Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Opioid receptors are seven transmembrane G-protein-coupled receptors. They are targets for opiate drugs and endogenous opioid neuropeptides that mediate analgesic, euphoria, and other important central and peripheral actions (Simon, 1991; Herz, 1993). Like other G-protein-coupled receptors, such as dopamine and adrenergic receptors, opioid receptors bind their ligands present in the extracellular medium and couple their binding to the activation of intracellular G-proteins (Evans et al., 1992; Chen et al., 1993; Wang et al., 1994). Three subtypes of opioid receptors (µ, , ) have been characterized and cloned (Evans et al., 1992; Chen et al., 1993; Wang et al., 1993, 1994) and among them, the µ-subtype mainly mediates analgesic and euphoria effects (Raynor et al., 1994). Thus, understanding the molecular mechanisms toward the interaction between opiate drugs, opioid peptide, and µ-receptor is of great importance.

Further understanding of the molecular mechanism for opioid receptor function depends on the knowledge of the molecular structure of this receptor. Early studies on receptor chimera and mutant receptors have provided a great deal of information about functionally important amino acid residues that are responsible for µ-receptor binding. Most of these studies revealed that Asp114, Asp147, and His297 within the transmembrane (TM) domains on the rat µ-receptor, as well as His223 in the putative second extracellular loop, are essential for µ-receptor binding (Surratt et al., 1994; Shahrestanifar et al., 1996; Spivak et al., 1997; Bot et al., 1998). In addition, studies found that point mutations of Trp318, His319, or Tyr326 in TM VII and Ile196 in TM IV resulted in a decreased affinity for a wide spectrum of µ-selective agonists (Mansour et al., 1997; Xu et al., 1999). DNA cloning of the sequences of rodent and human µ-opioid receptors revealed that there are many putative cysteine residues in the primary structure of the µ-opioid receptor (Wang et al., 1993, 1994). This suggests that sulfhydryl reagents, such as dithiothreitol and N-ethylmaleimide may be used to probe receptor structure and function. Evidence has been provided that sulfhydryl reagents, such as N-ethylmaleimide, iodoacetamide, and p-hydroxymercuribenzoate inhibit ligand binding to µ-opioid receptors (Pasternak et al., 1975; Simon and Groth, 1975; Shahrestanifar et al., 1996; Gaibelet et al., 1997). Therefore, residues that react with these reagents may be involved in the receptor-binding site or related structures. It is known that other G-protein-coupled receptors were susceptible to sulfhydryl reagents, such as D₁ and D₂ dopamine receptors (Sidhu et al., 1986; Chazot and Strange, 1992), and ₁- and ₂-adrenergic receptors (Reader et al., 1986). The sensitive cysteine residue in the TM III of the human D₂ dopamine receptor has been identified. The sensitive cysteine reacts with sulfhydryl reagents and results in the inhibition of the human D₂ dopamine receptor binding (Javitch et al., 1994). For the µ-opioid receptor, the possibility that cysteines are located within or near the ligand-binding site is supported by a "ligand protection" study, in which preincubation of the receptor with opioid ligands protected against the inhibitory effect of sulfhydryl reagents on µ-opioid receptor binding (Shahrestanifar et al., 1996). Another interesting fact is that analogs of leu-enkephalin and morphine, S-activated sulfhydryl morphine derivatives, activate opioid receptors persistently after extensive washing, implying that they activate the receptor via the formation of disulfide bonds with cysteines in or near the receptor-binding site (Kodama et al., 1989; Kanematsu et al., 1990). Results from a recent study of µ-/-receptor chimeras predict that the sensitive cysteine residues for µ-opioid receptor binding may be located between TM III and TM V (Shahrestanifar et al., 1996). However, site-directed mutagenesis of the cysteines in the µ-opioid receptor showed that none of the cysteine substitutions affected receptor binding or sensitivity toward N-ethylmaleimide (Shahrestanifar et al., 1996). This result could be partly due to the nonspecific effect of N-ethylmaleimide because it is known that N-ethylmaleimide not only reacts with the sulfhydryl group of cysteine but also alkylates the imidazole group of histidine (Mullikin-Kilpatrick et al., 1983).

