Effect of the A118G polymorphism on binding affinity, potency and agonist-mediated endocytosis, desensitization, and resensitization of the human mu-opioid receptor


Address correspondence and reprint requests to Thomas Koch, Department of Pharmacology and Toxicology, Otto-von-Guericke-University, 39120 Magdeburg, Leipziger Str. 44, Germany.
E-mail: Thomas.Koch@Medizin.Uni-Magdeburg.de


The most prevalent single-nucleotide polymorphism (SNP) A118G in the human µ-opioid receptor gene predicts an amino acid change from an asparagine residue to an aspartatic residue in amino acid position 40. This N40D mutation, which has been implicated in the development of opioid addiction, was previously reported to result in an increased β-endorphin binding affinity and a decreased potency of morphine-6-glucuronide. Therefore, in the present study we have investigated whether this mutation might affect the binding affinity, potency, and/or the agonist-induced desensitization, internalization and resensitization of the human µ-opioid receptor stably expressed in human embryonic kidney 293 cells. With the exception of a reduced expression level of N40D compared to human µ-opioid receptor (hMOR) in HEK293 cells, our analyses revealed no marked functional differences between N40D and wild-type receptor. Morphine, morphine-6-glucuronide and β-endorphin revealed similar binding affinities and potencies for both receptors. Both the N40D-variant receptor and hMOR exhibited robust receptor internalization in the presence of the opioid peptide [d-Ala2,N-MePhe4,Glyol5]enkephalin (DAMGO) and β-endorphin but not in response to morphine or morphine-6-glucuronide. After prolonged treatment with morphine, morphine-6-glucuronide or β-endorphin both receptors showed similiar desensitization time courses. In addition, the receptor resensitization rates were nearly identical for both receptor types.

Abbreviations used

Ca2+/calmodulin-dependent protein kinase II



G protein

guanine nucleotide-binding protein


G protein-coupled receptor kinase


human embryonic kidney 293


human µ-opioid receptor




single-nucleotide polymorphism

The characteristic effects of opioids, such as analgesia, sedation, euphoria/dysphoria, respiratory depression, reward and dependence, are mediated by the interaction with specific opioid receptors. Pharmacological studies have defined three opioid receptor types designated mu-, delta- and kappa-opioid receptor (Goldstein and Naidu 1989) which are activated by endogenous neuropeptides such as β-endorphin, enkephalin and dynorphin. However, most clinically used opioids, including morphine, methadone, fentanyl and related drugs, act through the µ-opioid receptor. Opioid receptors belong to the seven transmembrane protein family and couple via heterotrimeric guanine nucleotide-binding proteins (G proteins) to a variety of downstream effectors, including adenylate cyclase (Johnson et al. 1994), phospholipase C (Smart et al. 1994, 1995; Smith et al. 1999) and mitogen-activated protein kinase (Burt et al. 1996; Li and Chang 1996; Schulz and Hollt 1998; Schmidt et al. 2000). The agonist binding results in a receptor conformation state which facilitates the GDP/GTP exchange of the Gα-subunit and induces the dissociation of the activated Gα- and Gβγ-subunits (Neer 1995). Opioid receptors couple to an inhibitory G protein (Gi) that inhibits adenylate cyclase activity, resulting in a reduction of intracellular cAMP production.

After chronic agonist treatment the receptor activity is rapidly attenuated, a process which is also termed receptor desensitization. An important mechanism of G protein-coupled receptor desensitization is the phosphorylation of intracellular receptor domains by G protein-coupled receptor kinases (GRKs) or second messenger-regulated protein kinases such as protein kinase C, cAMP-dependent protein kinase or Ca2+/calmodulin-dependent protein kinase II. Receptor phosphorylation facilitates binding of β-arrestin molecules and results in an uncoupling of the receptor from G proteins. Furthermore β-arrestin binding induces receptor internalization by a clathrin-mediated endocytosis. Consecutively, the receptor is separated from the bound ligand, dephosphorylated, and recycled to the cell surface in a reactivated state (Koch et al. 1998).

