Characterization and neural differentiation of mouse embryonic and induced pluripotent stem cells on cadherin-based substrata

Biomaterials 33 (2012) 5094e5106

Contents lists available at SciVerse ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Characterization and neural differentiation of mouse embryonic and induced pluripotent stem cells on cadherin-based substrata Amranul Haque a, Xiao-Shan Yue b, Ali Motazedian a, Yoh-ichi Tagawa a, Toshihiro Akaike c, * a

Department of Biomolecular Engineering, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan b Department of Chemistry & Biochemistry, University of Notre Dame, Notre Dame, IN 46556, USA c Frontier Research Center, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 23 February 2012 Accepted 1 April 2012 Available online 19 April 2012

A suitable culture condition using advanced biomaterials has the potential to improve stem cell differentiation into selective lineages. In this study, we evaluated the effects of recombinant extracellular matrix (ECM) components on the mouse embryonic stem (mES) and induced pluripotent stem (miPS) cells’ self-renewal and differentiation into neural progenitors, comparing conventional culture substrata. The recombinant ECMs were established by immobilizing two chimera proteins of cadherin molecules, Ecadherin-Fc and N-cadherin-Fc, either alone or in combination. We report that the completely homogeneous population of mES and miPS cells could be maintained on E-cadherin-based substrata under feeder- and serum-free culture conditions to initiate neural differentiation. Using defined monolayer differentiation conditions on E-cadherin and N-cadherin (E-/N-cad-Fc) hybrid substratum, we routinely obtained highly homogeneous population of primitive ectoderm and neural progenitor cells. Moreover, the differentiated cells with higher expression of bIII-tubulin, Pax6, and tyrosine hydroxylase (TH) in absence of GFAP (a glial cell marker) expression suggesting the presence of a lineage restricted to neural cells. Our improved culture method should provide a homogeneous microenvironment for differentiation and obviate the need for protocols based on stromal feeders or embryoid bodies. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: ES/iPS cells Monolayer differentiation Cadherins Extracellular matrix (ECM) Neuroectoderm

1. Introduction The embryonic stem (ES) cells and induced pluripotent stem (iPS) cells represent a promising source to overcome many human diseases by providing an unlimited supply of differentiated cells, including cells with neural characteristics [1e3]. During embryogenesis in vivo, the neurons arise from the neuroectodermal precursors [4]. Efficient production of these ectoderm progenitors would allow on-demand production of neurons with different subtypes [5,6]. Efforts have been devoted to producing defined lineages of neural cells from ES and iPS cells. Most of the protocols for neural differentiation of ES cells relied on the formation of cluster of cells, so-called embryoid bodies (EBs) [7,8] or neurospheres [9], to initiate differentiation. Although initial studies on the generation of defined lineages of neural cells seemed encouraging, later work revealed that the neuronal clusters derived from ES and iPS cells contain a variety of neuronal subtypes as well as non-neuronal

* Corresponding author. Tel.: þ81 45 924 5790; fax: þ81 45 924 5815. E-mail address: [email protected] (T. Akaike). 0142-9612/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2012.04.003

lineages, including undifferentiated cells [7,10]. Since then, progress has been made to enrich neural subtypes with special emphasis on inductive signals [11,12], transcription factors [6], and cell adhesion molecules (CAMs) for extracellular matrix [7,13]. Among them, cell adhesion molecules could act as obvious candidates for guiding differentiation into uniform and defined lineages. The regulation of stem cell behavior and the formation of appropriate neural circuits depend on a complex interplay between extracellular guiding cues and intracellular signaling [14]. Two members of cadherin superfamily, E-cadherin and N-cadherin, are most-studied CAMs involved in the ES cell pluripotency and neural development, respectively [15,16]. Recent findings showed the important effect of E-cadherin to maintain pluripotency [17], reduce cellular heterogeneity [18,19], and improve iPS cell generation [20]. Moreover, N-cadherin acts as an important regulator of nervous system development by providing important molecular cues in many biological processes such as retina development, somite formation, and neurite outgrowth [15,21]. Both of these cadherins are differentially expressed, depending on developmental stage and cell type. Mouse (m) ES and iPS cells express high level of E-cadherin, which can act as pluripotent marker [17]. On the other hand, the neural differentiation of mES cells is associated

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

with an E- to N-cadherin switch, up-regulation of E-cadherin repressor molecules, and increased cellular motility [22]. The differential expression of these cadherin molecules can play a role in cell sorting in heterogeneous population of cells once engineered in EBs [23] or immobilized on extracellular substratum [24,25]. In our laboratory, we have derived enriched population of neural cells by directed differentiation of P19 embryonic carcinoma cells and neural stem cells using a chimera protein of N-cadherin as extracellular matrix [13]. To overcome the controversy of using carcinoma cell line and to increase the yield of differentiation we induced feeder-dependent mES and miPS cells into neuroectoderm progenitors. Both mES and miPS cells on cadherin-based artificial substratum showed better pluripotency compared to cells cultured on natural substrata. The homogenous population of undifferentiated mES and miPS cells made them ideal to generate neural progenitors under completely defined culture condition. Finally, we ensured an efficient monolayer differentiation condition by synergistic application of two chimera proteins, E-cadherin-Fc and N-cadherin-Fc, as artificial extracellular substrata.

5095

2.2. Preparation of natural and artificial substrata To prepare gelatinized surfaces, tissue culture dishes were treated with 0.1% (w/v) gelatin for 30 min at 37  C. Expression and purification of E-cadherin-Fc (E-cad-Fc) and N-cadherin-Fc (N-cad-Fc) fusion proteins and immobilization of these fusion proteins onto tissue culture polystyrene (PS) dishes have been described in detail elsewhere [13,25]. Briefly, to prepare E-cad-Fc and N-cad-Fc-coated surface, the purified fusion proteins were diluted into10 mg/ml, and the diluted solutions were separately added to non-treated PS plates and incubated at 37  C for 1 h. To prepare the E-cad-Fc and N-cad-Fc co-immobilized matrix, the optimized concentration of Ecad-Fc (5 mg/ml) and N-cad-Fc (5 mg/ml) was added to non-treated PS dish and incubated at 37  C for 1 h. The PS surface after immobilization was washed with PBS once and incubated with 0.25% BSA/PBS solution at 37  C for 2 h to inhibit nonspecific attachment of cells to the surface. 2.3. Cell adhesion and proliferation assay

2. Materials and methods

The attachment and proliferation of mES and miPS cells to various extracellular matrix (ECM) components were measured by MTT assay. Briefly, 24-well microplates were coated with each adhesion molecule at 37  C for 1 h. Cells were seeded at confluent density (1  104 cells per well) on adhesion molecule-coated plates for different time points in ES or iPS cell medium. After 4 h, medium and non-adhered cells were removed, and washed with fresh DMEM basal medium, and cultured in presence of fresh undifferentiating medium. 5 mg/ml MTT solution was added to each well and incubated for 4 h at 37  C. 0.5 ml of DMSO was added after removal of media and incubated for 5 min at 37  C. Absorbance was measured in a micro plate reader at 570 nm with a reference wavelength of 630 nm.

