µ-Opioid Receptors Expressed by Intrinsically Photosensitive Retinal Ganglion Cells Contribute to Morphine-Induced Behavioral Sensitization

Abstract

Opioid drugs are the most effective tools for treating moderate to severe pain. Despite their analgesic efficacy, long-term opioid use can lead to drug tolerance, addiction, and sleep/wake disturbances. While the link between opioids and sleep/wake problems is well-documented, the mechanism underlying opioid-related sleep/wake problems remains largely unresolved. Importantly, intrinsically photosensitive retinal ganglion cells (ipRGCs), the cells that transmit environmental light/dark information to the brain's sleep/circadian centers to regulate sleep/wake behavior, express μ-opioid receptors (MORs). In this study, we explored the potential contribution of ipRGCs to opioid-related sleep/circadian disruptions. Using implanted telemetry transmitters, we measured changes in horizontal locomotor activity and body temperature in mice over the course of a chronic morphine paradigm. Mice lacking MORs expressed by ipRGCs (McKO) exhibited reduced morphine-induced behavioral activation/sensitization compared with control littermates with normal patterns of MOR expression. Contrastingly, mice lacking MORs globally (MKO) did not acquire morphine-induced locomotor activation/sensitization. Control mice also showed morphine-induced hypothermia in both the light and dark phases, while McKO littermates only exhibited morphine-induced hypothermia in the dark. Interestingly, only control animals appeared to acquire tolerance to morphine's hypothermic effect. Morphine, however, did not acutely decrease the body temperature of MKO mice. These findings support the idea that MORs expressed by ipRGCs could contribute to opioid-related sleep/wake problems and thermoregulatory changes.

Keywords: addiction; behavioral sensitization; hypothermia; opioids; photoentrainment; retina; sleep/wake; μ-opioid receptor.

Figure 1

Prolonged but not acute morphine exposure differentially altered locomotor activity in Control, McKO, and MKO mice across light and dark phases. (A) Average behavioral activity plotted for Control (n = 14–15), McKO (n = 9–10), and MKO (n = 7) by hour (Zeitgeber Time) for different experimental days. Arrows indicate time of i.p. injection with either saline (white arrows) or 20 mg/kg of morphine (orange arrows). (B) Control and McKO animals exhibited increased horizontal activity compared with MKO animals. After prolonged morphine exposure, control mice exhibited increased locomotor activity compared with McKO littermates. Analysis performed using a three-way ANOVA (phase x genotype x experimental day) with a Holm–Bonferroni post hoc adjustment performed on all pairwise comparisons. (# p < 0.05, * p < 0.01, *** p<0.0001). Data presented as mean ± SEM.

Figure 2

McKO mice exhibited reduced morphine-induced locomotor activation compared with control littermates following protracted morphine exposure. (A) Quantification of locomotor activity 1–2 h following saline injections at ZT0 and ZT12 for Control (n = 15), McKO (n = 10), and MKO (n = 7) mice. (B) Quantification of horizontal activity 1–2 h following initial 20 mg/kg i.p. morphine injections at ZT0 and ZT12 for Control (n = 15), McKO (n = 10), and MKO (n = 7) mice. (C) Quantification of horizontal activity 1–2 h following 20 mg/kg i.p. morphine injections at ZT0 and ZT12 for Control (n = 14), McKO (n = 9), and MKO (n = 7) mice on day 6 of the morphine treatment paradigm. (D) Quantification of horizontal activity 1–2 h following 20 mg/kg i.p. morphine injections at ZT0 and ZT12 for Control (n = 14), McKO (n = 9), and MKO (n = 7) mice on day 10 of the morphine treatment paradigm. Three-way ANOVA with a Holm–Bonferonni post hoc adjustment was performed on all pairwise comparisons (# p < 0.05, * p < 0.01, ** p < 0.001, *** p < 0.0001). Data presented as mean ± SEM.

Figure 3

Morphine had differential effects on body temperature in control, McKO, and MKO mice across the morphine treatment paradigm. (A) Average body temperature plotted for control (n = 14–15), McKO (n = 9–10), and MKO (n = 7) mice by hour (Zeitgeber Time) for different experimental days. Arrows indicate time of i.p. injection with either saline (white arrows) or 20 mg/kg of morphine (orange arrows). (B) McKO mice had a lower mean body temperature than control littermates on all experimental days except for the first day of morphine treatment. Three-way ANOVA (phase × genotype × experimental day) with a Tukey post hoc adjustment was performed on all pairwise comparisons. (# p < 0.05, * p < 0.01, ** p < 0.001). Data presented as mean ± SEM.

Figure 4

Control mice, but not McKO or MKO mice, developed tolerance to the hypothermic effects of morphine. (A,B) Control mice (n = 14–15) treated with chronic morphine exhibited morphine-induced hypothermia following acute but not chronic morphine exposure. Body temperature was measured 2–3 h following a 20 mg/kg i.p. morphine injection at ZT0 and ZT12. (C,D) McKO mice (n = 9–10) treated with morphine-induced hypothermia in the dark but not in the light. Body temperature was measured 2–3 h following a 20 mg/kg i.p. morphine injection at ZT0 and ZT12. (E,F) MKO mice (n = 7) exhibited minimal body temperature changes following both acute and chronic morphine administration. Body temperature was measured 2–3 h following a 20 mg/kg i.p. morphine injection at ZT0 and ZT12. Three-way ANOVA (phase × genotype × experimental day) with a Holm–Bonferroni post hoc adjustment was performed on all pairwise comparisons (# p < 0.05, * p < 0.01). Timing of injections indicated by the white arrows. Data presented as mean ± SEM.