Proc. Natl. Acad. Sci. USA Vol. 74, No. 10, pp. 4685-4688, October 1977 Neurobiology
Affinity alkylation labels two subunits of the reduced acetylcholine receptor from mammalian muscle (muscle/affinity label/subunit structure)
STANLEY C. FROEHNER*t, ARTHUR KARLIN*, AND ZACH W. HALL§ * Department of Neurobiology, Harvard Medical School, Boston, Massachusetts 02115; * Department of Neurology, College of Ph sicians and Surgeons, Columbia University, New York, New York 10032; and § Department of Physiology, University of California, San Francisco, California 94143
Communicated by David H. Hubel, August 2,1977
ABSTRACT The acetylcholine receptor from denervated rat skeletal muscle was purified by affinity chromatography and, after reduction, was treated with the affinity alkylating agent 4(N-maleimido)benzyltri[3H]methylammonium iodide. The receptor specifically incorporated approximately 1 mol of alkylating agent per mol of "251-labeled a-bungarotoxin bound. Analysis of the labeled rece tor by polyacrylamide gel electrophoresis in sodium dodecyI sulfate showed that two subunits were labeled; their apparent molecular weights were 45,000 and 49,000. These results suggest that the affinity reagent labels a second site for acetylcholine binding in the muscle receptor that is not labeled in receptors from Electrophorus or Torpedo.
Receptors that bind the neurotransmitter acetylcholine (ACh) and mediate a change in ion permeability of the subsynaptic membrane have been isolated and purified from the electroplax tissues of the electric eel Electrophorus and the marine ray Torpedo. In both cases, the subunit composition of the purified receptors is complex, consisting of two to four different polypeptides (reviewed in ref. 1). One approach to the identification of the function of each of these subunits has been the use of affinity reagents directed against the ACh binding site associated with the ionic permeability change mediated by the receptor. One of these reagents, 4-(N-maleimido)benzyltri[3H]methylammonium iodide ([3H]MBTA), specifically alkylates a sulfhydryl group within about 1 nm of the binding site of the reduced Electrophorus receptor (2). This reagent binds to a single subunit in both Electrophorus and Torpedo receptors, thus identifying these subunits as bearing part or all of the ACh binding site. In each case, the subunit has an apparent molecular weight of approximately 40,000 and is the subunit of lowest molecular weight (3-5). Because of the small amounts of purified protein that can be obtained, little information is available concerning either the subunit structure or the ACh binding site(s) of the receptor from mammalian muscle. We report here the subunit composition of highly purified ACh receptor from denervated rat leg muscle and the affinity alkylation of this protein by [3H]MBTA. Our results suggest that two different subunits in the muscle receptor may bind ACh. MATERIALS AND METHODS
Assays. 125I-Labeled a-bungaratoxin (125I-a-BuTx) binding was determined by the DEAE-filter technique described by Brockes and Hall (6). Assays of a preparation of Torpedo receptor with 125I-a-BuTx and with [3H]methyltoxin 3 from Naja naja siamensis, prepared as described by Weill et al. (5), gave values within 15% of each other. Protein concentrations of the
purified muscle ACh receptor were determined by the method of Schultz and Wassarman (7). Purification of ACh Receptor from Denervated Rat Muscle. All procedures were performed at 0-4°. Lower leg muscles (100-150 g) of rats, denervated 10-20 days earlier by bilateral transection of the sciatic nerve, were homogenized in 700 ml of buffer A [50 mM Tris-HCl, pH 7.4/50 mM NaCI/1 mM (ethylenedinitrilo)tetraacetic acid (EDTA)/1 mM ethylene glycol-bis(fl-aminoethyl ether)-N,N'-tetraacetic acid (EGTA)/0. 1 mM phenylmethylsulfonyl fluoride (PMSF) containing Trasylol (Sigma Chemical Co.) at 20 units/ml, and Pepstatin at 0.1 ,ug/ml] in a Waring blender, passed through cheesecloth and centrifuged for 90 min at 40,000 X g. The pellets were resuspended in 100 ml of buffer A lacking EDTA and EGTA, and Triton X-100 was added to a final concentration of 1%. After incubation at 00 for 1 hr, insoluble material was removed by centrifugation for 45 min at 40,000 X g. The supernatant was adsorbed batchwise to 30 ml of concanavalin A (Con A)-Sepharose 4B [3 mg of Con A per ml Sepharose prepared by the cyanogen bromide activation technique (8)] by stirring at 40 for 90 min. The slurry was then poured into a glass chromatographic column. The column was sequentially washed with 50 ml of buffer B (50 mM Tris.HCI, pH 7.4/50 mM NaCI/0.1 mM PMSF/1% Triton), 100 ml of buffer B containing 1 M NaCl, and 50 ml of buffer B. Approximately 0.8 column volume of 0.4 M a-methylmannoside in buffer A containing 3% Triton was added to the column. After at least 24 hr, the column was eluted with 1 column volume of a-methylmannoside solution and the eluate was dialyzed overnight against 1 liter of buffer C (10 mM Tris-HCI, pH 7.4/1 mM EDTA/0.1 mM PMSF/1% Triton). A column of cobrotoxin-Sepharose 4B (6) (0.30 ml in a 1-ml plastic syringe) was washed with buffer B containing 1 M NaCl and equilibrated with buffer C. The Con A-Sepharose eluate was passed through the column twice at 5-8 ml/hr, which removed 80-90% of the toxin-binding activity from the solution. The column was then washed with 1 ml of buffer B, 1 ml of buffer B containing 1 M NaCl, and 1 ml of buffer B and then eluted for 30 min with 0.4 ml of 1 M carbamylcholine chloride in 50 mM Tris-HCI, pH 8.6/1 mM EDTA/1 mM EGTA/0.1 mM PMSF/1% Triton. Carbamycholine was removed by extensive dialysis against buffer D (0.2% Triton X-100/50 mM NaCI/10 mM sodium phosphate/I mM EDTA/3 mM NaN3, Abbreviations: ACh, acetylcholine; [3H]MBTA, 4-(N-maleimido)benzyltri[3H]methylammonium iodide; 125I-a-BuTx, 125-Ilabeled a-bungarotoxin; EDTA, (ethylenedinitrilo)tetraacetic acid; EGTA, ethylene glycol-bis(,B-aminoethyl ether)-NN'-tetraacetic acid; PMSF, phenylmethylsulfonyl fluoride; Con A, concanavalin A; NaDodSO4, sodium dodecyl sulfate.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate
t Present address:
Department of Neuroscience, Children's Hospital Medical Center, 300 Longwood Avenue, Boston, MA 02115.
