3.1. Effects of GSK-3 inhibitors SB216763 and AR-A014418 on morphine-induced hyperlocomotion
Mice were pretreated with SB216763 (5 mg/kg, s.c.), AR-A014418 (3 mg/kg, s.c.), or vehicle (0.3% Tween 80/0.3% DMSO in saline) for 30 min and then treated with 30 mg/kg (i.p.) morphine (r saline vehicle) and observed for 2 h. Figure 1A shows the time-course of horizontal locomotor activity observed in mice after saline challenge. There were no differences among the mice pretreated with SB216763, AR-A014418, or vehicle after saline challenge. A repeated-measures ANOVA (pretreatment x time) applied to the saline challenge data represented in Fig. 1A yielded a significant main effect of time (F(23,336) = 52, P < 0.0001) but no significant main effect of pretreatment (F(2,46) = 0.3, P = 0.76), nor any significant pretreatment x time interaction (F(46,336) = 1.1, P = 0.31).
Figure 1D shows the time course of the horizontal locomotor activity observed in mice after 30 mg/kg morphine challenge. Locomotor activity was increased in mice treated with morphine injection at 20 min post-injection, followed by a gradual increase in locomotion to maximal levels at 60 min. GSK3 inhibitor pretreatment completely eliminated the effects of morphine. A repeated-measures ANOVA (pretreatment x time) applied to the METH challenge data yielded significant main effects of time (F(23,336) = 1,7, P < 0.05) and pretreatment (F(2,46) = 27, P < 0.0001), and also a significant pretreatment x time interaction (F(46,336) = 1.1, P < 0.0001). Post hoc pair-wise comparisons showed significant differences in locomotor activity between vehicle-pretreated and GSK-3 inhibitor-pretreated groups challenged with morphine between 15- and 120-min post morphine treatment (Bonferroni/Dunn test, P < 0.05).
3.2. Effects of GSK-3 inhibitors SB216763 and AR-A014418 on morphine-induced Straub’s tail reaction (STR)
Mice were pretreated with SB216763 (5 mg/kg, s.c.), AR-A014418 (3 mg/kg, s.c.), or vehicle (0.3% Tween 80/0.3% DMSO in saline) for 30 min and then treated with 30 mg/kg (i.p.) morphine (or saline as vehicle) and observed for 2 h. Figure 1B shows the time-course of STR observed in mice after saline challenge. Very low levels of STR (based on the automated measure) were observed the three groups, and there was a slight reduction in those levels over the course of testing. Moreover, there were no differences resulting from treatment. A repeated-measures ANOVA (pretreatment x time) applied to the saline challenge data represented in Fig. 1B yielded a significant main effect of time (F(23,336) = 29, P < 0.0001) but no significant main effect of pretreatment (F(2,46) = 0.54, P = 0.60), nor any significant pretreatment x time interaction (F(46,336) = 0.80, P = 0.81).
Figure 1E shows the time course of STR observed in mice after 30 mg/kg morphine challenge. The STR was increased in mice treated with morphine injection alone at 25 min post-injection, followed by an increase in STR to maximal levels at 50 min post-injection. Pretreatment with the GSK-3 inhibitors completely eliminated STR. A repeated-measures ANOVA (pretreatment x time) applied to the morphine challenge data represented in Fig. 1E yielded significant main effects of time (F(23,336) = 1,8, P < 0.05) and pretreatment (F(2,46) = 5.3, P < 0.05), and also a significant pretreatment x time interaction (F(46,336) = 3.1, P < 0.0001). Post hoc pair-wise comparisons showed significant differences in STR between vehicle-pretreated and GSK-3 inhibitor-pretreated groups challenged with morphine (Bonferroni/Dunn test, P < 0.05) during between 25 100 min post morphine treatment (Fig. 1E).
3.3. Effects of GSK-3 inhibitors SB216763 and AR-A014418 on morphine-induced total duration of beam breaks
Mice were pretreated with SB216763 (5 mg/kg, s.c.), AR-A014418 (3 mg/kg, s.c.), or vehicle (0.3% Tween 80/0.3% DMSO in saline) and then treated with 30 mg/kg (i.p.) morphine (or saline as vehicle) 30 min later, followed by a 2 h observation period. Figure 1C shows the time-course of total duration of beam breaks observed in mice after saline challenge. Total duration of beam breaks was much higher than the duration of Straub tail (Fig. 1B), suggesting that this measure primarily reflected exploratory rearing under these conditions. The time course, high initial values followed by habituation, fits this interpretation. The maximal level occurred when mice were initially placed into the apparatus and decreased gradually to a minimal level after 120 min (Fig. 1C). There were no differences among the mice pretreated with SB216763, AR-A014418, or vehicle. A repeated-measures ANOVA (pretreatment x time) applied to the saline challenge data represented in Fig. 1C yielded a significant main effect of time (F(23,336) = 22, P < 0.0001) but no significant main effect of pretreatment (F(2,46) = 0.47, P = 0.64), nor any significant pretreatment x time interaction (F(46,336) = 0.91, P = 0.64).
