Chronic administration of aripiprazole activates GSK3β-dependent signalling pathways, and up-regulates GABAA receptor expression and CREB1 activity in rats

Abstract

Aripiprazole is a D2-like receptor (D2R) partial agonist with a favourable clinical profile. Previous investigations indicated that acute and short-term administration of aripiprazole had effects on PKA activity, GSK3β-dependent pathways, GABAA receptors, NMDA receptor and CREB1 in the brain. Since antipsychotics are used chronically in clinics, the present study investigated the long-term effects of chronic oral aripiprazole treatment on these cellular signalling pathways, in comparison with haloperidol (a D2R antagonist) and bifeprunox (a potent D2R partial agonist). We found that the Akt-GSK3β pathway was activated by aripiprazole and bifeprunox in the prefrontal cortex; NMDA NR2A levels were reduced by aripiprazole and haloperidol. In the nucleus accumbens, all three drugs increased Akt-GSK3β signalling; in addition, both aripiprazole and haloperidol, but not bifeprunox, increased the expression of Dvl-3, β-catenin and GABAA receptors, NMDA receptor subunits, as well as CREB1 phosphorylation levels. The results suggest that chronic oral administration of aripiprazole affects schizophrenia-related cellular signalling pathways and markers (including Akt-GSK3β signalling, Dvl-GSK3β-β-catenin signalling, GABAA receptor, NMDA receptor and CREB1) in a brain-region-dependent manner; the selective effects of aripiprazole on these signalling pathways might be associated with its unique clinical effects.

Figure 1. Effects of three antipsychotics on Akt activity.

The effects of aripiprazole (ARI), bifeprunox (BIF) and haloperidol (HAL) on Akt activity were measured in the prefrontal cortex (A), caudate putamen (B) and nucleus accumbens (C). The representative bands of Western blot are shown in (D). Akt was quantified at 60 kDa; p-Akt was quantified at 60 kDa. The data were normalised by taking the average value of the control group as 100% and expressed as mean ± S.E.M. (*p ≤ 0.05, **p < 0.01 vs the control).

Figure 2. Effects of three antipsychotics on GSK3β activity.

The effects of aripiprazole (ARI), bifeprunox (BIF) and haloperidol (HAL) on GSK3β activity were measured in the prefrontal cortex (A), caudate putamen (B) and nucleus accumbens (C). The representative bands of Western blot are shown in (D). GSK3β was quantified at 46 kDa; p-GSK3β was quantified at 46 kDa. The data were normalised by taking the average value of the control group as 100% and expressed as mean ± S.E.M. (*p ≤ 0.05, **p < 0.01 vs the control).

Figure 3. Effects of three antipsychotics on Dvl-3, β-catenin and GABA_A (β-1) receptor expression.

The effects of aripiprazole (ARI), bifeprunox (BIF) and haloperidol (HAL) on the expression of Dvl-3 and β-catenin were measured in the prefrontal cortex (A), caudate putamen (B) and nucleus accumbens (C). The representative bands of Western blot are shown in (D). Dvl-3 was quantified at 85 kDa; β-catenin was quantified at 92 kDa; GABA_A (β-1) receptor was quantified at 54 kDa. The data were normalised by taking the average value of the control group as 100% and expressed as mean ± S.E.M. (*p ≤ 0.05, **p < 0.01 vs the control).

Figure 4. Correlations between the ratio of p-GSK3β/GSK3β and the ratio of the expression of β-catenin, and the ratio of p-CREB1/CREB1 in the NAc.

The ratio of p-GSK3β/GSK3β was positively correlated with the expression of β-catenin (A), as well as the ratio of p-CREB1/CREB1 (B).

Figure 5. Effects of three antipsychotics on NMRA receptor subunits expression.

The effects of aripiprazole (ARI), bifeprunox (BIF) and haloperidol (HAL) on the expression of NMDA receptor subunits NR1 and NR2A were measured in the prefrontal cortex (A), caudate putamen (B) and nucleus accumbens (C). The representative bands of Western blot are shown in (D). NR1 was quantified at 105 kDa; NR2A was quantified at 165 kDa. The data were normalised by taking the average value of the control group as 100% and expressed as mean ± S.E.M. (*p ≤ 0.05, **p < 0.01 vs the control).

Figure 6. Effects of three antipsychotics on CREB1 activity.

The effects of aripiprazole (ARI), bifeprunox (BIF) and haloperidol (HAL) on CREB1 activity were measured in the prefrontal cortex (A), caudate putamen (B) and nucleus accumbens (C). The representative bands of Western blot are shown in (D). CREB1 was quantified at 40 kDa; p-CREB1 was quantified at 37 kDa. The data were normalised by taking the average value of the control group as 100% and expressed as mean ± S.E.M. (*p ≤ 0.05, **p < 0.01 vs the control).

Figure 7. A proposed schematic diagram illustrating the chronic effects of aripiprazole and haloperidol on cellular signalling in the nucleus accumbens.

D₂ receptor antagonists (e.g. haloperidol) and D₂ receptor partial agonists (e.g. aripiprazole) bind with the dopamine D₂-like receptor, resulting in the phosphorylation of Akt and the subsequent phosphorylation of GSK3β probably via biasedly antagonising D₂R-mediated β-arrestin2, together with the activation of the Dvl-GSK3β-β-catenin pathway (A), which might contribute to the therapeutic effect of antipsychotics. On the other hand, antagonism of the D₂-like receptors by aripiprazole and haloperidol leads to increased expression of GABA_A (β-1) receptors probably via the activation of PKA activity (indicated as a dashed arrow) (B), and enhanced CREB1 activity via the activation of GSK3β activity and PKA activity (C), both of which might also be involved in the therapeutic actions of antipsychotics.