Directed differentiation of human iPSC into insulin producing cells is improved by induced expression of PDX1 and NKX6.1 factors in IPC progenitors

Reagents

Unless specified otherwise, all chemicals were from Sigma-Aldrich (St. Louis, MO, USA). Cell culture media were purchased from Life Technologies (Carlsbad, CA, USA). Restriction endonucleases, polymerases and DNA modifying enzymes were from New England Biolabs (Ipswitch, MA, USA).

DNA constructs

To create pENTR/zeo-Pdx1-VP16 vector, Pdx1 coding region in frame with VP16 trans-acting coding sequence from the Herpes Simplex virus was synthesised by GeneArt AG (Regensburg, Germany) as a String™ DNA, and amplified with Q5 DNA polymerase using Pdx1-VP16 attB1 and Pdx1-VP16 attB2 oligonucleotides. The PCR product was shuttled with Gateway BP Clonase II (Life Technologies) into pDONR/zeo plasmid (Life Technologies) resulting in pENTR/zeo-Pdx1-VP16 construct. The sequence was confirmed by DNA sequencing with M13 Fwd (−20) and M13 Rev primers. To generate pENTR/zeo-Nkx6.1 construct, the total RNA was isolated from HEK293T cells and reverse-transcribed with random hexamer primers and M-MuLV reverse transcriptase. Nkx6.1 coding sequence was amplified from the cDNA with Q5 DNA polymerase with Nkx6.1 attB1 and Nkx6.1 attB2 primers and transferred into pDONR/zeo plasmid by means of the Gateway BP Clonase II enzyme to create pENTR/zeo-Nkx6.1 vector. In order to generate pLVX-TRE3G-DEST construct, pLVX-TRE3G-IRES plasmid (Clontech, Mountain View, CA, USA) was digested with EcoRI and BamHI restriction enzymes and filled with T4 DNA polymerase. The vector was isolated from the agarose gel with QIAquick DNA Gel Extraction Kit (Qiagen, Venlo, Limburg, The Netherlands) and ligated using T4 DNA ligase with Gateway Rf.A cassette (Life Technologies) that encompasses attR1 and attR2 recombination sequences, chloramphenicol resistance gene and ccdB negative selection marker. The correct orientation of the ligated insert was determined by restriction digest and DNA sequencing using primers Gateway 1 and Gateway 2. The transfer of Pdx1-VP16 fragment from the ENTRY construct into the final pLVX-TRE3G-Pdx1-VP16 vector was done using Gateway LR Clonase II enzyme (Life Technologies). The pLVX-TRE3G-Nkx6.1 plasmid was created by shuttling of Nkx6.1 sequence from the pENTR/zeo-Nkx6.1 vector into pLVX-TRE3G-DEST with Gateway LR Clonase II. The identity of the last two vectors was determined by restriction digest and DNA sequencing using pLVX-TRE3G Fwd and pLVX-TRE3G Rev primers. To generate pCXB-EmGFP vector, EmGFP coding sequence was PCR-amplified from pRSET-EmGFP plasmid (Life Technologies) with FP attB1 and FP attB2 primers. Obtained PCR product was shuttled into pDONR/zeo plasmid with application of Gateway BP Clonase II enzyme, resulting in pENTR/zeo-EmGFP construct. The transfer of the EmGFP fragment from the ENTRY construct into final expression vector was done by with Gateway LR Clonase II enzyme and pCXB-DEST plasmid (described in [17]). DNA sequences of oligonucleotides used for cloning purposes are included in Additional file 1: Table S1.

Generation and culture of induced pluripotent stem cells

Unless stated otherwise, iPS cells were generated as described previously [17]. Briefly, cells were seeded on Geltrex-coated cell culture plates and incubated overnight in a suitable primary medium. On the next day, cells were transfected with five episomal plasmids (pCE-hOct3/4, pCE-hSK pCE-hUL, pCE-mp53DD, and pCXB-EBNA1). After a 24-h incubation, medium was replaced with TeSR-E7 medium (StemCell Technologies), and the transfection was repeated. The medium was changed every other day for 14 days. After that, culture medium was replaced with Essential 8 and cells were cultivated until iPSC colonies started to appear. Then, they were transferred onto a new, Geltrex-coated culture dishes, expanded and maintained in Essential 8 medium.