this fact.
4685
4686 ' Neurobiology: Froehner et al.
pH 7.0). This procedure gives approximately 50,000-fold purification of the muscle ACh receptor and yields of 3-10%. [3H]MBTA Labeling. For determination of stoichiometric binding of [3H]MBTA to muscle and Torpedo receptors, the method of Karlin et al. (9) was used with minor changes. For experiments using sodium dodecyl sulfate (NaDodSO4)/ polyacrylamide gel electrophoresis, the procedure was modified as follows. A solution containing purified ACh receptor (50-100 pmol in 100-200 pAI) and, in some cases, PMSF and Pepstatin was incubated at 230 for 10 min (or, in one experiment, at 40 for 60 min) with an equal volume of 0.4 mM dithiothreitol in 0.2% Triton/150 mM NaCI/20 mM Tris-HCI, pH 8.3/1 mM EDTA/3 mM NaN3. One-tenth volume of 0.53 M sodium phosphate buffer, pH 6.7 was added, 400 ,l of 5 ,gM [3H]MBTA was added, and the solution was incubated for 1 min at 230. The sample was then chilled on ice and 20 ,Al 0.1 M 2-mercaptoethanol was added. In control experiments the receptor was incubated with cobrotoxin (4 ig/ml) after reduction. After addition of 50 jzg of polyaspartic acid, the sample was dialyzed against 1 liter of 10 mM Tris-HCI, pH 7.4/1% Triton for 1 hr at 4°. Protein was precipitated with 10 volumes of acetone at -20° for 15 min, collected by centrifugation, and redissolved in NaDodSO4/electrophoresis sample buffer (60,u) by boiling for 5 min. Electrophoresis and Fluorography. NaDodSO4/polyacrylamide gel electrophoresis was performed in a slab gel according to Laemmli (10) with minor modifications. The sample buffer contained 3.5% NaDodSO4 and 5% 2-mercaptoethanol and the acrylamide concentration in the separation gel was 9%. Standards used for molecular weight determinations were phosphorylase a (94,000), bovine serum albumin (68,000), pyruvate kinase (57,000), fumarase (49,000), aldolase (40,000), D-amino acid oxidase (37,000), and glyceraldehyde dehydrogenase (36,000). Because the polypeptide components of the ACh receptor are probably glycoproteins, the molecular weights determined by using these standards are provisional. Staining and destaining with Coomassie brilliant blue was performed according to Fairbanks et al. (11). After destaining, the gel was scanned with a densitometer or photographed and then prepared for fluorography as described by Bonner and Laskey
(12). RESULTS ACh receptor purified from denervated muscle by affinity chromatography on Con A-Sepharose and cobrotoxin-Sepharose had a specific activity of approximately 8-10 nmol of 125I-aBuTx bound per mg of protein, comparable to that of highly purified Electrophorus and Torpedo receptors (1). Analysis of the purified muscle receptor by NaDodSO4/gel electrophoresis showed the preparation to contain a number of different polypeptide chains of apparent molecular weights 45,000, 51,000, 56,000, 62,000, and 110,000 [Figs. 1 (lane a) and 2]. In addition, all preparations contained one or more polypeptides whose position corresponded to approximately 49,000 daltons. This component was not always clearly resolved and is seen in Fig. 2 as a shoulder(s) of the peak at 51,000 daltons. Analysis of preparations comparable to those in Figs. 1 and 2 indicates that over 90% of the protein corresponds to receptor (13). Further purification by sucrose density gradient centrifugation decreases but does not eliminate the 56,000 dalton component, and the relative amounts of the other polypeptides remain virtually constant (13). Initial experiments were performed with the method of Karlin et al. (9) to determine whether binding of [3H]MBTA
, t . * 'o
.,|' .l;
Proc. Nati. Acad. Sci. USA 74 (1977)
b
a
110
_
*
T
.23i N
:
t-
X :_,.
