Pii: s0014-2999(02)01823-x

European Journal of Pharmacology 446 (2002) 37 – 42 Antagonist-induced A-opioid receptor up-regulation decreases G-protein receptor kinase-2 and dynamin-2 abundance in mouse spinal cord Minesh Patel, Benedict Gomes, Chintan Patel, Byron C. Yoburn* Department of Pharmaceutical Sciences, College of Pharmacy and Allied Health Professions, St. John’s University, 8000 Utopia Parkway, Queens, NY 11439, USA Received 31 January 2002; received in revised form 7 May 2002; accepted 16 May 2002 Chronic treatment with opioid receptor antagonists has been shown to increase the density of A-, y- and n-opioid receptors in cell culture and in the intact animal. Although opioid receptor antagonist-induced up-regulation is a robust phenomenon, the mechanisms responsible forthe increase in receptor density remain unclear. In the present study, changes in a kinase and a GTPase that have been implicated in G-protein-coupled receptor regulation were examined following opioid receptor antagonist treatment. Mice were implanted s.c. with analtrexone pellet or placebo pellet. On the eighth day following implantation, spinal cord was removed and G-protein receptor kinase-2(GRK-2) and dynamin-2 abundance were determined using a quantitative immunoblot approach. Changes in A-opioid receptor density werealso determined. Naltrexone treatment produced a significant (145%) increase in A-opioid receptor density. Naltrexone treatment wasassociated with a significant 36% decrease in GRK-2 and 30% decrease in dynamin-2 abundance in spinal cord. These data raise thepossibility that opioid receptor antagonist-induced A-opioid receptor up-regulation in the intact animal may be due to a reduction inconstitutive internalization of opioid receptors. D 2002 Elsevier Science B.V. All rights reserved.
Keywords: A-Opioid receptor; Naltrexone; Receptor up-regulation; Constitutive internalization; Dynamin-2; G-protein receptor kinase-2 of the new receptors remains unclear. A concurrent changein receptor density and gene expression (i.e., steady state Treatment with opioid receptor agonists or antagonists mRNA level) has not been reliably observed following can regulate opioid receptor density as well as the potency antagonist treatment (Duttaroy et al., 1999; Castelli et al., of opioid agonists. For example, chronic in vivo opioid 1997; Jenab et al., 1995; Unterwald et al., 1995; however, receptor antagonist (e.g., naltrexone) treatment has been see Brodsky et al., 1995). The increase in receptor density shown to produce a robust increase in the density of A-, y- without a change in receptor mRNA levels in vivo suggests and n-opioid receptors (i.e., up-regulation) and a parallel that increases in receptor density may not require the syn- increase in the potency of opioid agonists (i.e., functional thesis of the new receptors. These results are consistent with supersensitivity) (Zukin et al., 1982, 1984; Yoburn et al., findings that an antisense oligodeoxynucleotide directed at 1986, 1995). Prolonged exposure (7 – 8 days) to an opioid the A-opioid receptor does not interfere with opioid receptor receptor antagonist is required to produce receptor up- antagonist-induced A-opioid receptor up-regulation in vivo regulation (Yoburn et al., 1986, 1989a,b), and functional (Shah et al., 1997). Similarly, an in vitro study showed that supersensitivity is proportional to the degree of up-regula- antagonist-induced up-regulation was not blocked by inhib- tion (Paronis and Holtzman, 1991; Yoburn et al., 1986, iting protein synthesis with cycloheximide (Tempel et al., 1986). Taken together, these data suggest that it is possible While the density of opioid receptors reliably increases that the opioid receptor antagonist-induced up-regulation after chronic opioid receptor antagonist treatment, the origin may be related to activation of a pre-existing pool of‘‘cryptic’’ receptors (Candido et al., 1992; Chan et al.,1995; Moudy et al., 1985; Unterwald et al., 1998) or a decrease in receptor degradation (Belcheva et al., 1991).
Corresponding author. Tel.: +1-718-990-1623; fax: +1-718-990-6036.
E-mail address: [email protected] (B.C. Yoburn).
