Proc. Nati. Acad. Sci. USA
Vol. 77, No. 8, pp. 4623-4627, August 1980
Solubilization of active opiate receptors
(opioid peptides/endorphins/zwitterionic detergent/neuroblastoma--glioma hybrid cells/brain membranes)
WILLIAM F. SIMONDS*, GREG KOSKI*, RICHARD A. STREATY*, LEONARD M. HJELMELANDt, AND
WERNER A. KLEE*
*Laboratory of General and Comparative Biochemistry, National Institute of Mental Health; and tDevelopmental Pharmacology Branch, National Institute of
Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20205
Communicated by Maxine F. Singer, May 12,1980
ABSTRACT Receptors that reversibly bind opiates and the time of planting with 5% C02/95% air, but not upon sub-
opioid peptides have been solubilized from brain and neuro- sequent feeding.
blastoma-glioma hybrid cell NG108-15 membranes. Active re- Brain Membrane Preparation. Crude mitochondrial frac-
ceptors are specifically solubilized with a new type of detergent tions (P2) from rat brain were prepared as described (8) and
3-[(3-cholamidopropyl)dimethylammonio}1-propanesulfonate, stored at -800C. Bovine brain membranes were prepared from
which is a zwitterionic derivative of cholic acid. The solubilized
receptor complexes behave as large molecules with a Stokes whole brain, without the cerebellum and medulla, and ho-
radius of 70 A and contain protein as an essential constit- mogenized in 10 vol of 50 mM Tris-HCl (pH 7.5). After filtra-
uent. tion through cheesecloth, the pellet from i centrifugation at
40,000 X g for 30 min was resuspended in a total of 2 vol of 50
A necessary first step in the study of the structure and mecha- mM Tris-HCl (pH 7.5) and frozen at -80'C until use.
nism of action of opiate receptors is their separation, in an active Solubilization. NG108-15 cell homogenate was centrifuged
form, from membranes. The many attempts made in the past at 50,000 rpm (160,000 X g) for 30 min at 40C. The pellet was
to solubilize active opiate receptors were uniformly unsuc- resuspended in 50 mM Tris-HCl (pH 7.5) to a protein concen-
cessful. Preformed complexes of opiate with receptor have been tration of 8-15 mg/ml. The suspension was homogenized on
solubilized, but the presumed receptor material in solution ice in the presence of 10 mM CHAPS with 25 strokes of a
could no longer reversibly bind opiates (1, 2). The present report ground-glass homogenizer, then centrifuged for 60 min at
describes the successful solubilization of opiate receptors in a 160,000 X g in cellulose nitrate tubes. The clear supernatant
state that reversibly binds opioid ligands and is therefore still was separated from a floating cloudy layer and the pellet by
active. A new type of detergent 3-[(3-cholamidopropyl)di- puncturing the tube with a 23-gauge needle at a point 1 cm
methylammonio]-l-propanesulfonate (CHAPS), a zwitterionic above the pellet and gently withdrawing the liquid into a 10-ml
derivative of cholic acid, is used for solubilization of active syringe.
opiate receptors from the membranes of neuroblastoma-glioma Aliquots of brain membrane suspensions were diluted with
hybrid, NG108-15, cells (3) and rat and beef brain (4-6). Op- 2 vol of 50 mM Tris-HCl (pH 7.5), 50 ,g of pancreatic trypsin
timal solubilization conditions, opiate-binding characteristics, inhibitor per ml (Worthington), 0.1 mM phenylmethylsulfonyl
and hydrodynamic properties are similar for the receptors fluoride (Calbiochem), and 0.01% dimethyl sulfoxide (East-
prepared from the three sources. The solubilized receptors are man). The suspension was centrifuged at 165,000 X g for 15
large complexes, with Stokes radii of approximately 70 A. min, and the pellet was resuspended in 2 vol of the above buffer
Conditions that would be expected to dissociate the receptor with 10 mM CHAPS to a protein concentration of approxi-
into smaller species result in the loss of receptor activity. mately 12 mg/ml. This suspension was agitated at 40C for 1 hr,
then centrifuged at 105,000 X g for 60 min. The clear, slightly
MATERIALS AND METHODS yellow supernatant fraction was carefully removed without
Materials. Chemicals were obtained from the following disturbing the pellet. This extract contained approximately 4
sources. [D-Ala2, tyrosyl-3,5-3H]Enkephalin(5-L-methionina- mg of protein per ml. Clear supernatants prepared in this way
mide) ([3H]DALAMID) (16 Ci/mmol; 1 Ci = 3.7 X 101 bec- were assayed either directly or after Sepharose 6B gel filtra-
querels) and [3H]etorphine (30.6 Ci/mmol) were from Amer- tion.
