Abstract
Dynorphin A is an endogenous opioid peptide that activates thekappa opioid receptor (KOR) with high potency. Some studies also showed that the distribution and functional activity of dynorphin A are not completely correlated with those of KOR, suggesting that dynorphin A may interact with other receptors. To investigate the possibility that dynorphin A may serve as an agonist for other opioid receptors, we took the advantage of the cloning of the three major types of opioid receptors, mu (MOR),delta (DOR) and KOR, and examined their affinity for and their activation by dynorphin A. We used mammalian cells transfected with each of the cDNA clones for the human receptors hMOR, hDOR, hKOR and showed that dynorphin A displaced [3H]-diprenorphine binding with Ki values in the nanomolar range at all three receptors. We also showed that, when hMOR, hDOR or hKOR was coexpressed with a G protein-activated potassium channel inXenopus oocytes, dynorphin A induced a potassium current with EC50 values in the nanomolar range for all three receptors. Furthermore, we showed that the human hORLl, an opioid receptor-like receptor that has been identified as a novel member of the opioid receptor gene family, displayed dynorphin A binding and functional activation. These results indicate that dynorphin A is capable of binding to and functional activation of all members of the opioid receptor family, suggesting that, as a potential endogenous agonist, its activity in humans may involve interaction with other members of the opioid receptor family in addition tokappa receptors.
Dynorphin A is an endogenous heptadecapeptide first isolated from pituitary glands by Goldstein and colleagues (Goldstein et al., 1979). Subsequent in vitro studies showed that it is a KOR ligand since it behaved like the prototypic kappa agonist ethylketocyclazocine in the assays of guinea pig ileum, mouse vas deferens, brain opioid receptor cross protection and brain tissue binding (Wuster et al., 1981; Huidobro-Toro et al., 1981; Chavkin et al., 1982; Rezvani et al., 1983; James et al., 1982, 1984). However, some in vivo studies showed that some of the physiological effects of dynorphins may not be mediated entirely through the KOR, such as the biphasic antinociception effects, motor effects, immunomodulation, inflammation response and modulation of respiration and body temperature (Tulunay et al., 1981; Lee, 1984; Walker et al., 1982b; Jhamandaset al., 1986; Chahl and Chahl, 1986; Smith and Lee, 1988). Studies using immunohistochemistry and in situ hybridization histochemistry showed that dynorphin A and KOR did not always coexist in some brain regions, and the distribution of dynorphin A was much more widespread throughout the brain than that of the KOR mRNA orkappa-specific ligand binding sites (DePaoli et al., 1994; Arvidsson et al., 1995). Binding studies using brain tissues have suggested that dynorphin A may interact with MOR and DOR (Quirion and Pert, 1981; Garzon et al., 1982;Young et al., 1983; Hewlett and Barchas, 1983; Garzonet al., 1984; Young et al., 1986). [3H]-dynorphin binding to brain membranes was completely displaced only by dynorphin itself, but not any other ligand even at micromolar concentrations (Smith and Lee, 1988). These results suggest that dynorphin A may possess other binding sites in addition to KOR, and thus may be a potential physiological ligand for other receptors.
In animal tissues, the complexity due to the coexistence of several types of opioid receptors and endogenous opioid ligands makes it difficult to accurately assess the interaction between a specific receptor and a given ligand such as dynorphin A. The cloning of the three major types of opioid receptors, mu, delta and kappa, and the identification of an ORL1 as a novel member of the opioid receptor gene family (Kieffer, 1995), made it possible to express each receptor in an exogenous cellular system without the presence of the others. This way, a homogenous receptor preparation can be used to evaluate ligand binding and receptor activation. In this study we used the opioid receptor cDNAs cloned from human source, and determined dynorphin A binding to and functional activation of these receptors.
Materials and Methods
Materials.
Dynorphin A was from Peninsula Laboratories Inc. (Belmont, CA), naloxone was from Research Biochemicals International (Natick, MA), radiolabeled diprenorphine and nociceptin/orphanin FQ were from Amersham (Arlington Heights, IL). AV12 cell line was from Eli Lilly and Company (Indianapolis, IN). HEK-293 cell line was from the American Type Culture Collection (Rockville, MD). Xenopus laevis were from Xenopus I (Ann Arbor, MI). Culture media were from HyClone Laboratories Inc. (Logan, UT) and Gibco BRL (Rockville, MD). In vitro transcription kit T7 mMessage mMachine was from Ambion (Austin, TX). Scintillation fluid was from ICN (Costa Mesa, CA). All other chemicals were from Sigma (St. Louis, MO).
Receptor expression.
