Atsushi Nakamura and Masahide Fujita contributed equally to this work.
G protein-gated inwardly rectifying potassium (KIR3) channels play a primary role in the antinociceptive effect of oxycodone, but not morphine, at supraspinal sites
Article first published online: 10 DEC 2013
© 2013 The British Pharmacological Society
British Journal of Pharmacology
Volume 171, Issue 1, pages 253–264, January 2014
How to Cite
Nakamura, A., Fujita, M., Ono, H., Hongo, Y., Kanbara, T., Ogawa, K., Morioka, Y., Nishiyori, A., Shibasaki, M., Mori, T., Suzuki, T., Sakaguchi, G., Kato, A. and Hasegawa, M. (2014), G protein-gated inwardly rectifying potassium (KIR3) channels play a primary role in the antinociceptive effect of oxycodone, but not morphine, at supraspinal sites. British Journal of Pharmacology, 171: 253–264. doi: 10.1111/bph.12441
- Issue published online: 10 DEC 2013
- Article first published online: 10 DEC 2013
- Accepted manuscript online: 7 OCT 2013 08:51PM EST
- Manuscript Accepted: 25 SEP 2013
- Manuscript Revised: 17 SEP 2013
- Manuscript Received: 12 JUL 2013
- μ-opioid receptor;
- G protein-gated inwardly rectifying potassium channel
Background and Purpose
Oxycodone and morphine are μ-opioid receptor agonists prescribed to control moderate-to-severe pain. Previous studies suggested that these opioids exhibit different analgesic profiles. We hypothesized that distinct mechanisms mediate the differential effects of these two opioids and investigated the role of G protein-gated inwardly rectifying potassium (KIR3 also known as GIRK) channels in their antinociceptive effects.
Opioid-induced antinociceptive effects were assessed in mice, using the tail-flick test, by i.c.v. and intrathecal (i.t.) administration of morphine and oxycodone, alone and following inhibition of KIR3.1 channels with tertiapin-Q (30 pmol per mouse, i.c.v. and i.t.) and KIR3.1-specific siRNA. The antinociceptive effects of oxycodone and morphine were also examined after tertiapin-Q administration in the mouse femur bone cancer and neuropathic pain models.
The antinociceptive effects of oxycodone, after both i.c.v. and i.t. administrations, were markedly attenuated by KIR3.1 channel inhibition. In contrast, the antinociceptive effects of i.c.v. morphine were unaffected, whereas those induced by i.t. morphine were attenuated, by KIR3.1 channel inhibition. In the two chronic pain models, the antinociceptive effects of s.c. oxycodone, but not morphine, were inhibited by supraspinal administration of tertiapin-Q.
Conclusion and Implications
These results demonstrate that KIR3.1 channels play a primary role in the antinociceptive effects of oxycodone, but not those of morphine, at supraspinal sites and suggest that supraspinal KIR3.1 channels are responsible for the unique analgesic profile of oxycodone.
femur bone cancer
G protein-gated inwardly rectifying potassium; i.t., intrathecal
maximal possible effect
Morphine and oxycodone are clinically prescribed μ-opioid receptor agonists that control moderate-to-severe pain. Although both opioids show potent analgesic effects against various types of pain (Moulin et al., 1996), they have different analgesic profiles (Koyyalagunta et al., 2012). For example, oxycodone has been reported in some cases to control cancer pain more effectively than morphine (Heiskanen and Kalso, 1997; Mercadante and Arcuri, 1998; Watson and Babul, 1998; Portenoy et al., 1999; Bercovitch and Adunsky, 2006; Silvestri et al., 2008). It is also known that switching from one opioid to another, commonly referred to as ‘opioid rotation’, often provides improved pain management, suggesting that the underlying analgesic mechanisms of oxycodone and morphine differ.
Both morphine and oxycodone produce analgesic effects by specifically acting on μ-opioid receptors, and in vitro experiments have demonstrated that oxycodone has lower agonist activity at μ-opioid receptors than morphine in the rodent spinal cord and brain (Lemberg et al., 2006; Narita et al., 2008). However, when these two opioids are s.c. administered, oxycodone shows equivalent or even more potent analgesic effects than morphine. One possible explanation for this paradoxical effect, in which the in vitro potency profiles do not reflect in vivo analgesic potencies, is a difference in the pharmacokinetics between the two opioids. Morphine and oxycodone are thought to exert their analgesic effects by acting on μ-opioid receptor in the CNS, and oxycodone passes through the blood–brain barrier more actively than morphine (Tunblad et al., 2003; Bostrom et al., 2006; 2008). Thus, different pharmacokinetic profiles in the CNS appear to account, at least in part, for the characteristic greater analgesic potency of oxycodone compared with morphine after systemic administration. This paradoxical effect has also been observed with local opioid administration into supraspinal sites. When oxycodone and morphine were applied by i.c.v. administration, oxycodone showed similar analgesic potency to morphine despite its weaker in vitro potency profile. These results suggest that the mechanisms underlying the antinociceptive effects at supraspinal sites differ between the two μ-opioid receptor agonists.
Multiple mechanisms mediate the opioid analgesic effect, including inhibition of voltage-gated Ca2+ channels, activation of voltage-gated potassium channels and activation of the G protein-gated inwardly rectifying potassium (KIR3 or GIRK) channels (Yoshimura and North, 1983; Vaughan et al., 1997). Among these, the KIR3 channels are directly activated by G proteins that act as important mediators of the morphine-induced analgesic effect at the spinal level (Marker et al., 2002; 2004). However, little is known about the role of KIR3 channels in producing an opioid-induced analgesic effect at supraspinal sites.
