Glycogen synthase kinase 3b inhibitor- CHIR 99021 augments the differentiation potential of mesenchymal stem cells

Kavitha Govarthanan1, Prasanna Vidyasekar1, Piyush Kumar Gupta1, Nibedita Lenka2, Rama Shanker Verma1,*

1 Stem Cell and Molecular Biology Lab, Bhupat and Jyoti Mehta School of Biosciences, Department of Biotechnology, Indian Institute of Technology Madras, Chen-nai, Tamilnadu, India

2 National Centre for Cell Science, Pune, Maharashtra, India

ARTICLE INFO

Article History:

Received 20 September 2019 Accepted 11 December 2019 Available online xxx

Keywords:

CHIR 99021

differentiation

mesenchymal stem cells

multipotency

regenerative therapy

ABSTRACT

Aim: Mesenchymal stem cells (MSCs) are immunomodulatory, non-teratogenic and multipotent alternatives to embryonic or induced pluripotent stem cells (ESCs or iPSCs). However, the potency of MSCs is not equiva-lent to the pluripotency of ESCs or iPSCs. We used CHIR 99021 to improve current protocols and methods of differentiation for the enhanced transdifferentiation potency of MSCs.

Main Methods: We used Flurescence activated cell sorter (FACS) for MSC immunophenotyping and biochemi-cal assay for demonstrating the trilineage potential of MSCs. We used real-time polymerase chain reaction, immunocytochemistry and Western blotting assay for analyzing the expression of lineage-specific markers.

Key Findings: CHIR 99021 treatment of MSCs resulted in enhanced transdifferentiation into neurological, hep-atogenic and cardiomyocyte lineages with standardized protocols of differentiation. CHIR 99021 treated MSCs showed increased nuclear localization of b-catenin. These MSCs showed a significantly increased depo-sition of active histone marks (H3K4Me3, H3K36Me3), whereas no change was observed in repressive marks (H3K9Me3, H3K27Me3). Differential methylation profiling showed demethylation of the transcription factor OCT4 promoter region with subsequent analysis revealing increased gene expression and protein content. The HLA-DR antigen was absent in CHIR 99021 treated MSCs and their differentiated cell types, indicating their immune-privileged status. Karyotyping analysis showed that CHIR 99021 treated MSCs were genomi-cally stable. Teratoma analysis of nude mice injected with CHIR 99021 treated MSCs showed the increased presence of cell types of mesodermal origin at the site of injection.

Significance: MSCs pretreated with CHIR 99021 can be potent, abundant alternative sources of stem cells with enhanced differentiation capabilities that are well suited to cell-based regenerative therapy.
© 2019 International Society for Cell and Gene Therapy. Published by Elsevier Inc. All rights reserved.

Introduction

Diseases or trauma that involve progressive degeneration or damage to tissue that results in loss of function of the organ can be potentially treated with regenerative therapy. Pathology like cardiovascular disease [1], diabetic nephropathy [2] or neural stroke [3] involves progressive, irreversible damage of cells that perform specific functions. Regenerative therapy using stem cells can be used to treat such pathologies by repo-pulating the damaged organ with the required functioning cell type.

Pluripotent cells such as embryonic or induced stem cells can become most adult somatic cells and regenerate damaged organs or tissue [4,5], but several ethical concerns and issues with immune

* Correspondence: Prof. Rama Shanker Verma, Stem Cell and Molecular Biology Laboratory, Bhupat and Jyoti Mehta School of Bioscience, Department of Biotechnol-ogy, IIT-Madras, Chennai- 600036, Tamilnadu, India.
E-mail address: [email protected] (R.S. Verma).

compatibility between donor and recipient have limited their appli-cation in humans [6,7]. Induced pluripotent cell (iPSC) technology uses retroviral-based vector systems to express the pluripotency fac-tors that reprogram the somatic cell to a state of pluripotency [8,9]. This has been observed to induce genomic instability from random genome integration, copy number variation and introduction of poly-adenylation signal in the promoter elements [10]. Several studies have raised concerns about the suitability of iPSCs generated using retroviral vector systems for certain clinical trials [11,12]. Mesenchy-mal stem cells (MSCs) are non-hematopoietic, multipotent cells that reside in most adult organs and are hypothesized to function in tissue homeostasis [13]. The immature nature, ability to self-renew and dif-ferentiation potential of MSCs make them ideal candidates for cell-based regenerative therapy [14 16]. However, the potency of human MSCs is not equivalent to the pluripotency of embryonic stem cells (ESCs) or iPSCs, and, therefore, their applicability in regenerative medicine and therapy is limited. The differentiation potential of

https://doi.org/10.1016/j.jcyt.2019.12.007

1465-3249/© 2019 International Society for Cell and Gene Therapy. Published by Elsevier Inc. All rights reserved.

ARTICLE IN PRESS

2 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

MSCs is also limited by current protocols and methods of lineage differentiation [16]. Manipulation of MSCs to enhance transdifferen-tiation potential and improve differentiation protocols and pharma-cological cues of induction will augment their clinical utility.

CHIR 99021 is a small molecule inhibitor that acts as an agonist of Wnt signaling by inhibiting the activity of GSK3b and maintaining the Wnt pathway in an active state. Several studies demonstrated that CHIR 99021 promotes self-renewal and improves efficiency of ESC isolation in both mouse and human ESCs [17 22]. In MSCs, CHIR 99021 has been shown to aid the induction of transdifferentiation into the hepatic lineage in vitro [23]. We hypothesized that CHIR 99021 treatment of MSCs could enhance potency or transdifferentia-tion into other lineages, in a similar manner.

In this study, we have tested the efficacy of CHIR 99021 as an enhancer of transdifferentiation of Wharton’s jelly (WJ) derived MSCs into hepatic, neuronal and cardiac lineages. WJ is a prolific source of highly plastic MSCs, and, by comparison with established pharmacological differentiation protocols using native (untreated) WJ-MSCs, we determined the efficacy of CHIR 99021 in aiding differ-entiation. We also examined and compared the differential methyla-tion status of CHIR 99021 treated WJ-MSCs with native MSCs to identify potential candidate genes that aid in this process.

Materials and Method

Collection and processing of human umbilical cord tissue

Human umbilical cords from cesarean delivery were aseptically col-lected in the sterile transport medium maintained at 4°C with prior con-sent from Seethapathy Hospital, Chennai, India. The collected cords were transported to the laboratory in medium consisting of 1x Dulbecco’s Phosphate Buffered Saline A (DPBSA- without Ca+2 and Mg+2 ions) containing 1% glucose, 1x antibiotics and anti-mycotic solution (Gibco, Thermo Fisher Scientific, Massachusetts, USA). All tissue specimens were processed within 30 min of collection. The blood remains in the cord tis-sue was removed by washing thrice with 1x DPBSA with 1x anti-mycotic and antibiotic solution, followed by a single wash with the serum-free a-Minimum essential medium, MEM medium (LONZA, Bassel, Switzerland). The cord tissue was then cut into small pieces followed by dissection of arteries and vein out from the tissues under the aseptic con-dition in the serum-free a-MEM medium. The human umbilical cord tis-sue (hUCT) was chopped mechanically into small fine pieces and kept for overnight enzymatic digestion (16 h) with Type 1 collagenase (1 mg/mL) (Worthington Biochemical Corporation, New Jersey, USA) in 1 mL of serum-free a-MEM at 37°C [24]. The finely chopped tissues were digested for 16 h at 37°C in a 5% CO2 incubator, and the resulting cell suspension was filtered through a cell strainer having 100-mm pore size (BD Biosciences, California, USA). The suspension was further washed twice with complete media by centrifuging at 800 g without break and acceleration. Cells were seeded at a density of 1 £ 107 nucleated cells/mL in a 25-cm2 flask with complete medium, which consists of a-MEM with 1X Non-Essential Amino acids (Gibco), 1x sodium pyruvate (Gibco), 1x L-Glutamine (Gibco) and 10% fetal bovine serum (FBS; South Ameri-can origin, Gibco, Massachusetts, USA). Cultures were incubated at 37°C in a humidified CO2 incubator (Forma Steri-Cycle CO2 incubator, Thermo Scientific, USA). The culture medium was changed at every 2-day inter-val. At 80% confluency, cells were harvested using accutase enzyme (Gibco, Thermo Fisher Scientific, MA, USA) and propagated at a ratio of 1:3 until fifth passage (P5) for the further experimental purpose.

Immunophenotyping

WJ-MSCs at P3 were harvested and washed twice with 1x DPBS. Cells were fixed with 2% paraformaldehyde for 40 min and permeabi-lized with 0.25% Triton x 100 for 20 min, followed by two times wash-ing with DPBS. Cells were further blocked with 5% normal goat serum

for 90 min and then stained with primary antibodies (Santa Cruz Biotechnology, CA, USA) targeting CD44, CD90, CD105, CD34 and CD45 cell surface markers for 16 h at 4°C. After incubation, cells were washed again with 1x DPBS and stained with appropriate secondary antibody conjugated to anti-rabbit Alexa Fluor 595 (Invitrogen, MA, USA) and anti-mouse Alexa Fluor 488 (Invitrogen), respectively, along with Hoechst 33342 staining (Sigma Aldrich, Merck, USA) for 2 h at room temperature (RT) under dark condition. For isotype control, mouse immunoglobulin (Ig)G1, rabbit IgG and goat IgG were used as per the heavy chain of primary antibody origin. The immunophenotyping was carried out with Fluorescent activated cell sorter (FACS) (BD FACSCanto II, BD Bioscience, CA, USA), and the data was analyzed using BD CellQuest Pro software (BD Bioscience, CA, USA). For HLA typing, CHIR-treated MSCs and their representative derivatives were stained with HLA-DR-FITC (Fluorescein isothiocyanate) conjugate and analyzed with their corresponding isotype control (Biolegend, USA).

