the Cholinergic Receptor Protein*

Size: px

Start display at page:

Download "the Cholinergic Receptor Protein*"

Blaze Ford
5 years ago
Views:

1 Proceeding8 of the National Academy of Science8 Vol. 67, No. 3, pp , November 1970 Use of a Snake Venom Toxin to Characterize the Cholinergic Receptor Protein* Jean-Pierre Changeuxt, Michiki Kasait, and Chen-Yuan Leet DEPARTEMENT DE BIOLOGIE MOLECULAIRE, INSTITUT PASTEUR, PARISt; AND PHARMACOLOGICAL INSTITUTE, COLLEGE OF MEDICINE, NATIONAL TAIWAN UNIVERSITY, TAIPEI, CHINAt Communicated by Jacques Monod, July 30, 1970 Abstract. a-bungarotoxin, a polypeptide of mol wt 8000 purified from the venom of Bungarus multicinctus, blocks irreversibly and specificallythe excitation by cholinergic agonists on the isolated electroplax and on purified membrane fragments in vitro. The toxin also blocks the in vitro binding of decamethonium to a protein recently isolated from electric tissue. This observation strengthens our earlier conclusion that this protein is the cholinergic receptor macromolecule. At a glance, the most evident and simple manner to characterize and identify the physiological receptor of acetylcholine in an excitable membrane is to use compounds that are structurally analogous to acetylcholine and thus present a high affinity for the cholinergic receptor site. However, it is now well established that in most excitable membranes there exist several -distinct classes of sites, all of which are able to bind cholinergic ligands. Among them are, in addition to the physiological receptor site, the catalytic and allosteric sites of the enzyme acetylcholinesterase (AcChE).l The use of cholinergic ligands thus meets with a difficult problem of specificity.2 Interestingly enough, in the course of the past few years it has been shown that certain toxins from snake venoms, although completely unrelated structurally to acetylcholine, nevertheless act much like curare. One of them is a-bungarotoxin (a-bgt), a basic polypeptide of mol wt 8000, which has been extensively studied by Lee and his associates.3-5 a-bgt is purified from the venom of an elapid snake from Taiwan (Bungarus multicinctus) and the purified toxin gives irreversible neuromuscular blocking effects. In addition, d-tubocurarine protects against the action of a-bgt. From these findings Lee and Chang4 have concluded that a-bgt combines irreversibly with the cholinergic receptor at the motor endplate. We report here some experiments carried out with purified a-bgt and various preparations derived from the electric organ of Electrophorus electricus. It is shown that in vivo, with the isolated electroplax, a-bgt irreversibly blocks the depolarization caused by bath application of carbamylcholine, a cholinergic agonist. The same effect is observed with purified electric-organ membrane fragments using the in vitro assay of Kasai and Changeux.6 In addition, in agreement with the early observation of Lee and associates, d-tubocurarine protects against a-bgt blockade. Finally we tested a-bgt on the binding of 1241