The present study was performed to investigate the mechanism involved in the unique in vivo antinociceptive effects of oxycodone at supraspinal sites. We examined whether inhibition of KIR3 channels affected the antinociceptive effects of morphine and oxycodone by i.c.v. and intrathecal (i.t.) administration in mice. There were marked differences in the importance of KIR3 channel function for the supraspinal antinociceptive effects between morphine and oxycodone.
Five hundred and forty C57BL/6J male mice (body wt of 18–23 g) (Charles River Laboratories Japan, Inc., Tokyo, Japan) and 42 C3H/HeN male mice (body wt of 18–23 g) (CLEA Japan, Tokyo, Japan) were used in the present study. Animals were housed in a room maintained at 23 ± 1°C under a 12 h light/dark cycle and allowed access to water and food ad libitum. All procedures for animal experiments were approved by the Animal Care and Use Committee of Shionogi Research Laboratories, Osaka, Japan, in agreement with the internal guidelines for animal experiments and in adherence to the ethics policy of Shionogi & Co., Ltd. (Osaka, Japan). The results of all studies involving animals are reported in accordance with the Animals in Research: Reporting In Vivo Experiments (ARRIVE) guidelines for reporting experiments involving animals (Kilkenny et al., 2010; McGrath et al., 2010).
Oxycodone hydrochloride and morphine hydrochloride were obtained from Shionogi & Co., Ltd. Fentanyl citrate was obtained from Tyco Healthcare (Tyco Healthcare, Tokyo, Japan). Tertiapin-Q was purchased from Alomone Labs Ltd. (Jerusalem, Israel), and pertussis toxin (PTX) was purchased from Sigma-Aldrich (Tokyo, Japan). All drugs were dissolved in 0.9% physiological saline (Otsuka Pharmaceutical Co. Inc., Tokyo, Japan) for in vivo experiments and dissolved in assay buffer for in vitro experiments.
The antinociceptive effects of oxycodone, morphine and fentanyl were determined using the tail-flick test (Ugo-Basile, Comerio, VA, Italy) in which a heat-intensity stimulus was adjusted so that the animal flicked its tail within 4–8 s after application of the stimulus. The antinociceptive effect was expressed as a percentage of the maximal possible effect (MPE) and calculated by (T1 – T0) × 100/(T2 – T0), where T0 and T1 are the tail-flick latencies before and after administration of the opioid agonist, respectively, and T2 is the cut-off time (set at 20 s) in the tests to avoid tail damage.
Drug administration i.c.v
Drugs were administered i.c.v. as described elsewhere (Nakamura et al., 2013). On the day before i.c.v. administration, a 2 mm double needle (tip: 27G, 2 mm; base: 22G, 10 mm; Natsume Seisakusyo, Tokyo, Japan) attached to a 25 μL Hamilton microsyringe was inserted into a unilateral injection site to make a hole in the skull for injection. The unilateral injection site was approximately 2 mm caudal and 2 mm lateral from the bregma, and the needle was inserted perpendicular to the skull. When drugs were administered, the injection volume was set at 2 μL for each mouse. Each solution was injected without injection cannulae.
For PTX treatments, a PTX solution (0.5 μg per mouse, i.c.v.) was administered once a day for six consecutive days. The tail-flick test was performed on the day after the last PTX dose. For tertiapin-Q treatment, mice were pretreated with tertiapin-Q (3–30 pmol per mouse, i.c.v.) 10 min before opioid administration, and oxycodone, morphine or fentanyl was administered 10 min before measurement of the tail-flick latency or paw withdrawal response.
Drug administration i.t
Drugs were administered i.t. as described previously (Narita et al., 2008) using a 25 μL Hamilton syringe with a 30 gauge needle. The injection volume was 2 μL for each mouse. Each solution was injected without injection cannulae. For tertiapin-Q treatments, mice were pretreated with tertiapin-Q (30 pmol per mouse, i.t.) 10 min before opioid administration, and either oxycodone or morphine was administered 10 min before measurement of the tail-flick latency.
Knockdown of KIR3.1 by siRNA
In vivo i.c.v. injection of KIR3.1-specific siRNAs was performed in C57BL/6J male mice. The day before i.c.v. treatment, a 2 mm double needle attached to a 25 μL Hamilton microsyringe (described earlier) was inserted into a unilateral injection site to make a hole in the skull for injection. Mixtures of three different siRNAs, each possessing a unique nucleotide sequence against the mouse KIR3.1 channel mRNA (Kcnj3: MSS 205697–205699, 500 ng per mouse; Invitrogen, Carlsbad, CA, USA), or a negative control siRNA were mixed with Invivofectamine Reagent (Invitrogen) and gently rotated for 30 min at room temperature, then diluted with 5% glucose, in accordance with the manufacturer's protocol. Injection was performed through the hole with a depth of 2 mm and repeated three times every 24 h. The tail-flick test was performed on the day following the last injection.