In this study, we used a set of specific sulfhydryl reagents to clarify the involvement of cysteine residues in the effect of sulfhydryl reagents on µ-opioid receptor binding. The specific sulfhydryl reagents are known as methanethiosulfonate (MTS) derivatives: ethylammonium MTS, CH₃SO₂SCH₂CH₂NH₃ 71 (MTSEA 71); trimethylammonium MTS, CH₃SO₂SCH₂CH₂N(CH₃)₃⁺ (MTSET⁺); and ethylsulfonate MTS, CH₃SO₂SCH₂CH₂SO₃ 71 (MTSES 71). The MTS derivatives are small, charged, highly water-soluble, sulfhydryl-specific reagents (Kenyon and Bruice, 1977; Akabas et al., 1992; Xu and Akabas, 1993; Stauffer and Karlin et al., 1994) that form disulfide bonds with water-accessible cysteines. The goals of this present study were 1) to examine the effect of MTS reagents on µ-opioid receptor binding and 2) to identify the sensitive cysteines that react with the MTS reagents and affect µ-opioid receptor-ligand interaction. In this study, we observed that MTS displayed irreversible inhibitory effects on human µ-opioid receptor binding of [³H][D-Ala²,N-MePhe⁴,Gly-ol⁵]-enkephalin (DAMGO). Preincubation of the receptor with DAMGO or naloxone effectively prevented the inhibition. Among eight mutated receptors, four mutants (C161S, C192S, C237S, and C332S) were significantly less sensitive to MTSEA treatment than the wild-type receptor, indicating that these cysteine residues cause the inhibitory effect of sulfhydryl reagents on µ-receptor binding.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Radioligand-Binding Assay. Chinese hamster ovary (CHO) cells stably expressing human µ-opioid receptor (HµCHO) were plated at 80% confluence in 150-mm plates, grown for 24 h in F-12 medium containing 10% fetal calf serum, 100 U/ml penicillin, and 100 µg/ml streptomycin. For receptor-binding assays, cells from a 150-mm plate were harvested and suspended in 4.8 ml 50 mM Tris-HCl. Binding of [³H]DAMGO to the receptor was assayed by the procedure reported previously (Wang et al., 1993). Briefly, a 400-µl cell suspension was incubated at room temperature for 90 min with different concentrations of [³H]DAMGO (37 Ci/mmol; NEN, Boston, MA) plus other reagents as indicated in a final volume of 500 µl. Each sample was measured in duplicate. For saturation binding, six concentrations of [³H]DAMGO ranged from 0.1 to 12 nM. The binding assay was terminated by filtration through Whatman GF/B filters (FP-100; Brandel, Bethesda, MD) with a Brandel filtration device. The filters were washed three times with 4 ml of 50 mM Tris-HCl, pH 7.4. Radioactivity was measured by a Beckman liquid scintillation counter at 40% efficiency. Specific binding was determined as total binding minus nonspecific binding in the presence of 1 µM nonradioactive naloxone. Data were analyzed by nonlinear regression analysis with the Prism program (GraphPad, San Diego, CA).

Treatment of HµCHO with MTS Reagents and Protection Experiments. HµCHO cells from 150-mm plates were suspended in 1.2 ml of 50 mM Tris-HCl buffer, pH 7.4. Aliquots (100 µl) of cell suspension were pretreated with varying concentrations of MTS reagents at room temperature for 20 min. Aliquots were then diluted to a final volume of 400 µl with 50 mM Tris buffer, and they were used for the [³H]DAMGO binding (1 nM) assay as described above. The concentrations of MTS reagents ranged from 10 µM to 2 mM. Protection experiments were performed by preincubation of the cell suspension with various concentrations of DAMGO (10-600 nM) or naloxone (0.1-10 µM) for 10 min at room temperature, followed by treatment with 0.5 mM MTSEA as described above. Cells were washed three times by centrifugation with 50 mM Tris-HCl buffer, pH 7.4, before performing [³H]DAMGO receptor-binding assays.