Tri-lineage differentiation of hMSCs

P3 MSCs were induced with adipocyte induction medium com-posed of MEM medium supplemented with 20% FBS, 1mmol/L dexa-methasone, 0.5 mmol/L isobutyl-methyl xanthine (IBMX), 100 ng/mL insulin and 60 mmol/L indomethacin (Sigma Aldrich, Missouri, USA). Cells were incubated for the next 28 days at 37°C in a 5% humidified CO2 incubator (Forma Steri-Cycle CO2 incubator, Thermo Scientific, USA) by changing the media twice in a week. The adipocyte differen-tiation was confirmed with oil red O staining (Sigma Aldrich, Mis-souri, USA) and quantitatively analyzed using semi-quantitative real time PCR (qRT-PCR) using Applied Biosystems 7500 Real-Time PCR instrument (Applied Biosystems, Thermo Fisher Scientific, MA, USA).

For chondrocyte induction, MSCs (P3) were cultured in chondro-cyte differentiation medium (Gibco, Thermo Fisher Scientific, USA) as a pellet culture system and incubated for the next 28 days at 37°C in

a 5% humidified CO2 incubator (Forma Steri-Cycle CO2 incubator, Thermo Scientific, USA) by changing the media every third day. The chondrocyte differentiation was studied using toluidine blue staining (Sigma Aldrich, Merck, USA) and quantitatively analyzed using qPCR using the Applied Biosystems 7500 Real-Time PCR instrument (Applied Biosystems, Thermo Fisher Scientific, MA, USA).
MSCs at P3 were induced with osteogenic induction medium com-posed of a-MEM medium supplemented with 10% FBS, 100 nmol/L dexamethasone (Sigma Aldrich, Missouri, USA), 10 mmol/L b-glycerol phosphate (Sigma Aldrich, Missouri, USA) and 50 mmol/L ascorbate-2-phosphate (Sigma Aldrich, Missouri, USA). Mineralization was studied using alizarin red S staining (Himedia, Bangalore, India) after 21 days [25], and osteoblast-specific genes were quantified using qPCR using the Applied Biosystems 7500 Real-Time PCR instrument (Applied Biosystems, Thermo Fisher Scientific, MA, USA) for analyzing osteoblast differentiation.

Pretreatment of WJ-MSCs with CHIR 99021

WJ-MSCs were pretreated with Medium-1 (PTM-1) for 5 days and incubated at 37°C in a 5% humidified CO2 incubator (Forma Steri-Cycle CO2 incubator; Thermo Scientific, USA). PTM-1 consists of high-glucose Dulbecco’s Modified Eagle’s Medium supplemented with 5% FBS, 10% knockout serum replacement (KOSR), 1x insulin-transferrin-sodium selenite, 50 mg/mL L-ascorbic acid 2-phosphate, 0.1 mmol/L b-mercap-toethanol, 2 mmol/L Glutamax, 1% non-essential amino acids and 10 mmol/L CHIR 99021 (Gibco, Thermo Fisher Scientific, MA, USA and Sigma Aldrich, Missouri, USA) and the medium was changed once after 2 days. Furthermore, the cells were then maintained in Rosewell park memorial institute (RPMI)-1640 medium, which consists of 2% B27 supplement without insulin, 2 mmol/L Glutamax, 400 mmol/L 1-thioglycerol, 12 mmol/L CHIR 99021, 10 ng/mL bone morphogenetic protein-4, 20 ng/mL activin A and 10 ng/mL vascular endothelial

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 3

growth factor (Gibco, Thermo Fisher Scientific, MA, USA and Sigma Aldrich, Missouri, USA) for 2 days. After 2 days, the CHIR 99021 pretreated cells were subjected to hepatocyte, neuronal and cardiac lineage differentiation induction, and the efficacy of differentia-tion was analyzed further. For Wnt inhibitor studies, 10 mmol/L CHIR 99021 was replaced with 3 mmol/L C59 (R&D Systems, MN, USA) in the medium, and a similar protocol was followed, as mentioned above.

Enhanced potency analysis of CHIR 99021 pretreated WJ-MSCs

Neuronal lineage differentiation

For neuronal lineage induction, the CHIR 99021 pretreated cells were induced with neuron induction medium composed of Dulbecco’s Modified Eagle’s Medium with nutrient mixture F12 containing 10% Knock out serum replacement (Gibco, Thermo Fisher Scientific, USA), 10 mmol/L All trans retinoic acid (Sigma Aldrich, Missouri, USA), 10 mmol/L Forskolin (Stemcell Technologies, Canada) and 100 mmol/L isobutyl-methyl xanthine (Sigma Aldrich, Missouri, USA) and incubated for up to 20 days at 37°C in a 5% humidified CO2 incubator (Forma Steri-Cycle CO2 incubator, Thermo Scientific, USA). Untreated MSCs were also subjected to similar differentiation protocols to evaluate the efficacy of the differen-tiation exhibited by CHIR 99021 treated MSCs. The medium was replen-ished every third day till 20 days, and the neuronal differentiation was assessed by comparing the expression of neuron-specific genes between CHIR 99021 treated MSCs and native untreated MSCs using semi-quantitative real time PCR and immunocytochemistry (ICC) analysis.

Myocyte lineage differentiation

For the induction of myocyte lineage, the CHIR 99021 pretreated cells were induced into myocyte-like cells using a PSC cardiomyocyte dif-ferentiation kit (Thermo Fisher Scientific, MA, USA). The differentiation assay was performed according to the manufacturer’s recommended protocol. Untreated MSCs were also subjected to similar differentiation protocols to evaluate the efficacy of the differentiation exhibited by CHIR 99021 treated MSCs. The muscle-specific markers were analyzed and compared between CHIR 99021 pretreated MSCs and untreated native MSCs using qRT-PCR and ICC after 30 days of induction.

Endoderm-hepatocyte lineage differentiation

For hepatocyte induction, the CHIR 99021 pretreated cells were cultured in a basal medium comprising Rose well park memorial insti-tute-1640 medium supplemented with B27 supplements, 100 U/mL penicillin and 100 mg/mL streptomycin (Gibco, Thermo Fisher Scien-tific, MA, USA) as described earlier [26]. The endoderm induction medium was prepared by adding 100 ng/mL activin A, and 3 mmol/L CHIR 99021in basal medium, and cells were incubated in this medium from the second to the fourth day of post-induction. The hepatocyte induction medium was prepared with 5 ng/mL basic fibroblast growth factor, 20 ng/mL BMP4 and 0.5% dimethyl sulfoxide (DMSO) and cells were incubated again from the fifth to ninth day of post-induction. Fur-thermore, cells were cultured in immature hepatocyte medium consist-ing of basal medium supplemented with 20 ng/mL hepatocyte growth factor and 0.5% DMSO from the 10th to the 14th day of post-induction. Finally, cells were cultured in mature hepatocyte medium comprising basal medium supplemented with 20 ng/mL hepatocyte-specific growth factor, 20 ng/mL oncostatin M, 100 nmol/L dexamethasone and 0.5% DMSO and cells were incubated for the next 12 days. Untreated MSCs were also subjected to similar differentiation protocols to evalu-ate the efficacy of the differentiation exhibited by CHIR 99021 treated MSCs. The hepatocyte-specific markers were analyzed and compared between CHIR 99021 pretreated MSCs and untreated native MSCs using qRT-PCR and ICC after 28 days of induction.

Periodic Acid Schiff staining

CHIR 99021 pretreated, followed by hepatocyte-induced cells, were stained biochemically for the glycogen storage potential using

the Periodic Acid Schiff (PAS) staining method. After 28 days of induc-tion, cells were washed with DPBS and fixed in 2% paraformaldehyde solution for 40 min. Furthermore, the fixed cells were incubated with 0.5% periodic acid staining solution (Hi-Media, Bangalore, India) for 5 min. Then, cells were washed again with DPBS and again incubated in Schiff’s staining solution (Hi-Media, Bangalore, India) for 15 min. For counterstaining, hematoxylin solution was used. Finally, stained cells were washed three times with DPBS and observed under the phase-contrast microscope at 10X magnification.

Albumin assay

The spent media from WJ-MSCs (untreated, differentiation induced) and CHIR 99021 pretreated hepatogenic lineage induced MSCs were collected on day 0 and on the 5th, 10th, 15th and 20th days and used for albumin (ALB) quantification using a mouse ALB enzyme-linked immunosorbent assay ELISA kit (catalog number ab108791; Abcam, Cambridge, UK) as per the manufacturer’s proto-col. Untreated cells were used as a negative control for the assay.

Urea assay

The spent media from WJ-MSCs (untreated, differentiation induced) and CHIR 99021 pretreated hepatogenic lineage induced MSCs were collected on day 0 and the 5th, 10th, 15th and 20th days and used for the estimation of urea concentration using a urea assay kit (catalog num-ber ab83362; Abcam, Cambridge, UK) according to the manufacturers protocol. Untreated cells were used as a negative control for the assay.

