Supplementary Methods

1. hESC and mESC culture

Human ESC line, VAL3 (Valbuena et al., 2006) was obtained from Príncipe Felipe Research Center, Valencia, Spain. They were maintained on Matrigel (BD Biosciences, US) coated plate with mTeSR1 (Stemcell technologies, Canada). The culture medium was replaced every 2 days and VAL3 were passaged every 4-5 days. VAL3 were passaged using Accutase (Stemcell technologies) and supplemented with ROCK inhibitor (Y-27632, Stemcell technologies) for the first day after cell seeding. Mouse embryonic stem cell (mESC) line, L4 was obtained from Department of Biochemistry, The University of Hong Kong. They were maintained on 0.1% gelatin (Sigma-Aldrich) coated plate with mESC medium [Knockout DMEM (Life Technologies) supplemented with 15% Knockout serum replacement (KOSR, Life Technologies), 4 mM L-glutamine (Life Technologies), 1x penicillin/streptomycin (Life Technologies), 1x non-essential amino acids (Life Technologies), 1 mM sodium pyruvate (Sigma-Aldrich), 0.1 mM 2-mercaptoethanol (Sigma-Aldrich), 1x N2 supplement (Life Technologies), 1x B27 supplement (Life Technologies), 1000 units/ml leukemia inhibitory factor (Life Technologies), 1 mM PD0325901 (Tocris, R&D Systems, US) and 3 mM CHIR99021 (Tocris)]. The culture medium was replaced every 2 days and L4 were passaged every 3 days using 0.05% trypsin-EDTA (Life Technologies).

2. Embryoid body (EBs) formation from hESC

VAL3 were digested into single cells using Accutase (Stemcell technologies) and seeded at 2.4 x 105 cells/well on AggreWell 400 (Stemcell technologies) with mTeSR1 (Stemcell technologies) and Y-27632 for embryoid body formation. After 24 hr, the EBs were transferred to 6-well ultra-low attachment plates (Corning, US) and cultured in DMEM low glucose (5.5 mM D-glucose, Life Technologies) supplemented with 15% FBS (Life Technologies) for 3 more days. The EBs were then transferred to 0.1% gelatin (Sigma-Aldrich) coated plates for attachment growth. For the hyperglycemic treatment, D-glucose (Sigma-Aldrich) was added to the differentiation medium to a final concentration of 25 mM or 50 mM from day 1 onwards. The area of EB outgrowth was assessed by Image-Pro Plus (MediaCybernetics, US) and data was presented as pixels.

3. DE and PP differentiation from ESCs and hyperglycemic treatments

In this study, we compared two protocols for DE differentiation of VAL3. First, the commercially available STEMdiff Definitive Endoderm Kit (STEMdiff DE kit, Stemcell Technologies) was used. In brief, VAL3 were seeded at 0.021 x 106 cells/cm2 and differentiated 24hr later in the DE differentiation medium supplemented with components A and B for 1 day followed by 3 days with component B only. Later on, we also adopted Pagliuca and coworkers’ protocols for DE differentiation in low glucose (5.5 mM D-glucose) basal medium (Pagliuca et al., 2014) with modifications. In brief, VAL3 were seeded at 0.021 x 106 cells/cm2 and cultured in mTeSR1 for 2 days. The culture media was changed to MCDB131 (5.5 mM D-glucose, Life Technologies) supplemented with 2% BSA (Sigma-Aldrich), Insulin-Transferrin-Selenium-Ethanolamine (ITS-X, Life Technologies) at 1:50 ratio, 0.25 mM ascorbic acid (Sigma-Aldrich) and 100 ng/ml Activin A (R&D Systems) for 3 days. CHIR99021 (3mM, Tocris) were added to the differentiation media on the first day only. For the PP differentiation of VAL3, STEMdiff Pancreatic Progenitor Kit (STEMdiff PP kit, Stemcell Technologies) was used. Briefly, VAL3 were seeded at 0.21 x 106 cells/cm2 and differentiated 24hr later in the PP differentiation medium supplemented with components 1A, 1B, 2A, 2B, 3 or 4 according to manufacturer’s instructions. The differentiation process lasted for 13 days. For the hyperglycemic treatment, D-glucose or L-glucose (Sigma-Aldrich) at different concentrations was added to the differentiation medium.

For the differentiation of the mouse L4 cells into DE, the protocol of Borowiak and coworkers was adopted (Borowiak et al., 2009). Briefly, L4 was seeded at 2500 cells/cm2 on gelatinized plates and cultured in mESC medium for 2 days. The cells were then cultured in L-glutamine (Life Technologies) supplemented with Advanced RPMI 1640 (Life Technologies) containing 11.11 mM glucose for 1 day. DE differentiation was initiated by 5 mM inducer of definitive endoderm 1 (IDE1, Tocris) and 0.2% FBS was then added for 2 days followed by IDE1 and 2% FBS for another 2 days. For the hyperglycemic treatment, D-glucose (Sigma-Aldrich) was supplemented to the differentiation medium to a final concentration of 25 or 50 mM from day 1 onwards.

