- Email: [email protected]
Evidence for a new type of opioid binding site in the brain of the frog Rana ridibunda
European Journal of Pharmacologp, 150 (1988) 75-84
75
Elsevier EJP 50246
Evidence for a new type of opioid binding site in the brain of the frog Rana ridibunda C a t h e r i n e M o l l e r e a u , A n n e P a s c a u d , G i l b e r t Baillat, H o n o r 6 M a z a r g u i l , A l a i n P u g e t a n d Jean-Claude Meunier * Laboratoire de Pharmacologic et de Toxicologic Fondamentales, CNRS, 205 Route de Narbonne, 31400 Toulouse, France
Received 29 July 1987, revised MS received 7 January 1988, accepted 16 February 1988
The crude membrane fraction from the brain of the frog Rana ridibunda was shown to contain 0.7-0.8 pmol/mg protein for a site with high (K D = 0.1 nM) and about 3.2 pmol/mg protein for a site with lower (K D = 10-15 nM) affinity for the opiate agonist [3H]etorphine and for the opiate antagonist [3H]diprenorphine. In addition to its very high affinity for the two tritiated oripavine derivatives, the high affinity site displayed (i) a considerably reduced ability to bind the agonist but not the antagonist in the presence of Na + ions and (ii) pronounced stereospecificity. These properties are all typical of an opioid receptor site. The lower affinity site, which was about four times as abundant as the other exhibited none of the aforementioned characteristics and is therefore probably not opioid in nature. Detailed testing of the potency of various unlabelled opioid ligands to inhibit the binding of [3H]etorphine at the high affinity site showed that the latter consists of a mixture of several types of opioid sites, including a major type with an apparent binding profile clearly different from those of mammalian brain /1, 8- and K-opioid sites. In particular, this major type of site, which accounted for about 70% of the opioid binding in frog brain membranes, bound /t ([D-Ala2,MePhea,GlyolS]enkephalin), 8 ([D-Thr2,LeuS]enkephalyl-Thr) and x (U50,488) selective ligands with much lower affinity than did /~-, 6- and r-opioid receptor sites, respectively. [3 H]Etorphine binding;
[3 H]Diprenorphine binding;
(High (opioid) and lower (non-opioid) affinity sites, Neither/~, nor 6 or x opioid receptor type, Amphibian brain)
1. Introduction Opioid substances elicit a wide range of pharmacological effects by interacting primarily with three major types (/~, 8 and x) of specific recognition sites in the nervous system. The /~, 8 and ~ sites display clearly distinct binding characteristics ( L o r d et al., 1977; C h a n g and Cuatrecasas, 1979; Kosterlitz et al., 1981; C h a n g et al., 1981) and regional distributions ( G o o d m a n et al., 1980; G o o d m a n and Snyder, 1982) indicating that they are carried by structurally distinct,
* To whom all correspondence should be addressed.
yet p r o b a b l y related, receptor subunits. A fourth type of opioid site, the fl-endorphin-preferring type has also been characterized in rat vas deferens (Schulz et al., 1981) and brain (Chang et al., 1984). Experimental access to the molecular basis of opioid receptor multiplicity will require separate purification of the various types in sufficient quantities. This is a difficult goal since opioid receptors are present in trace amounts and often as a mixture of several types in nerve tissue preparations. Moreover, opioid receptors, in particular those f r o m m a m m a l i a n sources, are not easily obtained as the active form in detergent solution: the yields obtained with C H A P S (3-[(3-cholami-
0014-2999/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
76 dopropyl)dimethylammonio]-l-propanesulfonate) are very low (Simonds et al., 1980) while digitonin has to be used under conditions (high salts) that interfere with the subsequent binding of agonists (Howells et al., 1982). It may prove to be advantageous, for these and other reasons, to use nonmammalian sources. For instance, amphibian brain membrane preparations display high levels of opiate binding (Ruegg et al., 1981) and opiate binding activity from amphibian brain is readily obtained in solution with reasonably high yields by means of digitonin alone (Ruegg et al., 1981). Therefore, the amphibian brain seems to be a suitable starting material for the purification of opioid receptors by affinity chromatography. We have now examined the equilibrium binding characteristics of the opiate agonist [3H]etorphine and of the opiate antagonist [3H]diprenorphine in a crude membrane fraction from the brain of the frog Rana ridibunda. [3H]Etorphine and [3H]diprenorphine specifically labelled two distinct classes of sites in frog brain membrane preparations. The two classes were clearly different on the basis of not only (i) affinity for the two radioligands but also (ii) sensitivity to allosteric effectors, Na + ions in particular and (iii) stereospecificity. The lower affinity site appeared not to be opioid in nature while the high affinity site consisted of a mixture of several types of opioid sites, including a prominent type which was neither ~, nor 8 or ~ as defined in mammalian nerve tissue.
2. Materials and methods
2.1. Animals Live frogs of the species R. ridibunda and weighing 50-100 g were purchased from Arbona and Novier (Bourg en Bresse, France). 2.2. Preparation of the crude membrane fraction The animals were killed by decapitation and their brains were rapidly excised and processed at 0-4°C. One gram of wet tissue ( - 10 brains) in 50 mM Tris-HC1, pH 7.4 (final volume: 12 ml) was
homogeneized in a Potter-Elvehjem tissue grinder with 20 strokes of a teflon pestle, motor-driven at 800-1000 r.p.m. The resulting homogenate was processed further without prior incubation at 3 7 ° C since omitting this step had no measurable effect on the subsequent binding properties of the final crude membrane fraction. The homogenate was centrifuged at 30000 r.p.m, in a Beckman rotor type 30. The supernatant was discarded and the pellet was briefly dispersed (Polytron) in a large excess of ice-cold buffer and centrifuged again as before. The resulting pellet was homogenized (Polytron) in buffer (final volume: 12 ml) to yield the crude membrane fraction with a protein concentration of about 5 m g / m l as estimated by the method of Lowry et al. (1951). 2.3. Equilibrium binding studies (25 ° C) Equilibrium binding studies on the crude membrane fraction from frog brain were carried out at a final protein concentration of 0.1 m g / m l in 50 mM Tris-HC1 pH 7.4. Each incubation mixture (0.5 ml) in triplicate contained 50/xg of membrane protein and the radioligand at the desired concentration. Non-specific binding was assayed in triplicate in the presence of 10 I~M diprenorphine and was found to be proportional to the radioligand concentration in the range 0.04-8 nM. A 60-min incubation was followed by rapid filtration of the contents of each tube on glass fiber discs (Whatman G F / B ) on Millipore model 1225 sampling manifolds. The filters were rinsed with three 3 ml portions of ice-cold buffer, dried for 15 min under an IR lamp and were counted for tritium radioactivity at about 50% efficiency in 3 ml of MP Ready-Solv (Beckman) cocktail. The counter was a Kontron model MR300 Automatic Liquid Scintillation System. Competition experiments were carried out essentially as described above except that total and non-specific binding of the radioligand were assayed in six replicates, fl-Endorphin, dynorphin(1-13) and [D-Prol°]dynorphin-(1-11) were tested for competition against [3H]etorphine binding in polypropylene tubes (to minimize adsorption to tube walls) in the presence of 10 /zM bestatin, 1 mM L-leucyl-L-leucine and 50 ktM N-carboxyme-
77 TABLE 1 Binding to be measured
In test tube at equilibrium
Retained on filter
Total qT = q s + q N s + q F (1) Q = qs + q N s + ~t'qF Non-specific q-r = qr~s + qF (2) Q ' = qNs + a.q~.
tyl-L-phenylalanyl-L-leucine (to prevent degradation by endogenous peptidases). In the filtration binding assay, radioligands bind to glass fiber. The contribution of this undesirable retention can easily be formulated in very general terms assuming that (i) bound radioligand is totally retained on the filter and (ii) free radioligand binds to the filter in direct proportion, a (0 < a < 1), to the amount added. q (Q) refers to the amount of radioligand: qv to total in test tube; qs and qNs to specifically and non-specifically bound, respectively; qF and q~ to free; Q and Q' to bound to filter. Combining (1) and (2) yields: qv = qv + qs. Specific binding is usually taken as: Q - Q' = qs + (qF - qv) = (1 a ) ' q s < qs, while free radioligand is calculated from: Q T - Q ' = q v - a ' q v = ( 1 - a ) ' q v
2.4. Chemicals [15,16-3H]etorphine (30-60 Ci/mmol; 1 Ci = 37 GBq), [15,16(n)-3H]diprenorphine (25-30 Ci/mmol) and [3,5-3H]Tyr-D-Ala-Gly-MePhe Glyol ([3H]DAGO, 30-60 Ci/mmol): Amersham (Amersham, England); etorphine and diprenorphine: Reckitt and Colman (Kingston upon Hull, England); levorphanol and dextrorphan: Hoffmann La Roche (Basel, Switzerland); naloxone: Endo (Garden City, NY, USA); ethylketocyclazocine: Sterling Winthrop (Rensselaer, NY,
USA); morphine: Francopia (Paris, France). TyrD-Ala-MePhe-Glyol (DAGO), Tyr-D-Thr-GlyPhe-Leu-Thr (DTLET) and bestatin: Cambridge Research (Harston, England); trans-3,4-dichloroN-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide (U50,488): Upjohn (Kalamazoo, MI, USA); fl-endorphin and L-leucyl-L-leucine: Bachem (Bubendorf, Switzerland); dynorphin-(113), [D-PrcC°]dynorphin-(1-11) and N-carboxymethyl-L-phenylalanyl-L-leucine were generously supplied by Drs. H. Mazarguil and J.-E. Gairin (Toulouse, France). 5'-Guanylylimidodiphosphate (GppNHp) was from Sigma (St Louis, MO, USA).
