Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Fentanyl and its structural analogs sufentanyl and alfentanyl are potent analgesics of the 4-anilidopiperidine class of opioids, which are clinically used in the management of pain. However, some fentanyl analogs, such as p-fluorofentanyl, are known as designer drugs and have been encountered in illicit drug traffic. The illicit fentanyl homologs encountered to date are believed to have potencies ranging between the extremes of (+)-cis-3-methylfentanyl and fentanyl (Cooper et al., 1986). Although many fentanyl analogs, including designer drugs, have been synthesized, knowledge of their pharmacology is often limited to their analgesic potencies as determined by in vivo tests (Van Bever et al., 1974; Casy and Huckstep, 1988). These results provide little pharmacological evidence with respect to their affinity and selectivity for µ-, -, and -opioid receptors, through which these ligands mediate their actions. In particular for fentanyl and its analogs, the question of the opioid receptor subtype involved in respiratory depression has been addressed. Comparative pharmacological characterization of fentanyl derivatives on rat brain homogenates (Yeadon and Kitchen, 1988), guinea pig whole brain membranes, guinea pig ileum, and mouse vas deferens (Maguire et al., 1992) revealed that some fentanyl analogs with high µ-affinity such as carfentanyl show low subtype selectivity. This raises the possibility that actions through - and -opioid receptors may contribute to analgesia, euphoria, and opioid-induced side effects such as respiratory depression.

Because these analogs have not been characterized on cloned opioid receptors thus far, we investigated the pharmacological profile of p-fluorofentanyl, a representative designer drug, on human µ-, -, and -opioid receptors, coexpressed with GIRK1/GIRK2 and RGS4 in Xenopus laevis oocytes. G protein-coupled inwardly rectifying K⁺ (GIRK) channels, consisting of GIRK1 and GIRK2 subunits, mimic the probable heteromultimeric state of native neuronal GIRK channels and represent important effectors by which opioids exert their actions at the cellular level (Kofuji et al., 1995). Signaling via the G protein-mediated pathway is regulated by a recently identified gene family, known to encode regulators of G protein signaling (RGS) proteins (Druey et al., 1996). These regulators act as GTPase-activating proteins, which resolve the existing discrepancy for GIRK channel gating kinetics when coexpressed in a heterologous expression system. Coexpression of the RGS4 protein, which is highly expressed in brain, strongly accelerates GIRK channel deactivation, thereby reconstituting the native gating kinetics (Doupnik et al., 1997).

Recent molecular modeling of opioid receptors has provided new information on key residues contributing to high binding affinity of opioid ligands, including fentanyl and its analogs (Tang et al., 1996; Pogozheva et al., 1998). (+)-cis-3-Methylfentanyl was fitted into the binding pocket of the µ-opioid receptor model, and it was postulated that the aromatic ring of the 4-phenylpropanamide moiety forms a - interaction by insertion of the phenyl ring between two aryl ring planes of Trp-318 and His-319 (Tang et al., 1996). However, compared with - and -opioid receptors, key residues involved in favorable interactions are identical to those in the µ-opioid receptor model with the exception of Trp-318 (TMVII) and His-319 (TMVII). In the -opioid receptor, Trp-318 is replaced by Leu-300 at the corresponding position and the favorable interaction is lost. In the -opioid receptor, Trp-318 and His-319 are both replaced by a Tyr residue (Tyr-312 and Tyr-313), causing sterical hindrance with fentanyl derivatives. While this manuscript was in preparation, it was reported that mutation of Trp-318 to Ala decreased opioid receptor binding to almost undetectable levels. The substitution of Ala for His-319 was shown to significantly decrease the binding affinity of ohmefentanyl (Xu et al., 1999). In our study, we addressed the question of whether Trp-318 and His-319 could provide the molecular basis for µ/ and µ/ selectivity. For this reason, we mutated Trp-318 and His-319 to the residues at the corresponding positions in the - and -opioid receptor, respectively, and determined the EC₅₀ values for GIRK1/GIRK2 channel activation through opioid receptors by fentanyl, sufentanyl, and p-fluorofentanyl. Because favorable interactions with Trp-318 and His-319 also may contribute to the high affinity of morphinan alkaloids for the µ-opioid receptor, we also determined the potency of morphine on both mutant receptors. These experiments allowed us to verify whether the mutation of Trp-318 and His-319 in the human µ-opioid receptor decreases potency of the 4-anilidopiperidine and morphinan alkaloid class of opioids to the - or -opioid receptor. Finally, the efficacy of the ligands used in this study was also addressed.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subcloning and In Vitro Transcription of cDNA Clones Encoding GIRK1/2 Channels; Human µ-, -, and -Opioid Receptors (hMOR, hKOR, and hDOR); and RGS4

