SEARCH

SEARCH BY CITATION

Keywords:

  • CREB;
  • stem cells;
  • RA;
  • neural progenitor cells;
  • differentiation

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Mouse embryonic stem (ES) cells can be induced by various chemicals to differentiate into a variety of cell types in vitro. In our study, retinoic acid (RA), one of the most important inducers, used at a concentration of 5 μM, was found to induce the differentiation of ES cells into neural progenitor cells (NPCs). During embryoid body (EB) differentiation, the level of active cyclic AMP response element-binding protein (CREB) was relatively high when 5 μM RA treatment was performed. Inhibition of CREB activity committed EBs to becoming other germ layers, whereas increased expression of CREB enhanced NPC differentiation. Moreover, RA increased the expression of active CREB by enhancing the activity of JNK. Our research suggests that CREB plays a role in RA-induced NPC differentiation by increasing the expression of active JNK. Anat Rec, 291:519–526, 2008. © 2008 Wiley-Liss, Inc.

Embryonic stem (ES) cells are able to differentiate into all three germ layers in vitro (Tonack et al.,2006). The ability to generate defined neural cell types from ES cells offers a powerful resource with which to study the early stage of differentiation (Yurek and Fletcher-Turner,2004; Buhnemann et al.,2006; Fukuda et al.,2006). Early exposure of embryoid bodies (EBs) to retinoic acid (RA), a vitamin A derivative, may drastically affect differentiated cell types (Yamada,1994; Pachernik et al.,2005; Salero and Hatten,2007). RA strongly promotes ectodermal differentiation at high concentrations, but induces cardiomyocyte differentiation at low concentrations (Niederreither et al.,2000; Canestro and Postlethwait,2007). In addition, cells derived from high-concentration RA-treated ES cells showed dorsal positional identities (Tsai et al.,2007). This finding suggests that high-level RA can be used as an effective inducer of neural differentiation. However, the exact mechanism by which RA induces ES cells along a neuronal pathway has not yet been elucidated.

Several studies have shown that the regulation of cellular differentiation by RA is a complex process that can be mediated in various cellular models by regulation of different signal molecules, including protein kinase A (PKA), protein kinase B, and mitogen-activated protein kinase (MAPKs; Crossthwaite et al.,2004; Riese et al.,2004; Flammer et al.,2006), which have also been identified to be cyclic AMP response element-binding protein (CREB) kinases. Active CREB is sufficient to induce gene expression in response to these signals; therefore, it plays a key role in controlling cell growth, survival, and cell cycle progression (Shankar and Sakamoto,2004; Dworet and Meinkoth,2006). For instance, CREB was reported to play a particularly important role in regulating neuronal function (Fukushima et al.,2007; Mao et al.,2007). CREB belongs to the family of basic leucine zipper transcription factors, which bind constitutively to specific DNA motifs. Its activity is triggered by phosphorylation on Ser133, which recruits the CREB binding protein to the initiator complex and thereby promotes transcription (Chrivia et al.,1993). A recent report of a transgenic mouse model has shown that CREB-deficient mice exhibit brain defects that affect the corpus callosum, anterior commissure, and lateral ventricles (Pittenger et al.,2006). This result suggests that CREB may be a key transcription factor in regulating neuronal differentiation. In addition, recent studies have shown that retinoic acid-induced differentiation of SH-SY5Y (Fernandes et al.,2007) and PC12 cells (Canon et al.,2004) requires the activation of CREB, independently of the transcriptional regulatory functions of RA receptors. This finding suggests that the activation of CREB may play an important role in neurogenesis.

In this study, we investigated the role of CREB in RA-induced ES cell commitment into neurons. The results showed that EB differentiation relies on the activation of CREB by RA in a dose- and time-dependent manner. Furthermore, CREB activation is regulated by RA by means of the JNK pathway. Thus, our data demonstrate that, at high concentrations, RA induces neural progenitor cell (NPC) differentiation by stimulating CREB signaling.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Vectors and Antibodies

The expression construct for CREB-VP16 (the transactivation domain of VP16 is fused to CREB's DNA binding domain so that CREB-VP16 exhibits constitutive CREB activity) was provided by Dr. Huang (Harbin Medical University, Harbin). The mCREB (containing a mutation at serine 133, which blocks phosphorylation and transactivation) expression vector was purchased from Clontech.