Early preliminary data also support opioid receptor antago- 0014-2999/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 1 4 - 2 9 9 9 ( 0 2 ) 0 1 8 2 3 - X M. Patel et al. / European Journal of Pharmacology 446 (2002) 37–42 nist-induced changes in constitutive receptor cycling (Evans then the incubation was terminated by filtration of samples et al., 1997). More recent studies with A-opioid receptor over GF/B glass fiber filters (Brandel, Gaithersburg, MD).
splice variants have supported these suggestions that antag- Filters were washed three times with cold phosphate buffer, onist treatment interferes with constitutive internalization of transferred to vials containing scintillation cocktail and then A-opioid receptor (Koch et al., 2001).
counted. Counts per minute (CPMs) were converted to G-protein receptor kinase-2 (GRK) and dynamin have disintegration per minute (DPMs) using the external standard been shown to play an important role in opioid agonist- method. Protein was assayed using the Bradford method induced receptor internalization (Whistler and von Zastrow, (Bradford, 1976) with reagent purchased from Bio-Rad 1998; Zhang et al., 1998). These data raise the possibility that opioid receptor antagonist-induced up-regulation maybe associated with alterations in the abundance or activity of GRK and dynamin. Therefore, in the present study, weexamined GRK-2 and dynamin-2 abundance following Mice (n=12/treatment) were sacrificed, spinal cords were chronic opioid receptor antagonist treatment. We focused removed and homogenized (Brinkman Polytron Homoge- on the regulation of A-opioid receptors in spinal cord, since nizer, 20,000 rpm, 30 s) in 2 ml lysis buffer [2% sodium studies demonstrate opioid receptor antagonist-induced A- dodecyl sulfate (SDS), 1 mM sodium orthovanadate, 12.5 opioid receptor up-regulation in spinal cord and there is mM Tris (pH 7.4)], boiled for 5 min and centrifuged at substantial support for spinal mediation of opioid effects 10,000 rpm for 10 min. The supernatant was removed for analysis and protein concentration was determined (Brad-ford, 1976). An aliquot (four spinal cords/tube) of thesample (8 Al, 0.6 – 12 Ag protein) was loaded on polyacry- lamide gels (Pager Gels 10% Tris – Glycine, BiowhittakerMolecular Applications, Rockland, ME) and subject to electrophoresis at 150 V for 60 min. Transfer of protein toPVDF membrane (Bio-Rad, Hercules, CA) was carried out Male, Swiss – Webster mice (24 – 28 g) from Taconic at 100 V for 75 min. Nonspecific binding sites on the Farms (Germantown, NY) were used in all experiments.
membrane were blocked by incubation (1 h, 24 jC) in The animals were housed 10 per cage for at least 24 h prior blocking buffer (1.5% BSA in Tris-buffered saline with to experimentation with free access to food and water. Mice Tween-20 (TBST): (25 mM Tris, 150 mM NaCl, 0.05% Tween-20) for colorimetric assays; or Aurorak BlockingBuffer (ICN Biomedicals, Costa Mesa, CA) for chemilumi- nescence assays), followed by incubation (1 h, 24 jC) withprimary antibody in appropriate blocking buffer [rabbit Mice were implanted subcutaneously at the nape of the polyclonal immunoglobulin G (IgG) for GRK-2 (1:200); neck with an inert placebo pellet or a 15 mg naltrexone goat polyclonal IgG for dynamin-2 (1:300); goat polyclonal pellet for 8 days while mice were lightly anesthetized with IgG for actin (1:300)] (Santa Cruz Biotechnology, Santa halothane/oxygen (4:96). At the end of 8 days, mice were Cruz, CA). Membranes were washed twice with blocking killed and spinal cords were collected for receptor binding buffer and then incubated (1 h, 24 jC) with secondary antibody in TBST or in blocking buffer [anti-rabbit IgG-APfor GRK-2 (1:5000); anti-goat IgG-AP for dynamin-2 (1:5000); anti-goat IgG-AP for actin (1:5000)] (Santa CruzBiotechnology). Membranes for colorimetric assays were Binding was performed as described previously (Yoburn washed twice with TBST buffer followed by two quick et al., 1995). Briefly, mice (n = 12/treatment) were sacrificed, rinses with Tris-buffered saline (25 mM Tris, 150 mM spinal cords were removed and homogenized in 80 volumes NaCl). Membranes for chemiluminescence assays