Plasmids containing the entire coding sequence for the mouse GIRK1 and the mouse GIRK2 channel were subcloned into the vector pSP35T and pBScMXT, respectively, and designated as pSP/GIRK1 (Kobayashi et al., 1995) and pBScMXT/GIRK2 (Kofuji et al., 1995). The polylinker in each of these vectors is flanked by X. laevis globin 5' and 3' untranslated regions, resulting in an enhanced protein expression after injection of in vitro transcribed cRNA (Kreig and Melton, 1984). For in vitro transcription, plasmids were first linearized with either EcoRI (for pSP/GIRK1) or SalI (for pBScMXT/GIRK2). Next, the cRNAs were synthesized from the linearized plasmids using the Riboprobe combination system (Promega, Madison, WI) with SP6 RNA polymerase (for pSP/GIRK1) or T3 RNA polymerase (for pBScMXT/GIRK2) in the presence of a cap analog diguanosine triphosphate (Boehringer-Mannheim Biochemica, Mannheim, Germany).

The hMOR (Raynor et al., 1995), hKOR (Zhu et al., 1995), hDOR (Knapp et al., 1994), and rat RGS4 (Doupnik et al., 1997) cDNA clones in their original vector, pcDNA3.1(+) (Invitrogen, San Diego, CA) in the case of MOR, DOR, and RGS4 and pBK-CMV (Stratagene, la Jolla, CA) in the case of KOR, were first subcloned into our custom-made high expression vector pGEMHE (Liman et al., 1992). The cDNAs encoding hMOR, hKOR, and RGS4 were isolated by a double restriction digest with HindIII, XmaI, and BamHI + XbaI, respectively. In the case of hDOR, a unique XbaI restriction site was introduced in the 3' untranslated region using the GeneEditor in vitro site-directed mutagenesis system (Promega). The cDNA encoding hDOR was subsequently isolated by a double restriction digest with BamHI + XbaI. cDNAs were then loaded onto an agarose gel, and fragments of interest were cut out, gene cleaned (Qiagen, Studio City, CA), and ligated with T4 DNA ligase (Promega) into the corresponding restriction sites of pGEMHE. For in vitro transcription, each ligation product, hMOR/pGEMHE, hKOR/pGEMHE, hDOR/pGEMHE, and RGS4/pGEMHE, was linearized with NheI.

Next, the capped cRNAs were synthesized from the linearized plasmids using the large-scale T7 mMESSAGE mMACHINE transcription kit (Ambion, Austin, TX).

Construction of Mutant hMORs

Trp-318 and His-319 in hMOR were mutated to the corresponding residues in hKOR (Tyr-312 and Tyr-313, respectively) and hDOR (Leu-300 and His-301, respectively) using the QuickChange site-directed mutagenesis kit (Stratagene). The primers for the first mutant, W318Y/H319Y, were 5'-CCAGAAACTACGTTCCAGACTGTTTCTTACTACTTCTGCATTGCTCTAGGT-3' and 5'-ACCTAGAGCAA TGCAGAAGTAGTAAGAAACAGTCTGGAACGTAGTTTCTGG-3' (codon and complementary codon are underlined). The primers for the second mutant, W318L, were designed in such way that a silent HindIII restriction site was introduced simultaneously: 5'-CCA- GAAACTACGTTCCAGACTGTAAGCTTGCACTTCTGCATTGCT- CTAGGT-3' and 5'-ACCTAGAGCAATGCAGAAGTGCAAGCTTACAGTCTGGAA CGTAGTTTCTGG-3' (codon and complementary codon are underlined, palyndromic sequence is in bold). Cycling parameters were set according to the manufacturer's guidelines. For the first mutant, W318Y/H319Y, a 313-base-pair fragment containing the desired mutation was isolated by a double restriction digest with NsiI and BglII. The mutant cDNA was then loaded on an agarose gel, and the fragment of interest was cut out, gene cleaned (Qiagen), and ligated with T4 DNA ligase (Promega) into the corresponding restriction sites of the wild-type (WT) hMOR/pGEMHE. The same mutant fragment was subcloned into pGEM11Zf(+) (Promega) for DNA sequencing. For the second mutant, W318L, the cDNAs from eight single colonies were digested with HindIII to identify possible mutants. All eight clones contained the HindIII restriction site, which was introduced by the mutant primers. The full open reading frame of this construct was sequenced to verify the presence of appropriately engineered mutations and the absence of inadvertent polymerase chain reaction errors. For in vitro transcription, each mutant, hMORW318YH319Y/pGEMHE and hMORW318L/pGEMHE, was linearized with NheI. Next, the capped cRNAs were synthesized from the linearized plasmids using the large-scale T7 mMESSAGE mMACHINE transcription kit (Ambion).