2.1. Cell culture

2.4. Alkaline phosphatase activity

The feeder-dependent mES cell line (ST1) and Nanog-GFP expressing miPS cell line (APS0001 iPS-MEF-Ng-20D-17) [26] were routinely cultured on mouse embryonic fibroblast (MEF) cells in 35 mm culture dish coated with gelatin in a humidified atmosphere of 5% CO2 at 37  C. ST1 cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM; SigmaeAldrich), supplemented with 20% (v/v) fetal bovine serum (FBS), 1 mM sodium pyruvate (nacalai tesque), 1 mM L-glutamine (Millipore), 1% nonessential amino acids (NEAA; Gibco, Invitrogen), 0.1 mM 2mercaptoethanol (Sigma Chemical), 1000 units/ml recombinant leukemia inhibitory factor (LIF; Chemicon). The miPS cells were cultured in medium supplemented with DMEM (high glucose without sodium pyruvate; Sigma); 15% FBS, 0.1 mM NEAA, 0.1 mM 2-mercaptoethanol, and 1000 units/ml LIF. All media contained 50 mg/ml penicillin and 50 mg/ml streptomycin (nacalai tesque). mES and miPS cells were passaged every third day with daily media change. The mouse embryonic carcinoma cell line (P19) and feeder-independent mES cells (EB3 cell line) were used as control in cell adhesion and mRNA expression analyses. The culture conditions of P19 and EB3 cells were described previously [13,19].

The alkaline phosphatase (AP) activity of mES and miPS cells cultured in presence of undifferentiating medium for 4 days on adhesion molecule-coated 12-well plates was detected according to manufacturer’s instruction (Sigma, Leukocyte Alkaline Phosphatase Kit, 85L3R). 2.5. Flow cytometry The cultured miPS cells were harvested with Accutase and analyzed. The dissociated 1 106 cells/ml were resuspended in cold PBS and centrifuged to remove enzyme. The Nanog-expressing cells were then analyzed using flow cytometer (Guava Technologies, Millipore). 2.6. Induction of differentiation The basal differentiation medium was identical to that described above for ES and iPS cell culture medium except that Glasgow minimum essential medium

Table 1 List of primer sequences used in this study. Genes analyzed Primitive Nanog Neural N-cadherin Sox1 Sox2 Nestin Ngn1 MAP2 BLBP bIII-tubulin GFAP Pax6 TH Mesendoderm Brachyury Goosecoid Endoderm Foxa2 Sox17 Gata6 Mesoderm Gata1 House keeping b-actin

Forward primer (50 e30 )

Reverse primer (50 e30 )

Annealing temp. ( C)

GAGGAAGCATCGAATTCTGG0

AAGTTATGGAGCGGAGCAGC0

58

CAGTCTTACCGAAGGATGTGC CCTCGGATCTCTGGTCAAGT GAACGCCTTCATGGTATGG GCTACATACAGGACTCTGCTG CGATCCCCTTTTCTCCTTTC TCAGACTTCCACCGAGCAG GGGTAAGACCCGAGTTCCTC AGCGATGAGCACGGCATAG GGAGAGGGACAACTTTGCAC TGCCCTTCCATCTTTGCTTG TCCTGCACTCCCGCTCAGAG

ATCAGCTCTCGATCCAGAGG TACAGAGCCGGCAGTCATAC AGCCGTTCATGTAGGTCTGC AAACTCTAGACTCACTGGATTCT TGCAGCAACCTAACAAGTGG AGGGGAAAGATCATGGCCC ATCACCACTTTGCCACCTTC CAGGTTCCAAGTCCACCAGA GCTCTAGGGACTCGTTCGTG TCTGCCCGTTCAACATCCTTAG CCAAGAGCAGCCCATCAAAGG

58 58 55 55 55 55 58 55 55 58 58

ATGCCAAAGAAAGAAACGAC ATGCTGCCCTACATGAACGT

AGAGGCTGTAGAACAGGATT CAGTCCTGGGCCTGTACATT

55 55

TATTGGCTGCAGCTAAGCGG TTTGTGTATAAGCCCGAGATGG ACCTTATGGCGTAGAAATGCTGAGGGTG

GACTCGGACTCAGGTGAGGT AAGATTGAGAAAACACGCATGAC CTGAATACTTGAGGTCACTGTTCTCGGG

55 55 60

CACCATCAGGTTCCACAGG

TTGAGGCAGGGTAGAGTGC

55

CCTAAGGCCAACCGTGAAAAG

TCTTCATGGTGCTAGGAGCCA

55

5096

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

Fig. 1. The effect of cadherin-based engineered extracellular matrix (ECM) components on mouse (m) ES and iPS cell pluripotency. (A) Culturing of mES and miPS cells on the different ECM components led to variations in cell morphologies and colony shapes. Feeder-dependent ES and iPS cells on 0.018% type I collagen formed spherical colonies, while cells on 5 mg/ml fibronectin or 0.1% gelatin spread out like differentiated cells. Cells on 10 mg/ml E-cad-Fc matrix formed single cells with scattering morphology. (B) The adhesion of mES cells on E-cad-Fc is integrin-independent. Cells were seeded on gelatin and E-cad-Fc immobilized substrata for 16 h in presence of 15% KSR or 15% FBS. The expression of total FAK (tFAK) and phosphorylated FAK (pFAK) in mES cells was analyzed by Western blotting using monoclonal anti-FAK and monoclonal anti-pFAK antibodies. (C) Adhesion of miPS