The development of opioid tolerance, dependence and addiction have been the subject of extensive studies. In particular, genetic variations in the hMOR (human µ-opioid receptor) gene affecting structure, regulation and expression of the receptor protein have been suggested to confer genetic vulnerability to drug addiction. Using large-scale sequencing, eight single-nucleotide polymorphisms (SNPs) have been identified in the human µ-opioid receptor gene, which affect the amino acid sequence (Hoehe et al. 2000; LaForge et al. 2000). Three polymorphisms (S4R, A6V and N40D) are localized in the extracellular N-terminus, two (S147C, N152D) in the third transmembrane domain and three (R360H, R365H and S268P) in the third intracellular loop of the human µ-opioid receptor. Allelic variation S268P affects the important Ca2+/calmodulin-dependent protein kinase II (CaMK II) phosphorylation site in the third intracellular loop of the hMOR and results in a resistance to CaMK II-mediated desensitization. It has been demonstrated that this mutant receptor revealed a slower agonist-induced desensitization compared with wild-type hMOR and a weaker G protein coupling after agonist treatment (Koch et al. 2000).

The most frequent single-nucleotide polymorphism A118G leads to a change in the amino acid sequence (N40D) in the extracellular N-terminal domain of the µ-opioid receptor (Bergen et al. 1997). For different ethnic groups the overall allelic frequency of A118G polymorphism varied between 10.5% and 32.1% (LaForge et al. 2000; Li et al. 2000). Interestingly, investigations of the functional properties of the N40D mutant led to inconsistent results. For β-endorphin an increased receptor binding affinity and potency as well as an increased activation of G protein-activated potassium channels has been described at the N40D-variant receptor in AV-12 cells and Xenopus oocytes (Bond et al. 1998). On the other hand, no differences in agonist binding and functional coupling between N40D variant and wild-type µ-opioid receptor have been observed in transiently expressed COS cells (Befort et al. 2001). Recently, in humans a decreased potency of morphine-6-glucuronide at the N40D receptor has been described (Lotsch et al. 2002a). Morphine-6-glucuronide is a metabolized product of morphine and still an highly active opioid agonist leading to toxic effects during chronic morphine treatment. However, the mechanistic basis for the observed differences in morphine-6-glucuronide potency between N40D and hMOR wild-type receptor remains unclear.

In the present study we examined whether the N40D mutation affect the ligand binding and activity as well as the agonist-induced endocytosis, desensitization and resensitization of the µ-opioid receptor by using the agonists morphine, morphine-6-glucuronide and β-endorphin in HEK293 cells. Unambiguously, our results indicate no functional differences between N40D-variant receptor and the hMOR wild-type receptor.

Materials and methods

Epitope tagging and subcloning of DNA

The human µ-opioid receptor (hMOR) and the N40D receptor mutant were tagged with an N-terminal HA epitope tag sequence MYPYNVPNYA and then subcloned into the pEAK10 expression vector (Edge Bio Systems, Gaithersburg, MD, USA). To introduce a HindIII restriction site, the forward primer 5′-CGTGAAAAGCTTACCATGTACCCATACGACGTCCCAGACTACGCTGAC-AGCAGCGCTGCCCCCACG-3′ was synthesized and used for PCR amplification of both receptors. The sequence for the reverse primer introducing an XbaI restriction site was 5′-TCGGAATGGCTCTAGACCCTGTTAGGG-3′.

Generation of cell lines expressing μ-opioid receptor or N40D receptor mutant

Transfection of HEK293 cells were performed according to the calcium phosphate precipitation method (Chen and Okayama 1988). Approximately 1.5 × 105 cells were transfected with 20 µg of plasmid DNA. Cells were selected in the presence of 1 µg/mL puromycin (Sigma), and resistant cells were grown in the presence of 1 µg/mL puromycin. Receptor expression was monitored using cAMP assays and confocal microscopy as described below.

Radioligand binding assay

Binding studies were performed as described previously (Koch et al. 1998). Fifty-microliter aliquots of the crude membrane suspension in 50 mm Tris-HCl buffer (pH 7.8, 1 mm EGTA/2 mm EDTA, 5 mm MgCl2) containing 150–250 µg protein were mixed with [3H]DAMGO ([d-Ala2,N-MePhe4,Glyol5]enkephalin) and were subsequently incubated for 40 min at 25°C. All assays were carried out at least in duplicate. Specific binding was calculated by subtracting non-specific binding, defined as that seen in the presence of 2.5 nm[3H]DAMGO plus 1 µm unlabelled DAMGO, from total binding obtained with 2.5 nm[3H]DAMGO alone. In the competition experiments for determination of IC50 for morphine, morphine-6-glucuronide and β-endorphin the drugs were incubated in the range 10−11 to 10−4m. In all binding experiments the reaction was terminated by rapid filtration under reduced pressure through 0.1% polyethyleneimine-treated GF 10 glass-fibre filters. Filters were washed with buffer and taken for liquid scintillation in a toluene-containing solvent. Data were calculated as femtomole-bound radioligand per milligram of protein.

