Figure 1.
Effects of baicalein on the HIF1α activity and expression of Epo and VEGF in cortical neurons.
(A) Primary cultured cortical neurons co-transfected with pHREEPO-Luc and pRL-TK were treated with baicalein (BA) at indicated concentrations (3.5 nM~35 μM) or CoCl2 (0.4 mM) for luciferase activity assay of HRE-driven gene expression as an index of HIF activity. Inset in A: EC50 of BA on HREEPO-driven luciferase. (B) qRT-PCR of Epo and VEGF mRNA of RNA extracted BA-treated neurons at indicated time or concentrations. (C) Upper panel: RT-PCR analysis of HIF1α mRNA in neurons transfected with scrambled RNA (Scr) or siHif1a. Lower panel: qRT-PCR analysis of Epo and VEGF mRNA in 3.5 μM BA-treated neurons transfected with Scr or siHif1a. (D) ELISA of Epo and VEGF of cell lysate of BA-treated neurons. Data represent means ± SEM (n=3). *p<0.05, **p<0.01 and ***p<0.001 versus vehicle-treated control by one-way ANOVA and Newman-Keuls multiple comparison posttest; # p<0.05 and # # # p<0.001 versus the Scr-Ctrl by unpaired t-test.
Figure 2.
Effects of 12/15-LOX knockdown on the baicalein-induced Epo and VEGF gene expression.
(A) qRT-PCR of 12/15-LOX mRNA in neurons and astrocytes. (B) RT-PCR analysis of 12/15-LOX mRNA in neurons transfected with scrambled RNA (Scr) or siAlox15 for 72 h. (C, D) qRT-PCR of VEGF (C) and Epo (D) mRNA in siAlox15 or Scr-transfected neurons with or without 3.5 μM BA treatment for 12 h. In (A), ***p<0.01 versus Astrocytes group (n=3). In (C) and (D), *p<0.05 and **p<0.01 versus control (Ctrl); # p<0.05 versus the Scr-Ctrl (n=3).
Figure 3.
Effects of PI3K inhibitors on the baicalein–activated HIF1α and Epo/VEGF gene transcription in neurons.
Neurons were treated with 3.5 μM BA with or without 1 h pretreatment with the pan PI3K inhibitor LY294002 (LY, 10 μM), PI3Kα/β inhibitor (PI3K α inhibitor-2, 50 nM), or PI3Kγ inhibitor (CAY10505, 200 nM). (A) Cells were harvested at 30 min after the BA treatment for Western blotting of pAkt and Akt. (B) Dual-luciferase activity assay of pHREEPO-Luc expression in cells 24 h after the treatments. (C) qRT-PCR of Epo and VEGF mRNA expression in cells 24 h after the treatments. (D) Anti-HIF1α-based ChIP assay was performed at 0, 1, 3 h after the BA treatment to analyze the HIF1α-associated rat Epo promoter fragment (Epo-p), HRE-containing Epo enhancer fragment (Epo-e) and HRE-containing Vegf promoter fragment (Vegf-p) by PCR (left panel) and qPCR (right panel). *p<0.05, **p<0.01, ***p<0.001 versus control; # p<0.05, # # p<0.01, # # # p<0.001 versus the BA-treated group (n=3).
Figure 4.
Effects of extracellular Epo/VEGF neutralization and PI3K inhibitor treatment on the baicalein neuroprotection against excitotoxicity, and baicalein effect on neuronal excitability.
(A, B) Neurons were pre-treated with baicalein (BA, 3.5 μM) with or without LY294002 (LY, 10 μM), anti-EPO, or anti-VEGF, or normal goat IgG antibodies at the concentrations as indicated for 12 h, followed by the glutamate (25 μM)/ NMDA (25 μM) (Glu/NMDA) treatment. Neurons were stained with DAPI to visualize nuclear condensation for cell apoptosis. (A) Representative fluorescent micrographs with anti-Epo and anti-VEGF antibodies at 5 μg/ml; (B) quantitative result of the percent of apoptotic cells. Scale bar: 20 μm. ***p<0.001 versus vehicle control; # # p<0.01 and # # # p<0.001 versus the BA + Glu/NMDA-treated group; + + + p<0.001 versus the BA/IgG + Glu/NMDA-treated group (unpaired t-test). (n=5). (C) Miniature GABAA-receptor-mediated currents with the example traces showing before and after bath application of BA (30 μM) (upper panel) and summary of BA effect on the amplitude and frequency (lower panel). Scale bars: 1 min/50 pA. (D) Left panel: Representative traces of membrane responses of CA1 pyramidal neurons evoked by the 1-s depolarizing (300 pA) and hyperpolarizing (-100 pA) current pulses before (Ctrl) and after BA application. Scale bars: 250 ms/50 mV. Right panel: Summary of the BA effect on membrane potential (Vm, n=3) and input resistance (Rin, n=4) of CA1 pyramidal neurons.