A. Haque et al. / Biomaterials 33 (2012) 5094e5106 (GMEM; SigmaeAldrich), stage specific differentiation induction factors, and 10% (v/ v) knockout serum replacement (KSR; Invitrogen) were added, while DMEM, LIF, and FBS were omitted. Before the induction of differentiation, mES and miPS cells were cultured on 10 mg/ml E-cad-Fc-coated plates to remove feeder cells. For monolayer differentiation, mES and miPS cells were seeded onto E-cad-Fc- and Ncad-Fc-coated surfaces or onto the surface co-immobilized with E-cad-Fc and N-cadFc. Cells were cultured for 24 h in undifferentiating ES or iPS cell media before induction to differentiation. The neural differentiation using monolayer protocol was induced in KSR differentiation medium supplemented with 10 ng/ml Dkk-1 and 500 ng/ml Lefty-A for five days [27]. From days 6e12, cells were culture in basal differentiation medium containing basic fibroblast growth factor (bFGF, 20 ng/ml; Promega). Medium was changed every two days during this process. Growth and changes in morphology were monitored daily. Differentiation was confirmed by monitoring the axon formation, RT-PCR method, and immunofluorescence staining. Spontaneous differentiation in embryoid bodies (EBs) was carried out using the hanging drop method. Briefly, cells on E-cad-Fc-coated dishes were dissociated with Accutase and diluted in undifferentiating ES and iPS cell medium in absence of LIF; 20 ml drops containing 600 cells were placed on the inside of a PS petri dish lid. On day 3 and day 5, five hanging drops containing embryoid bodies were transferred to each 35 mm 0.1% gelatin-coated dish and cultured for one more day in absence of LIF. The EBs were then collected in day 4 and 6 for mRNA expression analysis. For neural differentiation using hanging drop technique, same protocol was used except neural differentiation medium was used from first to fifth days of differentiation in hanging drop. The EBs were then transferred to gelatin-coated dishes and cultured for five more days in presence of bFGF. Growth and changes in morphology were monitored daily. At different differentiation time points, cells were collected for the analysis of stage specific markers. 2.7. Immunofluorescence Cells were fixed with Mildform 20 N (8% formaldehyde) for 15 min and permeabilized with 0.2% Triton X-100 (nacalai tesque) for 5 min. Fixed cells were blocked with Blocking one solution (nacalai tesque) for 1 h. The primary antibodies used include mouse anti-E-cadherin (BD Transduction Laboratories), rabbit antimouse N-cadherin (H-63, Santa Cruz Biotechnology), anti-mouse SSEA1 (Santa Cruz Biotechnology), rabbit anti-human Oct3/4 (Santa Cruz Biotechnology), rabbit anti-human Nestin (IBL Ltd. Japan), mouse anti-neuron specific bIII-tubulin antibody (TuJ-1; R&D Systems, Inc.), and mouse anti-GFAP (GA5; Cell Signaling). The secondary antibodies used include goat anti-mouse IgG F(ab’)2-TRITC (Santa Cruz), anti-rabbit IgG F(ab’)2 Alexa Fluor 555 conjugate (Cell Signaling), anti-mouse Cy3 (Fab’)2 secondary antibody conjugated with Alexa Fluorophore (Invitrogen), and anti-mouse IgG F(ab’)2 Alexa Fluor 488 conjugate (Invitrogen). 2.8. Reverse transcriptase polymerase chain reaction (RT-PCR) Total RNA was extracted using Trizol reagent (Invitrogen). RNA was reversetranscribed into cDNA with an oligo-dT primer using Moloney murine leukemia virus (M-MLV) reverse transcriptase (Invitrogen). PCR was performed with ExTaq polymerase (Takara) in PCR buffer (Qiagen) containing 0.2 mM dNTPs (Takara). Primers and PCR conditions used are listed in Table 1. The amount of mRNA for each marker was deduced from the fluorescent signal of PCR products using ImageQuant image analysis software (Version 5.2, Molecular Dynamics). 2.9. Western-blot analyses The total cellular protein was extracted with lysis buffer (10 mM TriseHCl, 150 mM NaCl, 1% Nonidet P-40, 10 mM EDTA, and protease inhibitor cocktail; PH 7.4), and cell lysates were centrifuged at 15,000  g for 15 min at 4  C. Samples were separated by electrophoresis on 7.5% polyacrylamide gels and electrophoretically transferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore). The primary antibodies were as follows: mouse anti-E-cadherin (BD Transduction Laboratories), rabbit anti-human N-cadherin (Santa Cruz Biotechnology), mouse anti-FAK (BD Transduction Laboratories), mouse anti-pFAK (BD Transduction Laboratories), and mouse anti-b-actin (Sigma). The membranes were then reacted with horseradish peroxidase (HRP)-conjugated secondary antibody (1: 10,000 dilution; Jackson ImmunoResearch Laboratories) for 1 h. HRP activity was assayed using Immobilon Western detection reagents (Millipore) according to the manufacturer’s instruction.

5097

2.10. Statistical analyses Data are presented as the mean  standard deviations (SD). Statistical analyses were performed with Student’s t-test for paired samples. A pevalue < 0.05 was considered statistically significant.

3. Results 3.1. Pluripotency of feeder-dependent mouse ES and iPS cells First, we confirmed the effect of E-cadherin substratum on the pluripotency of feeder-dependent mES (ST1) and miPS cells prior to the induction of neural differentiation. Culturing of mES and miPS cells on different ECM components led to variations in cell morphologies and shapes (Fig. 1A). Both ES and iPS cells on type I collagen formed compact spherical colonies which were comparable to typical undifferentiated cells cultured on MEF feeder layer (Suppl Fig. S1a, S1b). In contrast, undifferentiated cells on fibronectin or gelatin formed less compact colonies (Fig. 1A). The cells on E-cadherin substratum formed single cells with scattering morphology. To address whether the adhesion mechanism of ES and iPS cells on feeder-free artificial ECM is integrin-independent, we determined the state of integrin activation. The phosphorylation of FAK stimulated by gelatin and E-cad-Fc ECM components was examined by Western blotting (Fig. 1B). On gelatin, the level of phosphorylation at Tyr-397 of FAK was more prominent than in mES cells cultured on E-cadherin substratum. The phosphorylation of FAK was slightly higher in serum-containing medium which might be due to the possible presence of extracellular matrix molecules (e.g., fibronectin) in FBS. These results suggest that the initial adhesion of ES cells to E-cadherin matrix is not dependent on integrin. The faint band of activated FAK on Ecadherin substratum might be due to the mechanical stress in stretching cells generated from attaching to E-cadherin substratum [28]. In addition, the adhesion efficiency of miPS cells on E-cadherin substratum was not affected even in serum-free conditions, suggesting the preferential application of cadherinbased substratum for self-renewal and differentiation in serumfree culture conditions (Fig. 1C). On the basis of these results, we hypothesized that ES and iPS cells might assume better states of pluripotency depending on the E-cadherin-based extracellular components. We measured the proportion of ES and iPS cell colonies positive for alkaline phosphatase (AP) activity, which is widely used as an undifferentiated marker. High proportions of positive colonies were found on type I collagen, but a low proportion of positive colonies were found in cells cultured on fibronectin or gelatin (Fig. 1D). Interestingly, higher proportions of scattered single cells on E-cadherin substratum showed AP activity. Flow cytometric analysis of Nanog-GFP protein expression by miPS cells cultured for four days yielded results similar to those of AP staining with higher percentages of Nanog-expressing cells on Ecadherin (w77%) and collagen (w71%) substratum, but low proportion of positive cells was found on gelatin (w67%) and fibronectin (w54%) (Fig. 1E). Immunocytochemical analysis of stagespecific embryonic antigen 1 (SSEA1) protein expression, which is used as a marker