Mammalian cells were transfected with receptor cDNAs cloned in the pcDNA3 vector (Strategene, La Jolla, CA) by the calcium phosphate method (Chen and Okayama, 1987). Cells stably expressing hMOR or hDOR were isolated from AV12 cells, and cells stably expressing hORL1 were isolated from HEK-293 cells. Due to the instability of transfected hKOR in cells, cells transiently expressing the receptor were used in the experiment.
Opioid receptor binding.
[3H]-diprenorphine was used as radiolabeled ligand. Binding of membrane proteins (10–75 μg/reaction) was performed at 4°C for 2.5 hr in binding buffer (50 mM Tris.HCl, pH7.4 and 0.5% bovine serum albumin, 1 mM PMSF, 10 μg/ml leupeptin, 100 μg/ml benzamidine, 100 μg/ml trypsin inhibitor) in untreated glass tubes with a total volume of 0.5 ml/tube. Reactions were terminated by vacuum filtration with a Brandel M-24R cell harvester through Whatman GF/B filters that had been pretreated with 0.2% polyethylenimine. The filters were washed three times, with 3.5 ml of ice-cold 50 mM Tris.HCl, pH 7.4). The washed filters were then transferred to 20-ml scintillation tubes preloaded with 10 ml of scintillation fluid, and radioactivity was determined by a Beckman L55801 scintillation counter.
Receptor hORL1 binding.
Membrane proteins (50–100 μg) in 200 μl binding buffer were added to mixtures containing 25 μl of [3H]-nociceptin and 25 μl of cold ligand or buffer in 96-well plates and incubated on ice for 90 min. Reactions were terminated by vacuum filtration with a Brandel MPXR-96T harvester through GF/B filters that had been pretreated with 0.5% polyethylenimine and 0.1% bovine serum albumin. The filters were washed four times with 1 ml of ice-cold 50 mM Tris.HCl. 30 μl of Microscint-20 was added to each filter and radioactivity was determined by scintillation spectrometry in a Packard TopCount.
Oocyte injection and electrophysiology.
Xenopusoocytes were prepared as described (Dascal et al., 1993).In vitro transcribed RNA was injected into oocytes (1–2 ng/oocyte) by a Drummond automatic microinjector. Oocytes were incubated in 50% L-15 medium supplemented with 0.8 mM of glutamine and 10 μg/ml of gentamycin at 18°C. Three days after injection, oocytes were voltage-clamped at -80 mV with two glass electrodes filled with 3 M KCl and having a resistance of 2 to 3 MΩ, using an Axoclamp-2A (Axon Instruments) under the control of pCLAMP software (Axon Instruments). Oocytes were superfused with either ND96 (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1.5 mM CaCl2 and 5 mM HEPES, pH 7.5) or a high potassium solution (96 mM KCl, 2 mM NaCl, 1 mM MgCl2, 1.5 mM CaCl2 and 5 mM HEPES, pH 7.5). Membrane currents were recorded with the aid of the pCLAMP software and on a Gould chart recorder and analyzed using pCLAMP.
Results
Dynorphin A binding to the human mu, delta, and kappa opioid receptors.
By using cDNA transfected mammalian cells, we performed membrane binding experiments to determine the affinity values of dynorphin A for three major types of human opioid receptors. Because diprenorphine is a non-selective antagonist for MOR, DOR and KOR, we used [3H]-diprenorphine as the radiolabeled ligand in receptor binding. Specific binding was defined as the difference between bound radioactivity in the absence of naloxone and that in the presence of 10 μM naloxone.
As shown in figure 1, all three opioid receptors showed saturable binding to diprenorphine. TheKd values for MOR, DOR and KOR were 2.9 ± 0.7, 1.8 ± 0.4 and 0.8 ± 0.5 nM (mean ± S.E., n = 2–3), respectively. The maximum binding values (Bmax) were 4364 ± 216 fmol/mg protein and 4901 ± 186 fmol/mg protein for stably expressed hMOR and hDOR, respectively, and 1262 ± 44 fmol/mg protein for transiently expressed hKOR (mean ± S.E., n = 2–3). Because diprenorphine binds to mu, deltaand kappa receptors with nanomolar affinity values, it was chosen as the radioligand in subsequent displacement binding experiments.
Figure 2 shows the displacement curves determined by using dynorphin A as the competitive ligand. A 0.5 to 1 nM radiolabeled diprenorphine was used. The IC50values of dynorphin A for hMOR, hDOR and hKOR were 2.0 ± 0.3, 1.65 ± 0.5 and 0.11 ± 0.2 nM (mean ± S.E.,n = 2–3), respectively. Based on these values, the calculated Ki values of dynorphin A at hMOR, hDOR and hKOR were 1.60 ± 0.18, 1.25 ± 0.12 and 0.05 ± 0.01 nM (mean ± S.E., n = 2–3), respectively (table 1). These experiments indicate that in addition to being a high affinity ligand for the human KOR with subnanomolar affinity, dynorphin A is also capable of binding to the human MOR and DOR with affinity values in the nanomolar range.
