SEARCH

SEARCH BY CITATION

Keywords:

  • hnRNP A2/B1;
  • Human embryonic stem cells;
  • Self-renewal;
  • Pluripotency;
  • G1/S transition;
  • PI3K/Akt

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Self-renewal and pluripotency of human embryonic stem cells (hESCs) are a complex biological process for maintaining hESC stemness. However, the molecular mechanisms underlying these special properties of hESCs are not fully understood. Heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP A2/B1) is a multifunctional RNA-binding protein whose expression is related to cell proliferation and carcinogenesis. In this study, we found that hnRNP A2/B1 expression was localized to undifferentiated hESCs and decreased upon differentiation of hESCs. hnRNP A2/B1 knockdown reduced the number of alkaline phosphatase-positive colonies in hESCs and led to a decrease in the expression of pluripotency-associated transcription factors OCT4, NANOG, and SOX2, indicating that hnRNP A2/B1 is essential for hESC self-renewal and pluripotency. hnRNP A2/B1 knockdown increased the expression of gene markers associated with the early development of three germ layers, and promoted the process of epithelial-mesenchymal transition, suggesting that hnRNP A2/B1 is required for maintaining the undifferentiated and epithelial phenotypes of hESCs. hnRNP A2/B1 knockdown inhibited hESC proliferation and induced cell cycle arrest in the G0/G1 phase before differentiation via degradation of cyclin D1, cyclin E, and Cdc25A. hnRNP A2/B1 knockdown increased p27 expression and induced phosphorylation of p53 and Chk1, suggesting that hnRNP A2/B1 also regulates the G1/S transition of hESC cell cycle through the control of p27 expression and p53 and Chk1 activity. Analysis of signaling molecules further revealed that hnRNP A2/B1 regulated hESC proliferation in a PI3K/Akt-dependent manner. These findings provide for the first time mechanistic insights into how hnRNP A2/B1 regulates hESC self-renewal and pluripotency. STEM Cells 2013;31:2647–2658


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Human embryonic stem cells (hESCs) are undifferentiated and pluripotent stem cells with unlimited self-renewal capacity which can be differentiated into a variety of differentiated cells [[1, 2]]. Self-renewal and pluripotency of hESCs are controlled by the master transcription factors OCT4, NANOG, and SOX2 [[1, 3]]. The master transcription factors play their roles in regulating their own genes and large networks of downstream target genes, which are important for maintaining hESC stemness [[1, 3, 4]]. They usually activate a variety of target genes encoding critical components of self-renewal and pluripotency, while repressing the signaling pathways that promote differentiation processes [[1, 3]]. However, the molecular mechanisms and signaling pathways that control the self-renewal and pluripotency of hESCs are still largely unknown.

Heterogeneous nuclear ribonucleoproteins (hnRNP) are a family of proteins that play roles in cytoplasmic trafficking of mRNAs, mRNA stability and turnover, splicing regulation, and telomerase biogenesis [5]. hnRNP A2/B1 is a major component of the hnRNP core complex in mammalian cell nuclei. hnRNP A2 and B1 full length cDNAs were first cloned from HeLa cell cDNA expression library, and hnRNP B1, which has 12 amino acid insertion near its N-terminal end relative to hnRNP A2, is derived from a splicing variant of hnRNP A2 mRNA [[6, 7]]. hnRNP A2/B1 is considered to play major roles in RNA metabolism through direct interaction with mRNAs [[8-10]]. However, many other studies showed that hnRNP A2/B1 has multiple functions in many cellular processes. For example, hnRNP A2 binds telomeric DNA repeats and the RNA component of telomerase simultaneously, which suggests its role in telomere maintenance [11]. hnRNP A2/B1 is upregulated or mislocalized in human cancers such as lung, brain, colon, breast, pancreatic, and stomach carcinoma [[12-19]]. Furthermore, hnRNP A2/B1 colocalizes with c-myc, c-fos, p53, and Rb in undifferentiated neuroblastoma and gastric adenocarcinoma cells [[17, 18]]. As hnRNP A2/B1 is implicated in the splicing of many genes [8], the regulation of alternative splicing by upregulation of hnRNP A2/B1 may contribute to tumor cell proliferation and progression. Conversely, accumulating evidence suggested that hnRNPA2/B1 can directly bind to the promoters of c-myc, apolipoprotein E, vitamin D receptor, breast cancer 1, and gonadotropin-releasing hormone 1 genes [[20-24]]. Recent studies also showed that hnRNP A2 functions as a transcription coactivator by recruiting DNA binding factors to the promoter sites of mitochondrial respiratory stress-responsive genes [[25, 26]]. Thus, hnRNP A2/B1 is considered to be a multifunctional RNA- and DNA-binding protein.

In an attempt to find novel regulatory molecules involved in maintaining hESC self-renewal and pluripotency, hnRNP A2/B1 attracted our interest because hnRNP A2/B1 was highly expressed in undifferentiated hESCs and rapidly downregulated upon differentiation. hnRNP A2/B1 knockdown analysis further showed that hnRNP A2/B1 was required for maintaining hESC self-renewal and pluripotency. Subsequent studies revealed how hnRNP A2/B1 regulated hESC self-renewal and pluripotency.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Cell Culture

hESC line H9 was cultured on mouse embryonic fibroblast (MEF) feeder cells as described previously [[27, 28]]. For feeder-free culture, H9 cells were cultured in conditioned medium supplemented with the Rho-associated coiled kinase inhibitor Y27632 (10 μM, Sigma-Aldrich, Seoul, Korea, www.sigmaaldrich.com/korea.html), on tissue culture dish coated with Matrigel (BD Bioscience, Seoul, Korea, www.bdbiosciences.com/kr) as described previously [29]. Differentiation of H9 cells was induced by incorporating all-trans-retinoic acid (RA, Sigma-Aldrich) at 10−5 M into the medium and culturing the cells for at least 10 days.

Flow Cytometry

Flow cytometric analysis was performed using hESC marker-specific antibodies and anti-hnRNP A2/B1 antibodies as described previously [[27, 28, 30]]. Briefly, hESCs were detached with collagenase IV and further incubated with cell dissociation buffer (Invitrogen, Seoul, Korea, www.lifetechnologies.com) as described before [27]. The dissociated cells were then incubated with primary antibodies in PBA (1% bovine serum albumin, 0.02% NaN3 in phosphate-buffered saline [PBS]) for 30 minutes at 4°C. Primary antibodies used were anti-SSEA3, anti-SSEA4 (R&D Systems, Minneapolis, MN, www.rndsystems.com), anti-TRA-1-60, and anti-TRA-1-81 (Millipore, Billerica, MA, www.millipore.com). Then, the cells were further incubated with fluorescein isothiocyanate (FITC)-conjugated anti-mouse immunoglobulin (Ig) G (BD Biosciences, Seoul, Korea). Propidium iodide (PI)-negative cells were analyzed for the antibody binding using FACSCalibur (BD Biosciences). To detect apoptosis, hESCs were stained and analyzed with PI and FITC-conjugated Annexin V using an Annexin V-FITC/PI Apoptosis kit (BD Biosciences) according to the manufacturer's protocol.

