Insulin-like growth factor-1 promotes G(1)/S cell cycle progression through bidirectional regulation of cyclins and cyclin-dependent kinase inhibitors via the phosphatidylinositol 3-kinase/Akt pathway in developing rat cerebral cortex

Abstract

Although survival-promoting effects of insulin-like growth factor-1 (IGF-1) during neurogenesis are well characterized, mitogenic effects remain less well substantiated. Here, we characterize cell cycle regulators and signaling pathways underlying IGF-1 effects on embryonic cortical precursor proliferation in vitro and in vivo. In vitro, IGF-1 stimulated cell cycle progression and increased cell number without promoting cell survival. IGF-1 induced rapid increases in cyclin D1 and D3 protein levels at 4 h and cyclin E at 8 h. Moreover, p27(KIP1) and p57(KIP2) expression were reduced, suggesting downregulation of negative regulators contributes to mitogenesis. Furthermore, the phosphatidylinositol 3-kinase (PI3K)/Akt pathway specifically underlies IGF-1 activity, because blocking this pathway, but not MEK (mitogen-activated protein kinase kinase)/ERK (extracellular signal-regulated kinase), prevented mitogenesis. To determine whether mechanisms defined in culture relate to corticogenesis in vivo, we performed transuterine intracerebroventricular injections. Whereas blockade of endogenous factor with anti-IGF-1 antibody decreased DNA synthesis, IGF-1 injection stimulated DNA synthesis and increased the number of S-phase cells in the ventricular zone. IGF-1 treatment increased phospho-Akt fourfold at 30 min, cyclins D1 and E by 6 h, and decreased p27(KIP1) and p57(KIP2) expression. Moreover, blockade of the PI3K/Akt pathway in vivo decreased DNA synthesis and cyclin E, increased p27(KIP1) and p57(KIP2) expression, and prevented IGF-1-induced cyclin E mRNA upregulation. Finally, IGF-1 injection in embryos increased postnatal day 10 brain DNA content by 28%, suggesting a role for IGF-1 in brain growth control. These results demonstrate a mitogenic role for IGF-1 that tightly controls both positive and negative cell cycle regulators, and indicate that the PI3K/Akt pathway mediates IGF-1 mitogenic signaling during corticogenesis.

Figure 1.

Effects of IGF-1 on DNA synthesis, cell cycle progression, and cortical precursor cell number in vitro. A, DNA synthesis was measured in E14.5 rat cortical precursors using [³H]thymidine incorporation assay. Addition of IGF-1 (0.03–20 ng/ml) to media elicited a dose-dependent increase in DNA synthesis at 24 h. Data are representative of one of three experiments, four wells per group. Statistical analysis was performed with ANOVA followed by Dunnett's posttest. cpm, Absolute counts per minute per well. B, BrdU (10 μm) was added 2 h before culture termination at 24 h. BrdU-labeled cells and total cells (observed under phase microscopy) were counted in 10 randomly selected fields. IGF-1 (10 ng/ml) increased the BrdU labeling index by twofold, indicating the factor stimulated G₁/S progression. Data are representative of three experiments, three dishes per group per experiment. C, IGF-1 also increased the number of cells labeled for the M-phase marker phosphohistone H3 by 39%. Phosphohistone H3-labeled cells were counted in three 1 cm rows in three dishes per group. D, Proliferating cells were labeled during a 2 h BrdU (10 μm) pulse at 22 h, and the absolute number of BrdU-labeled cells was counted at 24 and 36 h. At 24 h, the number of BrdU-positive cells increased by twofold after IGF-1 exposure. At 36 h, the number of BrdU-labeled cells increased in both groups with twice as many in the IGF-1 group, indicating that cells that entered S phase between 22 and 24 h progressed through the cell cycle and divided in both groups. E, Quantification of cell number at 2, 24, and 36 h in control and IGF-1-treated cultures. The y-axis corresponds to the mean number of cells counted in 10 different fields per 35 mm dish. Whereas no difference in cell number among groups was observed between 2 and 24 h, a 19% increase was detected in IGF-1-treated cultures at 36 h. Note that no change in cell number was detected in control group at any time analyzed. Data are from three experiments. Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2.

Effects of IGF-1 on cortical precursor cell survival and apoptosis. Cells incubated in 35 mm dishes were counted in 10 randomly selected fields, in three dishes per group. A, PI (red signal) and FDA (green signal) were used to distinguish dead cells (arrows) and living cells at 2, 8, and 24 h. Percentage cell survival was determined as the ratio of FDA-stained cells over total cells stained for FDA and PI. This value was then normalized to the value of the control group at 2 h, which was set at 100%. No significant difference was observed at each time among groups. Data are expressed as percentage ± SEM of 2 h control value (p > 0.05). B, Apoptotic cells immunolabeled for activated caspase-3 were counted and are expressed as percentage of total cell number observed under phase microscopy. Whereas the percentage of activated caspase-3-labeled cells increased from 5% at 8 h to 13% at 24 h among groups, no difference between IGF-1 and control groups was observed at each time analyzed (p > 0.05). Data are from three experiments, three dishes per group per experiment. Values shown represent the mean ± SEM.