Differential DNA methylation analysis 180k DNA microarray

Genomic DNA (gDNA) was isolated from native MSCs, and CHIR 99021 treated MSCs as per the instruction by Zymo quick gDNA extraction kit (ZYMO Corporation, CA, USA). gDNA with OD260/OD280

> 1.8 and OD260/OD230 1.3 was used for 180k DNA microarray experiments. The isolated, purified DNA was sonicated to obtain the DNA fragments of 100 800 base pair (bp) size using bioruptor (Dia-genode). The sheared DNA was incubated with antibody-bound beads 5-methyl cytosine (Eurogenetec BI-Mecy-1000), and control Pan Mouse IgG Dyna1 beads (catalog number 110.41) in the presence of immunoprecipitation (IP) buffer. The IP reaction was set up by the gentle stirring of tubes overnight at 4°C on the tube rotator. After overnight incubation, the bound beads were washed with wash buffer, and immune-precipitated DNA was eluted in the elution buffer. The input DNA and IP samples were labeled using the Agilent Genomic DNA labeling kit. The random primers labeled Exo-Klenow was used to label IP and input DNA samples as per Agilent Methyla-tion recommended protocol. The random primer labeled DNA sam-ples were then hybridized on a 180K customized DNA microarray chip. During hybridization, Cy3-labeled input was mixed with Cy5-labeled IP-DNA. This step was performed with a hybridization kit (Agilent 5188-6420) in the Agilent hybridization chamber at 67°C for 40 h. The hybridized slide was washed using Agilent wash buffer (part number 5188-5226) and scanned using the Agilent microarray scanner G2505C at 3 mm resolution.

The methylated and unmethylated regions that were of signifi-cance were identified using a methylation status detection algorithm known as Batman Algorithm. Briefly, this algorithm is designed for analyzing the differentially methylated profiles in two-color based assays. Before analyzing the differential methylation data, we excluded the possible source of technical bias using the methylation algorithm that allows normalization by melting temperature status. After analyzing the melting temperature, each probe is applied with Gaussian fits using the log ratios as z scores and P values. Gaussian data finally gives the probabilities and confidence values for methyl-ated and unmethylated probe populations. Together with Gaussian data, the algorithm calculates a methylation log odd, which gives the

ARTICLE IN PRESS

4 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

relative probability that that probe is more likely to be methylated than unmethylated [27,28].

RNA Extraction and complementary DNA synthesis

Total RNA was extracted from native MSCs, CHIR 99021 treated MSCs and differentiation (Neuronal, Cardiac, and Hepatocyte) induced CHIR-treated MSCs using Trizol (Sigma Aldrich, Missouri, USA) according to the manufacturer’s recommended protocol. Approximately 2 mg of total RNA was reverse-transcribed into com-plementary DNA in a 20 mL reaction volume using MMLV-RT enzyme (Thermo Fisher Scientific, MA, USA) and oligo-dT primers (New Eng-land Biolabs, MA, USA). Semi-quantitative RT-PCR was performed with corresponding primers of specific cell types listed in Supple-mentary Table S1 (Eurofins, Bangalore, India) at 95°C for 5 min, and 30 sec at 60°C. Melting curve analysis was performed after amplifica-tion to ensure the specificity of products. Relative messenger RNA expression was quantified by normalizing with b-actin, and then fold change was calculated using the 2 DDct method [29].

ICC

Native MSCs and CHIR 99021 treated MSCs differentiated into neuronal, cardiac and hepatocyte lineages were fixed with 2% para-formaldehyde solution for 40 min and permeabilized with 0.25% Tri-ton X-100 in DPBS for 15 min. Cells were washed with DPBS three times and incubated with blocking buffer containing 5% normal goat serum for 1 h at room temperature. Furthermore, cells were incu-bated with primary antibody overnight at 4°C with gentle shaking (antibody details and their dilutions are listed in Supplementary Table 2). Cells were then washed with DPBS and incubated again for 2 h with appropriate secondary antibody (IgG Alexa flor-594/ IgG-FITC with 1:1000 dilutions) at room temperature. Nuclei were stained with Hoechst 33258 dye (Sigma Aldrich, Missouri, USA) and observed under fluorescence microscope (Nikon Tie, Japan) 10X magnification.

Immunoblotting

The crude 30-mg protein samples were isolated from native MSCs and CHIR 99021 treated MSCs and boiled with laemmli sample buffer (1x) for 5 min. Proteins were run in 12% Sodium docecyl sul-phate poly acryl amide gel electrophoresis and transferred onto a nitrocellulose membrane of 0.22 mmol/L size (BioRad Laboratories, PA, USA), followed by blocking for 1 h at room temperature using Tris-buffered saline with 0.2% Tween 20 containing 5% Bovine serum albumin. The nitrocellulose membrane was incubated with primary antibodies at 4°C overnight (antibody details and their dilutions are listed in Supplementary Table S2) and washed three times for 15 min with DPBS. The blots were further incubated with secondary antibod-ies conjugated with horseradish peroxidase (Sigma Aldrich, Missouri, USA) for 2 h at room temperature, and bands were finally detected using chemiluminescence using Enhanced chemi luminescence (Pierce, Thermo Fisher Scientific, MA, USA).

Teratoma assay

In this study, 2 £ 105 cells of both MSCs and CHIR 99021 treated MSCs in 200 mL volume of DPBS were subcutaneously injected beneath the skin on the dorsal rear flank region of 6- to 8-week-old male nude mice housed in the animal house (IISc, Bangalore, India). Then, mice were monitored for teratoma signs up to 6 weeks. Mice were then humanely killed after about 6 weeks; tissues were collected and fixed in 10% formalin followed by paraffin embedding and sectioned at 5-mm thickness for light microscopy and stained with hematoxy-lin and eosin following standard histopathological procedures.

Ethical approval

This study was ethically approved and performed in accordance with the Institutional Ethical Committee of the Indian Institute of Technology-Madras, India (IEC/2016/01/RSV/-4/20). All study proce-dures involving human participants were under the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Statistical analysis

All experiments were performed in biological triplicates (n = 3). The statistical analysis was performed by one-way and two-way analysis of variance test. For multiple comparisons, Bonferroni post-test was used. For all statistical analyses, P < 0.05 was considered sig-nificant. *P < 0.05; **P < 0.01; ***P < 0.001.

Results

Isolation and characterization of WJ-derived MSCs

MSCs were isolated from the WJ region of cord tissue and ana-lyzed for its characteristic property as defined by the International Society for Cellular Therapy [30]. The isolated population showed spindle-shaped cells with fibroblastoid morphology 2 4 days after the seeding of collagenase digested cells in complete medium. These cells expressed CD105 (97%), CD90 (99%) and CD44 (95%) markers on their surface, whereas lower expression of CD34 and CD45 markers ( 2%) was observed as shown in Figure 1 (A1 A6). Adipogenesis-induced MSCs (P3) exhibited lipid droplets accumulation (triglycer-ides), as confirmed using oil red O staining in differentiated culture (Figure 1B1). Real-time PCR analysis showed the two-fold upregula-tion of adipocyte-specific markers like PPAR-g, CEBP-a and FABP4 in differentiated cells (Figure 1B4). Osteoblast-induced MSCs exhibited the mineralized calcium nodules and stained positive with alizarin red stain (Figure 1B2). Furthermore, osteoblast-specific genes such as COL1A1, RUNX2 and OPN were found to be upregulated in differenti-ated cells (Figure 1B5). Chondrogenesis induction was confirmed with the presence of sulphated proteoglycans in differentiated cells, as evidenced by toluidine blue staining (Figure 1B3). Chondrocytes were confirmed with the upregulation of COL2A1 and SOX9 genes and the downregulation of COL1A1 gene expression using real-time PCR analysis (Figure 1B6). Negligible expression of CD133 and OCT4, as assessed using flow cytometry, indicated the absence of very small embryonic-like stem cells in initial passages of cultures (Supplemen-tary Figure S1).

Pretreatment with CHIR 99021

Treatment with CHIR 99021 induced noticeable morphological changes in MSCs, changing from spindle fibroblastoid to flat epithe-lia-like. Increased proliferation rate was observed from the fifth to eighth day of post-treatment (Figure 2A2). After the 13th day in cul-ture, an increased number of small, compact colonies began to appear (Figure 2A3). ICC analysis showed nuclear-stabilized b-CATENIN in CHIR 99021 treated MSCs, whereas cytoplasmic localization was more evident in native MSCs (Figure 2B1). Furthermore, CHIR 99021 treated MSCs showed the loss of VIMENTIN, a mesenchymal-specific marker with a gain of Epithelial cell adhesion molecule (EpCAM), an embryonic and epithelial marker (Figure 2B1). Also, Western blot analysis validated the significantly upregulated levels of b-CATENIN nuclear localization in CHIR 99021 treated MSCs com-pared with native MSCs (Figure 2B2). The pGSK3b-Ser9, phosphory-lated form associated with the inactive state of GSK3b, correlated

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 5

Fig. 1. Characterization of WJ-MSCs. (A) Immunophenotyping of isolated WJ-MSCs. (A1) Histogram showing the isotype control used for the study. (A2 A4) Representative histo-gram showing the positive expression of MSC-specific markers CD90 (99%, A2), CD105 (97%, A3) and CD44 (95%, A4). (A5 and A6) Representative histogram showing the negative expression of hematopoietic markers, CD34 (<2%, A5) and CD45 (<2%, A6) in cultured MSCs CD34 (<2%, A5) and CD45 (<2%, A6). A total of 10 000 events were recorded for data analysis, and control cells were used for gating. (B) Biochemical and gene expression analysis of tri-lineage differentiated MSCs. (B1 B3) Representative biochemical staining image showing tri-lineage potential of MSCs. (B1) Bright-field microscopic image showing oil red O stained lipid droplets in adipocyte differentiated MSCs at 10X magnification. (B2) Bright-field image showing Ca+2 mineralized nodules in osteocyte-differentiated MSCs using alizarin red stained at 10X magnification. (B3) Photographic image showing the presence of sulphated proteoglycans in chondrocyte-differentiated MSCs via Toluidine staining. (B4 B6) qRT-PCR analysis of tri-lineage specific gene expression. (B4) Representa-tive bar graph showing the upregulated expression of adipocyte-specific genes, such as CEBPa, PPARg and FABP4 in adipocyte-differentiated MSCs. (B5) Representative bar graph showing the osteoblast-specific gene expression with more than 2 folds of COL1A1, RUNX2 and OPN. (B6) Representative bar graph showing the downregulated expression of COL1A1 gene and upregulated expression of COL2A1 and SOX9 genes. Multiple comparisons were statistically done between native and lineage-induced differentiation groups (adi-pocytes, osteocytes and chondrocytes) using Bonferroni post-test. (*P < 0.05; **P < 0.01; ***P < 0.001).