The dissociation constant (Kd) and the number of [3H]DAMGO-binding sites (Bmax) were calculated by Scatchard analysis using at least seven concentrations of labelled DAMGO in a range of 0.25–10 nm.

Measurement of cyclic AMP levels

Approximately 1.5 × 105 cells/well were seeded in 22-mm 12-well dishes with Dulbecco's modified Eagle medium containing 10% fetal calf serum and incubated at 37°C for 24 h. For desensitization studies cells were then exposed to 1 µm DAMGO (Bachem, Heidelberg, Germany), morphine, morphine-6-glucuronide or β-endorphin for 0, 0.5, 1, 2, 4 or 6 h. For resensitization assays, cells were washed after 6 h of DAMGO exposure followed by an additional incubation period of either 0, 10, 20, 30 or 40 min in the absence of agonist.

For the measurement of cAMP accumulation, medium was removed and replaced by 0.5 mL serum-free RPMI medium (Seromed, Berlin, Germany) containing 25 µm forskolin (Biotrend, Köln, Germany) or 25 µm forskolin plus 1 µm DAMGO, morphine, morphine-6-glucuronide or β-endorphin. The cells were then incubated for 15 min at 37°C. The reaction was terminated by removing the medium and sonicating the cells in 1 mL of ice-cold HCl/ethanol (1 vol. of 1 N HCl/100 vol. of ethanol). After centrifugation, the supernatant was evaporated, the residue was dissolved in Tris-EDTA buffer, and the cAMP content measured using a commercial radioassay kit (Amersham, Braunschweig, Germany).

Confocal microscopy

HEK293 cells stable expressing hMOR or the N40D receptor mutant were grown onto poly-l-lysine-coated coverslips overnight. After washing, the cells were incubated with 1 µg/mL affinity-purified polyclonal rabbit anti-HA-tag antibody (Gramsch Laboratories, Schwabhausen, Germany) for 2 h at 4°C to label cell-surface receptors. The cells were subsequently exposed to 1 µm DAMGO, 1 µm morphine, morphine-6-glucuronide or β-endorphin for 0, 10 or 30 min at 37°C to induce receptor endocytosis. The cells were then fixed with 4% paraformaldehyde and 0.2% picric acid in 0.1 m phosphate buffer, pH 6.9, for 45 min at room temperature and subsequently washed several times in TPBS (10 mm Tris, 10 mm phosphate buffer, 137 mm NaCl, 0.05% thimerosal, pH 7.4) and TPBS containing 1% normal goat serum. Cells were then incubated for 3 min in 50% methanol and 3 min in 100% methanol and subsequently washed several times in TPBS. After 1 h pre-incubation in TPBS containing 3% normal goat serum, cells were incubated with Cyanine 3 labelled anti-rabbit secondary antibody at a dilution of 1 : 400 in TPBS containing 1% normal goat serum overnight at room temperature. Cells were then dehydrated, cleared in xylol and permanently mounted in DPX (Fluka, Neu-Ulm, Germany). Specimens were examined using a Leica TCS-NT laser scanning confocal microscope. Cyanine 3 was imaged with 568-nm excitation and 570- to 630-nm bandpass emission filters.