sham; [D-Ala2]enkephalin(5-L-methioninamide) (DALAMID) Opioid Binding. Aliquots of the final supernatant or Seph-
was from Calbiochem; f3-endorphin (camel) was from arose 6B eluate were incubated for 10 min at 37°C in a mixture
Boehringer Mannheim; etorphine-HCI was a gift from Robert containing 1 mM CHAPS, 100-500,ug of protein, 2 nM [3H]-
Willette (National Institute of Drug Abuse) and naloxone-HCI DALAMID (NG108-15 extracts), or 5 nM [3H]etorphine (brain
was donated by Endo Laboratories (New York). Other opiates extracts) with or without unlabeled opioid in a final volume of
were gifts from Arthur Jacobson (National Institute of Arthritis, 550 a1.. Dithiothreitol (0.5 mM) was included in the incubation
Metabolism, and Digestive Diseases). Sodium cholate and of brain extracts. After cooling in ice, 500 ,l of the mixture was
Zwittergent were from Calbiochem; CHAPS was synthesized applied to a Sephadex G-25M column (9.1 ml bed volume,
as described (7). Dithiothreitol and sucrose (ultra-pure reagent) Pharmacia Column PD-10) equilibrated with 0.32 M su-
were from Bethesda Research Laboratories (Rockville, MD). crose/10 mM Tris-HCI, pH 7.5, at 40C. The high molecular
Cell Growth. Neuroblastoma-glioma hybrid NG108-15 cells weight fraction, eluted in the inital 4.5 ml, was collected and
were grown as described (3), except that 850-cm2 tissue culture mixed with 12 ml of Aquasol, and radioactivity was determined.
roller bottles (Corning) were used. The bottles were gassed at No dissociation of bound ligand occurs during the gel filtration
at 4°C: binding is independent of dwell time of the complex
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked "ad- Abbreviations: CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-
vertisement" in accordance with 18 U. S. C. §1734 solely to indicate 1-propanesulfonate; DALAMID, [D-Ala2]enkephalin(5-L-methioni-
this fact. namide).
4624 Biochemistry: Simonds et al. Proc. Natl. Acad. Sci. USA 77 (1980)
on the column (between 5 and 30 min), opioid affinities for An estimate of the size of the detergent-solubilized receptors
soluble receptors measured by gel filtration are the same as for was obtained by gel filtration on a calibrated Sepharose 6B
particulate receptors measured by centrifugation, and binding column. Fig. 2 shows the elution profile of one such determi-
to particulate receptors measured by gel filtration equals that nation for NG108-15 receptors. The bulk of the opiate receptors
measured by centrifugation. Binding to NG108-15 or brain migrated on the column in a position between that of the
membranes with or without 1 mM CHAPS was determined by marker proteins 19S thyroglobulin (R, = 85 A) and phos-
the centrifugation assay (8). phorylase a (R8 = 64 A). Thus, the main peak exhibited a Stokes
Adenylate cyclase activity was determined by a modification radius of approximately 70 A. Adenylate cyclase activity was
of the procedure of Salomon et al. (9), as described (10). Protein solubilized along with opiate receptors from NG108-15 cell
was measured by the method of Lowry et al. (11), with bovine membranes. In the intact membranes of these cells opiate re-
serum albumin as standard. ceptors are coupled, as inhibitors, to adenylate cyclase (10).
RESULTS After solubilization with CHAPS, adenylate cylase activity
migrated slightly faster than did opiate receptors on the column
Solubilization of opiate receptors from NG108-15 hybrid cell (Fig. 2) and, thus, had a Stokes radius between that of opiate
membranes with CHAPS is demonstrated in Fig. 1A. The fig- receptors and 19S thyroglobulin. The enzyme was no longer
ure shows total and nonspecific binding of [3H]DALAMID to coupled to inhibitory (opiate) or stimulatory (prostaglandin E1
supernatant fractions after extraction at several detergent and adenosine) receptors. It was activated by fluoride ions and
concentrations and centrifugation at 160,000 X g for 1 hr. was conveniently assayed in the presence of 10 mM NaF.
