Activation of Astroglial Connexin is Involved in Concentration-Dependent Double-Edged Sword Clinical Action of Clozapine

Abstract

Clozapine (CLZ) is a gold-standard antipsychotic against treatment-refractory schizophrenia, but is one of the most toxic antipsychotic agents. Pharmacological mechanisms of the double-edged sword clinical action of CLZ remain to be clarified. To explore the mechanisms of CLZ, the present study determined the astroglial transmission associated with connexin43 (Cx43), which is the most principal expression in astrocytes and myocardial cells, and expression of Cx43 in primary cultured astrocytes. Both acute and subchronic administrations of CLZ concentration-dependently increased Cx43-associated astroglial release of l-glutamate and d-serine, whereas therapeutic-relevant concentration of CLZ acutely did not affect but subchronically increased astroglial release. In contrast, after the subchronic administration of therapeutic-relevant concentration of valproate (VPA), acute administration of therapeutic-relevant concentration of CLZ drastically increased Cx43-associated astroglial releases. VPA increased Cx43 expression in cytosol fraction without affecting plasma membrane fraction, whereas CLZ increased Cx43 expression in both fractions. Acute administration of therapeutic-relevant concentration of CLZ drastically increased Cx43 expression in the plasma membrane fraction of astrocytes subchronically treated with VPA. The present findings suggest that CLZ-induced the activation of Cx43-associated channel activity and transported Cx43 to plasma membrane, probably contribute to the double-edged sword clinical action of CLZ, such as improvement of cognitive dysfunction and CLZ-induced myocarditis.

Keywords: clozapine; connexin; hemichannel; schizophrenia.

Figure 1

Concentration-dependent effects of acute administration of CLZ (clozapine) on basal and K⁺-evoked astroglial releases of l-glutamate (A) and d-serine (B). After wash-out, astrocytes were incubated in 100 μL ACSF (artificial cerebrospinal fluid) containing CLZ (0, 1, 3, 10, 30, or 100 μM) for 60 min (pretreatment incubation). After pretreatment incubation, to determine the K⁺-evoked astroglial releases of L-glutamate and D-serine, astrocytes were incubated in ACSF (3.0 mM K⁺: opened circles), MK-ACSF (50.0 mM K⁺: closed circles) or HK-ACSF (100.0 mM K⁺: blue circles) containing the same concentration of CLZ of pretreatment incubation for 20 min. To clarify the astroglial releases of L-glutamate and D-serine associated with Cxs (connexin (Cx) composed transmembrane channels), astrocytes were incubated in HK-ACSF containing GAP19 (TAT-conjugated Gap19, 20 μM) (green circles) or CBX (carbenoxolone, 100 μM) (red circles) with the same concentration of CLZ during pretreatment incubation for 20 min, and then incubation medium (ACSF, MK-ACSF or HK-ACSF) was collected for analysis. Ordinate: mean ± SD (n = 6) of extracellular levels of l-glutamate and d-serine (μM). Abscissa: concentration of CLZ (μM). * p < 0.05 and ** p < 0.01 vs. CLZ free by MANOVA with Tukey’s post hoc test. @ p < 0.05 and @@ p < 0.01 vs. HK-ACSF by MANOVA with Tukey’s post hoc test.

Figure 2

Concentration-dependent effects of subchronic administration of CLZ on basal and K⁺-evoked astroglial releases of L-glutamate (A) and D-serine (B). Astrocytes were incubated in Dulbecco’s modified Eagle’s medium containing 10% fetal calf serum (fDMEM) containing CLZ (0, 1, 3, 10, 30, or 100 μM) for 7 days. After wash-out, astrocytes were incubated in 100 μL ACSF containing the same concentration of CLZ for 60 min (pretreatment incubation). After pretreatment incubation, to determine the K⁺-evoked astroglial releases of L-glutamate and D-serine, astrocytes were incubated in ACSF (3.0 mM K⁺: opened circles), MK-ACSF (50.0 mM K⁺: closed circles), or HK-ACSF (100.0 mM K⁺: blue circles) containing the same concentration of CLZ during pretreatment incubation for 20 min. To clarify the astroglial releases of L-glutamate and D-serine associated with Cxs, astrocytes were incubated in HK-ACSF containing GAP19 (20 μM) (green circles) or CBX (100 μM) (red circles) with the same concentration of CLZ during pretreatment incubation for 20 min, and then incubate medium (ACSF, MK-ACSF, or HK-ACSF) was collected for analysis. Ordinate: mean ± SD (n = 6) of extracellular levels of l-glutamate and d-serine (μM). Abscissa: concentration of CLZ (μM). * p < 0.05 and ** p < 0.01 vs. CLZ free by MANOVA with Tukey’s post hoc test. @@ p < 0.01 vs. HK-ACSF by MANOVA with Tukey’s post hoc test.