were of cold 50 mM Tris buffer (pH—7.4). Homogenates were washed thrice with blocking buffer followed by two quick centrifuged at 15,000 rpm for 15 min, supernatants were rinses with assay buffer (200 mM Tris – HCl, pH 9.8; 10 discarded and pellets were resuspended and incubated (30 mM MgCl2). Target proteins were visualized using the min, 25 jC) in Tris buffer. Homogenates were centrifuged Starlight alkaline phosphatase Chemiluminescent Assay again and the pellets were finally resuspended in 50 mM (Aurora, ICN Biomedicals) according to the manufacturer’s phosphate buffer (pH 7.2). An aliquot of homogenate was instructions. Target proteins were visualized for the colori- assayed in triplicate in tubes containing 0.03 – 5 nM [3H] [D- metric assay using 5-bromo-4-chloro-3-indolyl-phosphate Ala2, N-MePhe4, Gly5-ol] enkephalin (DAMGO) (A-opioid (BCIP) and nitroblue tetrazolium color development sub- receptor ligand, New England Nuclear, Boston, MA). Non- strate (Promega, Madison, WI) according to the manufac- specific binding was determined in the presence of 1 AM turer’s instructions. Chemiluminescence and colorimetric levorphanol. Tubes were incubated for 90 min at 25 jC and results were captured using a FluorChem ver. 2.0 Imaging M. Patel et al. / European Journal of Pharmacology 446 (2002) 37–42 Differences between means were assessed using Stu- dent’s t-test (P<0.05). BMAX and KD were estimated fromsaturation studies using nonlinear regression (Prism ver.
3.02, Graphpad Software, San Diego, CA). Binding datawas best fit by a one-site model.
3.1. Effects on GRK-2 and dynamin-2 abundance Changes in the abundance of GRK-2 and dynamin-2 in spinal cord were determined following chronic naltrexonetreatment. Western blotting of SDS-solubilized spinal cord Fig. 1. Typical standard curves for GRK-2 (A) and dynamin-2 (B).
Increasing amounts of total protein were loaded on polyacrylamide gels.
System (Alpha Innotech, San Leandro, CA) and imageswere then quantitated for optical density using GelPro ver.
3.0 (Media Cybernetics, Silver Spring, MD). All assays forGRK-2 and actin used chemiluminescence, while twoassays for dynamin-2 employed the colorimetric method.
A standard curve (minimum 4 points) using increasing amounts (0.2 – 32 Ag) of untreated spinal cord sample wasincluded in each gel assay (dynamin-2, GRK-2, actin). Thisallowed conversion of optical density into valid estimates ofpercent changes in protein. All data are expressed in theseprotein equivalents. The standard curves were linear from0.2 to 32 Ag of total protein, which included the range ofoptical densities employed for the unknowns. Typicalchemiluminescent standard curves are shown in Fig. 1A(GRK-2: mean r2=0.87F0.05; six assays); Fig. 1B (dyna-min-2: mean r2=0.89F0.05; five assays) (actin: meanr2=0.98F0.01; five assays, data not shown). Results fromcolorimetric standard curves were similar.
Naltrexone (30 mg) and corresponding placebo pellets Fig. 2. The effect of chronic naltrexone on GRK-2 and dynamin-2 were obtained from Research Triangle Institute (Research abundance in spinal cord. Naltrexone (15 mg) and placebo pellets were Triangle Park, North Carolina) through the Research Tech- implanted subcutaneously for 8 days. At the end of the treatment, spinal nology Branch of the National Institute on Drug Abuse. The cords were removed and levels of GRK-2 and dynamin-2 were determinedusing Western blotting. Insets are representative blots for GRK-2 (A) and pellets were cut in half (yielding 15 mg implants), weighed dynamin-2 (B) (C=control; N=naltrexone). Data are meanFS.E.M. from and then implanted. The pellets were wrapped in nylon five or six independent experiments. *Significantly different from placebo mesh before subcutaneous implantation.
M. Patel et al. / European Journal of Pharmacology 446 (2002) 37–42 cord (Shah et al., 1997). Although several mechanisms havebeen proposed, to date, the basis of this effect has remainedunknown. Among the possible substrates for opioid receptorantagonist-induced up-regulation are changes in transcrip-tion, recruitment of ‘‘cryptic’’ receptors and reduction inreceptor degradation in association with changes in receptorcycling.