.,._
62 56
51
49 45
c
_
w
^
:::
.s
_
.s
s
: ;..X _i.: *
'f
: 'pS.a .S
42 39
FIG. 1. NaDodSO4/polyacrylamide gel electrophoresis of [3H]MBTA-labeled ACh receptor. Lanes: a, Coomassie brilliant bluestained gel of 2.3 Ag of ACh receptor; b, fluorogram of 9.2 gg of [3H]MBTA-labeled ACh receptor; c, Coomassie brilliant blue-stained gel of 9.2 of jg [3H]MBTA-labeled ACh receptor (same sample as in b). Reduction and alkylation were performed at 230 for 20 min and 10 min respectively, in the absence of protease inhibitors. Molecular weights (X10-3) are shown at the sides.
to the muscle receptor could be detected. Specific binding was defined as that which could be blocked by a-BuTx. The values obtained in one such experiment (Table 1) show that over 90% of the binding can be blocked by toxin. In three experiments with two different preparations of ACh receptor from muscle, the ratio of 125I-a-BuTx binding to specific binding of [3H]MBTA was 1.5, 1.1, and 0.5. Corresponding values in parallel experiments on purified Torpedo receptor were 1.8, 2.5, 2.0, and, in a fourth experiment, 2.0, confirming values previously found by Weill et al. (5). To determine which of the polypeptides in the muscle receptor bind the affinity reagent, purified receptor was reduced with dithiothreitol, incubated with [3H]MBTA, and analyzed by gel electrophoresis and fluorography. In an initial experiment [Fig. 1 (lane b)], bands of radioactivity were observed corresponding to molecular weights of 49,000, 45,000, 42,000, and 39,000. All of these bands were eliminated by prior incubation with purified a-neurotoxin from cobra venom. It appears, however, from a comparison of gels stained with Coomassie brilliant blue before and after the reduction and alkylation reactions [Fig. 1 (lanes a and c)], that proteolysis had occurred during the reaction, resulting in the appearance of new bands corresponding to 42,000 and 39,000 daltons. In two subsequent experiments in which reduction was performed in
Proc. Nati. Acad. Sci. USA 74 (1977)
Neurobiology: Froehner et al.
44687
TOP I
I
I
I
I
I
I
51,000 I
(I II I
45,000
DYE FRONT
!11'
I
FIG. 2. Densitometer tracings of Coomassie brilliant blue-stained (- - -) gel of alkylated receptor and fluorogram of [3H]MBTA-labeled (-) receptor. The same sample was used for both analyses. Alkylation was performed for 1 min at 230 after reduction for 10min at 230 in the presence of protease inhibitors.
the presence of protease inhibitors, either at 40 for 60 min or at 23° for 10 min, and the time of alkylation was decreased to 1 min, no change in subunit composition after labeling was detected by staining with Coomassie brilliant blue. In both of these experiments, we obtained two peaks of radioactivity. A parallel experiment with receptor from Torpedo gave only one labeled subunit as previously observed. The results of one of the experiments with the muscle receptor are shown in Fig. 2. One peak of radioactivity coincided with the 45,000 dalton subunit, and the other occurred at a position corresponding to 49,000 daltons. The latter peak was clearly distinguished from the 51,000 dalton component and appeared most closely aligned with the shoulder of that peak. Similar results were obtained in the other experiment, except that the peak of radioactivity at the higher molecular weight was not sharp enough to allow unequivocal assignment of either Table 1. Specific labeling of reduced muscle and Torpedo ACh receptors by [3HJMBTA [3H]- Toxin Ratio, MBTA bind- toxin bound 3H in ppt., bound, ing,t [3HJMBTA pmol pmol bound cpm Sample 1115 Buffer D Muscle ACh receptor 1200 plus toxin 1.1 5.3 Muscle ACh receptor 4091 4.9 Torpedo ACh receptor 1035 plus toxin Torpedo ACh 6.7 2.5 2642 2.7 receptor * Reduction and alkylation with [3HJMBTA and filtration were carried out according to Karlin et al. (9) as described in Materials and Methods. Each value represents the average of two determinations. t -251-Labeled toxin binding activity was determined as described previously (6).
the 49,000 or the 51,000 dalton component. The ratio of radioactivity in the 45,000 dalton subunit to that in the subunit of higher molecular weight was 1.0 in the experiment shown. In a second experiment, performed on the same preparation 3 weeks later, the ratio was 0.4. Because the paramount consideration in these two experiments was to prevent proteolysis, we cannot be sure that the conditions used resulted in complete labeling. Thus, these values cannot be used to infer relative amounts of MBTA-reactive sites in these two subunits. DISCUSSION MBTA is one of a group of affinity reagents designed to react with a sulfhydryl group near the active site of the reduced ACh receptor from Electrophorus electroplax. The reactive sulfhydryl group is thought to be near the active site because alkylation with the affinity reagents can be blocked by cholinergic agonists and antagonists and because certain alkylating agents such as bromoacetylcholine produce a state of permanent depolarization of the membrane that can be reversed only by high concentrations of d-tubocurarine (14, 15). Incubation of reduced eel ACh receptor, either in intact electroplax or after purification, with [3H]MBTA results in specific labeling of a single polypeptide chain that is thus thought to contain all or part of the active site (3, 4). [3H]MBTA has been shown to label a subunit of the same size in purified Torpedo receptor (5). The experiments reported here demonstrate that, after reduction, purified muscle receptor, like that of Electrophorus and Torpedo, can also be labeled with [3H]MBTA. The reaction is highly specific because it is decreased more than 90% by a-bungarotoxin. These results are consistent with physiological experiments showing that there is a disulfide bond near the active site of vertebrate muscle receptors that can be reduced and alkylated (16, but see also 22). The ability of [3H]MBTA to label the active site of the reduced receptor from such disparate species as rat and marine ray suggests that at least part of the receptor structure has been strongly conserved throughout evolution. Our results differ in two ways from those found with the fish receptors. First, [3H]MBTA labels two polypeptide chains of
4688
Neurobiology: Froehner et al.
