Life Sciences, Vol. 51, pp. PL 89-94 Printed in the USA
Pergamon P r e s s
PHARMACOLOGY Accelerated
LETTERS
Communication
D I F F E R E N T I A L EFFEC"I~ OF C H R O N I C AGONIST ADMINISTRATION ON/.t,-OPIOID RECEPTOR- AND MUSCARINIC R E C E P T O R - R E G U L A T E D ADENYLATE CYCLASE IN RAT STRIATAL NEURONS Bernard J. Van Vliet, Corinne H. Dotman, George Wardeh, Arie H. Mulder and Anton N.M. Schoffelmeer" Department of Pharmacology, Free University, Medical Faculty van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands (Submitted April 15, 1992; accepted May 12, 1992; received in final form June 23, 1992)
Abstract In cultured rat striatal neurons exposed to 10/zM morphine or oxotremorine for 24 hours, we observed an increased (about 30 %) dopamine Di receptor-stimulated cyclic AMP production, whereas no desensitization of tz-opioid receptor or muscarinic cholinergic receptor was found. However, whereas upregulation of dopamine D 1 receptor-stimulated adenylate cyclase activity upon 7 days morphine exposure was at least as pronounced as observed after 24 hours of exposure to the opioid, this adaptive phenomenon was virtually absent following one week of oxotremorine treatment. This reduced adenylate cyclase sensitization upon 7 days oxotremorine exposure appeared to coincide with desensitization of muscarinic cholinergic receptors. A possible role of the resistance of lz receptors to desensitization and the (resulting) upregulation of the neuronal adenylate cyclase system upon chronic receptor activation in the development of opiate tolerance and dependence is suggested.
Introduction
Our knowledge of the molecular mechanisms that underlie opioid tolerance and the withdrawal phenomena is as yet very limited. It has been suggested that, at least in tumor-related cells such as 7315c pituitary tumor cells (1), desensitization and down-regulation of /~-opiate receptors inhibitorily linked to adenylate cyclase may play a role in morphine tolerance. On the other hand, increased neuronal levels of second messengers that play an important role in neuronal activity, such as calcium and cyclic AMP, may be involved in withdrawal effects observed following cessation of chronic opioid administration (2,3). A neuronal model system offering the advantage of investigating receptor-effector coupling in a fully defined medium is that of striatal neurons in primary culture (4). Studying this important issue in primary cultures of rat striatal neurons chronically exposed to morphine, we observed no desensitization of the/~-opioid receptors inhibitorily linked to adenylate cyclase (5), whereas a profound upregulation of dopamine D l receptor- and 13-adrenoceptor-stimulated adenylate cyclase activity was apparent upon morphine withdrawal. In this respect cultured striatal neurons appeared to be an excellent model for central neurons, since these effects were also observed in the striatum of morphine-dependent rats (6). Recently, sensitization of receptor-stimulated adenylate cyclase activity was suggested to represent a general effect occurring in cells upon chronic administration of agonists acting on receptors
" To whom coffespondence should be addressed. 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd All rights reserved.
PL-90
Differential Mechanisms of Drug Tolerance
Vol. 51, No. 10, 1992
inhibitorily linked to the enzyme (7). Since muscarinic cholinergic receptors have been shown to be inhibitorily linked to adenylate cyelase in striatal neurons (8) and desensitization as well as down-regulation of muscarinic receptors has consistently been demonstrated (9,10,11), we compared the effects of chronic activation of/~-opioid receptors with that of musearinic receptors in cultured striatal neurons in the present study. It was reasoned that the lack of receptor desensitization upon chronic morphine exposure may be unique for/~-opioid receptors, which might be important regarding the long-term changes in neurons underlying opiate dependence. Methods
Preparation of primary cultures of striatal neurons. Primary neuronal cultures were prepared as previously described (12), according to a method reported by Boekaert et al., (4) for the cultures of mouse striatal neurons. Briefly, striata were dissected from 17-day-old rat embryos (Wistar I-Isd/Cpb:WU) and were mechanically dissociated, using a fire-narrowed Pasteur pipet, in a serumfree medium. Cells were plated in 12-well Linbro culture dishes (6 x 105 cells/ml per well), previously coated with poly-L-ornithine (1.5 #g/ml) and medium containing 10 % supplemented calf serum. The culture medium was composed of a 1:1 mixture of Dulbecco's modified Eagles medium and F-12 nutrient and contained glucose (0.6 %), glutamine (2 mM), sodium bicarbonate (3 Mm), HEPES buffer (5mM), streptomycin (100/zg/ml) and penicillin (100 IU/ml). A defined hormone and salt mixture was added, consisting of insulin (25 /zg/ml), transferrin (100 /zg/ml), progesterone (20 nM), putrescine (60/zM), ILestradiol (1 pM) and selenium sodium salt (30 nM). Under our culture conditions, the neurons developed an extensive matrix of dendrites and synapses, as described by Bockaert et al. (4), and survived for more than 4 weeks. Since neurons were grown in a serum-free (hormone supplemented) medium, only a few glial cells were observed (less than 10 % of total cells, as determined by