The antibodies used in this study were goat anti-nestin (Santa Cruz), mouse anti-β-tubulin III (Chemicon), rabbit anti–glial fibrillary acidic protein (anti-GFAP; Santa Cruz), rabbit anti-pCREB (Cell Signaling), mouse anti-pJNK (Cell Signaling), mouse anti-pp38 (Cell Signaling), mouse anti-pErk (Cell Signaling), mouse anti–glyceraldehyde-3-phosphate dehydrogenase (anti-GAPDH; Chemicon), goat anti-rabbit conjugated to fluorescein isothiocyanate (Santa Cruz), goat anti-rabbit conjugated to tetrarhodamine isothiocyanate (TRITC; Santa Cruz), rabbit anti-goat conjugated to fluorescein isothiocyanate (Santa Cruz), and goat anti-mouse conjugated to TRITC (Santa Cruz).

Cell Culture and Differentiation

Experiments were performed using R1 mouse ES cells. Controlled neural differentiation was achieved by following a culture protocol. R1 cells were cultured on mouse embryonic fibroblast-coated dishes containing Dulbecco's modified Eagle's medium (DMEM) and leukemia inhibitory factor (LIF, 1,000 U/ml). EBs were formed in bacteriological dishes for a period of 8 days, with the addition of 5 μM all-trans RA (Sigma) during the last 4 days.

For transient transfection assays, RA-treated EBs were transfected with mCREB or CREBVP16 using Lipofectamine LTX™ (Invitrogen) according to the manufacturer's protocol. For neural differentiation, EBs were transferred to fibronectin-coated dishes with 80/20 (Stem Cell) medium (80/20 supplemented with N2 and B27, from Gibco) for an additional 6 days. Cells were trypsinized, triturated through flame-polished Pasteur pipettes, exposed to poly-L-ornithine/laminin (LPO) -coated dishes, and cultured in 80/20 medium to support neural differentiation for another 2 weeks.

Immunocytochemistry

Immunocytochemistry was carried out using standard procedures. Nestin is a neural progenitor marker, β-tubulin III is a neuron-specific marker, and GFAP is an astrocyte-specific marker. Cells were fixed with 4% paraformaldehyde for 20 min at room temperature, permeabilized with 0.1% Triton X-100, and blocked with 10% goat serum. After washing with PBS, the cells were incubated at room temperature for 1 hr with primary antibody. After washing with PBS three times, the cells were incubated for 1 hr at room temperature with secondary antibodies. Cell nuclei were stained with Hoechst 33258. Images were captured using an Olympus fluorescence microscope. A P value of less than 0.05 denoted the presence of a statistically significant difference.

RT-PCR Analysis

Total RNA was extracted from cells with Trizol reagent (Invitrogen) according to the manufacturer's protocol. One microgram of RNA was reverse transcribed into cDNA and amplified using an RT-PCR system (Promega). The primer sequences and product sizes are listed in Table 1. PCR products were separated on a 1% agarose gel. The expression levels were analyzed using multianalysis software and standardized to the expression levels of the GAPDH housekeeping gene.

Table 1. Primers used for PCR analysis
GeneSense primer (5′ [RIGHTWARDS ARROW] 3′)Antisense primer (5′ [RIGHTWARDS ARROW] 3′)Size (bp)Annealing Tm (°C)
BrachyuryCTGGGAGCTCAGTTCTTTCGTCACATAGATGGGGGTGACA20155
eomesCAACATAAACGGACTCAACCCGACCTCCAGGGACAATCT50565
mash1CACAAGTCAGCGGCCAAGCAGGATCCCTCGTCGGAGGAGTAG25052
nestinGGACAGGACCAAAGGAACATCCCACCTCTGTTGACTTCC59959
pax6AGACTTTAACCAAGGGCGGTTAGCCAGGTTGCGAAGAACT58856
sox1CCTCGGATCTCTGGTCAAGTTACAGAGCCGGCAGTCATAC59353
TTRAGTCCTGGATGCTGTCCGAGTCCTGAGCTGCTAACACGG44059
gapdhTGAAGGTCGGTGTGAACGGATCAGGGGGGCTAAGCAGTTGGT47065

Western Blot Analysis

Western blot analysis was performed using a previously established method. Proteins were extracted from EBs or cells using lysis buffer (Pierce). The protein concentrations were determined using a BCA protein assay kit (Bioteke). Twenty micrograms of protein was loaded, resolved by sodium dodecylsulfate-10% polyacrylamide gel electrophoresis (SDS-PAGE), and electroblotted onto a polyvinylidene difluoride membrane. Blocking was performed in TBS-T (0.1% Tween, 5% skimmed milk in TBS) for 1 hr. The membranes were incubated overnight at 4°C with primary antibodies. To normalize the amounts of protein applied to SDS-PAGE, the membranes were reprobed with anti-GAPDH monoclonal antibody. Signals were detected with horseradish peroxidase–conjugated secondary antibodies (Pierce) using an ECL kit (Pierce). Each experiment was repeated three times.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