For comparison of the functional properties of the N40D and wild-type µ-opioid receptor, both receptors were HA-epitope tagged and stable expressed in HEK293 cells. After transfection 5 stable receptor expressing clones were isolated from each of the transfected receptor types. Receptor expression was monitored by ligand-binding experiments. Saturation-binding experiments revealed no marked differences between N40D and hMOR expressing cells with respect to their affinities (Kd) to [3H]DAMGO (2.55 ± 0.2 nm and 2.26 ± 0.3 nm for N40D and hMOR, respectively). However, differences were determined in the number of [3H]DAMGO binding sites (Bmax) between N40D (627 ± 138 fmol/mg protein) and hMOR (4777 ± 842 fmol/mg protein). The determined average Bmax values were calculated from 5 stable N40D expressing clones (797, 322, 465, 1083 and 468 fmol/mg protein) and from 5 stable hMOR expressing clones (1834, 4396, 6431, 4823 and 6401 fmol/mg protein) indicating a general higher expression of hMOR compared to N40D in HEK293 cells for all the clones tested. The observed differences between N40D and hMOR cell surface receptor expression have been confirmed by using ELISA technique as previously described (Koch et al. 2003).

Effect of the N40D polymorphism on the ligand-binding affinity of the μ-opioid receptor in HEK293 cells

We performed radioligand-binding assays to determine whether the N40D mutation changes the receptor's ability to bind opioid ligands in HEK293 cells. According to previous studies demonstrating differences in the agonist-affinity and/or potency for both receptors, we used β-endorphin, morphine and morphine-6-glucuronide in our experiments. The competition for the binding of morphine, morphine-6-glucuronide or β-endorphin with the radiolabelled ligand ([3H]DAMGO) was tested in membranes of HEK293 cells stably expressing N40D-variant or hMOR receptor. Our results indicate that the N40D mutation did not alter the overall profile of the ligand binding for the tested agonists (Fig. 1). The IC50-values calculated from the binding curves for all ligands are presented in Table 1.

Figure 1.

Binding affinities of morphine, morphine-6-glucuronide and β-endorphin to the hMOR and N40D opioid receptor. Membrane preparations were used in the binding experiments to compete the [3H]DAMGO binding. The values represent means ± SEM from three separate experiments performed in triplicate.

Table 1.  Binding affinities and dose-responses for cAMP reduction of opioids on hMOR and N40D receptors expressed in HEK293 cells
  1. Competition binding experiments were performed on transfected HEK293 cell membranes expressing hMOR and N40D receptors, using variable concentrations of unlabelled ligands to compete [3H]DAMGO. IC50 values were calculated from the data presented in Fig. 1. Dose–responses for opioid-stimulated cAMP reduction at hMOR and N40D receptors were performed as described under ‘Materials and Methods’. EC50 values were calculated from the data presented in Fig. 2. Data are means ± SEM from three different experiments performed in triplicate.

IC50 (nm)26.85 ± 11.8923.50 ± 9.2932.67 ± 4.4650.67 ± 18.2715.20 ± 6.8411.70 ± 3.39
EC50 (nm)0.75 ± 0.081.10 ± 0.291.70 ± 0.141.76 ± 0.2413.70 ± 2.4514.30 ± 2.75

Effect of the N40D polymorphism on the agonist potency of β-endorphin, morphine and morphine-6-glucuronide for activating the μ-opioid receptor

After stimulation the µ-opioid receptor inhibits adenylate cyclase by an inhibitory G protein, resulting in a decrease of the intracellular cAMP concentration. To examine the effect of N40D mutation on the agonist-mediated reduction of cAMP level, we analysed the effect of the agonists morphine, morphine-6-glucuronide and β-endorphin on the inhibition of forskolin-stimulated cAMP accumulation in HEK293 cells. Therefore, HEK293 cells stably expressing the corresponding receptor type were incubated for 15 min at 37°C with variable agonist concentrations. The maximum reduction of forskolin-stimulated cAMP level has been defined as 100%. Our results revealed that the used agonists activated both receptors with similar potency (Fig. 2), suggesting that the N40D polymorphism has no influence on agonist-induced receptor response. The EC50 values calculated from the dose–response curves for all ligands are presented in Table 1.

Figure 2.

Morphine, morphine-6-glucuronide and β-endorphin dose–response relations of hMOR and N40D receptors expressed in HEK293 cells. Receptor response was measured as dose-dependent inhibition of forskolin-induced cAMP accumulation. Data are presented as means ± SEM from three different experiments performed in triplicate.