Site-Directed Mutagenesis. The human µ-opioid receptor cDNA was cloned into Bluescript vector at the EcoRI site, then subcloned into the expression vector pcDNA1/Amp with coding region Small and XhoI sites, and vector EcoRV and XhoI sites. The single-stranded cDNA was prepared with helper phage M13K07 according to Surratt et al. (1994). Oligonucleotides were synthesized to generate the appropriate serine mutation or alanine mutation, and the desired mutations were obtained with Amersham in vitro mutagenesis kits, following the manufacturer's manual. Each mutation site was confirmed by DNA sequencing. The DNA fragments containing mutation sites were reintroduced into wild-type µ-receptor cDNA in pcDNA1/Amp. Mutants are named by the single letter code for the wild-type amino acid, followed by the position number of the residue and the substituted residue (i.e., C161S refers to substitution of a cysteine at position 161 by serine).

Transient Expression of Mutant Receptors in COS7 Cells. COS7 cells were plated in 150-mm plates at 30% confluence, and grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1% L-glutamine, and 100 U/ml penicillin and 100 µg/ml streptomycin at 37°C and 5% CO₂ for 24 h. Then cells were transfected with 15 µg of wild-type or mutant plasmid cDNA with the calcium phosphate coprecipitation method. The medium was replaced with fresh medium 24 h after transfection. Then 72 h after transfection, the cells were harvested for binding assays according to the procedures described above.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of MTS Reagents on µ-Opioid Receptor Binding of [³H]DAMGO and Protection by Opioid Ligands. CHO cells expressing the human µ-opioid receptor (HµCHO) were incubated with three kinds of MTS reagents before evaluating the binding affinity of [³H]DAMGO. The three MTS reagents (MTSEA, MTSET, MTSES) all displayed dose-dependent inhibitory effects on µ-opioid receptor-specific binding of 1 nM [³H]DAMGO. The maximal inhibition was >90% of specific [³H]DAMGO binding compared with non-MTS-treated cells (Fig. 1). The IC₅₀ values of MTSEA, MTSET, and MTSES inhibition on the receptor binding were 0.12, 0.15, and 13 mM, respectively. We further examined the mechanism of the inhibitory effect of MTSEA, the most potent of the three MTS reagents, on [³H]DAMGO binding to the µ-opioid receptor. The B_max and the K_d values were found to be 2.06 ± 0.12 pmol/mg protein and 0.70 ± 0.18 nM for non-MTSEA-treated cells, and 0.88 ± 0.03 pmol/mg protein and 1.86 ± 0.26 nM for MTSEA-treated cells, respectively. These data suggest that the cells pretreated with 1 mM MTSEA lost a significant number of receptor-binding sites (P < .01, t test) and affinity for the µ-opioid receptor agonist (P < .01, t test) (Fig. 2). In the presence of either µ-opioid agonist or antagonist, the receptor was protected against the inhibitory effects of MTSEA (Fig. 3) and the protective effects become greater as the ligand concentrations increased. At high ligand concentrations of DAMGO (0.6 µM) or naloxone (10 µM), most of the receptor binding can be protected from the MTSEA treatment. These data indicate that the sulfhydryl reaction might occur within or near the receptor-binding pocket.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effects of MTS reagents on specific [3H]DAMGO binding to HµCHO cells. Dissociated cells in 50 mM Tris-HCl buffer were treated with various concentrations of MTS reagents ranging from 10 µM to 2 mM for 20 min. A binding assay with 1 nM [3H]DAMGO was performed as described in Materials and Methods. Results of three independent experiments are shown with a mean ± S.E.; each sample was tested in duplicate. IC50 values were obtained by nonlinear regression analysis (Prism program).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Effects of MTSEA on specific [3H]DAMGO binding to HµCHO cells. A, dissociated CHO cells in 50 mM Tris-HCl buffer were incubated with different concentrations of [3H]DAMGO ranging from 0.1 to 12 nM with or without (control) pretreatment with 0.1 mM MTSEA. The data from a representative experiment are shown, each was determined in duplicate. B, Scatchard plot analysis. Kd and Bmax values were calculated by nonlinear regression analysis with the Prism program. Non-MTSEA-treated cells have a Kd of 0.70 ± 0.18 nM and a Bmax of 2.06 ± 0.12 pmol/mg protein, and MTSEA-treated cells have a Kd of 1.86 ± 0.26 nM and a Bmax of 0.88 ± 0.03 pmol/mg protein. Both changes of Kd and Bmax between MTSEA-treated and nontreated cells were significant (P < .01, Student's t test).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   DAMGO and naloxone protection against the inhibitory effect of MTSEA on µ-opioid receptor binding of [3H]DAMGO. HµCHO cells were preincubated with varying concentrations of DAMGO ranging from 10 to 600 nM, or naloxone ranging from 0.1 to10 µM, for 10 min at room temperature, followed by the addition of 0.5 mM MTSEA for 20 min at room temperature. Cells were washed three times with Tris-HCl buffer by centrifugation. The cells were resuspended in Tris-HCl buffer and their ability to bind 1 nM [3H]DAMGO was assayed as described in Materials and Methods. The percentage of specific binding was defined as specific binding with pretreatment of MTSEA divided by specific binding without pretreatment of MTSEA. The dotted line represents the specific binding after 0.5 mM MTSEA treatment in the absence of protective drugs. The results were the average for three independent experiments, each with duplicate determination.