4. Western blotting and immunofluorescence staining

The cells were lysed by the cell disruption buffer from the mirVana PARIS Kit (Life Technologies). The cell lysate for protein analysis was mixed with protease inhibitor (Calbiochem, Millipore, US). Bradford assay was used to determine the protein concentration. Equal amounts (5 mg) of protein samples were denatured and subjected to SDS-PAGE. The antibodies used for Western blotting were as follows: goat anti-Sox17 (1:1000; AF1924, R&D Systems), rabbit anti-Oct4 (1:1000; sc-9081, Santa Cruz Biotechnology, US), mouse anti-active b-catenin (1:200; 05665, Millipore), mouse anti-total b-catenin (1:1000; BD610153, BD Biosciences), goat anti-Pdx1 (1:1000; AF2419, R&D Systems) and mouse anti-b-actin (1:10000; A5316, Sigma-Aldrich). For immunofluorescence staining, the cultured cells were fixed with 4% formaldehyde freshly prepared from paraformaldehyde (Sigma-Aldrich) and permeabilized with 0.1% Triton X-100 (Bio-Rad, US), followed by blocking. The cells were then incubated with antibodies including rabbit anti-Oct4 (1:50; sc-9081, Santa Cruz Biotechnology), mouse anti-TRA-1-81 (1:50; 90233, Millipore), goat anti-Sox17 antibody (1:100; AF1924, R&D Systems), rabbit anti-FoxA2 (1:100; AB4125, Millipore) and rabbit anti-Pdx1 (1:100; 061379, Millipore). The nucleus of the cells was stained with Hoechst 33342 (H1399, Life Technologies). The fluorescent signals were observed under a confocal microscope (LSM 700; Carl Zeiss AG, Germany).

5. Reverse transcription and real-time quantitative polymerase chain reaction (RT-qPCR)

Total RNAs were extracted by the mirVana PARIS Kit (Life Technologies). The extracted RNA was reverse transcribed by the TaqMan Reverse Transcription Reagents (Life Technologies). The cDNA obtained was used for quantification of gene expression by qPCR using a 7500 Real-Time PCR System (Life Technologies). The PCR system was programmed for 40 cycles of 15 sec. at 95oC and 1 min. at 60oC. The expression of mRNA was normalized with endogenous 18S ribosomal RNA using the 2-∆∆CT method, and the data were analyzed by the manufacturer’s software (Life Technologies).

6. Fluorescence-activated cell sorting (FACS)

VAL3 at undifferentiated or DE stages, and mouse embryos at E7.5 were dispersed to single cells using Accutase or trypsin. The dispersed cells were then resuspended in “flow buffer” (0.2% BSA in PBS), fixed in methanol (pre-cooled at -20oC) for 15 min., followed by incubation with primary and secondary antibodies for 30 min. FACS was performed in a BD LSR Fortessa Analyzer (BD Biosciences) and DIVA software (BD Biosciences) provided by the Faculty Core Facility at the University of Hong Kong. FACS data were analyzed by the FlowJo software (Tree Star, US).

7. Chromatin precipitation (ChIP) assay

Chromatin precipitations were performed as previously described with minor modifications (Bramswig et al., 2013). In brief, samples were cross-linked with 1.11% formaldehyde for 10 min. at room temperature and quenched with 0.14 M glycine. Samples were lysed and sonicated with Soniprep 150 (MES, UK) four times for 10 sec. each. The resulting chromatin fragments (200-500 bp) were then incubated with 2.5 mg antibody per 1x106 cells overnight at 4°C with rotation. The complexes were further incubated with Dynabeads Protein G (Life Technologies) for 1 hr at 4°C with rotation. The immunoprecipitated complexes were eluted, reverse cross-linked, and subjected to PCR amplification using primers listed in Table S1.

8. Bisulfite sequencing

Genomic DNA was extracted using DNeasy blood and tissue kit (Qiagen, Netherlands) according to the manufacturer’s instructions. The target genomic regions were subjected to PCR amplification using primers listed in Table S1. The PCR products were ligated into pGEM-T Easy Vector (Promega, US), followed by transformation into DH5a competent cells (Life Technologies). The transformed cells were plated onto LB agar (Sigma-Aldrich) supplemented with 100 mg/ml ampicillin (Sigma-Aldrich), 0.5 mM IPTG (Sigma-Aldrich) and 80 mg/ml X-Gal (Sigma-Aldrich). The single positive clones were sub-cultured. DNA was then extracted using QIAprep Spin Miniprep Kit (Qiagen). Sequencing service was provided by Tech Dragon Limited, Hong Kong.