Regulatory mechanisms involving increased transcrip- tion are unlikely, since antagonist-induced A-opioid receptorup-regulation is not associated with changes in A-opioidreceptor mRNA levels (Duttaroy et al., 1999; Castelli et al.,1997; Jenab et al., 1995; Unterwald et al., 1995; however,see Brodsky et al., 1995). Similarly, up-regulation is prob-ably independent of translational changes since an antisenseoligodeoxynucleotide to the A-opioid receptor does notblock opioid receptor antagonist-induced A-opioid receptorup-regulation or supersensitivity in vivo (Shah et al., 1997);and the protein synthesis inhibitor, cycloheximide, does not Fig. 3. Scatchard plot of specific [3H] DAMGO binding in spinal cord interfere with receptor up-regulation in tissue culture (Tem- homogenate from mice (n = 12) chronically treated subcutaneously with pel et al., 1986). Taken together, these data do not support a naltrexone or placebo pellet for 8 days. The KD and BMAX for placebo: 0.92 role for gene expression or protein synthesis in antagonist- nM, 247F7 fmol/mg protein, and for naltrexone: 1.38 nM, 605F4 fmol/mg induced opioid receptor up-regulation.
protein. The results are representative data from a single experiment.
Activation of a pre-existing pool of ‘‘cryptic’’ receptors (Candido et al., 1992; Chan et al., 1995; Moudy et al., preparation indicated that chronic naltrexone treatment sig- 1985; Unterwald et al., 1998) has also been suggested to nificantly (P<0.05) decreased the abundance of GRK-2 account for opioid receptor antagonist-induced up-regula- (36%) and dynamin-2 (30%) in spinal cord as compared tion. The cryptic receptor hypothesis was addressed using to placebo group (Fig. 2A and B). A representative blot for combined receptor autoradiography and immunohistochem- each GRK-2 and dynamin-2 is shown in the insets in Fig.
istry in brain slices (Unterwald et al., 1998). The results are 2A and B. Immunoblot assay indicated no significant generally consistent with a cryptic receptor hypothesis, changes in actin abundance (data not shown).
since antagonist treatment induced greater magnitudeincreases in radioligand binding in more brain areas than in immunoreactive protein. However, there were significantincreases in immunoreactive protein, and all changes in In saturation binding, chronic naltrexone treatment pro- binding could not be accounted for by unmasking of cryptic duced significant (P<0.05) A-opioid receptor up-regulation (BMAX=247F7 and 606F4, control and naltrexone, respec- A preliminary study of opioid receptor antagonist- tively) without altering affinity (KD=0.92 and 1.38, control induced up-regulation in cells expressing A-opioid recep- and naltrexone, respectively). A typical Scatchard plot is tors raised the possibility of a reduction in constitutive receptor cycling (Evans et al., 1997). More recently,studies of A-opioid receptor splice variants found thatnaloxone may act to stabilize membrane opioid receptors by interfering with constitutive internalization (Koch et al.,2001). If opioid receptor antagonist-induced receptor up- In the present study, naltrexone produced a significant regulation involves a change in constitutive internalization, increase in A-opioid receptor density in spinal cord and then it might be expected that trafficking proteins would decreased the abundance of GRK-2 and dynamin-2. These be regulated by chronic antagonist treatment. Both GRK data raise the possibility that down-regulation of cellular and dynamin have been shown to be involved in internal- trafficking proteins may mediate opioid receptor antagonist- ization of the constitutively active mutants of many G- protein-coupled receptors in vitro (Mhaouty-Kodja et al., Opioid receptor antagonist-induced receptor up-regula- 1999; Li et al., 2001; Pei et al., 1994). Furthermore, tion is a robust phenomenon in the whole animal and in cell numerous studies have shown that both GRK and dynamin culture (Zukin et al., 1982, 1984; Yoburn et al., 1985, 1986; play an important role in agonist-mediated receptor inter- Yoburn, 1988; Unterwald et al., 1995). Up-regulation of A- nalization of G-protein-coupled receptors including A- opioid receptors has been demonstrated in many CNS opioid receptors (Whistler and von Zastrow, 1998; Zhang regions (e.g., Unterwald et al., 1995), including the spinal M. Patel et al. / European Journal of Pharmacology 446 (2002) 37–42 Our data demonstrating down-regulation of GRK-2 and 1995. The effect of the irreversible mu-opioid receptor antagonist clo- dynamin-2 by naltrexone support the role of inhibition of cinnamox on morphine potency, receptor binding and receptor mRNA.
