Alcohol Inhibition of NK Cell Function Opioid Receptors in Splenocytes to Prevent Δ and Μ Feedback Interaction Between Opioid

Home , Naltriben

Opioid Antagonist Naltrexone Disrupts Feedback Interaction between µ and δ Opioid Receptors in Splenocytes to Prevent Alcohol Inhibition of NK Cell Function This information is current as of September 28, 2021. Nadka I. Boyadjieva, Kirti Chaturvedi, Michael M. Poplawski and Dipak K. Sarkar J Immunol 2004; 173:42-49; ; doi: 10.4049/jimmunol.173.1.42 http://www.jimmunol.org/content/173/1/42 Downloaded from

References This article cites 48 articles, 17 of which you can access for free at:

http://www.jimmunol.org/content/173/1/42.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 28, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Opioid Antagonist Naltrexone Disrupts Feedback Interaction between ␮ and ␦ Opioid Receptors in Splenocytes to Prevent Alcohol Inhibition of NK Cell Function1

Nadka I. Boyadjieva, Kirti Chaturvedi, Michael M. Poplawski, and Dipak K. Sarkar2

Naltrexone, an opioid antagonist, has been used in clinical trials to treat alcoholism. As the opioid peptides ␤-endorphin and enkephalin increase splenic NK cell function in laboratory animals, it is anticipated that naltrexone treatment will cause immu- nosuppression. However, we report in this study that chronic naltrexone administration in laboratory rats increases the cytolytic activity of NK cells. It also prevents alcohol’s suppressive effect on these cells. We identiﬁed that, in the splenocytes, ␦ opioid receptor expression is tightly controlled by negative feedback regulation of ␮ opioid receptors. Naltrexone disrupts this feedback control by reducing ␮ opioid receptor function, thereby up-regulating ␦ opioid receptor binding, which results in an enhanced NK Downloaded from cell cytolytic response to ␦ opioid receptor ligands. We conclude that naltrexone, which has been shown to be a promising agent for the clinical management of alcoholism, may have potential use in the treatment of immune deﬁciency in alcoholic and nonalcoholic patients. The Journal of Immunology, 2004, 173: 42–49.

atural killer cells are lymphocytes that participate in the cently, the opioid antagonist naltrexone was used in clinical trials body’s immune defense system by killing bacteria, vi- to treat alcoholism (21). Because naltrexone is an opioid receptor http://www.jimmunol.org/ N ruses, and cancer cells. NK cells quickly respond to im- antagonist, the concern was raised as to whether this drug, by mune activation signals, and activated NK cells circulate through blocking the functions of endogenous opioid ligands, increases the blood and lymph and accumulate at the sites of injuries, infections, severity of immune dysfunction in alcoholic patients. Therefore, and tumors (1–3). A reduced NK cell cytolytic activity and in- we investigated the effect of naltrexone on NK cell function using creased incidence of certain forms of cancers and infections have the Fischer-344 rat as an animal model. been documented in chronic alcoholics (3–10). NK cell cytolytic activity is increased by the lymphokine IFN-␥, Materials and Methods which has a number of opioid-like effects (11). Endogenous opioid Animals and treatments

peptides, such as enkephalins and endorphins, which control the by guest on September 28, 2021 Male Fischer-344 rats, 150–175 g body weight, were maintained in a con- functions of the neuroendocrine system, have been shown to stim- trolled environment (4), given free choice of water, and fed a liquid diet ulate NK cell cytolytic activity. These opioid peptides regulate NK containing alcohol or pair-fed an isocaloric liquid diet (Bio-Serv, French- cell cytolytic activity by acting directly on the spleen and altering town, NJ). Because the pair-fed group was calorie-matched to the alcohol- the neuroendocrine-immune system function (4, 12–15). In the fed group, they had some food restriction, but had no change in body CNS, ␤-endorphin and enkephalins are known to bind to classical weight for a period of 2 wk. Furthermore, we showed that pair-fed animals ␦ ␮ and ad libitum-fed rodent chow animals had similar NK cell cytolytic ac- and opioid receptors (16). Opioid receptors have been shown tivity, proliferative action to mitogens, and cytokine production (4, 19). to be present in spleen cells (15, 17). Cultured human lymphocyte After the first week of adjustment to the diet, each animal was s.c. im- lineages have been shown to express functional enkephalin recep- planted, for a period of 2 wk, with either a naltrexone or a naltrexone- tors using highly ␦ opioid receptor-selective enkephalin analogues placebo pellet (50 mg/pellet, 21-day release; Innovative Research of Amer- ␦ ica, Sarasota, FL). In vitro experiments were conducted using rats ad (18). opioid receptor-specific agonists have been shown to stim- libitum-fed rodent chow. Animal surgery and care were performed in ac- ulate NK cell cytolytic activity, and a ␦-specific antagonist blocks cordance with institutional guidelines and complied with National Insti- ␤-endorphin-stimulated NK cell function (4, 19). tutes of Health policy. The endogenous opioid peptides and NK cells respond to alco- NK cell cytolytic activity hol in both animals and humans (4, 12, 19, 20). Chronic alcohol exposure reduces ␤-endorphin levels in the hypothalamus and in- The cytolytic activity of NK cells was determined with a standard 4-h 51Cr 3 hibits the opioid’s ability to stimulate NK cell function (4). Re- assay using yeast artificial chromosome 1 (YAC-1) lymphoma cells as targets (4). Spleens of experimental animals were used to isolate splenocytes that were then cultured in the presence or the absence of rIL-2 (500 U/ml: Sigma-Aldrich, St. Louis, MO) for 18 h at 37°C with or without Program of Endocrinology, Center of Alcohol Studies, Department of Animal Sci- various opioid peptides (methionine-enkephalin, leucine-enkephalin, ences, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901 ␤-endorphin, and dynorphin A [1-13]; obtained from Sigma-Aldrich), and Received for publication September 26, 2003. Accepted for publication April opioid agonists (␦ opioid receptor agonist (ϩ)-4-[(␣R)-␣-((2S,5R)-4-allyl-2,5- 20, 2004. dimethyl-1-piperazinyl)-3-methoxybenzyl]-N,N-diethyl benzamide (SNC80) ␬ Ϫ ␮ The costs of publication of this article were defrayed in part by the payment of page (22), opioid receptor agonist ( )-U50488 (23), opioid receptor agonist, charges. This article must therefore be hereby marked advertisement in accordance and Tyr-D-Ala-Gly-Nme-Phe-Gly-ol (DAMGO; obtained from Tocris with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by National Institutes of Health Grants AA00220 and 3 Abbreviations used in this paper: CTAP, cyclic peptidergic analog of somatostatin; AA12642. 2 5 DADLE, [D-Ala ,D-Leu ]enkephalin; DAMGO, Tyr-D-Ala-Gly-Nme-Phe-Gly-ol; 2 Address correspondence and reprint requests to Dr. Dipak K. Sarkar, Endocrinology Bmax, maximum binding; SNC80, (ϩ)-4-[(␣R)-␣-((2S,5R)-4-allyl-2,5-dimethyl-1- Program, Rutgers, The State University of New Jersey, 84 Lipman Drive, New Bruns- piperazinyl)-3-methoxybenzyl]-N,N-diethyl benzamide; YAC, yeast artificial wick, NJ 08901. E-mail address: [email protected] chromosome.