Binding of [³H]DAMGO to Mutated HµCHO. According to the amino acid sequence derived from a cloned cDNA of the HµCHO, there are eight cysteine residues located within the seven TM domains (Fig. 4). To identify the MTS reagent-sensitive cysteine residues, we separately mutated each of the eight cysteines located in TM domains into serine, transiently expressed the mutant receptors in COS7 cells and examined their specific [³H]DAMGO binding. Each mutant receptor exhibited the ability to recognize the opioid ligand with similar affinity to that of the wild type (Table 1), indicating that substituting cysteines with serines did not cause a drastic change in receptor structure. B_max for each mutant receptor was 57.5 to 109.0% of that of the wild type (Table 1).

View larger version (119K): [in this window] [in a new window]   Fig. 4.   Schematic representation of structures of the human µ-opioid receptor. The receptor sequence is shown with the single-letter codes for amino acid residues. The seven putative TM domains (cylinders) are connected by extra- or intracellular loops, and the amino and carboxyl termini. The receptor's five potential N-linked glycosylation sites are marked with forks. The eight mutated cysteines are indicated with arrows. The number showed the location of the cysteine residue.

Effects of MTSEA on Mutant Receptor Binding of [³H]DAMGO COS7 cells transiently expressing a mutant receptor were pretreated with 0.1 mM MTSEA before testing in the binding assays. MTSEA irreversibly inhibited the specific binding of 1 nM [³H]DAMGO to both wild-type and four of the mutant receptors (C81S, C253S, C294S, and C323S) by ~60% (Fig. 5). However, the other four mutant receptors were significantly resistant to MTSEA inhibitory effects compared with the wild-type receptor. MTSEA inhibited 1 nM [³H]DAMGO binding of C161S, C192S, and C237S by only 15% and C332 by 29%, implying that these four cysteines are sensitive cysteines.

View larger version (46K): [in this window] [in a new window]   Fig. 5.   Effects of MTSEA on [3H]DAMGO binding in mutated human µ-opioid receptors transiently expressed in COS7 cells. Cells transiently expressing the mutant receptors were assayed for their binding of 1 nM [3H]DAMGO with or without pretreatment with 0.1 mM MTSEA for 20 min at room temperature. The percentage of specific binding was defined as specific binding after MTSEA treatment divided by specific binding without MTSEA. Each mutant was assessed by three independent experiments. An * indicated that the MTSEA-mediated inhibitory effect on mutant receptor binding was significantly decreased (P < .05, Student's t test) compared with that of the wild-type receptor.