For intracellular staining, hESCs were fixed in 2% paraformaldehyde (PFA) in PBS (pH 7.4) and permeabilized in 0.5% saponin (Sigma-Aldrich) in PBA (1% bovine serum albumin, 0.02% NaN3 in PBS) for 10 minutes at room temperature (RT). After being washed with PBA, the cells were then incubated with mouse monoclonal anti-hnRNP A2/B1 antibody (Abcam, Cambridge, U.K., www.abcam.com) and rabbit anti-NANOG (Santa Cruz Biotechnology, Santa Cruz, CA, www.scbt.kr), rabbit anti-OCT4 (Millipore), or rabbit anti-SOX2 (Santa Cruz Biotechnology) diluted in PBA containing 0.5% saponin for 30 minutes at RT, and subsequently incubated with FITC-conjugated rabbit IgG (Vector Laboratories, Burlingame, CA, www.vectorlab.com) and PE-Cy5.5-conjugated anti-mouse IgG1 (Invitrogen) in PBA for 30 minutes at RT, and analyzed on a Becton-Dickinson FACSCalibur.

siRNA Transfection of hESCs

Small interfering (siRNA) oligonucleotides targeting hnRNP A2/B1 (Santa Cruz Biotechnology) were used with AllStars negative control siRNA (Qiagen, Valencia, CA, www.qiagen.com). Feeder-free H9 cells (8 × 104 cells per well in a six-well plate) preincubated with 10 μM Y-27632 for 1 hour were dissociated with 0.05% trypsin-EDTA (WelGene, Daegu, Korea, welgene.com) and seeded into feeder-free plate. The next day, the cells were transfected with 100 nM AllStar control or hnRNP A2/B1 siRNA by RNAimax (Invitrogen). The cells were transfected the second time after 48 hours following the first transfection and then incubated for additional 24 hours before harvesting for analysis.

Colony-Forming Assay

Control or hnRNP A2/B1 siRNA-transfected H9 cells were preincubated with Y-27632 for 1 hour before collagenase IV treatment. To produce single-cell suspensions, H9 cells were incubated with 0.05% trypsin-EDTA at 37°C for 5 minutes and dissociated into single cells by filtering through a 40 μm strainer, before seeding at 0.5–1.5 × 104 cells on Matrigel-coated six-well plates in the presence of Y-27632. After 6–10 days, colonies were fixed with 3.7% PFA and stained with an alkaline phosphatase staining kit (Sigma-Aldrich) according to the supplier's protocol.

RT-PCR and Real-Time PCR

H9 cells were transfected with either control or hnRNP A2/B1 siRNAs as described above or treated with RA for at least 10 days. After transfection or RA treatment, total RNAs were extracted using the RNA iso plus reagent (TaKaRa, Otsu, Japan, www.takatra.com). cDNAs were generated from total RNAs by Prime Script RT Master Mix (TaKaRa). Quantitative real-time PCR was performed using SYBR green technology and Bio-Rad IQ5 detector (Bio-Rad, Hercules, CA, www.bio-rad.com). Levels of transcripts for specific genes were determined using specific primers for human transcripts encoding hnRNP A2B1 [31], FoxA2, AFP [32], GATA6, Brachyury T [33], KDR/FLK1, NES, HES1, NANOG, SOX2, OCT4, GAPDH [32], HAND1 [34], BAX, BAD, BCL2, NOXA [35], p21, and p27 [36]. Primers for BID cDNA were: sense, 5′-CAAAGTGGTTCCCTCTCTTAGG-3′ and antisense, 5′-CTGATTCCTGGGACATAGCTTAC-3′. Primers for PUMA cDNA were: sense, 5′-GTGACCACTGGCATTCATTTG-3′ and antisense, 5′-TCCTCCCTCTTCCGAGATTT-3′. PCR reactions were performed using a total of 30 cycles consisting of denaturation for 1-minute at 95°C, followed by annealing for 45 seconds at 45°C–56°C and extension for 1-minute at 72°C. Each sample was analyzed in triplicate for each target gene, and the levels of mRNA were quantified by the standard curve method.

Cell Proliferation Assay

Feeder-free H9 cells preincubated with Y-27632 for 1 hour were dissociated with 0.05% trypsin-EDTA and seeded into six-well plates (5 × 104 cells per well). The next day, cells were transfected with control or hnRNP A2/B1 siRNA as described above. Cell numbers were determined after 72 hours by trypan blue exclusion assay.

Cell Cycle Analysis

Feeder-free H9 cells were transfected with control or hnRNP A2/B1 siRNA as described above. Actively proliferating H9 cells were treated with 30 μM bromodeoxyuridine (BrdU) for 4 hours and detached by 0.05% trypsin-EDTA treatment. Cells (1–2 × 106) were fixed in 70% ethanol overnight at 4°C. After being washed twice with PBA, cells were suspended in denaturation buffer (2 M HCl/0.5% Triton X-100) for 30 minutes at RT. Cells were then suspended in 0.1 M sodium borate for 2 minutes after being washed twice with PBA. Then, cells were incubated with FITC-conjugated anti-BrdU antibody for 30 minutes at RT and incubated in RNase A (10 μg/ml) and PI (20 μg/ml) solution for 30 minutes at RT before analysis on a FACSCalibur.