Cookson, Ellisville, MO) (24)). Erythrocytes were removed by a 5-s The liquid diet regimen used for alcohol administration did not hypotonic shock with sterile distilled water. The percentages of cytolytic produce a nutritional deficit for a period of 2 wk (4). activity at E:T cell ratios of 200:1, 100:1, 50:1, and 25:1 were converted to 6 lytic units per 10 effector cells according to Pross et al. (25). Each assay Naltrexone’s effect on NK cell cytolytic activity was conducted in quadruplicate, and the cytolytic activity of the individual E:T cell ratios was used in the conversion. We have previously shown that alcohol treatment for a period of 2–4 wk significantly decreased NK cell cytolytic activity com- Opioid receptor binding assays pared with that in pair-fed or ad libitum-fed control animals (4). In Freshly prepared splenocytes (1–4 ϫ 106) were incubated with various this study we also found that in rats given an alcohol-containing concentrations of [3H]naltrindole or [3H]DAMGO (PerkinElmer, Boston, liquid diet for 2 wk, the NK cytolytic was significantly reduced 2ϩ 2ϩ MA) in a 96-well plate for 4.5 h at 20°CinCa - and Mg -deficient compared with the activity in pair-fed control animals (Fig. 1A). HBSS medium in the presence of protease inhibitor mixture (both from Sigma-Aldrich) to determine ␮ and ␦ opioid receptor binding (18). Unla- Rats treated with alcohol also showed reduced IL-2 ability to in- beled naltrindole (50 ␮M; Sigma-Aldrich) and cyclazocine (50 ␮M; Sig- crease NK cytolytic activity (Fig. 1B). Alcohol-fed animals treated ma-Aldrich) were used to assess nonspecific binding for [3H]naltrindole with chronic naltrexone had a marked increase in both basal and 3 and [ H]DAMGO, respectively. Incubation was terminated by harvesting cytokine-stimulated NK cytolytic activity. Chronic treatments with cell contents to Whatman glass-fiber filters (Clifton, NJ), which were then air-dried and counted in a beta counter. Receptor binding data were ana- naltrexone also improved NK cell function in control animals. lyzed by nonlinear regression analysis using PRISM 3.0 (GraphPad, San Diego, CA). For in vitro experiments, splenocytes were treated with alco- Naltrexone’s ability to alter NK cell response to opioid peptides hol (50 mM), naltrexone (10 ng/ml), naltrexone (10 ng/ml) and alcohol (50 In the brain, the endogenous opioid peptides, ␤-endorphin and en- mM), cyclic peptidergic analog of somatostatin (CTAP; 10 ng/ml), or nal- ␦ ␮ Downloaded from trindole (10 ng/ml) for 3 days in RPMI 1640 medium containing 10% FBS kephalin, bind to classical and opioid receptors, whereas (all from Sigma-Aldrich). Medium was changed at 12-h intervals, and the dynorphin binds to ␬ opioid receptors (16). These opioid receptors medium alcohol levels were maintained between 30 and 50 mM, as deter- are also present in spleens (15, 27). Naltrexone antagonizes the mined by a commercially available enzymatic assay (Sigma-Aldrich). Cells ligand binding primarily to ␮ opioid receptors, but also to ␦ and ␬ were washed three times to remove bound naltrexone and then assayed for opioid receptors, in brain tissues (28). The question arose of [3H]naltrindole or [3H]DAMGO binding. The specific binding at each concentration was calculated by subtracting nonspecific binding from total whether, after chronic administration, naltrexone’s ability to block http://www.jimmunol.org/ binding and was used for determination of maximum ligand binding (Bmax) opioid ligand binding in the spleen is altered. Dose-response stud- 3 and binding affinity (1/Kd). For [ H]naltrindole binding, the ratios of spe- ies of the effects of opioid peptides on NK cell function of control cific to nonspecific binding of a representative curve at 1.37, 2.74, 5.48, rats’ splenocytes revealed that ␤-endorphin and methionine-en- 10.96, and 44.0 nM concentrations were 2.30, 1.14, 0.65, 0.63, and 0.30, respectively. For [3H]DAMGO binding, the ratios of specific to nonspecific kephalin markedly stimulated, leucine-enkephalin moderately binding of a representative curve at 2.4, 4.8, 9.6, 22.0, and 44.0 nM con- stimulated, and dynorphin moderately inhibited NK cell cytolytic centrations were 0.96, 1.0, 0.79, 0.68, and 0.51, respectively. To avoid any influence of ligand uptake and internalization on the ligand binding assay, several controls were used. First, with each dose of radioactive ligand we included a nonspecific binding control group, which addressed the problem

of nonspecific or internalized nonreceptor-bound ligand binding. Further- by guest on September 28, 2021 more, we calculated specific binding at each concentration by subtracting nonspecific binding from total binding. Secondly, in the receptor assay, after transferring cell contents onto glass-fiber paper, we washed the filter paper to eliminate any free radioligand present in the cultures. For the in vitro assay, the binding values were not normalized using the total protein levels, but, instead, were normalized using the total cell number. For displacement studies, various concentrations (10Ϫ4–10Ϫ9 M) of opioid ligands were used to displace [3H]DAMGO (6 nM) or [3H]naltrindole (8 nM) binding. The concentrations used for [3H]naltrindole and 3 [ H]DAMGO were similar to their Kd values determined from the saturation curves (see above). The various ligands used were CTAP, SNC80, 2 5 naltrexone, and [D-Ala ,D-Leu ]enkephalin acetate salt (DADLE; Sigma-

Aldrich). The EC50 value of each ligand was determined by nonlinear regression analysis (PRISM Graph Software 4.0; GraphPad). The Ki (dis- sociation constant) values were calculated from the EC50 values by Cheng Prusoff equation using PRISM Graph Software 4.0.