RA Induces Early Priming of ESC Differentiation Into Neurons

The differentiation of ES cells into neurons was performed as shown in Figure 1A. EBs treated with 5 μM RA for 4 days were nestin-positive, unlike controls (Fig. 1Bc,d,g). This finding suggested significant priming of neural germ layer development by RA, consistent with previous reports that high concentrations of RA induce neural differentiation (Niederreither et al.,2000; Canestro and Postlethwait,2007). These cells were induced into greater than 90% β-tubulin III-positive neural cells after being replated on LPO substrates, which emanated from individual rosettes and extended lots of neurites (Fig. 1C). Only approximately 8% of the cells were positive for GFAP, an astrocyte marker, and no MBP-positive oligodendrocytes were observed, suggesting less glial cell differentiation. Thus, we have established an efficient model of neural differentiation of ES cells, and provided a good platform for further research into the RA-regulated neural progenitor differentiation of ES cells in vitro. Although previous reports have shown that neural differentiation of ES cells can be promoted by RA, the effect of RA at a concentration of 5 μM on germ layer differentiation had never been precisely described. Next, to determine how 5 μM RA promotes early neural differentiation, we analyzed the expression of germ layer-specific genes in EBs between day 4 (EB4d) and day 8 (EB8d). Treatment of EBs with RA strongly increased the expression of the neuroectoderm-specific gene nestin, but decreased the expression of genes specific for other layers (Fig. 1D). These results clearly demonstrated that RA is a potent inducer of neural progenitor differentiation.

thumbnail image

Figure 1. Retinoic acid (RA) promotes embryonic stem (ES) cell differentiation into neurons. A: The protocol for differentiation of ES cells into neurons. B: Sequential images of cells during differentiation. a: ES cells growing on a gelatin-coated dish. b: Embryoid bodies (EBs) aggregated in medium containing RA without leukemia inhibitory factor (LIF). c: EBs seeded on an fibronectin (FN) -coated dish. d: Nestin-positive cells observed by immunofluorscence (green). e,f: Neural-like cells growing on poly-L-ornithine/laminin (LPO) substrates. g: A significant enhancement of nestin-positive cells was seen among RA-treated EBs compared with untreated EBs (*P < 0.05). h: Reverse transcriptase-polymerase chain reaction analysis of the germ layer gene expression in EBs, with or without RA, from day 5 to day 8. Ca–d: Double immunostaining for β-tubulin III (red) and glial fibrillary acidic protein (GFAP, green) in ES-derived cells grown on LPO in 80/20 medium, at day 10. The nuclei were counterstained with Hoechst33258 (blue). e: The proportions of cells positive for β-tubulin III among all cells were determined by immunocytochemistry. (*P < 0.05). DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum. Data shown are representative of results from three independent experiments. Scale bars = 50 μm.

Download figure to PowerPoint

RA Promotes Neural Progenitor Differentiation by Inducing CREB Phosphorylation

CREB is a transcription factor that integrates numerous signals to control cell differentiation. The phosphorylation of CREB plays an important role in RA-induced PC12 cell differentiation (Canon et al.,2004). We analyzed whether this modification also occurs during the early differentiation of ES cells. For this purpose, we first detected the amount of pCREB by Western blot using an antibody specific for CREB phosphorylated at Ser133. As shown in Figure 2A, RA increased the phosphorylation of CREB in a dose-dependent manner. RA (5 μm) induced activation of CREB after 24 hr; CREB remained activated until day 8. The peak concentration time for activated CREB was after 6 days of RA incubation. However, because RA is required for the formation of neuroectoderm, the previous experiments were not designed to determine CREB phosphorylation levels response to RA during neuroectoderm differentiation in EBs. Therefore, we examined the relationship between neuroectoderm differentiation and the phosphorylation levels of CREB by immunofluorescence, using antibodies against differentiation markers and pCREB. As shown in Figure 2B, pCREB is activated both in the absence and presence of RA in EBs, and is localized to the nucleus. The percentage of pCREB+/Nestin+ cells was increased in RA-induced EBs (Fig. 2Bg). This result suggested that RA induces neuroectoderm differentiation by stimulating CREB phosphorylation. To determine whether CREB activation is necessary for the effects of RA on neural progenitor differentiation, we introduced CREB-VP16 or mCREB mutants into RA-induced EBs by transfection. Western blotting showed that mCREB inhibits the activation of pCREB in RA-treated EBs, blocking nestin expression (Fig. 2C). However, the expression of nestin was significantly increased after RA-induced EBs were transfected with constitutively active CREB-VP16. Taken together, the above results suggest that RA induces neural progenitor differentiation by means of the activation of CREB in EBs.