Electrophysiological Recordings

The isolation of X. laevis oocytes was conducted as previously described (Liman et al., 1992). Oocytes were coinjected with 0.5 ng/50 nl GIRK1, 0.5 ng/50 nl GIRK2, and 10 ng/50nl RGS4 cRNA, with the addition of 10 ng/50 nl hMOR, hKOR, hDOR, hMORW318L, or hMORW318Y/H319Y cRNA. Injected oocytes were maintained in ND-96 solution (composed of 2 mM KCl, 96 mM NaCl, 1 mM MgCl₂, 1.8 mM CaCl₂, 5 mM HEPES, pH 7.5) supplemented with 50 µg/ml gentamicin sulfate. Functional expression of hMOR (and hMOR mutants), hKOR, and hDOR was confirmed by the electrophysiological measurement of an agonist-gated increase of the K⁺ conductance on application of an opioid receptor subtype-selective agonist [100 nM [D-Ala²,N-Me-Phe⁴,Gly⁵-ol]-enkephalin (DAMGO), U50488-H, and [D-Pen²,D-Pen⁵]-enkephalin (DPDPE), respectively, data not shown].

Whole-cell currents from oocytes were recorded from 1 to 2 days after injection using the two-microelectrode voltage-clamp technique (GeneClamp 500; Axon Instruments, Burlingame, CA). Resistances of voltage and current electrodes were kept as low as possible (~200 k) and were filled with 3 M KCl. Currents were filtered at 10 or 200 Hz, depending on the protocol, using a 4-pole low-pass Bessel filter. To eliminate the effect of the voltage drop across the bath-grounding electrode, the bath potential was actively controlled. All experiments were performed at room temperature (19-23°C). At the start and the end of each experiment, oocytes were perfused with low-potassium (ND-96) solution (composed of 2 mM KCl, 96 mM NaCl, 1 mM MgCl₂, 1.8 mM CaCl₂, 5 mM HEPES, pH 7.5). During the application of increasing concentrations of ligands, oocytes were perfused with high-potassium (HK) solution (composed of 96 mM KCl, 2 mM NaCl, 1 mM MgCl₂, 1.8 mM CaCl₂, 5 mM HEPES, pH 7.5). In HK solution, the K⁺ equilibrium potential is close to 0 mV and enables K⁺ inward currents to flow through inwardly rectifying K⁺ channels at negative holding potentials (70 mV in all experiments). Each concentration was applied for as long as needed to achieve a steady-state GIRK1/GIRK2 current activation. Each ligand concentration was washed out by superfusion with HK solution. This opioid-induced increase of the GIRK channel conductance requires the coexpression of an opioid receptor and is antagonized by naloxone (data not shown). At the end of each experiment, the oocyte was perfused with HK solution containing 300 µM BaCl₂, causing block of the net GIRK1/2-gated inward current. Finally, the superfusion was switched back to ND-96 solution to confirm complete reversibility. Repeated receptor stimulation with 1 µM DAMGO did not reveal a decrease in the agonist-induced current increase, indicating that no receptor desensitization or GIRK1/GIRK2 channel inactivation occurred during the timeframe of a single experiment (Ulens et al., 2000). A gravity-controlled fast perfusion system (Warner Instruments, Hamden, CT) was used to ensure rapid solution exchanges. Analysis of uninjected cells (n = 3), under the same experimental conditions as injected oocytes, revealed an endogenous current that mounted maximally 1% compared with the current measured in injected oocytes. The application of opioid ligands did not evoke an increase in the conductance in uninjected oocytes.