2 1242 BIOCHEMISTRY: CHANGEUX ET AL. PROC. N. A. S. radioactive decamethonium to a protein recently isolated from the electric organ of E. electricus, which presents, in solution, several characteristic properties of the cholinergic receptor macromolecule.7 By means of an equilibrium dialysis assay, we show that a-bgt blocks in vitro the binding of decamethonium to this protein but has no detectable effect on the catalytic center of AcChE. a-bgt thus combines specifically with the cholinergic receptor site on the receptor protein. Materials and methods. a-bungarotoxin: The lyophilized venom of B. multicinctus was first fractionated on a column of CM-Sephadex (C-50) into eight major fractions by gradient elution8 with ammonium acetate buffer (from 0.05 M, ph 5.0 to 1.0 M, ph 6.8.). a-bungarotoxin was obtained from fraction 2 by repeated rechromatography on CM-cellulose columns. The purified a-bungarotoxin has been shown to be homogeneous by ultracentrifugal analysis, microzone electrophoresis, amino acid analysis, and end-group analysis (Narita and Lee, to be published). Its molecular weight, estimated by sedimentation equilibrium, is Isolated electroplax: The cell was dissected and mounted exactly as described by Higman et al.9 Membrane potentials were measured across the innervated membane by intracellular glass microelectrode filled with 3 M KCl. In vitro assay for excitation by cholinergic agonists: Excitable "microsacs" derived from the innervated membrane of the electroplaxes were purified from homogenates of electric tissues.10 Their response to cholinergic agonists was assayed6 by measuring the efflux of 22Na+. After overnight equilibration with 22Na+ the concentrated microsac suspension is rapidly diluted in a nonradioactive saline medium and then, at intervals, aliquots are filtered through a Millipore filter which is then washed with buffer. The dried Millipore filters are counted in a Packard scintillation counter. Extraction of the cholinergic receptor protein: Extracts were prepared from fresh or frozen electric tissue exactly as described by Changeux et al.7 Binding of [methyl-3h I- decamethonium at 178 Ci/mol (Radiochemical Centre, Amersham) was measured at 40C, as already described,7 in the presence of 5.6 X 10-7 M decamethonium in Ringer's solution, buffered at ph 8.0. AcChE was essayed with acetylthiocholine as substrate.' Results. In vitro effect of a-bgt on the isolated electroplax: Fig. 1 shows that a 10-min exposure of the innervated mv Carb 9/ b Carb Carb face of the isolated electroplax to a 1,ug/ml -8 solution of a-bgt in Ringer's medium is fol- Rlowed by an irreversible blocking of the elec- RVR R mv Carb Carb Carb trical response of the cell to bath application -80 of carbamylcholine. At this concentration -60[ t t R R of toxin, the residual depolarization repre- '0min. sents about 30% of the initial decrease of membrane potential; in addition, there is no FIG. 1. Irreversible blockade in tendency of recovery up to 80 min after exvivo by a-bungarotoxin (a-bgt) of posure to the toxin. At higher concentrathe electrical response of the isolated tos o thg the resposeei competely electroplax to a cholinergic agonist, ions of a-bgt the response is completely carbamylcholine (Carb). abolished. a-bgt thus irreversibly blocks the response to cholinergic agonists in vivo. In vitro effect of a-bgt on purified membrane fragments: The same effect is observed (Fig. 2) with membrane fragments purified from a crude homogenate of electric organ. The response to cholinergic agonists is now studied by measuring the permeability of the membrane microsacs to Na+ ions.6 Again, 1 /ug/ml a-

3 VOL. 67, 1970 CHOLINERGIC RECEPTOR control pg/ml n-bgt ko Xra~~82 0 C.D of+ 10-5M d-tubocurarine - 01\6\ \s ki pg/ml m-bungerotoxin 30 30x\ ~~~~. control. 104MCarbc o SM d-tc [minutes) [minutes] (Left) FIG. 2. In vitro effect of a-bgt on the response of purified electricorgan membrane fragments to cholinergic agonists. The 22Na+ content of the microsacs is plotted as a function of time. The dilution medium contained either 10-4 M carbamylcholine (Carb, X), 10-5 M d-tubocurarine (d-tc, 0) or no effector (o). Prior to the permeability assay the membrane fragments were incubated for 90 min at 220C with 1 pg/ml a-bgt. The concentration of protein in the undiluted microsacs suspension was 4.9 mg/ml. (Right) FIG. 3. Protective effect of 10-5 M d-tubocurarine against the irreversible action of a-bungarotoxin on the in vitro excitability of membrane fragments. Same techniques as for Fig. 2. ko is the first order rate constant for the initial outflow of Na+ in the absence of cholinergic effector and k is the same constant in the presence of 10-4 M carbamylcholine. The microsacs were exposed at zero time to 1 pg/ml a-bgt in the presence or the absence (control) of 10-5 M d-tubocurarine. The suspension was then diluted at the indicated time for the in vitro assay of Na+ efflux in the presence and the absence of carbamylcholine. Concentration of proteins in the undiluted suspension, 4.9 mg/ml. Bgt blocks the response to carbamylcholine, but the inhibition is now complete. Dilution experiments moreover show that in vitro, as in vivo, the effect of a-bgt is irreversible. We were also able to confirm the result of Lee and Chang4 that d-tubocurarine protects against the irreversible action of a-bgt (Fig. 3), and thus that a-bgt combines with the same macromolecule as does d-tubocurarine, carbamylcholine, or decamethonium, i.e. with the cholinergic macromolecular receptor. We then measured the dose-response curves for carbamylcholine at various concentrations of a-bgt (Fig. 4a). The midpoint of the curve does not change in the presence of a-bgt, but the maximal response is reduced. a-bgt thus seems to make an irreversible and stoichiometric complex with the cholinergic macromolecular receptor. in agreement with this conclusion, the maximal response to carbamylcholine decreases linearly with the concentration of a-bgt (Fig. 4b). Moreover, when half as much membrane is used for the titration curve, half as much a-bgt is needed to block the response completely (Fig. 4b). We then confirmed that the irreversible and stoichiometric inhibition of the cholinergic receptor by a-bgt is specific and exclusive. This was shown by supplementing the suspension of excitable membrane fragments with the same quantity (as estimated by protein assay) of nonexcitable membrane fragments derived from the noninnervated face of electroplaxes (Table 1). Approximately 30%0 more a-bgt is needed in the presence of nonexcitable fragments than in their absence to yield complete blockade; this 30% excess falls in the range of