Binding Properties of Mutant Receptors with Cysteines Substitution by Alanine. The effect of MTSEA on cysteine-to-serine mutant receptors suggested that C81, C253, C294, and C323 were not sensitive to the MTSEA inhibitory effect. This result implies that the locations of these four cysteines are not near the receptor-binding site and that the changes at or near these locations would not interfere with the receptor's ability to recognize the ligand. To further test this concept, we mutated these four (C81, C253, C294, and C323) cysteines to a nonpolar amino acid, alanine. The binding properties of the alanine-substituted receptors and the wild-type receptors were similar (Table 2). We also substituted four sulfhydryl-sensitive cysteines with alanine (C161A, C192A, C237A, and C332A) and did the corresponding binding experiment on these mutants (Table 3). Alanine substitution of C161, C192, C237, or C332 did not alter the binding capacity of the µ-receptor. This result supports the notion that these cysteines, even though probably located near or at binding site, are not essential for the receptor binding. The inhibitory effect of sulfhydryl reagents on the receptor binding is probably due to steric hindrance by the large moieties from these reagents attached to the sulfhydryl groups of the cysteine residues.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of MTS reagent treatment on µ-opioid receptor binding are consistent with previous studies of the effects sulfhydryl reagents on µ-receptor binding (Pasternak et al., 1975; Simon and Groth, 1975; Shahrestanifar et al., 1996; Gaibelet et al., 1997). Compared with other sulfhydryl reagents such as N-ethylmaleimide, MTS reagents are more specific for water-accessible cysteine residues (Kenyon and Bruice, 1977; Akabas et al., 1992; Xu and Akabas, 1993; Stauffer and Karlin et al., 1994). When hydrophilic MTS reagents are added extracellularly, they interact specifically with all water-accessible cysteine residues by forming mixed disulfide bonds. In the present study, DAMGO binding to the human µ-opioid receptor was irreversibly inhibited by MTS reagents, suggesting that at least some of the cysteines on the water-accessible side of the receptor are critical for the sulfhydryl reactions that impair receptor binding. Preincubation of the receptor with either an agonist or an antagonist prevented MTS-mediated inhibition of the µ-opioid receptor. This finding indicates that the reaction of cysteine residues with MTS reagents has an impact on receptor binding either directly at the receptor binding site or at a location important for the access to the binding site.

Both positively charged MTS reagents, MTSEA, MTSET, and the negatively charged MTSES prevent DAMGO from binding to the human µ-opioid receptor (Fig. 1), indicating that the different potency of MTS reagents on µ-receptor binding is not related to the charge carried by each sulfhydryl compound. In contrast, ligand binding to the dopamine D₂ receptor was only affected by the positively charged MTS (Javitch et al., 1994). For D₂ receptors, the result correlates well with the high affinity of positively charged dopamine and dopamine antagonists to this receptor (Javitch et al., 1994). This means that either the electrostatic potential near the sensitive cysteine is very negative or the diffusion pathway from the extracellular medium to the sensitive cysteine is cation-selective. Similarly, only the positively charged MTS reacts with the cysteines lining the acetylcholine receptor channel, consistent with the cation-selective properties of the channel (Akabas et al., 1992; Stauffer and Karlin, 1994). Although the ligand requirement of the µ-opioid receptor with regard to the charge is similar to the D₂ receptor, the µ-opioid receptor did not display a cation-selective pattern for MTS. This indicates that other mechanisms must be involved in µ-opioid receptor sensitivity for MTS reagents, such as steric hindrance by the large moieties transferred to the sulfhydryl groups of cysteine residues by MTS reagents.

There were several reasons to focus on the eight cysteines located in the µ-opioid receptor TM domains. Previous structural and functional studies showed that the large extracellular N-terminal segment is not necessary for µ-opioid receptor binding (Surratt et al., 1994). The two extracellular loop cysteines, which are highly conserved among the G-protein-coupled receptors, are likely to form a disulfide bond (Karnik et al., 1988; Dohlman et al., 1990; Karnik and Khorana 1990; Brandt et al., 1999) that would not react with sulfhydryl reagents. Intracellular cysteines are expected to be inaccessible to water-soluble reagents. Furthermore, the ligand-binding sites in most G-protein-coupled receptors are formed within the membrane-spanning segments (Strader et al., 1989; Savarese and Fraser, 1992; Samama et al., 1993; Strader et al., 1994), although peptide ligand may need additional extracellular domains to promote binding. Therefore, the cysteines located in the TM domains are reasonable targets to study with the hydrophilic, sulfhydryl-specific reagents.