Standardization of Experimental Model

Since Kovoor et al. (1998) reported that the opioid receptor expression level can affect the EC₅₀ value of the investigated agonist, careful attention was paid to standardization of the expression system. To achieve enhanced levels of expression, even the day after cRNA injection, all cDNAs were subcloned in our custom-made high-expression vector, pGEM-HE (Liman et al., 1992), in which the polylinker is flanked by X. laevis globin 5' and 3' untranslated regions. This results in high and reliable expression levels of the channels, receptors, and RGS4 proteins coexpressed in oocytes. Routinely, identical amounts of each receptor cRNA (hMOR, hKOR, hDOR, hMORW318L, or hMORW318Y/H319Y) were injected, and all concentration-response curves were constructed on the first day after injection. To make a valid comparison of opioid receptor agonist potencies via hMOR, hMORW318L, hMORW318Y/H319Y, hKOR, and hDOR, relatively equal numbers of receptors must be present under these different conditions. To assess the maximal level of receptor expression, each receptor cRNA (10 ng/50 nl) was coinjected with GIRK1 and GIRK2 cRNA (both 0.5 ng/50 nl). Next, maximal receptor activation was measured as the maximal GIRK1/GIRK2 channel activation on application of a saturating concentration of a full opioid receptor agonist (1 µM DAMGO for hMOR, 10 µM DAMGO for hMORW318L and hMORW318Y/H319Y, 1 µM U50488-H for hKOR, and 1 µM DPDPE for hDOR). The maximal GIRK1/GIRK2 current activation was 102 ± 7% (n = 9) via hMOR, 97 ± 11% (n = 8) via hMORW318L, 96 ± 4% (n = 8) via hMORW318Y/H319Y, 103 ± 16% (n = 9) via hKOR, and 99 ± 8% (n = 9) via hDOR (the basal GIRK1/GIRK2 conductance was taken as the 0% current level). These results suggest that relatively equal numbers of receptors are expressed under these conditions and that a comparison of potencies via different receptors is valid. To investigate whether an enhanced receptor density would yield "spare receptors" in our expression system and affect the EC₅₀ values of the agonists studied (Kovoor et al., 1998), a batch of oocytes was injected with a 10-fold dilution of opioid receptor cRNA. Under these conditions, no significant increase in EC₅₀ values was observed, confirming that no spare receptors are present. In addition, the presence of spare receptors would also reflect a large variability in EC₅₀ values obtained for different cells and batches of oocytes, which is not the case in our study (see Results).

Data Analysis

The pCLAMP program was used for data acquisition and data files (Axon Instruments) were directly imported, analyzed, and visualized with a custom-made add-in for Microsoft Excel. The percentage of activated current was calculated using the equation:

and 0% was taken as the control current level. Current percentages were then used for the calculation of concentration-response curves, using the Hill equation:
where I represents the current percentage, I_max is the maximal current percentage, EC₅₀ is the concentration of the agonist that evokes the half-maximal response, A is the concentration of agonist, and n_H is the Hill coefficient. Averaged data are indicated as mean ± S.E. and were calculated using n experiments, where n indicates the number of oocytes tested. For each experiment, current percentages were normalized to 100%, and an averaged concentration-response curve was drawn using the average EC₅₀ values and Hill coefficients of n experiments. Statistical analysis of differences between groups was carried out with Student's t test and a probability of .05 was taken as the level of statistical significance. The S.E. of quotients was calculated by taking the square root of the sum of the squares of the relative S.E., multiplied by the quotient of the means (Skoog et al., 1992).

Compounds

Synthesis, Purification, and Mass Spectrometrical Analysis of p-Fluorofentanyl. p-Fluorofentanyl (structure is shown in Fig. 1) was synthesized according to literature schemes, proceeding through the 1-substituted 4-piperidones with the formation of the Schiff base with aniline, followed by reduction and subsequent acylation of the 4-anilino moiety (Janssen, U.S. Patent 3,164,600).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Chemical structures of fentanyl analogs.

Other Compounds. Fentanyl HCl (kindly provided by National Institute on Drug Abuse, Bethesda, MD), sufentanyl (purchased from the Janssen Research Foundation, Beerse, Belgium) morphine HCl (Federa, Brussels, Belgium), DAMGO (Sigma Chemical Co., St. Louis, MO), and p-fluorofentanyl were dissolved in HK solution at a concentration of 1 mM, stored at 5°C until use, and extracellularly applied after appropriate dilution.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Pharmacological Profiles on Cloned hMOR, hKOR, and hDOR. Using heterologous expression in X. laevis oocytes, we investigated the potency of fentanyl, p-fluorofentanyl, sufentanyl, morphine, and DAMGO for hMOR, hKOR, and hDOR. Each receptor subtype was individually coexpressed with GIRK1/GIRK2 channels and RGS4, mimicking the native neuronal G protein-mediated pathway of K⁺ channel activation. The two-microelectrode voltage-clamp technique was then used to measure the opioid receptor-activated GIRK1/GIRK2 channel response as the increase of the inward K⁺ current at 70 mV, evoked by the application of increasing concentrations of opioid ligands. Representative traces of agonist-gated currents, evoked from oocytes expressing either hMOR, hKOR, or hDOR, are shown for fentanyl in Fig. 2, A to C, respectively. For comparison, representative experiments with p-fluorofentanyl on each receptor subtype are shown in Fig. 3, A to C. Concentration-response relationships, drawn using the average EC₅₀ values and Hill coefficients of four to seven experiments, are shown for fentanyl (Fig. 4A) and p-fluorofentanyl (Fig. 4B). EC₅₀ values for GIRK1/GIRK2 channel activation of sufentanyl, morphine, and DAMGO through hMOR, hKOR, and hDOR were determined in the same way. Table 1 ranks EC₅₀ values for the five ligands tested in order of increasing EC₅₀ value. These results show that p-fluorofentanyl activates GIRK1/GIRK2 channels through opioid receptors with a significantly higher potency than fentanyl. Moreover, p-fluorofentanyl is the agonist with the highest potency, not only on hMOR (4.2 ± 1.0 nM) but also on hDOR (96.2 ± 14.6 nM). As can be clearly seen in Fig. 4, the p-fluoro substitution also changes the potency profile from µ >  >  (Fig. 4A, fentanyl) to µ >    (Fig. 4B, p-fluorofentanyl).