Western blotting analysis
The effectiveness of siRNA knockdown was confirmed by Western blotting analysis. Whole brains of the siRNA-treated mice were homogenized in lysis buffer (640 mM sucrose, 0.1 M HEPES, containing a complete mini-EDTA-free tablet; Roche, Indianapolis, IN, USA) with a high-velocity revolution homogenizer and centrifuged at 400× g at 4°C for 10 min. Supernatants were centrifuged at 50 000× g at 4°C for 30 min. The pellets were dissolved with RIPA buffer (25 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% sodium deoxycholate, and 0.1% SDS), and 0.1% complete mini-EDTA-free tablet solution was added. After gentle rotation at 4°C for 30 min, samples were centrifuged again at 50 000× g at 4°C for 30 min, and the membrane fractions were collected from the resulting supernatants. The protein concentrations in the samples were determined using a BCA Protein Assay-Reducing Agent Compatible kit (Thermo Fisher Scientific K.K., Yokohama, Japan) according to the manufacturer's protocol. The protein samples were separated by 10% SDS-PAGE, transferred onto PVDF membranes and incubated in blocking buffer (5% skim milk in Tris-buffered saline, 0.1% Tween 20) for 60 min. The membranes were incubated with anti- KIR3.1 channel antibody (GIRK1, H-145, sc-50410, 1:200 dilution in blocking buffer; Santa Cruz Biotechnology, Santa Cruz, CA, USA) at 4°C for 16 h, and then incubated with peroxidase-labelled anti-rabbit IgG (111-036-003, 1:5000 dilution in blocking buffer; Jackson Immuno Research Laboratories, West Grove, PA, USA) at room temperature for 1 h. Membrane-bound antibody was visualized with the ECL Plus Western Blotting Detection Reagent (GE Healthcare Life Sciences, Tokyo, Japan). Signal intensities of each band were assessed by Multi Gauge software (Fujifilm, Tokyo, Japan).
Cloning of the KIR3.1 channel and μ-opioid receptor
The entire open reading frame (ORF) of the mouse KIR3.1 channel and the μ-opioid receptor 1C were amplified by PrimeSTAR HS DNA polymerase (Takara, Shiga, Japan) from mouse brain samples by PCR under the following cycling conditions (Table 3): initial denaturation at 98°C for 2 min, followed by 30 cycles at 98°C for 10 s, annealing and extension at 68°C for 1.5 min, and a final extension at 68°C for 5 min. PCR products were subcloned into the pCR 2.1-TOPO vector (Invitrogen) using the LigaFast Rapid DNA Ligation System (Promega, Madison, WI, USA). Following ligation, plasmids were transformed into competent Escherichia coli DH5α cells (Toyobo, Osaka, Japan) and plated on luria broth (LB) agar containing ampicillin and isopropylthio-β-galactoside. Positive colonies were selected and incubated in LB medium containing ampicillin. Plasmid DNA was extracted using an EndoFree plasmid mega kit (Qiagen, Tokyo, Japan).
A haemagglutinin-epitope tag sequence was added between the first methionine codon and the second aspartic acid codon of the μ-opioid receptor 1C ORF (Table 3). This included the Kozak sequence (5′-CACC-3′) following the restriction site in the 5′ region of the first codon. Amplification was performed with PrimeSTAR HS DNA polymerase (Takara) under the following PCR cycling conditions: initial denaturation at 98°C for 2 min, followed by 30 cycles at 98°C for 10 s, annealing and extension at 68°C for 1.5 min, with a final extension at 68°C for 5 min. The entire region of μ-opioid receptor 1C was digested from the vector by restriction enzymes and subcloned into the pcDNA3.1(+) vector (Invitrogen). Restriction enzyme mapping and nucleotide sequencing verified insert orientation and polymerase fidelity of the constructs.
Electrophysiological recordings in Xenopus oocytes
For in vitro transcription, plasmids containing the entire coding sequence of the mouse μ-opioid receptor 1C and KIR3.1 channel were first linearized with EcoRI for pcDNA3.1(+) and with KpnI for pCR2.1-TOPO respectively (Table 3). Capped cRNAs were synthesized from the linearized plasmids using the mMESSAGE mMACHINE T7 Ultra Kit (Ambion, Austin, TX, USA).
Mature female Xenopus laevis were anaesthetized by placing on ice and a small incision was made in the abdominal region. Small pieces of the ovarian lobes were dissected and gently shaken in Ca2+-free modified Barth's solution (88 mM NaCl, 1.0 mM KCl, 2.4 mM NaHCO3, 0.82 mM MgSO4 and 7.5 mM Tris-HCl) containing 0.6 mg·mL−1 collagenase (Yakult, Tokyo, Japan) for 40–60 min at room temperature. Oocytes were dissociated by trituration with a fire-polished Pasteur pipette and coinjected with 12 ng of μ-opioid receptor 1C and 12 ng of KIR3.1 channel cRNAs using an injector (Nanoject; Drummond Scientific, Broomall, PA, USA). Injected oocytes were incubated in sterile modified Barth's solution (88 mM NaCl, 1.0 mM KCl, 2.4 mM NaHCO3, 0.82 mM MgSO4, 7.5 mM Tris-HCl, 0.41 mM CaCl2 and 0.33 mM Ca[NO3]2) at 15°C for 7–10 days.
For electrophysiological recordings, the macroscopic GIRK channel current was recorded with a two-electrode voltage clamp technique. Oocytes were voltage-clamped at −70 mV using two glass electrodes with resistances of 0.5–1.5 MΩ filled with 3 M KCl. To enable inward K+ currents to flow through KIR3 channels, the oocytes were infused with a high-K+ solution (96 mM KCl, 2 mM NaCl, 1 mM MgCl2 and 1.5 mM CaCl2). The magnitudes of KIR3 channel current were measured with a GeneClamp 500B amplifier (Axon Instruments, Foster City, CA, USA) and the Digidata 1332A acquisition system (Axon Instruments) and analysed with pCLAMP 10 software (Axon Instruments). All experiments were performed at room temperature.