4 1244 BIOCHEMISTRY: CHANGEUX ET AL. PROC. N. A. S. k If control k, ~~~~~~~~~~~~~~EME ~ mg prot/mi b l_~~~~~~ 0.43pg/ml m-bgt * EME I*~ \ \ 4 85 mg prot/ml / pg/ml a-bgt c.ii- 1.0 pg/ml m-bgt - 10'M i0'm i0-m [carbamylcholine] [m-bungarotoxin] pg/ml FIG. 4. (a) Effect of cz-bgt on the dose-response curves of excitable microsacs to carbamylcholine. ko and k have the same significance as for Fig. 3. The microsacs were incubated with the indicated concentration of a-bgt for 1 hr at 220C. Concentration of proteins in the undiluted microsac suspension, 4.9 mg/ml. FIG. 4. (b) Titration curves obtained in vitro with a-bgt and excitable microsacs at two different concentrations of membrane. The maximal response to carbamylcholine measured at a given concentration of a-bgt was obtained in experiments similar to the one represented in Fig. 4a and have the same significance as for Fig. 3. EME, excitable membrane fragments. experimental error and is probably due to some contamination of the noninnervated membrane fragments by fragments from the innervated face. We also checked that, as already found by Lee and associates (unpublished), in the range of concentrations used in these experiments, a-bgt has absolutely no effect on the catalytic activity and affinity of AcChE present in the microsacs. From the number of a-bgt molecules needed for a complete inhibition of the in vitro response to carbamylcholine, it is possible to estimate the actual number of cholinergic receptor molecules present per milligram of membrane protein and compare it to the number of AcChE molecules. Let us make the following plausible assumptions regarding a-bgt and AcChE: (1) a-bgt is pure, has a mol wt of 8000, and binds irreversibly and exclusively to the cholinergic receptor macromolecule; (2) the specific activity of pure AcChE is 750 mol of acetylcholine esterified per g per hour, its mol wt is 260,000, and the catalytic activities of soluble and membrane-bound AcChE are not different. Then, we find that, in our membrane preparation which contains 5 nmol of AcChE per g of protein, there are 4.9 mol of a-bgt bound per mole of AcChE. If we further assume that TABLE 1. Specific effect of a-bgt on excitable membrane fragments. Excitable + Excitable nonexcitable microsacs microsacs* Proteins (mg/ml) a-bgt required for 100% block of response to carbamylcholine (pmol/ml) AcChE (pmol/ml) *Specific activity of AcChE (moles of acetylcholine per g per hr) a-bgt/acche * Preparation of excitable inicrosomies suppllelmented with an equal amount (in protein) of nonexcitable microsacs.

VOL. 67, 1970 CHOLINERGIC RECEPTOR 1245 the number of a-bgt molecules bound is equal to the number of cholinergic receptor sites and that the esterase is made up of four subunits, each with one

