Preclinical Testing of Nalfurafine as an Opioid-sparing Adjuvant that Potentiates Analgesia by the Mu Opioid Receptor-targeting Agonist Morphine

Abstract

Mu opioid receptor (MOR)-targeting analgesics are efficacious pain treatments, but notorious for their abuse potential. In preclinical animal models, coadministration of traditional kappa opioid receptor (KOR)-targeting agonists with MOR-targeting analgesics can decrease reward and potentiate analgesia. However, traditional KOR-targeting agonists are well known for inducing antitherapeutic side effects (psychotomimesis, depression, anxiety, dysphoria). Recent data suggest that some functionally selective, or biased, KOR-targeting agonists might retain the therapeutic effects of KOR activation without inducing undesirable side effects. Nalfurafine, used safely in Japan since 2009 for uremic pruritus, is one such functionally selective KOR-targeting agonist. Here, we quantify the bias of nalfurafine and several other KOR agonists relative to an unbiased reference standard (U50,488) and show that nalfurafine and EOM-salvinorin-B demonstrate marked G protein-signaling bias. While nalfurafine (0.015 mg/kg) and EOM-salvinorin-B (1 mg/kg) produced spinal antinociception equivalent to 5 mg/kg U50,488, only nalfurafine significantly enhanced the supraspinal analgesic effect of 5 mg/kg morphine. In addition, 0.015 mg/kg nalfurafine did not produce significant conditioned place aversion, yet retained the ability to reduce morphine-induced conditioned place preference in C57BL/6J mice. Nalfurafine and EOM-salvinorin-B each produced robust inhibition of both spontaneous and morphine-stimulated locomotor behavior, suggesting a persistence of sedative effects when coadministered with morphine. Taken together, these findings suggest that nalfurafine produces analgesic augmentation, while also reducing opioid-induced reward with less risk of dysphoria. Thus, adjuvant administration of G protein-biased KOR agonists like nalfurafine may be beneficial in enhancing the therapeutic potential of MOR-targeting analgesics, such as morphine.

Graphical abstract

Fig. 1.

Nalfurafine is a potent G protein-biased KOR agonist. Comparison of G protein and β-arrestin signaling outcomes with a variety of KOR ligands (at indicated concentrations) compared with the reference standard U50,488. (A) G protein signaling was assessed with a GloSensor assay of cyclic AMP inhibition. (B) β-arrestin recruitment to the activated KOR was assessed via the Tango assay. (C and D) Results from GloSensor and Tango assays used to calculate bias factors for each compound from observed maximal efficacy (E_max or “Max”) and potency (EC₅₀) values. ∆∆log(Max/EC₅₀) values, with their 95% confidence intervals (CIs), are plotted in (D) for each compound to indicate relative (to U50,488) bias toward G protein signaling. All compounds were tested in triplicate alongside the reference standard U50,488. N ≥ 3 for all compounds.

Fig. 2.

Evaluation of EOM salvinorin B as a potential dose-sparing adjuvant for morphine. (A) EOM salvinorin B (1 mg/kg) produced spinal antinociception as measured by warm water tail withdrawal latencies significantly increased over vehicle control (n = 12). Data were analyzed by paired t test, ***P < 0.001. (B) EOM salvinorin B did not significantly augment the analgesic effect of morphine, n = 8–15 for all groups. Data were analyzed by one-way ANOVA. (C and D) EOM salvinorin B produced significant conditioned place aversion (C) and significantly reduced morphine-induced conditioned place preference (D) (n = 6–11 for all groups). Data were analyzed by two-way ANOVA, **P < 0.01. (E and F) EOM salvinorin B suppressed both novelty-induced (E) and morphine-stimulated locomotion (F) (n = 7 or 8 for all groups). Data were analyzed by unpaired t test, **P < 0.01.

Fig. 3.

Rotarod assay of motoric effects of tested KOR agonists as adjuvants to morphine. The rotating rod was set to begin rotating at 4 rpm and accelerate to 40 rpm over the course of 5 minutes. Mice were trained on the rotarod on day 1 in two 5-minute sessions. On day 2, the baseline latency to fall off the rod was determined and then mice were injected with either vehicle or drug as indicated in the line-graph's legend. Mice were then assessed at 10, 30, and 60 minutes postinjection. Data are reported as percent of baseline performance. Adding U50,488 (5 mg/kg), nalfurafine (0.015 mg/kg), or EOM salvinorin B (1 mg/kg) to 5 mg/kg morphine each produced a decrease in the time on the rod 10 minute after coadministration, but not at 30 or 60 minutes postinjection. n = 8–11 for all groups. Data were analyzed by two-way ANOVA; **P < 0.01; ****P < 0.0001.

Fig. 4.

Spinal nociception in C57BL/6J mice is blunted by the KOR agonists U50,488 and nalfurafine, and both KOR agonists enhance the antinociceptive effect of morphine. (A) Mice were treated with vehicle 30 minutes prior to establishing baseline latencies, then treated with indicated doses of nalfurafine or U50,488 30 minutes prior to testing. Data were analyzed by one-way ANOVA with Dunnett’s multiple comparison post hoc test to assess for differences from the pooled vehicle. Holm-Sidak multiple comparison post hoc test was used to assess for differences between treatment groups (n = 10–22 for all treatment conditions; ns, nonsignificant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ###P < 0.001; ####P < 0.0001). (B) Mice were treated with 5 mg/kg morphine alone or in combination with either 5 mg/kg U50,488 or 0.015 mg/kg nalfurafine. Data were analyzed by one-way ANOVA with Dunnett’s multiple comparison post hoc test to assess for differences from morphine alone (n = 21 to 22 for all treatment groups). Data in Fig. 4 are also represented as percent of maximal possible effect (%MPE) in Supplemental Fig. S1.