Western Blot Analysis

Western blotting was performed as described previously [[27, 37]]. The following antibodies were used in Western blot analysis: mouse monoclonal antibodies against cyclin-dependent kinase 2 (Cdk2), p21, Chk2, and β-actin (all from Santa Cruz Biotechnology); rabbit polyclonal antibodies against OCT4, SOX2, NANOG, phospho-Cdk2/Cdc2 (T14/T15), Cyclin B1, Cyclin E, Cdc25A, Chk1, p27, p16, Akt1/2/3, Caspase-3 (all from Santa Cruz Biotechnology), phospho-Cdk2 (T160), phospho-Akt (S473), cyclin D1, phospho-Chk1 (S345), phospho-Chk2 (T68), p53, phospho-p53 (S20), phospho-p53 (S15) (all from Cell Signaling Technology, Beverly, MA, www.cellsignal.com), and PARP (poly(ADP-ribose) polymerase) (Roche, Mannheim, Germany, www.roche.com); and goat polyclonal antibody against actin (Santa Cruz Biotechnology). Cells were lysed in 0.5% Nonidet P40, 120 mM NaCl, 50 mM Tris-HCl (pH7.4), 1 mM EDTA (pH 8.0), and complete protease inhibitor mixture (Roche). Cell lysate (50–100 μg/lane) was electrophoresed through SDS-PAGE and proteins were transferred to PVDF (polyvinylidene difluoride) membrane (Millipore) and blocked in Tris-buffered saline Tween 20 with 5% skim milk or 5% bovine serum albumin. After incubation with primary antibodies, blots were incubated with horseradish peroxidase-conjugated secondary antibodies (GE Healthcare, Seoul, Korea, www.gehealthcare.co.kr). Signals were visualized using ECL system (GE Healthcare). Statistical analysis of protein expression was done using the public domain image program Image J. For normalization of protein levels in Western blot analysis, β-actin or actin were used as loading controls.

Immunocytochemistry

Cells were stained as described previously [27]. For double immunofluorescence staining for anti-hnRNP A2/B1 and anti-OCT4, anti-NANOG, or anti-SOX2 antibodies, H9 cells were fixed with 3.7% PFA, permeabilized with 0.1% Triton X-100, and then incubated with anti-hnRNP A2/B1, and anti-OCT4, anti-NANOG, or anti-SOX2 antibodies. The cells were further incubated with Alexa 660-conjugated anti-mouse IgG and FITC-conjugated anti-rabbit IgG (Vector Laboratories). Between each step, cells were washed with Ca2+- and Mg2+-PBS. Nuclei were stained with 4,6-diamidino-2-phenylindole. Fluorescence signals were detected with a Leica TCS SP5 confocal microscope.

Chemical Inhibitor Treatment

Feeder-free H9 cells were treated with dimethyl sulfoxide or 30 μM LY294002 (Santa Cruz Biotechnology) in MEF-conditioned medium for 5 days. One fraction of inhibitor-treated cells was then lysed with radio immunoprecipitation assay buffer and the levels of hnRNP A2/B1, Akt1/2/3, and phosphorylated Akt1/2/3 were examined by Western blot analysis. β-Actin was used as a loading control.

RESULTS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

hnRNP A2/B1 Is Highly Expressed in Undifferentiated hESCs and Downregulated During Differentiation

To study whether hnRNP A2/B1 is involved in the undifferentiated and pluripotent state of hESCs, we performed intracellular flow cytometric analysis with a mouse monoclonal antibody to hnRNP A2/B1 and rabbit polyclonal antibodies to pluripotency markers (Fig. 1A). Approximately 70%, 80%, and 86% of OCT4, SOX2, and NANOG-positive hESCs, respectively, were positive for the expression of hnRNP A2/B1, showing that hnRNP A2/B1 is well localized to OCT4-, SOX2-, and NANOG-positive hESCs. Immunocytochemical analysis also displayed that hnRNP A2/B1 was well localized to OCT4-, SOX2-, and NANOG-positive hESCs (Fig. 1B). To examine the protein expression pattern of hnRNP A2/B1 upon differentiation, hESCs were then subjected to the differentiation induced by RA treatment. When hESCs were treated with RA for 10 days, the expression level of OCT4 was drastically decreased and the expression level of hnRNP A2/B1 was also gradually decreased during the differentiation of hESCs (Fig. 1C, the first and second panels). Thus, it is concluded that the expression of hnRNP A2/B1 is highly expressed in undifferentiated hESCs and downregulated during the differentiation of hESCs.

image

Figure 1. Expression of hnRNP A2/B1 is localized to undifferentiated and pluripotent human embryonic stem cells (hESCs). (A): Multicolor intracellular flow cytometry of H9 cells with mouse monoclonal anti-hnRNP A2/B1 antibody and anti-OCT4, SOX2, or NANOG antibodies. H9 cells were fixed and incubated with anti-hnRNP A2/B1 and anti-OCT4, NANOG, or SOX2 antibodies in PBA containing 0.5% saponin. The cells were subsequently incubated with PE-Cy5.5-conjugated anti-mouse IgG1 and fluorescein isothiocyanate (FITC)-conjugated rabbit IgG in PBA containing 0.5% saponin, and analyzed by flow cytometry. Values in each quadrant indicate percentage of positive cells. (B): Immunocytochemical analysis of H9 cells with anti-hnRNP A2/B1 and anti-OCT4, NANOG, or SOX2 antibodies. H9 cells were fixed, permeabilized, and incubated with mouse monoclonal anti-hnRNP A2/B1 and rabbit anti-OCT4, anti-NANOG, or anti-SOX2 antibodies. The cells were incubated with Alexa 660-conjugated anti-mouse IgG and FITC-conjugated anti-rabbit IgG. Scale bar is 40 μm. (C): Western blot analysis of hnRNP A2/B1 during differentiation of hESCs. H9 cells were cultured in the presence of RA for 10 days, and cell lysates were separated by SDS-PAGE and subjected to Western blotting with anti-hnRNP A2/B1 or anti-OCT4 antibodies followed by HRP-conjugated anti-mouse IgG or anti-rabbit IgG, respectively. The signal intensities of Western blots were measured quantitatively using the Image J software, and actin was used as a normalization control. Data shown are representative of at least three independent experiments. Abbreviations: DAPI, 4,6-diamidino-2-phenylindole; hnRNP A2/B1, heterogeneous nuclear ribonucleoprotein A2/B1; RA, retinoic acid; PBA, 1% bovine serum albumin, 0.02% NaN3 in PBS.