Femur bone cancer (FBC) model
NCTC 2472 tumour cells (American Type Culture Collection, Manassas, VA, USA) were maintained in Dulbecco's Modified Eagle's Medium (Invitrogen), supplemented with 10% fetal bovine serum (Invitrogen), 100 units·mL−1 penicillin and 100 μg·mL−1 streptomycin (Invitrogen), and cultured at 37 ± 0.2°C in a humidified atmosphere of 5% CO2. To prepare the FBC model, NCTC 2472 tumour cells were injected as described previously (Honore et al., 2000; Minami et al., 2009). Briefly, C3H/HeN mice were anaesthetized with 3% isoflurane, and left knee arthrotomy was performed. Tumour cells (1 × 105 cells in 5 μL of Hank's balanced salt solution) were injected directly into the medullary cavity of the distal femur, and the hole drilled in the bone was closed with resin cement (ADFA; Shofu, Kyoto, Japan). In the sham group, 5 μL of Hank's balanced salt solution was injected instead of the tumour cells in the same manner. Mice showing guarding times that changed from 0–2 s before tumour implantation to 8–16 s and displaying limb use abnormalities with a score higher than 3 (0, normal limb use; 1, slight limp; 2, obvious limp; 3, partial non-use of the limb; and 4, complete non-use of the limb) on the ipsilateral side were used for the experiments (Minami et al., 2009). The effects of opioids were assessed 14 days after tumour implantation, which is the optimal time for evaluation of allodynia in this model. Allodynia-like behaviour was recognized as paw withdrawal in response to tactile stimuli using a series of von Frey monofilaments (pressure: 0.008, 0.02, 0.04, 0.07, 0.16, 0.4, 0.6 and 1 g). The up-down method of the von Frey monofilament test was used as described previously (Minami et al., 2009).
Neuropathic pain model
Mice were anaesthetized with 3% isoflurane, and partial sciatic nerve injury was produced by tying a tight ligature with 8-0 silk suture (Natsume Seisakusyo, Tokyo, Japan) around approximately one-half the diameter of the sciatic nerve located on the right ipsilateral side under a light microscope (SD30; Olympus, Tokyo, Japan) as described previously (Seltzer et al., 1990; Malmberg and Basbaum, 1998). In the sham group, the nerve was exposed, but no ligation was performed. The effects of opioids were assessed 7 days after sciatic nerve ligation, which is the optimal time for evaluation of allodynia in this model. Allodynia-like behaviour was evaluated using von Frey monofilaments as described previously (Minami et al., 2009).
The data are shown as the means ± SEM. Statistical analyses were performed using SAS software (version 8; SAS Institute, Cary, NC, USA) and GraphPad Prism 4.0 (GraphPad Software, San Diego, CA, USA). Dose-response curves for antinociception were fitted using GraphPad Prism 4.0. The significance of differences among groups was assessed by two-way anova followed by Dunnett's comparison test. Statistical analysis of differences between two groups was carried out by Student's t-test. In all analyses, P < 0.05 was taken to indicate statistical significance. The doses producing 50% of the MPE (ED50) and 80% of the MPE (ED80) were determined by inverse prediction based on regression analysis.
Our drug/molecular target and channel nomenclature conforms to British Journal of Pharmacology's Concise Guide to PHARMACOLOGY (Alexander et al., 2013).
Antinociceptive effects of morphine and oxycodone by i.c.v. administration
We initially examined the antinociceptive effects of morphine and oxycodone by i.c.v. administration using the tail-flick test. Figure 1 shows that both morphine and oxycodone increased tail-flick latency in a dose-dependent manner (0.3–10 nmol per mouse) and 100% MPE was achieved at the highest dose (10 nmol per mouse) of both opioids (Figure 1A). When the time course of antinociception was examined, maximal drug effects were observed 10 min after administration of both opioids (data not shown), and this time point was used to determine ED50 and ED80 values. Figure 1B shows that the ED50 and ED80 values were similar between morphine and oxycodone, indicating that both opioid agonists possess similar antinociceptive efficacy. This was consistent with a previous report (Narita et al., 2008) indicating that morphine and oxycodone have equipotent antinociceptive effects with i.c.v. administration.
The antinociceptive effect of morphine is mediated by a G protein subtype sensitive to PTX, a selective inhibitor of Gi/Go (Hoehn et al., 1988; Chang et al., 1991; Shah et al., 1997). However, little is known about the G protein mediating the effect of oxycodone. Therefore, we examined whether PTX-sensitive G proteins mediate the antinociceptive effect of oxycodone. In these experiments, a submaximal antinociceptive dose of 3 nmol oxycodone or morphine/mouse was administered i.c.v. Figure 1C and D show that pretreatment with PTX (0.5 μg per mouse, i.c.v.) completely blocked the antinociceptive effects of oxycodone and morphine. These results demonstrated that a PTX-sensitive G protein mediates the effects of both opioids when administered i.c.v.