apparent molecular weights 45,000 and 49,000. Second, the stoichiometry of the binding of [3H]MBTA relative to 125I-aBuTx binding activity appears to be different. Both fish receptors bind exactly twice as much a-neurotoxin as they do MBTA (1), whereas our experiments suggest that the muscle receptor binds equivalent amounts of [3H]MBTA and 125I-aBuTx. The small amounts of muscle receptor available and the corresponding imprecision in the results make this conclusion a tentative one. The specific activity (mol of 125I-BuTx per mg of protein) of the purified muscle preparations (13) is similar to that reported for Torpedo, and the physical properties of the muscle receptor suggest that it has a molecular weight similar to that of Torpedo (see ref. 1). Thus, our results lead to the suggestion that the muscle receptor binds twice as much [3H]MBTA as does the receptor from Torpedo. The observation that [3H]MBTA labels two subunits of the muscle receptor could be explained in several ways. One possible explanation is that the 45,000 dalton subunit is derived from the larger one by proteolysis; alternatively, both may be derived from an even larger common precursor. It is difficult to exclude this possibility except to note that, under proper conditions, proteolysis during reduction and alkylation can be prevented (Fig. 2). Thus, any proteolytic conversion must have occurred during extraction and purification of the receptor. In addition, it is unlikely that either of the labeled subunits could be derived from the 51,000 dalton component because, in the experiments shown in Figs. 1 and 2, this component is devoid of radioactivity. Another possibility is that [3H]MBTA labels two separate sites in the muscle receptor, only one of which corresponds to that labeled in the fish receptors. Because binding to both subunits in the muscle protein is blocked by toxin, the two sites would presumably correspond to different toxin binding sites as well. Evidence for two classes of toxin-binding sites has been found both in muscle and Torpedo receptors (6, 18), although in the fish receptors it has thus far been possible to identify only one subunit that binds a-neurotoxin (5, 19, 20). Two classes of binding sites for ACh have been observed for the Torpedo receptor (18, 21, 22). A final possibility is that [3H]MBTA labels two classes of receptor molecules, each with a single binding site. It is unlikely that these could correspond to junctional and extrajunctional receptors because isoelectric focusing analysis indicates that the receptor preparation studied in these experiments is at least 90%
extrajunctional (N. Nathanson, unpublished data). Further
Proc. Natl. Acad. Sci.USA 74 (1977) experiments will be required to distinguish the various alternatives. We thank Steve Rowe for expert technical assistance, Joe Gagliardi for help with the photography, and Dr. Regis Kelly for helpful discussions. This work was supported by National Institutes of Health Grants NS 09646 and NS 07065, National Science Foundation Grant BMS 75-03026, and grants from the Muscular Dystrophy Association. S.F. was a Muscular Dystrophy Association Fellow. 1. Karlin, A. (1977) Excerpta Medica, in press. 2. Karlin, A., Prives, J., Deal, W. & Winnik, M. (1971) J. Mol. Biol., 61, 175-188. 3. Reiter, M. J., Cowburn, D. A., Prives, J. M. & Karlin, A. (1972) Proc. Nati. Acad. Sci. USA 69,1168-1172. 4. Karlin, A. & Cowburn, D. A. (1973) Proc. Natl. Acad. Scd. USA
70,3636-3640.
5. Weill, C. L., McNamee, M. G. & Karlin, A. (1974) Blochem. Blophys. Res. Commun. 61, 997-1003. 6. Brockes, J. & Hall, Z. W. (1975) Biochemistry 14,2092-2099. 7. Schultz, R. M. & Wassarman, P. M. (1976) Anal. Biochem. 77, 25-32. 8. Cuatrecasas, P. (1970) J. Blol. Chem. 245,3059-3065. 9. Karlin, A., McNamee, M. G. & Cowburn, D, A. (1976) Anal. Biochem. 76,442-451. 10. Laemmli, U. K. (1970) Nature 227,680-685. 11. Fairbanks, G., Steck, T. L. & Wallach, D. F. H. (1971) Biochemistry 10, 2606-2617. 12. Bonner, W. M. & Laskey, R. A. (1974) Eur. J. Blochem. 46, 83-88. 13. Froehner, S. C., Reiness, C. G. & Hall, Z. W. (1977) J. Biol. Chem., in press. 14. Karlin, A. (1969) J. Gen. Physiol. 54, 245S-264S. 15. Silman, H. I. & Karlin, A. (1969) Science 164, 1420-1421. 16. Ben-Haim, D., Landau, E. M. & Silman, I. (1973) J. Physiol. (London) 234, 305-325. 17. Raftery, M. A., Vandlen, R. L., Reed, K. L. & Lee, T. (1975) Cold Spring Harbor Symp. Quant. Blol. XL, 193-202. 18. Biesecker, G. (1973) Biochemistry 12,4404-4409. 19. Gordon, A., Bandini, G. & Hucho, F. (1974) FEBS Lett. 47, 204-208. 20. Eldefrawi, M. E., Eldefrawi, A. T. & Shamoo, A. E. (1975) Ann. N.Y. Acad. Sci. 264,183-202. 21. Changeux, J.-P., Benedetti, L., Bourgeois, J.-P., Brisson, A., Cartaud, J., Devaux, P., Grunhagen, H., Moreau, M., Popot, J.-L., Sobel, A. & Weber, M. (1975) Cold Spring Harbor Symp. Quant. Biol. XL, 211-230. 22. Lindstrom, J. M., Singtr, S. J. & Lennox, E. S. (1973) J. Membr. Biol. 11,217-226.