We identified the sulfhydryl-sensitive cysteines involved in µ-opioid receptor binding by replacing cysteine residues with serine by site-directed mutagenesis. Mutagenesis is advantageous because any residue can be altered; however, mutation of residues outside of a binding site could alter binding by long-range perturbation of the receptor structure, thus confounding the identification of binding site residues. In our study, both cysteine and serine are polar, noncharged amino acid residues, and because serine is the most conservative substitution for cysteine among the naturally occurring amino acids, substitution is the least destructive for the structure of the receptors. The mild effects on ligand binding by the mutations (Table 1) indicated that the sulfhydryl groups of these cysteines were not important for µ-opioid receptor binding.

Our data showed that MTS reagents did not display their typical inhibitory effects in the presence of C161S, C192S, C237S, and C332S, indicating the corresponding cysteines are critical residues for the inhibitory effect of MTS reagents on µ-opioid receptor binding. There are two possible explanations for this result. First, these four cysteines are part of the receptor-binding pocket and the sulfhydryl reaction with these cysteines cause a significant loss of binding sites. Therefore, mutations of any of these four cysteines can minimize the MTS inhibitory effect on receptor binding, which results in a partial recovery of receptor binding. A second explanation is that these four cysteines are located near the binding site. Interactions of MTS with these cysteines would then block access of the ligand to the binding site of the receptor, which is consistent with the finding that mutations have no significant effect on receptor binding. We also cannot exclude other possibilities, such as conformational changes; MTS reagents may disrupt the conformational changes that normally occurred during receptor activation and thus affect receptor binding.

The sensitive cysteines C161S, C192S, C237S, and C332S are located in TM III, TM IV, TM V, and TM VII. Previous work on µ-/-receptor chimeras suggests that cysteines located between TM II and TM V contribute to the inhibitory effect of N-ethylmaleimide on µ-opioid receptor binding (Shahrestanifar et al., 1996). In other investigations with µ-/-receptor chimeras, the segments from TM VI to TM VII of the µ-opioid receptor were replaced by the corresponding part of the -receptor (Chen et al., 1995; Xue et al., 1995). These studies found that the specific binding of the receptor chimera to DAMGO, sufentanil, and morphine all decreased greatly. The locations of sensitive cysteine residues identified in our current study generally agree with the results from these receptor chimera studies. It has been reported that C81 and C332 may be responsible for the inhibition by N-ethylmaleimide of agonist binding to the µ-opioid receptor, but C161 is not involved in this effect (Gaibelet et al., 1997). However, the effects of other cysteines in the TM domains were not investigated in this study. Our results further confirm that C332 is one of the critical cysteines for the inhibitory effect of sulfhydryl reagents on µ-opioid receptor binding. Moreover, unlike MTS reagents, N-ethylmaleimide can alkylate other groups in addition to the sulfhydryl group of cysteine. This may account for the different results for C81 and C161 between our study and the previous study.

In the initial studies, all mutants we used had a cysteine replaced by serine, a polar amino acid. To determine that the four nonsensitive cysteines (C81, C253, C294, C323) are not in positions critical to µ-opioid receptor binding, we replaced the cysteines with a nonpolar amino acid, alanine, to see if removal of the hydroxyl group of these residues would affect receptor binding. The mutant receptors exhibited similar binding properties to that of the wild-type receptor, and further imply that cysteine residues located in the different domains of the µ-opioid receptors might play different roles in the function of the receptor.

In conclusion, cysteines in TM III (C161), TM IV (C192), TM V (C237), and TM VII (C332) of the µ-opioid receptor that cysteine residues do not have essential roles in DAMGO binding; however, the locations of these cysteines seem important, presumably near or at the binding site. Once large moieties are introduced into these locations through sulfhydryl reagents, DAMGO binding is affected. Identification of these sulfhydryl-sensitive cysteines could provide further information in understanding of the molecular mechanism of the interaction between the µ-opioid receptor and its ligands.