Download figure to PowerPoint

hnRNP A2/B1 Knockdown Impairs hESC Self-Renewal and Pluripotency

To study whether hnRNP A2/B1 is essential for the maintenance of undifferentiated hESCs, we investigated the role of hnRNP A2/B1 in undifferentiated hESCs using an RNA interference approach. hESCs were transfected twice with either nonsilencing control siRNA or siRNA against hnRNP A2/B1 gene. hnRNP A2/B1 protein expression was decreased by 74%–84%, depending on experimental conditions (Fig. 2A, first and third panels). The decreased hnRNP A2/B1 expression was also confirmed by RT-PCR and real-time PCR (Fig. 3). In morphology, hnRNP A2/B1 siRNA-treated hESCs exhibited a striking alteration with large, flat, and fibroblast-like cells and tended to be individually dispersed (Fig. 2B).

image

Figure 2. Knockdown of hnRNP A2/B1 impairs human embryonic stem cell (hESC) self-renewal and pluripotency. (A): Knockdown of hnRNP A2/B1 by a siRNA technology. Feeder-free H9 cells were twice transfected with control (Con) or hnRNP A2/B1 siRNAs (A2/B1 Si) for 72 hours. Cell lysates were subjected to Western blot analysis with mouse monoclonal anti-hnRNP A2/B1 antibody. The signal intensities of Western blot were measured quantitatively using the Image J software, and β-actin was used as a normalization control. Shown are two independent experiments. (B): Morphology of hESCs treated with control or hnRNP A2/B1 siRNAs at day 3 after double transfections. Scale bar is 100 μm. (C): Colony-forming assays of control or hnRNP A2/B1 siRNA-transfected hESCs. H9 cells were pretreated with Y-27632 for 1 hour and dissociated into single-cell suspensions and transfected with control or hnRNP A2/B1 siRNAs for 72 hours. After culturing for 8 days, visible colonies were stained with alkaline phosphatase according to the supplier's protocol. (D): Statistical analysis of alkaline phosphatase-positive colonies from control and hnRNP A2/B1 siRNA-transfected hESCs. The graph presents the mean values of at least three independent determinations ± SD. Statistical analysis used Student's t test, and asterisk indicates difference from control siRNA (*, p = .022). (E): Western blot analysis showing the expression levels of OCT4, NANOG, and SOX2 in control and hnRNP A2/B1 siRNA-transfected hESCs. Cell lysates were separated by SDS-PAGE, and Western-blotted with anti-OCT4, anti-NANOG, or anti-SOX2 antibodies followed by HRP-conjugated anti-rabbit IgG. The signal intensities of Western blots were measured quantitatively using the Image J software, and β-actin was used as a normalization control. Data shown are representative of at least three independent experiments. (F): Overlaid histograms showing expression levels of SSEA3, SSEA4, TRA-1-60, and TRA-1-81 in control and hnRNP A2/B1 siRNA-transfected hESCs. Control and hnRNP A2/B1 siRNA-transfected H9 cells were detached and analyzed by flow cytometry. Shown are representative overlaid histograms. Abbreviation: hnRNP A2/B1, heterogeneous nuclear ribonucleoprotein A2/B1.

Download figure to PowerPoint

image

Figure 3. hnRNP A2/B1 knockdown decreases pluripotency genes and increases differentiation genes of three germ layers in human embryonic stem cells (hESCs). RNA transcripts were isolated from hESCs transfected with control (Con) and hnRNP A2/B1 siRNAs (A2/B1 Si) 72 hours after transfection and subjected to RT-PCR and real-time PCR. (A): RT-PCR analysis of mRNA levels of pluripotency genes (OCT4, NANOG, and SOX2) and hnRNP A2/B1. GAPDH was used as the internal control. The data presented here are representative of three independent experiments. (B): Real-time PCR analysis of mRNA levels of pluripotency genes and hnRNP A2/B1 as described in (A). The graph presents the mean values of three independent determinations ± SD. Statistical analysis used Student's t test, and asterisks indicate the difference from control siRNA (*, p < .01; **, p < .001). (C): RT-PCR analysis of mRNA levels of early differentiation genes, FoxA2, AFP, GATA6, Brachyury T, KDR/FLK1, NES, HES1, and HAND1. (D): Real-time PCR analysis of mRNA levels of early differentiation genes as described in (C). The graph presents the mean values of at least three independent determinations ± SD. Statistical analysis used Student's t test, and asterisks indicate the difference from control siRNA (*, p < .05; **, p < .01; ***, p < .001). Abbreviation: hnRNP A2/B1, heterogeneous nuclear ribonucleoprotein A2/B1.

Download figure to PowerPoint

As the hnRNP A2/B1 knockdown hESCs showed the typical phenotypes of differentiating cells, we measured the number of alkaline phosphatase-positive colonies by the colony-forming assay, which is often used to evaluate the self-renewal ability of hESCs [38]. The colony formation efficiency of hnRNP A2/B1 knockdown hESCs was decreased by approximately 64% (p < .05), compared to control hESCs (Fig. 2C, 2D). The results suggest that hnRNP A2/B1 knockdown impairs hESC self-renewal. We next investigated the protein expression levels of key pluripotency genes such as OCT4, SOX2, and NANOG that are well-known factors associated with hESC self-renewal and pluripotency [[1, 3, 4]]. Western blot analysis showed that hnRNP A2/B1 knockdown led to a decrease in the expression levels of OCT4, SOX2, and NANOG (Fig. 2E). RT-PCR and real-time PCR analysis also showed the decreased levels of OCT4, SOX2, and NANOG genes in the hnRNP A2/B1 knockdown hESCs (Fig. 3A, 3B), indicating that hnRNP A2/B1 is necessary for hESC pluripotency. We then compared the expression levels of surface undifferentiated markers between control and hnRNP A2/B1 knockdown hESCs. The expression levels of SSEA3, TRA-1-60, and TRA-1-81 were obviously decreased in the hnRNPA2/B1 knockdown hESCs, although the expression level of SSEA4 was slightly decreased (Fig. 2F and Supporting Information Fig. S1). Taken together, the results suggest that hnRNP A2/B1 is necessary for maintaining hESC self-renewal and pluripotency.

hnRNP A2/B1 Knockdown Upregulates Differentiation Genes in hESCs

The ability of hESCs to self-renew is mediated through promotion of cell proliferation and prevention of apoptosis and differentiation [41]. In order to examine whether hnRNP A2/B1 depletion induces apoptosis in hESCs, hnRNP A2/B1 knockdown hESCs were stained with Annexin V and PI, and analyzed by flow cytometry (Supporting Information Fig. S2A). Annexin V-positive apoptotic cells were not significantly altered in hnRNP A2/B1 known hESCs, as compared to control knockdown hESCs. Western blot analysis also showed that the active forms of PARP and caspase-3 were not increased in hnRNP A2/B1 knockdown hESCs (Supporting Information Fig. S2B). To further confirm whether hnRNP A2/B1 depletion induces apoptosis, the expression levels of the known apoptosis regulators such as BAX, BAD, BID, BCL2, PUMA, and NOXA genes were also analyzed by real-time PCR (Supporting Information Fig. S2C). No significant increases were observed in the expression levels of apoptosis regulators except for the case of antiapoptotic gene BCL2, indicating that hnRNP A2/B1 knockdown does not lead to apoptosis in hESCs.