Eur. J. Pharmacol. 287, 135 – 143.
constitutive cycling in up-regulation. While opioid receptor Duttaroy, A., Shen, J., Shah, S., Chen, B., Sehba, F., Carroll, J., Yoburn, antagonist treatment may act to alter GRK-2 and dynamin-2 B.C., 1999. Opioid receptor upregulation in mu-opioid receptor defi- abundance, it is unknown if this is mediated by a change in cient CXBK and outbred Swiss Webster mice. Life Sci. 65, 113 – 123.
transcription or degradation rate of these proteins. Further- Evans, C.J., Keith Jr., D., Zaki, P., von Zastrow, M., 1997. Up-regulation of more, we recognize that several different mechanisms might surface A-opioid receptors by antagonists. Soc. Neurosci. 23, 1770.
Jenab, S., Kest, B., Inturrisi, C.E., 1995. Assessment of delta opioid anti- contribute simultaneously to opioid receptor antagonist- nociception and receptor mRNA levels in mouse after chronic naltrex- induced receptor up-regulation in vivo. For example, acti- one treatment. Brain Res. 691, 69 – 75.
vation of a pre-existing pool of cryptic receptors (Candido et Koch, T., Schulz, S., Pfeiffer, M., Klutzny, M., Schroder, H., Kahl, E., al., 1992; Chan et al., 1995; Moudy et al., 1985; Unterwald Hollt, V., 2001. C-terminal splice variants of the mouse mu-opioid et al., 1998) or a decrease in receptor degradation (Belcheva receptor differ in morphine-induced internalization and receptor resen-sitization. J. Biol. Chem. 276, 31408 – 31414.
et al., 1991) are also likely candidates that would support Li, J., Chen, C., Huang, P., Liu-Chen, L.Y., 2001. Inverse agonist up-reg- ulates the constitutively active D3.49 (164) Q mutant of the rat mu- In summary, down-regulation of GRK-2 and dynamin-2 opioid receptor by stabilizing the structure and blocking constitutive abundance as demonstrated in the present study may help internalization and down-regulation. Mol. Pharmacol. 60, 1064 – 1075.
mediate opioid receptor antagonist-induced up-regulation.
Mhaouty-Kodja, S., Barak, L.S., Scheer, A., Abuin, L., Diviani, D., Caron, M.G., Cotecchia, S., 1999. Constitutively active alpha-1b adrenergic In the future, it would be of interest to examine the effect of receptor mutants display different phosphorylation and internalization regulating GRK-2 and dynamin-2 either alone or together features. Mol. Pharmacol. 55, 339 – 347.
on A-opioid receptor density and functional potency of Moudy, A.M., Spain, J.W., Coscia, C.J., 1985. Differential up-regulation of microsomal and synaptic membrane mu opioid receptors. Biochem.
Biophys. Res. Commun. 132, 735 – 741.
Paronis, C.A., Holtzman, S.G., 1991. Increased analgesic potency of mu agonists after continuous naloxone infusion in rats. J. Pharmacol. Exp.
Pei, G., Samama, P., Lohse, M., Wang, M., Codina, J., Lefkowitz, R.J., 1994.
A constitutively active mutant beta 2-adrenergic receptor is constitu- These data were presented at the annual meeting of the tively desensitized and phosphorylated. Proc. Natl. Acad. Sci. U. S. A.
Mid-Atlantic Pharmacology Society, Philadelphia, PA, September 2001. These data represent a portion of a thesis Shah, S., Duttaroy, A., Chen, B.T., Carroll, J., Yoburn, B.C., 1997. The presented by Minesh Patel to the faculty of the College of effect of mu-opioid receptor antisense on morphine potency and antag-onist-induced supersensitivity and receptor upregulation. Brain Res.
Pharmacy, St. John’s University, in partial fulfillment of the requirements for the MS degree in Pharmaceutical Sciences.
Tempel, A., Crain, S.M., Peterson, E.R., Simon, E.J., Zukin, R.S., 1986.
Our thanks to Dr. Tom Turnock for 26 years of solace. This Antagonist-induced opiate receptor upregulation in cultures of fetal project was supported by the National Institutes of Health, mouse spinal cord-ganglion explants. Brain Res. 390, 287 – 291.