Cell culture

iPS cells derived from human fibroblasts, established as described earlier [17], were cultured in low oxygen conditions (5% O₂/5% CO₂), in chemically defined Essential 8 medium on cell culture plates coated with Geltrex (20 μg/cm²) or recombinant Vitronectin (0.5 μg/cm²). Cells were passaged twice a week when they reached 60–70% confluence. Pluripotent cells were incubated in Essential 8 medium supplemented with 10 μM Y-27632 for 1 h before passaging and colony expansion. Cells were dissociated with 0.5 mM EDTA in ion-free PBS for 5–8 min at room temperature. Afterwards, EDTA solution was removed, and the process of cell separation was terminated by addition of Essential 8 medium. Clumps of cells were then transferred into fresh culture vessels coated with Geltrex or Vitronectin. Renal epithelial cells were isolated from urine samples according to the protocol described previously [18]. Cells were cultured on gelatin-coated cell culture plates and maintained in REBM medium supplemented with REGM SingleQuot Kit (Lonza, Basel, Switzerland). HFF-1 (neonatal human foreskin fibroblasts) and HEK293T cell lines were obtained from ATCC (Manassas, VA, USA). Cells were cultured in DMEM supplemented with 10% FBS. HeLa cells were purchased from DSMZ (Braunschweig, Germany) and cultured in DMEM medium supplemented with 10% FBS. In addition, all the abovementioned cell culture media contained penicillin (100 U/mL), streptomycin (100 µg/mL), amphotericin B (0.25 µg/mL) and 2.5 µg/mL of Plasmocin (InvivoGen, San Diego, CA, USA).

Coating agents

Geltrex™ hESC-qualified basement membrane matrix and Coating Matrix Kit (human, recombinant Colagen I) were purchased from Life Technologies. Vitronectin XF™, full-length human recombinant protein was obtained from STEMCELL Technologies (Vancouver, BC, Canada). Laminin-511 and Laminin-521 was purchased from BioLamina AB (Sundbyberg, Sweden). Fibronectin from human plasma, Laminin from Engelbreth-Holm-Swarm murine sarcoma basement membrane, Collagen IV from human placenta and porcine Gelatin were from Sigma-Aldrich. All of the protein coating agents were diluted in DMEM/F12 basal medium to desired concentrations, applied on the surfaces of cell culture vessels and incubated for 1 h at 37 °C.

Preparation of L64-PEI copolymer

L64-PEI copolymer was synthesised as reported earlier [19, 20] with minor modifications. Briefly, 2 g of pluronic-L64 PEG-PPG-PEG block polymer, with average molecular weight of 2900 Da, was desiccated in a vacuum dryer (VacuCell 22, MMM Medcenter, Germany) at 40 °C overnight and activated with threefold molar excess of 1,1′-carbonyldiimidazole (CDI) in 10 mL of anhydrous acetonitrile. After 3 h of incubation on shaker platform at room temperature, the reaction mixture was diluted with equal volume of deionized water and dialyzed against 10% ethanol/H₂O for 24 h using 1 kDa molecular weight cut-off cellulose membrane tube. Following dialysis, the activated Pluronic-L64 derivate was desiccated in a vacuum dryer overnight at 40 °C and dissolved in 20% ethanol/H₂O. The activated polymer was conjugated with fivefold molar excess of polyethyleneimine (PEI, molecular weight 1200 Da) and the resulting solution was stirred overnight at room temperature. In order to remove unincorporated PEI, the reaction mixture was dialyzed for 48 h at room temperature against deionized water using 3.5 kDa cut-off cellulose membrane. To separate conjugated polymer from unreacted Pluronic-L64, the reaction product was processed with HiTrap SP cation exchange chromatography column (GE Healthcare Bio-Sciences, Uppsala, Sweden), washed with five column volumes of the 10% ethanol/H₂O and eluted with 500 mM citric acid. Following 24 h of dialysis with 3.5 kDa MWCO cellulose membrane against 20% ethanol/H₂O, the reaction product was desiccated for 24 h in vacuum dryer, heated at 65 °C in order to remove residual DNAse activity, weighted and dissolved in nuclease-free water.

Production of lentiviral vectors

Lentiviral particles were generated by the simultaneous transient transfection of HEK293T cells with packaging plasmid (pLV-HELP), envelope plasmid (pLV-iVSV-G, for pseudotyping viral particles with pantropic VSV-G protein) and the transfer vector carrying the gene of interest. Packaging and envelope plasmids were purchased from InvivoGen as a part of LENTI-Smart kit, a 2nd generation lentiviral system. HEK293T cells were seeded at a density 8 × 10⁶ cells per 10 cm dish and grown in DMEM (high glucose, 4.5 g/L) supplemented with 10% FBS. 12–16 h after the initial plating, culture media were discarded and replaced with 12 mL of pre-warmed growth medium (DMEM with 10% FBS). HEK293T cells were transfected with 30 μg of plasmid DNA (15 μg of transfer vector, 10 μg of packaging plasmid and 5 μg of envelope-encoding plasmid) using linear polyethylenimine (PEI, 25 kDa, Polysciences, Warrington, PA, USA) at 2:1 PEI to DNA ratio. One day after transfection, culture media were replaced with fresh ones, and additionally supplemented with 10 mM HEPES buffer (pH = 7.2). Cell supernatants containing lentiviral particles were harvested after 24 and 48 h. Cellular debris was removed from the supernatants by centrifugation (2000G, 5 min) and filtration through a 0.45 μm PES (low protein binding) filter. Afterwards, lentiviral particles were concentrated by ultrafiltration with 100 kDa cut-off Amicon centrifugal filter (EMD Millipore, Billerica, MA, USA), aliquoted and stored at −80 °C.