3. Results
3.1. Equilibrium binding properties of [ 3H]etorphine and of [3H]diprenorphine in the crude membrane fraction from the brain of R. ridibunda The equilibrium adsorption isotherms of the two non-selective opiates, the agonist [3H]etorphine and the antagonist [3H]diprenorphine were clearly biphasic. Scatchard transforms of the data (fig. 1A) indicated that the preparation contained (i) (0.76 _+ 0.16 p m o l / m g protein (n = 5) of a high affinity site that bound the two radioligands with nearly the same high affinity (K D = 0.13 + 0.02 nM) and, (ii) 3.2 +_ 0.2 p m o l / m g protein (n = 5) of a lower affinity site that bound the two radioligands with about the same lower affinity (K D = 13 _+ 4 nM). The fact that the low affinity site had a low affinity for the two tritiated oripavines was already indicative of its non-opioid nature. Nevertheless, we attempted to further differentiate the high and low affinity sites on the basis of other properties typical of opioid interactions, namely (i) the differential allosteric regulation of agonist and antagonist binding by sodium ions a n d / o r by guanine nucleotides (Pert et al., 1973; Childers and Snyder, 1980; Frances et al., 1985) and (ii) stereospecificity (Goldstein et al., 1971). The affinity of the antagonist [3H]diprenorphine for the high affinity site was slightly (1.5-fold) increased in the presence of sodium ions (120 mM NaC1) while high affinity binding of the
78
c B *120mMNaCI
+120ram NaCI + 5OpM GppNHp
+5OHM GppNHp
t ~
50~IHMT!~is"HCI
1
2
0
1
2 0 1 b, I~ol/mg protein
2
0
1
2
Fig. 1. Scatchard transforms of the equilibrium adsorption isotherms (25 o C) of [3H]etorphine (e) and of [3H]diprenorphine (m) in
the crude membrane fraction from the brain of Rana ridibunda. The concentrations of [3 H]etorphine and of [3H]diprenorphine were increased from 0.04 nM to 8 nM. b = amount of specificallybound radioligand, f(nM): concentration of free radioligand. The insert in fig. 1A shows the Scatchard transform of saturation data generated by isotopic dilution of 0.1 nM [3H]etorphine. Saturation experiments with isotopic dilution were aimed at measuring Bma x values with some precision ( B m a x - 4 pmol/mg protein). All other data shown here were obtained with the same membrane preparation.
agonist [3H]etorphine was still observed (K D = 0.13 + 0.04 nM, n = 4) but to a considerably reduced maximum number (0.17 + 0.09 p m o l / m g protein) of sites (fig. 1B). Under these conditions, equilibrium saturation binding of the two radioligands at the lower affinity sites did not seem to have been affected. Only the analogue of GTP, G p p ( N H ) p , (50 /~M) modified neither the high nor the low affinity binding of [3H]etorphine and of [3H]diprenorphine (fig. 1C). Yet, when present together with NaC1, G p p ( N H ) p appeared to prevent the sodium-induced increase in affinity of [3H]diprenorphine for the high affinity site and further decreased [3H]etorphine binding at this site. In the simultaneous presence of G p p ( N H ) p and of NaC1, high affinity binding of the tritiated opiate agonist was no longer detected while low affinity binding of the two radioligands (K D = 10.5 nM, Bmax = 3.1 p m o l / m g protein) was still observed (fig. 1D). Finally, we examined the ability of the two enantiomers, levorphanol ( - ) and dextrorphan ( + ) to inhibit the binding of [3H]etorphine at the two sites. When [3H]etorphine 0.1 nM was used so that, at equilibrium and in the absence of competing drug, > 90% of the radioligand was bound at
the high affinity site and < 10% at the lower affinity site, levorphanol ( K I = 3 nM) inhibited the binding of [3H]etorphine (fig. 2A) about 1 700 times more potently than dextrorphan (K l = 5 100 nM). When [3H]etorphine 2 nM was used, so that at equilibrium and in the absence of competing drug, about 62% of the radioligand was bound at the high affinity site and about 38% at the lower affinity site, the inhibition curves for the two enantiomers were both clearly biphasic yet with an inflexion at about 65% inhibition in the case of levorphanol and at about 35% inhibition in the case of dextrorphan (fig. 2B). These seemingly conflicting results could easily be reconciled if it was assumed that levorphanol displayed a higher apparent affinity ( I 3 2 . 5 = 56 nM; K I = 3.4 nM) for the [3H]etorphine high affinity site than (I82~ = 1600 nM, K I = 1400 nM) for the lower affinity site while, conversely, dextrorphan exhibited a higher apparent affinity 017.5 = 2000 nM; K I = 1 800 nM) for the [3H]etorphine lower affinity site than ( I 6 7 . 5 = 100000 nM; K l = 6 100 nM) for the high affinity site. Again, the [3H]etorphine high affinity site showed considerable stereospecificity, with 1 800-fold higher apparent affinity for levorphanol than for dextrorphan, whereas the
79 A
C 4
b
bo
• control V
~EX
LEV ~DEX
.5
:
\
'\ \
9 8 7 6 5 4
I
I
I "t~ i
8 7 6 5 luraabeled ligandl , -IogM
..ol~ 2
1
"1--
4
0
3
0
1 2 b, pmoUm 9 proton
Fig. 2. Inhibition of [3H]etorphine binding by levorphanol (*, LEV) and by dextrorphan (za, DEX) in the crude membrane fraction from the brain of Rana ridibunda. Competition studies. (A) [3H]Etorphine was used at the fixed concentration of 0.1 nM, so that at equilibrium, 90% of radioligand was bound at the high affinity site. (B) [ 3H]Etorphine was used as the fixed concentration of 2 nM so that, at equilibrium, about 62% of the radioligand was bound at the high affinity site and about 38% at the lower affinity site. In A and B, b 0 and b are the amounts of specifically bound radioligand in the absence and in the presence of unlabelled ligand, respectively. Saturation study. (C) Scatchard transform of the equilibrium adsorption isotherm of [3H]etorphine in the absence and in the presence of 10/~M levorphanol or of 10/~M dextrorphan. The concentration of [3H]etorphine was increased from 0.04 nM to 8 nM. b = amount of specifically bound radioligand, f(nM): concentration of free radioligand. There was no detectable high or low affinity specific binding with [3H]etorphine in the presence of 10 /~M levorphanol. Dextrorphan, 10 /~M, selectively suppressed the specific binding of [3H]etorphine at the lower affinity site while only slightly inhibiting (competitively) the specific binding of [ 3H]etorphine at the high affinity site.
[3H]etorphine lower affinity site did not. The apparent dissociation constants were nearly equal, 1 400 and 1600 nM for levorphanol and dextrorphan, respectively. Equilibrium saturation binding studies revealed that 10 /~M levorphanol totally prevented the interaction of [3H]etorphine, not only with the high, but also with the lower affinity sites while 10/~M dextrorphan selectively blocked binding of the radioligand at the lower affinity site (fig. 2C). Under these conditions (10/~M dextrorphan) there was only a moderate, competitive inhibition of high affinity [3H]etorphine binding with a slightly reduced K D value (0.35 nM) and an unchanged Bmax (0.72 pmol/mg protein).
3.2. ldentity of the opioid high affinity site in the crude membrane fraction from the brain of Rana ridibunda In order to identify the high affinity site as a /~-, 8- or x-opioid receptor site, we examined the ability of various unlabelled opioid ligands to compete with [3H]etorphine binding in frog brain
\Tx,l~i~k -
~
\. I\ 1~ IXX~_
b°
~, J'-\] ~ - \?"J ~. ~ ~ ~ , ~ \ ~
11
10
"~xl
• DIP X
~k
~ ~
9 8 7 6 lunlabeled I~ndl, -IogM
EKC
MOR '~ DAGO * DTLET ~
\ o
5
4
Fig. 3. Inhibition of [3H]etorphine binding by various unlabelled opioid ligands in the crude membrane fraction from the brain of Rana ridibunda. [3H]Etorphine was used at the fixed concentration of 0.1 nM. At equilibrium, total binding of the radioligand amounted to 1470+204 c.p.m. (n = 7) while non-specific binding (measured in the presence of 10 /~M diprenorphine) amounted to 129 + 17 c.p.m. (n = 7). b 0 and b are the amounts of radioligand specifically bound in the absence and in the presence of unlabelled ligand, respectively, Each b / b 0 value is the mean + S.D. of the values measured in three different preparations. ETO: etorphine, DIP: diprenorphine, NAL: naloxone, EKC: ethylketocyclazocine, MOR: morphine.