Figure 3.

Effects of IGF-1 on positive regulators of the cell cycle. A, Western blot showing that IGF-1 increased cyclin D1 protein levels by 30% as early as 4 h. Data are from five experiments. However, no changes in cyclin D1 mRNA levels, measured by Q-RT-PCR, were observed at 4 h. Data are representative of three experiments, three dishes per group. B, IGF-1 also increased cyclin D3 protein levels, by 45% at 4 h (4 experiments), but no changes in cyclin D3 mRNA levels were detected (data are representative of 3 experiments). C, Cyclin E protein levels increased by 49% at 8 h of culture, after IGF-1 exposure. Data are from six experiments. In addition, IGF-1 increased cyclin E mRNA levels by 25% at 8 and 24 h. Data are representative of three experiments. Autoradiographic signals were quantified by densitometric analysis. Levels of protein of interest were normalized to β-actin. Levels of cyclin mRNA were normalized to GAPDH. Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01.

Figure 4.

Effects of IGF-1 on negative regulators of the cell cycle. p27^KIP1 and p57^KIP2 protein levels were analyzed by Western blotting, and mRNA levels by Q-RT-PCR, at different times of culture. A, p27^KIP1 protein levels were decreased by 25% at 8 h, 70% at 12 h, and 40% at 24 h in IGF-1-treated cultures. Data are from three to four experiments. A 20% reduction in p27^KIP1 mRNA levels was measured at 12 and 24 h after IGF-1 treatment. Data are representative of three experiments. B, IGF-1 treatment elicited a 17% decrease in p57^KIP2 protein levels at 12 h (3 experiments), and a 20% decrease in mRNA levels at 12 and 24 h (data are representative of 3 experiments). p57^KIP2-specific signal (arrows) was determined by comparison to signal obtained from brain homogenates of E14.5 wild-type (p57^+/+; positive control) and p57^KIP2 gene knock-out (p57^−/−; negative control) mouse embryos. Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01.

Figure 5.

IGF-1 activates the PI3K/Akt and MEK/ERK signaling pathways. A, B, IGF-1 increases the phosphorylation of Akt at 30 min (A) and ERK at 10 min (B). IGF-1 treatment preferentially induced phosphorylation of ERK2 (p42), whereas phospho-ERK1 (p44) signal was barely detectable. C, IGF-1 treatment increased the phosphorylation of GSK-3β at 30 min. Data are from three experiments. Values shown represent the mean ± SEM. *p < 0.001.

Figure 6.

Blockade of PI3K/Akt signaling, but not the MEK/ERK pathway, inhibits IGF-1 stimulatory effects on DNA synthesis of cortical precursors. A, Cells were preincubated with PI3K inhibitor LY294002 (LY) for 30 min before IGF-1 addition, and DNA synthesis was assayed at 12 h. IGF-1 stimulation of DNA synthesis was markedly reduced by 10 μm LY294002. Data are representative of three experiments. B, Western blot showing that IGF-1-induced Akt phosphorylation was blocked by the PI3K inhibitor LY294002 (10 μm), at 30 min of incubation. C, Cell survival at 12 h was estimated by counting living and dead cells in 10 randomly selected fields using phase microscopy combined with PI (staining dead cells), or using DAPI staining to assess living and dead cells based on nuclear morphology. No differences were observed among groups (p > 0.05). Data are from three experiments. D–F, Cortical precursors were preincubated with different concentrations of MEK inhibitor PD98059 (D) or U0126 (F) for 30 min before the addition of IGF-1 for 12 h. Blockade of the MEK/ERK pathway with either drug resulted in an inhibitor dose-dependent decrease in DNA synthesis in both control and IGF-1-treated groups. Data are representative of three experiments. The Western blot in E shows that IGF-1-induced ERK phosphorylation is blocked by MEK inhibitor PD98059 (PD) (50 μm), at 10 min of culture. Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01.

Figure 7.

Effects of IGF-1 intracerebroventricular injections on DNA synthesis, S-phase entry, and apoptosis in the cerebral cortex of developing embryos. IGF-1 (30 ng) was injected into the lateral ventricles of E16.5 rat embryos. A, DNA synthesis was assessed in the cerebral cortex at 6 and 12 h after IGF-1 injection, using percentage [³H]thymidine incorporation assay. [³H]Thymidine was injected subcutaneously into pregnant dams 1 h before the killing. IGF-1 elicited a 12% increase in [³H]thymidine incorporation at 6 h (PBS-treated animals, n = 29; IGF-1-treated animals, n = 30; 5 pregnant dams) and 26% increase at 12 h (PBS-treated animals, n = 20; IGF-1-treated animals, n = 17; 4 pregnant dams) compared with the control group. B, C, Analysis of apoptosis 12 h after injection using activated caspase-3 as a marker. B, Rare cleaved caspase-3-labeled cells (arrows), exhibiting cytoplasmic signal, were revealed in the proliferating and postmitotic cortical zones of both PBS- and IGF-1-injected embryos. C, Western blotting analyses performed on the cortices of injected embryos revealed two bands (17 and 19 kDa) corresponding to the large fragments of activated caspase-3. No difference in protein levels assessed by densitometry was observed between control and IGF-1 treated animals (p > 0.05). Data are representative of three experiments, four animals per group. D, Analysis of cells in S-phase using BrdU immunohistochemistry. BrdU was injected into pregnant dams 1 h before the killing. BrdU-labeled cells were counted over total cells in the VZ of the dorsolateral cortex. IGF-1 intracerebroventricular injection elicited a 26% increase in the BrdU LI at 12 h. Data are from four animals per group. LV, Lateral ventricle. Values shown represent the mean ± SEM. *p < 0.01; **p < 0.001.