5 VOL. 67, 1970 CHOLINERGIC RECEPTOR 1245 the number of a-bgt molecules bound is equal to the number of cholinergic receptor sites and that the esterase is made up of four subunits, each with one catalytic site," the number of cholinergic receptor sites is estimated to be approximately the same as the number of acetylcholinesterase catalytic sites and thus of esterase subunits. Effects of a-bgt on the binding of decamethonium to the cholinergic receptor protein isolated in vitro. In a recent paper7 was described the extraction, from eel electric organ, of a protein which presents in vitro several characteristic properties of the cholinergic receptor macromolecule. In particular, this protein binds the cholinergic agonists and antagonists with a rather high selectivity. It then becomes of extreme interest to see if a-bgt blocks this binding in vitro. Fig. 5 FIG. 5. Antagonism by a-bung- 30[ arotoxin to decamethonium (Deca) binding to a preparation of choliner- 20- gic receptor protein. Before dialysis X the extract was incubated about 30 min at room temperature in the presence of the indicated concentration of toxin. The concentration of proteins 00 in the extract was 18 mg/ml [m-bungarotexin] Wgmi shows that it does. In this experiment we incubated a-bgt, at the indicated concentrations, with the protein extract for 30 min. The treated extract was then dialyzed7 against radioactive decamethonium in Ringer's saline medium at ph 8.0. Fig. 5 shows that a-bgt irreversibly blocks the in vitro binding of decamethonium to the receptor protein present in the extract. In agreement with what we had already observed with the membrane fragments, under the conditions where the binding of decamethonium is strongly inhibited by a-bgt, we could not detect any effect of ai-bgt on the activity of AcChE. a-bgt thus binds to the cholinergie receptor protein but not to the catalytic site of the enzyme. Fig. 5 further shows that the inhibition of decamethonium binding by a-bgt is partial. In the presence of an excess of a-bgt (50,ug/ml) and of a free concentration of decamethonium of 5.6 X 10-7M in the external dialysis buffer, about 28% of decamethonium bound in the absence of a-bgt remains associated with the extract. Such incomplete antagonism is interpreted as due to the presence, in our preparation, of molecules which carry binding sites for decamethonium but are not involved in the physiological action of cholinergic agonists. This 28% of bound decamethonium is completely displaced by phenyltrimethylammonium, hexamethonium, or carbamylcholine, which are known to be strong inhibitors of AcChE and are likely to be carried by AcChE. We are thus in a position to distinguish the receptor sites which belong to the cholinergic receptor protein from those which belong to the catalytic center of AcChE.'2 These last results are quantitatively consistent with those obtained in situ with the membrane fragments. About ten times as much protein is present in the

6 1246 BIOCHEMISTRY: CHANGEUX ET AL. PROC. N. A. S. soluble extract as in the membrane fragments, and we need about ten times as much a-bgt to get maximal antagonism to decamethonium binding. Discussion. a-bungarotoxin, a polypeptide of mol wt 8000, combines irreversibly with the cholinergic receptor of the eel electroplax both in situ and in solution: it blocks the binding of decamethonium to the class of site that binds both d-tubocurarine and decamethonium, without affecting the catalytic site of AcChE. a-bgt is thus a specific reagent for the physiological receptor of acetylcholine. From the number of a-bgt molecules irreversibly bound, we are in a position to give an estimate of the number of cholinergic receptor sites present in situ or in our extracts. We found that 4.9 molecules of a-bgt are bound per molecule of AcChE present in situ in the membrane fragments; since the molecule of AcChE is tetrameric and presumably carries one catalytic site per subunit" this means that there are, in the excitable membrane, approximately as many cholinergic receptor sites as AcChE catalytic centers. Two mechanisms might account for the effect of a-bgt on the cholinergic receptor: (1) the secondary and tertiary foldings of a-bgt are such that it presents a structure complementary to that of the cholinergic receptor site and hence prevents the binding of decamethonium by steric hindrance; or (2) a-bgt binds outside the cholinergic receptor site and acts as an allosteric effector: according to current models,"' the toxin might stabilize a conformation of the cholinergic receptor protein which possesses a low affinity for the cholinergic agonists-in other words, stabilize the receptor in its resting conformation. It is difficult to distinguish between these alternatives. Let us say, however, that the first hypothesis of a direct effect of a-bgt at the level of the cholinergic receptor site seems the more plausible. Indeed, the fact that d-tubocurarine protects against the binding of a-bgt is not simply explained on the basis of an allosteric effect since, according to the model considered, both d-tubocurarine and a-bgt would be expected to bind to the resting state of the cholinergic receptor. Competition between a polypeptide and a small ligand at a protein binding site is not unusual and several examples of similar antagonisms have been described with various proteases" or nucleases.'4 Such a possibility is not excluded by the consideration of the respective sizes of a-bgt and d-tubocurarine. On the assumption that a- Bgt is a sphere of density 1.3, the expected diameter of the molecule would be around 27 A. The distance between two quaternary nitrogens in d-tubocurarine is close to 14 A. There is, thus, no serious reason to exclude a binding of these two compounds to the same area of the cholinergic receptor macromolecule. Abbreviations: a-bgt, a-bungarotoxin; AcChE, acetylcholinesterase. * This investigation was supported by grants from the U.S. National Institutes of Health, the Centre National de la Recherche Scientifique, the D6l6gation G6n6rale A la Recherche Scientifique et Technique, and the Commissariat A l'energie atomique. We thank Simone Mougeon for her technical assistance. I Changeux, J-P., T. R. Podleski, J-C. Meunier, J. Gen. Physiol., 54, 225S (1969). 2 Chagas, C., E. Penna-Franca, K. Nishie, and E. J. Garcia, Arch. Biochem. Biophys., 75, 251 (1958); Changeux, J-P., T. R. Podleski, and L. Wofsy, Proc. Nat. Acad. Sci. USA, 58, 2063 (1967); Ehrenpreis, S., Biochim. Biophys. Acta, 44, 561 (1960); Karlin, A., J. Gen. Physiol., 54, 245S (1969); La Torre, J. L., G. S. Lunt, and E. De Robertis, Proc. Nat. Acad.