Effects of tertiapin-Q on the antinociceptive effects of morphine and oxycodone
The antinociceptive effect of morphine by i.t. administration is known to be mediated by KIR3 channels (Ikeda et al., 2000; Marker et al., 2002; 2004; Mitrovic et al., 2003), but the role of KIR3 channels in the antinociceptive effects at supraspinal sites is largely unknown. Therefore, we examined whether KIR3 channels are involved in the antinociceptive effects of morphine and oxycodone following i.c.v. administration. When mice were treated with the KIR3 channel blocker tertiapin-Q, the antinociceptive effect of oxycodone (3 nmol per mouse, i.c.v.) was markedly attenuated in a dose-dependent manner (Figure 2). In contrast, the antinociceptive effect of morphine (3 nmol per mouse, i.c.v.) was unaffected by tertiapin-Q treatment, and the effect of morphine remained at 92.7% MPE when the highest dose (30 pmol per mouse) of tertiapin-Q was administered (Figure 2). These results indicate that the underlying antinociceptive mechanisms differ between morphine and oxycodone at supraspinal sites, and that a tertiapin-Q-sensitive mechanism mediates the effect of oxycodone in an agonist-dependent manner.
We further investigated the importance of KIR3 channel activation in the antinociceptive effects of morphine and oxycodone by comparing dose-response curves of the opioid agonists in the presence or absence of tertiapin-Q. Both i.c.v. administration of oxycodone (0.3–100 nmol per mouse) and morphine (0.3–10 nmol per mouse) produced antinociceptive effects in a dose-dependent manner (Figure 3A, B), while tertiapin-Q treatment (30 pmol per mouse, i.c.v.) in the oxycodone treatment group produced a greater rightward shift in the dose-response curve compared with that in the morphine treatment group (Figure 3A, B). Table 1 shows the relative potency determined as the shift ratio (T/V) of the ED50 values between the tertiapin-Q-treated group (T) and the vehicle-treated group (V) for each opioid. While oxycodone produced a large difference in the ED50 values between the two groups with a shift ratio of 7.09, morphine exhibited only a small difference with a shift ratio of 1.43, confirming differences in the tertiapin-Q-sensitive mechanism between morphine and oxycodone at supraspinal sites (Table 1).
|ED50 (nmol)||Vehicle||Tertiapin-Q||Shift ratio (T/V)|
|Oxycodone||1.29 (0.82–2.03)||9.15 (6.12–13.68)||7.09|
|Morphine||1.57 (1.05–2.36)||2.24 (1.59–3.18)||1.43|
Since, we found for the first time that the importance of the tertiapin-Q-sensitive mechanism differed between oxycodone and morphine, we then examined the role of the tertiapin-Q-sensitive mechanism in the antinociceptive effect of fentanyl, another clinically used potent μ-opioid receptor agonist. As the maximal antinociceptive effect of fentanyl was observed 10 min after i.c.v. administration (data not shown), we evaluated the effect of fentanyl at this time point. As shown in Figure 4, i.c.v. administration of fentanyl (0.1–1.7 nmol per mouse) produced a dose-dependent antinociceptive effect. When mice were treated with fentanyl in the presence of tertiapin-Q (30 pmol per mouse, i.c.v.), the effect of fentanyl was attenuated. However, the attenuation was much smaller than that observed with oxycodone, and the shift ratio (T/V) of ED50 value was 1.46 (ED50 = 0.23 nmol per mouse in the vehicle group, 0.33 nmol per mouse in the tertiapin-Q group), which was similar to that of morphine. Thus, the tertiapin-Q-sensitive mechanism plays a primary role in the antinociceptive effect of oxycodone, but not morphine and fentanyl, at supraspinal sites.
We next examined the effect of tertiapin-Q on the antinociceptive effect of morphine and oxycodone induced by i.t. administration. As the time course experiment showed maximal drug effects 10 min after i.t. administration for both opioids (data not shown), the drug effects were determined at this time point. The results indicated that both oxycodone (1–100 nmol per mouse, i.t.) and morphine (0.1–10 nmol pe mouse, i.t.) exhibited dose-dependent antinociceptive effects (Figure 5A, B), and tertiapin-Q treatment (30 pmol per mouse, i.t.) significantly and similarly attenuated the effects of the respective opioids (Figure 5A, B). Table 2 shows the relative potency determined as the shift ratio of the ED50 values (T/V) between the tertiapin-Q-treated group (T) and the vehicle-treated group (V) for each opioid; the effects of both oxycodone and morphine were attenuated to a similar extent (shift ratio of oxycodone = 4.44, shift ratio of morphine = 5.29). This result indicates that a tertiapin-Q-sensitive mechanism plays an important role in for both the oxycodone- and morphine-induced antinociceptive effects at spinal sites.
|ED50 (nmol)||Vehicle||Tertiapin-Q||Shift ratio (T/V)|
|Oxycodone||3.51 (2.04–6.05)||15.58 (9.03–26.9)||4.44|
|Morphine||0.28 (0.17–0.48)||1.48 (0.73–3.01)||5.29|
|The following sequences were used for μ-opioid receptor 1C cloning|
|First PCR,|| |
|Nested PCR,|| |
|For the HA tag insertion in μ-opioid receptor 1C|
|First PCR,|| |
|Second PCR|| |
|The following sequences were used for KIR3.1 cloning|
Role of KIR3.1 channels in the antinociceptive effect of oxycodone administered i.c.v
Although tertiapin-Q is known to be a specific blocker of KIR3 channels, it is important to confirm whether the observed effect of tertiapin-Q is due to inhibition of KIR3 channels using a different approach. It has been reported that tertiapin-Q strongly inhibits KIR3.1 and KIR3.4 channel activities (Jin and Lu, 1998), and the level of KIR3.4 channel expression is lower than those of other KIR3 channels (KIR3.1–3.3) in the brain. Therefore, we focussed on KIR3.1 channels and employed an in vivo knockdown approach using siRNA to inhibit KIR3.1 channel activity.