Effect of N40D mutation on the agonist-induced μ-opioid receptor desensitization

Next, we determined the time courses of the agonist-induced desensitization of hMOR and N40D receptors in HEK293 cells. Receptor desensitization was measured as the decreased ability of the agonist to inhibit forskolin-stimulated adenylate cyclase after prolonged agonist pre-treatment. Therefore, HEK293 cells stably expressing the N40D-variant receptor or hMOR receptor were pre-treated with 1 µm agonist (morphine, morphine-6-glucuronide or β-endorphin) for various time intervals (0, 0.5, 1, 2, 4, 6 h), followed by determination of the agonist-induced inhibition of forskolin-stimulated adenylate cyclase. Forskolin treatment resulted in a 8-fold increase of intracellular cAMP level (up to 8 pmol) as compared with untreated HEK293 cells. For both hMOR and N40D receptor expressing HEK293 cells absolute inhibition levels of cAMP accumulation after morphine, morphine-6-glucuronide and β-endorphin treatment were 43 ± 4%, 45 ± 7% and 56 ± 5%, respectively. For each receptor type and for each agonist used, the maximum agonist-induced inhibition of cAMP accumulation without agonist pre-treatment has been defined as 100%. Prolonged agonist pre-incubation (4 h) resulted in a nearly complete desensitization of both receptors (Fig. 3). The morphine- and morphine-6-glucuronide-mediated desensitization of both receptors were nearly identical, whereas the time courses of β-endorphin-induced desensitization showed minor differences which were not statistically significant (Fig. 3). These data demonstrate that N40D and hMOR do not differ in their time courses of agonist-induced desensitization.

Figure 3.

Agonist-selective desensitization of µ-opioid receptor hMOR and N40D variant. For the analysis of receptor desensitization, HEK293 cells expressing hMOR and N40D receptor were exposed to 1 µm morphine, morphine-6-glucuronide or β-endorphin for the indicated time-periods. After washing, the cells were treated with 25 µm forskolin or 25 µm forskolin plus 1 µm of the respective agonist, and cAMP levels were determined as described under ‘Materials and Methods’. The values represent means ± SEM from three different measurements performed in triplicate.

Effect of N40D mutation on the agonist-induced μ-opioid receptor endocytosis and recycling

From the observed differences in the receptor expression rates between N40D and hMOR arises the question whether both receptors differ in their agonist-induced internalization and resensitization rate. Therefore we incubated HEK293 cells stably expressing the N-terminal HA-tagged N40D or hMOR receptors with anti-HA-specific antibodies at 4°C to label cell surface receptors. Then the cells were treated with the agonist (morphine, morphine-6-glucuronide, β-endorphin or DAMGO) for 30 min at 37°C. Control cells were incubated at 37°C for 30 min without agonists to examine whether these receptors may undergo constitutive internalization. The cells were subsequently fixed and permeabilized, and bound anti-HA antibody was immunofluorescently detected (Fig. 4). At low temperature (4°C), N40D and hMOR receptors were located at plasma membrane as assessed by confocal microscopy. After 30 min incubation at 37°C without agonists both receptors exhibited a slight constitutive internalization (Fig. 4). After exposure to morphine or morphine-6-glucuronide, neither the N40D variant receptors nor the hMOR receptors revealed additional endocytosis, whereas after 30 min DAMGO treatment, both receptors showed robust endocytosis. β-Endorphin was also able to induce endocytosis of both receptors but less than DAMGO. These findings indicate that N40D and wild-type µ-opioid receptor were internalized after β-endorphin and DAMGO incubation but not after morphine or morphine-6-glucuronide treatment.

Figure 4.

Agonist-selective endocytosis of hMOR and N40D receptor variant. HEK293 cells expressing HA epitope tagged hMOR or N40D receptors were incubated with anti-HA antibodies at 4°C to label cell surface receptors (control 4°C). The cells were then either not treated (control 37°C) or treated with 1 µm morphine, morphine-6-glucuronide, β-endorphin or DAMGO for 30 min at 37°C. The cells were subsequently fixed, fluorescently labelled and examined by confocal microscopy. Representative results from one of four independent experiments are shown.

In addition we tested whether N40D mutation would affect the resensitization rate of the µ-opioid receptor. Receptor resensitization was measured as increased ability of the previously desensitized receptor to inhibit forskolin-stimulated adenylate cyclase activity during a 40 min drug-free interval. Thus, HEK293 cells expressing N40D or hMOR were pre-treated with 1 µm DAMGO for 6 h, the medium was removed, and after an additional drug-free interval of 0, 10, 20, 30 or 40 min, agonist-induced inhibition of forskolin-stimulated adenylate cyclase was determined. As depicted in Fig. 5, no statistically significant differences in the resensitization rate after DAMGO treatment was observed between N40D and hMOR in HEK293 cells.