a
Affinity alkylation labels two subunits of the reduced acetylcholine receptor from mammalian muscle (muscle/affinity label/subunit structure)
STANLEY C. FROEHNER*t, ARTHUR KARLIN*, AND ZACH W. HALL§ * Department of Neurobiology, Harvard Medical School, Boston, Massachusetts 02115; * Department of Neurology, College of Ph sicians and Surgeons, Columbia University, New York, New York 10032; and § Department of Physiology, University of California, San Francisco, California 94143
Communicated by David H. Hubel, August 2,1977
ABSTRACT The acetylcholine receptor from denervated rat skeletal muscle was purified by affinity chromatography and, after reduction, was treated with the affinity alkylating agent 4(N-maleimido)benzyltri[3H]methylammonium iodide. The receptor specifically incorporated approximately 1 mol of alkylating agent per mol of "251-labeled a-bungarotoxin bound. Analysis of the labeled rece tor by polyacrylamide gel electrophoresis in sodium dodecyI sulfate showed that two subunits were labeled; their apparent molecular weights were 45,000 and 49,000. These results suggest that the affinity reagent labels a second site for acetylcholine binding in the muscle receptor that is not labeled in receptors from Electrophorus or Torpedo.
Receptors that bind the neurotransmitter acetylcholine (ACh) and mediate a change in ion permeability of the subsynaptic membrane have been isolated and purified from the electroplax tissues of the electric eel Electrophorus and the marine ray Torpedo. In both cases, the subunit composition of the purified receptors is complex, consisting of two to four different polypeptides (reviewed in ref. 1). One approach to the identification of the function of each of these subunits has been the use of affinity reagents directed against the ACh binding site associated with the ionic permeability change mediated by the receptor. One of these reagents, 4-(N-maleimido)benzyltri[3H]methylammonium iodide ([3H]MBTA), specifically alkylates a sulfhydryl group within about 1 nm of the binding site of the reduced Electrophorus receptor (2). This reagent binds to a single subunit in both Electrophorus and Torpedo receptors, thus identifying these subunits as bearing part or all of the ACh binding site. In each case, the subunit has an apparent molecular weight of approximately 40,000 and is the subunit of lowest molecular weight (3-5). Because of the small amounts of purified protein that can be obtained, little information is available concerning either the subunit structure or the ACh binding site(s) of the receptor from mammalian muscle. We report here the subunit composition of highly purified ACh receptor from denervated rat leg muscle and the affinity alkylation of this protein by [3H]MBTA. Our results suggest that two different subunits in the muscle receptor may bind ACh. MATERIALS AND METHODS
Assays. 125I-Labeled a-bungaratoxin (125I-a-BuTx) binding was determined by the DEAE-filter technique described by Brockes and Hall (6). Assays of a preparation of Torpedo receptor with 125I-a-BuTx and with [3H]methyltoxin 3 from Naja naja siamensis, prepared as described by Weill et al. (5), gave values within 15% of each other. Protein concentrations of the
purified muscle ACh receptor were determined by the method of Schultz and Wassarman (7). Purification of ACh Receptor from Denervated Rat Muscle. All procedures were performed at 0-4°. Lower leg muscles (100-150 g) of rats, denervated 10-20 days earlier by bilateral transection of the sciatic nerve, were homogenized in 700 ml of buffer A [50 mM Tris-HCl, pH 7.4/50 mM NaCI/1 mM (ethylenedinitrilo)tetraacetic acid (EDTA)/1 mM ethylene glycol-bis(fl-aminoethyl ether)-N,N'-tetraacetic acid (EGTA)/0. 1 mM phenylmethylsulfonyl fluoride (PMSF) containing Trasylol (Sigma Chemical Co.) at 20 units/ml, and Pepstatin at 0.1 ,ug/ml] in a Waring blender, passed through cheesecloth and centrifuged for 90 min at 40,000 X g. The pellets were resuspended in 100 ml of buffer A lacking EDTA and EGTA, and Triton X-100 was added to a final concentration of 1%. After incubation at 00 for 1 hr, insoluble material was removed by centrifugation for 45 min at 40,000 X g. The supernatant was adsorbed batchwise to 30 ml of concanavalin A (Con A)-Sepharose 4B [3 mg of Con A per ml Sepharose prepared by the cyanogen bromide activation technique (8)] by stirring at 40 for 90 min. The slurry was then poured into a glass chromatographic column. The column was sequentially washed with 50 ml of buffer B (50 mM Tris.HCI, pH 7.4/50 mM NaCI/0.1 mM PMSF/1% Triton), 100 ml of buffer B containing 1 M NaCl, and 50 ml of buffer B. Approximately 0.8 column volume of 0.4 M a-methylmannoside in buffer A containing 3% Triton was added to the column. After at least 24 hr, the column was eluted with 1 column volume of a-methylmannoside solution and the eluate was dialyzed overnight against 1 liter of buffer C (10 mM Tris-HCI, pH 7.4/1 mM EDTA/0.1 mM PMSF/1% Triton). A column of cobrotoxin-Sepharose 4B (6) (0.30 ml in a 1-ml plastic syringe) was washed with buffer B containing 1 M NaCl and equilibrated with buffer C. The Con A-Sepharose eluate was passed through the column twice at 5-8 ml/hr, which removed 80-90% of the toxin-binding activity from the solution. The column was then washed with 1 ml of buffer B, 1 ml of buffer B containing 1 M NaCl, and 1 ml of buffer B and then eluted for 30 min with 0.4 ml of 1 M carbamylcholine chloride in 50 mM Tris-HCI, pH 8.6/1 mM EDTA/1 mM EGTA/0.1 mM PMSF/1% Triton. Carbamycholine was removed by extensive dialysis against buffer D (0.2% Triton X-100/50 mM NaCI/10 mM sodium phosphate/I mM EDTA/3 mM NaN3, Abbreviations: ACh, acetylcholine; [3H]MBTA, 4-(N-maleimido)benzyltri[3H]methylammonium iodide; 125I-a-BuTx, 125-Ilabeled a-bungarotoxin; EDTA, (ethylenedinitrilo)tetraacetic acid; EGTA, ethylene glycol-bis(,B-aminoethyl ether)-NN'-tetraacetic acid; PMSF, phenylmethylsulfonyl fluoride; Con A, concanavalin A; NaDodSO4, sodium dodecyl sulfate.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate
t Present address:
Department of Neuroscience, Children's Hospital Medical Center, 300 Longwood Avenue, Boston, MA 02115.