We first evaluated KIR3.1 channel knockdown at supraspinal sites by i.c.v. administration of KIR3.1 channel siRNA. Western blotting analysis showed that KIR3.1 channel protein level was significantly decreased in the brains of mice treated with the KIR3.1 channel siRNA (Figure 6A). Using this knockdown protocol, we next examined whether the antinociceptive effects of oxycodone and morphine were attenuated in the tail-flick test. The results confirmed that knockdown of KIR3.1 channel expression had no significant effect on the basal tail-flick latency (control siRNA: 6.05 ± 0.66 s, KIR3.1 siRNA: 4.91 ± 0.32 s), and no obvious behavioural change was observed under our experimental conditions. When KIR3.1 channels were inhibited by siRNA treatment, the antinociceptive effect of oxycodone (3 nmol per mouse, i.c.v.) was significantly attenuated, whereas the effect of morphine (3 nmol per mouse, i.c.v.) was virtually unaffected (Figure 6B). These results were consistent with the observations from the earlier experiments using tertiapin-Q, and we concluded that the antinociceptive role of KIR3.1 channels at supraspinal sites is different between oxycodone and morphine.
Increase in K+ channel current by opioid agonists and inhibition by tertiapin-Q in Xenopus oocytes coexpressing both μ-opioid receptor 1C and KIR3.1 channels
We next investigated whether tertiapin-Q had the expected inhibitory effect on opioid-induced KIR3.1 channel activation using Xenopus oocyte-mediated gene expression. In oocytes coinjected with mouse μ-opioid receptor 1C and KIR3.1 channel cRNAs, oxycodone (0.1–100 μM) and morphine (0.01–10 μM) evoked inward currents in a concentration-dependent manner . The EC50 values of oxycodone and morphine were 3.5 and 0.18 μM, and the maximum currents of oxycodone and morphine were observed at 100 and 10 μM respectively (data not shown). When tertiapin-Q (100 nM) was applied simultaneously with either oxycodone (100 μM) or morphine (10 μM), the inward currents evoked by the respective opioids were significantly inhibited in a similar manner (Figure 7). Although tertiapin-Q (100 nM) alone changed the baseline current, this change was predicted under the experimental conditions due to the presence of high levels of potassium, and this has been reported to not affect interpretation of the opioid effect (Yow et al., 2011). These results indicate that tertiapin-Q produced the expected inhibitory effect on opioid-induced KIR3.1 channel activity.
Effects of tertiapin-Q on the antinociceptive effects of oxycodone in chronic pain models
The earlier experiments, using the tail-flick test in normal mice, indicated a significant role for supraspinal KIR3 channels in mediating the antinociceptive effects of oxycodone. However, it is also important to investigate the role of KIR3.1 channels in opioid-induced analgesia in chronic pain models. Therefore, we investigated the effects of tertiapin-Q on the antinociceptive effects of morphine and oxycodone in FBC and neuropathic pain models. We first determined optimal doses of systemically administered morphine and oxycodone in these two pain models and found that morphine at 30 mg·kg−1 and oxycodone at 5.6 mg·kg−1 exhibited similar submaximal antinociceptive effects in both models. These doses were subsequently used to evaluate the effects of tertiapin-Q. Administration of oxycodone (5.6 mg·kg−1, s.c.) significantly restored the decreased paw withdrawal threshold in the FBC model (Figure 8A), and pretreatment with tertiapin-Q (30 pmol per mouse, i.c.v.) significantly attenuated this antinociceptive effect of oxycodone. In contrast, the same tertiapin-Q treatment had no effect on the antinociceptive effect of morphine (30 mg·kg−1, s.c.) in the FBC model. A similar result was observed in the neuropathic pain model, where the antinociceptive effect of oxycodone (5.6 mg·kg−1, s.c.) was markedly attenuated by treatment with tertiapin-Q (30 pmol per mouse, i.c.v.), whereas that induced by morphine was little affected by tertiapin-Q (30 pmol per mouse, i.c.v.) (Figure 8B). These results indicate that, in addition to the differential antinociceptive effects in normal animals, antinociceptive effects in the chronic pain models are mediated by different signalling mechanisms at supraspinal sites and that KIR3.1 channels in the brain play a primary role in the effects of oxycodone, but not in those of morphine.
In the present study, we demonstrated that inhibition of KIR3.1 channels significantly attenuated the antinociceptive effects of i.c.v. oxycodone administration, whereas the antinociceptive effects of morphine were unaffected by either pharmacological or biochemical KIR3.1 channel inhibition. These results show that roles of KIR3.1 channels in the mechanisms of the antinociceptive effects differ between morphine and oxycodone at the supraspinal sites, and that activation of KIR3.1 channels is required for the intrinsic antinociceptive effect of oxycodone at supraspinal sites, but not for that of morphine.