80 TABLE 2 Parameters characterizing the inhibition of binding of [3H]etorphine (0.1 nM) by various unlabelled opioid ligands in the crude membrane fraction from the brain of R. ridibunda. Apparent Hill coefficient and IC50 (concentration of unlabelled ligand causing 50% inhibition) values were calculated by linear regression analysis of modified Hill transforms, log b / ( b 0 - b ) = f [log (concentration of inhibitor)] of the inhibition curves, most of which are shown in figs. 2 and 3. K I values were calculated from K I - I C 5 0 (I+L/Ko) 1, L being the concentration of free [3H]etorphine (0.07-0.08 nM) and K D (0.13 nM), the equilibrium dissociation constant of the [3 H]etorphine macromolecular complex. The K 1 values represent true dissociation constants only for those unlabelled ligands for which the modified Hill transform of inhibition curves had a slope close to unity. The values shown in parentheses ( _+S.D., n = 3) were obtained from inhibition experiments carried out in the presence of D A G O (10 nM) + DTLET (20 nM) + U50,488 (20 nM), a mixture ensuring nearly complete blockade of ~t, 8- and ~-opioid sites. Note that, under these conditions, the apparent Hill coefficients increased significantly, especially in the case of DAGO and of DTLET, to values close to 1. Unlabelled ligand Etorphine Diprenorphine Naloxone EKC Morphine
IC50 (nM)
Apparent Hill coefficient
Apparent K l (nM)
0.4 0.4 3.2 9.6 37
1.04 1.03 0.68 0.82 0.65
Levorphanol Dextrorphan
4.5 7900
0.71 0.83
3 5130
D A G O (/Q DTLET (8) U50,488 (K)
176 (88_+ 28) 333 (182_+ 58) 1 730 (1975 _+487)
0.47(0.93_+0.02) 0.57 (0.96_+0.07) 0.80 (0.88 _+0.11)
114 (57_+ 18) 216 (118_+ 38) 1 123 (1 280 _+320)
/3-Endorphin Dynorphin 1-13 D P D Y N (K)
44 179 115
membranes. [3H]Etorphine was used at the fixed concentration of 0.1 nM so that, at equilibrium, > 90% of the radioligand was bound at the high affinity site and < 10% at the lower affinity site. The results of these competition experiments are shown in fig. 3 and in table 2. The high affinity site was found to display high affinity not only for the two oripavine derivatives, etorphine and diprenorphine, but also for several other classical opiate ligands such as naloxone, levorphanol, ethylketocyclazocine and morphine. Yet, the high affinity [3H]etorphine site in frog brain membranes appeared to be neither ~t, nor 8 or J¢: 176 nM DAGO, a /~ selective ligand (Kosterlitz and Paterson, 1981), 333 nM DTLET, a ~ selective ligand (Zajac et al., 1983) and 1 730 nM U50,488, a K selective ligand (Von Voigtlander et al., 1983) were necessary to halve the binding of [3H]etorphine, while nanomolar concentrations either of D A G O or of D T L E T or of U50,488
0.79 0.74 0.83
0.26 0.26 2.1 6.2 24
28 116 75
would have sufficed had the high affinity site been either/~, or 6 or ~, respectively. The slopes of modified Hill transforms of the inhibition curves, especially those for D A G O and D T L E T were clearly less than unity (table 2) indicating that 0.1 nM [3H]etorphine had labelled a mixture of several types of opioid sites: 15-20% of [3 H]etorphine binding persisted in the presence of as much as 10 /xM D T L E T (fig. 3). This suggested that, in addition to the lower affinity site ( < 10%), there were very small amounts (5-10%) of 'enkephalin-resistant' opioid sites (Meunier et al., 1983) which could be K sites. The inhibition curve of U50,488, with a minor component (5-10%) that no longer bound [3H]etorphine at concentrations of the x selective ligand in the range 1-10 nM supported this possibility. Similarly, the curve for D A G O inhibition of [3H]etorphine binding showed a plateau around 10-15%. This plateau value was reached at concentrations of the/~ selec-
81
.10
m
b : 0.06 f
,,L
0.7 + 0 . 7 4
f f+60
.05
•01
--
I 0
I
.1 .2 b , pmol/mg protein
,3
Fig. 4. Scatchard transform of the equilibrium adsorption isotherm ( 2 5 ° C ) of [3H]DAGO in the crude m e m b r a n e fraction from the brain of R. ridibunda. The concentration of [3H]DAGO was increased from 0.1 n M to 10 nM. b = a m o u n t of specifically b o u n d radioligand, f(nM): concentration of free radioligand at equilibrium. The equation is that of the saturation function that best fits the experimental data, assuming that [3H]DAGO b o u n d to two non-interacting classes of sites: a high affinity site which is likely to represent a /~-opioid site (KD = 0.7 nM, Bmax = 0.06 p m o l / m g protein) and a lower affinity site (K n = 60 nM, Bm~ = 0.74 p m o l / m g protein).
tive ligand as low as 10 nM and could therefore reflect binding of [3H]etorphine at a /~-opioid receptor site. This conclusion was supported by the results of saturation binding studies with [3H]DAGO. The crude membrane fraction from frog brain bound about 0.06 p m o l / m g protein of the/~ selective radioligand with high affinity (K D = 0.7 nM) (fig. 4). The inhibition of [3H]etorphine (0.1 nM) binding by DAGO, DTLET and U50,488 was tested in the presence of 10 nM DAGO + 20 nM DTLET + 20 nM U50,488, a mixture that nearly completely blocked the/~, 6 and K sites. Under these conditions, (i) control binding of [3H]etorphine was inhibited 20-30% and (ii) the slopes of modified Hill transforms of the inhibition curves of the three unlabelled competitors, especially those of DAGO and of DTLET, were significantly increased to close to unity (table 2). These results indicated that the radioligand now interacted with a homogenous population of sites. Finally, in order to further characterize the high affinity site as well as to select probes that might be used subsequently in cross-linking ex-
periments, we tested fl-endorphin, dynorphin-(113) and the dynorphin analogue [D-Prd°]dy norphin-(1-11) for their ability to compete with the binding of [3H]etorphine (0.1 nM) in frog brain membranes. The three peptides showed only moderate apparent affinity for the high affinity site. However, [D-Prol°]dynorphin-(1-11), a highly selective probe (Gairin et al., 1984) displayed a considerably lower (1 000-2 000-fold) apparent affinity for the high affinity [3H]etorphine site in the frog brain membranes than it did for ~-opioid sites in guinea pig cerebellum membranes (Gairin et al., 1984). Taken together, these results indicated that, in the preparation under study, there was a major opioid receptor sites that accounted for about 70% of the binding of [3H]etorphine (0.1 nM). The binding characteristics of this predominant site were clearly different from those of mammalian t~-, 8- and ~-opioid sites. In this same preparation, 20-30% of the [3H]etorphine (0.1 nM) binding could be accounted for by a /~ site (10%), a ~ site (5-10%) and by the lower affinity site ( < 10%). There was no clear evidence for the presence of a 6 site as no high affinity binding of the 8 selective radioligand [3H]DTLET could be reliably demonstrated in this preparation (data not shown). 4. D i s c u s s i o n
The present study has shown that the brain of an amphibian, the frog Rana ridibunda, contains a high and a lower affinity site that bind specifically both the opiate agonist [3H]etorphine and the opiate antagonist [3H]diprenorphine. The high affinity site showed (i) high affinity not only for the two tritiated oripavine derivatives but also for other classical opiate ligands, including naloxone and morphine; (ii) a considerably reduced ability to bind the agonist [3H]etorphine but not the antagonist [3H]diprenorphine in the presence of Na + ions and (iii) stereospecificity, all properties that are typical of an opioid site. The lower affinity site which is about four times as abundant as the high affinity site showed none of the above characteristics and is therefore probably not opioid in nature. So far, nothing is known about the identity of the lower affinity site.
82 The data on amphibian opioid receptors are both scarce and conflicting. For instance, the brain of the toad (Bufo marinus) appears to lack the lower affinity site for [3H]diprenorphine. However, this discrepancy is only apparent, since toad brain membranes were selectively assayed by Ruegg et al. (1981) fo stereospecific binding of the tritiated opiate antagonist so that the lower affinity site, which is not stereospecific, was probably overlooked. Ruegg et al. (1981) had first suggested that 'the toad brain contains a predominance of morphine preferring receptors, designated ~ by Lord et al. (1977)'. Later, Simon et al. (1982) found, also in the toad brain, a ' p r e d o m i n a n t opiate receptor subclass.., of the benzomorphanpreferring type'. Interestingly, these authors also noted that the subclass in question, which they referred to as x / o , had ' a somewhat higher affinity for/~ and 6 ligands than its mammalian counterpart', namely, the x type of opioid receptor (Kosterlitz et al., 1981). More recently, Simon et al. (1984) claimed that the brain of R. esculenta contains opioid sites that are mostly of the K type. In this latter study however, the radioligand ([3H]naloxone) was first shown to bind to two classes of sites in frog brain membranes and was then used in competition studies with unlabelled opioid ligands at a concentration that labelled nearly equal amounts of high and of lower affinity sites. It is not clear how these authors came to the conclusion of the existence of a ~-opioid site in this preparation under these conditions. We have now shown, in agreement with the preliminary observation of Simon et al. (1982), that the brain of the frog R. ridibunda contains a predominant opioid site with an in vitro binding profile clearly distinguishable from those of m a m malian #-, 8- and ~-opioid sites. What makes this site (hereafter referred to as Op) so conspicuous is that while it exhibits a considerably lower affinity for the two t¢ selective ligands U50,488 and D P D Y N as Simon et al. (1982) found in toad brain, it has a higher apparent affinity for /~ ( D A G O ) and 6 (DTLET) ligands than do r-opioid sites in mammalian brain membranes (table 3). If the amphibian Op site is clearly not K, as this latter type has been well characterized in guinea pig (Kosterlitz et al., 1981) and rabbit (Meunier et