9. Cell viability assay

The CyQUANT NF Cell Proliferation Assay Kit (Life Technologies) was used according to the manufacturer’s instructions. In brief, cells were seeded in 96-well plates and differentiated using the STEMdiff kit. On day 5, the culture medium was aspirated, followed by washing with PBS. Dye binding solution was then added to the cells and incubated at 37°C for 30 min. The fluorescence intensity of each sample was measured using a fluorescence microplate reader, (Infinite 200; Tecan, Switzerland) with excitation at 485 nm and emission detection at 535 nm.

Supplementary figures

Supp. Figure 1: Spontaneous and directed differentiation of VAL3 into DE. (A) Relative mRNA levels of the DE markers along the course of embryoid body (EB) formation (n=3, *:P<0.05 compared to d0, one way ANOVA), (B) the effects of different glucose levels on relative mRNA levels of ectoderm and mesoderm markers on day 8 of EB formation, (C) the relative mRNA levels of DE markers (SOX17, FOXA2, CXCR4 and EOMES) and pluripotent markers (OCT4, NANOG and SOX2) upon DE differentiation (n=5, **:P<0.01, Mann-whitney rank sum test), (D) immunofluorescence staining of SOX17 upon DE differentiation. The scale bar represents 100mm.

Supp. Figure 2: Hyperglycemia did not alter the cell count. The cell count upon DE differentiation with or without D-/L-glucose treatment (n=3).

Supp. Figure 3: DE differentiation using Pagliuca’s protocol with low basal glucose background. (A) The relative mRNA levels of DE markers (SOX17, FOXA2, CXCR4 and EOMES) and pluripotent marker (OCT4) upon DE differentiation using Pagliuca’s protocol (n=6, **:P<0.01, Mann-whitney rank sum test), (B) immunofluorescence staining of SOX17 and OCT4 upon DE differentiation using Pagliuca’s protocol. The scale bar represents 100mm.

Supp. Figure 4: Hyperglycemia retained H3K27me3 mark and EZH2 on promoters of DE markers. (A) Effects of different glucose levels on the relative mRNA levels of DE markers at DE stage using STEMdiff DE kit (n=5, *:P<0.05, **:P<0.01, one way ANOVA), (B) The binding of H3K27me3 on the promoters of SOX17 and FOXA2 with or without L-glucose treatment using STEMdiff DE kit (n=3, **:P<0.01, #:P<0.001, one way ANOVA). [lane 1: VAL3, lane 2: DE (15 mM), lane 3: DE (L-glucose 25 mM), lane 4: DE (L-glucose 50 mM)], (C) The binding of H3K27me3 on the promoters of SOX17 and FOXA2 with or without L-glucose treatment using Pagliuca’s protocol (n=3, *:P<0.05, **:P<0.01, one way ANOVA). [lane 1: VAL3, lane 2: DE (5.5 mM), lane 3: DE (L-glucose 13.5 mM), lane 4: DE (L-glucose 25 mM)], (D) Effects of different glucose levels on the relative mRNA levels of EZH2 and KDM6B at DE stage using STEMdiff DE kit (n=5, **:P<0.01, one way ANOVA).

Supp. Figure 5: DNA methyltransferase inhibitor did not affect DE differentiation. (A) DNA methylation status of OCT4 proximal promoter upon 5-aza (0.4 mM) treatment. Black dots represented methylated CpG sites, while white dots represented unmethylated CpG sites. (B) The relative mRNA levels of DE markers (SOX17, FOXA2, CXCR4) and pluripotent markers (OCT4 and NANOG) upon 5-aza treatment (n=3).

Supp. Figure 6: Upsurge of mesendoderm markers during early DE differentiation. The temporal mRNA expression levels of mesendoderm markers (MIXL1 and BRACHYURY) upon DE differentiation (n=5, *:P<0.05, one way ANOVA).

Supp. Figure 7: DE differentiation of L4 using small molecule IDE1. Relative mRNA expressions of DE markers and pluripotent markers upon IDE1 treatment (n=4, *:P<0.05, Mann-whitney rank sum test).

Supp. Figure 8: Creation of diabetic mouse model using STZ. Mean blood glucose level of STZ-treated mice (n=14) and control (n=5) (**:P<0.01, t-test).

Supp. Figure 9: VAL3 efficiently induced into PP using STEMdiff PP kit. Relative mRNA expression of NKX6-1 at PP stage (n=4, *:P<0.05, Mann-whitney rank sum test).

Supplementary table

Table S1: Primers sequences

Primers used for chromatin immunoprecipitation (ChIP)
Gene / Primer sequence
SOX17 / Forward: gaatggacgctcggtatgtt
Reverse: gagactcgaaaagccgtctg
FOXA2 / Forward: gagcctccacatccaaacac
Reverse: cagcagctcttgggttcaa
CXCR4 / Forward: tccactttagcaaggatggac
Reverse: tcccagaggcatttcctaag
EOMES / Forward: tcaacttgaccgatgctttg
Reverse: ctcacacgctggaagaaggt
Primers used for bisulfite sequencing
OCT4 / Forward: AAGTTTTTGTGGGGGATTTGTAT
Reverse: CCACCCACTAACCTTAACCTCTA