cells in absence of FBS. iPS cells adhered to E-cad-Fc-coated dishes with higher efficiency as compared to 0.018% type I collagen-coated dishes after 6 h of incubation in 15% KSR containing undifferentiating medium. (D) Phase-contrast microphotograph of mES and miPS cells cultured for four days on various substrata. Cells were stained for alkaline phosphatase (AP) activity. (E) Flow cytometric profile of Nanog protein expression in miPS cells cultured for four days on four different ECM molecules. (F) Immunocytochemical analysis of SSEA1 protein expression in mES (i-iv) and miPS (v-viii) cells cultured on 0.1% gelatin- and 10 mg/ml E-cad-Fc-coated dishes for two days. (G) Markers for pluripotency (Oct3/4), ectoderm (Sox1), mesendoderm (Gsc, Bra), endoderm (Foxa2, Sox17, Gata6), mesoderm (Gata1) in spontaneously differentiated feeder-independent mES cells (EB3) and feeder-dependent iPS cells were analyzed by RT-PCR. Bar 50 mm. Abbreviations: Goosecoid, Gsc; Brachyury, Bra; EB3 cell-derived embryoid bodies, ES-EB; miPS cell-derived embryoid bodies, iPS-EB.

5098

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

Fig. 2. Expression pattern of cell surface E-cadherin and N-cadherin proteins in undifferentiated cells. mES (ieiv) and miPS (veviii) cells were cultured for two days in presence of LIF and assessed for total cells (DAPI, blue), E-cadherin (red) and N-cadherin (red) protein expression using immunofluorescent microscopy. P19 cells were used as control (ixexii). Bar 50 mm. (B) Western blot was performed to assess total cellular E-cadherin or N-cadherin proteins in mES cells differentiated for 4 and 10 days hanging drop protocol. mES cells were used as control.

of undifferentiated mES cells, yielded results similar to those of AP staining and Nanog expression on gelatin and E-cad-Fc (Fig. 1F). The above data indicates that undifferentiated state of feeder-dependent ES and iPS cells can be maintained for prolong period of time on cadherin-based substratum. We also showed that miPS cells can maintain their pluripotent state after repeated passage on E-cad-Fc surfaces. miPS cells on MEF at passage 13 were cultured on E-cad-Fc matrix and maintained in presence of undifferentiating medium for 7 more passage. These iPS cells at passage 20 were induced to form embryoid bodies (EBs) using handging-drop technique for three days in absence of LIF and then transferred to gelatin-coated surfaces. The cells on gelatin were cultured for additional two days and then lineage specific markers were checked to confirm the capability of these cells to undergo spontaneous differentiation, a strategy to assess the pluripotency. The iPS-derived EBs in absence of LIF expressed markers for ectoderm (Sox1), mesendoderm (Goosecoid and Brachyury), endoderm (Sox17, Foxa2, and Gata6), and mesoderm (Gata1) with reduced expression of undifferentiated iPS cell marker (Oct3/4) (Fig. 1G), suggesting that miPS cells can be maintained on E-cad-Fc without compromising their pluripotent capability to differentiate into all three germ layers.

3.2. Expression of E-cadherin and N-cadherin in ES and iPS cells It has been reported that neural differentiation of mES cells is associated with an E- to N-cadherin switch [22]. To establish cadherin-based engineered ECM, we evaluated the expression pattern of E-cadherin and N-cadherin during the differentiation of mES and miPS cells. Expression of E-cadherin was assessed in undifferentiated mES and miPS cells by immunofluorescence staining (Fig. 2A). The majority of cells expressed only E-cadherin at the cell surface. In contrast, undifferentiated mES and miPS cells lacked cell surface N-cadherin expression. The E-cadherin and Ncadherin expressing P19 cell line was used as positive control. To monitor E- to N-cadherin switch, we induced neural differentiation in presence of Dkk-1 and Lefty-A by hanging drop technique (explained in method section). Total E- and N-cadherin protein was assessed by Western blot analysis of whole cell lysates (Fig. 2B). Ncadherin protein was absent from undifferentiated cells and was detected within 4 days after differentiation, with increased levels observed up to day 10 of differentiation. In contrast, E-cadherin protein was detected in undifferentiated cells but was substantially down-regulated in presence of neural differentiation media. The expression of E-cadherin and N-cadherin were overlapped between 4 and 6 days of differentiation. Considering the role of E-cadherin

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

for adhesion of undifferentiated mES and miPS and N-cadherin for neural differentiation, we introduced two chimera proteins of cadherin superfamily (E-cadherin-Fc and N-cadherin-Fc) either alone or in combination to induce homogeneous population of neural progenitors from mES and miPS cells. 3.3. E-cadherin and N-cadherin co-immobilized matrix We first optimized the concentration of E-cad-Fc (Suppl. Fig. S2a) and N-cad-Fc (Suppl. Fig. S2b) fusion proteins to immobilize on PS surface using ELISA. The adsorption of individual fusion protein to PS surface increased in a dose-dependent manner reaching to a monolayer concentration at 10 mg/ml. We also confirmed the adhesion ratio of ES cells on E-cad-Fc matrix in a dose-dependent manner and found that 5 mg/ml immobilized Ecad-Fc can adhere a significant proportion of ES cells (about 85%) to matrix (data not shown). We estimated the monolayer concentration of E-cad-Fc and N-cad-Fc mixture keeping E-cad-Fc concentration fixed in 5 mg/ml. As we expected, 5 mg/ml N-cad-Fc together with 5 mg/ml E-cad-Fc (abbreviated E-/N-cad-Fc) was suitable for

