Protein kinase C regulates human pluripotent stem cell self-renewal

Abstract

Background: The self-renewal of human pluripotent stem (hPS) cells including embryonic stem and induced pluripotent stem cells have been reported to be supported by various signal pathways. Among them, fibroblast growth factor-2 (FGF-2) appears indispensable to maintain self-renewal of hPS cells. However, downstream signaling of FGF-2 has not yet been clearly understood in hPS cells.

Methodology/principal findings: In this study, we screened a kinase inhibitor library using a high-throughput alkaline phosphatase (ALP) activity-based assay in a minimal growth factor-defined medium to understand FGF-2-related molecular mechanisms regulating self-renewal of hPS cells. We found that in the presence of FGF-2, an inhibitor of protein kinase C (PKC), GF109203X (GFX), increased ALP activity. GFX inhibited FGF-2-induced phosphorylation of glycogen synthase kinase-3β (GSK-3β), suggesting that FGF-2 induced PKC and then PKC inhibited the activity of GSK-3β. Addition of activin A increased phosphorylation of GSK-3β and extracellular signal-regulated kinase-1/2 (ERK-1/2) synergistically with FGF-2 whereas activin A alone did not. GFX negated differentiation of hPS cells induced by the PKC activator, phorbol 12-myristate 13-acetate whereas Gö6976, a selective inhibitor of PKCα, β, and γ isoforms could not counteract the effect of PMA. Intriguingly, functional gene analysis by RNA interference revealed that the phosphorylation of GSK-3β was reduced by siRNA of PKCδ, PKCε, and ζ, the phosphorylation of ERK-1/2 was reduced by siRNA of PKCε and ζ, and the phosphorylation of AKT was reduced by PKCε in hPS cells.

Conclusions/significance: Our study suggested complicated cross-talk in hPS cells that FGF-2 induced the phosphorylation of phosphatidylinositol-3 kinase (PI3K)/AKT, mitogen-activated protein kinase/ERK-1/2 kinase (MEK), PKC/ERK-1/2 kinase, and PKC/GSK-3β. Addition of GFX with a MEK inhibitor, U0126, in the presence of FGF-2 and activin A provided a long-term stable undifferentiated state of hPS cells even though hPS cells were dissociated into single cells for passage. This study untangles the cross-talk between molecular mechanisms regulating self-renewal and differentiation of hPS cells.

Figure 1. An ALP activity-based high-throughput screening assay of chemical library for PKC inhibitors.

The ALP activity using 4-methylumbelliferyl phosphate in 201B7 hiPS cells in a 96-well plate was measured by fluorometry. Each dot on the graph represents the fluorescent intensity for each compound of the kinase inhibitor library. Dotted line indicates the level for DMSO as a control.

Figure 2. Effect of PKC inhibitor on FGF-2 signaling in hPS cells.

The phosphorylation levels in H9 hES cells were measured by AlphaScreen® SureFire® assay kit. The values of the y-axis are the ratio of each phosphorylation to each total signal protein. (A) The cells were stimulated with FGF-2 (100 ng/ml) in fresh medium without insulin after overnight starvation and incubated with (open square) or without GFX (5 µM, closed square) for 180 minutes. The data are represented as means ± SE (n = 3). *P<0.05. (B) The cells were stimulated with FGF-2 (100 ng/ml) in fresh medium without insulin after overnight starvation. Fifteen minutes after FGF-2 addition together with each inhibitor as indicated on the panel. The data are represented as means ± SE (n = 3). *P<0.05. (C) The cells were treated with FGF-2 (100 ng/ml), BMP-4 (100 ng/ml) or activin A (100 ng/ml) in fresh medium without insulin after overnight starvation. Fifteen minutes after the addition of each growth factor as indicated on the panel. The data are represented as means ± SE (n = 3). *P<0.05. (D) The cells after growth factor starvation were stimulated with FGF-2 (10 ng/ml) and activin A (10 or 100 ng/ml) together with U0126 (5 µM) and GFX (5 µM) or Gö6976 (5 µM) in fresh medium without insulin for 15 minutes. Fifteen minutes after the addition of each growth factor/inhibitor as indicated on the panel. The data are represented as means ± SE (n = 3). *P<0.05.

Figure 3. The effect of PKC on the morphologies of hPS cells with or without GFX.