Statistical analyses The mean and SE of the data are presented in the text and figures. The t test was used to compare differences between two experimental groups. Mul- tiple group data were analyzed using one- or two-way ANOVA as appro- priate. Post-hoc multiple comparisons were performed using the Student- Newman-Keuls test for one-way ANOVA and a Bonferroni corrected t test Ͻ for two-way ANOVA. A value of p 0.05 was considered significant. FIGURE 1. Effects of naltrexone on NK cell functions. A, The cytolytic activity of the NK cells, as measured by lytic unit, was determined in Results splenocytes obtained from spleens of naltrexone (NAL)- or placebo (PL)- Alcohol diet consumption and blood alcohol levels treated rats fed an alcohol-containing liquid diet (AF) or an isocaloric liq- Ͻ Ͻ As has been shown previously (26), the alcohol-containing liquid uid diet (PF). a, p 0.01, significantly different from the PL group; b, p 0.05, significantly different from the PF group with similar treatment (n ϭ diet consumption between alcohol-fed and naltrexone-treated al- ϳ 6–8). B, Splenocytes of all four groups were cultured in the absence cohol-fed animals was similar ( 70–95 ml/day). As expected (CONT) or the presence of IL-2, and cytolytic activity was measured. Note from our previous findings (4), the alcohol treatment regimen sig- that six of eight subjects in the NAL/PF-treated group had higher lytic units nificantly elevated the level of blood alcohol (0.13 Ϯ 0.01 g/dl; than those in the PL/PF-treated group. a, p Ͻ 0.001 significantly different n ϭ 8). The blood alcohol concentrations of rats treated with al- from the CONT/PF group; b, p Ͻ 0.01, significantly different from the PL cohol and naltrexone were not different (0.12 Ϯ 0.01 g/dl; n ϭ 7). group with similar treatment. 44 OPIOID RECEPTOR CONTROL OF NK CELL CYTOLYTIC FUNCTION

FIGURE 2. Effects of naltrexone on NK cell response to opioid peptides. Spleens from all four groups, as described in Fig. 1, were used to determine the in vitro effects of ␤-endorphin (A), methionine-enkephalin (B), leucine-enkephalin (C), and dynorphin A (D). Data are the mean Ϯ SE of five to seven animals. a, p Ͻ 0.05, significantly different from the zero opioid dose, but within the same treatment group; b, p Ͻ 0.05, significantly different from the rest of the groups with the same treatment; c, p Ͻ 0.05 significantly different from all but 10Ϫ6 M opioid dose with the same treatment; d, p Ͻ Downloaded from 0.05, significantly different from the similar dose of opioid in the PL group. http://www.jimmunol.org/ activity at only the higher dose (Fig. 2). The NK cell cytolytic Naltrexone’s ability to alter NK cell response to opioid receptor responses to all ligands were diminished in splenocytes obtained agonists from alcohol-fed animals. However, the NK cell responses to ␤-endorphin and met-enkephalin were markedly increased, and the To test the possibility that naltrexone treatment changed the activity of response to Leu-enkephalin was moderately increased in naltrex- specific types of opioid receptors, the effects of the specific opioid one-treated and naltrexone- and alcohol-treated animals. Naltrex- receptor agonists on NK cell cytolytic activity were compared among one treatment did not produce consistent effects on dynorphin- the four groups. The cytolytic activity of the control animals was by guest on September 28, 2021 modulated NK cell function. These data suggest that chronic stimulated by the ␦ agonist and the ␮ agonist, but was moderately naltrexone treatments altered NK cell responses to the opioid re- inhibited by the ␬ agonist (Fig. 3). The NK cell responses to these ceptor ligands ␤-endorphin and enkephalin. opioid receptor agonists were reduced in splenocytes of alcohol-fed

FIGURE 3. Effects of naltrexone on NK cell response to opioid receptor agonists. Spleens from all four groups, as described in Fig. 1, were used to determine the in vitro effects of the ␦ opioid receptor agonist SNC80 (A), the ␮ opioid receptor agonist DAMGO (B), and the ␬ opioid receptor agonist (Ϫ)-U50488 (C)on NK cell cytolytic activity. Data are the mean Ϯ SE of five to seven animals. a, p Ͻ 0.05, significantly different from the zero opioid dose, but within the same treatment group; b, p Ͻ 0.05, significantly different from the rest of the groups with the same treatment; c, p Ͻ 0.05 significantly different from all but the 10Ϫ6 or 10Ϫ4 M opioid doses within the same treatment groups; d, p Ͻ 0.05, significantly different from the similar dose of opioid in the PL group. The Journal of Immunology 45

FIGURE 4. Naltrexone modulations of opioid receptor binding in splenocytes: in vivo effects. Shown are representative saturation curves and the mean Ϯ ␦ 3 SE Bmax of opioid receptor-specific ligand [ H]naltrindole binding (A and B) and ␮ opioid receptor-specific ligand [3H]DAMGO binding (C and D)on splenocytes of all four groups described in Fig. 1: naltrexone (NAL)- or placebo (PL)-treated rats fed alcohol (AF) or pair-fed an isocaloric liquid diet (PF). Data are the mean Ϯ SE of three to five independent experiments. a, p Ͻ 0.05, significantly different from the PF/PL group; b, p Ͻ 0.01, significantly different from the AF/PL group. Downloaded from

animals. However, the NK cell responses of splenocytes from nal- NK cell response to the ␮ opioid ligand of the splenocytes, ex- ␮