Figure 3

Concentration-dependent effects of acute administration of VPA (valproate) on basal and K⁺-evoked astroglial releases of l-glutamate (A) and d-serine (B). After wash-out, astrocytes were incubated in ACSF containing VPA (0, 300, 1000, or 3000 μM) for 60 min (pretreatment incubation). After pretreatment incubation, to determine the K⁺-evoked astroglial releases of L-glutamate and d-serine, astrocytes were incubated in ACSF (3.0 mM K⁺: opened circles), MK-ACSF (50.0 mM K⁺: closed circles), or HK-ACSF (100.0 mM K⁺: blue circles) containing the same concentration of VPA during pretreatment incubation for 20 min. Concentration-dependent effects of subchronic administration of VPA on basal and K⁺-evoked astroglial releases of l-glutamate (C) and d-serine (D). Astrocytes were incubated in fDMEM containing VPA (0, 300, 1000, or 3000 μM) for 7 days. After wash-out, astrocytes were incubated in ACSF containing the same concentration of VPA (pretreatment incubation). After pretreatment, to clarify the K⁺-evoked astroglial releases of l-glutamate and d-serine, astrocytes were incubated in ACSF (opened circles), MK-ACSF (closed circles), or HK-ACSF (blue circles) containing the same concentration of VPA during pretreatment incubation for 20 min, and then incubate medium (ACSF, MK-ACSF, or HK-ACSF) was collected for analysis. Ordinate: mean ± SD (n = 6) of extracellular levels of l-glutamate and d-serine (μM). Abscissa: concentration of VPA (μM). * p < 0.05 and ** p < 0.01 vs. VPA free by MANOVA with Tukey’s post hoc test.

Figure 4

Acute effects of therapeutic-relevant concentration of VPA (1000 μM) on astroglial releases of L-glutamate (A) and D-serine (B) from astrocytes subchronically administrated with CLZ. Astrocytes were incubated in fDMEM containing CLZ (0, 1, 3, 10, 30, or 100 μM) for 7 days. After wash-out, astrocytes were incubated in ACSF containing the same concentration of CLZ without (control: opened circles) or with therapeutic-relevant concentration of VPA (1000 μM: closed circles) for 60 min (pretreatment incubation). After pretreatment, to determine the K⁺-evoked astroglial releases of L-glutamate and D-serine, astrocytes were incubated in MK-ACSF (50.0 mM K⁺: black circles) or HK-ACSF (100.0 mM K⁺: blue circles) containing the same concentrations of CLZ and VPA during pretreatment incubation for 20 min, and then incubate medium (MK-ACSF or HK-ACSF) was collected for analysis. Ordinate: mean ± SD (n = 6) of extracellular levels of l-glutamate and d-serine (μM). Abscissa: concentration of CLZ (μM). * p < 0.05 and ** p < 0.01 vs. CLZ free by MANOVA with Tukey’s post hoc test.

Figure 5

Acute effects of CLZ on astroglial releases of L-glutamate (A) and D-serine (B) from astrocytes subchronically administrated with therapeutic-relevant concentration of VPA (1000 μM). Astrocytes were incubated in fDMEM without (control: opened circles) or with VPA (1000 μM: closed circles). After wash-out, astrocytes were incubated in ACSF containing the same concentration of VPA with CLZ (0, 1, 3, 10, 30, 100 μM) for 60 min (pretreatment incubation). After pretreatment incubation, to determine the K⁺-evoked astroglial releases of L-glutamate and D-serine, astrocytes were incubated in MK-ACSF (50.0 mM K⁺: black circles) or HK-ACSF (100.0 mM K⁺: blue circles) containing the same concentration of CLZ and VPA during pretreatment incubation for 20 min, and then incubate medium (MK-ACSF or HK-ACSF) was collected for analysis. Ordinate: mean ± SD (n = 6) of extracellular levels of l-glutamate and d-serine (μM). Abscissa: concentration of CLZ (μM). * p < 0.05 and ** p < 0.01 vs. CLZ free, and @@ p < 0.01 vs. VPA free (control) by MANOVA with Tukey’s post hoc test.

Figure 6

Subchronic effects of VPA (1000 and 3000 μM) (A) and CLZ (3, 10, and 30 μM) (B) on Cx43 expression in cytosol (blue columns) and plasma membrane (green columns) of astrocytes. Acute effects of CLZ (3, 10, and 30 μM) on Cx43 expression in cytosol and plasma membrane of subchronically treated astrocytes with therapeutic-relevant concentration of VPA (1000 μM) (C). Ordinate: mean ± SD (n = 6) of relative protein level of Cx43. Abscissa: concentration of VPA and CLZ (μM). * p < 0.05 and ** p < 0.01 vs. agent free by MANOVA with Tukey’s post hoc test. Panels 6(D) and 6(E) pseudo-gel images using Simple Western results using anti-GAPDH and anti-Cx43 antibody for blotting of cytosol and plasma membrane fractions, respectively.