(A) Phase-contrast image of 201B7 hiPS cells cultured in feeder-free hESF9 defined medium on FN 24 hours after treatment with DMSO, PMA (10 nM), 4α-PMA (10 nM), GFX (5 µM), PMA (10 nM) with GFX (5 µM), PMA (10 nM) with Gö6976 (5 µM), PMA (10 nM) with LY-294002 (50 µM), PMA (10 nM) with LiCl (1 mM), PMA (10 nM) with Y-27632 (10 µM), or PMA (10 nM) with U0126 (20 µM). An inactive PMA analogue, 4α-PMA is used as negative control. Scale bars, 200 µm. (B) Quantitative ALP-based assay of 201B7 hiPS cells cultured in feeder-free hESF9 medium with GFX (closed circle) or Gö6976 (open circle) as indicated concentrations. (C) Colony forming efficiency of dissociated single hPS cells cultured with or without GFX. Dissociated single 201B7 cells seeded at 250,000 cells/well were grown on a 6-well plate coated with FN (2 µg/cm²) in hESF9 medium supplemented with and without 1 µM GFX. A in 5 days and stained with ALP fast-red substrate. (D) Phase-contrast image of 201B7 hiPS cells or H9 hES cells cultured in feeder-free hESF9 medium with DMSO (open square), GFX (5 µM, gray square), or Gö6976 (5 µM, closed square). (E) Growth of cell colony area of hPS cells in the presence of GFX or Gö6976. The whole images of 201B7 cell colonies grown in a 6-well-plate coated with FN in the presence of DMSO, GFX or Gö6976 in hESF9 medium was measured by an analysis software, Cell-Quant. The images were captured every 12 hours in live cell imaging system Biostation CT. The data are represented as means ± SD (n = 3). (F) Cell growth of hPS cells in the presence of GFX or Gö6976. The numbers of H9 (open bars) and 201B7 cells (closed bars) grown in a 6-well-plate coated with FN in the presence of DMSO, GFX or Gö6976 in hESF9 medium were counted on 5 days. The data are represented as means ± SD (n = 3).

Figure 4. Specific-isoform of PKCs function in FGF-2 signaling.

(A) RT-PCR analysis of PKC isoform expression. Total RNA was extracted from the undifferentiated 201B7 hiPS cells cultured on feeder cells (CF-1) with KSR-based medium or the feeder cells. Primers were listed in Table S3. (B) Phosphorylation of PKCδ, ε, or ζ isoforms induced by FGF-2 (open square) with GFX (closed square). 201B7 hiPS cells were stimulated with FGF-2 (100 ng/ml) after overnight starvation and incubated with or without GFX (5 µM) for 180 minutes. The cells were lysed and followed by western blot analysis using an antibody detecting the phosphorylation or total protein amount of PKCδ, PKCε, or PKCζ. Protein content quantified from the gel blot images (n = 3). The values of the y-axis are the ratio of each phosphorylation to each total signal protein. (C) FGF-2 signaling in hPS cells with specific PKC isoforms-targeting siRNA. 201B7 iPS cells were transfected with specific PKCδ, ε, or ζ isoforms-targeting siRNA or non-targeting siRNA. The phosphorylation levels of the cells treated with FGF-2(100 ng/ml) after overnight starvation were measured by AlphaScreen® SureFire® assay kit. The values of the y-axis are the ratio of each phosphorylation to each total signal protein. The data are represented as means ± SE (n = 3). *P<0.05.

Figure 5. Single cell culture of hPS cells in the hESF9a_2i medium.

(A) Cell growth of dissociated single H9 hES cells cultured in each indicated condition for three passages. Cells were reseeded at the cell density of 1×10⁶ cells/well every 5 days. When the cells were passages, cell numbers were counted. Cell growth in the hESF9a_2i medium was significantly different (P<0.05) from hESF9 (FGF-2), FGF-2 + activin A, FGF-2 + activin A + U0126. Cell growth in hESF9a + GFX was significantly different (P<0.05) from hESF9 (FGF-2), FGF-2 + activin A, FGF-2 + activin A + U0126, and FGF2 + U0126. The data are represented as means ± SE (n = 3). (B) Gene expression in the hPS cells cultured in each indicated condition for three passages. The gene expression levels of NANOG, OCT3/4, FOXA2, T in the cells were measured by real-time RT-PCR. On the y axis, the gene expression level in the cells cultured with FGF-2 in a experiment was taken as 1.0. The data are represented as means ± SE (n = 3). *P<0.05. (C) Phase-contrast image of hPS cells grown on FN in hESF9a_2i medium for 3 passages. The cells were dissociated into single cells for passage, and reseeded at a ratio of 1∶3 - 1∶5 every five days. Scale bars, 200 µm. (D) OCT3/4 expression in hPS cells grown on FN in hESF9a_2i. The cells grown in hESF9a_2i as described above in Figure 5C were reseeded on a 6-well-plate and cultured for 5 days. The cells stained with anti-OCT3/4 antibody were visualized with Alexa Fluor 488 (upper panels). Nuclei were stained with Hoechst 33342 (blue). Scale bars, 200 µm. Whole cell images in whole plate were captured and OCT3/4 expression profiles were analyzed by Image Analyzer (lower panels). Antigen histogram (red); control histogram (green); Y axis is cell numbers and X axis is fluorescence intensity for anti-OCT3/4 antibody.

Figure 6. Model for the molecular mechanism of PKCs regulating self-renewal or differentiation in hPS cells.

Our study suggested a model that FGF-2 activates PI3K/AKT, MEK/ERK-1/2, and PKCε/δ/ζ. PKCε, δ, and ζ inactivates directly or indirectly GSK-3β by phosphorylation which promotes differentiation of hPS cells. PKCε and ζ activates ERK-1/2 which promotes differentiation of hPS cells. Activin A activates SMAD-2/3 which controls self-renewal and differentiation while activin A together with FGF-2 activates both ERK-1/2 and PKCs. Inhibition of both ERK-1/2 and PKCs pathway provides a metastable undifferentiated state of hPS cells. Blue arrow indicated pathway promoting hPS cell self-renewal and black arrow indicated pathway promoting hPS cell differentiation.