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Representative current traces evoked from X. laevis oocytes coexpressing GIRK1/GIRK2 channels and RGS4 with hMOR (A), hKOR (B), or hDOR (C). Agonist-gated currents were evoked at 70 mV by the application of increasing concentrations of fentanyl.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Representative current traces evoked from X. laevis oocytes coexpressing GIRK1/GIRK2 channels and RGS4 with hMOR (A), hKOR (B), or hDOR (C). Agonist-gated currents were evoked at 70 mV by the application of increasing concentrations of p-fluorofentanyl.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Concentration-response curves for GIRK1/GIRK2 channel activation by increasing concentrations of fentanyl (A) and p-fluorofentanyl (B). Agonist-gated currents were evoked from X. laevis oocytes coexpressing GIRK1/GIRK2 and RGS4 with hMOR (), hKOR (), or hDOR (). The agonist-gated increase of GIRK current at each concentration was normalized to a maximal response of 100%. Zero percent was taken as the control current level. Each point represents the average current activation evoked from four to seven different oocytes.

Comparison of Channel Activation Efficacy through Cloned hMOR, hKOR, and hDOR. The efficacy by which morphine, fentanyl, and its analogs activate GIRK1/GIRK2 channels through cloned opioid receptors was investigated by application of saturating concentrations of these different agonists to the same oocyte. This experimental procedure allows compensation for variable expression levels of GIRK1/GIRK2 channels and RGS4 between oocytes. Representative traces of agonist-gated currents, evoked from oocytes coexpressing hMOR with GIRK1/GIRK2 channels and RGS4, are shown in Fig. 5A, left. Saturating concentrations of morphine (10 µM), DAMGO (1 µM), fentanyl (1 µM), sufentanyl (1 µM), and p-fluorofentanyl (1 µM) were applied in order of decreasing EC₅₀ values. Agonists with the lowest EC₅₀ value were applied at the end of each experiment because these agonists cause very slow deactivation of GIRK1/GIRK2 channels, thereby impeding total current deactivation during the time course of the experiment. Each agonist was applied for as long as needed to achieve a steady-state GIRK1/GIRK2 current activation and oocytes were perfused with HK solution after each application. For each experiment, currents were normalized as a function of the maximal DAMGO-evoked current percentage set at 100%. Averaged current percentages are shown in Fig. 5A, right (n = 8). Statistical analysis revealed that morphine, fentanyl, and its analogs activate GIRK1/2 channels through hMOR less efficiently than DAMGO. Similar protocols were used to compare the efficacy of morphine, fentanyl, and its analogs on hKOR (Fig. 5B, left) and hDOR (Fig. 5C, left). Because these ligands have higher EC₅₀ values on hKOR and hDOR, 10-fold higher concentrations were used to achieve the level of saturation. Again, agonists were applied in order of decreasing EC₅₀ values. U50488H and DPDPE were used as control agonists for oocytes coexpressing GIRK1/2 channels and RGS4 with hKOR and hDOR, respectively. For each experiment, currents were normalized as a function of the maximal U50488-H [trans-(±)-3,4-dichloro-N-methyl-N-(2-[1-pyrrolidinyl]cyclohexyl-benzeneacetamide] or DPDPE-evoked current percentage set at 100%. Averaged current percentages of are shown in Fig. 5B, right, and 5C, right, for hKOR (n = 5) and hDOR (n = 5), respectively. For hKOR, statistical analysis revealed no significant difference between sufentanyl and U50488-H. The efficacy of current activation was significantly lower for morphine (80.2 ± 3.9%), fentanyl (58.3 ± 4.7%), and p-fluorofentanyl (45.1 ± 5.3%). Morphine, fentanyl, and its analogs were less efficacious on hDOR than DPDPE. Furthermore, p-fluorofentanyl was significantly less efficacious on hDOR (49.2 ± 3.0%) than morphine, fentanyl, and sufentanyl.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Representative current traces illustrating the experimental protocol that was used to compare the efficacy of the agonists used in this study. Saturating concentrations of these different agonists were applied to the same oocyte coexpressing hMOR with GIRK1/GIRK2 channels and RGS4 (A, left). Morphine (10 µM), DAMGO (1 µM), fentanyl (1 µM), sufentanyl (1 µM), and p-fluorofentanyl (1 µM) were used. A, right, averaged current percentages (n = 8). Similar protocols were used to compare the efficacy of morphine, fentanyl and its analogs on hKOR (B, left) and hDOR (C, left). U50488H and DPDPE were used as control agonists for oocytes coexpressing GIRK1/2 channels and RGS4 with hKOR and hDOR, respectively. Averaged current percentages are shown in B, right, and C, right, for hKOR (n = 5) and hDOR (n = 5), respectively (**P < .05).