Specific opiate binding, as measured by the difference between The soluble receptors exhibited a high affinity for opioid
the two binding curves, was solubilized optimally at CHAPS peptides. The specific binding of [3H]DALAMID as a function
concentrations near 10 mM, and higher detergent concentra- of ligand concentration is shown in Fig. 3. Scatchard analysis
tions during solubilization were deleterious. Comparable ex- (12) (Fig. 3 Inset) revealed a Kd of 2.3 nM and 0.96 pmol of
periments with sodium cholate are presented in Fig. 1B. The receptor per mg of protein. The high stereoselectivity charac-
protein solubilized by cholate displayed little specific opiate teristic of membrane-bound opiate receptors (13) was retained
binding even though the total amounts of protein solubilized in the solubilized binding protein. The data shown in Fig. 4
in the experiments were comparable for the two detergents. As demonstrate that the natural, pharmacologically active (-)-
receptors were brought into solution with increased CHAPS enantiomer of morphine was at least 1000 times more potent
concentration, there was a loss of membrane binding activity than the inactive (+)-enantiomer (14) in displacing [3H]-
(data not shown). At 10 mM CHAPS, 10% of the original DALAMID from the solubilized receptors. The data also show
binding capacity was present in the pellet and 20% was solu- that, at saturation, (-)-morphine and nonradioactive DALA-
601 0 ~~~-- 6
40 0 - - 4 I
-o C a a
a be 0e
5 ,,-z cJ
200 _ 2
0.5 1 5 10 50
Detergent, mM 5 10 15 20- 25
FIG. 1. Effect of detergent concentration on the solubilization Fraction
of [3HDALAMID binding activity and protein. Aliquots of NG108-15 FIG. 2. Sepharose CL-6B gel filtration of a CHAPS extract of
membranes (14.8 mg of protein per ml) were incubated for 16 min at NG108-15 membranes. One milliliter of 10 mM CHAPS extract (5.0
room temperature with detergent at the concentrations shown. After mg of protein) was applied to a Sepharose CL-6B column (2.3 X 6 cm)
centrifugation at 160,000 X g for 1 hr at 40C, the supernatant fluid equilibrated with 0.32 M sucrose/10 mM Tris-HCl, pH 7.5/1 mM di-
was diluted 1:25 in 0.32 M sucrose/10 mM Tris-HCl, pH 7.5, with thiothreitol/1 mM CHAPS at 40.C. Eighteen-drop aliquots were col-
sufficient detergent added to maintain a final detergent concentration lected at a flow rate of 150 Al/min and weighed because drop size varies
of 2 mM in all cases. Aliquots (1 ml) of the diluted supernatant were with detergent concentration. Aliquots were assayed for specific
assayed for binding of 2 nM [3H]DALAMID. 0, Total binding of [3H]DALAMID binding (0), adenylate cyclase activity (0), and
[3H]DALAMID; 0, nonspecific binding assayed in the presence of protein concentration (--- -). Marker proteins were chromatographed
10 ,uM unlabeled ligand. Values are means I SEM of quintuplicate in the absence of dithiothreitol and CHAPS. Their peak positions
determinations. ---, Protein concentration in the undiluted super- correspond to fraction numbers as follwsw 19S thyroglobulin, 14.7
natant. (A) Solubilization with CHAPS; (B) solubilization with so- (Rs = 85 A); phosphorylase a, 18.6 (R8 = 64 A); and catalase, 19.8 (R.
= 52 A).
Biochemistry: Simonds et al. Proc. Natl. Acad. Sci. USA 77 (1980) 4625
Table 1. Binding affinities of opioids to receptors of NG108-15
cells in soluble and membrane-bound forms
Opioid 1 mM CHAPS 1 mM CHAPS No CHAPS
DALAMID 2 2 2
f3-Endorphin 2 2 2
Etorphine 2 2 0.4
Levorphanol 20 10 4
Naloxone 50 30 13
(-)-Morphine 80 70 40
(+)-Morphine >5000 >4000-. >4000
'06- Dextrorphan >5000 1000 500
Values shown are dissociation constants (Kd) in nM. Kd values for
[3H]DALAMID were determined from direct and Scatchard plots of
0.4- specific binding. The other Kd values were determined from the
concentrations required to lower specifically bound [3H]DALAMID
to 50% (EDro):
Kd = EDso + (1 + [[KH]DALAMID])
0.2 _ Binding to solubilized receptors was measured by gel filtration.