A previous study showed that hnRNPA2/B1 expression promotes smooth muscle cell (SMC) differentiation during guided differentiation of murine embryonic stem cells (mESCs) into SMCs [42]. Therefore, we compared the mRNA expression levels of SMC-specific differentiation markers (SMαA and SMMHC) and transcription factors (SRF and Myocardin) between hnRNP A2/B1 knockdown hESCs and control hESCs by RT-PCR and real-time PCR analyses (Supporting Information Fig. S3). hnRNP A2/B1 knockdown downregulated the SMC-specific differentiation markers and transcription factors in hESCs, as expected from the previous study [42]. Therefore, the result suggests that hnRNP A2/B1 expression may be required for SMC differentiation during guided differentiation of hESCs into SMC as well. However, hnRNP A2/B1 knockdown exhibited a large, flattened, and dispersed morphology in undifferentiated hESCs (Fig. 2B), which suggests that hnRNP A2/B1 expression is necessary for maintaining undifferentiated and epithelial phenotypes of hESCs. To examine the differentiation state of hnRNP A2/B1 knockdown hESCs, we compared the expression levels of early differentiation markers between hnRNP A2/B1 knockdown hESCs and control hESCs by RT-PCR and real-time PCR analyses (Fig. 3C, 3D). hnRNP A2/B1 knockdown led to an obvious increase in the mRNA levels of three germ layer markers, including FoxA2, GATA6, Brachyury T, KDR/FLK1, and NES, indicating that hnRNP A2/B1 knockdown hESCs tend to differentiate into multicell lineages. The results suggest that hnRNP A2/B1 expression is necessary for maintaining undifferentiated phenotype of hESCs.

hnRNP A2/B1 Expression Is Required for Maintaining the Epithelial Phenotype of hESCs

Previous studies suggested that epithelial-mesenchymal transition (EMT) events take place during hESC differentiation [[43, 44]]. A defining characteristic of EMT is the loss of E-cadherin-mediated cell-cell contacts and the acquisition of a mesenchymal phenotype [45]. hnRNP A2/B1 knockdown correlates with an increase of E-cadherin expression and downregulation of the E-cadherin inhibitors in nonepithelial lung cancer cell lines, suggesting that hnRNP A2/B1 expression is required for maintaining the mesenchymal phenotypes in nonepithelial lung cancer cell lines [46]. However, hnRNP A2/B1 knockdown hESCs displayed a flattened, scattered, and fibroblast-like morphology (Fig. 2B), which suggests that hnRNP A2/B1 knockdown, instead of hnRNP A2/B1 expression, may promote the process of EMT in hESCs. Flow cytometric analysis showed that hnRNP A2/B1 knockdown resulted in an obvious decrease in the expression level of E-cadherin in hESCs, although it did not alter the expression of N-cadherin (Fig. 4A). To determine the exact role of hnRNP A2/B1 during the process of EMT in hESCs, we examined protein expression of several genes known to play a role in EMT (Fig. 4B). hnRNP A2/B1 knockdown resulted in downregulation of E-cadherin and upregulation of vimentin, which represents typical EMT characteristics. Furthermore, hnRNP A2/B1 knockdown increased the expression of two EMT mediators Snail and Slug. The results suggest that hnRNP A2/B1 expression is required for maintaining the epithelial phenotype of hESCs.

image

Figure 4. hnRNP A2/B1 is required for maintaining the epithelial phenotype of human embryonic stem cells (hESCs). Feeder-free hESCs were twice transfected with control (Con) or hnRNP A2/B1 siRNAs (A2/B1 Si) for 72 hours. (A): Single-cell suspensions of transfected hESCs were subjected to flow cytometry. Shown is cell surface expression of E-cadherin and N-cadherin. Dashed lines represent control siRNA-transfected hESCs, while solid lines represent hnRNP A2/B1 siRNA-transfected hESCs. (B): Cell lysates from transfected hESCs were subjected to Western blot analysis with antibodies against hnRNP A2/B1, E-cadherin, vimentin, Snail, and Slug. Shown are representative Western blots from two independent experiments. The signal intensities of Western blot were measured quantitatively using the Image J software, and β-actin was used as a normalization control. Abbreviation: hnRNP A2/B1, heterogeneous nuclear ribonucleoprotein A2/B1.

Download figure to PowerPoint

hnRNP A2/B1 Knockdown Induces the G0/G1 Phase Arrest Before Cellular Differentiation

Previous studies showed that hnRNP A2/B1 is associated with cellular processes that affect the cell cycle and proliferation in human carcinoma cell lines [[47, 48]]. Therefore, we investigated whether hnRNP A2/B1 knockdown could affect hESC proliferation as well. hnRNP A2/B1 knockdown resulted in a drastic reduction in hESC proliferation (Supporting Information Fig. S4), suggesting that hnRNP A2/B1 expression is necessary for hESC proliferation. To address how hnRNP A2/B1 affects hESC proliferation, we then analyzed cell cycle distribution of hnRNP A2/B1 knockdown hESCs by BrdU and PI incorporations at 48, 60, and 72 hours after knockdown (Fig. 5A, 5B). hnRNP A2/B1 knockdown induced a substantial accumulation of hESCs in the G0/G1 phase at 60 and 72 hours, as compared to control knockdown. BrdU-positive cells represent the S-phase cells actively synthesizing DNA. As expected, hnRNP A2/B1 knockdown reduced the percentage of hESC population in the S-phase (Fig. 5A, 5B). Thus, hnRNP knockdown inhibits hESC proliferation by blocking cell cycle progression from G1 to S-phase.

image

Figure 5. Knockdown of hnRNP A2/B1 induces the G0/G1 phase arrest before cellular differentiation. (A): Cell cycle distribution analyzed by bromodeoxyuridine (BrdU) incorporation of control or hnRNP A2/B1 siRNA-transfected hESCs. Feeder-free hESCs were transfected with control or hnRNP A2/B1 siRNAs, and stained with BrdU and PI at 48, 60, and 72 hours after hnRNP A2/B1 knockdown. Shown are representative cell cycle distributions with the percentages of cells in the G0/G1, S, and G2/M phases. (B): Graphic presentation of cell cycle distribution analyzed by BrdU incorporation of control or hnRNP A2/B1 siRNA-transfected hESCs. Each bar represents mean value of three independent experiments ± SD. (C): Real-time PCR analysis of mRNA levels of hnRNP A2/B1, p21, p27, GATA6, FoxA2, KDR/FLK, and NES genes during 5 days in hnRNP A2B1 knockdown hESCs. RNA transcripts were isolated from hESCs transfected with control and hnRNP A2/B1 siRNAs at 0, 24, 48, 60, 72, and 120 hours after knockdown and subjected to real-time PCR. GAPDH was used as the internal control. The graph presents the mean values of at least three independent determinations ± SD. Abbreviation: hnRNP A2/B1, heterogeneous nuclear ribonucleoprotein A2/B1; hESCs, human embryonic stem cells.