ARTICLE IN PRESS

6 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

Fig. 2. Characterization of CHIR 99021 treated MSCs. (A) Morphological characterization of CHIR 99021 treated MSCs. (A1) Phase contrast microscopic image of untreated native MSCs showing its characteristic spindle shape before treatment (10X magnification). (A2) Phase contrast microscope image at 10X magnification showing the visible morphological changes from typical spindle shape to epithelial cell-like on fifth day of post-treatment. (A3) Phase contrast image at 10X magnification showing the gradual appearance of small compact mass of clumped cells on the 15th day of post-treatment. (B) Protein profiling of CHIR 99021 treated WJ-MSCs. (B1) ICC analysis of CHIR 99021 treated MSCs. Fluorescent microscopic image panel showing nuclear localization of b-CATENIN in CHIR-treated MSCs, whereas, in native MSCs, localization of b-CATENIN was prominently observed in the cytoplasm (magnification 60X). Mesenchymal marker, Vimentin, was significantly lost after 6 days of post-treatment and appearance of EpCAM, an embryonic epithelial marker in CHIR 99021 treated MSCs (scale bar 100 mm). (B2) Immunoblotting of CHIR 99021 treated MSCs. Nuclear fraction showed a more profound localization of b-CATENIN in CHIR 99021 treated MSCs, whereas the native MSCs showed negligible expression of b-CATENIN in the nucleus. Cytoplasm fraction showed more cytoplasmic localization of b-CATENIN

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 7

with the increased nuclear stabilized b-catenin levels, as observed in CHIR 99021 treated MSCs. Native MSCs showed relatively less nuclear localization of b-CATENIN with less phosphorylated pGSK3b-Ser9 deposition, suggesting that CHIR 99021 treatment favored the inactivation of GSK3b followed by stabilization of b-CAT-ENIN and its nuclear translocation (Figure 2B2). Furthermore, the lev-els of downstream targets of b-CATENIN, such as c-MYC and CYCLIN D, were also significantly upregulated in CHIR 99021 treated MSCs as compared with native MSCs (Figure 2B2). The absence of chromo-somal abnormalities was demonstrated by G band analysis showing 46 + XY chromosomes in CHIR 99021 treated MSCs (Figure 2C).

Enhanced transdifferentiation potential of CHIR 99021 treated MSCs

Ectoderm-neuronal lineage induction

CHIR 99021 treated MSCs were induced to neuronal lineage using All trans retenoic acid Forskolin supplementation resulted in rosette-like structures on the third day post-induction followed by its expan-sion during the later phase of induction (Figure 3A3 and 3A4). Native MSCs treated with induction medium showed varied phenotypes but no rosette-like structures (Figure 3A1). Further treatment showed a predominant presence of stress fibers during the later phase of induc-tion (Figure 3A2). Increased expression of neuron-specific genes such as MAP2 and SYN (more than 30 folds), TUJ1 (19 folds) and NSE (nearly 100 folds) were observed using qRT-PCR analysis on the 18th day of culture in differentiated CHIR 99021 treated MSCs (Figure 3B). Native MSCs were shown to express increased expression of neuron-specific genes. However, the levels were not as signifi-cantly increased as in differentiated CHIR 99021 treated MSCs (Figure 3B). ICC analysis of neuron-specific proteins, such as NESTIN, MAP2, TUJ1 and GFAP, showed an observable increase in expression in differentiated CHIR 99021 treated MSCs as compared with differ-entiated native MSCs (Figure 3C1 4).

Endoderm-hepatocyte lineage induction

Polygonal-shaped cells were observed using bright-field micros-copy on the fourth day of hepatogenesis induced in CHIR 99021 treated MSCs (Figure 4A2), whereas native MSCs showed similar phenotype only on the eighth day of induction (Figure 4A1). Semi-quantitative analysis of hepatocyte-specific transcripts on the 16th day of culture showed more than a 4-fold increase in expression of ALB and HNF-4a, with a >40-fold increase in expression of the AFP gene in differentiated CHIR 99021 treated MSCs compared with dif-ferentiated native MSCs (Figure 4C1). Nuclear localization of HNF3b followed by upregulated expression of CK18, CK19 and ALB were observed with ICC of differentiated CHIR 99021 treated cells (Figure 4B1 4). Native MSCs were found to have a weak expression of CK18, CK19 and HNF3b (Figure 4B2 4), but no observable differ-ence in albumin expression was found in differentiated native MSCs from differentiated CHIR 99021 treated MSCs (Figure 4B1). PAS staining in differentiated CHIR 99021 treated MSCs exhibited intense glycogen accumulation (pink stained) after 20 days of induc-tion (Figure 4A4), as compared with differentiated native MSCs (Figure 4A3). Increased levels of urea with the maximum being 31 mg/L/106 cells at the 20th day of culture were obtained in differenti-ated CHIR 99021 treated MSCs as compared with native differenti-ated MSCs (8.2 mg/L/106cells; Figure 4C3). Similarly, the levels of ALB secreted in the spent medium showed a maximum increase of 2.3 g/ dL/106 cells on the 20th day of post-induction CHIR 99021 treated

MSCs compared with differentiated native MSCs (0.96 g/dL/106; Figure 4C2).

Mesoderm-myoblast lineage induction

Mesoderm lineage induced CHIR-treated MSCs showed a compact mass of cells clumped together to form visible organoid-like structures on fifth day post-induction (Figure 5A3 and 5A4). Native MSCs showed elongated morphology after 5 days of induction (Figure 5A1 and 5A2). Real-time PCR for gene expression in differentiated CHIR 99021 treated MSCs for cardiac-specific genes GATA4, NKX 2.5 and SERCA showed >3.5-fold upregulation compared with differentiated native MSCs (Figure 5C). Differentiated native MSCs expressed cardiac-specific genes GATA4 (0.7 fold), NKX 2.5 (2.6 folds) and SERCA (1.5 folds) more than control (Figure 5C). The expression of structural genes, such as MYH6 and MYL7, was increased by 1.5 folds after 28 days in culture (Figure 5C) of differentiated CHIR 99021 treated MSCs. Differentiated native MSCs showed negligible levels of expression of the same markers (0.2 fold in MYH6 and 0.1 fold in MYL7; Figure 5C). GATA4 showed increased nuclear localization and prominent expression of the structural gene MYLC2V was observed in differentiated CHIR 99021 treated MSCs (Figure 5B). GATA4 nuclear localization was not evident in differentiated native MSCs, and MYLC2V expres-sion was also absent (Figure 5B).

Immuno-expression profiling of HLA-DR antigen in CHIR 99021 treated MSCs and its subsequent derivatives

The immune reactivation status of the HLA-DR antigen was ana-lyzed in CHIR 99021 treated MSCs and its subsequent derivatives. FACS analysis showed that native MSCs were HLA-DR antigen-nega-tive by default as per minimal criteria to define MSCs [30] (Figure 6). Also, CHIR 99021 treated MSCs were found to be negative for HLA-DR antigen as well (Figure 6). Furthermore, subsequent derivatives of differentiated, CHIR 99021 treated MSCs were also negative (Figure 6). From this experiment, we confirmed that no HLA-DR anti-gen reactivation occurs before and after treatment with CHIR 99021, ensuring that such MSCs and cells differentiated from them are clini-cally safe for transplantation.

Methylation microarray analysis

Hierarchical clustering of differentially methylated core plurip-otent marker genes is represented in the form of a Heat map between MSCs and CHIR 99021 treated MSCs (Figure 7A). Scatter plot analysis showed that CHIR 99021 treated MSCs had a meth-ylation profile distinct from native MSCs (Figure 7B). The custom-ized human 180K differential methylation assay showed an increase in global Cytosine Guanine sequence (CpG) unmethy-lated percentage from 49.52% (n = 89 233 CpGs) to 55.25% (n = 95 693 CpGs), as well as in methylated CpG residues from 28.56% (n = 51 470) to 32.5% (n = 21 220) in CHIR 99021 treated MSCs compared with native MSCs (Figure 7C1 2). A similar pat-tern was obtained in the core pluripotent gene methylation pro-file, where methylated CpG residues were increased from 32.6 36.16% (Figure 7C3 4). Interestingly, there was a significant loss of methylation (from 72 60%) in the OCT4 promoter region, following CHIR 99021 treatment (Supplementary Table 1). The observed change in OCT4 promoter regions correlated with upre-gulation in OCT4 (Figure 8A1, 8B and 8C1 2) transcriptional and translational levels, whereas other pluripotent markers showed

in native MSCs. Thus, profound nuclear translocation of b-CATENIN was well evident in CHIR 99021 treated MSCs, compared with native MSCs. Similarly, the cytosolic pool of b-CATENIN is also higher in CHIR 99021 treated MSCs compared with control MSCs. Increased levels of p-GSK3b-ser9, a predictive marker for the inactive form of GSK3b, was found in CHIR 99021 treated MSCs, suggesting its pharmacological inhibition after treatment. The downstream targets of b-CATENIN, such as C-MYC and CYCLIN D, were also found to be upregulated in CHIR 99021 treated MSCs as compared with native MSCs. (B3) Representative bar graph showing the relative densitometry analysis of corresponding immunoblotting data of Figure B1 (*P < 0.05; **P < 0.01; ***P < 0.001). (C) Karyotyping analysis of CHIR 99021 treated MSCs. (C1 and C2) Karyogram showing intact 44+XY chro-mosomal banding pattern in CHIR 99021 treated MSCs, and, hence, no significant chromosomal abnormalities were observed after treatment.