0.5 1.0 Binding to membrane receptors was determined by measurement of
the amount of ligand associated with the 20,000 rpm pellet (8);
therefore, any solubilized activity generated by 1 mM CHAPS will
not contribute to the measured membrane activity.
0 1 2 3 are dissociation constants for membrane-bound receptors in the
DALAMID, M X 108 presence and absence of 1 mM CHAPS. The affinities of the
opioids for the soluble receptor were similar or identical to the
FIG. 3. Specific binding of [3H]DALAMID to solubilized recep- affinities for the membrane receptors measured in the presence
tors as a function of DALAMID concentration. A 10 mM CHAPS
extract of NG108-15 membranes was diluted to 2 mM CHAPS and of detergent. It appears that the affinities of alkaloid, but not
0.4 mg of protein per ml, and aliquots were assayed for binding of 2 peptide, opioids were reduced somewhat by the zwitterionic
nM [3H]DALAMID. (Inset) Scatchard analysis of the same data. detergent. The binding affinites of the alkaloid agonists tested,
Ordinate, ratio of specifically bound DALAMID to free DALAMID, etorphine, levorphanol, and (-)-morphine, closely paralleled
250 units per division; abscissa, specifically bound DALAMID in their analgesic activity in humans which, on a relative scale, are
pmol/mg of protein. 200, 5, and 1, respectively (15). Dextrorphan, the inactive en-
antiomer of levorphanol, had a reduced affinity for the solu-
MID displaced the same amount of bound opioid. Thus, the two bilized receptor. The opioid antagonist naloxone bound to sol-
ligands appear to be interacting with the same receptor popu- uble and membrane receptors with the same affinity.
lation. Treatment of the solubilized receptors from NG108-15
The ability of several alkaloid and peptide opioids to compete membranes with trypsin destroyed their ability to bind both
with [3H]DALAMID for occupancy of the soluble receptor- DALAMID and etorphine (Table 2). Thus, protein is an es-
binding sites was measured. The dissociation constants calcu- sential constituent of the solubilized receptors, as has been
lated from these experiments are listed in Table 1. Also listed shown for membrane-bound receptors (3, 6, 16). Table 2 also
shows that treatment of the solubilized receptors with N-eth-
300 ylmaleimide resulted in their inactivation, presumably through
alkylation of an essential thiol group on the protein (6).
There is evidence that the solubilized receptor may not be
C.l Table 2. Effects of trypsin and N-ethylmaleimide treatment on
activity of opioid receptors solubilized from
.0 NG108-15 cell membranes
fmol bound fmol bound
Control* 8.3 8.2
Trypsin* 0.3 1.9
I I I$
* Aliquots (1 ml) of a CHAPS extract of NG108-15 membranes di-
7e) luted 1:10 (with 0.32 M sucrose/10 mM Tris-HCl, pH 7.5) were in-
0 10-8 10-7 10-6 10-5
Competing ligand, M cubated for 30 min at 370C with or without 100 ug of trypsin treated
with N-tosylphenylalanine chloromethyl ketone (Worthington) and
FIG. 4. Displacement of bound [3H]DALAMID from solubilized assayed for specific binding with either [3H]DALAMID or [3H]-
NG108-15 receptors by different concentrations of (-)-morphine (0) etorphine (2 nM of each).
and (+)-morphine (-) and by a saturating amount of unlabeled t Aliquots of a diluted CHAPS extract of NG108-15 membranes were
DALAMID (-). Assay conditions were as described in the legend to assayed for specific [3H]DALAMID binding in the presence or ab-
Fig. 3. Data are means ± SEM of quadruplicate determinations. sence of 1 mM N-ethylmaleimide.
