Insulin promotes glycogen storage and cell proliferation in primary human astrocytes

Abstract

Introduction: In the human brain, there are at least as many astrocytes as neurons. Astrocytes are known to modulate neuronal function in several ways. Thus, they may also contribute to cerebral insulin actions. Therefore, we examined whether primary human astrocytes are insulin-responsive and whether their metabolic functions are affected by the hormone.

Methods: Commercially available Normal Human Astrocytes were grown in the recommended medium. Major players in the insulin signaling pathway were detected by real-time RT-PCR and Western blotting. Phosphorylation events were detected by phospho-specific antibodies. Glucose uptake and glycogen synthesis were assessed using radio-labeled glucose. Glycogen content was assessed by histochemistry. Lactate levels were measured enzymatically. Cell proliferation was assessed by WST-1 assay.

Results: We detected expression of key proteins for insulin signaling, such as insulin receptor β-subunit, insulin receptor substrat-1, Akt/protein kinase B and glycogen synthase kinase 3, in human astrocytes. Akt was phosphorylated and PI-3 kinase activity increased following insulin stimulation in a dose-dependent manner. Neither increased glucose uptake nor lactate secretion after insulin stimulation could be evidenced in this cell type. However, we found increased insulin-dependent glucose incorporation into glycogen. Furthermore, cell numbers increased dose-dependently upon insulin treatment.

Discussion: This study demonstrated that human astrocytes are insulin-responsive at the molecular level. We identified glycogen synthesis and cell proliferation as biological responses of insulin signaling in these brain cells. Hence, this cell type may contribute to the effects of insulin in the human brain.

Figure 1. mRNA expression in human astrocytes (black bars) in comparison to human myotubes (light grey bars) and human adipocytes (dark grey bars).

(A) Insulin receptor, (B) IRS-1, (C) IRS-2, (D) GLUT1, (E) GLUT3, (F) GLUT4. mRNA expression was normalized for mRNA of the housekeeping gene Rps13. Bars represent means + SEM. N = 3. There were significant differences between the groups in all mRNA expressions analyzed (ANOVA, all p≤0.0254) except for GLUT3 mRNA (ANOVA, p = 0.2). * indicates significant difference from astrocytes (Tukey Kramer post hoc test p<0.05).

Figure 2. mRNA expression of insulin receptor isoforms A and B as well as GLUT 2 in human astrocytes.

(A) mRNA expression of insulin receptor isoform A and B in human astrocytes. Given are means + SEM, N = 3. *  =  there were significant differences between groups (Student's t-test, p = 0.0181). (B) mRNA expression of GLUT2 in human astrocytes (black bar), human myotubes (white bar), human adipocytes (dark grey bar), and HepG2 (human hepatocellular carcinoma cell line, light grey bar). mRNA expression was normalized for mRNA of the housekeeping gene Rps13. Bars represent means + SEM. N = 3. There were significant differences between the groups (ANOVA, p = 0.0006). * indicates significant difference from astrocytes (Tukey Kramer post hoc test p<0.05).

Figure 3. Protein expression, insulin-stimulated Akt phosphorlyation, and insulin-stimulated phosphatidylinositiol 3′-phosphorylation in human astrocytes.

(A) Equal amounts of total cell lysates of two independent experiments were loaded onto a SDS-PAGE gel. The indicated proteins were detected using specific antibodies. (B) Cells were stimulated with the indicated insulin concentrations for 15 minutes and lysed afterwards. Akt phosphorylation was detected by a phospho-specific antibody. A representative western blot is shown in the lower part. Bars in the diagram represent means + SEM. N = 6. There were significant differences between the groups (ANOVA, p = 0.0108). * indicates significant difference from 0 nM insulin (Tukey Kramer post hoc test p<0.05). (C) Following insulin stimulation with insulin for 30 minutes, cells were harvested, and lysates were immunoprecipitated with anti IRS-1 antibodies and a PI-3 kinase assay was performed using L-α-phosphatidylinositol as substrate. A representative image is shown in the lower part. Bars in the diagram represent means + SEM. N = 3. There were significant differences between the groups (ANOVA, p = 0.0267). * indicates significant difference from 0 nM insulin (Tukey Kramer post hoc test p<0.05).

Figure 4. Effects of insulin stimulation and glucose concentration on lactate secretion of human astrocytes.

Prior to experiment, NHAs were either starved in medium containing 1 g/l glucose or kept in medium with 4.5 g/l. Medium was than replaced by new medium with the same glucose concentrations without (white bars) or with 50 nM insulin (grey bars). After 8 hours, lactate concentration in the supernatant was measured. Bars represent means of five independent experiments + SEM. There were no significant differences between groups (ANOVA, p = 0.4).

Figure 5. Effects of insulin stimulation on glycogen synthesis (A) and glycogen content (B) of human astrocytes.

(A) The indicated cells were preincubated with the PI-3 kinase inhibitor LY294002 for 1 hour. Cells were stimulated with the indicated concentrations of insulin for 3 hours. Glycogen synthesis in the absence of insulin and LY294002 was set as 100%. Bars represent means of at five independent experiments + SEM. There were significant differences between the groups (ANOVA, p<0.0001). * indicates significant difference (Tukey Kramer post hoc test p<0.05). (B) Prior to experiment, NHAs were either starved in medium containing 1 g/l glucose (left lane) or kept in medium with 4.5 g/l (right lane). Medium was than replaced by new medium with the same glucose concentrations without (upper panels) or with 50 nM insulin (lower panels). After three hours of stimulation, cells were stained for glycogen (pink). Shown are representative examples of at least three independent experiments.

Figure 6. Effects of insulin stimulation on human astrocytes' proliferation.

After starvation for at least 24 hours, cells were treated with the indicated concentrations of insulin for 3 days. Thereafter, cell proliferation was assessed by WST-1 assay. Bars represent means of six independent experiments + SEM. There were significant differences between the groups (ANOVA, p = 0.0007). * indicates significant difference from 0 nM (Tukey Kramer post hoc test p<0.05).