Clinically employed opioid analgesics produce antinociception via μ-δ opioid receptor heteromers in Rhesus monkeys

Abstract

Morphine and related drugs are widely employed as analgesics despite the side effects associated with their use. Although morphine is thought to mediate analgesia through mu opioid receptors, delta opioid receptors have been implicated in mediating some side effects such as tolerance and dependence. Here we present evidence in rhesus monkeys that morphine, fentanyl, and possibly methadone selectively activate mu-delta heteromers to produce antinociception that is potently antagonized by the delta opioid receptor antagonist, naltrindole (NTI). Studies with HEK293 cells expressing mu-delta heteromeric opioid receptors exhibit a similar antagonism profile of receptor activation in the presence of NTI. In mice, morphine was potently inhibited by naltrindole when administered intrathecally, but not intracerebroventricularly, suggesting the possible involvement of mu-delta heteromers in the spinal cord of rodents. Taken together, these results strongly suggest that, in primates, mu-delta heteromers are allosterically coupled and mediate the antinociceptive effects of three clinically employed opioid analgesics that have been traditionally viewed as mu-selective. Given the known involvement of delta receptors in morphine tolerance and dependence, our results implicate mu-delta heteromers in mediating both antinociception and these side effects in primates. These results open the door for further investigation in humans.

Figure 1

Chemical structures of opioid agonists used in the study.

Figure 2

Methadone and fentanyl selectively activate mu-delta opioid receptor heteromers. Intracellular Ca²⁺ ion release mediated by increasing concentrations of either (a) fentanyl or (b) methadone, were measured in HEK-293 cells stably expressing opioid receptors and transiently transfected with the chimeric G-protein (Δ6-G_αqi4-myr). We have previously shown that morphine also selectively activates mu-delta opioid receptor heteromers (ref (29)). Response was measured as change in relative fluorescence units (ΔRFU = RFU_max – RFU_min). All points in the panels represent mean ± SEM from triplicate biological replications.

Figure 3

Antagonism of morphine, methadone, and fentanyl by naltrindole (NTI) occurs only at mu-delta heteromers. (a–c) NTI (1 μM) did not produce significant antagonism of morphine, methadone, or fentanyl (all 1 μM) in HEK-293 cells expressing only mu opioid receptors. However, NTI (d–f) significantly antagonized all three agonists (0.1 μM and 1 μM) in HEK-293 cells coexpressing both mu and delta opioid receptors. The antagonism of NTI (1 μM) could be surmounted by increasing the concentration of methadone (f), but not morphine (d) or fentanyl (e). The Y-axis represents % change in relative fluorescence units (RFU), and all the data are plotted as mean ± SEM.

Figure 4

Effects of pretreatment with NTI on (a) morphine-, (b) fentanyl-, and (c) methadone-induced antinociception in the assay of thermal nociception using a 50 °C thermal stimulus in rhesus monkeys (n = 3–4). Naltrindole was administered 30 min before the start of the behavioral session. Abscissae for panels (a–c): Dose of test drug in mg/kg (log scale). Ordinates for panels (a–c): Percent maximal possible effect (% MPE). Dashed lines indicate the antinociceptive effects of each mu agonist alone. Panel (d) shows dose ratios (ED₅₀ of mu agonist in the presence of naltrindole ÷ ED₅₀ of the mu agonist alone) for morphine (triangles), fentanyl (circles), and methadone (squares) as a function of the naltrindole dose. Closed symbols show no significance, while open symbols in panel (d) represent points significantly different from methadone within a naltrindole dose as indicated by nonoverlapping confidence limits shown in Table S2 in the Supporting Information. All drugs were administered via intramuscular injections. All points in all panels represent mean ± SEM.

Figure 5

Effect of intrathecal administration of NTI on the antinociceptive activity of morphine in mice. Antinociceptive dose–response curves were established for morphine in the absence and presence of NTI (5 nmol/mouse). NTI and morphine were administered 20 and 10 min, respectively, before the start of the tail-flick experiment. The dose–response curve of morphine (ED₅₀ = 25.6 pmol/mouse; CI, 21.3–35.7) was 9.7-fold right-shifted by NTI (ED₅₀ = 267.0 pmol/mouse; CI, 155.9–457.1).