thumbnail image

Figure 2. Cyclic AMP response element-binding protein (CREB) is required for retinoic acid (RA) -induced differentiation of nestin-positive neural progenitors. A: The effects of RA on the phosphorylation of CREB in embryoid bodies (EBs). a: Western blot analysis of dose (0–5 μM) effects of RA on phospho-CREB (pCREB) expression in EBs; extracts were subjected to immunoblotting with anti-pCREB. b: Western blot analysis of activation of CREB in EBs treated with RA for the indicated times (EB4d–EB8d), and the extracts were subjected to immunoblotting with anti-pCREB. B,a,b,d,e: Nuclear localization of pCREB (b,e) in nestin-immunoreactive cells (a,d) was confirmed by three-dimensional reconstruction of confocal microscopic images. c,f: Merged images. g: The proportions of cells positive for both nestin and pCREB among all cells in EBs were determined by immunocytochemistry, with the number of immunoreactive cells being expressed as a percentage of the total number of cells counted on the basis of nuclear staining with Hoechst33258 (*P < 0.05 vs. control). The percentages of cells expressing both nestin and pCREB were higher in EBs treated with RA than in untreated EBs. C: Protein extracts from EBs transfected with CREB-VP16 or mCREB for 48 hr were subjected to Western blot analyses using antibodies against pCREB and nestin. The data shown are representative of results from three independent experiments. Results were normalized to glydderaldehyde-3-phosphate dehydrogenase (GAPDH) expression. Scale bars = 20 μm.

Download figure to PowerPoint

RA Regulates CREB Activation by Means of the JNK Pathway

Given that CREB can be phosphorylated on serine133 by multiple kinases, such as protein PKA and MAPKs, which are also downstream of RA, we next asked which signaling pathway was responsible for RA-induced CREB activation. To determine the signaling pathways mediating CREB activation during RA-induced neural progenitor differentiation, we investigated the expression of pCREB and nestin in EBs treated with PKA by Western blot. As shown in Figure 3, in RA-induced EBs, the expression of pCREB remained unchanged with or without H89 (a PKA inhibitor) treatment; the same pattern was also detected for the expression of nestin, suggesting that the PKA pathway is not required for RA-mediated EB differentiation. Interestingly, H89 changed the expression of pCREB and nestin in the absence of RA. CREB was initially identified as a substrate of PKA that plays a key role in cAMP-regulated gene expression (Ryu et al.,2005; Aggarwal et al.,2006). However, our result showed that RA induced CREB activation in a PKA-independent manner.

thumbnail image

Figure 3. JNK-dependent activation of cyclic AMP response element-binding protein (CREB)by retinoic acid (RA) in embryoid bodies (EBs). A: A protein kinase A (PKA) inhibitor, H89 (10 μM), was applied to both RA− and RA+ EBs for 1 hr; after 24 hr, protein extracts from EBs were subjected to Western blot analyses using antibodies against pCREB. Protein extracts from EBs at EB8d were subjected to Western blot analyses using antibodies against nestin. B,a: Protein extracts from EBs with or without RA at EB5d or EB8d were subjected to Western blot analyses using antibodies against phospho-JNK (pJNK), pCREB, phospho-p38 (pp38), and phospho-ERK (pERK). b: RT-PCR analysis of the sequential expression of germ layer differentiation genes in EBs treated with RA, with or without sp600125 treatment. C: The effect of sp600125 on CREB activation in RA-induced EBs. A JNK inhibitor, sp600125, was used at different concentrations (0–10 μM) in RA-treated EBs for 30 min. After 24 hr, protein extracts from EBs were subjected to Western blot analyses using antibodies against pJNK, pCREB; after 96 hr, protein extracts from EBs were subjected to Western blot analyses using antibodies against nestin. D,a,b: Nuclear localization of pCREB(a,b) in EBs treated with RA, in the absence or presence of sp600125, as detected by confocal microscopy. c: pCREB-positive cells as a percentage of the total number of cells counted on the basis of nuclear staining with Hoechst33258 (*P < 0.05 vs. control). The percentages of cells expressing pCREB were higher in RA-induced EBs without sp600125 than in those stimulated by sp600125. RA+, RA-treated; RA−, without RA-treated. The data shown are representative of the results from three independent experiments. Results were normalized to glyceraldehydes-3-phosphate dehydrogenase expression. Scale bars = 20 μm.