anti-GFAP staining) after 7-14 days in vitro (DIV), when cultures were used for experiments. Although the presence of type II astrocytes cannot be excluded, immunocytochemical staining of the cells with 200,000 MW neurofilament (NF) supports the fact that neurons represent more than 90 % of the cells. At the microscopical level we did not detect any morphological change of the cultured neurons chronically exposed to oxotremorine or morphine. Moreover, [3H]ATP and [3H]yaminobutyric acid (GABA) accumulation by the neurons remained unaffected following drug treatment, suggesting no apparent effect of the drugs on cell growth. Determination of cyclic AMP production. Cyclic AMP production was determined in 7 DIV-old cultured neurons (10 #M morphine pretreatment) or 14 DIV-old cultured neurons (10 /zM oxotremorine pretreatment) by measuring the conversion of [3H]adenine nucleotide precursors (mainly [3H]ATP) to [3H]cyclic AMP. As shown previously for morphine (13), oxotremorine did not appear to be fully stable under tissue culture conditions. Thus, studying the efficacy of the culture medium at presynaptic muscarinic receptors (14), we found that the oxotremorine concentration appeared to be reduced from 10 /~M to 1-3 /~M within 7 days (data not shown) as found for morphine (13). Thus, although some breakdown of the ligand did occur, the final concentration of oxotremorine would still maximally activate muscarinic receptors in the absence of receptor desensitization. Following chronic treatment (24 hours or 7 days), the cultures were washed with phosphatebuffered s~iline (PBS) containing (raM) 137 NaCI, 2.7 KCI, 8 Na2HPO4, 1.5 KH2PO4, 0.5 MgCI2, 1.2 CaCI2 and 5 glucose (pH 7.3) before incubation at 37°C with 2/~Ci [3H]adenine in absence or presence of 10 /~M morphine or oxotremorine. After 2 h, the preparations were rapidly washed with 6 volumes of PBS in order to remove morphine or oxotremorine (for details about the efficiency of the washing procedure, see (5)), and exposed for 20 rain to a PBS solution containing drugs (SKF 38393, oxotremorine, DAGO) and 3-isobutyl-l-methylxanthine (IBMX, 1 mM). The reaction was stopped by aspiration of the media and addition of 1 ml ice-cold
Vol.
51, No.
i0,
1992
Differential
Mechanisms
of D r u g T o l e r a n c e
PL-91
triehioroacetic acid (5 %). [3I-l]eyclic AMP was separated from [3H]ATP through sequential chromatography on Dowex and alumina columns (15). Cyclic AMP production was expressed as: [3H]cyclic AMP % conversion = x 100 [3H]ATP + [3H]cyclic AMP This prelabelling technique, originally described by Shimizu et al. (16), has yielded results completely consistent with those based on measurement of endogenous levels of cyclic AMP (17), and has been used in various subsequent studies with cell cultures (4). Statistics. Statistical significance of differences was determined by one-way analysis of variance (ANOVA) followed by the Student's t-test. Chemicals. The following drugs were obtained commercially: [3I-l]adenine (15-25 Ci/mmol) from Amersham; atropine, oxotremorine and naloxone hydroehloride from Sigma; morphine hydroehloride from O.P.G. (Utrecht, the Netherlands); D A G O from Bachem. Results
Effect of muscarinic receptor activation on dopamine D 1 receptor-stimulated adenylate cyclase activity. Activation of dopamine DI receptors by SKF 38393 (1 /zM) resulted in an increase of adenylate cyclase activity, which was inhibited by simultaneous activation of muscarinic receptors with oxotremorine (1 p.M). The inhibitory effect of oxotremorine was completely abolished by a low concentration (0.01/zM) of the muscarinic receptor antagonist atropine (Table 1), which alone did not affect the stimulatory effect of SKF 38393 on cyclic AMP production. Following washing of the cultured neurons, the stimulatory effect of SKF 38393 (1 /zM) on cyclic AMP production in neurons exposed to morphine (10 /zM) or oxotremorine (10 /~M) for 24 h was significantly higher than that in untreated neurons (Figure 1). Whereas the increase of dopamine D~ receptorstimulated cyclic AMP production was still observed in neurons exposed to morphine for 7 days, it was virtually absent in neurons exposed for one week to oxotremorine (Figure 1). TABLE I Inhibitory Effect of Muscarinic Receptor Activation on Dopamine D I Receptor-Stimulated Cyclic AMP Production. Drugs SKF 38393 1 /~M " + oxotremorine 1 /zM " + " + atropine 0.01 /zM
Adenylate cyclase activity (% conversion ATP in cAMP) 0.52 _+ 0.01 0.41 -+ 0.01 * 0.52 --. 0.01
In the absence of drugs, conversion of ATP in cyclic AMP amounted to 0.13 _+ 0.01 %. Atropine alone had no effect on dopamine D 1 receptor-stimulated cyclic AMP production. Data are mean -+ S.E.M. values of 4 observations. * P < 0.01. Effect of chronic morphine and oxotremorine treatment on receptor-sensitivity. Figure 2 shows the dose-dependent inhibition by the/~-opioid receptor selective agonist [D-AIa 2, MePhe 4, Gly-olS] enkephalin (DAGO) and oxotremorine of dopamine D 1 receptor-stimulated cyclic AMP production in untreated neurons and in neurons exposed to 10/~M morphine or oxotremorine for 24 h or 7 days. Analysis of data, obtained in three separate experiments showed that the ECs0 value of oxotremorine (0.09 _+ 0.02 /zM) was not significantly affected by exposure of the neurons to
PL-92
Differential
i
i
Mechanisms
of D r u g T o l e r a n c e
Vol.
51, No.
140
"6 120
~
~00
o C
24h
7
d
HG. 1 Time-dependent effect of morphine and oxotremorine exposure on dopamine D] receptor-stimulated cyclic AMP production in rat striatal neurons. Neurons were exposed to 10 /~M morphine (black bars) or 10 /~M oxotremorine (hatched bars) for 24 hours or 7 days and subsequently washed to remove morphine or oxotremorine before stimulation of cyclic A M P production with SKF 38393 (1 /zM). Dopamine D l receptorstimulated conversion of ATP in cyclic AMP amounted to 0.47 -4- 0.04 % and 0.63 _+ 0.03 % in neurons used to study the effect of morphine and oxotremorine, respectively. Data are mean _+ S.E.M. (bars) values of three experiments performed in quadruplicate. C: control, i.e. no morphine or oxotremorine exposure; h: hours; d: days. * Significantly higher than in untreated neurons (P < 0.01).
° lif
loo
~
~o
~ h~
75
d
o,6
6o 50 ,o
~
A C
9
8
7
6
- log [DAGO] ~v0
C -
8
7
6
5
log [ o x o t r e r n ~ i n e ] (Iv0
FIG. 2 Effect of chronic agonist exposure on the inhibition by morphine or oxotremorine of dopamine D 1 receptor-stimulated cyclic A M P production in rat striatal neurons. Neurons were exposed to 10/~M morphine (A) or oxotremorine (B) for various periods of time ( - - o : no drug exposure; - - e - - : 24 hours; • . i . • : 7 days) and subsequently washed to remove morphine or oxotremorine before stimulation of cyclic AMP production with SKF 38393 (1 /~M), in the presence of increasing concentrations of D A G O (A) or oxotremorine (B). Data are mean -+ S.E.M. (bars) values from 3 experiments performed in quadruplicate.
i0, 1992
Vol.
51, No.
10, 1992
oxotremorine for 24 significant (P < 0.01) = 1.25 -+ 0.04 #M). and was independent
Differential
Mechanisms
of Drug Tolerance
PL-93
h (0.15 -+ 0.04 /zM), whereas an exlx~ure period of 7 days caused a 10-fold shift to the right of the dose response curve of oxotremorine (ECs0 On the other hand the ECs0 value of D A G O amounted to 5.9 -+ 1.2 nM of pretreatment of the neurons with morphine. Discussion
In agreement with other studies (8), our present data show that in addition to the/~-opioid receptors, muscarinic receptors may mediate inhibition of adenylate cyclase activity in cultured striatal neurons, although the type (M 2 or M4) of muscarinie receptors involved is as yet unclear (8,18). This enabled us to investigate the importance of the absence of /~-opioid receptor desensitization regarding the enhanced effect of dopamine D l receptor stimulation on adenylate cyclase following chronic morphine treatment. Upon exposure of the neurons to either morphine or oxotremorine for 24 hours, dopamine D I receptor-stimulated cyclic AMP production was enhanced and no receptor-desensitization was observed. However, upon more prolonged activation (7 days) of these receptors, striking differences were observed at opioid receptor- and muscarinic receptor-regulated adenylate cyclase. Thus, whereas upregulation of dopamine D 1 receptorstimulated adenylate cyclase upon 7 days morphine treatment was at least as pronounced as after 24 hours opioid administration, the enhanced activity of this second messenger system was virtually absent following one week oxotremorine treatment. Interestingly, the strongly reduced upregulation of adenylate cyclase activity upon 7 days oxotremorine administration appeared to coincide with a 10-fold shift to the right of the dose-response curve of the muscarinic receptor agonist, indicating a profound muscarinic receptor desensitization. In contrast, as shown previously (5), no ~-opioid receptor desensitization appeared to occur. Although in Figure 2 only the effect of the most efficacious and selective p. agonist D A G O is shown, the inhibitory effect of less effective agonists (morphine, [D-Ala2-D-LeuS]enkephalin) also appeared to be completely unchanged in morphine-treated neurons. Moreover, chronic treatment of the neurons with D A G O also caused a profound increase of dopamine Dt receptor-stimulated cyclic AMP production without the occurence of p.-opioid receptor desensitization (data not shown). Possible reasons for the absence of a significant desensitization of the adenylate cyclase-coupled kt-opioid receptors could be that (i) partial uncoupling of opioid receptors from G-proteins, phosphorylation of the receptors or some down-regulation of the number of opioid binding sites does occur but has no significant effect on opioid agonist efficacy due to the presence of a very large receptor reserve or (ii) adenylate cyclase coupled/.r-opioid receptors are structurally resistant to desensitization. Obviously, these possibilities deserve to be further investigated. It should be emphasized here that we do not rule out the possibility that desensitization of opioid receptors (coupled to other effector systems) may occur in the brain (13). However, this does not seem to play an important role at /~-opioid receptors coupled to adenylate cyclase in rat striatum in view of the fact that similar results were recently obtained in rat striatal slices obtained from morphine-dependent rats (6). Taken together, the data presented here suggest that tolerance to an agonist may be due to a loss of agonist affinity upon receptor desensitization (muscarinic receptors) or to the fact that the agonist is acting against an upregulated second messenger system counteracting the acute effect of the agonist (p-opioid receptors, see also 5 and 6). Moreover, we tentatively suggest that there is a causal relationship between the lack of significant/~-opioid receptor desensitization and the persistent increase of dopamine D~ receptor-stimulated adenylate cyclase activity. In view of the role of dopamine in opiate-dependence (19), the lack of/.~-opioid receptor desensitization and the (resulting) increase of dopamine D~ receptor-stimulated cyclic AMP production upon chronic morphine administration may play an important role in the development and/or maintainance of opiate dependence. Although the type of neurons containing the/.r-opioid receptors studied here is unknown, it is likely that GABA neurons are involved, given (i) the vast majority of these neurons with cell bodies in the striatum (20) and which as a result will survive in the cultures (4, unpublished observations), and (ii) the well-known inhibitory effect of/.~ receptor activation on the activity of GABA neurons (19).
PL-94
Differential Mechanisms of Drug Tolerance
Vol. 51, No. i0, 1992
Acknowledgements This work was supported by a Senior Fellowship of the Royal Netherlands Academy of Science and Arts (K.N.A.W.) awarded to Anton N.M. Sehoffelmeer. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
P.S. PUTrFARCKEN and B.M. COX, Life Sci. 45 1937-1942 (1989). A.P. SMITH, P.-L LAW and H.H. LOH, The opiate receptors, G.W. Pastemak (ed), 441485, The Human Press, Clifton, New Jersey (1988). S.M. JOHNSON and W.W. FLEMMING, Pharmacol. Rcv. 41 435-488 (1989). J. BOCKAERT, J. GABRION, F. SLADECZEK, J.-P. PIN, M. RECASENS, M. SEBBEN, D. KEMP, A. DUMUIS and S. WEISS, J. Physiol., Pads 81 219-227 (1986). B.J. VAN VLIET, T.J. DE VRIES, G. WARDEH, A.H. MULDER and A.N.M. SCHOFFELMEER, Eur. J. Pharmacol. Molecular Pharmacol. Section 208 105-111 (1991). T.J. DE VRIES, B.J. VAN VLIET, F. HOGENBOOM, G. WARDEH, J.W. VAN DER LAAN, A.H. MULDER and A.N.M. SCHOFFELMEER, Eur. J. Pharmacol. 208 97-104 (1991). J.M. THOMAS and B.B. HOFFMAN, Trends in Pharmacol. Sci. _8 308-311 (1987). J. ELLIS, J.H. HUYLER, D.E. KEMP and S. WEISS, Brain Res. 51__1234-240 (1990). D.A. GREEN and R.B. CLARK, J. Neurochem. 39 1125-1131 (1982). O. DILLON-CARTER and D.-M. CHUANG, J. Neurochem. 52 598-604 (1989). F. FUKAMAUCHI, C. H O U G H and D.-M. CHUANG, J. Neurochem. 56 716-719 (1991). B.J. VAN VLIET., A.H. MULDER and A.N.M. SCHOFFELMEER, J. Neurochem. 55, 12741280 (1990). C. KENNEDY and G. HENDERSON, Mol. Pharmacol. 40 1000-1005 (1991). A.N.M. SCHOFFELMEER, B.J. VAN VLIET, G. WARDEH and A.H. MULDER, Eur. J. Pharmacol. 128 291-294 (1986). Y. SALOMON, C. LONDOS and M. RODBELL, Analytical Biochem. 58 541- (1974). H. SHIMIZU, J.W. DALY and C.R. CREVELING, J. Neurochem. 16 1609- (1969). J.W. DALY, Cyclic nucleotides in the nervous system, Plenum Press, New York (1977). M. McKINNEY, J.H. HILLER, V.A. GIBSON, L. NICKELSON and S. AKSOY, Mol. Pharmacol. 40 1014-1022 (1991). G.D. DI CHIARA and R.A. NORTH, Trends Pharmacol. Sci. 13 185-193 (1992). A.M. GRAYBIEL, Trends Neurosci. 13 233-244 (1992).
Pergamon P r e s s
PHARMACOLOGY Accelerated
LETTERS
Communication
D I F F E R E N T I A L EFFEC"I~ OF C H R O N I C AGONIST ADMINISTRATION ON/.t,-OPIOID RECEPTOR- AND MUSCARINIC R E C E P T O R - R E G U L A T E D ADENYLATE CYCLASE IN RAT STRIATAL NEURONS Bernard J. Van Vliet, Corinne H. Dotman, George Wardeh, Arie H. Mulder and Anton N.M. Schoffelmeer" Department of Pharmacology, Free University, Medical Faculty van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands (Submitted April 15, 1992; accepted May 12, 1992; received in final form June 23, 1992)
Abstract In cultured rat striatal neurons exposed to 10/zM morphine or oxotremorine for 24 hours, we observed an increased (about 30 %) dopamine Di receptor-stimulated cyclic AMP production, whereas no desensitization of tz-opioid receptor or muscarinic cholinergic receptor was found. However, whereas upregulation of dopamine D 1 receptor-stimulated adenylate cyclase activity upon 7 days morphine exposure was at least as pronounced as observed after 24 hours of exposure to the opioid, this adaptive phenomenon was virtually absent following one week of oxotremorine treatment. This reduced adenylate cyclase sensitization upon 7 days oxotremorine exposure appeared to coincide with desensitization of muscarinic cholinergic receptors. A possible role of the resistance of lz receptors to desensitization and the (resulting) upregulation of the neuronal adenylate cyclase system upon chronic receptor activation in the development of opiate tolerance and dependence is suggested.
Introduction
Our knowledge of the molecular mechanisms that underlie opioid tolerance and the withdrawal phenomena is as yet very limited. It has been suggested that, at least in tumor-related cells such as 7315c pituitary tumor cells (1), desensitization and down-regulation of /~-opiate receptors inhibitorily linked to adenylate cyclase may play a role in morphine tolerance. On the other hand, increased neuronal levels of second messengers that play an important role in neuronal activity, such as calcium and cyclic AMP, may be involved in withdrawal effects observed following cessation of chronic opioid administration (2,3). A neuronal model system offering the advantage of investigating receptor-effector coupling in a fully defined medium is that of striatal neurons in primary culture (4). Studying this important issue in primary cultures of rat striatal neurons chronically exposed to morphine, we observed no desensitization of the/~-opioid receptors inhibitorily linked to adenylate cyclase (5), whereas a profound upregulation of dopamine D l receptor- and 13-adrenoceptor-stimulated adenylate cyclase activity was apparent upon morphine withdrawal. In this respect cultured striatal neurons appeared to be an excellent model for central neurons, since these effects were also observed in the striatum of morphine-dependent rats (6). Recently, sensitization of receptor-stimulated adenylate cyclase activity was suggested to represent a general effect occurring in cells upon chronic administration of agonists acting on receptors
" To whom coffespondence should be addressed. 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd All rights reserved.
PL-90
Differential Mechanisms of Drug Tolerance
Vol. 51, No. 10, 1992
inhibitorily linked to the enzyme (7). Since muscarinic cholinergic receptors have been shown to be inhibitorily linked to adenylate cyelase in striatal neurons (8) and desensitization as well as down-regulation of muscarinic receptors has consistently been demonstrated (9,10,11), we compared the effects of chronic activation of/~-opioid receptors with that of musearinic receptors in cultured striatal neurons in the present study. It was reasoned that the lack of receptor desensitization upon chronic morphine exposure may be unique for/~-opioid receptors, which might be important regarding the long-term changes in neurons underlying opiate dependence. Methods
Preparation of primary cultures of striatal neurons. Primary neuronal cultures were prepared as previously described (12), according to a method reported by Boekaert et al., (4) for the cultures of mouse striatal neurons. Briefly, striata were dissected from 17-day-old rat embryos (Wistar I-Isd/Cpb:WU) and were mechanically dissociated, using a fire-narrowed Pasteur pipet, in a serumfree medium. Cells were plated in 12-well Linbro culture dishes (6 x 105 cells/ml per well), previously coated with poly-L-ornithine (1.5 #g/ml) and medium containing 10 % supplemented calf serum. The culture medium was composed of a 1:1 mixture of Dulbecco's modified Eagles medium and F-12 nutrient and contained glucose (0.6 %), glutamine (2 mM), sodium bicarbonate (3 Mm), HEPES buffer (5mM), streptomycin (100/zg/ml) and penicillin (100 IU/ml). A defined hormone and salt mixture was added, consisting of insulin (25 /zg/ml), transferrin (100 /zg/ml), progesterone (20 nM), putrescine (60/zM), ILestradiol (1 pM) and selenium sodium salt (30 nM). Under our culture conditions, the neurons developed an extensive matrix of dendrites and synapses, as described by Bockaert et al. (4), and survived for more than 4 weeks. Since neurons were grown in a serum-free (hormone supplemented) medium, only a few glial cells were observed (less than 10 % of total cells, as determined by anti-GFAP staining) after 7-14 days in vitro (DIV), when cultures were used for experiments. Although the presence of type II astrocytes cannot be excluded, immunocytochemical staining of the cells with 200,000 MW neurofilament (NF) supports the fact that neurons represent more than 90 % of the cells. At the microscopical level we did not detect any morphological change of the cultured neurons chronically exposed to oxotremorine or morphine. Moreover, [3H]ATP and [3H]yaminobutyric acid (GABA) accumulation by the neurons remained unaffected following drug treatment, suggesting no apparent effect of the drugs on cell growth. Determination of cyclic AMP production. Cyclic AMP production was determined in 7 DIV-old cultured neurons (10 #M morphine pretreatment) or 14 DIV-old cultured neurons (10 /zM oxotremorine pretreatment) by measuring the conversion of [3H]adenine nucleotide precursors (mainly [3H]ATP) to [3H]cyclic AMP. As shown previously for morphine (13), oxotremorine did not appear to be fully stable under tissue culture conditions. Thus, studying the efficacy of the culture medium at presynaptic muscarinic receptors (14), we found that the oxotremorine concentration appeared to be reduced from 10 /~M to 1-3 /~M within 7 days (data not shown) as found for morphine (13). Thus, although some breakdown of the ligand did occur, the final concentration of oxotremorine would still maximally activate muscarinic receptors in the absence of receptor desensitization. Following chronic treatment (24 hours or 7 days), the cultures were washed with phosphatebuffered s~iline (PBS) containing (raM) 137 NaCI, 2.7 KCI, 8 Na2HPO4, 1.5 KH2PO4, 0.5 MgCI2, 1.2 CaCI2 and 5 glucose (pH 7.3) before incubation at 37°C with 2/~Ci [3H]adenine in absence or presence of 10 /~M morphine or oxotremorine. After 2 h, the preparations were rapidly washed with 6 volumes of PBS in order to remove morphine or oxotremorine (for details about the efficiency of the washing procedure, see (5)), and exposed for 20 rain to a PBS solution containing drugs (SKF 38393, oxotremorine, DAGO) and 3-isobutyl-l-methylxanthine (IBMX, 1 mM). The reaction was stopped by aspiration of the media and addition of 1 ml ice-cold
Vol.
51, No.
i0,
1992
Differential
Mechanisms
of D r u g T o l e r a n c e
PL-91
triehioroacetic acid (5 %). [3I-l]eyclic AMP was separated from [3H]ATP through sequential chromatography on Dowex and alumina columns (15). Cyclic AMP production was expressed as: [3H]cyclic AMP % conversion = x 100 [3H]ATP + [3H]cyclic AMP This prelabelling technique, originally described by Shimizu et al. (16), has yielded results completely consistent with those based on measurement of endogenous levels of cyclic AMP (17), and has been used in various subsequent studies with cell cultures (4). Statistics. Statistical significance of differences was determined by one-way analysis of variance (ANOVA) followed by the Student's t-test. Chemicals. The following drugs were obtained commercially: [3I-l]adenine (15-25 Ci/mmol) from Amersham; atropine, oxotremorine and naloxone hydroehloride from Sigma; morphine hydroehloride from O.P.G. (Utrecht, the Netherlands); D A G O from Bachem. Results
Effect of muscarinic receptor activation on dopamine D 1 receptor-stimulated adenylate cyclase activity. Activation of dopamine DI receptors by SKF 38393 (1 /zM) resulted in an increase of adenylate cyclase activity, which was inhibited by simultaneous activation of muscarinic receptors with oxotremorine (1 p.M). The inhibitory effect of oxotremorine was completely abolished by a low concentration (0.01/zM) of the muscarinic receptor antagonist atropine (Table 1), which alone did not affect the stimulatory effect of SKF 38393 on cyclic AMP production. Following washing of the cultured neurons, the stimulatory effect of SKF 38393 (1 /zM) on cyclic AMP production in neurons exposed to morphine (10 /zM) or oxotremorine (10 /~M) for 24 h was significantly higher than that in untreated neurons (Figure 1). Whereas the increase of dopamine D~ receptorstimulated cyclic AMP production was still observed in neurons exposed to morphine for 7 days, it was virtually absent in neurons exposed for one week to oxotremorine (Figure 1). TABLE I Inhibitory Effect of Muscarinic Receptor Activation on Dopamine D I Receptor-Stimulated Cyclic AMP Production. Drugs SKF 38393 1 /~M " + oxotremorine 1 /zM " + " + atropine 0.01 /zM
Adenylate cyclase activity (% conversion ATP in cAMP) 0.52 _+ 0.01 0.41 -+ 0.01 * 0.52 --. 0.01
In the absence of drugs, conversion of ATP in cyclic AMP amounted to 0.13 _+ 0.01 %. Atropine alone had no effect on dopamine D 1 receptor-stimulated cyclic AMP production. Data are mean -+ S.E.M. values of 4 observations. * P < 0.01. Effect of chronic morphine and oxotremorine treatment on receptor-sensitivity. Figure 2 shows the dose-dependent inhibition by the/~-opioid receptor selective agonist [D-AIa 2, MePhe 4, Gly-olS] enkephalin (DAGO) and oxotremorine of dopamine D 1 receptor-stimulated cyclic AMP production in untreated neurons and in neurons exposed to 10/~M morphine or oxotremorine for 24 h or 7 days. Analysis of data, obtained in three separate experiments showed that the ECs0 value of oxotremorine (0.09 _+ 0.02 /zM) was not significantly affected by exposure of the neurons to
PL-92
Differential
i
i
Mechanisms
of D r u g T o l e r a n c e
Vol.
51, No.
140
"6 120
~
~00
o C
24h
7
d
HG. 1 Time-dependent effect of morphine and oxotremorine exposure on dopamine D] receptor-stimulated cyclic AMP production in rat striatal neurons. Neurons were exposed to 10 /~M morphine (black bars) or 10 /~M oxotremorine (hatched bars) for 24 hours or 7 days and subsequently washed to remove morphine or oxotremorine before stimulation of cyclic A M P production with SKF 38393 (1 /zM). Dopamine D l receptorstimulated conversion of ATP in cyclic AMP amounted to 0.47 -4- 0.04 % and 0.63 _+ 0.03 % in neurons used to study the effect of morphine and oxotremorine, respectively. Data are mean _+ S.E.M. (bars) values of three experiments performed in quadruplicate. C: control, i.e. no morphine or oxotremorine exposure; h: hours; d: days. * Significantly higher than in untreated neurons (P < 0.01).
° lif
loo
~
~o
~ h~
75
d
o,6
6o 50 ,o
~
A C
9
8
7
6
- log [DAGO] ~v0
C -
8
7
6
5
log [ o x o t r e r n ~ i n e ] (Iv0
FIG. 2 Effect of chronic agonist exposure on the inhibition by morphine or oxotremorine of dopamine D 1 receptor-stimulated cyclic A M P production in rat striatal neurons. Neurons were exposed to 10/~M morphine (A) or oxotremorine (B) for various periods of time ( - - o : no drug exposure; - - e - - : 24 hours; • . i . • : 7 days) and subsequently washed to remove morphine or oxotremorine before stimulation of cyclic AMP production with SKF 38393 (1 /~M), in the presence of increasing concentrations of D A G O (A) or oxotremorine (B). Data are mean -+ S.E.M. (bars) values from 3 experiments performed in quadruplicate.
i0, 1992
Vol.
51, No.
10, 1992
oxotremorine for 24 significant (P < 0.01) = 1.25 -+ 0.04 #M). and was independent
Differential
Mechanisms
of Drug Tolerance
PL-93
h (0.15 -+ 0.04 /zM), whereas an exlx~ure period of 7 days caused a 10-fold shift to the right of the dose response curve of oxotremorine (ECs0 On the other hand the ECs0 value of D A G O amounted to 5.9 -+ 1.2 nM of pretreatment of the neurons with morphine. Discussion
In agreement with other studies (8), our present data show that in addition to the/~-opioid receptors, muscarinic receptors may mediate inhibition of adenylate cyclase activity in cultured striatal neurons, although the type (M 2 or M4) of muscarinie receptors involved is as yet unclear (8,18). This enabled us to investigate the importance of the absence of /~-opioid receptor desensitization regarding the enhanced effect of dopamine D l receptor stimulation on adenylate cyclase following chronic morphine treatment. Upon exposure of the neurons to either morphine or oxotremorine for 24 hours, dopamine D I receptor-stimulated cyclic AMP production was enhanced and no receptor-desensitization was observed. However, upon more prolonged activation (7 days) of these receptors, striking differences were observed at opioid receptor- and muscarinic receptor-regulated adenylate cyclase. Thus, whereas upregulation of dopamine D 1 receptorstimulated adenylate cyclase upon 7 days morphine treatment was at least as pronounced as after 24 hours opioid administration, the enhanced activity of this second messenger system was virtually absent following one week oxotremorine treatment. Interestingly, the strongly reduced upregulation of adenylate cyclase activity upon 7 days oxotremorine administration appeared to coincide with a 10-fold shift to the right of the dose-response curve of the muscarinic receptor agonist, indicating a profound muscarinic receptor desensitization. In contrast, as shown previously (5), no ~-opioid receptor desensitization appeared to occur. Although in Figure 2 only the effect of the most efficacious and selective p. agonist D A G O is shown, the inhibitory effect of less effective agonists (morphine, [D-Ala2-D-LeuS]enkephalin) also appeared to be completely unchanged in morphine-treated neurons. Moreover, chronic treatment of the neurons with D A G O also caused a profound increase of dopamine Dt receptor-stimulated cyclic AMP production without the occurence of p.-opioid receptor desensitization (data not shown). Possible reasons for the absence of a significant desensitization of the adenylate cyclase-coupled kt-opioid receptors could be that (i) partial uncoupling of opioid receptors from G-proteins, phosphorylation of the receptors or some down-regulation of the number of opioid binding sites does occur but has no significant effect on opioid agonist efficacy due to the presence of a very large receptor reserve or (ii) adenylate cyclase coupled/.r-opioid receptors are structurally resistant to desensitization. Obviously, these possibilities deserve to be further investigated. It should be emphasized here that we do not rule out the possibility that desensitization of opioid receptors (coupled to other effector systems) may occur in the brain (13). However, this does not seem to play an important role at /~-opioid receptors coupled to adenylate cyclase in rat striatum in view of the fact that similar results were recently obtained in rat striatal slices obtained from morphine-dependent rats (6). Taken together, the data presented here suggest that tolerance to an agonist may be due to a loss of agonist affinity upon receptor desensitization (muscarinic receptors) or to the fact that the agonist is acting against an upregulated second messenger system counteracting the acute effect of the agonist (p-opioid receptors, see also 5 and 6). Moreover, we tentatively suggest that there is a causal relationship between the lack of significant/~-opioid receptor desensitization and the persistent increase of dopamine D~ receptor-stimulated adenylate cyclase activity. In view of the role of dopamine in opiate-dependence (19), the lack of/.~-opioid receptor desensitization and the (resulting) increase of dopamine D~ receptor-stimulated cyclic AMP production upon chronic morphine administration may play an important role in the development and/or maintainance of opiate dependence. Although the type of neurons containing the/.r-opioid receptors studied here is unknown, it is likely that GABA neurons are involved, given (i) the vast majority of these neurons with cell bodies in the striatum (20) and which as a result will survive in the cultures (4, unpublished observations), and (ii) the well-known inhibitory effect of/.~ receptor activation on the activity of GABA neurons (19).
PL-94
Differential Mechanisms of Drug Tolerance
Vol. 51, No. i0, 1992
Acknowledgements This work was supported by a Senior Fellowship of the Royal Netherlands Academy of Science and Arts (K.N.A.W.) awarded to Anton N.M. Sehoffelmeer. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
P.S. PUTrFARCKEN and B.M. COX, Life Sci. 45 1937-1942 (1989). A.P. SMITH, P.-L LAW and H.H. LOH, The opiate receptors, G.W. Pastemak (ed), 441485, The Human Press, Clifton, New Jersey (1988). S.M. JOHNSON and W.W. FLEMMING, Pharmacol. Rcv. 41 435-488 (1989). J. BOCKAERT, J. GABRION, F. SLADECZEK, J.-P. PIN, M. RECASENS, M. SEBBEN, D. KEMP, A. DUMUIS and S. WEISS, J. Physiol., Pads 81 219-227 (1986). B.J. VAN VLIET, T.J. DE VRIES, G. WARDEH, A.H. MULDER and A.N.M. SCHOFFELMEER, Eur. J. Pharmacol. Molecular Pharmacol. Section 208 105-111 (1991). T.J. DE VRIES, B.J. VAN VLIET, F. HOGENBOOM, G. WARDEH, J.W. VAN DER LAAN, A.H. MULDER and A.N.M. SCHOFFELMEER, Eur. J. Pharmacol. 208 97-104 (1991). J.M. THOMAS and B.B. HOFFMAN, Trends in Pharmacol. Sci. _8 308-311 (1987). J. ELLIS, J.H. HUYLER, D.E. KEMP and S. WEISS, Brain Res. 51__1234-240 (1990). D.A. GREEN and R.B. CLARK, J. Neurochem. 39 1125-1131 (1982). O. DILLON-CARTER and D.-M. CHUANG, J. Neurochem. 52 598-604 (1989). F. FUKAMAUCHI, C. H O U G H and D.-M. CHUANG, J. Neurochem. 56 716-719 (1991). B.J. VAN VLIET., A.H. MULDER and A.N.M. SCHOFFELMEER, J. Neurochem. 55, 12741280 (1990). C. KENNEDY and G. HENDERSON, Mol. Pharmacol. 40 1000-1005 (1991). A.N.M. SCHOFFELMEER, B.J. VAN VLIET, G. WARDEH and A.H. MULDER, Eur. J. Pharmacol. 128 291-294 (1986). Y. SALOMON, C. LONDOS and M. RODBELL, Analytical Biochem. 58 541- (1974). H. SHIMIZU, J.W. DALY and C.R. CREVELING, J. Neurochem. 16 1609- (1969). J.W. DALY, Cyclic nucleotides in the nervous system, Plenum Press, New York (1977). M. McKINNEY, J.H. HILLER, V.A. GIBSON, L. NICKELSON and S. AKSOY, Mol. Pharmacol. 40 1014-1022 (1991). G.D. DI CHIARA and R.A. NORTH, Trends Pharmacol. Sci. 13 185-193 (1992). A.M. GRAYBIEL, Trends Neurosci. 13 233-244 (1992).