Figure 5.
Baicalein effect on astrocytic Epo/VEGF expression and its HIF1α dependency and correlation with the PHD inhibitor effect in astrocytes.
(A, C) Primary cultured astrocytes were treated with baicalein (BA) at the indicated concentrations for 24 h, followed by mRNA extraction for qRT-PCR analysis of Epo, VEGF,(A) and TNFα (C) transcripts. (B) Culture medium of astrocytes was collected 24 h after the BA treatment for ELISA analysis of Epo and VEGF. (D, E) For the HIF1α dependency experiment, the scrambled RNA- or siHif1a-transfected astrocytes were treated with BA (35 μM) for 24 h, followed by qRT-PCR analysis of VEGF (D) and Epo (E) mRNA. (F) Protein levels of PHD2 in neurons versus astrocytes as analyzed by Western blotting. (G) qRT-PCR of Epo and VEGF mRNA in 0.5 mM or 1.5 mM DMOG-treated neurons and astrocytes *p<0.05, **p<0.01, ***p<0.001 versus control; # p<0.05 and # # p<0.01 versus the Scr-BA-treated group in (D), and versus Neurons-Ctrl or Neurons-DMOG group in G (n=3).
Figure 6.
Effects of PI3K inhibitors on the baicalein-induced Akt phosphorylation, astrocyte-mediated neuroprotection, and astrocytic Epo/VEGF expression.
(A, E, F) Astrocytes were treated with baicalein (BA) at indicated concentrations (3.5, 10 or 35 μM) or pretreated with LY294002 (LY, 10 μM), PI3K α β inhibitor (PI3K α inhibitor-2, 50 nM), or PI3Kγ inhibitor (CAY10505, 200 nM) for 1 h, followed by the BA treatment. (A) Total proteins were harvested 30 min after the treatment for Western blotting of pAkt and Akt. (B, C) Cortical neurons were incubated with astrocyte-conditioned medium (ACM) from astrocytes treated with 35 μM BA in the presence or absence of LY pretreatment. Three hours after the ACM incubation, neurons were treated with glutamate (25 μM)/NMDA (25 μM) (Glu/NMDA) treatment for 21 h, followed by TUNEL assay to visualize (B) and quantify (C) apoptotic cells. Scale bar in (B): 20 μm. (D) ELISA analysis of Epo and VEGF concentrations in ACM used in (B) and (C). (E, F) qRT-PCR of Epo (E) and VEGF (F) transcripts in astrocytes 24 h after the treatment. *p<0.05 and ***p<0.001 versus the respective control group; # # # p<0.001 versus the BA- or BA-ACM-treated group (n=3)..
Figure 7.
Cell type-specific signaling mechanism of baicalein-induced endogenous Epo and VEGF production from neurons and astrocytes for neuroprotection.
Baicalein treatment in neurons activates class I PI3K/Akt to induce HIF1α-mediated Epo/VEGF expression with a minimal effective concentration at 35 nM. Its direct inhibition of 12/15-LOX additionally contributes to the induction of neuronal VEGF, but not Epo. In astrocytes where PHD2 is in low abundance and 12/15-LOX is lacking, high concentration of baicalein (35 μM in minimum) is required to activate class I PI3K/Akt. However, baicalein-induced upregulation of astrocytic Epo/VEGF is sensitive to LY294002 (LY) but not the selective class I PI3K inhibitors, suggesting that other LY294002-inhibitable protein kinases (PKs) mediate the effect. Furthermore, baicalein-induced astrocytic Epo expression is HIF1α-independent and possibly by HIF2α that reportedly mediates astrocytic Epo gene transcription and can also be stabilized when PHD2 is inhibited. Since brain neurons and astrocytes may access nanomolar and micromolar concentration of baicalein respectively when the compound is applied peripherally, the increased production of Epo and VEGF from both cell types may thus be converged to provide neuroprotection against excitotoxic or other neurotoxic insults.