ARTICLE IN PRESS

8 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

Fig. 3. Characterization of CHIR 99021 treated neurogenic-induced WJ-MSCs. (A) Morphological observation of CHIR-treated neuron lineage-induced MSCs. (A1 A2) Bright-field image showing neurogenesis induced native MSCs on the fifth and 20th day of induction (magnification 100 mm). (A3 A4) Bright-field image showing the formation of rosette-like structure, a characteristic feature of neuron progenitor cells on the third and 15th day of post-induction in CHIR treatment MSCs (10X and 20X magnifications). (B) Neuron-specific gene expression in neurogenic lineage-induced MSCs. Representative bar graph showing a nearly 100-fold increase for NSE gene, a more than 30-fold increase for SYN and MAP2 genes and a 19-fold increase in TUJ1 gene expression in 15 days’ culture of differentiated CHIR-treated MSCs, whereas native MSCs showed comparatively less expression of neu-ron-specific genes NSE (5 folds), SYN (8 folds), TUJ1 (5 folds) and NSE (24 folds). (*P < 0.05; **P < 0.01; ***P < 0.001). (C) Neuron-specific marker analysis in neurogenic lineage-induced MSCs. ICC for neuron-specific marker expression in CHIR 99021 treated MSCs showed the profound expression of neuron-specific markers, such as NES, TUJ1, MAP2 and GFAP, on the 20th day of induction. Differentiated native MSCs showed weak expression in NES, GFAP and TUJ1, whereas negligible expression was found in MAP2. Neuron-specific markers were negative in untreated native MSCs.

fewer changes with a >5% decrease in methylated CpG residues in SOX2, NANOG, BMI1, DNMT3A, CYCLIN-D, KLF4, LEF1 and DPPA5 in CHIR 99021 treated hMSCs.

Histone modification after CHIR 99021 treatment

Nuclear lysate of MSCs and CHIR 99021 treated MSCs were probed for detecting bivalent histone marks (Figure 7D1 2). The deposition of

active histone marks, such as H3K4Me3 and H3K36Me3, were increased significantly after CHIR 99021 treatment of MSCs compared with native MSCs. In contrast, there was no significant change in the repressive histone marks like H3K9Me3 and H3K27Me3 in CHIR 99021 treated MSCs compared with native MSCs (Figure 7D1 2). Thus, our analysis of histone modification suggests that abundant expression of transcriptionally active histone marks like H3K4Me3 and H3K36Me3 in MSCs after CHIR 99021 treatment may contribute to the

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 9

Fig. 4. Characterization of CHIR 99021 treated hepatogenic-induced WJ-MSCs. (A) Morphological observation of CHIR-treated hepatogenic-induced MSCs. (A1) Phase contrast microscopic image showing the polygonal-shaped epithelial-like cells on the eighth day of post-induction in differentiated native MSCs (Magnification 10X). (A2) Phase contrast microscopic image showing the more prominent polygonal-shaped epithelial-like cells at the fifth day of post-induction in differentiated CHIR 99021 treated MSCs (magnification 10X). (A3) PAS staining confirmed the glycogen storing potential in differentiated native MSCs (magnification10X). (A4) Differentiated CHIR-treated MSCs showed more intense gly-cogen stored evident in PAS staining compared with differentiated native MSCs (magnification10X), suggesting enhanced differentiation of CHIR-treated MSCs. (A5) PAS-stained untreated native MSCs (magnification10X). (B) ICC for hepatocyte-specific markers expression in CHIR 99021 treated MSCs showed a more profound expression of hepatocyte markers, such as HNF3b, CK18, CK19 and ALB, after 28 days as compared with differentiated native MSCs. Lower panel showing the absence of hepatocyte markers in untreated native MSCs (scale bar 100 mm). (C) Hepatocyte-specific gene expression and functional analysis in differentiated CHIR-treated MSCs. (C1) Representative bar graph showing the expression of hepatocyte-specific genes using qRT-PCR assay. Differentiated CHIR 99021 treated MSCs showed statistically significant upregulation of AFP gene and increased expression in HNF4a, and ALB genes were observed, whereas native differentiated MSCs showed weak expression of hepatocyte-specific genes. *P < 0.05; **P < 0.01; ***P < 0.001. (C2 and C3) Representative assays for analyzing hepatocyte functions. (C2) The albumin assay showed the significantly increased levels of albumin on days 5, 10, 15 and 20 of induc-tion in differentiated CHIR-treated MSCs in spent medium with maximum levels on day 20 (2.4 g/dL/106 cells) of post-induction. The maximum concentration of albumin was obtained in native differentiated MSCs as 0.8 g/dL/106 cells, which were comparatively less than differentiated CHIR-treated MSCs. (C3) The urea assay also showed a significantly higher level of urea on days 5, 10, 15 and 20 in spent medium, and the levels reached maximum on day 20 (as 30.8 mg dL/106 cells) of post-induction in differentiated CHIR 99021 treated MSCs. Differentiated native MSCs showed very low levels of urea (0.5 mg dL/106 cells) on day 20 of post-induction, thus confirming the enhanced differentiation in CHIR 99021 pretreated MSCs followed by differentiation induction. *P < 0.05; **P < 0.01; ***P < 0.001.

expression of stemness genes (Figure 8A C). Moreover, the deposition of both H4K4Me3 and H3K27Me3 in CHIR 99021 treated MSCs were suggestive of a bivalent domain similar to that seen in most pluripo-tent cells (ESCs and iPSCs) [55 60].

Analysis of significant stem cell gene expression in CHIR 99021 treated MSCs

Quantitative RT-PCR analysis showed upregulation of certain signifi-cant stem cell genes with a >50-fold increase in NANOG, >10-fold

ARTICLE IN PRESS

10 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

Fig. 5. Characterization of CHIR 99021 treated cardiomyocyte-induced WJ-MSCs. (A) Morphological observation of CHIR-treated hepatogenic-induced MSCs. (A1 A2) Phase-con-trast microscopic images showing elongated spindle morphology in MSCs treated with cardiac differentiation medium (A1, day 3 of induction; A2, day 18 of induction). (A3 A4) Phase contrast microscopic image showing visible organoid-like structures on day 5 of post-induction (A3) and day 20 of induction (magnification 10X). (B C) Cardiac-specific marker analysis in CHIR-treated cardiomyocyte-induced MSCs. (B) ICC analysis showed increased expression of cardiac-specific markers, such as GATA4 and MYLC2V, in differenti-ated CHIR-treated MSCs compared with differentiated native MSCs (scale bar 100 mm). Differentiation-induced cells showed nuclear localization of GATA4 more evident in CHIR-pretreated induced MSCs than native differentiated MSCs, whereas MYLC2V was expressed only in the CHIR-pretreated induced condition and was absent in native differentiated MSCs. (C) qRT-PCR gene expression analysis showed significantly increased expression (more than 3.5 folds) of cardiac markers, such as GATA4, NKX-2.5 and SERCA, in differenti-ated CHIR-treated MSCs as compared with native differentiated MSCs. Differentiated CHIR 99021 treated MSCs showed slightly increased expression in MYH6 and MYL7 (more than 1.5 folds) compared with native differentiated MSCs.

increase in OCT4 and b-CATENIN and a nearly 10-fold increase in SOX2 in CHIR 99021 treated MSCs compared with native MSCs (Figure 8B). WNT3A, a canonical Wnt pathway activating ligand, was also found to be upregulated (~2.4 folds) in CHIR 99021 treated MSCs (Figure 8B). Immunoblot analysis showed significantly increased expression of OCT4, SOX2 and NANOG in CHIR 99021 treated MSCs (Figure 8C1 and 8C2), whereas their negligible expression was found in native MSCs (Figure 8C1 and 8C2). The nuclear localization of OCT4, SOX2 and NANOG and the cell surface localization of TRA1-60 and TRA 1-80 in CHIR 99021 treated MSCs were further verified using ICC analysis (Figure 8A1 5). Interestingly, MSCs were found to have TRA1-60 (Figure 8A4) expression before being subjected to CHIR 99021 treat-ment. However, except for TRA1-60, other primitive markers were found to be absent in native MSCs (Figure 8A1 5).

Absence of teratoma formation from CHIR 99201 induced MSCs confers its clinical utility

Injection of native MSCs and CHIR 99021 treated MSCs into nude mice displayed no visible teratoma formation at the site of injection after 6 weeks of maintenance. Histochemical studies showed the predominant presence of mesenchymal origin cell types, such as fat cells, bone-forming cells and the absence of significant features of cells from both ectodermal and endodermal lineages suggesting CHIR 99021 treated MSCs were clinically safe (Figure 9Ba). A similar observation was also seen in control mice, and no significant changes were observed (Figure 9Bb and c).

Discussion

MSCs possess multilineage differentiation potential [31], but their utility is limited in regenerative therapy because they are not as

potent as ESCs or iPSCs. MSCs, however, have very few ethical con-cerns regarding their use in clinical applications. There are no reports of immune incompatibility, and MSCs can be isolated from a wide variety of sources, some of which harbor populations that are highly plastic and sensitive to differentiation induction cues/protocols [32,33]. A standardized and frequently used source of MSCs is bone marrow. However, the extraction and isolation of bone marrow from donors is invasive and causes discomfort [34,35]. Recently, a source of MSCs that is prolific, easily accessible without being invasive to the donor and hosts highly plastic populations has been identified the WJ [36]. WJ is an extra-embryonic tissue [37] from which populations of MSCs have been isolated that have increased differentiation poten-tial compared with MSCs isolated from other sources. WJ-MSCs have been observed to possess epigenomic traits of primitive stem cell lin-eages, and its epigenome is vastly different from that of an adult stem cell [38 40]. WJ-MSCs have a short cell doubling time, they are non-immunogenic and, because WJ is considered medical waste, they have an abundant source [41]. Treating a potent population such as WJ-MSCs with an effective pharmacological cue can further increase the efficiency of protocols for differentiation of MSCs [42]. CHIR 99021 is one such small molecule that has been extensively studied for its role in retaining pluripotency in ESCs and iPSCs by tar-geting and inhibiting GSK3b, activating the Wnt signaling pathway and expression of downstream b-catenin target genes [17 22].