TABLE 3 Apparent affinity (KI) of/~, 8 and g selective opioid ligands for several types of opioid receptor in mammalian (~, 8, r and '~2') and frog (Op) nerve tissue preparations. The values in parentheses are for apparent affinity at the preferred site. Data are from a Paterson et al. (1984), b Gairin et al. (1984), c Gouarderes et al. (1984) and d the present study. K 1 (nM) ~t-Site a 6-Site a x-Site a ,r2,_Site c Op site d DAGO (~t) (1.8) 423 1927 41 57 DTLET (8) 34 (2.7) 14500 330 118 U50,488 (r) 890 9800 (7.3) 890 1280 DPDYN (K) 2.0 b 7.5 b (0.03)b 75 _
al., 1983) brain membranes, its binding characteristics are reminiscent of those of the so-called '•2' subtype of the K-opioid receptor. The 'x2'-opioid sites were first identified in rat (Gouard~res et al., 1982) and in guinea pig (Attali et al., 1982) lumbo-sacral cord because they had much higher affinity t h a n x-opioid sites for /~ ([DAla2,MePhe4,Met(O)olS]enkephalin) and for 8 ([D-Ala2,D-AlaS]enkephalin and [D-Ser2,LeuS]en kephalyl-Thr) ligands. Interestingly, Weyhenmeyer and Mack (1985) have presented preliminary evidence that the rat brain contains a minor population of opioid sites that appear to be neither /~, nor 6 or x. However, these authors have not reported the apparent affinities of D A G O and of D T L E T for these additional sites thus precluding any direct comparison with the Op site from amphibian brain. The existence of a particular type of opioid binding site in the brain of an amphibian was not totally unexpected because digitonin extracts from amphibian brain membranes display opiate binding activity (see Introduction) whereas digitonin extracts from m a m m a l i a n brain membranes do not. This suggests structural differences between amphibian and m a m m a l i a n opioid receptor types. The physiological significance of the frog brain Op site remains to be elucidated. Pharmacological data on opiates in amphibians are scarce but there is recent evidence ' n o t only for similarities but also for important differences in the pharmacological actions of opiates and antagonists in amphibians and m a m m a l s ' (Pezalla and Stevens,
83 1984). This c o u l d b e r e s t a t e d as ' s i m i l a r i t i e s a n d i m p o r t a n t differences b e t w e e n o p i o i d r e c e p t o r types in a m p h i b i a n s a n d m a m m a l s ' . F o r instance, systemic m o r p h i n e a p p e a r s to b e a w e a k e r analgesic in frogs t h a n in rats while d e x t r o r p h a n , which is essentially inactive in rats elicits a n a l g e s i a in frogs. H o w e v e r , d e x t r o r p h a n - i n d u c e d analgesia in frogs is, unlike m o r p h i n e - i n d u c e d a n a l g e s i a in rats a n d frogs, insensitive to n a l o x o n e (Stevens a n d Pezalla, 1984). These a n d o t h e r d a t a (Pezalla, 1983) c o u l d be t a k e n to i n d i c a t e that responses m e d i a t e d b y o p i o i d receptors c o u l d b e m e d i a t e d via different r e c e p t o r types in m a m m a l s a n d amphibians. Finally, it is w o r t h y of n o t e that the O p site p r e d o m i n a t e s early in the course of v e r t e b r a t e evolution, i.e. in a m p h i b i a n nerve tissue. Also, a l t h o u g h it d i s p l a y s an original b i n d i n g profile it also shares with m a m m a l i a n /~-, 3- a n d K-opioid sites several basic, well preserved, p r o p e r t i e s which i n c l u d e (i) e x t r e m e l y high affinity for certain o r i p a v i n e derivatives, n a m e l y e t o r p h i n e a n d d i p r e norphine, (ii) differential allosteric r e g u l a t i o n of agonist a n d a n t a g o n i s t b i n d i n g b y N a ÷ ions a n d (iii) stereospecificity. It is therefore t e m p t i n g to suggest that the a m p h i b i a n O p site represents the a r c h e t y p e of the o p i o i d site at the origin of the various types of o p i o i d site f o u n d in m a m m a l i a n nerve tissue. T h e r e is i n d e e d p r o o f for the evolution of o p i o i d systems: t o a d (Xenopus laevis) p r o e n k e p h a l i n lacks the [LeuS]enkephalin sequence f o u n d in higher v e r t e b r a t e s ( M a r t e n s et al., 1984). However, there is no such direct evidence in the case of o p i o i d r e c e p t o r types a n d f i r m p r o o f of o u r p r o p o s a l of the a r c h e t y p i c n a t u r e of the a m p h i b i a n O p site awaits sequencing of their genes.
References Attali, B., Ch. Gouardrres, H. Mazarguil, Y. Audigier and J. Cros, 1982, Evidence for multiple 'kappa' binding sites by use of opioid peptides in the guinea-pig lumbo-sacral spinal cord, Neuropeptides 3, 53. Chang, K.-J., S.G. Blanchard and P. Cuatrecasas, 1984, Benzomorphan sites are ligand recognition sites of putative c-receptors, Mol. Pharmacol. 26, 484. Chang, K.-J. and P. Cuatrecasas, 1979, Multiple opiate recep-
tors: enkephalins and morphine bind to receptors of different specificity, J. Biol. Chem. 254, 2610. Chang, K.-J., E. Hazum and P. Cautrecasas, 1981, Novel opiate binding sites selective for benzomorphan drugs, Proc. Natl. Acad. Sci. U.S.A. 78, 4141. Childers, S.R. and S.H. Snyder, 1980, Differential regulation by guanine nucleotides of opiates agonist and antagonist interactions, J. Neurochem. 34, 583. Frances, B., Ch. Moisand and J.C. Meunier, 1985, Na + ions and Gpp(NH)p selectively inhibit agonist interactions at /~ and x-opioid receptor sites in rabbit and guinea-pig cerebellum membranes, European J. Pharmacol. 117, 223. Gairin, J.-E., Ch. Gouard~res, H. Mazarguil, P. Alvinerie and J. Cros, 1984, [D-Prol°]Dynorphin-(1-11) is a highly potent and selective ligand for x-opioid receptors, European J. Pharmacol. 106, 457. Goldstein, A., L.I. Lowney and B.K. Pal, 1971, Stereospecific and nonspecific interactions of the morphine congener levorphanol in subcellular fractions of mouse brain, Proc. Natl. Acad. Sci. U.S.A. 68, 1742. Goodman, R.R. and S.H. Snyder, 1982, ~ Opiate receptors locahzed by autoradiography to deep layers of cerebral cortex: relation to sedative effects, Proc. Natl. Acad. Sci. U.S.A. 79, 5703. Goodman, R.R., S.H. Snyder, M.J. Kuhar and W.S. Young, III, 1980, Differentation of 8 and t~ opiate receptor localizations by light microscopic autoradiography, Proc. Natl. Acad. Sci. U.S.A. 77, 6239. Gouard~res, Ch., Y. Audigier and J. Cros, 1982, Benzomorphan binding sites in rat lumbo-sacral spinal cord, European J. Pharmacol. 78, 483. Howells, R.D., T. Gioannini, J.M. Hiller and E.J. Simon, 1982, Solubilization and characterization of active opiate binding sites from mammalian brain, J. Pharmacol. Exp. Ther. 222, 629. Kosterlitz, H.W. and S.J. Paterson, 1981, Tyr-D-Ala-GlyMePhe-NH(CH2)aOH is a selective ligand for the p-opiate binding sites, Br. J. Pharmacol. 73, 299P. Kosterlitz, H.W., S.J. Paterson and L.E. Robson, 1981, Characterization of the kappa subtype of the opiate receptor in the guinea-pig brain, Br. J. Pharmacol. 73, 939. Lowry, D.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193, 265. Lord, J.A.H., A.A. Waterfield, J. Hughes and H.W. Kosterlitz, 1977, Endogenous opioid peptides: multiple agonists and receptors, Nature (London) 267, 495. Martens, G.J.M. and E. Herbert, 1984, Polymorphism and absence of Leu-enkephalin sequences in proenkephalin genes in Xenopus laevis, Nature (London) 310, 251. Meunier, J.C., Y. Kouakou, A. Puget and Ch. Moisand, 1983, Multiple opiate binding sites in the central nervous system of the rabbit. Large predominance of a mu subtype in the cerebellum and characterization of a kappa subtype in the thalamus, Mol. Pharmacol. 24, 23. Paterson, S.J., L.E. Robson and H.W. Kosterlitz, 1984, opioid receptors, in: The Peptides: Analysis, Synthesis, Biology,
84 eds. S. Udenfriend and J. Meienhofer, (Academic Press, London) 5, 147. Pert, C.B., G. Pasternak and S.H. Snyder, 1973, Opiate agonists and antagonists discriminated by receptor binding in brain, Science (Washington D.C.) 182, 1359. Pezalla, P.D., 1983, Morphine-induced analgesia and explosive motor behavior in a amphibian, Brain Res. 273, 297. Pezalla, P.D. and C.W. Stevens, 1984, Behavioral effects of morphine, levorphanol, dextrorphan and naloxone in the frog Rana pipiens, Pharmacol. Biochem. Behav. 21,213. Ruegg, U.T., S. Cuenod, J.M. Hiller, T. Gioannini, R.D. Howells and E.J. Simon, 1981, Characterization and partial purification of solubilized active opiate receptors from toad brain, Proc. Natl. Acad. Sci. U.S.A. 78, 4635. Schulz, R., M. Wuster and A. Herz, 1981, Pharmacological characterization of the c-opiate receptor, J. Pharm. Exp. Ther. 216, 604. Simonds, W.F., G. Koski, R.A. Streaty, L.M. Hjelmeland and W.A. Klee, 1980, Solubilization of active opiate receptors, Proc. Natl. Acad. Sci. U.S.A. 77, 4623. Simon, E.J.J.M. Hiller, J. Groth, Y. Itzhak, M.J. Holland and
S.G. Eck, 1982, The nature of opiate receptors in toad brain, Life Sci. 31, 1367. Simon, J., M. Szucz, S. Benyhe, A. Borsodi, P. Zeman and M. Wolleman, 1984, Solubilization and characterization of opioid binding sites from frog (Rana esculenta) brain. J. Neurochem. 43, 957. Stevens, C.W. and P.D. Pezalla, 1984, Naloxone blocks the analgesic action of levorphanol but not of dextrorphan in the leopard frog, Brain Res. 301, 171. Yon Voigtlander, P.F., R.A. Lahti and J.H. Ludens, 1983, U-50488: a selective and structurally novel non-/~ (~¢) opioid agonist, J. Pharmacol. Exp. Ther. 224, 7. Weyhenmeyer, J.A. and K.J. Mack, 1985, Binding characteristics of kappa opioids in rat brain. A comparison of in vitro binding paradigms, Neuropharmacology 24, 111. Zajac, J.M., G. Gacel, F. Petit, P. Dodey, P. Rossignol and B.P. Roques, 1983, Deltakephaiin Tyr-D-Thr-Gly-Phe-LeuThr: a new highly potent and fully specific agonist for opiate delta receptors, Biochem. Biophys. Res. Commun. 111, 390.