this fact.
4685
4686 ' Neurobiology: Froehner et al.
pH 7.0). This procedure gives approximately 50,000-fold purification of the muscle ACh receptor and yields of 3-10%. [3H]MBTA Labeling. For determination of stoichiometric binding of [3H]MBTA to muscle and Torpedo receptors, the method of Karlin et al. (9) was used with minor changes. For experiments using sodium dodecyl sulfate (NaDodSO4)/ polyacrylamide gel electrophoresis, the procedure was modified as follows. A solution containing purified ACh receptor (50-100 pmol in 100-200 pAI) and, in some cases, PMSF and Pepstatin was incubated at 230 for 10 min (or, in one experiment, at 40 for 60 min) with an equal volume of 0.4 mM dithiothreitol in 0.2% Triton/150 mM NaCI/20 mM Tris-HCI, pH 8.3/1 mM EDTA/3 mM NaN3. One-tenth volume of 0.53 M sodium phosphate buffer, pH 6.7 was added, 400 ,l of 5 ,gM [3H]MBTA was added, and the solution was incubated for 1 min at 230. The sample was then chilled on ice and 20 ,Al 0.1 M 2-mercaptoethanol was added. In control experiments the receptor was incubated with cobrotoxin (4 ig/ml) after reduction. After addition of 50 jzg of polyaspartic acid, the sample was dialyzed against 1 liter of 10 mM Tris-HCI, pH 7.4/1% Triton for 1 hr at 4°. Protein was precipitated with 10 volumes of acetone at -20° for 15 min, collected by centrifugation, and redissolved in NaDodSO4/electrophoresis sample buffer (60,u) by boiling for 5 min. Electrophoresis and Fluorography. NaDodSO4/polyacrylamide gel electrophoresis was performed in a slab gel according to Laemmli (10) with minor modifications. The sample buffer contained 3.5% NaDodSO4 and 5% 2-mercaptoethanol and the acrylamide concentration in the separation gel was 9%. Standards used for molecular weight determinations were phosphorylase a (94,000), bovine serum albumin (68,000), pyruvate kinase (57,000), fumarase (49,000), aldolase (40,000), D-amino acid oxidase (37,000), and glyceraldehyde dehydrogenase (36,000). Because the polypeptide components of the ACh receptor are probably glycoproteins, the molecular weights determined by using these standards are provisional. Staining and destaining with Coomassie brilliant blue was performed according to Fairbanks et al. (11). After destaining, the gel was scanned with a densitometer or photographed and then prepared for fluorography as described by Bonner and Laskey
(12). RESULTS ACh receptor purified from denervated muscle by affinity chromatography on Con A-Sepharose and cobrotoxin-Sepharose had a specific activity of approximately 8-10 nmol of 125I-aBuTx bound per mg of protein, comparable to that of highly purified Electrophorus and Torpedo receptors (1). Analysis of the purified muscle receptor by NaDodSO4/gel electrophoresis showed the preparation to contain a number of different polypeptide chains of apparent molecular weights 45,000, 51,000, 56,000, 62,000, and 110,000 [Figs. 1 (lane a) and 2]. In addition, all preparations contained one or more polypeptides whose position corresponded to approximately 49,000 daltons. This component was not always clearly resolved and is seen in Fig. 2 as a shoulder(s) of the peak at 51,000 daltons. Analysis of preparations comparable to those in Figs. 1 and 2 indicates that over 90% of the protein corresponds to receptor (13). Further purification by sucrose density gradient centrifugation decreases but does not eliminate the 56,000 dalton component, and the relative amounts of the other polypeptides remain virtually constant (13). Initial experiments were performed with the method of Karlin et al. (9) to determine whether binding of [3H]MBTA
, t . * 'o
.,|' .l;
Proc. Nati. Acad. Sci. USA 74 (1977)
b
a
110
_
*
T
.23i N
:
t-
X :_,.
