Distinct Profiles of Desensitization of µ-Opioid Receptors Caused by Remifentanil or Fentanyl: In Vitro Assay with Cells and Three-Dimensional Structural Analyses

Abstract

Remifentanil (REM) and fentanyl (FEN) are commonly used analgesics that act by activating a µ-opioid receptor (MOR). Although optimal concentrations of REM can be easily maintained during surgery, it is sometimes switched to FEN for optimal pain regulation. However, standards for this switching protocol remain unclear. Opioid anesthetic efficacy is decided in part by MOR desensitization; thus, in this study, we investigated the desensitization profiles of REM and FEN to MOR. The efficacy and potency during the 1st administration of REM or FEN in activating the MOR were almost equal. Similarly, in β arrestin recruitment, which determines desensitization processes, they showed no significant differences. In contrast, the 2nd administration of FEN resulted in a stronger MOR desensitization potency than that of REM, whereas REM showed a higher internalization potency than FEN. These results suggest that different β arrestin-mediated signaling caused by FEN or REM led to their distinct desensitization and internalization processes. Our three-dimensional analysis, with in silico binding of REM and FEN to MOR models, highlighted that REM and FEN bound to similar but distinct sites of MOR and led to distinct β arrestin-mediated profiles, suggesting that distinct binding profiles to MOR may alter β arrestin activity, which accounts for MOR desensitization and internalization.

Keywords: desensitization; fentanyl; in silico simulation; opioid receptor; remifentanil.

Figure 1

Concentration-response curves of remifentanil (REM), fentanyl (FEN), and DAMGO with cells stably expressing µ opioid receptors (MORs) with CellKeyTM, cAMP, and β arrestin assays. Cells expressing MOR were treated with REM, FEN, or DAMGO (10⁻¹² to 10⁻⁵ M), and changes in impedance (ΔZiec) were measured. Concentration-response curves were prepared by calculating ΔZiec relative to the data obtained for the positive control (10⁻⁵ M DAMGO) with CellKey^TM (A). Intracellular cAMP levels were measured using the GloSensor^® cAMP assay. Data were prepared by calculating cAMP levels relative to the data obtained with 10⁻⁵ M DAMGO and are presented as mean ± SEM for three independent experiments (n = 6) (B). β arrestin assay was performed in the cells treated with each of the compounds (10⁻¹¹ to 10⁻⁵ M), and data were prepared by calculating intracellular β arrestin levels relative to the data obtained for the positive control (10⁻⁵ M DAMGO) (C). All data are presented as mean ± SEM for three independent experiments (n = 6).

Figure 2

Repetitive administration of REM or FEM in the cells expressing MOR with CellKeyTM. Cells expressing MOR were treated with REM or FEN twice for each indicated concentration (10⁻¹⁰ to 10⁻⁸ M). Changes in impedance (ΔZiec) were measured with the CellKey^TM (A). REM was repetitively administrated as shown in (A), and data are reported (B). FEN was also repetitively administrated (A), and data are reported (C). Summary of the concentration-response curves of REM or FEN is depicted in (D). Data are presented as mean ± SEM for three independent experiments (n = 6). Relative ratio of 2nd response to 1st response (%) is reported. * p < 0.05, *** p < 0.001, **** p < 0.0001 vs. 1st responses by REM or FEN. ++ p < 0.01 vs. data obtained with 10⁻⁹ M FEN.

Figure 3

Overall analyses to find concentrations of secondary-administered REM or FEN to achieve responses equal to the 1st responses in cells expressing MOR with the CellKey^TM. Cells expressing MOR were treated with varying concentrations of REM or FEN (10⁻⁹ to 10⁻⁶ M). After measuring changes in impedance (ΔZiec) by the 1st administration (10⁻⁹, 3 × 10⁻⁹, and 10⁻⁸ M) of analgesics, 2nd administration (10⁻⁹ to 10⁻⁶ M) was conducted, and changes in impedance were measured (ΔZiec) in cells. The 1st administration of REM at 10⁻⁹ M (A), 3 × 10⁻⁹ M (B), or 10⁻⁸ M (C) and the 1st administration of FEN at 10⁻⁹ M (D), 3 × 10⁻⁹ M (E), or 10⁻⁸ M (F) are independently depicted. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 vs. the 1st responses by FEN or REM.

Figure 4

Opioid switching from REM to FEN and FEN to REM in cells expressing MOR assessed with CellKey^TM. Cells expressing MOR were treated with REM or FEN at specific concentrations. In the protocol, analgesics of the 2nd administration were changed from REM to FEN (A) or FEN to REM (B), and the changes in impedance (ΔZiec) were measured. Data are presented as mean ± SEM for three independent experiments (n = 6). ** p < 0.01, *** p < 0.001, **** p < 0.0001 vs. 1st responses by REM or FEN.

Figure 5

Internalization of MOR induced by REM or FEN in HEK293 cells stably expressing Halotag^®MOR. HEK293 cells stably expressing Halotag^®MOR were stained with Hoechst 33342 (blue) and the Halotag pH sensor ligand (red) and treated with REM, FEN, or vehicle for up to 180 min, images were taken after 60 min of incubation (A,B), and they were observed at the indicated time points (C–E). To quantify the internalization levels, the numbers and intensities of red spots/cell were counted using MetaMorph^® 7.7. Data were quantified by the sum of intensity/cell and normalized as % of sum of intensity/cell before REM or FEN administration (C–E). All data are presented as mean ± SEM (n = 3–5). Bar = 20 µm. ** p < 0.01,*** p < 0.001, **** p < 0.0001 vs. vehicle, n.s.: not significant vs. vehicle, + p < 0.05 vs. REM at the same concentration.

Figure 6

Determination of molecular dynamics (MD) simulation of REM- or FEN-bound MOR complex. Molecular structures of REM and FEN (A). Determination of MD simulation was conducted with REM- or FEN-bound human MOR. Fluctuation analysis during MD simulation revealed the difference in the influence of the ligand docking (B). The difference between REM (blue) and FEN (red) based on the fluctuation analysis is visualized (C). Green indicates the difference over 0.5 Å between the fluctuations of REM and FEN. TM: transmembrane, ICL: intracellular region, RMSF: root mean square fluctuation, AA: amino acids.