Shorter dynorphin fragments such as dynorphin A-(l-8) have relatively higher affinity for MOR and DOR compared to dynorphin A. Dynorphin is subjected to enzymatic degradation and the presence of the degraded fragments in the binding reactions may complicate the interpretation of the results. To minimize the proteolysis, a combination of protease inhibitors was used in our binding assays. In the absence of inhibitors, the apparent affinity of dynorphin A for these opioid receptors was lower than that in the presence of inhibitors, suggesting that not all the peptides produced by degradation have high affinity for the receptors. Further studies will be required to determine the affinity of dynorphin fragments for different opioid receptors, and to assess whether some of these dynorphin fragments may have potential relevance as ligands for the opioid receptors.
Dynorphin A activation of the human mu, delta and kappa opioid receptors.
To determine whether dynorphin A could serve as an agonist to activate the opioid receptors, we chose to useXenopus oocytes for functional expression. Each opioid receptor was coexpressed in oocytes with a GIRK1 channel that has been shown to be activated by G protein-coupled receptors such as muscarinic acetylcholine receptors and opioid receptors (Dascal et al., 1993; Kubo et al., 1993; Chen and Yu, 1994). Functional coupling of the receptor to the K+ channel was assessed by measuring inwardly rectifying K+currents with a two-electrode voltage clamp. We observed that dynorphin A activated an inwardly rectifying K+ current through all three human opioid receptors. As controls, oocytes injected with cRNA of either GIRK1 alone or any of the receptors alone did not show any response to dynorphin A (data not shown). This excluded the possibility of endogenous K+ currents in oocytes being activated by dynorphin A.
Due to the variability of individual oocytes in expressing exogenous proteins, we normalized the receptor-mediated response by taking the ratio of the receptor-activated current (Ia) to the basal current in hK solution (Is) in each cell (Zhang and Yu, 1995). Using this method, the dose response relations were determined for each of the human opioid receptors. As shown in figure 3, dynorphin A activated all three opioid receptors in a dose-dependent manner. A sigmoid curve was fitted to the data for each receptor, and the calculated EC50 values for hMOR, hDOR and hKOR are 30 ± 5, 84 ± 11 and 0.43 ± 0.08 nM (mean ± S.E.,n = 2–3), respectively (table 1). These results indicated that in addition to being a highly potent agonist at the KOR, dynorphin A can also serve as an agonist at the MOR and DOR.
Dynorphin A binding to and activation of the human opioid receptor-like receptor hORL1.
We previously showed that the XOR1 can be activated by dynorphin A in Xenopus oocytes (Zhang and Yu, 1995). Its endogenous peptide ligand nociceptin/orphanin FQ has been isolated from brain (Meunier et al., 1995; Reinscheidet al., 1995). Because this peptide binds ORL1 with high affinity, we used its tritiated form [3H]-nociceptin/orphanin FQ as the radioligand to determine the dynorphin A binding to the human receptor hORL1. We transfected hORL1 into HEK-293 cells and used the membrane from one of the cell lines stably expressing this receptor in our binding assay. The results showed that the receptor has saturable binding to nociceptin/orphanin FQ (fig. 4A). The dissociation constant (Kd ) was 0.81 ± 0.12 nM (mean ± S.E., n = 3), and the maximum binding (Bmax) was 739 ± 38 fmol/mg protein (mean ± S.E., n = 3). These results indicate that nociceptin/orphanin FQ binds hORL1 with nanomolar affinity value, thus can serve as a suitable radioligand for receptor binding to dynorphin A. As shown in figure 4B, the binding of [3H]- nociceptin/orphanin FQ to the receptor can be displaced by dynorphin A competitively. The IC50 value was 870 ± 48 nM (mean ± S.E., n = 3), and the calculatedKi value was 386 ± 47 nM (mean ± S.E., n = 3) based on the radiolabeled nociceptin/orphanin FQ concentration of 0.6 to 1 nM.
Coexpression of hORL1 with GIRK1 in oocytes showed that dynorphin A could functionally activate the human ORL1 receptor, resulting in an increased K+ conductance in a dose-dependent manner (fig. 4C). The EC50 value calculated from the dose response curve was 30 ± 6 nM (mean ± S.E.,n = 2), comparable to the EC50values of dynorphin A for human MOR and DOR. These results suggest that dynorphin A can also serve as an agonist for the human ORL1 receptor.
