 190 and 150 kDa, and five smaller polypeptides of 60, 55, 35, 30, and 25 kDa. An additional polypeptide of either 79 or 55 kDa is crosslinked with the 190 kDa component to form 250270 kDa (~270kDa) to the extent of 15-20% through disulfide bond(s). The 190 kDa component (al) is responsible for photoaffinity labeling with [3H]diazipine, since minor photolabeled ~270kDa was converged to the major labeled 190 kDa component when electrophoresed under reducing conditions. The 150 kDa component (a2) was derived by reduction of disulfide bonds from another 190 kDa component of glycopolypeptide which was separated from the channel complex in 1% Triton X-100 and capable of binding to WGA-Sepharose. The four smaller components of 60,35,30, and 25 kDa were not covalentiy associated with the large components through disulfide bonds, whereas the 55 kDa polypeptide was suggested to be a mixture of two kinds of peptides with respect to the disulfide bond: one was crosslinked with al through disulfide linkage and the other was not covalentiy associated with any other component. None of the polypeptides in the purified preparation or the sarcolemmal membranes which were phosphorylated by the cAMPdependent protein kinase, was immunoprecipitated by either of the antibodies.
Voltage-dependent Ca2+-channels in excitable cells play essential roles in many cellular functions. Four types of Cal+-channels at least, namely L, T, N, and P, have been distinguished on the basis of their electrophysiological and pharmacological properties (for a review, see Ref. 1). In the L-type CaI+-channels, cardiac channels show higher affinity (reviewed in Ref. 2) than those from skeletal muscles, brains, or smooth muscles toward 1,4-dihydropyTidines (DHPs), one of the typical ligands for the L-type Ca2+channels, suggesting that cardiac DHP receptor of the Ca1+-channel is different from those of other tissues.
Indeed, this has been proved by recent results of cDNA cloning in rabbit heart (3), skeletal muscle (4), smooth muscle (5, 6), and brain (7). It is believed that L-type Ca2+-channels are composed of several different polypeptides including the DHP receptor, like the well-characterized skeletal muscle channel which is composed of al, a2, p, y, and tfl-3 (8, 9). It was reported that immunologically related +-channel. Samples were treated either with 20 mM NEM (non-reducing conditions, odd-numbered lanes) or with 20 mM DTT (reducing conditions, even-numbered lanes) before being separated by SDS-PAGE using a 5-15% gradient gel and stained with silver: cardiac sarcolemmal membranes (3 n%, lanes 1 and 2), the first heparin-agarose column pool (1.5 ftg, lanes 3 and 4), and the second heparin-agarose column pool (0.5//g, lanes 5 and 6). The skeletal muscle Ca*+-channel (0.3 //g) which was purified according to the method described (32) was analyzed on the same gel for comparison.
1500 "D
ss
1000 -
8 5 8
I I2 H-
f 2
500 •
Slice Number Fig. 2. Photoaffinity labeling of cardiac DHP receptor. Purified cardiac Cal+-channel preparations which were prelabeled with [*H]diazipine were photolabeled as described in 'MATERIALS AND METHODS." Irradiated samples (15 ng protein) were treated with either 20 mM NEM (non-reducing conditions, • ) or 20 mM DTT (reducing conditions, O) and electrophoresed on a 5-15% gradient polyacrylamide gel. After SDS-PAGE, gels were stained with Coomasaie Brilliant Blue. Individual gel lanes were cut into 3 mm slices which were solubilized with 30% H,O! and counted for radioactivity. Photolabeled 270- and 190-kDa bands under non-reducing conditions (filled arrows) and 190 kDa band under reducing conditions (open arrow) are indicated. A£ standards are also shown (myoain, 200 kDa; /3-galactosidase, 116 kDa; phosphorylase b, 97 kDa; bovine serum albumin, 66 kDa; ovalbumin, 45 kDa).
DHP receptor deduced from the cDNA sequence (3). The antibody immunoprecipitated the [sH]PN200-110-binding component of the porcine cardiac Ca2+-channel in a dosedependent manner, up to 70% of the applied radioactivity (Fig. 3). The immunoprecipitation was sequence specific, as it was blocked to the control level in the presence of antigen peptide CCP5. The other antibody is a monoclonal antibody (MCC-1) specific for ar2cfsubunit complex of Ca2+-channels
10
10'
10'
Ig G Log [ pg ]
Fig. 3. Immunoprecipitation activity of anti-CCP6 antibody against solnblllzed cardiac DHP receptors prelabeled with [*H]PN200-110. With antibody beads which were prepared by addition of various amounts of purified IgG (4-200 /ig) to protein ASepharose (25 //I gel), solubilired and ['H]PN200-110-boundcardiac DHP receptors (3,000 dpm) were incubated in the absence (•) or presence (O) of peptide CCP5 (10 ^M). After incubating for 2 h at 4'C and washing with TBS-0.2% digitonin, the radioactivity of the beads was determined. As a control, non-immune IgG (A) was similarly treated and measured.