Effects of hMORW318L and hMORW318Y/H319Y Mutations. To explore the role of favorable interactions of opioid ligands with Trp-318 and His-319 in hMOR, we mutated Trp-318 to the residue at the corresponding position in hDOR (Leu-300). The adjacent His is already present in hDOR (His-301). For hMORW318Y/H319Y, a double mutant, Trp-318 and His-319 were simultaneously mutated to the residues at the corresponding positions in hKOR (Tyr-312 and Tyr-313, respectively). Each mutant receptor was individually coexpressed with GIRK1/GIRK2 channels and RGS4 in X. laevis oocytes, and EC₅₀ values for p-fluorofentanyl, fentanyl, sufentanyl, morphine, and DAMGO were determined in the same way as for the WT receptors. For clarification, EC₅₀ values on hMOR, hKOR versus hMORW318YH319Y, and hDOR versus hMORW318L are compared in the same diagram (Fig. 6) for fentanyl (Fig. 6A), p-fluorofentanyl (Fig. 6B), sufentanyl (Fig. 6C), and morphine (Fig. 6D). Changes in EC₅₀ values for the W318L and W318Y/H319Y µ-opioid receptors show a partial contribution of these residues to the decreased GIRK1/GIRK2 channel activation by fentanyl analogs through - and -opioid receptors. The most pronounced effect was observed for p-fluorofentanyl, suggesting that an interaction between the 4-fluorophenylpropanamide moiety of the drug and residues Trp-318 and His-319 is important for the resulting enhanced GIRK1/GIRK2 channel activation through the µ-opioid receptor. Interestingly, the mutation W318L completely confers -like potency for morphine on the mutant µ-opioid receptor. The potency of the ligands for the mutant receptors was compared with the potency for the WT hMOR by calculation of the ratio of EC₅₀ MUTANT over EC₅₀ hMOR, yielding a µ/mutant selectivity. To evaluate the efficiency of the mutation to convert µ-potency of GIRK1/GIRK2 channel activation to its -potency (W318L, Table 2) or -potency (W318Y/H318Y, Table 3), the "mutation efficiency" was calculated as the percentage (EC₅₀ W318L/EC₅₀ hDOR) × 100 (Table 2) or (EC₅₀ W318Y/H319Y)/EC₅₀ hKOR) × 100 (Table 3). Calculated EC₅₀ values of each ligand for hMORW318L and hMORW318Y/H319Y, their µ/mutant selectivities, and mutation efficiencies are summarized in Tables 2 and 3, respectively. Ligands in Tables 2 and 3 were ranked in order of increasing mutation efficiency.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Comparison of EC50 values for hMOR, hKOR versus hMORW318YH319Y (µ), and hDOR versus hMORW318L (µ), in the same diagram for fentanyl (A), p-fluorofentanyl (B), sufentanyl (C), and morphine (D). EC50 values for WT and mutant receptors are shown as open and filled columns, respectively. Error bars indicate S.E., and each EC50 value is the average of four to seven experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

To investigate the pharmacological profile of the ligands used in this study, heterologous expression of hMOR, hKOR, or hDOR in X. laevis oocytes was used. To achieve enhanced levels of expression, even the day after cRNA injection, all cDNAs were subcloned in a high-expression vector in which the polylinker is flanked by X. laevis globin 5' and 3' untranslated regions. Each receptor was individually coexpressed with GIRK1/GIRK2 channels and RGS4. The two-microelectrode voltage-clamp technique was then used to measure the opioid receptor-activated GIRK1/GIRK2 channel response as the increase of the inward K⁺ current at 70 mV, evoked by the application of increasing concentrations of opioid ligands.