5099

co-immobilization for mES and miPS cells’ initial adhesion and induction into neural differentiation (Fig. 3A). 3.4. Adhesion and growth on co-immobilized substratum To confirm the efficacy of chimera protein-based ECMs (E-cadFc and N-cad-Fc), we monitored adhesion capability of ES and iPS cells on these recombinant ECMs under undifferentiating and differentiating culture conditions. Both ES and iPS cells adhered to the 10 mg/ml E-cad-Fc-coated surface with the same efficiency as to the conventional gelatin- and collagen-coated surfaces (Fig. 3B and C). However, they detached from the surface on the way of differentiation (Suppl. Fig. S3a). On the other hand, N-cad-Fc-coated surface did not support the adhesion of undifferentiated cells (Fig. 3B). However, differentiated neural cells in neurosphere adhered onto N-cad-Fc-coated surface with neurite outgrowth (Suppl. Fig. S3b, S3c). The expression of neurite outgrowth was confirmed by bIII-tubulin expression. Therefore, we applied the surface co-immobilized with N-cad-Fc and E-cad-Fc to culture and differentiate mES and miPS cells into neural lineages. We checked

Fig. 3. mES and miPS cell adhesion and morphology on the fusion-protein immobilized surfaces. (A) The co-adsorption of E-cad-Fc and N-cad-Fc on polystyrene surface was detected by ELISA, as described in materials and methods. The data indicates means  SD of experiments (n ¼ 3). Adhesion of mES (B) and miPS (C) cells on various extracellular matrix components. ES cells (ST1) adhered to E-cad-Fc-coated dishes with equivalent efficiency as to 0.1% gelatin-, 0.018% type I collagen-coated dishes after 6 h of incubation. Both ES and iPS cells showed almost similar adhesion efficiency (w85%) on E-/N-cad-Fc co-immobilized substratum. (D) Morphological observation of ES and iPS cells on the two different matrices. ES cells were cultured on polystyrene surfaces coated with 0.1% gelatin, 10 mg/ml N-cad-Fc, and (5 mg/ml E-cad-Fc þ 5 mg/ml N-cad-Fc) E-/N-cad-Fc fusion proteins. P19 cells were used as control. Bar 50 mm.

5100

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

the adhesion capability of mES and miPS cells to co-immobilized ECM (E-/N-cad-Fc). The plating efficiency of ES and iPS cells on E-/N-cad-Fc was about 80% (Fig. 3B and C) and all cells showed singlecell scattering morphology with growth and proliferation (Fig. 3D). On the conventional gelatin-coated surface, both of these cell lines formed aggregated colonies. As we introduced N-cadherin as a component of ECM in neural differentiation protocol, the effect of immobilized N-cadherin matrix to induce neural differentiation at an early stage of differentiation was evaluated. We checked the expression of undifferentiated (Oct3/4 and E-cadherin) and neural differentiation (N-cadherin and nestin) specific markers for mES and miPS cells. Collagen I-coated matrix was used as control for conventional extracellular matrix. In presence of LIF, the undifferentiated ES cells and iPS cells did not express N-cadherin or nestin on E-/N-cad-Fc matrix within 2 days of differentiation, while almost all cells showed expression of pluripotent markers, Oct3/4, E-cadherin (Fig. 4A and B), and Nanog (Suppl. Fig. S4). 3.5. Differentiation of ES and iPS cells into neural precursors The induction of neural differentiation using monolayer culture condition is illustrated in Fig. 5A. We mainly focused on the generation of neurons and for this purpose mES and miPS cells were culture for 12 days in presence of neural differentiation medium. The viability and proliferative ability of these cells cultured on E-/N-cad-Fc-coated surfaces in presence of neural differentiation medium was assessed by MTT assay. The growth curve of mES cells (Fig. 5B) and miPS cells (Fig. 5C) on E-/N-cad-Fc-

coated surfaces showed increased proliferation over those on gelatin-, collagen-, and fibronectin-coated dishes suggesting the possibility to generate a large number of differentiated cells on engineered artificial matrix. In addition, the exposure of mES and miPS cells to neural differentiation medium induced distinguishable changes in morphology of scattered single mES (Fig. 5D) and miPS cells (Suppl. Fig. 5) on E-/N-cad-Fc substratum. Using cadherin-based ECM, it was possible to maintain homogeneous population of adherent cells throughout all stages of differentiation with visible neurite outgrowth within 10 days of differentiation. To confirm the progression of neural differentiation, we checked stage specific markers for primitive ectoderm, primitive neural stem cells, neural stem cells, neural progenitor cells, and finally cells with phenotypic and genotypic characteristics of neurons and glial cells. Nanog and Oct3/4-expressing mES and miPS cells were induced into primitive ectoderm-like cells, which expressed brain lipid binding protein (BLBP). The mRNA transcript level of Sox2 was elevated with decreasing level of Nanog expression within two days of differentiation (Fig. 5E). The differentiated cells at this stage expressed low level of N-cadherin and neurogenin1 (Ngn1). These cells were then induced into primitive neural stem cells (NSCs), characterized by the expression of moderate levels of intermediate filament protein nestin and N-cadherin, and a rounded morphology within four days of differentiation (Fig. 6A and B). The cells with primitive NSC-like morphology adopt a distinct, spindle-shape morphology reminiscent of the shape of radial-glial cells with elevated expression of BLBP and Pax6 (Fig. 5, D and Fig. 6C). The transcript level of both BLBP is initially expressed by most cells but

Fig. 4. The effect of immobilized N-cadherin chimera on the undifferentiated state of mES and iPS cells. Immunocytochemical analysis showing the presence of undifferentiated ES cell markers (Oct3/4 and E-cadherin) and absence of neural progenitor cell markers (N-cadheirn and nestin) in mES (i-iv) and iPS (v-viii) cells cultured for two days on E-/N-cad-Fc matrix in presence of LIF. Bar 50 mm.