Download figure to PowerPoint

hnRNP A2/B1 knockdown resulted in increased expression of various differentiation genes in hESCs (Fig. 3) and also caused the G0/G1 phase arrest (Fig. 5A, 5B). To clarify whether the primary effect of hnRNP A2/B1 is to change cell cycle progression, or whether hnRNP A2/B1 induces cell differentiation and this in turn is accompanied by inhibition of the cell cycles, we examined the expression changes of differentiation genes in hESCs at 0, 24, 48, 60, 72, and 120 hours after knockdown by real-time PCR (Fig. 5C). The expression of hnRNP A2/B1 reached minimum at 48 hours and gradually increased. The increased expression of p21 was obvious even at 24 hours while the increased expression p27 was observed after 72 hours. The expression of hnRNP A2/B1 decreased by less than 50% at 72 hours, and the increased expression of GATA6, FoxA2, KDR/FLK, and NES was observed as expected from Figure 3C, 3D. Thus, the G0/G1 phase arrest and cellular differentiation took place simultaneously at 72 hours after hnRNP A2/B1 knockdown (Fig. 5A--5C). The expression of hnRNP A2/B1 also decreased by less than 50% at 48 and 60 hours after hnRNP A2/B1 knockdown, but the increased expression of differentiation genes was not observed (Fig. 5A--5C). However, the G0/G1 phase arrest was observed at 60 hours after hnRNP A2/B1 knockdown (Fig. 5A, 5B). Taken together, the results indicate that the primary effect of hnRNPA2/B1 expression is to change cell cycle progression before the induction of cell differentiation.

hnRNP A2/B1 Knockdown Induces the G0/G1 Phase Arrest Through the Degradation of Cyclin D1, Cyclin E, and Cdc25A

In order to investigate how hnRNP A2/B1 knockdown prevents the G1/S transition in hESCs, we next examined cyclin expression in hnRNP A2/B1 knockdown hESCs by Western blot analysis (Fig. 6A). Cyclin B1 expression was not altered in hnRNP A2/B1 knockdown hESCs. However, cyclin D1 and E expression were decreased in the hnRNP A2/B1 knockdown hESCs. Cdc25A, required for the G1/S transition, also appeared to be degraded (Fig. 6A, the fourth panel), suggesting that degradation of cyclin D1, cyclin E, and Cdc25A proteins is involved in the G0/G1 phase arrest. Cdk2 makes complexes with cyclin E and bears significant impacts on the maintenance of hESC pluripotency [[49, 50]]. As Cdk2 is a major regulatory molecule of G1/S boundary and Cdk2 activation requires phosphorylation of threonine 160 and dephosphorylation of threonine 14 and tyrosine 15 [[49-52]], we examined Cdk2 phosphorylation by Western blot analysis (Fig. 6A, 6B). As compared to those of control hESCs, Cdk2 activating phosphorylation was not altered in the hnRNP A2/B1 knockdown hESCs (Fig. 6B). As expected from the degradation of Cdc25A protein (Fig. 6A, the fourth panel), however, Cdk2 inhibitory phosphorylation was slightly increased. Taken together, the results suggest that degradation of cyclin D1, cyclin E, and Cdc25A proteins is closely associated with the G0/G1 phase arrest in the hnRNP A2/B1 knockdown hESCs.

image

Figure 6. Heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP A2/B1) knockdown induces the G0/G1 phase arrest through the degradation of cyclin D1, cyclin E, and Cdc25A, the expression of p27, and the phosphorylation of p53 and Chk1. (A, B): Western blot analyses of expression and phosphorylation of cyclins and Cdks in control (Con) and hnRNP A2/B1 siRNA-transfected human embryonic stem cells (hESCs) (A2/B1 Si). Cell lysates from control or hnRNP A2/B1 siRNA-transfected hESCs were subjected to Western blot analysis with indicated antibodies to analyze the expression of various cyclins and Cdks. Shown are representative of at least two independent experiments. The signal intensities of Western blot were measured quantitatively using the Image J software, and β-actin was used as a normalization control. (C): Western blot analysis of p16, 21, and p27 in control and hnRNP A2/B1 siRNA-transfected hESCs. Shown are representative of at least two independent experiments. (D): Expression and phosphorylation analysis of p53 in control and hnRNP A2/B1 siRNA-transfected hESCs. Cell lysates from control and hnRNP A2/B1 siRNA-transfected hESCs were subjected to Western blot analysis to analyze the expression of p53 and phosphorylation of p53 at serine 15 and 20. Shown are representative Western blots from two independent experiments. (E): Expression and phosphorylation analysis of Chk1 and Chk2 in control and hnRNP A2/B1 siRNA-transfected hESCs. Cell lysates from control and hnRNP A2/B1 siRNA-transfected hESCs were subjected to Western blot analysis to analyze the expression of Chk1 and Chk2 and the phosphorylation of Chk1 serine 345 and Chk2 threonine 68. Shown are representative Western blots from two independent experiments.

Download figure to PowerPoint

hnRNP A2/B1 Knockdown Increases p27 Expression

Passage through G1 into S-phase is a key regulatory point in the cell cycle of mammalian cells and is regulated by cyclin-dependent kinase inhibitors (CKIs) [53]. Previous studies suggested hnRNP A2 regulates the transcript levels of p16, p21, and p27 in cancer cells and fibroblasts [[48, 54]]. In order to examine whether the G0/G1 phase arrest induced by hnRNP A2/B1 knockdown is also regulated by CKIs, the expression levels of p16, p21, and p27 proteins were examined. As compared to control hESCs, the protein level of p27 was increased approximately fourfold in the hnRNP A2/B1 knockdown hESCs, but the p16 protein was not increased. Interestingly, despite the increased expression of p21 mRNA (Fig. 5C), the p21 protein was hardly detectable in the hnRNP A2/B1 knockdown hESCs (Fig. 6C), which is consistent with the previous observation [55]. Thus, the result suggests that hnRNP A2/B1 promotes the G1/S transition through suppression of p27 expression.