trexone-treated animals and naltrexone- and alcohol-treated animals posed chronically to naltrexone, was caused by altered levels of http://www.jimmunol.org/ showed differential responses to opioid receptor agonists. The ␦ opi- opioid receptor binding. To test this possibility, we compared opioid agonist markedly increased NK cell activity in pair-fed, naltrex- oid receptor ligand binding in splenocytes of all four groups. The one-treated and naltrexone- and alcohol-treated animals; the ␮ opioid ␮ opioid receptor binding was reduced in splenocytes of naltrex- agonist increased NK cell cytotoxicity only in pair-fed animals, not in one-treated, naltrexone- and alcohol-treated, and alcohol-treated ␬ ␮ naltrexone-fed or naltrexone- and alcohol-fed animals. The opioid animals (Fig. 4B). Additionally, the Kd values of opioid receptor Ͻ agonist moderately inhibited it, but in naltrexone-treated animals only. binding were higher ( p 0.001) in these animals (Kd: control, These results suggest that naltrexone treatment increased the NK cy- 13.9 Ϯ 1.6; alcohol, 30.9 Ϯ 5.8; naltrexone, 39.7 Ϯ 0.8; naltrexone tolytic activity in both pair-fed and alcohol-fed animals. Furthermore, and alcohol, 24.5 Ϯ 2.3; n ϭ 3). These data suggest that chronic

the increased NK cell function caused by naltrexone may have been naltrexone down-regulated ␮ opioid receptor activity. by guest on September 28, 2021 primarily the result of increased sensitivity to ␦ opioid receptors in the ␮ ␦ spleen. Characterization of and receptor binding sites in splenocytes ␦ Naltrexone’s ability to alter opioid receptor binding to ligands To further characterize ␮ and ␦ opioid receptor binding in the in splenocytes spleen, we performed displacement studies to determine Ki (dis- To identify whether the increased ␦ opioid receptor response in sociation constant) values for various ␮ opioid receptor and ␦ opi- splenocytes of animals treated with naltrexone was caused by an oid receptor ligands. We used SNC80 (a high affinity ␦ opioid alteration in receptor binding and affinity, we compared ␦ opioid receptor agonist) (22), CTAP (a high affinity ␮ opioid receptor receptor ligand ([3H]naltrindole) binding in splenocytes of all four antagonist) (29), DADLE (an agonist that can bind to ␦ opioid groups (Fig. 4A). Splenocytes of alcohol-fed animals showed re- receptor with higher affinity than ␮ opioid receptor) (30), and nal- ␦ ␮ duced ligand binding (Bmax), but showed no changes in binding trexone (an antagonist with higher affinity for opioid receptor ␦ affinity (1/Kd), compared with control animals (Kd: alcohol-fed, than opioid receptor) (28). As shown in Table I and Fig. 5, the 11.7 Ϯ 0.1; control, 14.1 Ϯ 0.2; n ϭ 3). The ␦ ligand binding was ␦ receptor ligand SNC80 completely displaced [3H]naltrindole markedly increased in animals treated with naltrexone with or binding, whereas the ␮ opioid receptor ligand CTAP completely ␦ 3 without alcohol. The Kd value of ligand binding was moderately displaced [ H]DAMGO binding, indicating the specificity of ra- increased ( p Ͻ 0.05) in naltrexone-treated animals (27.2 Ϯ 3.9; n dioligand bindings. CTAP and SNC80 could not displace ϭ 3), but not in naltrexone-treated, alcohol-fed animals (15.5 Ϯ [3H]DAMGO and [3H]naltrindole binding, respectively, even at 3.6; n ϭ 3). As splenocytes derived from naltrexone-treated animals showed an increased NK cell response to ␦ opioid ligands and ␦ an elevated level of opioid ligand binding, the possibility arose 3 3 ␦ Table I. Pharmacological characterization of [ H]Naltrindole and H that naltrexone up-regulated opioid-like receptor activity. DAMGO binding sites on rat splenocytesa Naltrexone’s ability to alter ␮ opioid receptor binding to ligands K1 (nM) The NK cell cytolytic response to the ␮ opioid receptor ligand was Ligands [3H]Naltrindole [3H]DAMGO reduced in splenocytes derived from naltrexone-treated and nal- SNC80 7.72 Ϯ 2.49 Ͼ1000 trexone- and alcohol-treated animals. NK cell cytolytic assays CTAP Ͼ1000 1.3 Ϯ 0.6 were conducted in vitro using splenocytes exposed to the opiate DADLE 27.26 Ϯ 10.99 242 Ϯ 74 Ϯ Ϯ antagonist in vivo. During the assay, splenocytes were not exposed Naltrexone 183 76 5.3 1.5 to naltrexone. Hence, the question arose of whether the reduced a n ϭ 3–5. 46 OPIOID RECEPTOR CONTROL OF NK CELL CYTOLYTIC FUNCTION Downloaded from

3 3 ␮

FIGURE 5. Representative displacement curves of [ H]DAMGO and [ H]naltrindole binding on splenocytes by a opioid receptor ligand, CTAP (A); http://www.jimmunol.org/ a ␦-ligand, SNC80 (B); a mixed ␮ and ␦ ligand, naltrexone (C); and a ␦ and ␮ ligand, DADLE (D). [3H]naltrindole and [3H]DAMGO were used at 8 and

6 nM, respectively. The Ki values for binding of these ligands are shown in Table I. high concentrations (10Ϫ4 M). Naltrexone, an opioid receptor an- with naltrexone, we tested the possibility that there may exist an tagonist, displaced binding of [3H]naltrindole and [3H]DAMGO, intrinsic feedback inhibition of constitutively active ␮ receptors on 3 ␦ but the Ki value for [ H]naltrindole was higher than that for receptors. In vitro studies were conducted to determine the effect [3H]DAMGO. Similarly, DADLE displaced the binding of both of naltrexone on the interaction between ␮ and ␦ receptors in 3 radioligands, but showed high Ki for [ H]DAMGO. These data splenocytes. Treatment with naltrexone for 3 days significantly by guest on September 28, 2021 suggest that [3H]naltrindole and [3H]DAMGO binding sites on increased ␦ ligand binding (Fig. 6, A and C) and reversed alcohol’s splenocytes are classical ␦ and ␮ opioid receptor binding sites, inhibitory effects on this ligand binding without affecting the af- ␦ respectively. finity of the receptor (data not shown). The Kd of opioid receptors in splenocytes was between 6 and 11 nM. In contrast, naltrex- ␮ ␦ Identification of negative feedback interaction between and one treatment with or without alcohol decreased ␮ ligand binding receptors in splenocytes: effect of naltrexone (Fig. 6, B and D), suggesting that long term treatment of the an- As we noted an inverse relationship between the levels of ␦ opioid tagonist inhibits ␮ opioid receptor activity. Alcohol by itself also receptor and ␮ opioid receptors in the spleen of animals treated decreased the ␮ ligand binding, confirming the in vivo data with