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

5101

Fig. 5. Neural differentiation of mES and miPS cells can be achieved under homogeneous culture condition on E-/N-cad-Fc. (A) Schematic representation of the strategy of differentiation of mES and miPS cells into neural cells. Proliferation of mES (B) and miPS (C) cells cultured on various ECM components. Cells were seeded in 24-well dish at 1  104 cells per well on each ECM component in neural differentiating medium supplemented with Dkk-1 and Lefty-A. The data indicate means  SD (n ¼ 3). *:p < 0.05 for ES cells cultured on E-cad-Fc-coated plates versus collagen-coated plate. (D) Bright field microscopic image shows the morphological changes of mES and miPS cells on 0.1% gelatinized or 10 mg/ml E-/N-cad-Fc-coated surfaces. The prominent morphological changes were observed within four days of differentiation with radial-glial cell-like phenotype. By 10 days of differentiation, the characteristic neurite outgrowth was visible. Cells on gelatin showed heterogeneous population with neurite outgrowth from the cluster of cells. Bar 50 mm. (E) Semiquantitative RT-PCR analysis for selected markers of pluripotency (Nanog) and lineage commitment (BLBP, N-cad, Sox2, and Ngn1) in second days of differentiation; mRNA from undifferentiated ES cells was used as control. The expression level was normalized using house-keeping gene, b-actin. Abbreviations: days of differentiation, d; brain lipid binding protein, BLBP; N-cadherin, N-cad; neurogenin1, Ngn1.

5102

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

Fig. 6. Defining the primitive neural stem cells using immunocytochemistry and transcript analysis. (A) Immunofluorescent images shows differentiated mES cells cultured for four days on surfaces coated with E-/N-cad-Fc and gelatin have low expression of N-cadherin, moderate expression of nestin, and no expression of bIII-tubulin. Bar 50 mm (B) miPS cells cultured for four days in presence of differentiation medium expressed reduced expression of pluripotent markers (Oct3/4 and Nanog), low expression of N-cadherin, and moderate expression of nestin. (C) Semi-quantitative RT-PCR analysis for selected markers of neural progenitors (BLBP, N-cad, Sox2, Ngn1, Pax6) in four days of differentiation; mRNA from spontaneously differentiated ES cells was used as control. The expression level was normalized using house-keeping gene, b-actin.

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

rapidly disappears within six days of differentiation (Fig. 7A). Moreover, the expression level of N-cadherin, Ngn1, Sox2 was upregulated compared to primitive ectoderm-like cells. All these results suggest that the switch from ES identity (Oct3/4þ, Nanogþ,

5103

Sox2þ, Nestin, Ngn1) to that of neuroectoderm progenitor cells (N-cadþ, Nestinþ, Ngn1þ, BLBP) seems to be complete by day six [7]. For comparative study, we checked time dependent analysis of nestin expression in mES cells (Fig. 7B and C) and miPS cells (Fig. 7B

Fig. 7. Differentiation of mES and miPS cells into neural progenitor cells. (A) Transcript expression of BLBP, N-cad, Sox2, Ngn1, and Pax6 was determined by RT-PCR in 6 days of differentiated ES and iPS cells and the fluorescent intensity of amplified product was quantified using ImageQuant software. mRNA from spontaneously differentiated ES cells was used as control. (B) Marker for neural progenitor cells (nestin) was used for transcription analysis in mES and miPS cells induced for 6 days on gelatin and E-/N-cad-Fc substrata. Nestin gene expression in differentiated mES (C) and miPS (D) cells was quantified using ImageQuant software by measuring the fluorescent intensity of amplified bands. The expression level was normalized using house-keeping gene, b-actin. (E) Immunofluorescent images shows differentiated mES (i-vi) and iPS (vii-xii) cells cultured for four days on surfaces coated with E-/N-cad-Fc have higher expression of N-cadherin and nestin compared to differentiated cells on gelatin (xiii-xviii) substratum. Differentiated ES cells on E-/Ncad-Fc matrix showed low level of bIII-tubulin (v, vi). iPS cells in the same stage of differentiation lack expression of Naong (xi, xii), though some cells in colony escaped differentiation and expressed pluripotent marker, Nanog (xvii, xviii). Bar 50 mm.

5104

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

and D) on gelatin and E-/N-cad-Fc. In contrast to cells on gelatincoated substratum, the level of expression of mRNA transcript for nestin was higher in cells on E-/N-cad-Fc throughout all stages of neural progenitor cell differentiation. Moreover, immunofluorescence images showed similar results of nestin and N-cadherin expression in cells cultured on E-/N-cadherin- and gelatin-coated substrata (Fig. 7E). In contrast, markers for endoderm (hepatocyte nuclear factor 4a, HNF4a) and mesoderm (Gata1) lineages were undetectable in six days of differentiated cells (data not shown). We monitored the timedependent down-regulation of Nanog protein expression in miPS cell-derived neural progenitor cells to determine if there was preferential differentiation to the neural lineage. On E-/N-cadherin hybrid substratum, the expression of Nonog was down-regulated dramatically in homogeneous population of adherent cells. The expression of Nanog was undetectable within six days in differentiated miPS cells (Fig. 7E). In contrast, the expression of Nanog was detectable in cellular aggregates on gelatin-coated matrix supporting the findings by Ying et al. that the loss of undifferentiated ES cell marker is asynchronous within the cluster of cells and some clusters can escape differentiation and retain undifferentiated through all stages of neural differentiation. Similar phenomenon was also monitored by generating moderate homogeneity of miPS cells on E-/N-cad-Fc substratum. The expression of Nanog gene was prominent in regions with compact cellular aggregation, which was clearly distinguishable from monolayer of non-aggregated cells with neural progenitor cell-like morphology suggesting the suitability of homogenous culture conditions to eliminate contaminated undifferentiated cells (Fig. 8). 3.6. Neural differentiation mES and miPS cell-derived neural progenitors were cultured in presence of bFGF to induce cells with neural characteristics. Induction of neural progenitors into neural cells was assessed in two different ways. First, we checked the changes in morphology upto 12 days of differentiation. Second, the markers of neural cells such as microtubule associated protein 2 (MAP2), tyrosine hydroxylase (TH), bIII-tubulin (Tuj), and glial cell marker (GFAP) were analyzed using RT-PCR. Cells with neuronal morphology began to appear within eight days of differentiation and were