Kim et al. showed the induction of a bi-potent progenitor cell state from human hepatocytes by small molecules A83-10 and CHIR 99021 [43]. Studies also reported that CHIR 99021 induced efficient hepato-cyte differentiation from human PSCs [44,45]. CHIR 99021 is a highly specific inhibitor of GSK3b, which is a central regulator of the Wnt/ b-catenin pathway. The Wnt pathway is integral to embryonic devel-opment, and it determines cell fate by modulating the shuttle of

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 11

Fig. 6. Immunoexpression analysis of HLA-DR antigen in native, CHIR-treated MSCs, and its subsequent derivatives. The representative histograms showing absence of HLA-DR expression in native (A2), CHIR-treated MSCs (A3) and its subsequent derivatives (A4-6) by flow cytometry analysis (A1) Histogram of isotype controls used in the FACS analysis.

cytosolic b-catenin activity to the nucleus and thus regulating gene expression [46,47]. In the absence of an activated Wnt signal, GSK3b mediates phosphorylation at the N-terminus of b-catenin and targets its proteasome-mediated degradation [48,49]. The Wnt/b-catenin signaling has been shown to maintain pluripotency in mouse ESCs [18,50 54]. An increase in b-catenin levels or the addition of Wnt3A to the culture medium promotes pluripotency and aids the expres-sion of NANOG and OCT4 [20]. Chemical inhibition of GSK3b by CHIR

99021 regulates gene expression by increasing cytosolic levels of b-CATENIN, leading to its nuclear translocation and prolonged inter-action with the Tcf family of transcription factors [47]. In this study, we observed nuclear localization of b-CATENIN in MSCs after treat-ment with CHIR 99021, suggesting a similar agonistic activity on Wnt signaling. We also observed the predominant loss of mesen-chymal marker (vimentin) and gain of EpCAM marker in CHIR 99021 treated MSCs, suggesting Mesenchymal epithelial

ARTICLE IN PRESS

12 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

Fig. 7. Nuclear status modification analysis in CHIR 99021 treated MSCs. (A) Hierarchical clustering of differentially methylated core pluripotent marker genes in the form of heat map between native MSCs and CHIR 99021 treated MSCs. (B) Scatter plot for CpG methylation values between control vs. CHIR 99021 treated MSCs. Each dot represents the indi-vidual CpG of the selected genes of interest. (C1 C2) Pie chart representation of the percentage methylation of CpGs in native MSCs (C1) and CHIR-treated MSCs (C2). The whole methylation data showed a significant increase in unmethylated CpG percentage levels from 49.5 55.25% after CHIR 99021 treatment. The methylation status of pluripotent core genes in native MSCs (C3) and CHIR-treated MSCs (C4) showed increased methylated CpG residue levels from 32.64 36.16% after treatment. (D1 D2) Histone modification analysis in CHIR 99021 treated MSCs. (D1) Nuclear lysate showed significantly increased deposition of active histone marks, such as H3K4Me3 and H3K36Me3, in CHIR 99021 treated MSCs, whereas there were those showing no significant change in the repressive histone marks (H3K9Me3 and H3K27Me3). Blot assay substantiated the existence of a bivalent chromatin state (a characteristic feature of pluripotent cells) in CHIR 99021 treated hMSCs. (D2) Representative bar graph showing the relative densitometry analysis of various histone-H3 methylation marks in CHIR 99021 treated hMSCs with respect to total H3, keeping the values in MSCs as 1. *P < 0.05; **P < 0.01; ***P < 0.001.

transition could have induced the enhanced transdifferentiation ability after being treated with CHIR 99021. CHIR 99021 inhibi-tion of GSK3b resulted in the accumulation of inactive Ser9 phos-phorylated form, which stabilized b-catenin levels and maintained the Wnt pathway in the active state, as evident in Figure 2. Furthermore, the sustained b-catenin translocation to the nucleus leads to the upregulated expression of OCT4, SOX2 and NANOG, as described in other studies [47] (Figure 8).

CHIR 99021, being an effective Wnt agonist, has been used in some studies in various concentrations. Different concentrations of CHIR 99021 are required to induce differentiation but this is depen-dent on the specific cell type. For instance, 3 mmol/L concentration of CHIR 99021 resulted in hepatic lineage differentiation and 12 mmol/L concentration was required to induce cardiac specification in iPSCs and ESCs respectively [26,70]. In our study, we observed CHIR 99021

mediated morphological events at its 10 mmol/L concentration, and enhanced the differentiation potential in MSCs. Additionally, the existence of bivalent chromatin consisting of active and repressive histone marks at the same location is essential to the maintenance of pluripotency in ESCs and iPSCs. Studies have shown that enriched H3 tri-methylated and di/trimethylated at lysine 4 and 27 (H3K27Me3 and H3K4Me3) were found to be associated with transcriptionally inactive (H3K27Me3) and active chromatin (H3K4Me3), respectively. These histone marks are essential to regulate bivalent genes in main-taining a pluripotent state or favoring differentiation upon stimula-tion [55 60]. A similar chromatin bivalent state was observed after CHIR 99021 treatment of MSCs with significantly increased deposi-tion of activating marks, H3K4Me3 and H3K36Me3, with no signifi-cant change in the levels of repressing marks, H3K9Me3 and H3K27Me3.

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 13

Fig. 8. Characterization of stemness marker expression in CHIR-treated MSCs. (A) Fluorescent microscopic image showing nuclear localization of stemness markers OCT4 (A1), SOX2 (A2) and NANOG (A3), and cell surface expression of TRA1-60 (A4),and TRA1-80 (A5) in CHIR 99021 treated MSCs. Native MSCs were found to express only TRA 1-60 among those stemness markers before being subjected to treatment (magnification 10X). (B) Stemness gene expression analysis showing nearly 10-fold increased expression of stemness markers, such as OCT4, SOX2 and b-CATENIN, and >50-fold increased expression of NANOG were observed in CHIR 99021 treated MSCs. Also, WNT3A expression was 2.4-fold higher in CHIR 99021 treated MSCs *P < 0.05; **P < 0.01; ***P < 0.001. (C1 C2) Immunoblotting of stemness markers (C1) showed profound expression of OCT4, SOX2 and NANOG proteins in CHIR 99021 treated MSCs. (C2) Representative bar graph showing the relative densitometry analysis of corresponding immunoblot data (C1) *P < 0.05; **P < 0.01; ***P

< 0.001.

Fig. 9. Nude mice teratoma assay. (a) Histochemical analysis showed the predominant presence of mesenchymal origin cell types, such as bone cells and fat cells, in MSCs and CHIR 99021 treated samples at the injected site, (b c) suggested that the derived cell types still functioned similarly to the control cell type and non-tera-togenic cell type.

CHIR 99021 promoted self-renewal of ESCs in mouse strains through stabilization of b-CATENIN and C-MYC at the protein level, by nuclear translocation of b-CATENIN and by activation of its down-stream targets [61,62]. Similarly, we also demonstrated the stabiliza-tion of b-CATENIN and upregulation in its downstream targets c-MYC and CYCLIN-D in CHIR 99021 treated MSCs on the sixth day of culture with immunoblotting. Differential methylation profiling using 180k DNA microarray data showed hypo-methylation in core-pluripotent genes. Importantly, the DNA microarray data showed the hypo-methylation of the OCT4 promoter region (Supplementary Table S3), correlating with a study on the upregulation of the OCT4 transcript level in CHIR 99021 pretreated MSCs [62]. Also, immuno-blotting for Wnt downstream targets after C59 (an antagonist of the Wnt pathway) treatment confirmed the agonistic activity of CHIR 99021 on the Wnt pathway (Supplementary Figure S2).

Several reports have shown that MSCs natively express immuno-modulatory cytokines, such as Leukemia inhibitory factor, Interleukin (IL)-2, IL-6, IL-8 and vascular endothelial growth factor [63,64]. MSCs are positive for HLA class I and negative for HLA class II, helping them overcome graft versus host rejection during transplantation [65,66]. CHIR 99021 treated MSCs were also observed to be HLA-DR nega-tive, thus ensuring no reactivation of HLA antigens after treatment (Figure 6). Chen et al. have demonstrated the non-teratoma forming potential of spontaneously formed embryoid bodies from extra-embryonic origin pig amniotic fluid derived MSCs [67]. Our study

ARTICLE IN PRESS

14 K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15

also reports that CHIR 99021 treated MSCs did not show any tera-toma formation in nude mice monitored up to 6 weeks, suggesting that MSCs after CHIR 99021 treatment retained their non-terato-ma forming property (Figure 9).