.,._
62 56
51
49 45
c
_
w
^
:::
.s
_
.s
s
: ;..X _i.: *
'f
: 'pS.a .S
42 39
FIG. 1. NaDodSO4/polyacrylamide gel electrophoresis of [3H]MBTA-labeled ACh receptor. Lanes: a, Coomassie brilliant bluestained gel of 2.3 Ag of ACh receptor; b, fluorogram of 9.2 gg of [3H]MBTA-labeled ACh receptor; c, Coomassie brilliant blue-stained gel of 9.2 of jg [3H]MBTA-labeled ACh receptor (same sample as in b). Reduction and alkylation were performed at 230 for 20 min and 10 min respectively, in the absence of protease inhibitors. Molecular weights (X10-3) are shown at the sides.
to the muscle receptor could be detected. Specific binding was defined as that which could be blocked by a-BuTx. The values obtained in one such experiment (Table 1) show that over 90% of the binding can be blocked by toxin. In three experiments with two different preparations of ACh receptor from muscle, the ratio of 125I-a-BuTx binding to specific binding of [3H]MBTA was 1.5, 1.1, and 0.5. Corresponding values in parallel experiments on purified Torpedo receptor were 1.8, 2.5, 2.0, and, in a fourth experiment, 2.0, confirming values previously found by Weill et al. (5). To determine which of the polypeptides in the muscle receptor bind the affinity reagent, purified receptor was reduced with dithiothreitol, incubated with [3H]MBTA, and analyzed by gel electrophoresis and fluorography. In an initial experiment [Fig. 1 (lane b)], bands of radioactivity were observed corresponding to molecular weights of 49,000, 45,000, 42,000, and 39,000. All of these bands were eliminated by prior incubation with purified a-neurotoxin from cobra venom. It appears, however, from a comparison of gels stained with Coomassie brilliant blue before and after the reduction and alkylation reactions [Fig. 1 (lanes a and c)], that proteolysis had occurred during the reaction, resulting in the appearance of new bands corresponding to 42,000 and 39,000 daltons. In two subsequent experiments in which reduction was performed in
Proc. Nati. Acad. Sci. USA 74 (1977)
Neurobiology: Froehner et al.
44687
TOP I
I
I
I
I
I
I
51,000 I
(I II I
45,000
DYE FRONT
!11'
I
FIG. 2. Densitometer tracings of Coomassie brilliant blue-stained (- - -) gel of alkylated receptor and fluorogram of [3H]MBTA-labeled (-) receptor. The same sample was used for both analyses. Alkylation was performed for 1 min at 230 after reduction for 10min at 230 in the presence of protease inhibitors.
the presence of protease inhibitors, either at 40 for 60 min or at 23° for 10 min, and the time of alkylation was decreased to 1 min, no change in subunit composition after labeling was detected by staining with Coomassie brilliant blue. In both of these experiments, we obtained two peaks of radioactivity. A parallel experiment with receptor from Torpedo gave only one labeled subunit as previously observed. The results of one of the experiments with the muscle receptor are shown in Fig. 2. One peak of radioactivity coincided with the 45,000 dalton subunit, and the other occurred at a position corresponding to 49,000 daltons. The latter peak was clearly distinguished from the 51,000 dalton component and appeared most closely aligned with the shoulder of that peak. Similar results were obtained in the other experiment, except that the peak of radioactivity at the higher molecular weight was not sharp enough to allow unequivocal assignment of either Table 1. Specific labeling of reduced muscle and Torpedo ACh receptors by [3HJMBTA [3H]- Toxin Ratio, MBTA bind- toxin bound 3H in ppt., bound, ing,t [3HJMBTA pmol pmol bound cpm Sample 1115 Buffer D Muscle ACh receptor 1200 plus toxin 1.1 5.3 Muscle ACh receptor 4091 4.9 Torpedo ACh receptor 1035 plus toxin Torpedo ACh 6.7 2.5 2642 2.7 receptor * Reduction and alkylation with [3HJMBTA and filtration were carried out according to Karlin et al. (9) as described in Materials and Methods. Each value represents the average of two determinations. t -251-Labeled toxin binding activity was determined as described previously (6).
the 49,000 or the 51,000 dalton component. The ratio of radioactivity in the 45,000 dalton subunit to that in the subunit of higher molecular weight was 1.0 in the experiment shown. In a second experiment, performed on the same preparation 3 weeks later, the ratio was 0.4. Because the paramount consideration in these two experiments was to prevent proteolysis, we cannot be sure that the conditions used resulted in complete labeling. Thus, these values cannot be used to infer relative amounts of MBTA-reactive sites in these two subunits. DISCUSSION MBTA is one of a group of affinity reagents designed to react with a sulfhydryl group near the active site of the reduced ACh receptor from Electrophorus electroplax. The reactive sulfhydryl group is thought to be near the active site because alkylation with the affinity reagents can be blocked by cholinergic agonists and antagonists and because certain alkylating agents such as bromoacetylcholine produce a state of permanent depolarization of the membrane that can be reversed only by high concentrations of d-tubocurarine (14, 15). Incubation of reduced eel ACh receptor, either in intact electroplax or after purification, with [3H]MBTA results in specific labeling of a single polypeptide chain that is thus thought to contain all or part of the active site (3, 4). [3H]MBTA has been shown to label a subunit of the same size in purified Torpedo receptor (5). The experiments reported here demonstrate that, after reduction, purified muscle receptor, like that of Electrophorus and Torpedo, can also be labeled with [3H]MBTA. The reaction is highly specific because it is decreased more than 90% by a-bungarotoxin. These results are consistent with physiological experiments showing that there is a disulfide bond near the active site of vertebrate muscle receptors that can be reduced and alkylated (16, but see also 22). The ability of [3H]MBTA to label the active site of the reduced receptor from such disparate species as rat and marine ray suggests that at least part of the receptor structure has been strongly conserved throughout evolution. Our results differ in two ways from those found with the fish receptors. First, [3H]MBTA labels two polypeptide chains of
4688
Neurobiology: Froehner et al.