Figure 1F shows the time course of the total duration of beam breaks observed in mice after 30 mg/kg morphine challenge. In mice pretreated with vehicle followed by morphine, the total duration of beam breaks was 130 sec after 5 min, followed by a decrease to a minimal level by 10 min post-injection. At this time the total duration of beam breaks increased, reaching a maximal level at 50 min post-injection, after which time beam break duration decreased gradually to basal levels. The duration of beam breaks corresponds almost exactly with the duration of STR (Fig. 1E), suggesting that the overall duration of beam breaks in this group reflects Straub tail. In contrast, mice treated with either of the GSK3 antagonists showed a very different pattern; an initial increase in beam break duration followed by a reduction to minimal levels by 25–35 min post-injection. The increases in this measure corresponded to a time where minimal levels of STR were observed (Fig. 1E) so this initial increase in beam break duration likely reflects exploratory rearing, which disappears and is not followed by STR as is observed in the saline-morphine treated mice. A repeated-measures ANOVA (pretreatment x time) applied to the morphine challenge data represented in Fig. 1F were consistent with these observations, yielding significant main effects of time (F(23,336) = 2,8, P < 0.0001), but no significant main effect of pretreatment (F(2,46) = 3.4, P = 0.06), although there was a significant pretreatment x time interaction (F(46,336) = 4.7, P < 0.0001). Post hoc pair-wise comparisons showed significant differences in STR between vehicle-pretreated and GSK-3 inhibitor-pretreated mice groups challenged with morphine (Bonferroni/Dunn test, P < 0.05) at time points of 10 and 15 min and during the time period of 30–110 min post-injection (Fig. 1F).
3.4. GSK-3β and phosphorylated GSK-3β protein expression in the striatum and nucleus accumbens
Western blotting was employed to observe protein expression levels of GSK-3β, phosphorylated GSK-3βSer9, and phosphorylated GSK-3βTyr216 in the striatum and nucleus accumbens. These brain regions were chosen because morphine-induced excitation is induced mainly in the striatum and nucleus accumbens [34, 35].
As shown in Fig. 2, the protein expression levels of GSK-3β and the phosphorylated GSK-3βSer9 were unchanged across the 4 treatment groups examined (GSK-3β: F(3,11) = 0.2, P = 0.89; GSK-3βSer9: F(3,11) = 1.3, P = 0.33). As shown in Fig. 3, the protein expression level of the phosphorylated GSK-3βTyr216 were affected by the treatments (F(3, 11) = 4.1, P = 0.050). GSK-3βTyr216 was slightly reduced with or without treatment with morphine.
3.5. Effect of SB216763 pretreatment on morphine-induced constipation
To investigate the effect of SB216763 on morphine-induced constipation, fecal boli produced by freely fed mice were counted and weighed. As shown in Fig. 4, morphine administration (30 mg/kg, i.p.) decreased the number of fecal boli (Fig. 4A) as well as the weight of the fecal boli (Fig. 4B) at all time points observed. Pretreatment with SB216763 (5 mg/kg, s.c.) had no effect either along or on morphine-induced constipation. The number of fecal boli also decreased over time. These effects were confirmed by ANOVA analysis followed by post hoc Bonferroni/Dunn test: three-way ANOVA (time x pretreatment x challenge) applied to the data represented in Fig. 4A yielded significant main effects of time (F(2,32) = 5.5, P < 0.01) and challenge (F(1,16) = 106, P < 0.0001), but no significant main effect of pretreatment (F(1,16) = 0.2, P = 0.69). There is no significant pretreatment x challenge interaction (F(1,16) = 0.006, P = 0.94), time x challenge interaction (F(2,32) = 0.2, P = 0.97), time x pretreatment interaction (F(2,32) = 1.6, P = 0.22), nor time x challenge x pretreatment interaction (F(2,32) = 0.6, P = 0.57). Three-way ANOVA (time x pretreatment x challenge) applied to the data represented in Fig. 4B yielded significant main effects of time (F(2,32) = 6.7, P < 0.01) and challenge (F(1,16) = 18, P < 0.001), but no significant main effect of pretreatment (F(1,16) = 0.006, P = 0.94). There is no significant pretreatment x challenge interaction (F(1,16) = 3.3, P = 0.09), time x challenge interaction (F(2,32) = 2.8, P = 0.07), time x pretreatment interaction (F(2,32) = 1.4, P = 0.25), nor time x challenge x pretreatment interaction (F(2,32) = 0.58, P = 0.57).