The differential role of KIR3.1 channels in the antinociceptive effects of morphine and oxycodone was observed not only with nociceptive pain in normal mice but also during allodynia in the chronic pain models. We chose the FBC and neuropathic pain models as chronic pain models as μ-opioid receptor agonists have frequently been prescribed to control bone cancer pain and neuropathic pain in clinical settings (Mercadante and Arcuri, 1998; Portenoy et al., 1999). The FBC model (Honore et al., 2000; Minami et al., 2009) mimics several clinical features of human bone cancer pain (Komiya et al., 1999; Pandit-Taskar et al., 2004), and the neuropathic pain model produces abnormal pain characterized by hyperalgesia and allodynia. Using these two pain models, we found that the antinociceptive effect of oxycodone, but not morphine, was strongly attenuated in the presence of tertiapin-Q, indicating the importance of functional KIR3.1 channels in this opioid-dependent antinociceptive effect. One alternative interpretation of the present data is that systemically administered morphine was not distributed at sufficiently high concentrations in the brain to produce an antinociceptive effect at the supraspinal level in these two pain models, and hence, the effect of morphine was unaffected by the i.c.v. administration of tertiapin-Q. However, our preliminary data showed that when an antinociceptive dose of morphine was administered systemically in these two models, the levels of morphine were similar in the spinal cord and several brain regions (data not shown). It is unlikely, therefore, that the level of morphine in the brain was too low to produce the antinociceptive effect after systemic administration. Thus, the differential antinociceptive role of KIR3.1 channels at supraspinal sites was not only specific to nociceptive pain in normal animals but was also applicable to chronic pain conditions.
In contrast to the effects induced by i.c.v. administration, the antinociceptive effects of i.t. administration of oxycodone and morphine were similarly inhibited by tertiapin-Q. These observations indicate that KIR3.1 channels play a comparable role in the antinociceptive effects of both opioids at spinal sites. It has been reported that KIR3.1/2 channel complexes at the spinal cord modulate thermal nociception and mediate the analgesic effect of morphine (Marker et al., 2002; 2004). Our data were consistent with these results in terms of the importance of KIR3.1 channels for the morphine-induced antinociceptive effect at spinal sites.
Oxycodone and morphine are μ-opioid receptor agonists (Narita et al., 2008; Nakamura et al., 2013). We showed here that both opioids exhibited their antinociceptive effects via a PTX-sensitive G protein, but the importance of KIR3.1 channels was different between the antinociceptive effects of oxycodone and morphine at supraspinal sites. Currently, little is known about an underlying mechanism for this opioid-dependent role of KIR3.1 channels in the antinociceptive effects. We also showed that KIR3.1 channels were differently involved in the antinociceptive effect of morphine between the spinal cord and the brain. This is consistent with findings from previous studies observed in KIR3.1channel-knockout mice. In these KIR3.1 channel-knockout mice the morphine-induced K+ channel current in locus ceruleus slices was abolished, but the antinociceptive effect of morphine at supraspinal sites was shown to be preserved, whereas the morphine withdrawal syndrome was strongly attenuated (Cruz et al., 2008), indicating that supraspinal KIR3.1 channel activation does not play a primary role in the antinociceptive effect of morphine. In contrast, morphine-induced antinociception at spinal sites was attenuated in other KIR3 channel knockout mice (Marker et al., 2002; 2004). Taken together, these results suggest that morphine activates KIR3.1 channels at both spinal sites and the brain, but that the importance of KIR3.1 channel function in its nociceptive effects differs and is dependent on the site of action of morphine. In the present study, we demonstrated such region-dependent difference in the roles of KIR3 channels for the morphine antinociception in comparable experimental settings between spinal sites and the brain. This tissue-specific signalling mechanism is of great interest in the elucidation of the physiological effects of opioids.
Among the various metabolites of oxycodone, oxymorphone is produced by CYP2D6 and is known to possess high affinity and potent efficacy as an agonist at μ-opioid receptors (Lemberg et al., 2006; Peckham and Traynor, 2006). While the amount of oxymorphone produced by i.c.v. and i.t. administrations of oxycodone appears to be little in our experimental scheme, systemic administration of oxycodone, such as s.c., might produce a significant amount of oxymorphone. This suggests that the observed antinociceptive effects induced by s.c. administration of oxycodone in the pain models were the sum of the effects of both oxycodone and oxymorphone. It appears that there are species differences in the extent that oxymorphone contributes to the effects of oxycodone after systemic administration. Although the contribution of oxymorphone to oxycodone-induced analgesia is minimal in humans (Klimas et al., 2013), oxymorphone does contribute to the antinociceptive effects of oxycodone in rats (Lemberg et al., 2006). As little is known about how much oxymorphone is produced in mice, we cannot completely rule out the possibility that oxymorphone induced some of the antinociceptive effects attributed to oxycodone in the present study, especially after s.c. administration.
In conclusion, this study demonstrated that KIR3.1 channels play a primary role in the antinociceptive effect of oxycodone, but not those of morphine at supraspinal sites. To our knowledge, this is the first study demonstrating differential roles of KIR3.1 channels in the antinociceptive effects of oxycodone and morphine. The data suggest that these differences in KIR3.1 channel function at supraspinal sites are responsible for the paradoxically potent antinociceptive effect of oxycodone compared with morphine. In addition, we also demonstrated differential roles of KIR3.1 channels in mediating the antinociceptive effects of morphine at the spinal cord and in the brain. Further studies are required to determine how this opioid-dependent at supraspinal sites, and region-specific function of KIR3 channels were originated.
Conflict of interest
None of the authors have any conflicts of interest to disclose relating to this submission. A N, M F, H O, Y H, T K, K O, Y M, A N, G S, A K and M H are employees of Shionogi Co., Ltd, the manufacturer of oxycodone and morphine.
