Gen. Pharmac. Vol. 21, No. 5. pp. 605-611, 1990 Printed in Great Britain

0306-3623/90$3.00+ 0.00 PergamonPress pie

MINIREVIEW PHARMACOLOGICAL PROPERTIES OF NEWLY SYNTHESIZED DERIVATIVES OF (-)-6fl-ACETYLTHIONORMORPHINE AND THEIR INTERACTIONS WITH OPIOID RECEPTORS [SSEI TAKAYANAGI,KATSUOKOIKEand TSUTOMUSUZUKI~ Department of Chemical Pharmacology, Toho University School of Pharmaceutical Sciences, Funabashi, Chiba 274 and ~Department of Applied Pharmacology, School of Pharmacy, Hoshi University, Tokyo 142, Japan [Tel. (0474) 72-II41]

( Receired 24 January 1990) Abstract--l. Some pharmacological properties of newly synthesizedderivatives of (-)-6fl-acetylthionormorphine, AcS-morphine (KT-88), KT-89 and KT-90 and their interactions with opioid receptors were studied. 2. AcS-morphine was about twice as potent as morphine in the inhibitory action of the twitch response of the guinea-pig ileal preparation to electrical stimulation and about 5 times as potent as morphine in the analgesic action in the rats. Both the effects of AcS-morphine were inhibited by naloxone, suggesting that the site of action of AcS-morphine is p-receptors. 3. Analgesic activities of KT-89 and KT-90 were about 6 and 10 times as potent as morphine in an acetic acid-induced writhing test and about 4 and 5 times as potent as in a pressure test. 4. As the analgesic action of KT-90 was antagonized by norbinaltorphimine, the site of action of KT-90 is concluded to be K-receptors. Furthermore, KT-89 and KT-90 behaved as antagonists on g-receptors. 5. The present results suggest that AcS-morphine which is N-methyl derivative, as well as morphine, has a selectivelyhigh affinity to ,u-receptors, and that KT-89 and KT-90 which are N-cyclopropylmethyl derivatives have nonselectively high affinities to p-, h- and 6-receptors.

INTRODUCTION

the compounds that exchange the N-methyl group of AcS-morphine for the N-cyclopropylmethyi group, and KT-90 is synthesized by the acetylation of the hydroxy group of the 3-position of KT-89.

The existence of multiple opioid receptors in the brain and peripheral tissues has been documented on the basis of biochemical and pharmacological studies (Martin et al., 1976; Lord et al., 1977; Magnan et al., 1982). It is well known that the functioning of many enzymes depends on free sulfhydryl (-SH) groups. The affinity of opioid agonists to their receptors, however, was found to be influenced by an SHreagent, N-ethylmaleimide (Simon, 1981). The role of the sulfhydryl group in the opioid receptors was also discussed (Bowen et al., 1981). Therefore, (-)-6flacetylthiomorphine (AcS-morphine, Fig. 1) (Fujii et al., 1988) and N-cyclopropylmethyl derivatives of (-)-6fl-acetylthionormorphine, KT-89 and KT-90 (Fig. I) were newly synthesized. This short review will focus on the interactions of the three newly synthesized compounds with opioid receptors to study their pharmacological properties.

PHARMACOLOGICAL PROPERTIES OF AcS-MORPHINE (KT.-Sg, N-METHYL DERIVATIVE) AND ITS INTERACTION WITH OPIOID RECEPTOR

CHEMICAL STRUCTURES OF NEWLY SYNTHESIZED DERIVATIVES OF (--)-6~-ACETYLTH1ONORMORPHINE

Chemical structures of AcS-morphine (KT-88), KT-89 and KT-90 are shown in Fig. I. AcS-morphine is the compound that exchanges the alcoholic hydroxy group of the 6-position of morphine for the acetylthio groups. AcS-morphine, however, has the N-methyl group as morphine. KT-89 and KT-90 are

Effect o f AcS-morphine on twitch response o f electricall)' stimulated ileal preparation o f guinea-pig The electrically stimulated ileal preparation from guinea-pig was used as a model to investigate the mode of action of narcotic analgesics in the central nervous system. Though it is more interesting to compare AcS-morphine with 6-acetylmorphine to study a structure-activity relationship, we used morphine as a reference drug which is easily obtained. AcS-morphine and morphine, a reference drug, inhibited in a concentration-dependent manner the twitch response to field stimulation. The pD2-value (the negative logarithm of the molar concentration required to produce 50% of the maximum response to the drug) for AcS-morphine indicated that it was about twice as potent as that for morphine (Table 1). Naloxone shifted the concentration-response curves of AcS-morphine and morphine towards a higher concentration. Schild plots (Arunlakshana and Schild, 1959) of these results gave straight lines with a slope of one (Table I). The pA2-values (the negative

605

606

ISSEI TAKAYANAGI el al. Table 2. The inhibitory effects (the plC~,-values) of AcS-morphine and morphine on the specific binding of three ~H-labeled opioid ligands to rat brain synaptosomal fraction plC~-values

A c S ~ H AC S-rno rph i ne (KT-88)

Naloxone Morphine AcS-morphine

[~HINal

[aHIEKC

[~H]DADLE

8.81 ~_0.03 8.48 _+0.5 9.18 + 0.04

8.23 + 0.05 7.73 _+ 0.08 8.43 + 0.06

7.83 + 0.1 I 7.13 + 0.20 7.92 + 0.15

Each value represents the mean + SE of 3 experiments. [3H]NaI: [ 3H]naloxone (0.5 nM); [3H]-EKC: [ ~H]ethylketocyclazocine (0.3 nM); and [~HIDADLE: [~H]D-Ala:-t~-Leu'-enkephalin (0.5 nM).

Binding assay A c SH~'~,~/

K T - 89

"c°

L'

H

KT - 90 Fig. I. Chemical structures of AcS-morphine (KT-88), KT-89 and KT-90. Ac: CH3CO-, Me: C H ~ - . (KT: Kanematsu and Takayanagi). logarithm of the dissociation constant of an antagonist) obtained by the Schild plot analysis of the data were not significantly different from each other (Table I), suggesting that the site of action of AcSmorphine is /,-receptors (Takayanagi et al., 1988). Analgesic actit'ity Antinociceptive activity was measured after subcutaneous injection in conscious male Wistar rats with Randal-Sellito apparatus. Pressure was given to a hind paw and the maximal pressure measured was 250 g. Analgesic effect was expressed as an analgesic index which has shown as a ratio of threshold pressures (g) 30 min before and after injection of one of the test drugs. A parallel line assay was employed tbr thc analgesic potency ratio using morphine as a standard. Analgesic activity of AcS-morphine, given s.c. to rats was about 5 times as potent as that of morphine. Analgesic activities of AcS-morphine (0.5 mg/kg, s.c.) and morphine (2.5 mg/kg, s.c.) were abolished by naloxonc (1 mg/kg, s.c.). These observations also suggest that their analgesic activity is mediated through /t-receptors. In the analgesic activity of morphinc the 6:~-isomer is much more potent that the 6/J-isomer. It is, however, interesting that (-)-6/1-AcS-morphine is 5 times as potent as ( - ) morphine (Takayanagi et al., 1988). Table I. The pD_,-values of AcS-morphine and morphine, the pA_,-values of naloxone against AcS-morphine and morphine, and the slopes of Schild plot for antagonism between naloxon¢ and AcS-morphine and morphine in the guinea-pig ileal preparation pD2-~alue M0rpfiine AcS-morphine

pA,-value (naloxone)

7. II .+ 0.0~----8.43~- 0.1,1 7.43 ± 0.04 8.36 + 0.15

Slope I. 14 -- 0.09 1.09 + 0.1 I

Slope: slope of Schild plot for antagonism between naloxone and morphine or AcS-morphine. Each value represents the mean _4-SE of 4-8 experiments

The synaptosomal fraction from brain is commonly used as a means of assessment of the multiple opioid receptors binding spectra of narcotic analgesics (Lord et al., 1977; Magnan et al., 1982; Konno et al., 1985; Takayanagi et al., 1985). Therefore, we examined the interactions of AcS-morphine with multiple opioid receptor subtypes using radioligand binding assay (Takayanagi et al., 1988). The binding of [3H]naloxone (Nal, /a-selective ligand), [3H]ethylketocyclazocine (EKC, x-selective ligand) and [3H]D-Ala2-D-LeuS-enkephalin (DADLE, f-selective ligand) to synaptosomal fraction from rat brain was displaced in a concentration-dependent manner and completely by the simultaneous addition of increasing concentration of nonradioactive AcSmorphine, morphine and naloxone. The plCs0-values as the negative logarithm of 50% inhibitory concentrations estimating from their displacement curves are shown in Table 2. The ability of AcS-morphine to displace the three labeled ligand bindings which were estimated as potency ratios relative to morphine were about 5 times more potent than those of morphine, because a difference between plC~0-values of AcSmorphine and morphine is 0.7 (Table 2). The ability of AcS-morphine and morphine to displace [3H]naloxone was the most effective among three labeled ligands. These facts suggest that AcS-morphine as well as morphine had a selectively high atfinity to /J-receptors. In order to determine the effects of sodium ion or guanosine-5'-triphosphate (GTP) on the interactions of the test drugs with opioid receptors, the binding reaction was performed in the presence and absence of 100 mM of NaCI and/or 50t(M of GTP with [3H]naloxone. The "sodium effect" and the "GTP effect" of AcS-morphine were almost the same as those of morphine, It is well known that both the effects are large in a pure /~-agonist and small in a pure antagonist (Pert and Snyder, 1974: Childers and Snyder, 1980). Dependence liability The dependence liability of AcS-morphine was preliminary tested in the guinea-pig ileal preparations treated with a high concentration of AcS-morphine for 24 hr. After the longitudinal muscle had been incubated for 24 hr with the concentration listed in Table 3, the electrically stimulated preparation was challenged by naloxone ( 1 0 -6 M ) . Addition of naloxone produced a large contraction of the muscle preparation treated with morphine and increased the twitch response. In the preparation treated with AcS-morphine, only the twitch response was slightly increased by naloxone. The ratio ( y ' / y ) of the twitch

( - )-6//-Acetylthionormorphine derivatives Table 3. Effects of naloxone (10-6M) on the test drug-treated longitudinal muscle of guinea-pig ileum stimulated electrically in the presence of the test drug.

Morphine AcS-morphine

Concentration in incubation medium (M)

Concentration in assay medium (M)

y'/y

2.0x 10 -5 1.0 x 10 ~

2.0x 10 .5 1.0 x 10 s

2.19+0.28 1.26 _+0.04*

y and y': twitch responses before and after naloxone (10 6 M). *Significantly different from the corresponding value of morphine at P < 0.05.

responses before (y) and after ( y ' ) naloxone (10 -6 M) was calculated. The ratio ( y ' / y ) obtained in the muscle preparation treated with AcS-morphine was significantly larger (P < 0.05) than that in the preparation treated with morphine (Table 3). When the longitudinal muscle, incubated for 24hr in Krebs solution without morphine, was challenged by naloxone (10 -6 M) in the presence of morphine (at tenth of ICso, 2.0 x 10 8 M), no contraction of the preparation was observed and the twitch response to field stimulation was not influenced by morphine (2.0 x 10 -s M). Ehrenpreis et al. (1975) found that the small intestine of chronically morphinized guinea-pig characteristically contracted upon naloxone exposure. Furthermore, myenteric plexus-longitudinal muscle strips prepared from tolerant and dependent guineapig and continuously exposed to morphine, displayed a contraction upon naloxone challenge (Schulz and Herz, 1976; Kaneto and Watanabe, 1980; Takayanagi et al., 1981). These phenomena are considered to represent a sign of abstinence from narcotics, since naloxone does not cause change in tension of preparations obtained from naive guineapigs. Therefore, if the changes in the base line tension after naloxone challenge reflect dependence liability of narcotic-analgesics, the liability of AcS-morphine may be less than that of morphine. In the ileum of the guinea-pig treated with the opiate for 24 hr and then washed out, naloxone increased the twitch response to electrical stimulation in the presence of a tenth of the IC5o, this concentration having no influence on the untreated ileum. This may be one of the characteristics of the opioid-dependent ileum of guinea-pig (Kaneto and Watanabe, 1980; Takayanagi et al., 1981). If such is indeed the case, the )"/3'-value for a drug having a low degree of dependence liability would be smaller than that for a drug with high dependence. The present observations that the y ' / y value for AcS-morphine was significantly (P < 0.05) smaller than that for morphine also suggest that AcS-morphine may be weaker than morphine is dependence liability. It seems almost certain, however, that AcS-morphine does have the dependence liability, though further studies were required (Takayanagi et al., 1988). PHARMACOLOGICAL PROPERTIES OF KT-89 AND KT-90 (N-CYCLOPROPYLMETHYL DERIVATIVES) AND THEIR INTERACTIONS WITH OPIOID RECEPTORS

Effects o f the K T - 8 9 and K T - 9 0 on twitch responses o f isolated smooth muscle preparations to electrical stimulation

KT-89 and KT-90, as well as morphine inhibited the twitch response to electrical stimulation of the GP 21 "5~C

607

guinea-pig ileal preparation which contains both/~and x-receptors. The pD2-values of KT-89 and KT90 indicate that the drugs are about 6.5 times and 10 times as potent as morphine, respectively (Table 4). Naloxone shifted the concentration-response curves of KT-89, KT-90 and morphine towards a higher concentration. Schild plots of these results gave straight lines with a slope of one (Table 4). The pA2-values of naloxone against KT-89 and KT-90 obtained by the Schild plot analysis of the data were significantly smaller than that of naloxone against morphine (Table 4), but equal to that against dynorphin, which was reported by Takemori et al. (1981). These results suggest that KT-89 and KT-90 interact with r-receptors to inhibit the twitch response of ileal preparation. Rabbit vas deferens has been reported to contain only g-receptors (Oka et al., 1981) and to respond to x-agonist with high intrinsic activity (Hayes and Kelly, 1985; Konno et al., 1986). KT-89 and KT-90, as well as dynorphine (1-13), a prototype x-agonist, inhibited the twitch response of rabbit vas deferens to electrical stimulation and their maximum responses were the same, suggesting that KT-89 and KT-90 are also ~c-agonists with the high intrinsic activity. The pD2-values indicate that KT-89 (7.18_+0.13) and KT-90 (7.22 _+ 0.05) are about 6 times as potent as dynorphin (I-13) (6.34 + 0.17). To study whether KT-89-KT-90 have the properties of an antagonist, their antagonistic potencies against morphine were tested in the guinea-pig ileal preparation. KT-89 and KT-90 behaved as an agonist-antagonist in the guinea-pig ileal preparation and acted as a competitive antagonist of morphine, a prototype/~-agonist. Their pA2-values against morphine indicate that the activities of KT-89 (8.72 _+0.05) and KT-90 (8.89 _+ 0.20) as /a-antagonists are more potent than naloxone (8.54_+0.13). It was recently reported that activation of central g-receptors may play a role in controlling pain mechanisms (Ward and Takemori, 1983; Porreca et al., 1984; Kov~.cs et al., 1988) and that this activation is followed by a rapid development of a tolerance to analgesic action (Kovfics et a/., 1988). In order to study the mode of action of KT-89 and KT-90 on g-receptors, their effects on a concentration-response curve of Leu-enkephalin were tested in the electrically stimulated vas deferens of mouse. KT-89 and KT-90 are suggested to be antagonists of g-receptors with the electrically stimulated vas deferens of mouse. Further experiments are needed to study effects of their properties as a 6-receptor antagonist on analgesic action and development of tolerance to this action (Takayanagi et al., 1990). Table 4. The pD2-values of KT-89, KT-90 and morphine, the pA2-values of naloxone against KT-89. KT-90 and morphine, and the slopes of Schild plot for antagonism between naloxone and KT-89, KT-90 and morphine in the guinea-pig ileal preparation

Morphine KT-89 KT-90

pD2-value

pA2-value (naloxone)

7.32__ 0.13 8.13__.0.04 8.49_+0.08

8.54+_0.13 7.75+_0.12 7.80__ 0 . 1 1

Slope

1.03+_.0.09 0.93_+0.09 0.86+0.11

Slope: slope of Schild plot for antagonism between naloxone and morphine. KT-89 or KT-90. Each value represents the mean _+SE of 5 experiments.

608

ISSEt T A K A Y A N A G ]

100

Analgesic' actit'ity

It has frequently been assumed that narcotics and antinarcotics exert their analgesic actions in any given analgesic test through the interaction with /~- and x-receptors. When a parallel line assay was employed for the analgesic potency ratios using morphine as a standard, analgesic activities of KT-89 and KT-90, given s.c. to rats and mice were 4.39 (3.28-6.08) to 5.08 (3.93-7.08) times as potent as morphine in the pressure test and 5.76 (3.50-7.92) to 10.1 (5.21--14.9) times as potent in the acetic acid-induced writhing test, respectively (Takayanagi et al., 1990). Using the hot-plate test, the analgesic effect of KT-90 was also determined in mice. KT-90 (10pg, i.c.v.) produced approximately 75% analgesic effect. This analgesic effect was antagonized by nor-binaltorphimine (5/~g, i.c.v.), a K-antagonist but not by [J-funaltrexamine (1/ag, i.c.v.), a /~-antagonist. The results suggest that the analgesic action of KT-90 is mediated through x-receptors. Pretreatment with KT-90 (I/~g, i.c.v.) antagonized the analgesic effect of morphine (10/~ g, i.c.v.). The/l-receptor antagonistic activity of KT-90 (1 #g, i.c.v.) was more potent than naloxone (1/~g, i.c.v.). Binding assay The binding of [~H]Nal, [3HIEKC and [~H]DADLE to the synaptosomal fraction from rat brain was displaced in a concentration-dependent manner and completely by the simultaneous addition of an increasing concentration of nonradioactive KT-89, KT-90, naloxone and morphine. The plC~0-values estimated from their displacement curves are shown in Table 5. The abilities of morphine and naloxone to displace [~H]Nal were the most effective among the three labeled ligands, while KT-89 and KT-90 had nonselectively high affinities to/~-, x- and f-receptors. However, the affinities of KT-89 and KT-90 to the three opioid receptors tested were more potent than morphine and naloxone. These results coincide with those obtained in the electrically stimulated preparations. The potencies of KT-89 and KT-90 to displace [~H]EKC were 6 and 10 times more than that of morphine. These results are equal to those obtained in the guinea-pig ileal preparation. The pA2-values of KT-89 and KT-90 against morphine were larger than that of naloxone against morpine in the electrically stimulated ileal preparations. KT-90 antagonized the analgesic action of morphine and antagonistic activity of KT-90 was more potent than that of naloxone. These facts were supported by the results that the abilities of KT-89 and KT-90 to displace [3H]Nal tend to be stronger than that of morphine (Table 5). Table 5. The inhibitory effects (the plC~-values) of KT-89, KT-90 and morphine on the specific binding of three ~H-labeled opioid ligands to rat brain synaptosomal fraction

['HINal Naloxone Morphine KT-89 KT-90

8.66 8.42 9.24 8.99

-_ 0.23 _- 0.25 _+ 0.22 ± 0.22

et al.

plCS0-values ['HIEKC ['H1DADLE 7.92 7.72 8.53 8.82

-_ 0.05 _- 0.08 - 0.11 -,- 0.05

7.75 7.27 8.96 8.54

+ + + +

0.05 0.04 0.22 0.17

Each value represents the mean + SE of 4 experiments. [~H]NaI: [ 'H]naloxone (0.3 nM ). {~H]EKC: [ ~H]ethylketocyclazocine (0.5 nM) and [~H]DADLE: [rH]D-Ala-LD-Leu~-enkephalin (0.7 nM).

90

-11

-10

-9

-8

-7

-6

-5

-4

Log M

Fig. 2. Concentration-response curves for KT-90, dynorphin (I- 13) and morphine in 137 mM K-stimulated 4SCa-uptake in the synaptosomes from rat brain and the curves for KT-90 in the presence of morphine. The experiments were carried out according to Toru et al. (1989). 0 : dynorphin (I--13), O: KT-90 alone and A: KT-90 with morphine (10 -7 M). Ordinate: % of maximal Ca amount taken up, and abscissa: log of concentration (M). Each point is presented as a mean + SE of 4 experiments. Abilities of [3H]DADLE were not different from one another (Table 5). These results were supported by the facts that there were no difference between their pA2-values against Leu-enkephalin in the mouse vas deferens (Takayanagi et al., 1990). Ca-uptake (inhibition o f synaptosomal Ca-uptake) Recently the actions of various opioids were examined on Ca-action potentials in the cell somata of guinea-pig myenteric neurons and on the release of acetylcholine at synapses onto these cells. The results obtained suggest that x-receptor antagonists may depress the release of acetylcholine by directly reducing Ca-entry into the nerve terminals, while presynaptic inhibition caused/~-receptor activation probably as a result of an increase in K conductance (Cherubini and North, 1985). However, dynorphin (1-13), nalorphine, nalorphine epoxide and morphine were reported to inhibit 45Ca-uptake by rat brain synaptosomes in a concentration-dependent manner (Toru et al., 1989). The maximum inhibitory response to dynorphin (I-13), nalorphine and nalorphine epoxide, which were reported as selective ligands to x-receptors (Takayanagi et al., 1985; Konno et al., 1986), was not significantly different among them. The maximum inhibitory response to morphine was significantly smaller than those to other x-agonist, suggesting that morphine might be a partial agonist. However, morphiceptin, a selective ligand to /~receptors and DADLE (D-Ala2-D-LeuS-enkephalin), a selective ligand to f-receptors were without any effect on 45Ca-uptake. These results indicate the opioids inhibit the synaptosomal Ca-uptake in the rat brain through x-receptors (Toru et al., 1989). Therefore, we tested the mode of action of KT-90 on synaptosomal Ca-uptake in rat brain. KT-90 as well as dynorphin ( I - I 3) inhibited stimuliinduced 4~Ca-uptake in a concentration-dependent manner. The maximum inhibitory response to KT-90 and dynorphin (I-13) was not significantly different from each other (Fig. 2). The plC~-values of KT-90, dynorphin (I-13) and other drugs were summerized in Table 6. With lower concentrations of KT-90,

( -- )-6fl-Acetylthionormorphine derivatives Table 6. The plC~-values of test drugs on synaptosomal Ca-uptake in rat brain. plC~-values Dynorphin (1-13) 9.07 + 0.17 KT-90 8.76 _ 0.29 Nalorphine 8.08 +_0.08 Nalorphine epoxide 7.49 + 0.30 Morphine 6.95 + 0.13 Morphiceptin N.D. DADLE N.D. Each value represents the mean + SE of 4--6 experiments. DADLE: o-Ala2-D-LeuL enkephalin. N.D.: not detected. morphine (10 7 M) acted as a synergist, but with higher concentrations of KT-90, a competitive antagonism between KT-90 and morphine was observed (Fig. 2), indicating that morphine behaves as a partial agonist on K-receptors. These results suggest that KT-90 inhibit the synaptosomal Ca-uptake in the rat brain through r-receptors.

Dependence liability of KT-90 Physical dependence. Physical dependence on opioids can quickly be induced in rodents by a variety of techniques, including implantation of morphine pellets, infusion of opioids, treatment with a slow release emulsion of opioids, implantation of a reservoir of morphine solution and treatment with drug-admixed food (Suzuki, 1990). The intermittent infusion of opioids through an implanted cannula produced rapid development of physical dependence on opioids in rats (Koga, 1976; Suzuki et al., 1980). This procedure provides a useful system for estimating physical dependence potentials of analgesics, such as morphine, codeine, meperidine, pentazocine and cyclazocine. We examined on utilization of this procedure in rats to evaluate a physical dependence liability of KT-90. Rats were intermittently medicated for 1-5 days at 1 hr intervals through an implanted intravenous cannula. Rats treated with morphine or pentazocine (48mg/kg/day) showed marked weight loss, diarrhea and ptosis after the termination of drug treatment. However, rats treated with KT-90 (96 mg/kg/day) did not show any withdrawal signs. In a naloxone-precipitated withdrawal test, rats treated with morphine or pentazocine also showed marked weight loss, body shakes, vocalization, diarrhea, ptosis, salivation, nose-bleed, etc., when they were given naloxone (3 mg/kg, s.c.). However, rats treated with KT-90 showed only body shakes and ptosis after the naloxone challenge. These results suggest that the dependence-producing capacity of KT-90 is to a very low degree, if any. It has been demonstrated in the whole animal and man that chronic treatment with a partial opioid agonist, buprenorphine, induces only a very low degree, if any, of dependence (Cowan et al., 1977; Mello and Mendelson. 1980). These reports are based upon experiments which exprapolate the degree of dependence from the intensity of withdrawal seen upon interruption of buprenorphine supply or direct precipitation of withdrawal with an antagonist after chronic treatment with buprenorphine. The results of such experiments must be interpreted with caution since buprenorphine is known to have very slow receptor kinetics (Cowan et al., 1977) and the disso-