Fig. 5.

Nalfurafine coadministration potentiates morphine-induced supraspinal analgesia in C57BL/6J mice. Mice were treated with indicated doses of morphine, nalfurafine, U50,488, or a combination of morphine (5 mg/kg) with either nalfurafine or U50,488. Latency to nociceptive response was recorded and compared with the response to saline [%MPE = (test response − baseline)/(30 seconds − baseline); thus, 100% MPE = 30 seconds with no sign of nociception]. All data sets were analyzed by one-way ANOVA with Dunnett’ multiple comparison post hoc test to assess differences from 5 mg/kg morphine. *P < 0.05; ***P < 0.001; ****P < 0.0001. (A) Tests of the reference standard U50,488: n = 18 for 5 mg/kg morphine; n = 7–16 for all other conditions. (B) Tests of the G protein-biased nalfurafine: n = 27 for 5 mg/kg morphine; n = 8–18 for all other conditions. Inset shows %MPE values for administration of 0.015 mg/kg nalfurafine with 1.25, 2.5, or 5 mg/kg morphine normalized to 5 mg/kg morphine alone (n = 8–15 for all groups).

Fig. 6.

Nalfurafine administration affects both spontaneous and morphine-induced locomotion. Assessment of open-field locomotion as elicited by environmental novelty or morphine administration. (A) For assessment of spontaneous locomotion, mice were administered vehicle or indicated single drug and immediately placed into the open-field chamber for 60 minutes. (B and C) To acclimate mice to drug administration, on day 2, mice were injected with vehicle only and immediately placed into the open-field chamber for 60 minutes. (D and E) For assessment of morphine-induced locomotion, mice were again habituated to the chamber for 30 minutes (on day 3) before receiving 5 mg/kg morphine with vehicle, or 5 mg/kg morphine with 1.25, 2.5 or 5 mg/kg U50,488 (D), or 5 mg/kg morphine with 0.015, 0.03, or 0.06 mg/kg nalfurafine (E) and assessed for locomotor behavior for an additional 60 minutes. Locomotion was measured by number of IR-beam breaks per 5 minutes. Data are presented as percent of pooled vehicle locomotion over the 60 minutes postinjection for suppression of novelty-induced locomotion. Total locomotion data were analyzed by one-way ANOVA with a Dunnett’s post hoc analysis to compare treatment groups with the vehicle alone or vehicle + morphine condition and Tukey’s post hoc analysis to assess difference between treatment conditions for novelty locomotion. Novelty-induced locomotion n: vehicle = 41, all other conditions, n = 7–16. Morphine suppressed locomotion n = 7–11 for all conditions. ns, not significant; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Fig. 7.

Dose-dependent production of conditioned place aversion (CPA) in C57BL/6J mice by nalfurafine administration. Mice were acclimated to the two-chambered apparatus and allowed to display a side preference. Then, mice were treated with saline (in their nonpreferred chamber) or drug (in their preferred chamber) and confined to that chamber. Vehicle and drug were given on alternate days for 4 days. On test day, the mice were allowed free access to either chamber. The preference for the drug-paired chamber was measured as the time spent in the drug-paired chamber minus the time spent in the vehicle-paired chamber. (A) Mice treated with the reference standard U50,488 displayed a significantly reduced preference for the drug-paired chamber (**P < 0.01; ****P < 0.0001, ***P < 0.0004, n = 8–15 for all groups). (B) Mice treated with 0.03 mg/kg nalfurafine, but not 0.015 or 0.06 mg/kg nalfurafine, displayed reduced preference for the drug-paired chamber. Data were analyzed by one-way ANOVA with a Dunnett’s multiple comparisons test comparing with vehicle (*) or a Sidak test to compare between doses of U50,488 or nalfurafine. **P < 0.01; ***P < 0.001; ****P < 0.0001, n = 8–26 for all groups).

Fig. 8.

Nalfurafine, like the conventional KOR agonist U50,488, reduces morphine-induced conditioned place preference (CPP) in C57BL/6J mice. Mice were acclimated to the two-chambered apparatus and allowed to display a side preference. Then, mice were treated with vehicle (in their preferred chamber) or drug (in their nonpreferred chamber) and confined to that chamber for 30 or 50 minutes (for U50,488 or nalfurafine, respectively). Vehicle or morphine was given on alternate days for 4 days. On test day, the mice were allowed free access to both chambers. The preference for the drug-paired chamber was measured as the time spent in the drug-paired chamber minus the time spent in the vehicle-paired chamber. (A) Mice coadministered morphine (5 mg/kg) and U50,488 (1.25 or 5 mg/kg) displayed reduced preference for the drug-paired chamber (n = 18–25 for all groups). (B) Mice coadministered morphine (5 mg/kg) and nalfurafine (0.015 or 0.06 mg/kg) displayed reduced preference for the drug-paired chamber (n = 18–25 for all groups). Data were analyzed by two-way ANOVA with a Sidak’s multiple comparisons test comparing with preconditioning (*), and a Dunnett’s test comparing with morphine alone (#): #P < 0.05; ##P < 0.01; ****P < 0.0001.