from bovine heart as well as skeletal muscle and brain, as reported previously (12). (1) Immunoblot analysis: The purified cardiac Ca2+-channel was first investigated by immunoblot analysis. As shown in Fig. 4A, anti-CCP5 antibody recognized polypeptides of 190 and ~270 kDa in approximately 4:1 ratio under the non-reducing conditions (lane 1). Two polypeptides were specifically reacted with the anti-CCP5, since the immunoblot was blocked with peptide CCP5. After reduction of disulfide bonds, the immunoblotted band of ~ 270 kDa diminished while the 190 kDa and a new band of 66 kDa were immunoblotted (lane 2). The 66 kDa band immunoreacted nonspecifically, as it was observed even in the presence of the competitive peptide CCP5 (lane 4). The results demonstrate that the polypeptides 190 kDa and minor ~ 270 kDa, which is converged to 190 kDa by reduction, possess sequences homologous to those of rabbit DHP receptor of the cardiac CaJ+-channel. Therefore, the 190 kDa polypeptide is considered to be the DHP receptor of porcine cardiac CaI+-channel. Similar experiments with anti-CCP5 were performed for Ca1+-channels from rabbit skeletal muscle, since the cardiac sequence CCP5 is almost identical to the corresponding region of the skeletal muscle channel (3). Figure 4B shows that the size-unchanged band of 170 kDa is specifically immunoreacted with the antibody under non-reducing and reducing conditions. Apparently the major immunoreactive 190 kDa peptide of cardiac DHP receptor is of higher molecular muss than the corresponding skeletal muscle counterpart. The cardiac 190 kDa polypeptide can be termed al subunit by analogy with the skeletal muscle 170 kDa. The purified cardiac preparation was also analyzed by immunoblotting using the monoclonal antibody MCC-1. Figure 4A (lane 5) shows that MCC-1 reacted with 190 kDa band under non-reducing conditions. After reductive treatment of the purified cardiac preparation, MCC-1 could not react with any polypeptides (lane 6) as expected from the J. Biochem.
Downloaded from https://academic.oup.com/jb/article-abstract/112/2/235/784340 by Tulane University Medical Library user on 23 January 2019
116 — 97 —
X
219
Dihydropyridine Receptor-Linked Calcium Channel
A ) Porcine Cardiac
B) Rabbit Skeletal
Origin—
«-> 200— o
-270 190
-170
* lies'^ 9 7 66 — 45 — Dye—
1 2 3 4 1 2 3 4 5 6 Fig. 4. Immnnoblot analysis of the purified cardiac (A) and skeletal muscle (B) Ca2+-channel preparations. The purified preparations from either porcine hearts (A) or rabbit skeletal muscles (B) were electrophoresed on 4-12% gradient polyacrylamide gels under non-reducing (odd-numbered lanes) or reducing conditions (even-numbered lanes), followed by electrotransfer onto an Immobilon-P membrane, as described in 'MATERIALS AND METHODS." The membrane was immunoblotted with anti-CCP5 antibody (lanes 1 and 2, in A and B), anti-CCP5 antibody plus 10^M peptide CCP5 (lanes 3 and 4, in A and B), or MCC-1 (lanes 5 and 6 in A) and stained with ['"I]protein A to make autoradiograms. Note that the band 66 kDa under reducing conditions (lane 2 in A) was immunoblotted nonspedfically by anti-CCP5 antibody, as it was also observed in the presence of competitive peptide CCP5 (lane 4 in A).
immunoprecipitated in the presence of competitive peptide CCP5 (lane 3). Under reducing conditions, three new bands of 150, 79, and 66 kDa appeared and the 55 kDa band increased, while the ~270 kDa band diminished (lane 2). Immunoprecipitation of the 66 kDa peptide was nonspecific, since it was also observed in the presence of CCP5 peptide (lane 4). Almost identical results were obtained with antibody MCC-1 (Fig. 5B), except that several additional nonspecific bands between 57 and 30 kDa were observed under reducing conditions (Fig. 5B, lanes 2 and 4). Although the 30 kDa band increased greatly after reduction (lane 2), it also increased in the preimmune serum (lane 4). We, therefore, did not take it into account as a reductiongenerated component. By surveying the polypeptides which are commonly observed in the immunoprecipitates with the two antibodies, the polypeptides of -270,190, 60, 55, 35, 30, and 25 kDa under non-reducing conditions and those of 190, 150, 79, 60, 55, 35, 30, and 25 kDa under reducing conditions, are considered as components of the cardiac Ca2+-channel. al and a2 Subunits—In rabbit skeletal muscles, al subunits are associated with a 2 subunits in a digitonin solution, but separate from each other in a Triton X-100 solution (8). It has also been demonstrated that the a2 subunit is heavily glycosylated and binds to lectins, such as WGA (8). We examined the mode of association of al and a2 subunits in porcine cardiac muscles. Figure 5C (lanes 14) shows the polypeptide composition of the immunoprecipitates of the Triton X-100-treated Ca2+-channel by the two antibodies. Anti-CCP5 antibody, which is a l subunit sequence-directed, immunoprecipitated —270 and 190 kDa peptides under non-reducing conditions, but 190, 79, and 55 kDa peptides were immunoprecipitated under reducing conditions (lanes 1 and 2). On the other hand,