4626 Biochemistry: Simonds et al. Proc. Natl. Acad. Sci. USA 77 (1980)
completely disaggregated. As shown in Fig. 5, optimal binding
activity required the presence of detergent in the assay, and VO
high concentrations of detergent were inhibitory. The re-
quirement of detergent for receptor activity suggests that lipid 800F
may normally be associated with the functional receptor. Loss s
of binding activity at high detergent concentrations may rep-
resent dissociation into inactive subunits or direct competition 600 V
at the binding site. The complete loss of all binding activity as 0
the concentration of CHAPS was raised from 5 to 10 mM makes r
a competitive mechanism unlikely. .0
Active opiate receptors were also solubilized from rat and
bovine brain membranes by treatment with CHAPS. Because
of the high peptidase activity associated with brain membranes, 200 F
pancreatic trypsin inhibitor and phenylmethylsulfonyl fluoride
were included during the solubilization. Pilot experiments with
rat and beef brain membranes showed that 10 mM CHAPS was II I
nearly optimal for solubilization of active receptors, as it was 10 15 20 25
for the hybrid cell membranes. Opiate receptors solubilized Fraction
from rat brain exhibited the same stereospecificity that char- FIG. 6. Sepharose CL-6B gel filtration of the 130,000 X g su-
acterizes membrane receptors. The Kd for (-)-morphine was pernatant fraction of a CHAPS extract of rat brain membranes. One
near 80 nM, whereas there was no evidence of (+)-morphine milliliter of 10 mM CHAPS extract (2.7 mg of protein) was applied
binding at concentrations up to 10 ,M. Solubilized receptors to the column. Conditions were as described in the legend to Fig. 2
from rat and bovine brain migrated on Sepharose 6B gel fil- except that the flow rate was appreciably slower. Total (0) and
nonspecific (&) binding of [3H]etorphine (5 nM) were determined on
tration as complexes with Stokes radii near 70 A. By this crite- 350-gl aliquots.
rion, receptors from these two sources are similar to the receptor
solubilized from NG108-15 membranes. The Sepharose 6B
elution profile of a solubilized receptor preparation from rat Sepharose 6B. The solubilized binding activity has the char-
brain membranes is shown in Fig. 6. The figure shows total and acteristics expected of true opiate receptors: (i) reversible
nonspecific [3H]etorphine binding in each fraction. Essentially binding of ligands; (ii) high affinity, with binding constants in
all the specific binding, which is the difference between the two the nanomolar range for the best ligands; (i) stereospecificity
curves, was included in the column after the void volume and such that the (+)-enantiomers of morphine and levorphanol
was therefore truly soluble. There was no evidence for size display affinities many orders of magnitude weaker than the
heterogeneity among brain receptors. Far more discriminating (-)-enantiomers; (iv) pharmacological relevance, because af-
techniques, however, will be necessary for an examination of finities of opiate agonists tested are in line with the known
the question of receptor heterogeneity. pharmacological potencies (15); and (v) chemical specificity,
DISCUSSION because both opioid peptides and opiate alkaloids bind as ex-
pected for the opiate receptor.
The results demonstrate that opiate-binding material has been We do not know the extent to which the detergent used here,
solubilized in an active state by CHAPS treatment of mem- CHAPS, is unique in its ability to solubilize active opiate re-
branes prepared from NG108-15 hybrid cells and from rat and ceptors. The detergent is unique in that it is a zwitterionic de-
bovine brain. The binding activity is judged to be truly soluble rivative of the steroid cholic acid. As such, it presumably ex-
because it does not sediment when subjected to centrifugation hibits the high critical micelle concentration and low aggre-
at 160,000 X g for 1 hr and because the activity is included gation number of bile salt-detergents and lacks the denaturing
between two soluble protein markers upon gel filtration with action of ionic detergents. Trial solubilizations of beef brain
membranes with digitonin, a nonionic steroid detergent, or with
the n-decyl zwitterionic detergent, Zwittergent (3-10), pro-
I duced no active receptors, whereas CHAPS solubilized active
receptors under the same conditions (data not shown). Thus,
400 the combination of the zwitterionic group and steroid nucleus
in the same molecule may be critical. Sodium cholate is also
incapable of solubilizing active opiate receptors. However,
^ 300 sodium cholate and CHAPS solubilize protein with comparable
0 efficiencies, and electrophoretic analysis of the solubilized
proteins shows that both detergents solubilize each of the major
protein bands visible in the untreated membranes. The state
of the receptors after solubilization with CHAPS is apparently
more nearly native than after treatment with sodium cholate
or the many other detergents that have hitherto been tested as
opiate receptor-solubilization agents.