75
Elsevier EJP 50246
Evidence for a new type of opioid binding site in the brain of the frog Rana ridibunda C a t h e r i n e M o l l e r e a u , A n n e P a s c a u d , G i l b e r t Baillat, H o n o r 6 M a z a r g u i l , A l a i n P u g e t a n d Jean-Claude Meunier * Laboratoire de Pharmacologic et de Toxicologic Fondamentales, CNRS, 205 Route de Narbonne, 31400 Toulouse, France
Received 29 July 1987, revised MS received 7 January 1988, accepted 16 February 1988
The crude membrane fraction from the brain of the frog Rana ridibunda was shown to contain 0.7-0.8 pmol/mg protein for a site with high (K D = 0.1 nM) and about 3.2 pmol/mg protein for a site with lower (K D = 10-15 nM) affinity for the opiate agonist [3H]etorphine and for the opiate antagonist [3H]diprenorphine. In addition to its very high affinity for the two tritiated oripavine derivatives, the high affinity site displayed (i) a considerably reduced ability to bind the agonist but not the antagonist in the presence of Na + ions and (ii) pronounced stereospecificity. These properties are all typical of an opioid receptor site. The lower affinity site, which was about four times as abundant as the other exhibited none of the aforementioned characteristics and is therefore probably not opioid in nature. Detailed testing of the potency of various unlabelled opioid ligands to inhibit the binding of [3H]etorphine at the high affinity site showed that the latter consists of a mixture of several types of opioid sites, including a major type with an apparent binding profile clearly different from those of mammalian brain /1, 8- and K-opioid sites. In particular, this major type of site, which accounted for about 70% of the opioid binding in frog brain membranes, bound /t ([D-Ala2,MePhea,GlyolS]enkephalin), 8 ([D-Thr2,LeuS]enkephalyl-Thr) and x (U50,488) selective ligands with much lower affinity than did /~-, 6- and r-opioid receptor sites, respectively. [3 H]Etorphine binding;
[3 H]Diprenorphine binding;
(High (opioid) and lower (non-opioid) affinity sites, Neither/~, nor 6 or x opioid receptor type, Amphibian brain)
1. Introduction Opioid substances elicit a wide range of pharmacological effects by interacting primarily with three major types (/~, 8 and x) of specific recognition sites in the nervous system. The /~, 8 and ~ sites display clearly distinct binding characteristics ( L o r d et al., 1977; C h a n g and Cuatrecasas, 1979; Kosterlitz et al., 1981; C h a n g et al., 1981) and regional distributions ( G o o d m a n et al., 1980; G o o d m a n and Snyder, 1982) indicating that they are carried by structurally distinct,
* To whom all correspondence should be addressed.
yet p r o b a b l y related, receptor subunits. A fourth type of opioid site, the fl-endorphin-preferring type has also been characterized in rat vas deferens (Schulz et al., 1981) and brain (Chang et al., 1984). Experimental access to the molecular basis of opioid receptor multiplicity will require separate purification of the various types in sufficient quantities. This is a difficult goal since opioid receptors are present in trace amounts and often as a mixture of several types in nerve tissue preparations. Moreover, opioid receptors, in particular those f r o m m a m m a l i a n sources, are not easily obtained as the active form in detergent solution: the yields obtained with C H A P S (3-[(3-cholami-
0014-2999/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
76 dopropyl)dimethylammonio]-l-propanesulfonate) are very low (Simonds et al., 1980) while digitonin has to be used under conditions (high salts) that interfere with the subsequent binding of agonists (Howells et al., 1982). It may prove to be advantageous, for these and other reasons, to use nonmammalian sources. For instance, amphibian brain membrane preparations display high levels of opiate binding (Ruegg et al., 1981) and opiate binding activity from amphibian brain is readily obtained in solution with reasonably high yields by means of digitonin alone (Ruegg et al., 1981). Therefore, the amphibian brain seems to be a suitable starting material for the purification of opioid receptors by affinity chromatography. We have now examined the equilibrium binding characteristics of the opiate agonist [3H]etorphine and of the opiate antagonist [3H]diprenorphine in a crude membrane fraction from the brain of the frog Rana ridibunda. [3H]Etorphine and [3H]diprenorphine specifically labelled two distinct classes of sites in frog brain membrane preparations. The two classes were clearly different on the basis of not only (i) affinity for the two radioligands but also (ii) sensitivity to allosteric effectors, Na + ions in particular and (iii) stereospecificity. The lower affinity site appeared not to be opioid in nature while the high affinity site consisted of a mixture of several types of opioid sites, including a prominent type which was neither ~, nor 8 or ~ as defined in mammalian nerve tissue.
2. Materials and methods
2.1. Animals Live frogs of the species R. ridibunda and weighing 50-100 g were purchased from Arbona and Novier (Bourg en Bresse, France). 2.2. Preparation of the crude membrane fraction The animals were killed by decapitation and their brains were rapidly excised and processed at 0-4°C. One gram of wet tissue ( - 10 brains) in 50 mM Tris-HC1, pH 7.4 (final volume: 12 ml) was
homogeneized in a Potter-Elvehjem tissue grinder with 20 strokes of a teflon pestle, motor-driven at 800-1000 r.p.m. The resulting homogenate was processed further without prior incubation at 3 7 ° C since omitting this step had no measurable effect on the subsequent binding properties of the final crude membrane fraction. The homogenate was centrifuged at 30000 r.p.m, in a Beckman rotor type 30. The supernatant was discarded and the pellet was briefly dispersed (Polytron) in a large excess of ice-cold buffer and centrifuged again as before. The resulting pellet was homogenized (Polytron) in buffer (final volume: 12 ml) to yield the crude membrane fraction with a protein concentration of about 5 m g / m l as estimated by the method of Lowry et al. (1951). 2.3. Equilibrium binding studies (25 ° C) Equilibrium binding studies on the crude membrane fraction from frog brain were carried out at a final protein concentration of 0.1 m g / m l in 50 mM Tris-HC1 pH 7.4. Each incubation mixture (0.5 ml) in triplicate contained 50/xg of membrane protein and the radioligand at the desired concentration. Non-specific binding was assayed in triplicate in the presence of 10 I~M diprenorphine and was found to be proportional to the radioligand concentration in the range 0.04-8 nM. A 60-min incubation was followed by rapid filtration of the contents of each tube on glass fiber discs (Whatman G F / B ) on Millipore model 1225 sampling manifolds. The filters were rinsed with three 3 ml portions of ice-cold buffer, dried for 15 min under an IR lamp and were counted for tritium radioactivity at about 50% efficiency in 3 ml of MP Ready-Solv (Beckman) cocktail. The counter was a Kontron model MR300 Automatic Liquid Scintillation System. Competition experiments were carried out essentially as described above except that total and non-specific binding of the radioligand were assayed in six replicates, fl-Endorphin, dynorphin(1-13) and [D-Prol°]dynorphin-(1-11) were tested for competition against [3H]etorphine binding in polypropylene tubes (to minimize adsorption to tube walls) in the presence of 10 /zM bestatin, 1 mM L-leucyl-L-leucine and 50 ktM N-carboxyme-
77 TABLE 1 Binding to be measured
In test tube at equilibrium
Retained on filter
Total qT = q s + q N s + q F (1) Q = qs + q N s + ~t'qF Non-specific q-r = qr~s + qF (2) Q ' = qNs + a.q~.
tyl-L-phenylalanyl-L-leucine (to prevent degradation by endogenous peptidases). In the filtration binding assay, radioligands bind to glass fiber. The contribution of this undesirable retention can easily be formulated in very general terms assuming that (i) bound radioligand is totally retained on the filter and (ii) free radioligand binds to the filter in direct proportion, a (0 < a < 1), to the amount added. q (Q) refers to the amount of radioligand: qv to total in test tube; qs and qNs to specifically and non-specifically bound, respectively; qF and q~ to free; Q and Q' to bound to filter. Combining (1) and (2) yields: qv = qv + qs. Specific binding is usually taken as: Q - Q' = qs + (qF - qv) = (1 a ) ' q s < qs, while free radioligand is calculated from: Q T - Q ' = q v - a ' q v = ( 1 - a ) ' q v
2.4. Chemicals [15,16-3H]etorphine (30-60 Ci/mmol; 1 Ci = 37 GBq), [15,16(n)-3H]diprenorphine (25-30 Ci/mmol) and [3,5-3H]Tyr-D-Ala-Gly-MePhe Glyol ([3H]DAGO, 30-60 Ci/mmol): Amersham (Amersham, England); etorphine and diprenorphine: Reckitt and Colman (Kingston upon Hull, England); levorphanol and dextrorphan: Hoffmann La Roche (Basel, Switzerland); naloxone: Endo (Garden City, NY, USA); ethylketocyclazocine: Sterling Winthrop (Rensselaer, NY,
USA); morphine: Francopia (Paris, France). TyrD-Ala-MePhe-Glyol (DAGO), Tyr-D-Thr-GlyPhe-Leu-Thr (DTLET) and bestatin: Cambridge Research (Harston, England); trans-3,4-dichloroN-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide (U50,488): Upjohn (Kalamazoo, MI, USA); fl-endorphin and L-leucyl-L-leucine: Bachem (Bubendorf, Switzerland); dynorphin-(113), [D-PrcC°]dynorphin-(1-11) and N-carboxymethyl-L-phenylalanyl-L-leucine were generously supplied by Drs. H. Mazarguil and J.-E. Gairin (Toulouse, France). 5'-Guanylylimidodiphosphate (GppNHp) was from Sigma (St Louis, MO, USA).