609

ciation of the drug from the receptor might affect the intensity of withdrawal. For this reason, Dum et al. (1981) developed the following new method for evaluation of physical dependence liability. Rats were pretreated twice daily, 4 days, with 0.5 mg/kg (s.c.) buprenorphine. On day 5, l0 mg/kg (s.c.) morphine was substituted for buprenorphine so that the degree of dependence could be judged by the intensity of withdrawal following the displacement of morphine, rather than of buprenorphine, from the receptor by antagonist, 10mg/kg (i.p.) naloxone. Rats treated with buprenorphine for 4 days showed only very weak signs of withdrawal upon cessation of buprenorphine treatment or upon challenge with naioxon¢. However, more intense withdrawal could be induced when buprenorphine treatment was followed by substitution treatment with morphine. Using this method, physical dependence liability of KT-90 and buprenorphine was examined. Rats chronically treated with 5 mg/kg (s.c.) KT-90 or 0.5 mg/kg (s.c.) buprenorphine for 4 days showed only very weak signs of withdrawal upon challenge with naloxone. The intensity of withdrawal signs in KT-90-treated rats was similar to that in buprenorphine-treated rats (Table 7). However, naloxone-precipitated withdrawal signs were potentiated in both groups when KT-90 or buprenorphine treatment was followed by substitution treatment with 10mg/kg (s.c.) morphine. Buprenorphine-treated rats showed diarrhea, jumping and salivation, but KT-90-treated rats did not (Table 7). These results indicate that physical dependence liability of KT-90 may be weaker than that of buprenorphine. Psychic dependence. The place conditioning procedure is used to evaluate the motivational properties of drugs. It was introduced early this decade to compensate for methodological and interpretive difficulties associated with the self-administration technique, the conventional method for assessing reinforcing properties of drugs (Capell and Le Blanc, 1981; Schuster and Johanson, 1981). Using this paradigm, the primary reinforcing properties of a drug are conditioned to salient environmental stimuli which by association, acquire secondary reinforcing properties. Saline injections are paired with different stimuli and, on the test day, the animal, in the undrugged state, is allowed to choose between the stimuli. This technique only requires that the animals Table 7. Naloxone-precipitatedwithdrawalsignsin rats treated with KT-90 and buprenorphine, and the effect of substitute treatment with morphine on naloxone-precipitatedwithdrawalsigns Morphine substitution Withdrawal signs KT-90 Buprenorphine KT-90 Buprenorphine Body shakes 1.4 4.4 0.8 1.2 Vocalization 2/5 I/5 4/5 I/5 Chewing 3/5 4/5 5/5 5,,'5 Diarrhea 0/5 0/5 0/5 I/5 Ptosis 5/5 3/5 5/5 3/5 Jumping 0/5 0/5 0/5 2/5 Salivation 0/5 0/5 0/5 2/5 Lacrimation 0/5 0/5 0/5 I/5 Scratching 0/5 0/5 0/5 1/5 Erection 0/5 I/5 4/5 I/5 Writhing 0/5 0/5 1/5 3/5 Body shakes are the mean number of 5 rats and other behavioral data are expressedas the number of subjects presenting a particular sign over the total number of subjectsobserved.

ISSEITAKAYANAGIet al.

610

Table 8. Pharmacologicalproperties, potency, affinityand selectivity of AcS-morphine, KT-89 and KT-90 N-methyl N-cyclopropylmethyl derivative derivatives Ac~-mor-Phine KT-89. . . . . . K T --90..... Pharmacological l~-agonist ~,'-agonist ~c-agonist properties ~u-antagonist /z-antagonist Potency >Morphine > Morphine > Morphine Affinity >Morphine :, Morphine > Morphine Selectivity g-selective nonselective nonselective