prominent within 12 days (Fig. 5D). To test whether neural progenitor cells undergo neuron-producing and later gliaproducing progenitors, we quantified the production of neurons and glia throughout the monolayer differentiation protocol. We found that the number of bIII-tubulinþ neurons is increases up to day 12, while the GFAP expressing glial cells were undetectable supporting the notion that neurogenesis is an earlier event than gliogenesis (Fig. 9A). In addition, we induced mES and miPS cells under low cell density culture condition (explained in Fig. 8). Both of these cell lines were induced into bIII-tubulin expressing neural cells with comparatively shorter neurites (Fig. 9B) than differentiated cells under high cell density. The higher expression of MAP2, Pax6 and TH transcript level with lack of GFAP transcript suggests predominate presence of neural subtypes rather than glial cells (Fig. 9C) within 10 days of differentiation. In contrast to gelatincoated matrix, the cells on cadherin-based substratum formed extended neurite outgrowth supporting the findings that cellecell contact or cluster of cells are not necessary for extended neurite outgrowth [14]. 4. Discussion Derivation of neuroectodermal progenitors from ES and iPS cells is necessary toward generation of neural circuits. In this study, we used completely homogeneous population of pluripotent mES and miPS cells on our cadherin-based substratum to initiate differentiation. The neuroectodermal progenitors were induced in presence of antagonists to Wnt (Dkk-1) and Nodal (Lefty-A) signaling using adherent monolayer culture condition. This protocol does not require formation of EBs, but culturing of cells in a monolayer in absence of cellular aggregates (Fig. 4A). Maintaining the complete homogeneity of pluripotent ES and iPS cells is pre-requisite to achieve efficient differentiation. Prior to induce neural differentiation, we determined the effects of ECM components on the self-renewal of mES cell and miPS cells under defined conditions (Fig. 1). Collagen (type I), gelatin, and recombinant ECM molecules were evaluated for long-term maintenance of mES and miPS cells while retaining high pluripotency and functionality. Interestingly, the cells cultured on E-cad-Fc showed better pluripotency than cells on type I collagen, gelatin, or fibronectin,

Fig. 8. The loss of Nanog is asynchronous within the heterogeneous culture due to cellular aggregates. Two different culture conditions (low cell density and high cell density) were induced on E-/N-cad-Fc substratum. iPS cells in high cell density (initial seeding density: 2  104 cells in 35 mm culture dish) formed cellular aggregates in which some clusters of cells (region 1 and 2) escaped differentiation and retained Nanog expression even in 6 days of differentiation (i-iv). iPS cells in low cell density (initial seeding density: 5  103 cells in 60 mm culture dish) on cadherin-based co-immobilized substratum formed homogeneous culture condition with complete absence of Nanog expression (v, vi). Bar 50 mm.

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

5105

Fig. 9. Differentiation of neural progenitor cells into bIII-tubulin expressing neural cells. mES and miPS cell-derived neural cells were stained with bIII-tubulin (red) after differentiation induction for 12 days. Two different culture conditions (low cell density and high cell density) were used. (A) mES cells in high cell density (initial seeding density: 2  104 cells in 35 mm culture dish) showed confluent growth with elongated neurites covering all the surfaces coated with E-/N-cad-Fc. Cells on gelatin formed elongated neurites from the cluster. Bar 200 mm (B) mES and miPS cells in low cell density (initial seeding density: 5  103 cells in 60 mm culture dish) on gelatin and cadherin-based co-immobilized substratum expressed bIII-tubulin. Bar 50 mm. (C) Transcript expression for Pax6, MAP2, TH, and GFAP was determined by RT-PCR in 8 and 10 days of differentiation. The expression level was normalized using house-keeping gene, b-actin. Abbreviation: microtubule associated protein 2, MAP2; tyrosine hydroxylase, TH, glial fibrillary acidic protein, GFAP; undifferentiated, UD.

supporting the findings that E-cadherin can modulate STAT3 activation, a down-stream target molecule for LIF mediated signal transduction pathway (data not shown) [29]. However, low proportion of miPS cells on all of these matrices showed no expression of Nanog-GFP expression as determined by FACS analysis (Fig. 1E). We have also succeeded in maintaining undifferentiated miPS cells on E-cad-Fc for more than 7 subcultures (Fig. 1G). Furthermore, on chimera protein of cadherin molecules the higher proliferation and lower dependency on serum-containing molecules was a step-forward to develop feeder-/serum-free monolayer differentiation protocol for neural cells (Fig. 5B and C). Although the EB-derived neural differentiation protocols are comparatively easy and takes relatively short period of time, the variability of the embryoid body (EB) size and the asynchronous distribution of morphogens to reach the innermost layers of the EBs affected the yield of differentiation. In addition, being an aggregate of many cells, EBs also present difficulties in monitoring cell morphology during differentiation [30]. These issues can be overcome when ES cells are differentiated in an adherent culture system wherein cells are differentiated as a monolayer in absence of

cellular aggregates and expose uniformly to morphogens. Considering these advantageous features, an initial attempt had taken by Ying et al. to generate neuroectoderm progenitors from mES cells using monolayer differentiation system [11]. Using this monolayer protocol, one can routinely obtain an enriched population of neural progenitors after 4e6 days in culture, although the presence of contaminating undifferentiated ES cells is still observed, probably due to cluster of cells which seems unavoidable on natural extracellular matrices such as gelatin and laminin. In this work, we report that the use of a new cadherin-based substratum allows more efficient production of neural progenitor cells from ES and iPS cells in monolayer culture might be due to highly homogeneous population of neural progenitors and less contamination with nonneuronal cells (Figs. 7E and 8). Using this substratum it was also possible to visualize the cellular behavior and cells after four to six days of differentiation showed characteristic radial-glial cell-like morphology with expression of BLBP (Fig. 5D) [7]. The cells with radial-glial phenotype quickly lost their typical spindle-shape morphology within eight days of differentiation. These neural progenitor cells had potentiality to generate neurons