3. Results
3.1. Equilibrium binding properties of [ 3H]etorphine and of [3H]diprenorphine in the crude membrane fraction from the brain of R. ridibunda The equilibrium adsorption isotherms of the two non-selective opiates, the agonist [3H]etorphine and the antagonist [3H]diprenorphine were clearly biphasic. Scatchard transforms of the data (fig. 1A) indicated that the preparation contained (i) (0.76 _+ 0.16 p m o l / m g protein (n = 5) of a high affinity site that bound the two radioligands with nearly the same high affinity (K D = 0.13 + 0.02 nM) and, (ii) 3.2 +_ 0.2 p m o l / m g protein (n = 5) of a lower affinity site that bound the two radioligands with about the same lower affinity (K D = 13 _+ 4 nM). The fact that the low affinity site had a low affinity for the two tritiated oripavines was already indicative of its non-opioid nature. Nevertheless, we attempted to further differentiate the high and low affinity sites on the basis of other properties typical of opioid interactions, namely (i) the differential allosteric regulation of agonist and antagonist binding by sodium ions a n d / o r by guanine nucleotides (Pert et al., 1973; Childers and Snyder, 1980; Frances et al., 1985) and (ii) stereospecificity (Goldstein et al., 1971). The affinity of the antagonist [3H]diprenorphine for the high affinity site was slightly (1.5-fold) increased in the presence of sodium ions (120 mM NaC1) while high affinity binding of the
78
c B *120mMNaCI
+120ram NaCI + 5OpM GppNHp
+5OHM GppNHp
t ~
50~IHMT!~is"HCI
1
2
0
1
2 0 1 b, I~ol/mg protein
2
0
1
2
Fig. 1. Scatchard transforms of the equilibrium adsorption isotherms (25 o C) of [3H]etorphine (e) and of [3H]diprenorphine (m) in
the crude membrane fraction from the brain of Rana ridibunda. The concentrations of [3 H]etorphine and of [3H]diprenorphine were increased from 0.04 nM to 8 nM. b = amount of specificallybound radioligand, f(nM): concentration of free radioligand. The insert in fig. 1A shows the Scatchard transform of saturation data generated by isotopic dilution of 0.1 nM [3H]etorphine. Saturation experiments with isotopic dilution were aimed at measuring Bma x values with some precision ( B m a x - 4 pmol/mg protein). All other data shown here were obtained with the same membrane preparation.
agonist [3H]etorphine was still observed (K D = 0.13 + 0.04 nM, n = 4) but to a considerably reduced maximum number (0.17 + 0.09 p m o l / m g protein) of sites (fig. 1B). Under these conditions, equilibrium saturation binding of the two radioligands at the lower affinity sites did not seem to have been affected. Only the analogue of GTP, G p p ( N H ) p , (50 /~M) modified neither the high nor the low affinity binding of [3H]etorphine and of [3H]diprenorphine (fig. 1C). Yet, when present together with NaC1, G p p ( N H ) p appeared to prevent the sodium-induced increase in affinity of [3H]diprenorphine for the high affinity site and further decreased [3H]etorphine binding at this site. In the simultaneous presence of G p p ( N H ) p and of NaC1, high affinity binding of the tritiated opiate agonist was no longer detected while low affinity binding of the two radioligands (K D = 10.5 nM, Bmax = 3.1 p m o l / m g protein) was still observed (fig. 1D). Finally, we examined the ability of the two enantiomers, levorphanol ( - ) and dextrorphan ( + ) to inhibit the binding of [3H]etorphine at the two sites. When [3H]etorphine 0.1 nM was used so that, at equilibrium and in the absence of competing drug, > 90% of the radioligand was bound at
the high affinity site and < 10% at the lower affinity site, levorphanol ( K I = 3 nM) inhibited the binding of [3H]etorphine (fig. 2A) about 1 700 times more potently than dextrorphan (K l = 5 100 nM). When [3H]etorphine 2 nM was used, so that at equilibrium and in the absence of competing drug, about 62% of the radioligand was bound at the high affinity site and about 38% at the lower affinity site, the inhibition curves for the two enantiomers were both clearly biphasic yet with an inflexion at about 65% inhibition in the case of levorphanol and at about 35% inhibition in the case of dextrorphan (fig. 2B). These seemingly conflicting results could easily be reconciled if it was assumed that levorphanol displayed a higher apparent affinity ( I 3 2 . 5 = 56 nM; K I = 3.4 nM) for the [3H]etorphine high affinity site than (I82~ = 1600 nM, K I = 1400 nM) for the lower affinity site while, conversely, dextrorphan exhibited a higher apparent affinity 017.5 = 2000 nM; K I = 1 800 nM) for the [3H]etorphine lower affinity site than ( I 6 7 . 5 = 100000 nM; K l = 6 100 nM) for the high affinity site. Again, the [3H]etorphine high affinity site showed considerable stereospecificity, with 1 800-fold higher apparent affinity for levorphanol than for dextrorphan, whereas the
79 A
C 4
b
bo
• control V
~EX
LEV ~DEX
.5
:
\
'\ \
9 8 7 6 5 4
I
I
I "t~ i
8 7 6 5 luraabeled ligandl , -IogM
..ol~ 2
1
"1--
4
0
3
0
1 2 b, pmoUm 9 proton
Fig. 2. Inhibition of [3H]etorphine binding by levorphanol (*, LEV) and by dextrorphan (za, DEX) in the crude membrane fraction from the brain of Rana ridibunda. Competition studies. (A) [3H]Etorphine was used at the fixed concentration of 0.1 nM, so that at equilibrium, 90% of radioligand was bound at the high affinity site. (B) [ 3H]Etorphine was used as the fixed concentration of 2 nM so that, at equilibrium, about 62% of the radioligand was bound at the high affinity site and about 38% at the lower affinity site. In A and B, b 0 and b are the amounts of specifically bound radioligand in the absence and in the presence of unlabelled ligand, respectively. Saturation study. (C) Scatchard transform of the equilibrium adsorption isotherm of [3H]etorphine in the absence and in the presence of 10/~M levorphanol or of 10/~M dextrorphan. The concentration of [3H]etorphine was increased from 0.04 nM to 8 nM. b = amount of specifically bound radioligand, f(nM): concentration of free radioligand. There was no detectable high or low affinity specific binding with [3H]etorphine in the presence of 10 /~M levorphanol. Dextrorphan, 10 /~M, selectively suppressed the specific binding of [3H]etorphine at the lower affinity site while only slightly inhibiting (competitively) the specific binding of [ 3H]etorphine at the high affinity site.
[3H]etorphine lower affinity site did not. The apparent dissociation constants were nearly equal, 1 400 and 1600 nM for levorphanol and dextrorphan, respectively. Equilibrium saturation binding studies revealed that 10 /~M levorphanol totally prevented the interaction of [3H]etorphine, not only with the high, but also with the lower affinity sites while 10/~M dextrorphan selectively blocked binding of the radioligand at the lower affinity site (fig. 2C). Under these conditions (10/~M dextrorphan) there was only a moderate, competitive inhibition of high affinity [3H]etorphine binding with a slightly reduced K D value (0.35 nM) and an unchanged Bmax (0.72 pmol/mg protein).
3.2. ldentity of the opioid high affinity site in the crude membrane fraction from the brain of Rana ridibunda In order to identify the high affinity site as a /~-, 8- or x-opioid receptor site, we examined the ability of various unlabelled opioid ligands to compete with [3H]etorphine binding in frog brain
\Tx,l~i~k -
~
\. I\ 1~ IXX~_
b°
~, J'-\] ~ - \?"J ~. ~ ~ ~ , ~ \ ~
11
10
"~xl
• DIP X
~k
~ ~
9 8 7 6 lunlabeled I~ndl, -IogM
EKC
MOR '~ DAGO * DTLET ~
\ o
5
4
Fig. 3. Inhibition of [3H]etorphine binding by various unlabelled opioid ligands in the crude membrane fraction from the brain of Rana ridibunda. [3H]Etorphine was used at the fixed concentration of 0.1 nM. At equilibrium, total binding of the radioligand amounted to 1470+204 c.p.m. (n = 7) while non-specific binding (measured in the presence of 10 /~M diprenorphine) amounted to 129 + 17 c.p.m. (n = 7). b 0 and b are the amounts of radioligand specifically bound in the absence and in the presence of unlabelled ligand, respectively, Each b / b 0 value is the mean + S.D. of the values measured in three different preparations. ETO: etorphine, DIP: diprenorphine, NAL: naloxone, EKC: ethylketocyclazocine, MOR: morphine.