hnRNP A2/B1 Knockdown Induces Phosphorylation of p53 and Chk1

The regulation of p53 activity by phosphorylation is closely involved in hESC survival, differentiation, and cell cycle [[50, 56]]. Therefore, we examined the phosphorylation of p53 in hnRNP A2/B1 knockdown hESCs (Fig. 6D). The expression of p53 protein was slightly increased in the hnRNP A2/B1 knockdown hESCs. Interestingly, the phosphorylation of p53 at serine 15 was markedly increased, while the phosphorylation of p53 at serine 20 was not altered. A previous study showed that irradiated hESCs are arrested in the G1 phase and p53, Chk1, and Chk2 are closely involved in the G1 phase arrest [55]. Therefore, we further examined the expression and phosphorylation of Chk1 and Chk2 in the hnRNP A2/B1 knockdown hESCs (Fig. 6E). The expression of Chk1 was not significantly altered, while the expression of Chk2 was slightly increased in the hnRNP A2/B1 knockdown hESCs. The phosphorylation of Chk1 at serine 345 was markedly increased, while the phosphorylation of Chk2 at threonine 68 was slightly increased. Chk1-mediated protein degradation of Cdc25A has been known as a critical control point for the G1/S transition [57]. As expected, the degradation of Cdc25A protein was also observed in the hnRNP A2/B1 knockdown hESCs (Fig. 6A, the fourth panel). Thus, the results suggest that hnRNP A2/B1 knockdown prevents the G1/S transition of hESCs through the regulation of p53 and Chk1 phosphorylation.

hnRNP A2/B1 Knockdown Inhibits PI3K/Akt Signaling

In an attempt to find the signaling pathways involved hnRNP A2/B1-mediated hESC self-renewal and pluripotency, we found that hnRNP A2/B1 knockdown dramatically reduced Akt phosphorylation in hESCs, as compared to control hESCs (Fig. 7A). The result suggests that hnRNP A2/B1 expression is closely associated with activation of the PI3K/Akt signaling pathway, which is critical for maintenance of hESC pluripotency and self-renewal [[58-61]]. Recent studies showed that hnRNP A2/B1 is a downstream target of Akt1 during mitochondrial respiratory stress [[25, 26]]. To confirm whether the PI3K/Akt signaling pathway also regulates the expression of hnRNP A2/B1, therefore, we inhibited the PI3K/Akt signaling activity using LY294002, a specific PI3K inhibitor, and examined the effect on the expression of hnRNP A2/B1. LY294002 treatment obviously decreased the expression of hnRNP B1, although it marginally decreased the expression of hnRNP A2 (Fig. 7B), indicating that the PI3K/Akt signaling also regulates hnRNP A2/B1 expression in hESCs. Taken together, the results suggest that hnRNP A2/B1 expression is closely involved in the PI3K/Akt signaling pathway in hESCs.

image

Figure 7. Knockdown of hnRNP A2/B1 downregulates phosphorylation of Akt in human embryonic stem cells (hESCs). (A): Western blot analysis showing phosphorylation of Akt in hESCs transfected with control (Con) or hnRNP A2/B1 siRNAs (A2/B1 Si). Cell lysates were separated by SDS-PAGE and Western-blotted with indicated antibodies. Shown are representative Western blots from three independent experiments. The signal intensities of Western blot were measured quantitatively using the Image J software. (B): Expression and phosphorylation analysis of Akt1/2/3 and hnRNP A2/B1 in dimethyl sulfoxide (DMSO)- (Con) or LY294002-treated hESCs (LY). Feeder-free H9 cells were treated with DMSO or LY294002 for 5 days and the levels of hnRNP A2/B1, Akt1/2/3, and phosphorylated Akt1/2/3 were examined by Western blot analysis. Shown are representative Western blots from two independent experiments. β-Actin was used as a normalization control. (C): Proposed model of hnRNP A2/B1-mediated maintenance of hESC self-renewal and pluripotency. Abbreviation: hnRNP A2/B1, heterogeneous nuclear ribonucleoprotein A2/B1.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Previous many microarray data suggested that hnRNP A2/B1 is highly expressed in undifferentiated hESCs and downregulated in somatic tissues [62]. Recent phosphoproteomic analysis also suggests that hnRNP A2/B1 may be highly expressed in undifferentiated hESCs [63]. Other membrane proteomic analyses also suggested that hnRNP A2/B1 might be expressed in the membrane fraction of undifferentiated hESCs [[63-66]]. Chromatin immunoprecipitation and microarray-based analysis of hESCs showed that hnRNP A2/B1 is a target gene of NANOG and SOX2 [3]. Thus, it seems that hnRNP A2/B1 is required for hESC self-renewal and pluripotency. Until this study, however, the functional role of hnRNP A2/B1 in hESCs has not been studied. In this study, we demonstrate for the first time that hnRNP A2/B1 is required to maintain hESC self-renewal and pluripotency.

hnRNP A2/B1 plays a role in the regulation of the cell cycle and cell proliferation in some human cancer cells [[47, 48, 67]]. In this study, we showed that hnRNP A2/B1 knockdown inhibited hESC proliferation and induced cell cycle arrest in the G0/G1 phase (Fig. 5A, 5B). The G0/G1 arrest was accompanied by an increase in the protein expression of Cdk inhibitor p27 (Fig. 6C). p27 is a negative regulator of protein kinases Cdk2/cyclin E and Cdk2/cyclin A, which drives the G1/S transition in cell cycle [[53, 68]]. p27 may be essential for the in vitro differentiation process of mESCs and hESCs as well [[69-71]], because mESCs lacking p27 undergo apoptosis before completing their differentiation program [[69, 72]]. hnRNP A2/B1 knockdown hESCs underwent extensive differentiation into multicell lineages without any signs of apoptosis (Fig. 3 and Supporting Information Fig. S2). p27 protein expression was too low to detect in undifferentiated hESCs, but markedly increased in hnRNP A2/B1 knockdown hESCs (Fig. 6C). The results suggest that hnRNP A2/B1 promotes the G1/S transition in hESCs through suppression of p27 expression. Interestingly, p27 mRNA is immunoprecipitated by an hnRNP A2 antibody [51], which suggests that hnRNP A2/B1 modulates p27 expression through a direct association between p27 mRNA and hnRNP A2 protein in vivo.

Recent studies suggested that p53 may be directly associated with hnRNP A2/B1 in some cancer cell lines [[17, 18]], and other studies also suggested that hnRNP A2 gene itself may be a target of p53, since p53 represses the expression of hnRNP A2 mRNA [12]. However, it seems that hnRNP A2 does not modulate p53 expression directly, because p53 mRNA and protein levels are not significantly altered in hnRNP A2 knockdown cancer cells [[47, 48]]. This study showed that the protein level of p53 was slightly increased in hnRNP A2/B1 knockdown hESCs (Fig. 6D). Furthermore, the level of phosphorylated p53 at serine 15 was markedly increased (Fig. 6D). The regulation of p53 activity by phosphorylation is closely involved in hESC survival, differentiation, and cell cycle [[50, 56]]. The phosphorylation of p53 at serine 15, serine 315, and serine 392 activates p53 activity and downregulates NANOG expression during differentiation in mESCs [73]. It is also known that NANOG accelerates the G1/S transition in hESCs and binds to hnRNP A2/B1 gene [[3, 74]]. This study showed that NANOG expression was approximately sevenfold decreased in hnRNP A2/B1 knockdown hESCs (Fig. 2E), suggesting that hnRNP A2/B1 promotes the G1/S transition of hESCs through promotion of NANOG expression and inhibition of p53 activity.

The expression level of hnRNP A2/B1 was gradually decreased during differentiation of hESCs induced by RA (Fig. 1C). When hnRNP A2/B1 was knocked down, hESCs were differentiated into cell types of all three germ layers. During the differentiation process, hESCs underwent the process of EMT, displaying a scattered and more mesenchymal-like morphology (Fig. 2B). The detailed analysis further showed that E-cadherin was downregulated and vimentin and E-cadherin repressor molecules were increased in hnRNP A2/B1 knockdown hESCs (Fig. 4). Thus, hnRNP A2/B1 was required to maintain the undifferentiated and epithelial phenotypes of hESCs. Our findings are also supported by two recent studies in which hnRNP A2/B1 expression was decreased when undifferentiated and epithelial neuroblastoma and gastric adenocarcinoma were induced to differentiate [[17, 18]]. However, the previous study reported that hnRNP A2/B1 knockdown is correlated with an increase of E-cadherin expression and downregulation of E-cadherin repressor molecules in the mesenchymal type of lung cancer cells [46]. The latest study also shows the inverse correlation between the emergence of hnRNP A2/B1 and the loss of E-cadherin expression in pancreatic cancer tissues [48]. These results suggest that hnRNP A2/B1 may be required to maintain the mesenchymal phenotype of some cancer cells. Thus, the exact role of hnRNP A2/B1 is controversial between undifferentiated epithelial types and mesenchymal types. In this regard, it is interesting to note that hnRNP A2/B1 functions in maintaining the expression of differentiation markers of SMCs, one of mesenchymal lineage cell type, during guided differentiation of mESCs into SMCs [42]. The levels of the SMC-specific markers were also decreased in hnRNP A2/B1 knockdown hESCs (Supporting Information Fig. S3), suggesting that hnRNP A2/B1 expression may be also necessary for SMC differentiation during guided differentiation of hESCs into SMCs. Therefore, it would be interesting to examine whether hnRNP A2/B1 expression is increased during guided differentiation of hESCs into SMCs.

We demonstrate for the first time that hnRNP A2/B1 regulates the self-renewal and pluripotency of hESCs. hnRNP A2/B1 knockdown hESCs showed nonapoptotic-related decrease in cell proliferation (Supporting Information Figs. S2, S4), indicating that hnRNP A2/B1 is required for control of cell proliferation in hESCs. Some signaling pathways required for self-renewal are necessary for regulating the cell cycle [[70, 75]], suggesting a direct relationship between control of cell proliferation and self-renewal. In an attempt to find the signaling pathways involved in hnRNP A2/B1-mediated cell proliferation, we found that hnRNP A2/B1 knockdown severely reduced the level of phosphorylated Akt in hESCs (Fig. 7A), suggesting that hnRNP A2/B1 regulates hESC proliferation through the PI3K/Akt signaling pathway. The decreased level of cyclin D1 also supports that hnRNP A2/B1 regulates hESC proliferation through the PI3K/Akt signaling pathway (Fig. 6A), because it is known that expression of cyclin D1 is regulated by the PI3K/Akt-dependent pathway [76]. The increased level of p27 expression also supports above finding, because Akt induces ubiquitin-dependent degradation of p27 [77]. Interestingly, inhibition of PI3K/Akt also decreased the expression of hnRNP A2/B1 (Fig. 7B), suggesting that hnRNA A2/B1 is a downstream target of the PI3K/Akt signaling pathway in hESCs. The result is supported by recent studies showing that hnRNP A2/B1 is a downstream target of Akt1 during mitochondrial respiratory stress [[25, 26]]. Conversely, the PI3K/Akt signaling pathway is necessary for the maintenance of hESC self-renewal and pluripotency [[58-60]]. Therefore, our findings suggest that hnRNP A2/B1 regulates hESC self-renewal and pluripotency through the control of cell proliferation in a PI3K/Akt-dependent manner.

CONCLUSION

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

In conclusion, we demonstrate for the first time that expression of hnRNP A2/B1 is necessary for maintaining hESC self-renewal and pluripotency. hnRNP A2/B1 knockdown impairs hESC self-renewal and pluripotency, which subsequently leads to differentiation of hESCs into multicell lineages. During the differentiation process, hnRNP A2/B1 knockdown also promotes the process of EMT. hnRNP A2/B1 knockdown decreases hESC proliferation via inhibition of the G1/S transition before cellular differentiation, and inhibition of the G1/S transition is partly due to degradation of cyclin D1, cyclin E, and Cdc25A. Subsequent studies further show that the G1/S transition is also controlled by the expression of p27 protein and the phosphorylation of p53 and Chk1. In addition, analysis of signaling molecules reveals that hnRNP A2/B1 regulates hESC proliferation in a PI3K/Akt-dependent manner. Proposed model of hnRNP A2/B1-mediated maintenance of hESC self-renewal and pluripotency is shown (Fig. 7C). These findings provide for the first time mechanistic insights into how hnRNP A2/B1 regulates hESC self-renewal and pluripotency.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

This study was supported in part by the National Research Foundation of Korea (2009-0053111, 2011-0002659, and 2012-0006144) and the Converging Research Center Program funded by the Ministry of Education, Science and Technology (Project No. 2012K001480).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
stem1366-sup-0001-suppfig1.tif64KSupporting Information Figure 1.
stem1366-sup-0002-suppfig2.tif129KSupporting Information Figure 2.
stem1366-sup-0003-suppfig3.tif84KSupporting Information Figure 3.
stem1366-sup-0004-suppfig4.tif58KSupporting Information Figure 4.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.