Figure 5.

Resensitization of hMOR and N40D receptor variant. Transfected HEK293 cells were cultured in the presence of 1 µm DAMGO for 6 h, which results in complete desensitization of both receptor types. Medium was removed and cells were washed three times with phosphate-buffered saline followed by an additional incubation for 10, 20, 30 and 40 min in agonist-free medium. Cells were then treated with 25 µm forskolin or 25 µm forskolin plus 1 µm DAMGO for 15 min and inhibition of intracellular cAMP levels were determined. Treatment with forskolin resulted in an eight-fold increase of intracellular cAMP (up to 8 pmol) compared to untreated HEK293 cells. Values represent means ± SEM of triplicate determinations of three independent experiments.


Large-scale sequencing of the human µ-opioid receptor (hMOR) gene has revealed polymorphic mutations that occur within the coding region (Hoehe et al. 2000). One of the most extensively investigated polymorphisms is the mutation N40D in the extracellular N-terminal region of the human µ-opioid receptor. This N40D mutation with a high allelic frequency has been suggested to be involved in alcohol and opioid dependence. However, despite a multitude of studies and extensive effort, no consistent data for a significant association between N40D polymorphism and substance dependence could be determined. Moreover, in vitro studies have produced mixed results concerning the β-endorphin binding and activity at the N40D receptor type. In AV-12 cells β-endorphin has been shown to bind three times more tightly at the N40D receptor than at the wild-type receptor (Bond et al. 1998), whereas in COS-7 cells no differences in the β-endorphin-binding affinity between both receptors were detected (Befort et al. 2001). Recently, it has been demonstrated that N40D mutation reduces the potency of morphine-6-glucuronide in humans (Lotsch et al. 2002a). Accumulation of the active metabolite morphine-6-glucuronide (M6G) causes toxic effects in patients with renal dysfunction after chronic morphine treatment. Interestingly, patients with renal failure who are homozygous carriers of the N40D mutation tolerated morphine and high plasma M6G concentrations (Lotsch et al. 2002b). Thus, it has been speculated that N40D mutation of the µ-opioid receptor is among the protective factors against M6G-related opioid toxicity (Lotsch et al. 2002b).

In the present study we have investigated whether the mutation N40D alters functional properties (ligand binding, functional coupling, receptor desensitization, internalization and resensitization) of the µ-opioid receptor expressed in the mammalian cell line HEK293. First we analysed the binding affinity of the agonists morphine, morphine-6-glucuronide and β-endorphin in membranes of HEK293 cells stably expressing hMOR or N40D variant receptor. For both receptor types the tested ligands revealed similar binding affinities, indicating that N40D mutation might not affect the conformation of the N-terminal extracellular region. The observed similar binding affinity of β-endorphin to both receptors in HEK293 cells is in good agreement with previously published findings in COS-7 cells (Befort et al. 2001) but is in direct contrast to the finding in hamster subcutaneous tumour AV-12 cells (Bond et al. 1998). Such discrepancy could stem from the fact that different cell lines were used in the studies.

Next we investigated whether the µ-opioid receptor-mediated inhibition of adenylate cyclase activity is affected by the N40D mutation. After agonist binding, the µ-opioid receptor couples to an inhibitory G protein (Gi), which leads to an adenylate cyclase inhibition and to a decrease in the intracellular cAMP level. EC50 values for morphine, morphine-6-glucuronide and β-endorphin are comparable for the wild-type and the N40D receptor, indicating that the tested agonists activated both receptor types with similar potency.