80 TABLE 2 Parameters characterizing the inhibition of binding of [3H]etorphine (0.1 nM) by various unlabelled opioid ligands in the crude membrane fraction from the brain of R. ridibunda. Apparent Hill coefficient and IC50 (concentration of unlabelled ligand causing 50% inhibition) values were calculated by linear regression analysis of modified Hill transforms, log b / ( b 0 - b ) = f [log (concentration of inhibitor)] of the inhibition curves, most of which are shown in figs. 2 and 3. K I values were calculated from K I - I C 5 0 (I+L/Ko) 1, L being the concentration of free [3H]etorphine (0.07-0.08 nM) and K D (0.13 nM), the equilibrium dissociation constant of the [3 H]etorphine macromolecular complex. The K 1 values represent true dissociation constants only for those unlabelled ligands for which the modified Hill transform of inhibition curves had a slope close to unity. The values shown in parentheses ( _+S.D., n = 3) were obtained from inhibition experiments carried out in the presence of D A G O (10 nM) + DTLET (20 nM) + U50,488 (20 nM), a mixture ensuring nearly complete blockade of ~t, 8- and ~-opioid sites. Note that, under these conditions, the apparent Hill coefficients increased significantly, especially in the case of DAGO and of DTLET, to values close to 1. Unlabelled ligand Etorphine Diprenorphine Naloxone EKC Morphine
IC50 (nM)
Apparent Hill coefficient
Apparent K l (nM)
0.4 0.4 3.2 9.6 37
1.04 1.03 0.68 0.82 0.65
Levorphanol Dextrorphan
4.5 7900
0.71 0.83
3 5130
D A G O (/Q DTLET (8) U50,488 (K)
176 (88_+ 28) 333 (182_+ 58) 1 730 (1975 _+487)
0.47(0.93_+0.02) 0.57 (0.96_+0.07) 0.80 (0.88 _+0.11)
114 (57_+ 18) 216 (118_+ 38) 1 123 (1 280 _+320)
/3-Endorphin Dynorphin 1-13 D P D Y N (K)
44 179 115
membranes. [3H]Etorphine was used at the fixed concentration of 0.1 nM so that, at equilibrium, > 90% of the radioligand was bound at the high affinity site and < 10% at the lower affinity site. The results of these competition experiments are shown in fig. 3 and in table 2. The high affinity site was found to display high affinity not only for the two oripavine derivatives, etorphine and diprenorphine, but also for several other classical opiate ligands such as naloxone, levorphanol, ethylketocyclazocine and morphine. Yet, the high affinity [3H]etorphine site in frog brain membranes appeared to be neither ~t, nor 8 or J¢: 176 nM DAGO, a /~ selective ligand (Kosterlitz and Paterson, 1981), 333 nM DTLET, a ~ selective ligand (Zajac et al., 1983) and 1 730 nM U50,488, a K selective ligand (Von Voigtlander et al., 1983) were necessary to halve the binding of [3H]etorphine, while nanomolar concentrations either of D A G O or of D T L E T or of U50,488
0.79 0.74 0.83
0.26 0.26 2.1 6.2 24
28 116 75
would have sufficed had the high affinity site been either/~, or 6 or ~, respectively. The slopes of modified Hill transforms of the inhibition curves, especially those for D A G O and D T L E T were clearly less than unity (table 2) indicating that 0.1 nM [3H]etorphine had labelled a mixture of several types of opioid sites: 15-20% of [3 H]etorphine binding persisted in the presence of as much as 10 /xM D T L E T (fig. 3). This suggested that, in addition to the lower affinity site ( < 10%), there were very small amounts (5-10%) of 'enkephalin-resistant' opioid sites (Meunier et al., 1983) which could be K sites. The inhibition curve of U50,488, with a minor component (5-10%) that no longer bound [3H]etorphine at concentrations of the x selective ligand in the range 1-10 nM supported this possibility. Similarly, the curve for D A G O inhibition of [3H]etorphine binding showed a plateau around 10-15%. This plateau value was reached at concentrations of the/~ selec-
81
.10
m
b : 0.06 f
,,L
0.7 + 0 . 7 4
f f+60
.05
•01
--
I 0
I
.1 .2 b , pmol/mg protein
,3
Fig. 4. Scatchard transform of the equilibrium adsorption isotherm ( 2 5 ° C ) of [3H]DAGO in the crude m e m b r a n e fraction from the brain of R. ridibunda. The concentration of [3H]DAGO was increased from 0.1 n M to 10 nM. b = a m o u n t of specifically b o u n d radioligand, f(nM): concentration of free radioligand at equilibrium. The equation is that of the saturation function that best fits the experimental data, assuming that [3H]DAGO b o u n d to two non-interacting classes of sites: a high affinity site which is likely to represent a /~-opioid site (KD = 0.7 nM, Bmax = 0.06 p m o l / m g protein) and a lower affinity site (K n = 60 nM, Bm~ = 0.74 p m o l / m g protein).
tive ligand as low as 10 nM and could therefore reflect binding of [3H]etorphine at a /~-opioid receptor site. This conclusion was supported by the results of saturation binding studies with [3H]DAGO. The crude membrane fraction from frog brain bound about 0.06 p m o l / m g protein of the/~ selective radioligand with high affinity (K D = 0.7 nM) (fig. 4). The inhibition of [3H]etorphine (0.1 nM) binding by DAGO, DTLET and U50,488 was tested in the presence of 10 nM DAGO + 20 nM DTLET + 20 nM U50,488, a mixture that nearly completely blocked the/~, 6 and K sites. Under these conditions, (i) control binding of [3H]etorphine was inhibited 20-30% and (ii) the slopes of modified Hill transforms of the inhibition curves of the three unlabelled competitors, especially those of DAGO and of DTLET, were significantly increased to close to unity (table 2). These results indicated that the radioligand now interacted with a homogenous population of sites. Finally, in order to further characterize the high affinity site as well as to select probes that might be used subsequently in cross-linking ex-
periments, we tested fl-endorphin, dynorphin-(113) and the dynorphin analogue [D-Prd°]dy norphin-(1-11) for their ability to compete with the binding of [3H]etorphine (0.1 nM) in frog brain membranes. The three peptides showed only moderate apparent affinity for the high affinity site. However, [D-Prol°]dynorphin-(1-11), a highly selective probe (Gairin et al., 1984) displayed a considerably lower (1 000-2 000-fold) apparent affinity for the high affinity [3H]etorphine site in the frog brain membranes than it did for ~-opioid sites in guinea pig cerebellum membranes (Gairin et al., 1984). Taken together, these results indicated that, in the preparation under study, there was a major opioid receptor sites that accounted for about 70% of the binding of [3H]etorphine (0.1 nM). The binding characteristics of this predominant site were clearly different from those of mammalian t~-, 8- and ~-opioid sites. In this same preparation, 20-30% of the [3H]etorphine (0.1 nM) binding could be accounted for by a /~ site (10%), a ~ site (5-10%) and by the lower affinity site ( < 10%). There was no clear evidence for the presence of a 6 site as no high affinity binding of the 8 selective radioligand [3H]DTLET could be reliably demonstrated in this preparation (data not shown). 4. D i s c u s s i o n
The present study has shown that the brain of an amphibian, the frog Rana ridibunda, contains a high and a lower affinity site that bind specifically both the opiate agonist [3H]etorphine and the opiate antagonist [3H]diprenorphine. The high affinity site showed (i) high affinity not only for the two tritiated oripavine derivatives but also for other classical opiate ligands, including naloxone and morphine; (ii) a considerably reduced ability to bind the agonist [3H]etorphine but not the antagonist [3H]diprenorphine in the presence of Na + ions and (iii) stereospecificity, all properties that are typical of an opioid site. The lower affinity site which is about four times as abundant as the high affinity site showed none of the above characteristics and is therefore probably not opioid in nature. So far, nothing is known about the identity of the lower affinity site.
82 The data on amphibian opioid receptors are both scarce and conflicting. For instance, the brain of the toad (Bufo marinus) appears to lack the lower affinity site for [3H]diprenorphine. However, this discrepancy is only apparent, since toad brain membranes were selectively assayed by Ruegg et al. (1981) fo stereospecific binding of the tritiated opiate antagonist so that the lower affinity site, which is not stereospecific, was probably overlooked. Ruegg et al. (1981) had first suggested that 'the toad brain contains a predominance of morphine preferring receptors, designated ~ by Lord et al. (1977)'. Later, Simon et al. (1982) found, also in the toad brain, a ' p r e d o m i n a n t opiate receptor subclass.., of the benzomorphanpreferring type'. Interestingly, these authors also noted that the subclass in question, which they referred to as x / o , had ' a somewhat higher affinity for/~ and 6 ligands than its mammalian counterpart', namely, the x type of opioid receptor (Kosterlitz et al., 1981). More recently, Simon et al. (1984) claimed that the brain of R. esculenta contains opioid sites that are mostly of the K type. In this latter study however, the radioligand ([3H]naloxone) was first shown to bind to two classes of sites in frog brain membranes and was then used in competition studies with unlabelled opioid ligands at a concentration that labelled nearly equal amounts of high and of lower affinity sites. It is not clear how these authors came to the conclusion of the existence of a ~-opioid site in this preparation under these conditions. We have now shown, in agreement with the preliminary observation of Simon et al. (1982), that the brain of the frog R. ridibunda contains a predominant opioid site with an in vitro binding profile clearly distinguishable from those of m a m malian #-, 8- and ~-opioid sites. What makes this site (hereafter referred to as Op) so conspicuous is that while it exhibits a considerably lower affinity for the two t¢ selective ligands U50,488 and D P D Y N as Simon et al. (1982) found in toad brain, it has a higher apparent affinity for /~ ( D A G O ) and 6 (DTLET) ligands than do r-opioid sites in mammalian brain membranes (table 3). If the amphibian Op site is clearly not K, as this latter type has been well characterized in guinea pig (Kosterlitz et al., 1981) and rabbit (Meunier et