Download figure to PowerPoint

It has been reported that RA induces ES cell differentiation by activating the MAPK signal transduction pathway. Further study has been concerned with the role of the MAPK family in the neural progenitor differentiation of ES cells induced by RA, with results showing that the phosphorylation/activation of JNK is increased in 5 μM RA-treated EBs, whereas phosphorylation of p38MAPK and ERK1/2 did not change after the same treatment (Fig. 3Ba). Gene expression analysis showed that, in RA-induced EBs, the inhibition of JNK decreased the expression levels of NPC markers, but increased the expression levels of markers of mesoderm and endoderm (Fig. 3Bb). These results suggested that JNK may act as a downstream effector of RA and that it is involved in RA-induced neural progenitor differentiation. Because JNK and CREB are both involved in RA-mediated differentiation and the expression pattern of pJNK is consistent with that of pCREB, we next examined the relationship between JNK and CREB activation in EBs. We pretreated EBs with sp600125, a JNK-specific inhibitor, for 30 min. Sp600125 blocked RA-induced activation of CREB in a dose-dependent manner (Fig. 3C). Moreover, as shown in Figure 3D, phosphorylated CREB was expressed at a lower level in EBs pretreated with sp600125, indicating that RA-induced CREB activation requires activation of the JNK signaling pathway. Therefore, these data suggest that RA-induced CREB activation is mediated by a JNK-dependent mechanism.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The transcription factor CREB has been extensively studied for its roles in cell differentiation and protection from apoptosis, as well as its role in learning and memory (Mayr et al.,2001). In the central nervous system, the functions of CREB were first described for the adrenergic system, in which generation of cAMP by β-adrenergic receptors activates protein kinase A, which in turn phosphorylates CREB leading to the induction of gene transcription. Nowadays, several signaling cascades, including a large number of receptors and kinases, are known to converge at CREB, thus underlining its importance for neuronal activity. CREB-deficient mice die immediately after birth and exhibit brain defects. It has been suggested that CREB may play a key role in early neural development. Previous data have shown that CREB phosphorylation is induced through different mechanisms in different types of cells during RA-mediated differentiation (Kim et al.,2007). Therefore, we investigated the roles of CREB in mediating differentiation induced by RA in ES cells.

First, we established a novel and feasible in vitro neural differentiation model of ES cells. The results from our study suggest that the differentiation of ES cells into neurons could be promoted by RA, especially by early exposure to the relatively high concentration of 5 μM. Our data are consistent with previous evidence that RA strongly triggers neural differentiation (Rohwedel et al.,1999). Furthermore, our data demonstrate that RA induces the activation of CREB in EBs during their differentiation, which is consistent with recent studies showing that RA induces CREB phosphorylation and CREB-mediated transcription in PC12 cell clones that also undergo morphological differentiation upon incubation with RA (Radhakrishnan et al.,1998; Ronda et al.,2007). The activation of CREB by RA may result in increased DNA binding of active CREB to its cognate CRE binding sequence. CRE-mediated transcription is a convergence point for multiple pathways initiated by differentiation signals. The transcriptional activity of CREB is regulated by phosphorylation at Ser133 in the P-box or KID of the protein; this domain includes several consensus phosphorylation sites for a variety of kinases, such as PKA, CaMKIV, and MAPKs (Lim et al.,2005). Our result demonstrated that CREB was mediated by RA independent of the conventional PKA pathway. Moreover, JNK functions as the intermediate between RA and the activation of CREB kinases, which up-regulates brain-derived neurotrophic factor and fibroblast growth factor expression (Hossain et al.,2002; Soto et al.,2006) and results in the differentiation of neural progenitor populations. JNKs are key components of networks that mediate cell fate decisions and differentiation. Mice lacking both jnk1 and jnk2 undergo mid-gestational embryonic lethality associated with defects in neural tube closure and deregulated neural apoptosis (Sabapathy et al.,1999). Thus, that CREB is one of the main downstream targets of JNK supports a broad role for JNK in neural development and morphogenesis. To our knowledge, this is the first report to show that RA induces JNK-dependent CREB activation during the differentiation of ES cells. Further study is necessary to examine the mechanism by which JNK regulates the activation of CREB.

In conclusion, the present findings suggest that RA regulates the neural differentiation of ES cells. RA induces the phosphorylation of CREB in a JNK-dependent manner, and this phorphorylation advances the differentiation of neural ectoderm. However, the direct target genes or upstream regulators of CREB during the neural differentiation of ES cells have not yet been identified. Therefore, it will be of interest to further study the function and regulation of CREB during neurogenesis in ES cells.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The authors thank Dr. Yu Li (Harbin Institute of Technology, China) and Dr. Yang Yu (Baylor College of Medicine, America) for their technical assistance, and Dr. Yan Wang for his help with statistical analysis.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED