Renewal - Wiley Online Library

4 downloads 14 Views 771KB Size Report
out feeder cells, even in chemically defined xeno-free ...... tory cascades underlying the unique hESC molecular program. .... 41 Lee YI, Seo M, Kim Y et al.

EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS Optimal Suppression of Protein Phosphatase 2A Activity Is Critical for Maintenance of Human Embryonic Stem Cell Self-Renewal BYUNG SUN YOON,a,b EUN KYOUNG JUN,a GYUMAN PARK,c SEUNG JUN YOO,a JAI-HEE MOON,a CHEONG SOON BAIK,d AEREE KIM,e HYUNGGEE KIM,a JONG-HOON KIM,f GOU YOUNG KOH,g HOON TAEK LEE,h SEUNGKWON YOUa Laboratories of Cell Function Regulation and fStem Cell Biology, College of Life Sciences and Biotechnology and bInstitute of Life Science and Natural Resources, Korea University, Seoul, Korea; cResearch Institute for Skin Image and eDepartment of Pathology, College of Medicine, Korea University Guro Hospital, Seoul, Korea; d SamKwang Medical Laboratories, Seoul, Korea; gNational Research Laboratory of Vascular Biology, Department of Biological Sciences, Korea Advanced Institute of Science and Technology, Daejeon, Korea; hDepartment of Animal Biotechnology, Bio-Organ Research Center/Animal Resources Research Center, Konkuk University, Seoul, Korea a

Key Words. Human embryonic stem cells • Self-renewal • Protein phosphatase 2A • Okadaic acid • Xeno-free

ABSTRACT The self-renewal of embryonic stem cells involves a balance between processes governed by crosstalk between intrinsic and extrinsic factors. We hypothesized that protein serine/threonine phosphatase 2A (PP2A) may play a central role in the signaling pathways that regulate human embryonic stem cell (hESC) self-renewal. Biochemical analyses revealed that PP2A activity gradually increases over the course of hESC differentiation; PP2A/C and PP2A/A levels also increased. The overexpression of PP2A/C or the addition of PP2A activator C2-ceramide promoted hESC differentiation. Accordingly, the addition of PP2A inactivator okadaic acid (OA) maintained hESC self-renewal in the absence of basic fibroblast growth fac-

tor (bFGF). The hESCs maintained with OA expressed pluripotency markers and exhibited substantial telomerase activity with normal karyotypes. The hESCs were able to differentiate into derivatives of the three germ layers, both in vitro and in vivo. Furthermore, the addition of OA and bFGF enabled the maintenance of hESC self-renewal without feeder cells, even in chemically defined xeno-free media. These findings shed a light on the role of PP2A in hESC differentiation and provide a novel strategy for maintaining the self-renewal capability of hESC in bFGFfree, feeder cell-free, and xeno-free media through the optimal suppression of PP2A activity using OA. STEM CELLS 2010;28:874–884

Disclosure of potential conflicts of interest is found at the end of this article.

INTRODUCTION The self-renewal of embryonic stem cells (ESCs) involves a balance between proliferative potential, inhibition of differentiation, and the prevention of senescence or apoptosis. Each of these processes is governed by intrinsic and extrinsic factors intimately linked by crosstalk between signaling pathways [1]. Human embryonic stem cells (hESCs) were initially isolated on mouse fibroblast feeder layers in medium containing serum [2], conditions similar to those first used to isolate mouse embryonic stem cells (mESCs) [3]. Despite these initial culture similarities, the factors mediating the self-renewal of hESCs and mESCs appear to be distinct. In the presence of serum and leukemia inhibitory factor (LIF), the activation of the Janus kinase (JAK)/signal transducer and activator of tran-

scription-3 (STAT3) pathway supports the feeder-independent growth of mESCs [4]. Under comparable culture conditions, LIF does not maintain hESCs, and the JAK/STAT3 pathway does not appear to become activated in conditions that maintain hESCs [5]. Likewise, although bone morphogenic protein-4 (BMP4) has been shown to block mESC differentiation along the neuroectoderm pathway [6], it induces trophectoderm differentiation in hESCs [7]. In contrast to JAK/STAT3 and BMP signaling, Wnt/b-catenin signaling participates in pluripotency control in both species [8], but Wnt activity alone is not sufficient to maintain hESC pluripotency [9]. Consequently, the signaling pathways implicated in the maintenance of hESC pluripotent status are not well understood. Basic fibroblast growth factor (bFGF), which has been known to activate pathways involving protein kinase B/AKT, extracellular signal-regulated kinase (ERK), and mothers

Author contributions: B.S.Y.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing; E.K.J., G. P.: collection and/or assembly of data, data analysis and interpretation; S.J.Y., J.H.M., C.S.B., A.K.: collection and/or assembly of data; H.K., J.H.K.: data analysis and interpretation; G.Y.K., H.T.L.: data analysis and interpretation, manuscript writing; S.Y.: conception and design, administrative support, collection and/or assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript. Correspondence: Seungkwon You, Ph.D., Laboratory of Cell Function Regulation, College of Life Sciences and Biotechnology, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-701, Korea. Telephone: 82-2-3290-3057; Fax: 82-2-3290-3507; e-mail: [email protected] Received September 22, 2009; accepted for publication March 9, 2010; first published online in STEM CELLS EXPRESS March C AlphaMed Press 1066-5099/2009/$30.00/0 doi: 10.1002/stem.412 19, 2010. V

STEM CELLS 2010;28:874–884

Yoon, Jun, Park et al.


against decapentaplegic (MAD), human homologue of Drosophila MAD (SMAD) [10, 11], appears to play a key role in sustaining hESC self-renewal and is included in nearly all reported medium formulations for hESC derivation and propagation. The activation of the Wnt pathway using glycogen synthase kinase-3b (GSK-3b) inhibitor 6-bromoindirubin-30 oxime (BIO) also promotes the self-renewal of mouse and human ESCs grown without LIF or Wnt [8]. Other factors suggested to play a role in supporting the self-renewal of hESCs include transforming growth factor-b1 (TGF-b1), activin A, and platelet-derived growth factor (PDGF) [9, 12, 13]. Thus, hESC self-renewal may be maintained by redundant signal transduction pathways. Although these studies have focused on identifying growth factors and conditions that support hESC self-renewal, little is known about the receptor kinases and phosphatases involved in the signal transduction pathways when hESCs are exposed to conditions favorable for self-renewal. The serine/threonine phosphatase protein phosphatase 2A (PP2A) is an important regulator of numerous target proteins in diverse signaling pathways that control cell growth and proliferation, apoptosis, transcription, and translation [14]. This phosphatase exists as a heterotrimer consisting of a 36kDa catalytic subunit (PP2A/C), a 65-kDa scaffolding subunit (PP2A/A), and a variable regulatory subunit designated as PP2A/B, B0 , B00 , and B000 . The AC catalytic complex exerts phosphatase activity, whereas the different B subunits recruit PP2A/C to distinct subcellular locations and define the substrate target [14]. For example, PP2A modulates cellular proliferation or apoptosis by regulating the phosphoinositide 3-kinase (PI3K)/AKT, mitogen-activated protein kinase kinase (MEK)/ERK, and GSK-3b pathways [15, 16] and the treatment of cells with the PP2A inhibitor okadaic acid (OA) activates the PI3K/AKT and MEK/ERK family kinases [17, 18] and inactivates GSK-3b [15]. In addition, PP2A appears to directly dephosphorylate c-Myc, and OA-mediated inhibition of PP2A stabilizes c-Myc [19]. Because the PI3K/AKT, MEK/ERK, GSK-3b, and c-Myc pathways are also known to be essential for maintaining ESC self-renewal [8, 10, 20], we hypothesized that PP2A may play a central role in the signaling pathways that regulate hESC self-renewal. In this study, we simultaneously investigated the regulated expression of AKT, GSK-3b, and c-Myc by PP2A in hESCs. We used OA to uncover a critical role of PP2A in regulating hESC self-renewal and developed a defined medium for hESC culture.




hESC Culture Two hESC lines, HSF-6 and Miz-hES4, were maintained on mitomycin C-treated CF1 mouse feeder layers in Dulbecco’s modified Eagle medium (DMEM)-F12 containing 20% knockout serum replacement (SR; Invitrogen, Carlsbad, CA, http:// and 4 ng/mL recombinant bFGF (R&D Systems, Minneapolis, MA, as previously described [21, 22]. To examine the effect of PP2A on hESC differentiation, varying concentrations of OA were added to hESCs cultured on the feeder layer without bFGF. To examine the effect of signaling pathways on hESC differentiation, various concentrations of LY294002, U0126 (Sigma-Aldrich, St. Louis, MO, http://www.sigmaaldrich. com), 10058-F4, C2-ceramide, and inactive analog C2-dihydroceramide [23] (Calbiochem, San Diego, CA, http:// were added to the medium. To induce embryoid body (EB) formation, hESCs were

cally dissociated into small clumps and cultured in suspension with EB medium without bFGF or OA for 14 days [21, 22]. To determine a novel culture method for hESC self-renewal under feeder-free or xeno-free conditions, the hESCs were cultured onto a gelatin- or CELLstart (Invitrogen: humanized substrate)-coated plate in DMEM/F12/SR or DMEM/F12/B27 medium supplemented with 1 nM OA, 4 ng/mL bFGF, or 1 nM OA plus 4 ng/mL bFGF.

Western Blot Analysis Total protein was extracted using radioimmunoprecipitation buffer containing protease inhibitor cocktail (Roche, Mannheim, Germany, The proteins were separated by 4%–12% gradient-precast sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Invitrogen). The membrane was incubated with the appropriate primary antibody (Supporting Information Table 1), and then with horseradish peroxidase-conjugated goat antibodies against mouse or rabbit immunoglobulin G. The secondary antibodies were detected by using a Super Signal West Pico kit (Pierce, Rockford, IL, http://www.

PP2A/C Activity Assay The enzymatic activity of PP2A was assessed after PP2A/C immunoprecipitation using a malachite green-based phosphatase assay (Upstate, Chicago, IL, The hESCs were lysed with 0.3 mL phosphatase extraction buffer and the supernatants incubated with protein A agarose slurry in the presence of an anti-PP2A/C antibody (upstate) at 4 C with constant rocking for 2 hours. After intense washing with 700 lL Tris-buffered solution and 500 lL optimized Ser/Thr buffer, the complexes were resuspended in 20 lL Ser/Thr buffer and the K-R-pT-I-R-R phosphopeptide added. The samples were incubated at 30 C in a shaking incubator for 10 minutes. The supernatants (25 lL) were transferred to a 96-well plate and the released phosphate measured by adding 100 lL malachite green phosphate detection solution. After allowing color development for 15 minutes, the plate was read at 650 nm.

Rho-Associated Kinase Inhibition and Transfection For Rho-associated kinase inhibition [24], 10 lM Y-27632 (Calbiochem) was added to the culture medium for 1 hour before detaching the hESCs from the feeder layer. The cells were harvested and incubated in trypsin-ethylenediaminetetraacetic acid at 37 C for 4 minutes. The dissociated cells were seeded the day before transfection on Matrigel-coated 100-mm plates at an initial seeding density of 2  106 cells. Transfection was carried out with 18 lg plasmid DNA (pWPXL-IRESEGFP and pShuttle-PP2A/C-IRES-hrGFP-2A) using TurboFect (MBI Fermentas, St. Leon-Rot, Germany, or small interfering RNA (siRNA) (mock siRNA and siPP2A/C siRNA) using siRNA transfection kit (Dharmacon Research, Inc., Lafayette, CO, according to the manufacturer’s protocol.

Coimmunoprecipitation Cells were solubilized in ice-cold 0.3% CHAPS lysis buffer, and 500 lg of the cell lysate was precleared with 40 lL protein A/G beads. The precleared lysates were incubated with PP2A/C antibody-bound beads for 5 hours, after which the beads were washed with 1 mL 0.3% CHAPS buffer and boiled in sample buffer. The immunoprecipitates were analyzed by Western blotting.

Immunostaining Cells were fixed for 30 minutes at room temperature in phosphate-buffered saline (PBS) containing Ca2þ, Mg2þ, and 4%

Self-Renewal of hESCs by PP2A


paraformaldehyde. The EBs were fixed in 4% paraformaldehyde, transferred to 20% sucrose, frozen in the optimum cutting temperature compound (Tissue Tek, Sakura, Japan), and sectioned into 10-lm thick slices. Immunostaining was performed using standard protocols [21]. The primary antibodies are listed in Supporting Information Table 1.

Reverse Transcription-Polymerase Chain Reaction Total RNAs were prepared using Trizol reagent (Invitrogen). Standard reverse transcription (RT) was performed on 500 ng of total RNA using oligo d(T)12–18 primer and superscriptase II (Invitrogen). The RT-polymerase chain reaction (PCR) was carried out with 1 lL of cDNA template, 10 pmol of primers (Supporting Information Table 2), and PCR premix (Bioneer, Seoul, Korea, in a thermocycler T3000 (Biometra, Goettingen, Germany, http://www.biometra. de) using a 5-minute denaturation at 94 C; 30 cycles of 94 C for 30 seconds, 62 C for 30 seconds, and 72 C for 30 seconds; and a final 10-minute extension at 72 C.

Telomeric Repeat Amplification Protocol Assay Telomeric repeat amplification protocol (TRAP) assays were performed using the TRAPEZE telomerase detection kit (Chemicon International, Temecula, CA, http://www. according to the manufacturer’s protocol. Briefly, the cells were lysed in 200 lL cold TRAP lysis buffer. A quantity of extract representing 1  103 cells was incubated with 5 lL TRAP reaction buffer, 1 lL dNTP Mix, 1 lL TS primer, 1 lL TRAP primer mix, 0.4 lL Taq polymerase, and 39.6 lL H2O. The 50 lL reaction mix was incubated at 30 C for 30 minutes, after which PCR (94 C for 30 seconds, 55 C for 30 seconds, and 72 C for 1 minute) was performed for 30 cycles in a thermocycler T3000 (Biometra). The PCR products were separated by electrophoresis on 10% Novex TBE gels (Invitrogen) and stained with ethidium bromide (Fluka, Buchs, Switzerland,

Karyotype Analysis Karyotype analysis was performed by Samkwang Medical Laboratories (Seoul, Korea, using standard G banding methods. Briefly, hESCs were treated with 0.1 lg/mL colcemid (Invitrogen) for 2–3 hours. The

cells were trypsinized and resuspended in hypotonic KCl solution (Sigma-Aldrich) at 37 C for 20 minutes and then fixed in 3:1 methanol and acetic acid.

Teratoma Formation Serially passaged hESCs (>25 passages) were injected into the testes of nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice. Seven to eight weeks later, the mice developed teratomas, which were removed, immediately rinsed in PBS, fixed in 4% formalin, and embedded in paraffin. Tissue sections 5–6 lm-thick were cut and processed for hematoxylin-eosin staining.

Fluorescence-Activated Cell Sorting Analysis The cells were trypsinized and placed into fluorescence-activated cell sorting (FACS) tubes at 1  106 cells/tube (BD Biosciences, San Diego, CA, After two rinses with cold buffer solution, the cells were incubated on ice for 30 minutes with primary antibody (Supporting Information Table 1), washed with 1% fetal bovine serum (FBS) in PBS, resuspended in 100 lL Cy3-labeled secondary antibody, and incubated for an additional 30 minutes on ice. The cells were then washed, fixed with fixative solution, and resuspended in PBS containing 1% FBS for FACS analysis.

Statistical Analysis The data were analyzed by analysis of variance using the general linear model procedures of the Statistical Analysis System (9.13 Package). The data were expressed as means 6 SD. p < .05 was considered significant.

RESULTS PP2A Activity Is Inversely Related to hESC Self-Renewal Human ESC undifferentiation was maintained in the presence of bFGF, whereas differentiation was induced by the withdrawal of bFGF. First, we examined the phosphorylation status of key signaling cascades involved in cellular proliferation and differentiation in undifferentiated and differentiated

Figure 1. Self-renewing hESCs exhibit lower PP2A activity than differentiated hESCs. (A): Undifferentiated hESCs and hESCs induced to differentiate spontaneously by withdrawing bFGF (DIFF) were lysed and their whole cell lysates immunoblotted and probed with antibodies against the indicated proteins. The results indicate that ERK1/2, AKT, and GSK-3b participate in maintaining the self-renewal capability of hESCs. (B): After withdrawing bFGF, hESC cultures were subjected to Western blot analysis on days 0, 3, and 6 using the indicated antibodies. The expression of PP2A/C was upregulated and its degradable form, pPP2A/C, was downregulated during hESC differentiation. (C): After withdrawing bFGF, hESC cultures were subjected to PP2A activity assays on days 0, 2, and 4. PP2A activity was found to be upregulated during hESC differentiation. Error bars indicate 6 SD; **, p < .01 versus day 0; n ¼ 3. (D): Transfection indicated that PP2A leads to hESC differentiation. Single hESCs treated with Y-27632 were transfected with pWPXL-GFP (left) or pSuttle-PP2A/C-IRES-hrGFP-HA (center). The SSEA-4 marker expression was measured in GFP-positive cells after 4 days of transfection (right). Error bars indicate 6 SD; *, p < .05 versus mock control (GFP); n ¼ 3. (E, F): Transfection of PP2A small interfering RNA decreased the expression of PP2A, the number of differentiated colonies, and PP2A activity with the increased Oct-4, Nanog, and Sox2 expression compared with those of mock treatment. Error bars indicate 6 SD; *, p < .05; **, p < .01 versus mock control; n ¼ 3. (G): PP2A activation induced hESC differentiation. The hESCs were treated with the PP2A activator C2-ceramide (lower left) or the inactive analog C2-dihydroceramide (upper left). The differentiation and PP2A activity on day 6 after ceramide treatments were determined by analyzing colony morphology and alkaline phosphatase (AP) immunostaining (upper graph) and immunoprecipitated PP2A activity (lower graph). Error bars indicate 6 SD; *, p < .05; **, p < .01 versus control (FGF alone); n ¼ 4. (H): The morphology of OAtreated hESCs (A–D). The bFGF-maintained hESCs were cultured without bFGF for 2 days and then treated with 0 (A), 0.1 (B), one (C), or 10 (D) nM OA. The cells underwent either differentiation over three passages (P3) in the absence (A) or presence (B) of 0.1 nM OA or cell death by 10 nM OA (d, Day 2). The cells, however, but the cells expressed SSEA-4 (E–G) and Oct-4 (H–J) and remained pluripotent for up to 80 passages in the presence of 1 nM OA (c, P80). (I): Western blot indicated that 1 nM OA alters the expression of pPP2A/C but not PP2A/C, PP2A/ A, PP1, and PP2B. The hESCs were incubated without bFGF for 2 days and then stimulated for 1 day with OA as described in (E) at the indicated concentration. Scale bar ¼ 200 lm. Abbreviations: bFGF, basic fibroblast growth factor; DAPI, 4,60 -diamidino-2-phenylindole; DIFF, differentiation; ERK, extracellular signal-regulated kinase; FGF, fibroblast growth factor; GSK-3b, glycogen synthase kinase-3b; GFP, green fluorescent protein; hESC, human embryonic stem cell; OA, okadaic acid; pAKT, phospho AKT; pERK, phospho ERK; pGSK-3b, phospho GSK-3b; pPP2A, phospho PP2A; pPP1, phospho PP1; PP2A, protein phosphatase 2A; SSEA, stage specific embroynic antigen.

Yoon, Jun, Park et al.


Figure 1.

hESCs by Western blot (Fig. 1A). Compared with differentiated hESCs, undifferentiated hESC had higher levels of Oct-4 and phosphorylated AKT (active), ERK (active), GSK-3b (degradable), and PP2A/C (inactive). In particular, phosphorylated PP2A/C gradually declined over 6-day differentiation,

whereas levels of PP2A/C and PP2A/A increased, but levels of PP2A/B, PP2B, and phosphorylated protein phosphatase 1 (PP1) were unchanged (Fig. 1B). Conversely, PP2A activity gradually increased during differentiation (100 6 3.75% at day 0, 141 6 4.63% at day 2, and 237 6 38.27% at day 4;


Fig. 1C). The level of Sox2 and phosphorylation of AKT and GSK-3b declined during differentiation, whereas ERK phosphorylation was unchanged (Fig. 1B). Thus, of the major protein phosphatase-related proteins, only PP2A appears to be actively, although negatively, involved in bFGF signaling maintaining hESC self-renewal. To further elucidate the role of PP2A in hESCs, dissociated single hESCs were transfected with PP2A/C-GFP or green fluorescent protein (GFP) (mock control) expression vectors. The marker of undifferentiated hESCs, SSEA-4, was significantly decreased in PP2A/C-GFPtransfected cells compared with control (Fig. 1D), indicating that the forced increase in PP2A protein induced differentiation. The marker of undifferentiated hESCs, SSEA-4, was significantly decreased in PP2A/C-GFP-transfected cells compared with control (Fig. 1D). On the other hand, knockdown of PP2A/C by siRNA significantly decreased expression of PP2A/C, rate of differentiated colonies, and PP2A activity whereas significantly increased expression of Oct-4, Nanog, and Sox2 (Fig. 1E, 1F) in the absence of bFGF. To validate the role of PPA2 in hESC differentiation, undifferentiated hESCs were treated with either the PP2A activator C2-ceramide and inactive analog C2-dihydroceramide. As shown by colony morphology and alkaline phosphatase (AP) activity, C2-ceramide, but not C2-dihydroceramide, increased PP2A activity and induced hESC differentiation in a dose-dependent manner, even in the presence of bFGF (Fig. 1G), indicating that PP2A activation induces hESC differentiation. On the basis of these observations, we sought to determine whether hESCs could be maintained in an undifferentiated state by inactivating PP2A. In the absence of bFGF, 1 nM OA maintained undifferentiated, SSEA-4- and Oct-4positive hESCs for over 80 passages, whereas 0 or 0.1 nM OA induced differentiation to flattened cells after three passages, and 10 or 100 nM OA gradually damaged the hESCs and feeder cells over time (Fig. 1H). Supporting these morphological changes, biochemical analyses revealed that 1 nM OA did not alter Oct-4, Nanog, and Sox2 levels, but 0 or 0.1 nM OA resulted in decreased Oct-4 and Sox2 levels (Fig. 1I). Notably, when the hESCs were treated with 1 nM OA plus bFGF, they differentiated with significant reductions in AP activity, Oct-4 protein, and undifferentiated colony formation after two passages (Supporting Information Fig. 1). Because OA is known to regulate not only PP2A but also PP1 and PP2B [25], we investigated the affect of 1 nM OA on these molecules. The levels of phosphorylated PP2A/C were elevated by OA, but had no effect on PP2A/C, PP2A/A, PP1, and PP2B protein levels (Fig. 1I). Thus, optimal inhibition of pPP2A/C, but not PP1 and PP2B, is capable of maintaining hESC self-renewal, and PP2A could be a key regulator in ESC self-renewal-related signaling pathways.

Self-Renewal of hESCs by PP2A

Although AKT, ERK, c-Myc, and GSK-3b have been reported to be targets of PP2A, PP2A-mediated dephosphorylation seems to occur in a cell type-specific manner [15, 18, 19]. Consequently, we analyzed whether PP2A interacts with these proteins by detecting their presence in PP2A/C complexes immunoprecipitated from hESCs cultured with or without 1 nM OA for 3 days. All of the proteins coimmunoprecipitated untreated cells, and these associations, with the exception of ERK, were much less evident in the treated cells; these patterns correlated strongly with the levels of phosphorylated AKT, ERK, c-Myc, and GSK-3b (Fig. 2B). Overall, these results suggest that PP2A regulates selfrenewal-related signaling pathways by directly interacting with AKT, c-Myc, and GSK-3b. Because c-Myc is an important regulator of mESC self-renewal [20], we examined the expression of c-Myc in undifferentiated and differentiated hESCs. Compared with undifferentiated hESCs, c-Myc protein levels were very low, but c-Myc Thr58 phosphorylation was higher in differentiated hESCs (Fig. 2C). Compared with undifferentiated hESCs, the levels of c-Myc RNA, as well as Oct-4, Nanog, Sox2, and human telomerase reverse transcriptase (hTERT), were lower in differentiated hESCs and days 14 and 28 EBs derived from hESCs (Supporting Information Fig. 2A). Immunofluorescence analysis detected abundant cMyc protein in the nuclei of undifferentiated hESCs, and its level decreased during differentiation (Supporting Information Fig. 2B and data not shown). To test whether PP2A maintains hESC self-renewal by regulating downstream molecules, we cultured hESCs with bFGF or OA in the presence of the c-Myc inhibitor 10058F4 (F4), which belongs to a class of small molecule inhibitors that disrupt the formation of the c-Myc/Max heterodimer [26]; the PI3K inhibitor LY294002 (LY); or the MEK inhibitor U0126 (U). All three chemicals reduced AP activity (Fig. 2D) and significantly induced differentiation (Fig. 2E). In particular, bFGF þ LY enhanced differentiation more than OA þ LY as shown by morphology and AP activity (Fig. 2D). These findings may be the result of LYinduced AKT phosphorylation suppression being more effective in the presence of bFGF than OA (Fig. 2F). Thus, both bFGF and OA appear to maintain hESC self-renewal via the AKT pathway. Moreover, ERK also appears to be involved in hESC self-renewal, at least in part, because ERK inhibition induced hESC differentiation (Fig. 2D, 2E). Treatment with F4 not only increased differentiation and decreased AP activity (Fig. 2D, 2E) but increased the expression of markers of the three germ layers (Supporting Information Fig. 3). Thus, c-Myc-mediated transcriptional activation also plays an important role in the maintenance of hESC undifferentiation.

Characterization of OA-Mediated Undifferentiation Inhibition of PP2A Maintains hESC Self-Renewal by Regulating AKT, GSK-3b, and C-Myc Stability To elucidate the mechanism by which OA maintains hESCs in an undifferentiated state, we examined the effect of OA on the signaling pathways known to participate in hESC selfrenewal [8, 10]. Compared with differentiating hESCs treated with neither OA nor bFGF, hESCs treated with 1 nM OA expressed higher levels of phosphorylated AKT and GSK-3b but not phosphorylated ERK. Moreover, compared with differentiating hESCs, 1 nM OA-treated hESCs expressed a higher level of the stable form, Ser62-phosphorylated, and lower level of the Thr58-phosphorylated degradation-susceptible form, of c-Myc (Fig. 2A).

To characterize OA-mediated hESC undifferentiation, we investigated cells passaged 30–80 times. Similar to hESCs maintained with bFGF, hESCs maintained with OA expressed the pluripotency markers Nanog, SSEA-4, Oct-4, Tra-1-60, cMyc, Tra-1-81, Sox-2, and hTERT but not SSEA-1 (Fig. 3A, 3B). The hESCs maintained with OA, but not those passaged three times without OA or bFGF, exhibited telomerase activity, which is required for indefinite proliferation (Fig. 3C). Because cultured hESCs can lose genetic integrity over passages, karyotyping was performed on OA-maintained hESCs. Normal karyotypes were observed: Miz-hES, 44 þ XY; HSF6, 44 þ XX (Fig. 3D). Moreover, the OA-maintained hESCs efficiently formed EBs in suspension and differentiated into derivatives of the three germ layers, as shown by the

Yoon, Jun, Park et al.


Figure 2. Inhibition of PP2A activates AKT signaling and enhances c-Myc stability, promoting the self-renewal capability of hESCs. (A): The effect of OA and bFGF on the levels of key signaling molecules. The hESCs were cultured for 2 days in the absence of bFGF and then stimulated with bFGF or the indicated concentration of OA for 1.5 hours. Western blot analysis was performed using antibodies against the indicated proteins. (B): AKT, ERK, c-Myc, and GSK-3b were coimmunoprecipitated with PP2A/C. hESCs were cultured for 3 days with or without OA. (C): Whole cell lysates from undifferentiated hESCs and hESCs induced to differentiate spontaneously by the withdrawal of bFGF (DIFF) were immunoblotted and probed with antibodies against the indicated proteins (left). Alternatively, bFGF was withdrawn from hESC cultures and the cells subjected to Western blot analysis on days 0, 3, and 6 using the indicated antibodies (right). The results indicate that c-Myc helps maintain the self-renewal capability of hESCs. (D, E): The effect of inhibiting the AKT and ERK pathways or c-Myc/Max heterodimerization on OA- or bFGF-treated hESC self-renewal. As shown in (D), hESCs maintained with OA or bFGF were treated with or without LY294002 (10 lM; LY), U0126 (10 lM; U), or 10058-F4 (100 lM; F4) for 6 days, after which they were stained with alkaline phosphatase (AP). Scale bar ¼ 200 lm. (E): The undifferentiated colonies distinguished by their morphology and the extent of AP staining are shown. The frequency of AP-positive undifferentiated hESCs was determined by counting at least 300 colonies. Error bars indicate 6 SD; **, p < .01 between the indicated pair; n ¼ 4. (F): LY294002 and U0126 reduce pAKT and pERK1/2 levels in bFGF- and OA-maintained hESCs, respectively. The bFGF-maintained hESCs were incubated without bFGF for 2 days and then stimulated with OA or bFGF in the presence of each inhibitor (30 lM) for 1.5 hours. Western blot analysis was performed using antibodies specific for the total and phosphorylated forms of GSK-3b, AKT, and ERK. Abbreviations: bFGF, basic fibroblast growth factor; DIFF, differentiation; ERK, extracellular signal-regulated kinase; FGF, fibroblast growth factor; GSK-3b, glycogen synthase kinase-3b; hESC, human embryonic stem cell; IP, immunoprecipitation; OA, okadaic acid; pAKT, phospho AKT; pERK, phospho ERK; pGSK-3b, phospho pGSK-3b; PP2A, protein phosphatase 2A.


Self-Renewal of hESCs by PP2A

Figure 3. Characterization of OA-maintained hESCs. (A): The hESCs were maintained with OA for over 30 passages, after which they were stained with antibodies specific for the undifferentiation markers AP, Oct-4, Nanog, c-Myc, SSEA-4, TRA-1-60, TRA-1-81, and SSEA-1. (B): hESCs at the indicated passage numbers were subjected to RT-polymerase chain reaction analysis. OA-maintained hESCs did not differ from bFGFmaintained hESCs in the expression of pluripotency markers Oct-4, Nanog, Sox2, and hTERT. (C): The hESCs maintained with OA for over 30 passages exhibit telomerase activity, as determined by the telomeric repeat amplification protocol assay. A total of 10,000 cells were used for each assay. HI samples served as a control. (D): The karyotypes of OA-maintained Miz-hES4 (left panel: 44 þ XY) and HSF-6 (right panel: 44þXX) cultures at the 30th and 35th passages, respectively, were analyzed by using the G-banding method. OA-maintained hESCs were found to be genetically stable over long-term culture. (E): OA-maintained hESCs readily differentiate into EBs and ectoderm, mesoderm, and endoderm derivatives. OA was withdrawn from OA-maintained hESCs and the cultures were examined 14 days later. Phase-contrast microscopy showing EB formation (top left) and immunofluorescence images of frozen EB sections stained with antibodies against ectodermal (nestin, Pax6; lower right), mesodermal (desmin, brachuary; lower left), and endodermal (glucagon, Sox17; top right) markers. Scale bars ¼ 200 lm. Abbreviations: AP, alkaline phosphatase; bFGF, basic fibroblast growth factor; DAPI, 4,60 -diamidino-2-phenylindole; EB, embryoid body; hESC, human embryonic stem cell; HI, heat inactivated; hTERT, human telomerase reverse transcriptase; OA, okadaic acid; RT, reverse transcription; SSEA, stage specific embryonic antigen.

expression of endodermal (glucagon, Sox17, GATA4, and Foxa2), mesodermal (desmin, brachuary, and kallikrein) and ectodermal (nestin, Pax6, and keratin) markers (Fig. 3E and Supporting Information Fig. 4).

hESCs Maintained by OA 1 bFGF Retain Self-Renewal Capacity Under Feeder- and Xeno-Free Conditions The hESCs were cultured on gelatin- or CELLstart-coated plates because those coating consistently supports hESC growth. Morphological analyses showed that hESCs grown in DMEM/F12/SR or DMEM/F12/B27 medium containing a combination of 1 nM OA and 4 ng/mL bFGF maintained the undifferentiated state for over 50 passages, whereas hESCs grown in the same medium containing only 1 nM OA or 4 ng/ mL bFGF lost the undifferentiated state within 3–6 passages (Fig. 4A, 4D). The hESCs grown in the combination of 1 nM OA and 4 ng/mL bFGF had a prominent, high ratio of nucleus to cytoplasm (Fig. 4A, 4D), which is typically characteristic of undifferentiated hESCs. Moreover, hESCs grown for 20 passages in a combination of 1 nM OA and 4 ng/mL bFGF retained telomerase and AP activities and were immunopositive

for Oct-4, SSEA-4, Tra-1-60, and Tra-1-81 but negative for SSEA-1. The cells also had a normal karyotype (Fig. 4E, 4F and Supporting Information Fig. 5A, 5B). Furthermore, hESCs efficiently formed EBs in suspension upon the withdrawal of OA þ bFGF for 14 days (Fig. 4G); these EBs also expressed endoderm, mesoderm, and ectoderm-specific markers (data not shown). When the hESCs grown in the combination of 1 nM OA and 4 ng/mL bFGF for 25 passages were injected into the testes of NOD/SCID mice, a typical teratoma formed that contained multiple linage cell types, including neuroepithelium (ectoderm), adipocytes (mesoderm), and ciliated epithelium (endoderm) (Fig. 4G and data not shown). Thus, hESCs cultured under feeder-free and xeno-free conditions maintain their pluripotency in vivo. Biochemical analyses revealed that the combination of OA and bFGF kept PP2A in the inactive state similar to hESCs cultured with bFGF on the feeder layer, but OA or bFGF alone activated PP2A in feeder-free culture (Fig. 4B, 4C). The former was accompanied by an elevation in AKT and GSK-3b phosphorylation, whereas the latter was accompanied by a reduction in phosphorylation (Fig. 4B, 4C). However, there were no changes in ERK phosphorylation for either condition. These results further support the idea that hESC selfrenewal requires a low, but optimal, level of PP2A activity.

Yoon, Jun, Park et al.


Figure 4. OA and bFGF-treated hESCs maintain their self-renewal capability under feeder- and xeno-free culture conditions. (A): Bright-field images of hESCs cultured in feeder-free conditions. hESCs cultured with OA and bFGF, but not with OA or bFGF alone, remained undifferentiated for over 40 passages (P). The inset image (lower left) shows that bFGF and OA-treated hESCs have a prominent nucleus structure at P40, as indicated by a high ratio of nucleus to cytoplasm. (B): OA and bFGF-maintained hESCs were cultured for 2 days without bFGF and OA, stimulated with OA and/or bFGF for 1.5 hours, and then subjected to Western blot analysis with antibodies specific for the indicated proteins. OA and bFGF-treated hESCs, but not hESCs cultured with OA or bFGF alone, maintained PP2A and GSK-3b in an inactivated state and AKT in an activated state. (C): OA and/or bFGF was withdrawn from hESC cultures and the cells subjected on to assays measuring PP2A activity on day 7. OA and bFGF were found to maintain PP2A in an inactive state. Error bars indicate 6 SD; *, p < .05; **, p < .01 versus bFGF þ feeder control; n ¼ 3. (D): The morphology of hESCs cultured with OA and/or bFGF in xeno-free culture conditions in which SR was replaced with B27. (E): The telomeric repeat amplification protocol assay indicated that bFGF-maintained hESCs cultured on feeder layers, with OA in xenofree conditions, or under feeder-free SR-supplemented conditions continue to show telomerase activity at the 20th passage. (F): The hESCs cultured with OA and bFGF under xeno-free conditions continue to express the undifferentiation markers AP, Oct-4, SSEA-4, TRA-1-60, and TRA1-81 but not SSEA-1 at the 20th passage. (G): Passage 25 hESCs cultured with OA and bFGF under xeno-free conditions differentiate readily in vitro and in vivo. The top left image is a phase-contrast image of EB formation after withdrawing OA and bFGF for 14 days. The top right and bottom images are after the testes of nude mice were injected with hESCs cultured with OA and bFGF in xeno-free conditions. The teratomas were harvested after 7–8 weeks, sectioned, and stained with hematoxylin and eosin. Tissues derived from all three germ layers were observed, including the neuroepithelium (ectoderm: upper right), adipocytes (mesoderm: lower left), and gut epithelium (endoderm: lower right). Scale bar ¼ 200 lm (g ¼ 100 lm). (H): Proposed mechanism by which PP2A acts in hESC self-renewal. The withdrawal of bFGF or feeder cells activates PP2A, which decreases AKT activity and c-Myc stability and activates GSK-3b. This results in hESC differentiation. However, in hESCs cultured on feeders with OA or bFGF, PP2A is inactivated, which results in AKT activation, increased c-Myc stability, and the inactivation of GSK3b. This is also observed in hESCs cultured under feeder-free conditions with both OA and bFGF. The outcome is hESC self-renewal. Abbreviations: AP, alkaline phosphatase; bFGF, basic fibroblast growth factor; DAPI, 4,60 -diamidino-2-phenylindole; EB, embryoid body; ERK, extracellular signal-regulated kinase; FGF, fibroblast growth factor; GSK-3b, glycogen synthase kinase-3b; HI, heat inactivated; hESC, human embryonic stem cell; OA, okadaic acid; pAKT, phospho AKT; pERK, phospho ERK; pGSK-3b, phospho pGSK-3b; pPP2A, phospho PP2A; PP2A, protein phosphatase 2A; SR, serum replacement; SSEA, stage specific embryonic antigen.

DISCUSSION This study highlights PP2A as a previously unappreciated molecule essential for hESC self-renewal, enabling the development of defined culture conditions for hESCs. We demonstrated that PP2A is negatively regulated by bFGF, which has been known to be an upstream self-renewal factor in hESCs. Indeed, we found PP2A activity to be elevated upon concomitant differentiation with the removal of bFGF, and that its

activity was inhibited by bFGF-induced signaling with the maintenance of hESC self-renewal. There is ample evidence for the role of bFGF signaling in maintaining the pluripotency of hESCs. The first indication of such a role emerged from the prevalent use of bFGF in the media used to derive and maintain existing hESCs, suggesting that this pathway has an important function in the mechanisms regulating hESC growth and self-renewal [27]. Moreover, bFGF was recently shown to be sufficient for supporting hESC growth on Matrigel in the absence of feeders or feeder-


conditioned medium [28]. Recently, hESCs and the human fibroblasts on which they were plated have been reported to send insulin-like growth factor (IGF) and fibroblast growth factor, respectively, as reciprocal paracrine signals sufficient to maintain hESC pluripotency [29]. These findings suggest that various signals help establish a local microenvironment in vitro, and presumably in vivo, that helps maintain pluripotency. In addition, the significance of the PI3K/AKT and Wnt/GSK-3b signaling pathways, as well as c-Myc, on ESC self-renewal by the effect of factors such as bFGF and IGF have been described [8, 10, 29-31]. Critically, the inhibition of signaling using small molecule antagonists induced hESC differentiation. Finally, we and others have confirmed the important role of this pathway by demonstrating that hESC treatment with LY294002 is associated with a rapid decrease in AKT and GSK-3b phosphorylation and significantly promoted hESC differentiation in short-term culture [30]. We demonstrated that the overexpression of PP2A or treatment with a PP2A activator leads to hESC differentiation, even in the presence of bFGF. On the other hand, PP2A inactivation by PP2A siRNA or OA maintained hESC selfrenewal in the absence of bFGF. Under these culture conditions, PP2A interacts with AKT, c-Myc, and GSK-3b, resulting in the decreased phosphorylation of AKT, GSK-3b, and c-Myc and the expression of the transcription factors Oct-4, Sox2, and c-Myc. However, OA almost completely blocks the interaction of PP2A with AKT, c-Myc, and GSK-3b, resulting in the enhanced expression of Oct-4, Sox2, AKT, c-Myc, and the inactive form of GSK-3b. It is known that Oct-4 and Sox2 are critical transcription factors for ESC self-renewal. First, Oct-4 and Sox2 co-occupy the promoters or enhancers of a substantial portion of the genes required for the maintenance of hESCs [32]. In fact, the sustained expression of Oct4 and Sox2 is closely related with mESC self-renewal and pluripotency controlled by LIF/STAT3 in a c-Myc-dependent manner [20]. Furthermore, induced pluripotent stem cells can be directly generated from fibroblast cultures in the presence of c-Myc plus Oct-4, Sox2, and Klf4 under ESC culture conditions. Moreover, under these conditions, the removal of cMyc results in the emergence of colonies with flattened nonESC-like morphology [33, 34]. Recent evidence indicates that c-Myc is a direct target of PP2A regulation [35]. The levels of c-Myc are tightly regulated by phosphorylating c-MycSer62, which stabilizes c-Myc, and the subsequent phosphorylation of c-MycThr58, which is required for its degradation. The PP2A enzyme dephosphorylates c-MycSer62, targeting c-Myc for ubiquitination and subsequent degradation. Conversely, PP2A inhibition increased the levels of phosphorylated Ser62 and led to c-Myc stabilization and increased activity. Interestingly, a c-Myc mutant, c-MycT58A, that is not capable of being degraded was functionally active in the transformation of primary human cells [19]. In agreement, we found that cMyc mRNA expression is downregulated and c-MycThr58 is phosphorylated in a PP2A-dependent manner. Also, c-Myc gene expression decreased during hESC differentiation, and OA increased c-Myc activity. Moreover, a c-Myc inhibitor induced hESC differentiation in short-term culture. The Wnt proteins form a large family of signaling molecules that participate in the control of gene expression, proliferation, differentiation, and cell polarity [36]. The binding of Wnt proteins to a cognate receptor, Frizzled, results in GSK3b inhibition, which leads to the stabilization of b-catenin and subsequent changes in gene expression. The Wnt signaling pathway, which regulates Oct-4, Sox2, and Nanog, has been shown to connect directly to the core transcriptional regulatory circuitry of ESCs [37]. The c-Myc protein is also a well-established target of the Wnt pathway [38]. In ESCs,

Self-Renewal of hESCs by PP2A

Tcf3 occupies the c-Myc promoter, and Wnt3a positively contributes to its expression [37]. It has also been suggested that GSK-3b inhibition is necessary for maintaining ESC pluripotency with increased c-Myc stability [20]. Additionally, Wnt could serve to directly activate endogenous c-Myc, substituting for exogenous c-Myc [39], thereby enhancing the reprogramming efficiency. Thus, the maintenance of hESC selfrenewal requires elevated c-Myc levels, which is achieved, at least in part, by the suppression of GSK-3b activity. These results could explain the positive effect of BIO, a GSK-3b inhibitor, on the maintenance of ESC self-renewal [8]. It has been reported that PP2A directly dephosphorylates GSK-3bSer9 purified from rabbit skeletal muscle, thus reactivating GSK-3b [40]. In addition, PP2A was involved in modulating GSK-3b phosphorylation induced by KCl depolarization in SHSY5Y neuroblastoma cells. However, treatment with a PP2A inhibitor blocked the depolarization-induced dephosphorylation of GSK-3b [41]. Therefore, PP2A is believed to be involved in the activation of GSK-3b after hESC differentiation. Recent studies also indicate that PP2A can act as a positive regulator of p53 signaling. The phosphatase directly dephosphorylates p53 at either Ser37 or Thr55, resulting in increased p53 stabilization and increased apoptosis after DNA damage [42]. In agreement, the targeted inhibition of PP2A has been shown to abolish Thr55 dephosphorylation and reduce overall p53 levels [43]. Recently, it was demonstrated that p53 binds to the Nanog gene promoter, a gene required for ESC self-renewal, and suppresses Nanog expression after DNA damage. The rapid downregulation of Nanog mRNA during differentiation correlates with the induction of p53 transcriptional activity and its phosphorylation [44]. Overall, it is possible to assume a potential link between ESC self-renewal and the PP2A/p53/Nanog pathway. Common features of more recently developed hESC culture conditions include the presence of bFGF, the absence of serum, and the presence of a serum substitute [45, 46]. Several defined medium systems have been described for hESCs, including TGF-b1, Activin/Nodal [9, 12], PDGF [13], BIO [8], and IGF [31]. These results imply that alternative and/or redundant factors or pathways are involved in hESC selfrenewal. First, Activin/Nodal/TGF-b signaling pathways are essential for maintaining ESC self-renewal. For example, both Activin and TGF-b1 are expressed by mouse feeder cells, which are commonly used with hESCs [12]. Activin and Nodal would be expected to act through identical receptors and the Smad two-third signaling pathway, and TGF-b1 would act through independent receptors, Alk5 and TbR-II, but also the same Smad two-third pathway [47], indicating that Activin, Nodal, and TGF-b1 sustain pluripotency marker expression by similar mechanisms, although independent of each other. Furthermore, several lines of evidence have indicated possible links between PP2A and Activin/Nodal and/or Wnt signaling [48]. For example, it was recently demonstrated that PPP2R2D, one of the related members of PP2A/B subunits, negatively modulates the TGF-b/Activin/Nodal signaling pathways. Indeed, overexpression of PPP2R2D inhibits active levels of Smad2, whereas knocking down PPP2R2D, or using OA, enhances these responses [48]. It has also been suggested that PDGF is an essential factor for maintaining hESC pluripotency [13]. A possible link between PDGF and PP2A was demonstrated by PDGF stimulating AKT phosphorylation, predominantly at Ser-473, whereas LY294002 prevented PDGF-induced AKT phosphorylation and its downstream effector molecules, p70S6K, ribosomal protein S6, 4E-BP1, and eIF4E. However, N-ethylmaleimide (NEM), a thiol alkylating agent, activates PP2A followed by the elimination of

Yoon, Jun, Park et al.


the phosphorylation induced by PDGF. Furthermore, the inhibition of PDGF-induced AKT phosphorylation by NEM is completely reversed by PP2A inhibitors [49].

CONCLUSION In conclusion, our results provide new insight into the regulatory cascades underlying the unique hESC molecular program. We also propose a possible link between the signaling pathways of factors supplemented in hESC cultures, PP2A, and hESC self-renewal. Finally, our defined culture system in which signaling pathways are transiently activated by a PP2A inhibitor indicated clearly that the undifferentiated state is readily reversible upon withdrawal of the inhibitor. Further studies will lead to a better understanding of the dynamic interactions of PP2A within the diverse signaling cascades, leading to a more complete understanding of which holoenzyme complexes regulate hESC self-renewal.

REFERENCES 1 2 3 4 5 6 7 8

9 10


12 13 14 15

16 17

Stewart R, Stojkovic M, Lako M. Mechanisms of self-renewal in human embryonic stem cells. Eur J Cancer 2006;42:1257–1272. Thomson JA, Itskovitz-Eldor J, Shapiro SS et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature 1981;292:154–156. Williams RL, Hilton DJ, Pease S et al. Myeloid leukaemia inhibitory factor maintains the developmental potential of embryonic stem cells. Nature 1988;336:684–687. Daheron L, Opitz SL, Zaehres H et al. LIF/STAT3 signaling fails to maintain self-renewal of human embryonic stem cells. Stem Cells 2004;22:770–778. Ying QL, Nichols J, Chambers I et al. BMP induction of Id proteins suppresses differentiation and sustains embryonic stem cell selfrenewal in collaboration with STAT3. Cell 2003;115:281–292. Xu RH, Chen X, Li DS et al. BMP4 initiates human embryonic stem cell differentiation to trophoblast. Nat Biotechnol 2002;20:1261–1264. Sato N, Meijer L, Skaltsounis L et al. Maintenance of pluripotency in human and mouse embryonic stem cells through activation of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat Med 2004;10:55–63. James D, Levine AJ, Besser D et al. TGFbeta/activin/nodal signaling is necessary for the maintenance of pluripotency in human embryonic stem cells. Development 2005;132:1273–1282. Armstrong L, Hughes O, Yung S et al. The role of PI3K/AKT, MAPK/ERK and NFkappabeta signalling in the maintenance of human embryonic stem cell pluripotency and viability highlighted by transcriptional profiling and functional analysis. Hum Mol Genet 2006;15:1894–1913. Greber B, Lehrach H, Adjaye J. Fibroblast growth factor 2 modulates transforming growth factor beta signaling in mouse embryonic fibroblasts and human ESCs (hESCs) to support hESC self-renewal. Stem Cells 2007;25:455–464. Beattie GM, Lopez AD, Bucay N et al. Activin A maintains pluripotency of human embryonic stem cells in the absence of feeder layers. Stem Cells 2005;23:489–495. Pebay A, Wong RC, Pitson SM et al. Essential roles of sphingosine-1phosphate and platelet-derived growth factor in the maintenance of human embryonic stem cells. Stem Cells 2005;23:1541–1548. Janssens V, Goris J. Protein phosphatase 2A: A highly regulated family of serine/threonine phosphatases implicated in cell growth and signalling. Biochem J 2001;353:417–439. Lin C-F, Chen C-L, Chiang C-W et al. GSK-3beta acts downstream of PP2A and the PI 3-kinase-Akt pathway, and upstream of caspase-2 in ceramide-induced mitochondrial apoptosis. J Cell Sci 2007;120: 2935–2943. Sontag E, Fedorov S, Kamibayashi C et al. The interaction of SV40 small tumor antigen with protein phosphatase 2A stimulates the map kinase pathway and induces cell proliferation. Cell 1993;75:887–897. Sonoda Y, Kasahara T, Yamaguchi Y et al. Stimulation of interleukin-8 production by okadaic acid and vanadate in a human promyelocyte cell line, an HL-60 subline. Possible role of mitogen-activated

ACKNOWLEDGMENTS We thank Drs. Didier Trono and Alexander D. Verin for kindly providing the pWPXL-GFP and pShuttle-PP2A/C-Ires-hrGFP2-HA constructs. This research was supported by a grant (SC5150) from the Stem Cell Research Center of the 21st Century Frontier Research Program funded by the Ministry of Education, Science and Technology, Republic of Korea; a grant (08172KFDA527) from the Korea Food and Drug Administration, Republic of Korea; a grant from the second phase of Brain Korea 21 project, Republic of Korea; and a Korea University Grant, Republic of Korea.





The authors indicate no potential conflicts of interest.

18 19 20 21

22 23 24 25 26 27


29 30

31 32 33 34

protein kinase on the okadaic acid-induced NF-kappaB activation. J Biol Chem 1997;272:15366–15372. Zhao JJ, Gjoerup OV, Subramanian RR et al. Human mammary epithelial cell transformation through the activation of phosphatidylinositol 3-kinase. Cancer Cell 2003;3:483–495. Yeh E, Cunningham M, Arnold H et al. A signalling pathway controlling c-Myc degradation that impacts oncogenic transformation of human cells. Nat Cell Biol 2004;6:308–318. Cartwright P, McLean C, Sheppard A et al. LIF/STAT3 controls ES cell self-renewal and pluripotency by a Myc-dependent mechanism. Development 2005;132:885–896. Yoon BS, Yoo SJ, Lee JE et al. Enhanced differentiation of human embryonic stem cells into cardiomyocytes by combining hanging drop culture and 5-azacytidine treatment. Differentiation 2006;74: 149–159. Yoo SJ, Yoon BS, Kim JM et al. Efficient culture system for human embryonic stem cells using autologous human embryonic stem cellderived feeder cells. Exp Mol Med 2005;37:399–407. Ruvolo PP, Deng X, Ito T et al. Ceramide induces Bcl2 dephosphorylation via a mechanism involving mitochondrial PP2A. J Biol Chem 1999;274:20296–20300. Watanabe K, Ueno M, Kamiya D et al. A ROCK inhibitor permits survival of dissociated human embryonic stem cells. Nat Biotechnol 2007;25:681–686. Mumby M. The 3D structure of protein phosphatase 2A: New insights into a ubiquitous regulator of cell signaling. ACS Chem Biol 2007;2: 99–103. Arabi A, Wu S, Ridderstrale K et al. c-Myc associates with ribosomal DNA and activates RNA polymerase I transcription. Nat Cell Biol 2005;7:303–310. Amit M, Carpenter MK, Inokuma MS et al. Clonally derived human embryonic stem cell lines maintain pluripotency and proliferative potential for prolonged periods of culture. Dev Biol 2000;227: 271–278. Wang L, Li L, Menendez P et al. Human embryonic stem cells maintained in the absence of mouse embryonic fibroblasts or conditioned media are capable of hematopoietic development. Blood 2005;105: 4598–4603. Bendall SC, Stewart MH, Menendez P et al. IGF and FGF cooperatively establish the regulatory stem cell niche of pluripotent human cells in vitro. Nature 2007;448:1015–1021. McLean AB, D’Amour KA, Jones KL et al. Activin a efficiently specifies definitive endoderm from human embryonic stem cells only when phosphatidylinositol 3-kinase signaling is suppressed. Stem Cells 2007;25:29–38. Wang L, Schulz TC, Sherrer ES et al. Self-renewal of human embryonic stem cells requires insulin-like growth factor-1 receptor and ERBB2 receptor signaling. Blood 2007;110:4111–4119. Boyer LA, Lee TI, Cole MF et al. Core transcriptional regulatory circuitry in human embryonic stem cells. Cell 2005;122:947–956. Takahashi K, Tanabe K, Ohnuki M et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 2007; 131:861–872. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006;126:663–676.

Self-Renewal of hESCs by PP2A


35 Arnold HK, Sears RC. Protein phosphatase 2A regulatory subunit B56alpha associates with c-myc and negatively regulates c-myc accumulation. Mol Cell Biol 2006;26:2832–2844. 36 Huelsken J, Birchmeier W. New aspects of Wnt signaling pathways in higher vertebrates. Curr Opin Genet Dev 2001;11:547–553. 37 Cole MF, Johnstone SE, Newman JJ et al. Tcf3 is an integral component of the core regulatory circuitry of embryonic stem cells. Genes Dev 2008;22:746–755. 38 He TC, Sparks AB, Rago C et al. Identification of c-MYC as a target of the APC pathway. Science 1998;281:1509–1512. 39 Marson A, Foreman R, Chevalier B et al. Wnt signaling promotes reprogramming of somatic cells to pluripotency. Cell Stem Cell 2008; 3:132–135. 40 Sutherland C, Cohen P. The alpha-isoform of glycogen synthase kinase-3 from rabbit skeletal muscle is inactivated by p70 S6 kinase or MAP kinase-activated protein kinase-1 in vitro. FEBS Lett 1994;338: 37–42. 41 Lee YI, Seo M, Kim Y et al. Membrane depolarization induces the undulating phosphorylation/dephosphorylation of glycogen synthase kinase 3beta, and this dephosphorylation involves protein phosphatases 2A and 2B in SH-SY5Y human neuroblastoma cells. J Biol Chem 2005;280:22044–22052.

42 Li HH, Cai X, Shouse GP et al. A specific PP2A regulatory subunit, B56gamma, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J 2007;26:402–411. 43 Manning BD, Cantley LC. AKT/PKB signaling: Navigating downstream. Cell 2007;129:1261–1274. 44 Lin T, Chao C, Saito S et al. p53 induces differentiation of mouse embryonic stem cells by suppressing Nanog expression. Nat Cell Biol 2005;7:165–171. 45 Xu C, Rosler E, Jiang J et al. Basic fibroblast growth factor supports undifferentiated human embryonic stem cell growth without conditioned medium. Stem Cells 2005;23:315–323. 46 Levenstein ME, Ludwig TE, Xu RH et al. Basic fibroblast growth factor support of human embryonic stem cell self-renewal. Stem Cells 2006;24:568–574. 47 Miyazawa K, Shinozaki M, Hara T et al. Two major Smad pathways in TGF-beta superfamily signalling. Genes Cells 2002;7:1191–1204. 48 Batut J, Schmierer B, Cao J et al. Two highly related regulatory subunits of PP2A exert opposite effects on TGF-{beta}/Activin/Nodal signalling. Development 2008;135:2927–2937. 49 Yellaturu CR, Bhanoori M, Neeli I et al. N-Ethylmaleimide inhibits platelet-derived growth factor BB-stimulated Akt phosphorylation via activation of protein phosphatase 2A. J Biol Chem 2002;277:40148–40155.

See for supporting information available online.