Discussion
Our understanding of the relationship between endogenous opioid peptides and opioid receptors is still evolving. Although dynorphin has been demonstrated to be a highly potent agonist at the KOR, not all reports agree as to the extent of its selectivity. Whereas several studies showed high selectivity for the KOR (Wuster et al., 1981; Huidobro-Toro et al., 1981; Chavkin et al., 1982; Rezvani et al., 1983; James et al., 1982;James et al., 1984), some reports suggested interactions of dynorphin with MOR and DOR (Quirion and Pert 1981; Garzon et al., 1982; Young et al., 1983; Hewlett and Barchas, 1983; Garzon et al., 1984; Young et al., 1986). These studies used brain tissue binding techniques using either dynorphin peptides to compete against the radiolabeled “selective” compounds, or radiolabeled dynorphin peptides to bind to “selectively” blocked receptor preparations. The presence of multiple receptor types in tissue preparations complicated the interpretation of the results. It is possible that part of the controversy about dynorphin A selectivity to the KOR is due to a lack of complete specificity of those radiolabeled compounds and blockers. Therefore, in this study we attempted to use homogenous populations of each of the human opioid receptors to characterize the dynorphin A binding and functional activation.
Based on our binding experiments using the cells transfected with human MOR, DOR, KOR or opioid receptor-like receptor hORL1, we conclude that dynorphin A binds to kappa receptor with subnanomolar affinity, and it also binds to the other receptors with high affinity. In biological tissues, the relative abundance of the opioid receptors will affect the binding profile of dynorphin, which may lead to the differences across brain regions and species, as noticed in some studies (Young et al., 1983, 1986).
Physical binding does not necessarily mean functional activation. To determine the significance of the binding between dynorphin A and these human receptors, functional assays were performed in Xenopusoocytes coexpressing each of the receptors and the GIRKl channel. Our results illustrated that dynorphin A is able to activate an inwardly rectifying K+ current by activating all four receptors. Consistent with binding experiments, our data showed that dynorphin A possesses the highest potency at the KOR. However, the action of any endogenous ligand depends not only on its affinity and potency for a particular receptor but also on the availability of those receptors at the site of ligand release. The fact that dynorphin A is capable of activating each of the opioid receptors suggests the possibility that, in places where KOR are not present, dynorphin A may bind to other receptors and activate them. Whether it acts at the KOR or other opioid receptors will depend in part on the relative abundance of receptors available in the vicinity of dynorphin-containing nerve terminals, which will vary as a function of brain region.
Animal studies have shown that dynorphin A does not always behave in the same manner as other prototypic kappa agonists. It does not always display analgesic activity in mice (Friedman et al., 1981); however, it modulates the analgesic effect of other drugs (Tulunay et al., 1981; Aceto et al., 1982;Friedman et al., 1981; Rezvani and Way, 1984). Under certain conditions dynorphin can either enhance or inhibit the analgesic effect of morphine and beta endorphin. It tends to inhibit morphine analgesia in morphine-naive animals but potentiate it in morphine-tolerant animals (Smith and Lee, 1988). This suggests that nociception regulation may be a reflection not of a single opioid, but of the interaction of several. Dynorphin may play a role in maintaining the endogenous opioid system in a state of homeostatic balance by interacting with multiple opioid receptors.
Some studies suggested that in addition to acting on an opiate site, the dynorphin molecule may have a second biologically active site that is nonopioid but capable of potent physiological effects (Walkeret al., 1982a). These effects were defined as nonopioid based on the fact that they could not be reversed by the opioid antagonist naloxone. Given the low affinity of naloxone for ORL1 (Zhang and Yu, 1995), and the ability of dynorphin A to functionally activate the ORL1 (fig. 4), this new member of the opioid receptor gene family may mediate some of the dynorphin A effects that were shown not to be blocked by naloxone.
In summary, we showed that when tested with homogenous populations of cloned receptors, dynorphin A is capable of binding to and activating the human mu, delta and ORL1 receptors, in addition to being a high-affinity agonist at the kappareceptor. Because in the synaptic region during an action potential the transient concentration of dynorphin A (for that matter, any neurotransmitter) is not known, the physiological relevance of dynorphin A binding to opioid receptors remains to be determined. Nevertheless, the nanomolar EC50 values of dynorphin A for these receptors (table 1) suggest the possibility that at synapses containing any of these non-KORs, dynorphin A is certainly a candidate agonist.
Footnotes
-
Send reprint requests to: Dr. Lei Yu, Department of Cell Biology, Neurobiology and Anatomy, University of Cincinnati, Cincinnati, OH 45267-0521.
-
↵1 This work was supported in part by the National Institutes of Health Grants DA09444 and DA11891.
- Abbreviations:
- MOR
- mu opioid receptor
- DORdelta opioid receptor
- KOR, kappa opioid receptor
- ORL1
- opioid receptor-like receptor 1
- GIRK1
- G protein-activated inwardly rectifying potassium channel
- PMSF
- phenylmethylsulfonyl fluoride
- XOR1
- rat opioid receptor-like receptor ORL1
- Received December 8, 1997.
- Accepted March 23, 1998.
- The American Society for Pharmacology and Experimental Therapeutics