7 VOL. 67, 1970 CHOLINERGIC RECEPTOR 1247 Sci. USA, 65, 716 (1970); O'Brien, R. D., and L. P. Gilmour, Proc. Nat. Acad. Sci. USA, 63, 496 (1969); O'Brien, R. D., L. P. Gilmour, and M. E. Eldefrawi, Proc. Nat. Acad. Sci. USA, 65, 438 (1970); Podleski, T., J-C. Meunier, and J-P. Changeux, Proc. Nat. Acad. Sci. USA, 63, 1239 (1969); Rang, H. P., and J. M. Ritter, Mol. Phamactl., 5, 394 (1969). 8 Chang, C. C., and C. Y. Lee, Arch. Int. Pharmacodyn., 144, 241 (1963). 4Lee, C. Y., and C. C. Chang, Mem. Inst. Butantan, Simp. Internac., 33, 555 (1966). 6 Lee, C. Y., L. F. Tseng, and T. H. Chiu, Nature, 215, 1177 (1967). 6Kasai, M., and J-P. Changeux, C. R. Acad. Sci., Paris, 270, 1400 D (1970). 7Changeux, J-P., M. Kasai, M. Huchet, and J-C. Metnier, C. R. A cad. Sci., Paris, 270, 2864 D (1970). 8 Lee, C. Y., C. C. Chang, T. H. Chiu, P. J. S. Chiu, T. C. Tseng, and S. Y. Lee, Nauyn- Schmiedebergs Arch. Exp. Path. Pharmakol., 259, 360 (1968). 9 Higman, H., T. R. Podleski, and E. Bartels, Biochim. Biophys. Acta, 79, 138 (1964). 10 Changeux, J-P., M. Gautron, M. Israel, and T. R. Podleski, C. R. Acad. Sci., Paris 269, 1788 D (1969). 11 Kremzner, L., and I. B. Wilson, Biochemistry, 3, 1902 (1964). 13 This conclusion is supported by recent experiments of selective heat denaturation and of ultracentrifugation of the soluble extract of receptor (Meunier and Changeux, unpublished results); it becomes possible to demonstrate that these two classes of site are carried by distinct protein units which are easily separated in vitro. 3 Laskowski, M., and M. Laskowski, Advan. Protein Chem., 9, 203 (1954). "4Lesca, P., and C. Paoletti, Proc. Nat. Acad. Sci. USA, 64, 913 (1969).

mp) irradiation. The potential difference across the excitable membrane may be

mp) irradiation. The potential difference across the excitable membrane may be PHOTOREGULA TION OF BIOLOGICAL ACTIVITY BY PHOTOCHROMIC REAGENTS, III. PHOTOREGULATION OF BIOELECTRICITY BY ACETYLCHOLINE RECEPTOR INHIBITORS* BY WALTER J. DEAL, BERNARD F. ERLANGER, AND DAVID NACHMANSOHN