There is an optimal detergent concentration for solubilization
0 2 4 6 8 10 of active receptors, which suggests that higher detergent con-
CHAPS, mM centrations may disaggregate essential components of the
FIG. 5. Specific binding of [3HJDALAMID to solubilized binding complex or that an inhibitor is unmasked. Assay of
NG108-15 receptors as a function of CHAPS concentration in the solubilized receptor also requires detergent and is inhibited at
assay. The data are means I SEM of quintuplicate determinations high detergent concentrations. Inhibition of binding may be
of cpm per 230 ;ig of protein. due to disaggregation of the receptor complex or direct com-
Biochemistry: Simonds et al. Proc. Natl. Acad. Sci. USA 77 (1980) 4627
petition with ligand. The latter is unlikely because of the W.F.S. is a Medical Student Research Fellow of the Pharmaceutical
complete loss of binding activity with only a 2-fold change in Manufacturer's Association Foundation. G.K. is a Pharmacology Re-
detergent concentration. Thus, the solubilized receptors may search Associate, National Institute of General Medical Sciences.
not be in their ultimate state of dissociation, and complete
dissociation may lead to inactivation. Approximately 20% of
the receptors originally present in the membranes are recovered 1. Simon, E. J., Hiller, J. M. & Edelman, I. (1975) Science 190,
as solubilized material even though the residual membranes 389-390.
retain only 10% of their binding activity. It seems likely that 2. Zukin, R. S. & Kream, R. M. (1979) Proc. Natl. Acad. Sci. USA
a large part of the remaining activity has been lost due to a more 76, 1593-1597.
complete, and poorly reversible, dissociation of the receptor 3. Klee, W. A. & Nirenberg, M. (1974) Proc. Natl. Acad. Sci. USA
4. Pert, C. B. & Snyder, S. H. (1973) Science 179, 1011-1014.
The receptor complex isolated in these studies is quite large. 5. Terenius, L. (1973) Acta Pharmacol. Toxicol. 32,317-320.
The measured Stokes radius of 70 A indicates that the molecular 6. Simon, E. J., Hiller, J. M. & Edelman, I. (1973) Proc. Natl. Acad.
weight of the complex of receptor and detergent is probably Sci. USA 70, 1947-1949.
in the range of 1-5 X 105. Protein is an essential component of 7. Hjelmeland, L. M. (1980) Proc. Natl. Acad. Sci. USA, in press.
this complex because binding activity is lost after treatment with 8. Klee, W. A. & Streaty, R. A. (1974) Nature (London) 248,61-
trypsin or alkylation with N-ethylmaleimide. The observation 63.
that CHAPS is necessary for binding activity suggests that the 9. Salomon, Y., Londos, C. & Rodbell, M. (1974) Anal. Biochem.
opiate receptor is an integral membrane protein (17). 10. Sharma, S., Nirenberg, M. & Klee, W. A. (1975) Proc. Nat!. Acad.
Our experiments with the solubilized receptors have shown Sci. USA 72,590-594.
that their concentration is approximately 1 pmol/mg of soluble 11. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.
protein. Purification of functional receptors is thus an attainable (1951) J. Biol. Chem. 193,265-275.
goal. The purification of active receptors will furnish the means 12. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672.
to directly answer the question of the number of different types 13. Goldstein, A., Lowney, L. I. & Pal, B. K. (1971) Proc. Nat!. Acad.
of opiate receptors (18) in brain and other tissues. Scd. USA 68,1742-1747.
14. Jacquet, Y. F., Klee, W. A., Rice, K. C., Iijima, I. & Minamikawa,
Although opiate receptors are normally coupled to adenylate J. (1977) Science 198, 842-845.
cyclase in NG108-15 cell membranes, the solubilized receptors 15. Kosterlitz, H. W., Waterfield, A. A. & Berthand, V. (1974) in
are uncoupled from the enzyme. Opportunities are thus created Narcotic Antagonsts, eds. Braude, M. C., Harris, L. S., May, E.
for studying the mechanism of opiate receptor-adenylate cy- L., Smith J. P. & Villarreal, J. E. (Raven, New York), pp. 319-
clase coupling by suitable reconstitution experiments with 334.
purified, soluble receptors and purified, soluble enzyme. Such 16. Pasternak, G. W. & Snyder, S. H. (1973) Mol. Pharmacol. 10,
studies would shed light on the mechanism of acute opiate ac- 17. Singer, S. J. (1974) Annu. Rev. Biochem. 43,805-833.
tions as well as on the phenomena of tolerance and depen- 18. Lord, J. A. H., Waterfield, A. A., Hughes, J. & Kosterlitz, H. W.
dence. (1977) Nature (London) 267,495-499.