TABLE 3 Apparent affinity (KI) of/~, 8 and g selective opioid ligands for several types of opioid receptor in mammalian (~, 8, r and '~2') and frog (Op) nerve tissue preparations. The values in parentheses are for apparent affinity at the preferred site. Data are from a Paterson et al. (1984), b Gairin et al. (1984), c Gouarderes et al. (1984) and d the present study. K 1 (nM) ~t-Site a 6-Site a x-Site a ,r2,_Site c Op site d DAGO (~t) (1.8) 423 1927 41 57 DTLET (8) 34 (2.7) 14500 330 118 U50,488 (r) 890 9800 (7.3) 890 1280 DPDYN (K) 2.0 b 7.5 b (0.03)b 75 _
al., 1983) brain membranes, its binding characteristics are reminiscent of those of the so-called '•2' subtype of the K-opioid receptor. The 'x2'-opioid sites were first identified in rat (Gouard~res et al., 1982) and in guinea pig (Attali et al., 1982) lumbo-sacral cord because they had much higher affinity t h a n x-opioid sites for /~ ([DAla2,MePhe4,Met(O)olS]enkephalin) and for 8 ([D-Ala2,D-AlaS]enkephalin and [D-Ser2,LeuS]en kephalyl-Thr) ligands. Interestingly, Weyhenmeyer and Mack (1985) have presented preliminary evidence that the rat brain contains a minor population of opioid sites that appear to be neither /~, nor 6 or x. However, these authors have not reported the apparent affinities of D A G O and of D T L E T for these additional sites thus precluding any direct comparison with the Op site from amphibian brain. The existence of a particular type of opioid binding site in the brain of an amphibian was not totally unexpected because digitonin extracts from amphibian brain membranes display opiate binding activity (see Introduction) whereas digitonin extracts from m a m m a l i a n brain membranes do not. This suggests structural differences between amphibian and m a m m a l i a n opioid receptor types. The physiological significance of the frog brain Op site remains to be elucidated. Pharmacological data on opiates in amphibians are scarce but there is recent evidence ' n o t only for similarities but also for important differences in the pharmacological actions of opiates and antagonists in amphibians and m a m m a l s ' (Pezalla and Stevens,
83 1984). This c o u l d b e r e s t a t e d as ' s i m i l a r i t i e s a n d i m p o r t a n t differences b e t w e e n o p i o i d r e c e p t o r types in a m p h i b i a n s a n d m a m m a l s ' . F o r instance, systemic m o r p h i n e a p p e a r s to b e a w e a k e r analgesic in frogs t h a n in rats while d e x t r o r p h a n , which is essentially inactive in rats elicits a n a l g e s i a in frogs. H o w e v e r , d e x t r o r p h a n - i n d u c e d analgesia in frogs is, unlike m o r p h i n e - i n d u c e d a n a l g e s i a in rats a n d frogs, insensitive to n a l o x o n e (Stevens a n d Pezalla, 1984). These a n d o t h e r d a t a (Pezalla, 1983) c o u l d be t a k e n to i n d i c a t e that responses m e d i a t e d b y o p i o i d receptors c o u l d b e m e d i a t e d via different r e c e p t o r types in m a m m a l s a n d amphibians. Finally, it is w o r t h y of n o t e that the O p site p r e d o m i n a t e s early in the course of v e r t e b r a t e evolution, i.e. in a m p h i b i a n nerve tissue. Also, a l t h o u g h it d i s p l a y s an original b i n d i n g profile it also shares with m a m m a l i a n /~-, 3- a n d K-opioid sites several basic, well preserved, p r o p e r t i e s which i n c l u d e (i) e x t r e m e l y high affinity for certain o r i p a v i n e derivatives, n a m e l y e t o r p h i n e a n d d i p r e norphine, (ii) differential allosteric r e g u l a t i o n of agonist a n d a n t a g o n i s t b i n d i n g b y N a ÷ ions a n d (iii) stereospecificity. It is therefore t e m p t i n g to suggest that the a m p h i b i a n O p site represents the a r c h e t y p e of the o p i o i d site at the origin of the various types of o p i o i d site f o u n d in m a m m a l i a n nerve tissue. T h e r e is i n d e e d p r o o f for the evolution of o p i o i d systems: t o a d (Xenopus laevis) p r o e n k e p h a l i n lacks the [LeuS]enkephalin sequence f o u n d in higher v e r t e b r a t e s ( M a r t e n s et al., 1984). However, there is no such direct evidence in the case of o p i o i d r e c e p t o r types a n d f i r m p r o o f of o u r p r o p o s a l of the a r c h e t y p i c n a t u r e of the a m p h i b i a n O p site awaits sequencing of their genes.
References Attali, B., Ch. Gouardrres, H. Mazarguil, Y. Audigier and J. Cros, 1982, Evidence for multiple 'kappa' binding sites by use of opioid peptides in the guinea-pig lumbo-sacral spinal cord, Neuropeptides 3, 53. Chang, K.-J., S.G. Blanchard and P. Cuatrecasas, 1984, Benzomorphan sites are ligand recognition sites of putative c-receptors, Mol. Pharmacol. 26, 484. Chang, K.-J. and P. Cuatrecasas, 1979, Multiple opiate recep-
tors: enkephalins and morphine bind to receptors of different specificity, J. Biol. Chem. 254, 2610. Chang, K.-J., E. Hazum and P. Cautrecasas, 1981, Novel opiate binding sites selective for benzomorphan drugs, Proc. Natl. Acad. Sci. U.S.A. 78, 4141. Childers, S.R. and S.H. Snyder, 1980, Differential regulation by guanine nucleotides of opiates agonist and antagonist interactions, J. Neurochem. 34, 583. Frances, B., Ch. Moisand and J.C. Meunier, 1985, Na + ions and Gpp(NH)p selectively inhibit agonist interactions at /~ and x-opioid receptor sites in rabbit and guinea-pig cerebellum membranes, European J. Pharmacol. 117, 223. Gairin, J.-E., Ch. Gouard~res, H. Mazarguil, P. Alvinerie and J. Cros, 1984, [D-Prol°]Dynorphin-(1-11) is a highly potent and selective ligand for x-opioid receptors, European J. Pharmacol. 106, 457. Goldstein, A., L.I. Lowney and B.K. Pal, 1971, Stereospecific and nonspecific interactions of the morphine congener levorphanol in subcellular fractions of mouse brain, Proc. Natl. Acad. Sci. U.S.A. 68, 1742. Goodman, R.R. and S.H. Snyder, 1982, ~ Opiate receptors locahzed by autoradiography to deep layers of cerebral cortex: relation to sedative effects, Proc. Natl. Acad. Sci. U.S.A. 79, 5703. Goodman, R.R., S.H. Snyder, M.J. Kuhar and W.S. Young, III, 1980, Differentation of 8 and t~ opiate receptor localizations by light microscopic autoradiography, Proc. Natl. Acad. Sci. U.S.A. 77, 6239. Gouard~res, Ch., Y. Audigier and J. Cros, 1982, Benzomorphan binding sites in rat lumbo-sacral spinal cord, European J. Pharmacol. 78, 483. Howells, R.D., T. Gioannini, J.M. Hiller and E.J. Simon, 1982, Solubilization and characterization of active opiate binding sites from mammalian brain, J. Pharmacol. Exp. Ther. 222, 629. Kosterlitz, H.W. and S.J. Paterson, 1981, Tyr-D-Ala-GlyMePhe-NH(CH2)aOH is a selective ligand for the p-opiate binding sites, Br. J. Pharmacol. 73, 299P. Kosterlitz, H.W., S.J. Paterson and L.E. Robson, 1981, Characterization of the kappa subtype of the opiate receptor in the guinea-pig brain, Br. J. Pharmacol. 73, 939. Lowry, D.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193, 265. Lord, J.A.H., A.A. Waterfield, J. Hughes and H.W. Kosterlitz, 1977, Endogenous opioid peptides: multiple agonists and receptors, Nature (London) 267, 495. Martens, G.J.M. and E. Herbert, 1984, Polymorphism and absence of Leu-enkephalin sequences in proenkephalin genes in Xenopus laevis, Nature (London) 310, 251. Meunier, J.C., Y. Kouakou, A. Puget and Ch. Moisand, 1983, Multiple opiate binding sites in the central nervous system of the rabbit. Large predominance of a mu subtype in the cerebellum and characterization of a kappa subtype in the thalamus, Mol. Pharmacol. 24, 23. Paterson, S.J., L.E. Robson and H.W. Kosterlitz, 1984, opioid receptors, in: The Peptides: Analysis, Synthesis, Biology,
84 eds. S. Udenfriend and J. Meienhofer, (Academic Press, London) 5, 147. Pert, C.B., G. Pasternak and S.H. Snyder, 1973, Opiate agonists and antagonists discriminated by receptor binding in brain, Science (Washington D.C.) 182, 1359. Pezalla, P.D., 1983, Morphine-induced analgesia and explosive motor behavior in a amphibian, Brain Res. 273, 297. Pezalla, P.D. and C.W. Stevens, 1984, Behavioral effects of morphine, levorphanol, dextrorphan and naloxone in the frog Rana pipiens, Pharmacol. Biochem. Behav. 21,213. Ruegg, U.T., S. Cuenod, J.M. Hiller, T. Gioannini, R.D. Howells and E.J. Simon, 1981, Characterization and partial purification of solubilized active opiate receptors from toad brain, Proc. Natl. Acad. Sci. U.S.A. 78, 4635. Schulz, R., M. Wuster and A. Herz, 1981, Pharmacological characterization of the c-opiate receptor, J. Pharm. Exp. Ther. 216, 604. Simonds, W.F., G. Koski, R.A. Streaty, L.M. Hjelmeland and W.A. Klee, 1980, Solubilization of active opiate receptors, Proc. Natl. Acad. Sci. U.S.A. 77, 4623. Simon, E.J.J.M. Hiller, J. Groth, Y. Itzhak, M.J. Holland and
S.G. Eck, 1982, The nature of opiate receptors in toad brain, Life Sci. 31, 1367. Simon, J., M. Szucz, S. Benyhe, A. Borsodi, P. Zeman and M. Wolleman, 1984, Solubilization and characterization of opioid binding sites from frog (Rana esculenta) brain. J. Neurochem. 43, 957. Stevens, C.W. and P.D. Pezalla, 1984, Naloxone blocks the analgesic action of levorphanol but not of dextrorphan in the leopard frog, Brain Res. 301, 171. Yon Voigtlander, P.F., R.A. Lahti and J.H. Ludens, 1983, U-50488: a selective and structurally novel non-/~ (~¢) opioid agonist, J. Pharmacol. Exp. Ther. 224, 7. Weyhenmeyer, J.A. and K.J. Mack, 1985, Binding characteristics of kappa opioids in rat brain. A comparison of in vitro binding paradigms, Neuropharmacology 24, 111. Zajac, J.M., G. Gacel, F. Petit, P. Dodey, P. Rossignol and B.P. Roques, 1983, Deltakephaiin Tyr-D-Thr-Gly-Phe-LeuThr: a new highly potent and fully specific agonist for opiate delta receptors, Biochem. Biophys. Res. Commun. 111, 390.