FIGURE 6. Naltrexone modulations of opioid receptor binding in splenocytes: in vitro effects. Shown are representative saturation curves and the mean Ϯ SE Bmax of [3H]naltrindole binding (A and B) and [3H]DAMGO binding (C and D)on splenocytes obtained from ad libitum-fed rats and maintained in cultures in the absence (CONT) or the presence of naltrexone (NAL) and alcohol (AL), either alone or in combination (NAL/AL). a, p Ͻ 0.05, signiﬁcantly different from the CONT group; b, p Ͻ 0.01, signiﬁcantly different from the AL group. The Journal of Immunology 47

Table II. Effects of ␮-opioid receptor-speciﬁc antagonist CTAP or ␦- data suggest that chronic inhibition of one of these two opioid opioid receptor-speciﬁc antagonist naltrindole on [3H]DAMGO binding receptors up-regulated the ligand binding ability of the other. Fur- 3 a and [ H]naltrindole binding on splenocytes in primary cultures thermore, the data provide support to the concept that there may exist a negative feedback interaction between ␮ and ␦ opioid re- 3 3 [ H]DAMGO binding [ H]Naltrindole binding ceptors in the spleen (Fig. 7). ϫ 6 ϫ 6 Treatment Bmax/4 10 cells Kd (nM) Bmax/1 10 cells Kd (nM) Discussion Ϯ Ϯ Ϯ Ϯ Control 42.1 3.5 3.9 0.1 124.5 5.0 6.5 0.8 The data obtained from in vivo studies presented here demon- Naltrindole 53.0 Ϯ 1.3b 6.7 Ϯ 1.0 87.3 Ϯ 11.9b 3.6 Ϯ 0.2 CTAP NDc ND 162.8 Ϯ 3.8b 9.6 Ϯ 2.1 strated that alcohol consumption for 2 wk suppressed NK cell cytolytic activity, and that naltrexone treatment prevented alcohol’s a n ϭ 3–6. b p Ͻ 0.05, compared to control. inhibitory action on NK cells. The opioid antagonist also improved c ND, not detectable in the assay. NK cell function in control animals. Naltrexone treatment enhanced NK cell responses to endorphin, enkephalin, and the ␦ opioid agonist and blocked alcohol’s ability to inhibit NK cell ␮ the ligand binding and the NK cell response. The Kd values of responses to these opioid receptor agonists. However, naltrexone ␮ ligand binding were increased ( p Ͻ 0.01) in alcohol-treated inhibited the NK cell response to the ␮ opioid receptor agonist and animals (control, 4.39 Ϯ 2.08; alcohol, 11.7 Ϯ 0.45; naltrexone, potentiated alcohol’s ability to block ␮ agonist action on NK cells. 3.49 Ϯ 0.8; naltrexone and alcohol, 13.09 Ϯ 2.35; n ϭ 3). This The opioid antagonist increased ␦ ligand binding and prevented

␦ Downloaded from could explain in part the lower Bmax in the alcohol-treated groups. alcohol’s inhibitory action on the ligand binding in splenocytes, As with naltrexone, exposure for 3 days to the ␮-specific an- whereas it decreased ␮ ligand binding and affinity and potentiated tagonist CTAP (29) decreased ␮ ligand binding, but increased ␦ alcohol’s inhibitory action on the ␮ ligand binding and affinity in ligand binding (Table II). The affinity of ␮ or ␦ ligand binding in splenocytes. Displacement studies using ␮- and ␦-specific ligands splenocytes did not change after CTAP treatments. However, treat- revealed that the classical ␦ and ␮ opioid receptor binding sites ment with the ␦-specific antagonist naltrindole increased ␮ ligand exist on splenocytes. The opioid antagonist also increased ␦ opioid

binding and decreased ␦ ligand binding in splenocytes. Also, nal- ligand binding, but reduced ␮ ligand binding in splenocytes in http://www.jimmunol.org/ trindole did not change the afﬁnity of either ␮ or ␦ receptors. These vitro. Like naltrexone, the ␮ opioid receptor-speciﬁc ligand CTAP by guest on September 28, 2021

FIGURE 7. Schematic representation of the pro- posed mechanisms by which naltrexone and ethanol affect negative interaction between ␦ opioid receptor (DOR) and ␮ opioid receptor (MOR). We propose that a negative interaction between MOR and DOR exists in the spleen, possibly due to heterodimerization of MOR and DOR or other unknown mechanisms. We also propose that an increased negative interaction between DOR and MOR might be a mechanism by which ethanol alters the NK cell response to the endogenous opioid peptide (OP). Also we propose that the action of naltrexone on NK cells might be related to prevention of the negative interaction between MOR and DOR, because the antagonist prevents MOR binding. 48 OPIOID RECEPTOR CONTROL OF NK CELL CYTOLYTIC FUNCTION increased ␦ ligand binding, but reduced ␮ ligand binding, in the spinal cord (38). The antagonist-induced micro-opioid receptor splenocytes. In contrast, treatment with the ␦-specific antagonist up-regulation decreases G protein receptor kinase-2 and dy- naltrindole increased ␮ ligand binding and decreased ␦ ligand namin-2 abundance in the mouse spinal cord, and this may be due binding. These data suggest that naltrexone increases NK cell cy- to a reduction in constitutive internalization of opioid receptors. tolytic function and blocks alcohol’s suppressive action on NK Our data presented in this study indicated that naltrexone prefer- cells. They also suggest that there exists a feedback interaction entially prevented ␮ opioid receptor activity, but increased ␦ opi- between ␦ and ␮ opioid receptors in splenocytes. Naltrexone dis- oid receptor activity. It was somewhat surprising to find that nal- rupts this feedback control by reducing ␮ opioid receptor function, trexone, which acts more as a general antagonist of opioid thereby up-regulating the endogenous ␦ opioid receptor ligand- receptors in the brain tissue, acted as a preferential antagonist of ␮ regulated enhancement of NK cell cytolytic activity. opioid receptors in the spleen. The ␦ opioid receptor stimulatory The action of alcohol on NK cells has not been well studied. We action of chronic naltrexone in the spleen is also a novel finding. have recently shown a time-dependent suppressive effect of alcohol These findings are consistent with the observations that chronic on NK cell cytolytic function in rats given an alcohol-containing naltrexone increases the ␦ opioid receptor-mediated antinocicep- (35% of the total calories) liquid diet (4). With careful attention to tion response as well as levels of ␦ receptor binding in brain tissues nutritional status and alcohol intake, studies in which mice were given (39). Hence, ␦ opioid receptors may have distinct characteristics high levels of alcohol (20%, w/v) in their drinking water showed that adapt to chronic naltrexone in tissues, including those of brain inhibition of NK cell cytolytic activity (7). Wu and Pratt (31, 32) in a and spleen. binge alcohol model also showed inhibition of NK cell cytolytic ac- The mechanisms underlying the response of opioid receptors to tivity and a decrease in cell percentages and numbers in mice. As- endogenous and exogenous ligands after prolonged exposure to Downloaded from sessments of NK cell activity in humans is complicated by the fact opioid receptor antagonist were not well studied. In the cellular that many patients use tobacco and other drugs that are known to model, chronic exposure to ␮ or ␦ opioid agonists results in several affect NK cell activity (33, 34). Furthermore, alcoholics are frequently adaptive responses, including receptor desensitization (40), recep- malnourished (35). Charpentier et al. (36) found that NK cell activity tor down-regulation (41), receptor internalization (42), and adeny- was impaired only in alcoholic subjects with inactive cirrhosis. Ad- lyl cyclase supersensitization (43). Chronic treatment with opioid