apparent molecular weights 45,000 and 49,000. Second, the stoichiometry of the binding of [3H]MBTA relative to 125I-aBuTx binding activity appears to be different. Both fish receptors bind exactly twice as much a-neurotoxin as they do MBTA (1), whereas our experiments suggest that the muscle receptor binds equivalent amounts of [3H]MBTA and 125I-aBuTx. The small amounts of muscle receptor available and the corresponding imprecision in the results make this conclusion a tentative one. The specific activity (mol of 125I-BuTx per mg of protein) of the purified muscle preparations (13) is similar to that reported for Torpedo, and the physical properties of the muscle receptor suggest that it has a molecular weight similar to that of Torpedo (see ref. 1). Thus, our results lead to the suggestion that the muscle receptor binds twice as much [3H]MBTA as does the receptor from Torpedo. The observation that [3H]MBTA labels two subunits of the muscle receptor could be explained in several ways. One possible explanation is that the 45,000 dalton subunit is derived from the larger one by proteolysis; alternatively, both may be derived from an even larger common precursor. It is difficult to exclude this possibility except to note that, under proper conditions, proteolysis during reduction and alkylation can be prevented (Fig. 2). Thus, any proteolytic conversion must have occurred during extraction and purification of the receptor. In addition, it is unlikely that either of the labeled subunits could be derived from the 51,000 dalton component because, in the experiments shown in Figs. 1 and 2, this component is devoid of radioactivity. Another possibility is that [3H]MBTA labels two separate sites in the muscle receptor, only one of which corresponds to that labeled in the fish receptors. Because binding to both subunits in the muscle protein is blocked by toxin, the two sites would presumably correspond to different toxin binding sites as well. Evidence for two classes of toxin-binding sites has been found both in muscle and Torpedo receptors (6, 18), although in the fish receptors it has thus far been possible to identify only one subunit that binds a-neurotoxin (5, 19, 20). Two classes of binding sites for ACh have been observed for the Torpedo receptor (18, 21, 22). A final possibility is that [3H]MBTA labels two classes of receptor molecules, each with a single binding site. It is unlikely that these could correspond to junctional and extrajunctional receptors because isoelectric focusing analysis indicates that the receptor preparation studied in these experiments is at least 90%
extrajunctional (N. Nathanson, unpublished data). Further
Proc. Natl. Acad. Sci.USA 74 (1977) experiments will be required to distinguish the various alternatives. We thank Steve Rowe for expert technical assistance, Joe Gagliardi for help with the photography, and Dr. Regis Kelly for helpful discussions. This work was supported by National Institutes of Health Grants NS 09646 and NS 07065, National Science Foundation Grant BMS 75-03026, and grants from the Muscular Dystrophy Association. S.F. was a Muscular Dystrophy Association Fellow. 1. Karlin, A. (1977) Excerpta Medica, in press. 2. Karlin, A., Prives, J., Deal, W. & Winnik, M. (1971) J. Mol. Biol., 61, 175-188. 3. Reiter, M. J., Cowburn, D. A., Prives, J. M. & Karlin, A. (1972) Proc. Nati. Acad. Sci. USA 69,1168-1172. 4. Karlin, A. & Cowburn, D. A. (1973) Proc. Natl. Acad. Scd. USA
70,3636-3640.
5. Weill, C. L., McNamee, M. G. & Karlin, A. (1974) Blochem. Blophys. Res. Commun. 61, 997-1003. 6. Brockes, J. & Hall, Z. W. (1975) Biochemistry 14,2092-2099. 7. Schultz, R. M. & Wassarman, P. M. (1976) Anal. Biochem. 77, 25-32. 8. Cuatrecasas, P. (1970) J. Blol. Chem. 245,3059-3065. 9. Karlin, A., McNamee, M. G. & Cowburn, D, A. (1976) Anal. Biochem. 76,442-451. 10. Laemmli, U. K. (1970) Nature 227,680-685. 11. Fairbanks, G., Steck, T. L. & Wallach, D. F. H. (1971) Biochemistry 10, 2606-2617. 12. Bonner, W. M. & Laskey, R. A. (1974) Eur. J. Blochem. 46, 83-88. 13. Froehner, S. C., Reiness, C. G. & Hall, Z. W. (1977) J. Biol. Chem., in press. 14. Karlin, A. (1969) J. Gen. Physiol. 54, 245S-264S. 15. Silman, H. I. & Karlin, A. (1969) Science 164, 1420-1421. 16. Ben-Haim, D., Landau, E. M. & Silman, I. (1973) J. Physiol. (London) 234, 305-325. 17. Raftery, M. A., Vandlen, R. L., Reed, K. L. & Lee, T. (1975) Cold Spring Harbor Symp. Quant. Blol. XL, 193-202. 18. Biesecker, G. (1973) Biochemistry 12,4404-4409. 19. Gordon, A., Bandini, G. & Hucho, F. (1974) FEBS Lett. 47, 204-208. 20. Eldefrawi, M. E., Eldefrawi, A. T. & Shamoo, A. E. (1975) Ann. N.Y. Acad. Sci. 264,183-202. 21. Changeux, J.-P., Benedetti, L., Bourgeois, J.-P., Brisson, A., Cartaud, J., Devaux, P., Grunhagen, H., Moreau, M., Popot, J.-L., Sobel, A. & Weber, M. (1975) Cold Spring Harbor Symp. Quant. Biol. XL, 211-230. 22. Lindstrom, J. M., Singtr, S. J. & Lennox, E. S. (1973) J. Membr. Biol. 11,217-226.
a