Our results show that p-fluorofentanyl more potently activates GIRK1/GIRK2 channels through opioid receptors than fentanyl, with EC₅₀ values ~7-, ~2-, and ~8-fold lower on hMOR, hKOR, and hDOR, respectively. In addition, the p-fluoro substitution also changes the potency profile from µ >  >  (fentanyl) to µ >    (p-fluorofentanyl). As previously shown for other fentanyl analogs with high affinity for µ-opioid receptor (Yeadon and Kitchen, 1988; Maguire et al., 1992), our study shows that p-fluorofentanyl also potently activates GIRK1/GIRK2 channels through -opioid receptors. Calculated EC₅₀ values for GIRK1/GIRK2 channel activation trough hMOR and hDOR were 28.8 and 790 nM, respectively, for fentanyl and 4.2 and 96.2 nM, respectively, for p-fluorofentanyl. By analogy to fentanyl, blood concentrations of fentanyl designer drugs are believed to range between 3 and 15 nM, with initial plasma concentrations exceeding 300 nM after a high dose (Bovill and Sebel, 1980). This raises the possibility that the contribution of GIRK1/GIRK2 channel activation through - and -opioid receptors may be relevant in opioid-induced effects, including analgesia, euphoria, and respiratory depression.

In our study, µ/ selectivity was not statistically different for p-fluorofentanyl, fentanyl, and sufentanyl. Only for p-fluorofentanyl was a significantly higher µ/ selectivity calculated. Compared with a previous study on the rat µ-opioid receptor, heterologously coexpressed with GIRK1/GIRK4 channels in X. laevis oocytes, EC₅₀ values calculated for fentanyl, sufentanyl, morphine, and DAMGO are similar (Kovoor et al., 1998). In our study, the EC₅₀ values for fentanyl at hDOR (790 ± 107 nM) and hKOR (291 ± 23 nM) show good similarities with previously obtained IC₅₀ values using guinea pig ileum and mouse vas deferens (700 ± 480 nM for  and 760 ± 460 nM for ). It should be noted that EC₅₀ values at the µ-opioid receptor are several orders higher than binding potencies (K_i values). However, binding potencies at the cloned -opioid receptor have been reported to show poor correlation with previously obtained results (e.g., K_i for fentanyl at  >1000 nM) (Raynor et al., 1994). In this respect, our EC₅₀ values obtained at hDOR, heterologously expressed in X. laevis oocytes, show a better correlation.

To compare the efficacy of the agonists used in this study, saturating concentrations of the different ligands were applied to the same oocyte, coexpressing either hMOR, hKOR, or hDOR with GIRK1/GIRK2 channels and RGS4. Our results show that morphine, fentanyl, and its analogs less efficiently activate GIRK1/GIRK2 channels through hMOR than DAMGO. Results on hKOR show no significant difference between U50488-H and sufentanyl, whereas morphine, fentanyl, and p-fluorofentanyl have a significantly lower efficacy. Results on hDOR illustrate that morphine, fentanyl, and its analogs have a significantly lower efficacy than DPDPE. Furthermore, p-fluorofentanyl shows a significantly lower efficacy than its mother compound fentanyl. The finding of decreased efficacy of morphine and fentanyl compared with DAMGO is supported by direct measures of receptor-activated G proteins (Selley et al., 1998). In addition, it can be noted that these experiments clearly illustrate that an agonist with a low EC₅₀ value causes slow deactivation of GIRK1/GIRK2 channels (e.g., see Fig. 5). Assuming that oocytes do not express spare receptors, the EC₅₀ values determined in our study should reflect the K_d value of the active receptor conformation. In this case, there is an inverse relationship between the EC₅₀ value and the deactivation time constant.

A recently published three-dimensional model of opioid receptors has provided information on favorable interactions contributing to the enhanced binding affinity of fentanyl analogs and morphinan alkaloids (Pogozheva et al., 1998). Trp-318 and His-319 are key residues, present only in transmembrane domain VII of the µ-opioid receptor, that have been postulated to form - interactions with the 4-phenylpropanamide moiety of fentanyl analogs and the aromatic ring of morphine (Tang et al., 1996). In our study, we addressed the question whether Trp-318 and His-319 could provide the molecular basis for µ/ and µ/ selectivity. For this reason, we constructed two mutant hMOR receptors, namely hMORW318L (µ to ) and hMORW318Y/H319Y (µ to ). These experiments allowed us to verify whether mutation of Trp-318 and His-319 in the hMOR completely decreases potency of the 4-anilidopiperidine and morphinan alkaloid class of opioids to - or -opioid receptor.