Esfandiari et al. showed GSK-3 inhibition promoted the prolifera-tion of the neuronal progenitors by activating b-catenin and NOTCH-related cell cycle genes in the human iPSC model system [68]. Neuro-genic differentiated CHIR 99021 treated MSCs in our study showed strong expression of neurogenic markers, such as Nestin, GFAP, TUJ1 and MAP2, compared with native differentiated induced MSCs. Kha-nabdali et al. have reported pretreatment with 5-azacytidine and zebularine induced differentiation of rat bone marrow MSCs into cardiomyocytes whereas that with b-mercaptoethanol would result in differentiation into neuronal-like cells [69]. In our study, pretreat-ment with CHIR 99021 showed enhanced transdifferentiation of WJ-MSCs into cardiomyocyte-like cells and neuron-like cells. Previously, Huang et al. treated human adipocyte-derived MSCs with CHIR 99021 and differentiated them into functional hepatocytes [23]. We also observed that pretreatment with CHIR 99021 enhanced the transdifferentiation of WJ-MSCs into hepatocytes compared with dif-ferentiation-induced native MSCs.

Additionally, we noticed that prolonged maintenance of primed CHIR 99021 treated MSCs in normal maintenance medium caused loss of the primed characteristic profile. However, immediate removal of CHIR 99021 from the medium conferred stable potent population and did not affect the efficiency of differentiation. Thus, the previously reported activity of CHIR 99021 as a potent inducer of differentiation into multiple lineages is corroborated in our study on MSCs.

Conclusion

MSCs can be a relevant and potent alternative to PSCs when a highly plastic source and specific inducers of differentiation, such as CHIR 99021, are identified and used in regenerative therapy. We observed nuclear localization of b-CATENIN in MSCs after treatment with CHIR 99021, inhibition of GSK3b resulting in accumulation of inactive Ser9 phosphorylated form, which stabilizes b-catenin levels and, in such manner, maintenance of an active Wnt pathway. We observed the sustained translocation of b-catenin to the nucleus resulting in upregulated gene expression of OCT4, SOX2 and NANOG. We also observed instances of chromatin bivalency after CHIR 99021 treatment of MSCs that are similar to chromatin states existent in PSCs. Also, differential methylation analysis demonstrated hypome-thylation of significant genes, such as OCT4, suggesting an epigenetic status that may respond to differentiation protocols and induction cues effectively. This increased transdifferentiation potential was demonstrated by efficiently differentiating CHIR 99021 treated MSCs into neuronal, cardiac and hepatic lineages. Thus, the Wnt ago-nist CHIR 99021 can be used to “prime” MSCs before differentiation into a state of increased potency. Although it is possible that WJ-MSCs may be more responsive to CHIR 99021 treatment given its extraembryonic origin, other sources of MSCs may show similar results because the Wnt signaling pathway is pivotal to stem cell maintenance and differentiation.

Declaration of Competing Interest

The authors declare no conflict of interest.

Author Contributions

G.K. and R.S.V. designed the study. G.K. performed all the experi-ments. G.K., P.K.G. and P.V. assisted in data analysis. G.K., P.V. and P.K. G. assisted in manuscript preparation. N.L. contributed to the neuro-nal differentiation study. R.S.V. conceived the whole study and was in charge of the overall direction, planning and execution of the planned

work. All authors provided critical feedback and contributed to man-uscript preparation.

Acknowledgment

G.K. and P.K.G. would like to thank University Grants Commission and Ministry of Human Resource and Development, Government of India, for their fellowship and acknowledge the Department of Bio-technology, IIT Madras, for providing infrastructure and facility. We are thankful to Mr. Bamadeb Patra, Senior Research Fellow, IIT Madras, for heterogeneity analysis by flow cytometry. We are also thankful to Prof. Ravi Sundaresan Nagalingam, Indian Institute of Sci-ence, Bangalore, for the nude mice teratoma assay and Dr. S. Meenak-shi, Associate Professor, Madras Medical College, Tamil Nadu, India, for HLA-DR-B1 genotyping analysis.

Supplementary Materials

Supplementary material associated with this article can be found in the online version at doi:10.1016/j.jcyt.2019.12.007.

References

[1] Faiella W, Atoui R. Therapeutic use of stem cells for cardiovascular disease. Clin-Transl Med. 2016;5:34.
[2] Liu Y, Tang SC. Recent Progress in Stem Cell Therapy for Diabetic Nephropathy. Kidney Dis (Basel) 2016;2:20–7.
[3] Hsuan YC, Lin CH, Chang CP, Lin MT. Mesenchymal stem cell-based treatments for stroke, neural trauma, and heat stroke. Brain Behav. 2016;6:e00526.
[4] Rosenthal N. Prometheus’s vulture and the stem-cell promise. N Engl J Med. 2003;349:267–74.
[5] Rossant J. Stem cells and early lineage development. Cell. 2008;132:527–31.

[6] Daley GQ, Scadden DT. Prospects for stem cell-based therapy. Cell. 2008;132: 544–8.

[7] Klimanskaya I, Rosenthal N, Lanza R. Deriveand conquer: sourcing and differenti-ating stem cells for therapeutic applications. Nat Rev Drug Discov. Cell. 2008;7
(2):131–42.

[8] Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embry-onic and adult fibroblast cultures by defined factors. Cell. 2006;126(4):663–76.
[9] Huangfu D, Maehr R, Guo W, Eijkelenboom A, Snitow M, Chen AE, et al. Induction of pluripotent stem cells by defined factors is greatly improved by small-molecule compounds. Nat Biotechnol. 2008;26(7):795–7.
[10] Schumann GG, Fuchs NV, Tristan-Ramos P, Sebe A, Ivics Z, Heras SR. The impact of transposable element activity on therapeutically relevant human stem cells. Mobile DNA 2019;10:9.
[11] Takahashi K, Tanabe K, Ohnuki M, Ichiasaka T, Tamoda K, Yamanaka S. Induction of pluripotent stem cells from adult human fibroblasts by defined growth factors. Cell. 2007;131(5):861–72.
[12] Kim JB, Zaehres H, Wu G, Gentile L, Ko K, Sebastiano V, Arauzo-Bravo MJ, et al. Pluripotent stem cells induced from adult neural stem cells by reprogramming with two factors. Nature 2008;454(7204):646–50.

[13] Veltieri M, Sorrentino A. The Mesenchymal Stromal Cell contribution to homeo-stasis. J Cell Physiol. 2008;217:296–300.
[14] Kim HJ, Park JS. Usage of human mesenchymal stem cells in cell-based therapy: Advantages and disadvantages. Dev Reprod. 2017;21:1–10.
[15] Jeevani T. Stem cell transplantation- types, risks and benefits. J Stem Cell Res Ther. 2011;1:114.
[16] Sepuvelda H, Aguilar R, Prieto CP, Bustos F, Aedo S, Lattus J, et al. Epigenetic signa-tures at the RUNX2-P1 and Sp7 gene promoters control osteogenic lineage com-mitment of umbilical cord-derived Mesenchymal Stem cells. Journal of Cellular Physiology. 2017;232:2519–27.

[17] Sato N, Meijer L, Skaltsounis L, Greengard P, Brivanlow AH. Maintenance of pluri-potency in human and mouse ESCs through activation of Wntsignalling by phar-macological GSK-3-specific inhibition. Nat Med. 2004;10:55–63.
[18] Ying QL, Wray J, Nicholas J, Batle-mocera L, Doble B, Woodgett J, et al. The ground state of embryonic stem cell self-renewal. Nature 2008;453:519–23.
[19] Sineva GS, Pospelov VA. Inhibition of GSK3b enhances both adhesive and signal-ling activities of beta-catenin in mouse embryonic stem cells. Bio Cell. 2010;102:549–60.

[20] Wray J, Kalkan T, Gomez-Lopez S, Eckardt D, Cook A, Kemler R, et al. Inhibition of glycogen synthase kinase-3 alleviates Tcf3 repression of the pluripotency net-work and increases embryonic stem cell resistance to differentiation. Nat Cell Biol. 2011;13(7):838–45.

[21] Kelly KF, Ng DY, Jayakumaran G, Wood GA, Koide H, Doble BW. b-Catenin enhan-ces OCT4 activity and reinforces pluripotency through a TCF-independent mecha-nism. Cell Stem Cell. 2011;8:214–27.

[22] Kirby LA, Schott JT, Nobel BL, Mendez DC, Caseley PS, Peterson SC, et al. Glycogen synthase kinase 3(GSK-3) inhibitors SB-216763 promotes pluripotency in mouse embryonic stem cells. Plos One. 2012;7:e39329.

ARTICLE IN PRESS

K. Govarthanan et al. / Cytotherapy 00 (2019) 1 15 15

[23] Huang J, Guo X, Li W, Zhang H. Activation of Wnt/b-catenin signaling via GSK3 inhibitors direct differentiation of human adipose stem cells to functional hepato-cytes. Sci Rep 2017;7:40716.

[24] Seshareddy K, Troyer D, Weiss ML. Method to isolate mesenchymal-like cells from Wharton’s jelly of umbilical cord. Methods Cell Biol. 2008;86:101–19.
[25] Standford CM, Jacobson PA, Eanes ED, Lembke LA, Midura RJ. Rapidly forming apatitic mineral in an osteoblastic cell line (UMR 106-01 BSP). J Biol. Chem 1995;270:9420–8.

[26] Peter DT, Henderson CA, Warren CR, Friesen M, Xia F, Becker CE, et al. Asialo-gly-coprotein receptor 1 is a specific cell surface marker for isolating hepatocytes derived from human pluripotent stem cells. Development. 2016;143:1475–81.
[27] Straussman R, Nejman D, Roberts D, Steinfeld I, Blum B, Benvenisty N, et al. Devel-opmental programming of CpG island methylation profiles in the human genome. Nature Structural and Molecular Biology 2009;16:564–71.

[28] Bentivegna A, Roversi G, Riva G, Paoletta L, Redaelli S, Miloso M, et al. The effect of culture on human bone marrow Mesenchymal stem cells; Focus on DNA Methyla-tion profiles. Stem Cells Int. 2016;2016:5656701.