ditionally, the decrease was more pronounced in cirrhotic patients agonists converts the antagonists naloxone and naltriben into in- http://www.jimmunol.org/ with severe malnutrition. Cook et al. (6) showed a trend toward lower verse agonists on ␦ opioid receptors (44). Both ␮ and ␦ opioid NK cell activity in alcoholics without liver disease and a loss of NK receptors have constitutive activity (31, 45, 46), and this may con- cells in alcoholics with liver disease. NK cell activity was also de- tribute to the development of adaptive responses. creased in depressed patients with alcoholism (8). Although additional Opioid receptors are members of the large superfamily of G research is needed in human alcoholics, a majority of these studies protein-coupled receptors (47). Like many G protein-coupled re- suggest reduced NK cell function. ceptors, ␦ and ␮ opioid receptors are constitutively active, produc- Alcohol’s effect on NK cells may be related to the alteration of ing spontaneous regulation of G protein and effectors in the ab- various neuroendocrine factors or local factors regulating NK cell sence of agonists (45, 48). It could be that chronic opioid functions (4, 12, 37). Of the neuroendocrine factors, opioid pep- antagonist treatment increases the affinity of agonists for those by guest on September 28, 2021 tides play significant regulatory roles in controlling NK cell func- opioid receptors that are not occupied by the antagonist, thereby tion. It has been shown that chronic treatment with alcohol reduces increasing the responsivity of the receptors to endogenous ligands. central and peripheral levels of ␤-endorphin, and that the reduction However, this view cannot explain the differential responses of the of hypothalamic and plasma ␤-endorphin after chronic alcohol in- ␦ and ␮ opioid receptors to naltrexone. As we noted an inverse take correlates with the reduction of NK cell cytolytic activity. relationship between the expression of these opioid receptors in the Furthermore, exogenous ␤-endorphin raises the reduced NK cell spleen after naltrexone treatment, we tested the possibility that activity to a normal level in alcohol-treated animals (4). The data there may exist an intrinsic feedback inhibition of constitutively presented in this study are consistent with ␤-endorphin involve- active ␮ opioid receptors on ␦ opioid receptors. This view is sup- ment and reveal for the first time that alcohol modulation of NK ported by our data showing down-regulation of ␮ opioid receptors cells involves down-regulation of ligand binding and responses of and up-regulation of ␦ opioid receptors after chronic naltrexone or ␦ and ␮ opioid receptors in the spleen. The expression of opioid ␮ opioid receptor-specific ligand CTAP treatment. Our studies de- receptors on NK cells has not been demonstrated, primarily due to termining the effects of ␦ opioid receptor-specific ligand naltridone low expression of the receptors and the lack of specific opioid also supported the concept that a feedback inhibition may exist receptor antibodies. We have seen both [3H]naltrindole and between constitutively active ␦ opioid receptors on ␮ opioid [3H]DAMGO binding on the rat-derived NK cell line, RNK-16 receptors. (data not shown). Additionally, [3H]naltrindole and [3H]DAMGO Recently, it has been shown that opioid receptors form dimers to binding was observed on splenocytes in cultures. In the spleno- function in various tissues (49). As ␦ and ␮ opioid receptors are cytes, 5–6% of the cells are NK cells. Hence, the possibility exists present in splenocytes, it is possible that ␦ and ␮ heterodimers also that the action of opioid ligands on NK cell cytolytic activity is due exist in this tissue. We found that an increase in binding of the at least partly to a direct action on NK cells. ␦-selective agonist occurred in the presence of a ␮ antagonist. This study also focused on the chronic effects of the opioid Also, the binding of a ␮-selective agonist increased in the presence receptor antagonist naltrexone on NK cell activity. The present of a ␦ antagonist. Previously, it has been observed that the binding data demonstrated, for the first time, that chronic administration of of a ␮ agonist increases in the presence of a ␦-selective antagonist naltrexone reversed the suppressive effect of alcohol on NK cell in cells expressing ␮ and ␦ dimers (49). Hence, it could be inter- activity. Moreover, our data indicated, for the first time, that the preted that by decreasing the ␮ opioid receptors, naltrexone un- opioid receptor antagonist naltrexone enhanced the effects of en- coupled the ␮ and ␦ dimers, thereby leading to increased ␦ ligand kephalins, ␤-endorphin, and ␦ opioid receptor agonists on NK cell binding sites and increased endogenous opioid-activated NK cell activity in both pair-fed and alcohol-fed rats. Previously it has been function in splenocytes (Fig. 6). An alternative explanation is that shown that treatment of mice with a naltrexone pellet for 8 days in the spleen, constitutively active ␮ receptors inhibited the ligand produced a significant increase in micro-opioid receptor density in binding ability of the ␦ receptors, and therefore naltrexone, by The Journal of Immunology 49 down-regulating ␮ receptor expression, increased ␦ receptor func- 23. Rothman, R. B., C. P. France, V. Bykov, B. R. De Costa, A. E. Jacobson, tion. This novel interaction between ␮ and ␦ opioid receptors also J. H. Woods, and K. C. Rice. 1989. Pharmacological activities of optically pure enantiomers of the ␬ opioid agonist, U50,488, and its cis diastereomer, evidence provides a general mechanism for differential responses of these for three ␬ receptor subtypes. Eur. J. Pharmacol. 167:345. receptors after chronic treatment with opiate agonists and antago- 24. Fang, F. G., C. M. Haws, K. Drasner, A. Williamson, and H. L. Fields. 1989. Opioid peptides (DAGO-enkephalin, dynorphin A1–13, BAM 22P) microinjected nists (39, 44). into the rat brainstem, comparison of their antinociceptive effect and their effect In summary, we have demonstrated that endogenous opioid pep- on neuronal firing in the rostral ventromedial medulla. Brain Res. 501:116. tides have binding sites in the cellular immune system, and acti- 25. Pross, H. F., M. G. Baines, P. Rubin, P. Sharagge, and M. S. Patterson. 1981. Spontaneous human lymphocyte-mediated cytotoxicity against tumor cells: the vation of these receptors leads to increased NK cell function. We quantitation of natural killer cell activity. J. Clin. Immunol. 1:51. have also shown that NK cells of alcohol-fed animals were less 26. Emanuele, N. V., N. LaPaglia, J. Steiner, L. Kirsteins, and M. A. Emanuele. responsive to opioid activation. Our results indicated that naltrex- 1999. Reversal of chronic ethanol-induced testosterone suppression in peripuber- tal male rats by opiate blockade. Alcohol. Clin. Exp. Res. 23:60. one administration during alcohol intake reverses the suppressive 27. Madden, J. J., R. M. Donahoe, J. Zwemer-Collins, D. A. Shafer, and A. Falek. effects of alcohol on NK cell function by preventing negative feed- 1987. Binding of naloxone to human T lymphocytes. Biochem. Phamacol. back action of ␮ opioid receptors on ␦ opioid receptor and by 36:4103. ␦ 28. Kreek, M. J. 1996. Opiates, opioids and addiction. Mol. Psychiatry 1:232. increasing the function of receptors in the spleen. Hence, nal- 29. Abbruscato, T. J., S. A. Thomas, V. J. Hruby, and T. P. Davis. 1997. Blood-brain trexone may have potential therapeutic value for the treatment of barrier permeability and bioavailability of a highly potent and ␮-selective opioid immune deficiency in alcoholic and nonalcoholic patients. receptor antagonist, CTAP: comparison with morphin. J. Pharmacol. Exp. Ther. 280:402. 30. Raynor, K., H. Kong, Y. Chen, K. Yasuda, L. Yu, G. L. Bell, and T. Reisine. References 1994. Pharmacological characterization of cloned ␬-, ␦- and ␮-opioid receptors.