Substitution of Trp-318 to Leu significantly increased the EC₅₀ value of fentanyl and p-fluorofentanyl (~6- and ~7-fold, respectively), whereas the EC₅₀ value of sufentanyl was affected to a lesser extent (~2-fold). For 4-anilidopiperidines, the rank order of mutation efficiency parallels that of µ/mutant selectivity. Our results also show that Trp-318 only has a partial contribution to the enhanced potency of fentanyl analogs for the µ-opioid receptor and that other structural elements also might be involved. In this regard, a previous study using chimeric receptors demonstrated that transmembrane domains I to III and the first extracellular loop in the µ-opioid receptor confer binding selectivity for sufentanyl on the -opioid receptor (Zhu et al., 1996). With an EC₅₀ value of morphine on hMORW318L being ~4-fold increased, the EC₅₀ values calculated on hMORW318L and hDOR are not statistically different. This implicates that a single mutation of Trp-318 to Leu decreases the potency via the µ-opioid receptor, thereby producing a -opioid-like receptor with a calculated mutation efficiency of 114 ± 25%. This provides experimental proof for the hypothesis that a - interaction between Trp-318 and the aromatic ring of morphine completely determines the higher potency of the ligand for the µ-opioid receptor.

Similar to the W318L mutation, EC₅₀ values on hMORW318Y/H319Y demonstrate that sufentanyl is not as sensitive as fentanyl for these mutations (~3- and ~5-fold increase of EC₅₀ values, respectively). However, mutation efficiencies are significantly higher than those calculated for hMORW318L, suggesting a larger contribution of these residues to the reduced potency of fentanyl analogs via the -opioid receptor. A mechanism of sterical hindrance by Tyr-313 in the -opioid receptor was previously suggested, explaining reduced binding affinity for this receptor subtype (Pogozheva et al., 1998). However, our results show that the mutated residues only partially contribute and that other elements might also be involved. Our finding indeed corroborates with previous data showing that regions in transmembrane domains VI and VII and the third extracellular loop are important for the selective binding of sufentanyl and lofentanyl to the µ- over the -opioid receptor (Zhu et al., 1996). The mutation hMORW318Y/H319Y has its most pronounced effect on p-fluorofentanyl, causing a 25-fold decrease of its potency via the µ-opioid receptor. Based on this result, we suggest that an interaction of the 4-fluorophenyl moiety with His-319 is important for its higher affinity for the µ-opioid receptor compared with fentanyl. It was previously postulated that the 4-phenylpropanamide moiety of fentanyl analogs is inserted into two aryl ring planes of Trp-318 and His-319, forming - interactions (Tang et al., 1996). p-Fluorosubstitution of the 4-phenylpropanamide moiety causes a redistribution in electron density, creating a partially negative charge at the fluor atom and partially positive charges in ortho and para positions, relative to the fluor atom. We hypothesize that this charge distribution can account for favorable electrostatic interactions, such as with the adjacent His-319. Such a mechanism could explain the enhanced potency of p-fluorofentanyl via the µ-opioid receptor compared with fentanyl. Morphine is the least affected by the mutation (1.4-fold reduction of the affinity), suggesting that morphine compounds are not affected by sterical hindrance.

With all of these structure-function results taken together, in the absence of any high-resolution three-dimensional structure of opioid receptors available, it should, however, be kept in mind that the results could merely represent distortion of the receptor due to detrimental influence of the mutation or construct in casu. At the present, we can only say that the G protein-coupled receptor models (Pogozheva et al., 1997; Lomize et al., 1999) that are proposed are consistent with a large body of experimental data that were not used in deriving the models and that therefore can serve as an independent control.

In conclusion, we demonstrated that p-fluorofentanyl more potently activates GIRK1/GIRK2 channels through opioid receptors than fentanyl and that the p-fluoro substitution also changes the potency profile from µ >  >  (fentanyl) to µ >    (p-fluorofentanyl). EC₅₀ values on two mutant receptors, hMORW318L and hMORW318Y/H319Y, demonstrate that these residues partially contribute to the reduced potency of fentanyl analogs via - and -opioid receptors. Sufentanyl is relatively insensitive to the mutations, suggesting that additional favorable interactions contribute to its enhanced potency via µ-opioid receptors. The EC₅₀ value of p-fluorofentanyl is strongly affected by the mutations, with the most pronounced effect on hMORW318Y/H319Y, suggesting that an interaction between His-319 and the 4-fluorophenylpropanamide moiety is important for its enhanced potency via the µ-opioid receptor. Finally, we demonstrated that mutation of Trp-318 to Leu confers -like potency for morphine on the mutant µ-opioid receptor.