- 2013). The Concise Guide to PHARMACOLOGY 2013/14: Overview. Br J Pharmacol 170: 1449–1867. et al. (
- 2006). High dose controlled-release oxycodone in hospice care. J Pain Palliat Care Pharmacother 20: 33–39. , (
- 2006). In vivo blood-brain barrier transport of oxycodone in the rat: indications for active influx and implications for pharmacokinetics/pharmacodynamics. Drug Metab Dispos 34: 1624–1631. , , (
- 2008). Blood-brain barrier transport helps to explain discrepancies in in vivo potency between oxycodone and morphine. Anesthesiology 108: 495–505. , , (
- 1991). Dissociation of opioid receptor upregulation and functional supersensitivity. Pharmacol Biochem Behav 38: 853–859. , , , (
- 2008). Absence and rescue of morphine withdrawal in GIRK/Kir3 knock-out mice. J Neurosci 9: 4069–4077. , , , , , et al. (
- 1997). Controlled-release oxycodone and morphine in cancer related pain. Pain 73: 37–45. , (
- 1988). Pertussis toxin inhibits antinociception produced by intrathecal injection of morphine, noradrenaline and baclofen. Eur J Pharmacol 27: 65–72. , , (
- 2000). Osteroprotegerin blocks bone cancer-induced skeletal destruction, skeletal pain and pain-induced neurochemical reorganization of the spinal cord. Nat Med 6: 521–528. , , , , , et al. (
- 2000). Involvement of G-protein-activated rectifying K (GIRK) channels in opioid-induced analgesia. Neurosci Res 38: 113–116. , , , , (
- 1998). A novel high-affinity inhibitor for inward-rectifier K+ channels. Biochemistry 37: 13291–13299. , (
- 2010). Animal research: reporting in vivo experiments: the ARRIVE guidelines. Br J Pharmacol 160: 1577–1579. , , , , (
- 2013). Contribution of oxycodone and its metabolites to the overall analgesic effect after oxycodone administration. Expert Opin Drug Metab Toxicol 9: 517–528. , , , (
- 1999). Bone tumors in the pelvis presenting growth during pregnancy. Arch Orthop Trauma Surg 119: 22–29. , , (
- 2012). A systematic review of randomized trials on the effectiveness of opioids for cancer pain. Pain Physician 15: ES39–ES58. , , , , , et al. (
- 2006). Antinociception by spinal and systemic oxycodone: why does the route make a difference? In vitro and in vivo studies in rats. Anesthesiology 105: 801–812. , , , , , et al. (
- 2010). Guidelines for reporting experiments involving animals: the ARRIVE guidelines. Br J Pharmacol 160: 1573–1576. , , , , (
- 1998). Partial sciatic nerve injury in the mouse as a model of neuropathic pain: behavioral and neuroanatomical correlates. Pain 76: 215–222. , (
- 2002). Hyperalgesia and blunted morphine analgesia in G protein-gated potassium channel subunit knockout mice. Neuroreport 20: 2509–2513. , , , , (
- 2004). Spinal G-protein-gated K+ channels formed by GIRK1 and GIRK2 subunits modulate thermal nociception and contribute to morphine analgesia. J Neurosci 17: 2806–2812. , , (
- 1998). Breakthrough pain in cancer patients: pathophysiology and treatment. Cancer Treat Rev 24: 425–432. , (
- 2009). Morphine, oxycodone, and fentanyl exhibit different analgesic profiles in mouse pain models. J Pharmacol Sci 111: 60–72. , , , , , et al. (
- 2003). Contribution of GIRK2-mediated postsynaptic signaling to opiate and alpha2-adrenergic analgesia and analgesic sex differences. Proc Nalt Acad Sci USA 100: 271–276. , , , , , (
- 1996). Randomised trial of oral morphine for chronic non-cancer pain. Lancet 347: 143–147. , , , , , (
- 2013). Differential activation of the μ-opioid receptor by oxycodone and morphine in pain-related brain regions in a bone cancer pain model. Br J Pharmacol 168: 375–388. , , , , , et al. (
- 2008). Comparative pharmacological profiles of morphine and oxycodone under a neuropathic pain-like state in mice: evidence for less sensitivity to morphine. Neuropsychopharmacology 33: 1097–1112. , , , , , et al. (
- 2004). Radiopharmaceutical therapy for palliation of bone pain from osseous metastases. J Nucl Med 45: 1358–1365. , , (
- 2006). Comparison of the antinociceptive response to morphine and morphine-like compounds in male and female Sprague-Dawley rats. J Pharmacol Exp Ther 316: 1195–1201. , (
- 1999). Breakthrough pain: characteristics and impact in patients with cancer pain. Pain 81: 129–134. , , (
- 1990). A novel behavioral model of neuropathic pain disorders produced in rats by partial sciatic nerve injury. Pain 43: 205–218. , , (
- 1997). Time-dependent effects of in vivo pertussis toxin on morphine analgesia and G-proteins in mice. Pharmacol Biochem Behav 56: 465–469. , , , , , (
- 2008). Oxycodone controlled-release as first-choice therapy for moderate-to-severe cancer pain in Italian patients: results of an open-label, multicentre, observational study. Clin Drug Investig 28: 399–407. , , , , , et al. (
- 2003). Morphine blood-brain barrier transport is influenced by probenecid co-administration. Pharm Res 20: 618–623. , , (
- 1997). How opioids inhibit GABA-mediated neurotransmission. Nature 390: 611–614. , , , (
- 1998). Efficacy of oxycodone in neuropathic pain: a randomized trial in postherpetic neuralgia. Neurology 50: 1837–1841. , (
- 1983). Substantia gelatinosa neurons hyperpolarized in vitro by encephalin. Nature 305: 529–530. , (
- 2011). Naringin directly activates inwardly rectifying potassium channels at an overlapping binding site to tertiapin-Q. Br J Pharmacol 163: 1017–1033. , , , , , et al. (