Figure 8.

Effects of anti-IGF-1 antibody on [³H]thymidine incorporation in rat cortical precursor cultures and in mouse embryonic cortex in vivo. A, The neutralizing activity of anti-IGF-1 antibody on IGF-1-induced mitosis was defined in E14.5 rat cortical precursor cultures. Mouse IGF-1 (3 ng/ml) or bFGF (10 ng/ml) were preincubated alone or in the presence of anti-IGF-1 (10 μg/ml) or anti-IgG (10 μg/ml) antibody under agitation for 2 h in 1 ml of culture medium. Then, growth factors or mixtures were applied to cultures, and [³H]thymidine incorporation was measured at 24 h. Mouse IGF-1 increased DNA synthesis by 70%, an effect that was abolished when the growth factor was preincubated with anti-IGF-1, but not with anti-IgG antibody. In the presence of bFGF, [³H]thymidine incorporation increased by fourfold, and the effect was not altered when preincubated with either anti-IGF-1 or anti-IgG antibody. Antibodies alone had no effects on [³H]thymidine incorporation in vitro. B, After intracerebroventricular injection of antibodies into E14.5–E16.5 mouse embryos, [³H]thymidine incorporation was assayed in the cortices at 4 h. IGF-1 neutralizing antibody (anti-IGF-1; 0.5, 1, or 2 μg using a 0.5 μg/μl solution) reduced DNA synthesis by 28% in the cortex of E14.5–E16.5 mice, compared with control (anti-IgG) (anti-IgG-treated animals, n = 17; anti-IGF-1-treated animals, n = 13; 4 pregnant dams; the data from the different doses were combined because effects were overlapping). Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 9.

Effects of IGF-1 intracerebroventricular injections on the PI3K/Akt and MEK/ERK pathways. A, IGF-1 injections into the lateral ventricles of E16.5 rat embryos (30 ng per embryo) resulted in a fourfold increase in the phosphorylation of Akt by 30 min, revealed by Western blotting. B, The phosphorylation of GSK-3β increased by twofold at 30 min. C, IGF-1 injection induced a 40% increase in ERK phosphorylation by 10 min. Densitometric values of phosphoproteins were normalized to values of respective total proteins. Data are representative of three experiments, three to four animals per group. Values shown represent the mean ± SEM. *p < 0.01.

Figure 10.

Effects of IGF-1 intracerebroventricular injections on the expression of cyclins D1 and E, and the CKIs p27^KIP1 and p57^KIP2. A, IGF-1 injection (30 ng per embryo) produced a 56% increase in cyclin D1 protein level at 6 h. Values represent the mean of seven embryos per group obtained from two pregnant dams. However, no changes in mRNA levels were observed using Q-RT-PCR. B, Injection of IGF-1 elicited a twofold increase in cyclin E protein levels at 6 h, and a 60% increase in cyclin E transcript at 6 and 12 h. C, IGF-1 injection decreased p27^KIP1 and p57^KIP2 mRNA levels by 25 and 22%, respectively, in the same samples at 6 h. Data are representative of three experiments, four to five animals per group. Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 11.

Effects of blocking PI3K/Akt signaling on DNA synthesis, expression of cell cycle regulators, and IGF-1 mitogenesis in the cortex of developing embryos. A, Injection of the PI3K/Akt inhibitor LY294002 (LY) (184 ng per embryo; 3 μl of 200 μm solution) into E16.5 rat embryos efficiently decreased ongoing Akt phosphorylation at 30 min. B, C, Intracerebroventricular injection of LY294002 elicited a 13% decrease in DNA synthesis at 6 h (data are representative of 3 experiments, 4 animals per group) (B), and 23% reduction in cyclin E mRNA levels at 6 h (data are from 3 experiments, 4 animals per group) (C). D, Conversely, injection of LY294002 increased p27^KIP1 and p57^KIP2 mRNA levels by 36 and 27%, respectively, at 6 h. Data are from three experiments, four animals per group. E, Western blot showing that intracerebroventricular injection of LY294002 30 min before IGF-1 injection blocked IGF-1-induced phosphorylation of Akt at 30 min. F, Whereas IGF-1 injection increased cyclin E mRNA levels by 30% at 6 h, preinjection of LY294002 prevented IGF-1 stimulatory effect on cyclin E expression (Con group, n = 4; IGF group, n = 3; LY294002 group, n = 4; LY294002 plus IGF-1 group, n = 4; data are from 3 experiments). Values shown represent the mean ± SEM. *p < 0.05; **p < 0.01.