Given that we could not detect any differences concerning ligand binding and dose–responses, we next examined the time-courses of agonist-induced receptor desensitization for both wild-type and N40D receptor. After chronic agonist treatment the µ-opioid receptor is rapidly phosphorylated and uncoupled from G proteins. Potential phosphorylation sites are serine and threonine residues in the C-terminal region and in the third intracellular loop of the receptor, which are substrates for the GRKs (Lefkowitz et al. 1993; Lohse 1993; Capeyrou et al. 1997). This phosphorylation changes the receptor conformation, facilitating β-arrestin binding and subsequent uncoupling from G proteins (Ferguson et al. 1996). For the present desensitization studies HEK293 cells stably expressing hMOR or N40D variant receptor were pre-incubated with morphine, morphine-6-glucuronide or β-endorphin for different time intervals. For all three agonists no significant differences in the desensitization rates between hMOR and N40D receptors were detected in HEK293 cells. However, during the first 2 h of agonist treatment β-endorphin revealed a slower time course of desensitization for both hMOR and N40D receptors compared with morphine and morphine-6-glucuronide. After 1 h of agonist treatment of both hMOR and N40D receptors, maximum inhibition of intracellular cAMP accumulation was ∼28%, ∼30% and ∼60% for morphine, morphine-6-glucuronide and β-endorphin, respectively. The slower β-endorphin-induced receptor desensitization might be explained by a rapid receptor internalization and recycling/reactivation (Koch et al. 1998).

It is well demonstrated that morphine fails to induce µ-opioid receptor endocytosis (Koch et al. 1998; Whistler and von Zastrow 1998), but morphine-6-glucuronide-induced endocytosis has not been analysed so far. Therefore, we investigated whether hMOR and N40D receptor differ in the agonist-selectivity of receptor endocytosis. Agonist-induced receptor phosphorylation and subsequent binding of β-arrestin is an initial step of receptor internalization (Bohm et al. 1997; Zhang et al. 1998). Morphine, which does not lead to a robust receptor phosphorylation (Koch et al. 2001), seems to trap the receptor in a conformation that prevents it from β-arrestin binding and endocytosis (Whistler and von Zastrow 1998).

For confocal analyses of µ-opioid receptor endocytosis HEK293 cells expressing the corresponding receptor type were pre-incubated with the primary antibody and next stimulated with morphine, morphine-6-glucuronide, β-endorphin or DAMGO. After morphine and morphine-6-glucuronide treatment both receptor types (hMOR and N40D) revealed stable membranal localization (Fig. 4), indicating that both agonists failed to induce receptor endocytosis. On the other hand, after incubation with β-endorphin and DAMGO both receptor types indicate robust endocytosis. Taken together, our results strongly confirm that in HEK293 cells wild-type µ-opioid receptor and N40D receptor variant do not differ in their agonist-selectivity of endocytosis.

After endocytosis, receptors will be dephosphorylated, separated from bound ligand and become subject to endosomal sorting either by passing through a process of receptor degradation or being recycled to the cell membrane. The observed differences between hMOR and N40D in the receptor expression rate in HEK293 cells might be due to distinct recycling/resensitization rates of the receptor after endocytosis. We therefore tested the time courses of receptor resensitization after complete receptor desensitization. Consistent with the results from the desensitization studies we obtained no differences in the resensitization rate of hMOR and N40D receptors (Fig. 5). Another possible explanation for the different receptor expression rates might be that N40D mutation results in the loss of one of five potential N-linked glycosylation sites of the µ-opioid receptor (Mestek et al. 1995) which might affect the transport and/or the incorporation of µ-opioid receptor into the plasma membrane. However, in previous studies no differences in the receptor expression rates were determined in COS cells transiently expressing hMOR or N40D receptors (Befort et al. 2001). Moreover, truncation of the N-terminal tail harbouring the glycosylation sites does not prevent expression of the µ-opioid receptor (Wang et al. 1993; Surratt et al. 1994).

In summary, our results indicate that N40D mutation has no effect on µ-opioid receptor function or ligand binding after expression in HEK293 cells. Thus, our in vitro findings do not provide any mechanistic basis for the previously described effect that the N40D mutation in the µ-opioid receptor reduces the potency of morphine-6-glucuronide in humans (Lotsch et al. 2002b). Remarkably, six control patients, five heterozygous patients, but only one homozygous N40D patient were tested in that study. Therefore, it cannot be ruled out that other genetic variations in the genome of the homozygous patient might be the reason for the observed effects on the morphine-6-glucuronide potency. Moreover, due to the lack of functional differences between hMOR and N40D receptors, it remains a moot point whether the N40D mutation in the human µ-opioid receptor plays a role in the development of opioid addiction and dependence.


We thank Evelyn Kahl, Sabrina Sattelkau and Michaela Böx for excellent technical assistance.