[29] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta DeltaC(T)) Method. Methods 2001;25:402–8.
[30] Dominici M, Blanc LK, Muller I, Cortenbach S-I, Marini F, Krause D, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy. 2016;8(4):315–7.
[31] Pittenger MF, Macay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of adult mesenchymal stem cells. Science 1999;284(5411):143–7.
[32] Via AG, Frizziero A, Oliva F. Biological properties of mesenchymal stem cells from different sources. Muscles Ligaments Tendons J. 2012;2(3):154–62.
[33] Fitzsimmons REB, Mazurek MS, Soos A, Simmons CA. Mesenchymal stromal/stem cells in regenerative medicine and tissue engineering. Stem Cells International. 2018;8031718:1–16. 2018.

[34] Steinert AF, Rackwitz L, Gilbert F, Noth U, Tuan RS. Concise review: the clinical

application of mesenchymal stem cells for musculoskeletal regeneration: current status and perspectives. Stem Cells Transl Med. 2012;1(3):237–47.
[35] Marcin D, Joseph JP, Pillai AA, Devi A. Eminent sources of adult mesenchymal stem cells and their therapeutic imminence. Stem Cell Rev and Rep. 2017;13(6):741–56.
[36] Bharti D, Shivakumar SB, Park J-K, Ullah I, Subbarao B, Park J-S, et al. Comparative analysis of human Wharton’s Jelly MSCs derived from different parts of the same umbilical cord. Cell and tissue research 2018;372:51–65.
[37] Stubbendorf M, Deuse T, Hua X, Phan TT, Bieback K, Atkinson K, et al. Immunolog-ical properties of extra embryonic human mesenchymal stromal cells derived from gestational tissue. Stem Cells Dev. 2013;22(19):2619–29.
[38] Nekanti U, Rao VB, Bahirvani AG, Jan M, Totey S, Ta M. Long-term expansion of Wharton’s jelly-derived mesenchymal stem cells. Stem Cells Dev. 2010;19
(1):117–30.

[39] Fong CY, Chak LL, Biswas A, Tan JH, Gauthaman K, Chan WK, et al. Human Wharton’s Jelly stem cells have unique transcriptome profiles compared with human embry-onic stem cells and other mesenchymal stem cells. Stem Cell Rev. 2011;7(1):1–16.
[40] Wang Q, Yang Q, Wang Z, Tong H, Ma L, Zhang Y. Comparative analysis of human mesenchymal stem cells from fetal-bone marrow, adipose tissue and Wharton’s jelly as sources of cell immunomodulatory therapy. Hum Vaccin. Immunother. 2016;12(1):85–96.

[41] Watson N, Divers R, Kedar R, Mehindru A, Borlongan MC, Borlongon CV. Dis-carded Wharton jelly of the human umbilical cords: a viable source for mesen-chymal stromal cells. Cytotherapy 2015;17(1):18–24.

[42] Song H, Chang W, Song BW, Hwang KC. Specific differentiation of mesenchymal stem cells by small molecules. American journal of stem cells 2011;1(1):22–30.
[43] Kim Y, Kang K, Lee SB, Sea D, Yoon A, Kim SJ, et al. 2019. Small molecule-mediated reprogramming of human hepatocytes into bipotent progenitor cells. 70(1), 97-107.

[44] Siller R, Greenhough S, Naumovska E, Sullivan GJ. Small-molecule-driven hepato-cyte differentiation of human pluripotent stem cells. Stem Cell Reports 2015;4
(5):939–52.

[45] Du C, Feng Y, Qiu D, Xu Y, Pang M, Cai N, et al. Highly efficient and expedited hepatic differentiation from human pluripotent stem cells by pure small-mole-cule cocktails. Stem cell research & therapy 2018;9(1):58.

[46] Angers S, Thorpe CJ, Biechele TL, Goldenberg SJ, Zheng N, MacCoss MJ, et al. The KLHL12- Cullin-3 ubiquitin ligase negatively regulates the Wnt-beta-catenin path-way by targeting Dishevelled for degradation. Nature Cell Biol. 2006;8(4):348–57.
[47] Wu D, Pan W. GSK3 a multifaceted kinase in Wnt signaling. Trends in biochemical sciences 2010;35(3):161–8.

[48] Logan CY, Nusse R. The Wnt signaling pathway in development and disease. Annu. Rev. Cell Dev Biol. 2004;20:781–810.
[49] MacDonald BT, Tamai K, He X. Wnt/b Catenin signaling: components, mecha-nisms, and diseases. Dev Cell. 2009;17(1):9–26.

[50] Ogawa K, Nishinakamura R, Iwamatsu Y, Shimosato D, Niwa H. Synergistic action of Wnt and LIF in maintaining pluripotency of mouse ES cells. Biochem. Biophys. Res. Commun. 2006;343(1):159–66.

[51] Singla DK, Schneider DJ, LeWinter MM, Sobel BE. Wnt3a but not wnt11 supports self-renewal of embryonic stem cells. Biochem. Biophys. Res. Commun 2006;345
(2):789–95.

[52] Takao Y, Yokota T, Koide H. Beta-catenin up-regulates Nanog expression through interaction with Oct-3/4 in embryonic stem cells. Biochem. Biophys. Res. Com-mun. 2007;353(3):699–705.

[53] Wagner RT, Xu X, Yi F, Merrill BJ, Cooney AJ. Canonical Wnt/b-catenin regulation of liver receptor homolog-1 mediates pluripotency gene expression. Stem Cells. 2010;28(10):1794–804.

[54] Sokol SY. Maintaining embryonic stem cells pluripotency with Wnt signaling. Development 2011;138(20):4341–50.
[55] Bernstein BE, Mikkelsen TS, Xie X, Kamal M, Huebert DJ, Cuff J, et al. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell. 2006;125(2):315–26.
[56] Mikkelsen TS, Ku M, Jaffe DB, Issac B, Liberman E, Giannoukos G, et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 2007;448(7153):553–60.

[57] Pan G, Tian S, Nie J, Yang C, Ruotti V, Wei H, et al. Whole genome analyzing of his-tone H3 lysine 4 and lysine 27 methylation in human embryonic stem cells. Cell Stem Cell. 2007;13(1):299–312.

[58] Bilodeau S, Kagey MH, Frampton GM, Rahl PB, Young RA. SetDB1 contributes to repression of genes encoding developmental regulators and maintenance of ES cell state. Genes Dev 2009;23(21):2484–9.

[59] Landeira D, Sauer S, Poot R, Dvorkina M, Mazzarella L, Jorgensen HF, et al. Jarid2 is a PRC component in embryonic stem cell required for multi lineage differentia-tion and recruitment of PRC1 and RNA polymerase II to developmental regulators. Nat.Cell. Biol. 2010;12(6):618–24.

[60] Alder O, Lavial F, Helness A, Brookes E, Pinho S, Chandrashekaran A, et al. Ring 1B and Suv 39h1 delineate distinct chromatin state at bivalent genes during early mouse lineage commitment. Development 2010;137(15):2483–92.

[61] Shoudong Y, Tan L, Yang R, Fang B, Qu S, Schulze EN, et al. Pleiotropy of Glycogen Synthase Kinase-3 inhibition by CHIR 99021 promotes self-renewal of embryonic stem cells from refractory mouse strains. PLoS One 2012;7(4):e35892.
[62] Meng X, Su R-J, Baylink DJ, Neisses A, Kirayon JB, Lee WY-W, et al. Rapid and effi-cient reprogramming of human fetal and adult blood CD34+ cells into mesenchy-mal stem cells with a single factor. Cell Res. 2013;23(5):658–72.
[63] Than NG, Romero R, Erez O, Weckle A, Tarca AC, Hotra J, et al. Emergence of hor-monal and redox regulation of galectin-1 in placental mammals: implication in maternal-fetal immune tolerance. Proc. Natl. Acad. Sci. USA. 2008;105 (41):15819–24.

[64] Marti M, Mulero L, Pardu C, Murera C, Carrio M, Laricchia-Robbio L, et al. Charac-terization of Pluripotent Stem Cells. Nature Prot 2013;8(2):223–52.
[65] Djouad F, Charbonnier LM, Bouffi C, Louis-Plence P, Bony C, Apparailly F, et al. Mesenchymal stem cells inhibit the differentiation of dendritic cells through an interleukin-6-dependent mechanism. Stem Cells. 2007;25(8):2025–32.
[66] Omar RE, Beroud J, Stoltz JF, Menu P, Velof E, Decot V. Umbilical cord MSCs: The new gold standard for MSC-base therapies. Tissue Engineering Part B Rev. 2014;20(5):523–44.
[67] Chen J, Lu Z, Cheng D, Peng S, Wang H. Isolation and characterization of porcine amniotic fluid derived multipotent stem cells. PLoS ONE 2011;6(5):e19964.
[68] Esfandiari F, Fathi A, Gourabi H, Kiani S, Nemati S, Baharvand H. Glycogen syn-thase kinase-3 inhibition promotes proliferation and neuronal differentiation of human-induced pluripotent stem cell-derived neural progenitors. Stem Cells Dev. 2012;21(17):3233–43.

[69] Khanabdali R, Saadat A, Fazilah M, Bazli KF, Qazi RE, Khalid RS, et al. Promoting effect of small molecules in cardiomyogenic and neurogenic differentiation of rat bone marrow-derived mesenchymal stem cells. Drug design, development and therapy 2015;10:81–91.

[70] Qiu XX, Liu Y, Zhang XF, Guan YN, Jia QQ, Wang C, et al. Rapamycin and CHIR 99021 coordinated robust cardiomyocyte differentiation from human pluripotent stem cells via reducing p53 dependent apoptosis. J.Am.Heart. Assoc. 2017;6(10). pii e005295.CHIR-99021