5106

A. Haque et al. / Biomaterials 33 (2012) 5094e5106

with elongated neurites and prominent cell bodies [4]. Over time, almost all cells formed neurite networks covering all the surfaces of cadherin-coated dish (Figs. 5D and 9A). By contrast, on gelatincoated dish long elongated neurite was migrated from the cluster of cells, which made the cellular condition heterogeneous. The cells on E-/N-cad-Fc substratum showed expression of transcript for neuron specific genes with absence of GFAP (marker for glial cells) supporting the previous findings on the use of Dkk-1 and Lefty-A for induction into neural progenitor cells [27]. In addition, using our artificial ECM it was possible to monitor the cellular behavior in lowcell density and highcell density. Our high-celldensity culture system with homogeneity should be suitable for derivation of large number of differentiated cells for preclinical investigations. On the other hand, lowcell-density culture system may allow the identification of growth factors that regulate the proliferation and differentiation of neural cells, and it may therefore be useful to gain a better understanding of the mechanism of nerve regeneration. Additional evidence will be necessary to prove this hypothesis. 5. Conclusion The results reported here describe an improved ES and iPS cell culture and differentiation strategy using cadherin-based extracellular matrix in order to increase the yield of neuronal precursors with possible elimination of non-neuronal cells. This opens the possibility of using undifferentiated cells with complete homogeneity prior to differentiation. Furthermore, the absence of cellular aggregates should allow us to monitor the cellular behavior and to conduct electrophysiological properties of stem cell-derived neurons. Acknowledgment This work was supported by Grants-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan; and Global-COE (Center of Excellence). Appendix A. Supplementary material Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.biomaterials.2012.04. 003. References [1] Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature 1981;292:154e6. [2] Martin GR. Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci U S A 1981;78:7634e8. [3] Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006;126: 663e76. [4] Bertrand N, Castro DS, Guillemot FS. Proneural genes and the specification of neural cell types. Nat Rev Neurosci 2002;3:517e30.

[5] Erceg S, Ronaghi M, Stojkovic M. Human embryonic stem cell differentiation toward regional specific neural precursors. Stem Cells 2009;27:78e87. [6] Chung S, Sonntag KC, Andersson T, Bjorklund LM, Park JJ, Kim DW, et al. Genetic engineering of mouse embryonic stem cells by Nurr1 enhances differentiation and maturation into dopaminergic neurons. Eur J Neurosci 2002;16:1829e38. [7] Bibel M, Richter J, Schrenk K, Tucker KL, Staiger V, Korte M, et al. Differentiation of mouse embryonic stem cells into a defined neuronal lineage. Nat Neurosci 2004;7:1003e9. [8] Bain G, Kitchens D, Yao M, Huettner JE, Gottlieb DI. Embryonic stem cells express neuronal properties in vitro. Dev Biol 1995;168:342e57. [9] Eiraku M, Sasai Y. Mouse embryonic stem cell culture for generation of threedimensional retinal and cortical tissues. Nat Protoc 2011;7:69e79. [10] Li M, Pevny L, Lovell-Badge R, Smith A. Generation of purified neural precursors from embryonic stem cells by lineage selection. Curr Biol 1998;8: 971e4. [11] Ying QL, Stavridis M, Griffiths D, Li M, Smith A. Conversion of embryonic stem cells into neuroectodermal precursors in adherent monoculture. Nat Biotechnol 2003;21(2):183e6. [12] Kawasaki H, Mizuseki K, Nishikawa S, Kaneko S, Kuwana Y, Nakanishi S, et al. Induction of midbrain dopaminergic neurons from ES cells by stromal cellderived inducing activity. Neuron 2000;28:31e40. [13] Yue XS, Murakami Y, Tamai T, Nagaoka M, Cho CS, Ito Y, et al. A fusion protein N-cadherin-Fc as an artificial extracellular matrix surface for maintenance of stem cell features. Biomaterials 2010;31:5287e96. [14] Cavallaro U, Dejana E. Adhesion molecule signalling: not always a sticky business. Nat Rev Mol Cell Biol 2011;12:189e97. [15] Miyatani S, Shimamura K, Hatta M, Nagafuchi A, Nose A, Matsunaga M, et al. Neural cadherin: role in selective cell-cell adhesion. Science 1989;245:631e5. [16] Halbleib JM, Nelson WJ. Cadherins in development: cell adhesion, sorting, and tissue morphogenesis. Genes Dev 2006;20:3199e214. [17] Redmer T, Diecke S, Grigoryan T, Quiroga-Negreira A, Birchmeier W, Besser D. E-cadherin is crucial for embryonic stem cell pluripotency and can replace OCT4 during somatic cell reprogramming. EMBO Rep 2011;12:720e6. [18] Nagaoka M, Hagiwara Y, Takemura K, Murakami Y, jixuan L, Stephen A, et al. Design of the artificial acellular feeder layer for the efficient propagation of mouse embryonic stem cells. J Biol Chem 2008;283:26468e76. [19] Haque A, Hexig B, Meng Q, Hossain S, Nagaoka M, Akaike T. The effect of recombinant E-cadherin substratum on the differentiation of endodermderived hepatocyte-like cells from embryonic stem cells. Biomaterials; 2011:322032e42. [20] Chen T, Yuan D, Wei B, Jiang J, Kang J, Ling K, et al. E-cadherin-mediated cellecell contact is critical for induced pluripotent stem cell generation. Stem Cells 2010;28:1315e25. [21] Hansen SM, Berezin V, Bock E. Signaling mechanisms of neurite outgrowth induced by the cell adhesion molecules NCAM and N-cadherin. Cell Mol Life Sci 2008;65:3809e21. [22] Spencer HL, Eastham AM, Merry CL, Southgate TD, Perez-Campo F, Soncin F, et al. E-cadherin inhibits cell surface localization of the pro-migratory 5T4 oncofetal antigen in mouse embryonic stem cells. Mol Biol Cell 2007;18: 2838e51. [23] Bratt-Leal AM, Carpenedo RL, McDevitt TC. Engineering the embryoid body microenvironment to direct embryonic stem cell differentiation. Biotechnol Prog 2009;25:43e51. [24] Nose A, Nagafuchi A, Takeichi M. Expressed recombinant cadherins mediate cell sorting in model systems. Cell 1988;23(54):993e1001. [25] Nagaoka M, Koshimizu U, Yuasa S, Hattori F, Chen H, Tanaka T, et al. E-cadherin-coated plates maintain pluripotent ES cells without colony formation. PloS One 2006;1. e15. [26] Okita K, Ichisaka T, Yamanaka S. Generation of germline-competent induced pluripotent stem cells. Nature 2007;448:313e7. [27] Osakada F, Ikeda H, Sasai Y, Takahashi M. Stepwise differentiation of pluripotent stem cells into retinal cells. Nat Protoc 2009;4(6):811e24. [28] Yano Y, Geibel J, Sumpio BE. Tyrosine phosphorylation of pp125FAK and paxillin in aortic endothelial cells induced by mechanical strain. Am J Physiol 1996;271:C635e49. [29] Arulanandam R, Geletu M, Feracci H, Raptis L. Activated Rac1 requires gp130 for Stat3 activation, cell proliferation and migration. Exp Cell Res 2010;316: 875e86. [30] Shin S, Dalton S, Stice SL. Human motor neuron differentiation from human embryonic stem cells. Stem Cells Dev 2005;14:266e9.