Central nervous system delivery of the antipsychotic olanzapine induces hepatic insulin resistance

Abstract

Objective: Olanzapine (OLZ) is an atypical antipsychotic whose clinical efficacy is hampered by side effects including weight gain and diabetes. Recent evidence shows that OLZ alters insulin sensitivity independent of changes in body weight and composition. The present study addresses whether OLZ-induced insulin resistance is driven by its central actions.

Research design and methods: Sprague-Dawley rats received an intravenous (OLZ-IV group) or intracerebroventricular (OLZ-ICV group) infusion of OLZ or vehicle. Glucose kinetics were assessed before (basal period) and during euglycemic-hyperinsulinemic clamp studies.

Results: OLZ-IV caused a transient increase in glycemia and a higher rate of glucose appearance (R(a)) in the basal period. During the hyperinsulinemic clamp, the glucose infusion rate (GIR) required to maintain euglycemia and the rate of glucose utilization (R(d)) were decreased in OLZ-IV, whereas endogenous glucose production (EGP) rate was increased compared with vehicle-IV. Consistent with an elevation in EGP, the OLZ-IV group had higher hepatic mRNA levels for the enzymes glucose-6-phosphatase and phosphoenolpyruvate carboxykinase. Phosphorylation of hypothalamic AMP-activated protein kinase (AMPK) was increased in OLZ-IV rats compared with controls. Similarly, an intracerebroventricular infusion of OLZ resulted in a transient increase in glycemia as well as a higher R(a) in the basal period. During the hyperinsulinemic period, OLZ-ICV caused a decreased GIR, an increased EGP, but no change in R(d). Furthermore, OLZ-ICV rats had increased hepatic gluconeogenic enzymes and elevated hypothalamic neuropeptide-Y and agouti-related protein mRNA levels.

Conclusions: Acute central nervous system exposure to OLZ induces hypothalamic AMPK and hepatic insulin resistance, pointing to a hypothalamic site of action for the metabolic dysregulation of atypical antipsychotics.

FIG. 1.

Time course (means ± SE) of plasma levels of glucose (A), insulin (B), and FFAs (C) during basal and hyperinsulinemic-euglycemic clamp periods. Minute 0 represents sampling before a prime-continuous (a bolus of 1.5 mg/kg and 1.0 mg/kg/h) intravenous (IV) infusion of OLZ (OLZ-IV, n = 10) or vehicle (n = 8) *Statistics denote comparison to respective 0 min and vehicle time point. Two-way ANOVA followed by Duncan test, P < 0.05.

FIG. 2.

Effects of intravenous infusion of OLZ (OLZ-IV, n = 10) or vehicle (n = 8) on (means ± SE) glucose appearance rate (R_a) (A), GIR (B), glucose disappearance rate (R_d) (C), and EPG (D) before and during stead-state hyperinsulinemic-euglycemic clamp condition. Liver (E), G6Pase, and PEPCK (F) mRNA levels (means ± SE) of rats intravenously (iv) infused with OLZ following the hyperinsulinemic-euglycemic clamp. *Statistics denote comparison to vehicle group; t test, P < 0.05.

FIG. 3.

Time course (means ± SE) of plasma levels of glucose (A), insulin (B), and FFAs (C) during basal and hyperinsulinemic-euglycemic clamp periods. Minute 0 represents sampling before a prime-continuous (a bolus of 110 μg and 73.4 μg/h) intracerebroventricular (icv) infusion of OLZ (OLZ-ICV, n = 8) or vehicle (n = 10). *Statistics denote comparison to respective 0 min and vehicle time-point. Two-way ANOVA followed by Duncan test, P < 0.05.

FIG. 4.

Effects of intracerebroventricular (icv) infusion of OLZ (OLZ-ICV, n = 8) or vehicle (n = 10) on (means ± SE) glucose appearance rate (R_a) (A), GIR (B), glucose disappearance rate (R_d) (C), and EPG (D) before and during stead-state hyperinsulinemic-euglycemic clamp condition. Liver (E), G6Pase, and PEPCK (F) mRNA levels (means ± SE) of rats intracerebroventricularly infused with OLZ following the hyperinsulinemic-euglycemic clamp. *Statistics denote comparison to vehicle group; t test, P < 0.05.

FIG. 5.

Immunoblot for phospho-AMPK, AMPK, and β-tubulin (A) and optical densitometry of the phospho-AMPK/AMPK ratio (B) from hypothalamus of intravenously infused rats with Vehicle (Veh-IV, n = 8) or OLZ (OLZ-IV, n = 10). Control samples from insulin-induced hypoglycemia (Positive; Pos) or ad libitum–fed (Negative; Neg) rats were also included. C: Relative gene expression (means ± SE) in the hypothalamus of rats intracerebroventricularly infused with OLZ (OLZ-ICV, n = 8) or vehicle (n = 10) following the hyperinsulinemic-euglycemic clamp. *Statistics denote comparison to vehicle group; t test, P < 0.05.