1. Trinchieri, G. 1991. Biology of natural killer cells. In Advancement in Immunol- Mol. Pharmacol. 45:330. Downloaded from ogy, Vol. 47. F. J. Dixon, ed. Academic Press, San Diego, p. 187. 31. Wu, W. J., and S. B. Pruett. 1994. Ethanol decreases the number and activity of 2. Hercend, T., and R. E. Schmidt. 1988. Characteristics and uses of natural killer splenic natural killer cells in a mouse model for binge drinking. J. Pharmacol. cells. Immunol. Today 9:291. Exp. Ther. 271:722. 3. Herberman, R. B., and J. R. Ortaldo. 1981. Natural-killer cells: their role in 32. Wu, W. J., and S. B. Pruett. 1996. Suppression of splenic natural killer cell activity defense against disease. Science 214:24. in a mouse model for binge drinking. II. Role of the neuroendocrine system. J. Phar- 4. Boyadjieva, N., M. Dokur, J. P. Advis, G. G. Meadows, and D. K. Sarkar. 2001. macol. Exp. Ther. 278:1331. Chronic ethanol inhibits NK cell cytolytic activity: role of opioid peptide ␤-en- 33. Ferson, M., A. Edwards, A. Lind, G. W. Milton, and P. Hersey. 1979. Low natural killer-cell activity and immunoglobulin levels associated with smoking in

dorphin. J. Immunol. 167:5645. http://www.jimmunol.org/ 5. Szabo, G. 1998. Monocytes, alcohol use and altered immunity. Alcohol. Clin. human subjects. Int. J. Cancer 23:603. Exp. Res. 22:216S. 34. Klimas, N. G., R. Morgan, N. T. Blaney, D. Chitwood, J. B. Page, K. Milles, and 6. Cook, R. T., F. Li, D. Vandersteen, Z. K. Ballas, B. L. Cook, and M. A. Fletcher. 1990. Alcohol and immune function in HIV-1 seronegative, D. R. LaBrecque. 1997. Ethanol and natural killer cells. I. Activity and immu- HTLV-I/II seronegative and positive men on methadone. In Progress in Clinical nophenotype in alcoholic humans. Alcohol. Clin. Exp. Res. 21:974. and Biological Research, Vol. 325: Alcohol, Immunomodulation and AIDS. 7. Blank, S. E., D. A. Duncan, and G. G. Meadows. 1991. Suppression of natural D. Seminara, R. R. Watson, and A. Pawlowski, eds. Alan R. Liss, New York, killer cell activity by ethanol consumption and food restriction. Alcohol. Clin. p. 103. Exp. Res. 15:16. 35. Shaw, S., and C. S. Lieber. 1978. Plasma amino acid abnormalities in the alco- 8. Irwin, M., C. Caldwell, T. L. Smith, S. Brown, M. A. Schuckit, and J. C. Gillin. holic: respective role of alcohol, nutrition and liver injury. Gastroenterology 1990. Major depressive disorder, alcoholism, and reduced natural killer cell cy- 74:677. totoxicity, role of severity of depressive symptoms and alcohol consumption. 36. Charpentier, B., D. Franco, L. Paci, M. Charra, B. Martin, D. Vuitton, and