Unterwald, E.M., Rubenfeld, J.M., Imai, Y., Wang, J.B., Uhl, G.R., Kreek, M.J., 1995. Chronic opioid antagonist administration upregulates muopioid receptor binding without altering mu opioid receptor mRNAlevels. Brain Res. Mol. Brain Res. 33, 351 – 355.
Unterwald, E.M., Anton, B., To, T., Lam, H., Evans, C.J., 1998. Quantita- tive immunolocalization of mu opioid receptors: regulation by naltrex-one. Neuroscience 85, 897 – 905.
Belcheva, M.M., Barg, J., McHale, R.J., Gao, X.M., Chuang, D.M., Coscia, Whistler, J.L., von Zastrow, M., 1998. Morphine-activated opioid receptors C.J., 1991. Up-regulation of delta opioid receptors in neuroblastoma elude desensitization by beta-arrestin. Proc. Natl. Acad. Sci. U. S. A.
hybrid cells: evidence for differences in the mechanisms of action of sodium butyrate and naltrexone. J. Pharmacol. Exp. Ther. 259, 302 – Yoburn, B.C., 1988. Opioid antagonist-induced upregulation and functional supersensitivity. Rev. Clin. Basic Pharmacol. 7, 109 – 128.
Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of Yoburn, B.C., Goodman, R.R., Cohen, A.H., Pasternak, G.W., Inturrisi, microgram quantities of protein utilizing the principle of protein-dye C.E., 1985. Increased analgesic potency of morphine and increased binding. Analytical Biochem. 72, 248 – 254.
brain opioid binding sites in the rat following chronic naltrexone treat- Brodsky, M., Elliott, K., Hynansky, A., Inturrisi, C.E., 1995. CNS levels of mu-opioid receptor (MOR-1) mRNA during chronic treatment with Yoburn, B.C., Nunes, F.A., Adler, B., Pasternak, G.W., Inturrisi, C.E., morphine or naltrexone. Brain Res. Bull. 38, 135 – 141.
1986. Pharmacodynamic supersensitivity and opioid receptor upregula- Candido, J., Lutfy, K., Billings, B., Sierra, V., Duttaroy, A., Inturrisi, C.E., tion in the mouse. J. Pharmacol. Exp. Ther. 239, 132 – 135.
Yoburn, B.C., 1992. Effect of adrenal and sex hormones on opioid Yoburn, B.C., Paul, D., Azimuddin, S., Lutfy, K., Sierra, V., 1989a. Chronic analgesia and opioid receptor regulation. Pharmacol. Biochem. Behav.
opioid antagonist treatment increases mu and delta receptor mediated spinal opioid analgesia. Brain Res. 485, 176 – 178.
Castelli, M.P., Melis, M., Mameli, M., Fadda, P., Diaz, G., Gessa, G.L., Yoburn, B.C., Sierra, V., Lutfy, K., 1989b. Chronic opioid antagonist treat- 1997. Chronic morphine and naltrexone fail to modify mu-opioid re- ment: assessment of receptor upregulation. Eur. J. Pharmacol. 170, ceptor mRNA levels in the rat brain. Brain Res. Mol. Brain Res. 45, Yoburn, B.C., Shah, S., Chan, K., Duttaroy, A., Davis, T., 1995. Super- Chan, K., Brodsky, M., Davis, T., Franklin, S., Inturrisi, C.E., Yoburn, B.C., sensitivity to opioid analgesics following chronic opioid antagonist M. Patel et al. / European Journal of Pharmacology 446 (2002) 37–42 treatment: relationship to receptor selectivity. Pharmacol. Biochem. Be- Zukin, R.S., Sugarman, J.R., Fitz-Syage, M.L., Gardner, E.L., Zukin, S.R., Gintzler, A.R., 1982. Naltrexone-induced opiate receptor supersensitiv- Zhang, J., Ferguson, S.G., Barak, L.S., Bodduluri, S.R., Laporte, S.A., Law, P.Y., Caron, M.G., 1998. Role for G-protein-coupled receptor kinase in Zukin, R.S., Tempel, A., Gardner, E.L., 1984. Opiate receptor upregulation agonist-specific regulation of A-opioid receptor responsiveness. Proc.
and functional supersensitivity. NIDA Res. Monogr. 54, 146 – 161.
Natl. Acad. Sci. U. S. A. 95, 7157 – 7162.

Source: http://www.yoburnlab.org/Documents/patel%20ejp%202002.pdf