Arch. Gen. Psychiatry 47:713. D. Fries. 1984. Deficient natural killer cell activity in alcoholic cirrhosis. Clin. by guest on September 28, 2021 9. MacGregor, R. R. 1986. Alcohol and immune defense. JAMA 256:1474. Exp. Immunol. 58:107. 10. Sarkar, D. K., J. G. Liehr, and K. W. Singletary. 2001. Role of estrogen in alcohol 37. Blank, S. E., R. M. Gallucci, J. H. Wang, and G. G. Meadows. 1994. Factors promotion of breast cancer and prolactinomas. Alcohol. Clin. Exp. Res. 25:230S. involved in modulation of NK cell activity by ethanol consumption. Alcohol 11. Dafny, N. 1983. Modification of morphine withdrawal by interferon. Life Sci. Alcohol. 2(Suppl.):439. 32:303. 38. Patel, M, B. Gomes, C. Patel, and B. C. Yoburn. 2002. Antagonist-induced mi- 12. Pruett, S. B., Y. C. Han, and W. J. Wu. 1994. A brief review of immunomodu- cro-opioid receptor up-regulation decreases G-protein receptor kinase-2 and dy- lation caused by acute administration of ethanol, involvement of neuroendocrine namin-2 abundance in mouse spinal cord. Eur. J. Pharmacol. 446:37. pathways. Alcohol Alcohol. Suppl. 2:431. 39. Kest, B., C. E. Lee, S. Jenab, B. Kao, and C. E. Inturrisi. 1998. Chronic naltrex- ␤ 13. Mandler, R. N., W. E. Biddison, R. Mandler and S. A. Serrate. 1986. -Endor- one differentially affects supraspinal ␦-opioid receptor-mediated antinociception. phin augments the cytolytic activity and interferon production of natural killer Eur. J. Pharmacol. 345:47. cells. J. Immunol. 136:934. 40. Kovoor, A., V. Nappey, B. L Kieffer, and C. Chavkin. 1997. Mu and ␦ opioid 14. Puente, J., P. Maturana, D. Miranda, C. Navarro, M. E. Wolf, and receptors are differentially desensitized by the coexpression of ␤-adrenergic re- A. D. Mosnaim. 1992. Enhancement of human natural killer cell activity by ceptor kinase 2 and ␤-arrestin 2 in Xenopus oocytes. J. Biol. Chem. 272:27605. ␤ opioid peptides, similar response to methionine-enkephalin and -endorphin. 41. Chakrabarti, S., W. Yang, P. Y. Law, and H. H Loh. 1997. The ␮opioid receptor Brain Behav. Immun. 6:32. down-regulates differently from the ␦-opioid receptor: requirement of a high af- 15. McCarthy, L., M. Wetzel, J. K. Sliker, T. K. Eisenstein, and T. J. Rogers. 2001. finity receptor/G protein complex formation. Mol. Pharmacol. 52:105. Opioids, opioid receptors, and the immune system. Drug Alcohol Depend. 42. Keith, D. E., S. R. Murray, P. A. Zaki, P. C. Chu, D. V. Lissin, L. Kang, 62:111. C. J. Evans, and M. von Zastrow. 1996. Morphine activates opioid receptors 16. Brownstein, M. J. 1993. A brief history of opiates, opioid peptides, and opioid without causing their rapid internalization. J. Biol. Chem. 271:19021. receptors. Proc. Natl. Acad. Sci. USA 90:5391. 17. Shahabi, N. A., K. McAllen, S. G. Matta, and B. M. Sharp. 2000. Expression of 43. Sharma, S. K., W. A. Klee, and M. Nirenberg. 1975. Dual regulation of adenylate ␦ opioid receptors by splenocytes from SEB-treated mice and effects on phos- cyclase accounts for narcotic dependence and tolerance. Proc. Natl. Acad. Sci. phorylation of MAP kinase. Cell. Immunol. 205:84. USA 72:3092. 18. Heagy, W., E. Teng, P. Lopez, and R. W. Finberg. 1999. Enkephalin receptors 44. Liu, J.-G., and P. L. Prather. 2002. Chronic agonist treatment converts antagonists ␦ and receptor-mediated signal transduction in cultures human lymphocytes. Cell. into inverse agonists at -opioid receptors. J. Pharmacol. Exp. Ther. 302:1070. Immunol. 191:34. 45. Merkouris, M., I. Mullaney, Z. Georgoussi, and G. Milligan. 1997. Regulation of ␦ 19. Boyadjieva, N. I., M. Dokur, J. P. Advis, G. G. Meadows, and D. K. Sarkar. 2002. spontaneous activity of the -opioid receptor: studies of inverse agonism in intact ␤-Endorphin modulation of lymphocyte proliferation, effects of ethanol. Alcohol. cells. J. Neurochem. 69:2115. Clin. Exp. Res. 26:1719. 46. Liu, J. G., M. B. Ruckle, and P. L. Prather. 2001. Constitutively active ␮opioid 20. Gianoulakis, C., X. Dai, and T. Brown. 2003. Effect of chronic alcohol consump- receptors inhibit adenylyl cyclase activity in intact cells and activate G-proteins 2 4 5 tion on the activity of the hypothalamic-pituitary-adrenal axis and pituitary ␤-en- differently than the agonist [D-Ala ,N-MePhe ,Gly-ol ]enkephalin. J. Biol. Chem. dorphin as a function of alcohol intake, age, and gender. Alcohol. Clin. Exp. Res. 276:37779. 27:410. 47. Law, P. Y., Y. H. Wong, and H. H. Loh. 2000. Molecular mechanisms and 21. O’Malley, S. S., and J. C. Froehlich. 2003. Advances in the use of naltrexone, an regulation of opioid receptor signalling. Annu. Rev. Pharmacol. Toxicol. 40:389. integration of preclinical and clinical findings. Recent Dev. Alcohol. 16:217. 48. Burford, N. T., D. Wang, and W. Sadee. 2000. G-protein coupling of ␮opioid 22. Bilsky, E. J., S. N. Calderon, T. Wang, R. N. Bernstein, P. Davis, V. J. Hruby, receptors (OP3): elevated basal signalling activity. Biochem. J. 348:531. R. W. McNutt, R. B. Rothman, K. C. Rice, and F. Porreca. 1995. SNC 80, a 49. Gomes, I., B. A. Jordan, A., Gupta, N. Trapaidze, V. Nagy, and L. A. Devi. 2000. selective, nonpeptidic and systemically active opioid ␦ agonist. J. Pharmacol. Heterodimerization of ␮- and ␦-opioid receptors, a role in opiate synergy. J. Neu- Exp. Ther. 273:359. rosci. 20:RC110.