Lentiviral transduction of iPS cells

In order to create pluripotent lines expressing Tet-On transactivator gene, iPS cells were seeded in low density on Geltrex-coated (20 μg/cm²) 6-well plates and cultured in Essential 8 medium. After 24 h, the medium was changed and approximately 100 transduction units of lentiviral particles prepared from pLVX-EF1α-Tet3G transfer plasmid (Clontech) was added to the culture medium. After 2 days, the medium of transduced iPS cells was changed to Essential 8 medium supplemented with 50 μg/mL of G418 (Life Technologies). Cells were maintained in selection medium until non-transduced iPS cells in control well were no longer detected on the plate. G418 resistant iPS cell colonies carrying Tet-On 3G transactivator gene were picked and expanded around day 10 post transduction.

To establish iPSC lines carrying Pdx1 and Nkx6.1 transgene under the control of doxycycline-inducible promoter, iPSC line with stably integrated Tet-on 3G regulatory gene were seeded at low density on Geltex-coated 6-well plate. Next day, cells were transduced with lentiviral vectors carrying either Pdx1-VP16 or Nkx6.1, or the combination of both under the control of doxycycline-regulated TRE3GV promoter. 2 days after the transduction culture medium was supplemented with 20 ng/mL of Puromycin (InvivoGen), and iPS cell colonies were maintained in antibiotic-containing media until non-transduced cells were no longer observed in control wells. 7 days post transduction, puromycin-resistant iPSC colonies were isolated, expanded and examined for doxycycline-induced transgene expression.

Alkaline phosphatase staining

Cells were fixed with 4% PFA in PBS for 20 min. Then, they were washed with TBS twice and stained with 5-bromo-4-chloro-3-indolylphosphate (BCIP) and nitro-blue tetrazolium (NBT) substrate solution (0.02% BCIP, 0.03% NBT, 5 mM MgCl₂ in 150 mM TBS, pH 9.5) at 37 °C until the purple colour emerged, washed with distilled water, and dried.

Flow cytometry

HeLa cells transfected with pCXB-EmGFP plasmid were analysed for green fluorescent protein expression by flow cytometry. 2 days after transfection, cells were detached from cell culture vessels to a single-cell suspension with TrypLE Select (Life Technologies) and analysed using Amnis FlowSight flow cytometer (EMD Millipore). The acquisition was set for 70,000 events per sample, and the data were analysed using IDEAS software (EMD Millipore).

xCELLigence impedance analysis of cells growth and survival

The RTCA xCELLigence (Roche, Basel, Switzerland) was used to determine cell viability. 150 μL of prepared culture medium was added into each well of E-plate 16. The background impedance was measured for 60 s. Fibroblasts and renal epithelial cells were grown on tissue culture vessels prior to the experiment. After reaching 50% confluence, they were rinsed with PBS, and detached from plates by treating them with Accutase. Single cell suspension was prepared and the cells were counted on ADAM-MC Cell System (NanoEnTek, Seoul, Korea). The cell suspension was added to medium-containing wells on E-plate 16 to final density 10,000 cells per well. The adhesion and cell viability was monitored every 60 min for a period of up to 72 h. The electrical impedance was quantified by the RTCA system as Cell index (CI) values. As the electrical impedance depends on the number of cells attached to the electrode and the dimensional changes of the attached cells, CI values represented physiological state of analysed cells.

Immunofluorescence analysis

For immunofluorescence studies, cells were fixed for 20 min at room temperature with 4% paraformaldehyde (PFA) in PBS, and washed once with TBS-T (Tris-buffered 0.9% saline with 0.05% of Tween-20, pH = 7.2). After fixation, cells were permeabilized (for 10 min with TBS-T/0.25% Triton X-100) and blocked with 10% donkey serum in TBS-T. Cells were incubated with primary antibody (1 h, room temperature), washed three times with TBS-T and incubated for 1 h at room temperature with Alexa-conjugated secondary antibodies (Life Technologies) and 4′,6-diamidino-2-phenylindole (DAPI) (50 ng/ml). Coverslips were mounted in 10% polyvinylalcohol in 25 mM Tris–HCl (pH 8.7) with 5% glycerol and 2.5% 1,4-diazobicyclo[2,2,2]-octane. Samples were examined on an Eclipse Ci-S epifluorescence microscope equipped with a DS-5Mc colour CCD camera (Nikon, Tokyo, Japan). Images in blue channel (350/50ex, 400lp, 460/50em), green channel (480/40ex, 510lp, 535/50em), and red channel (560/40ex, 585lp, 630/75em) were acquired with NIS-Elements 4.0 and processed with ImageJ software. Antibodies used for immunofluorescence analysis are listed in Additional file 2: Table S2.

Dithizone stainings

Dithizone stock solution was prepared from 10 mg of dithizone dissolved in 10 mL of DMSO. Cells on the culture dish were fixed with 4% PFA for 15 min at room temperature, rinsed with TBS-T and stained with dithizone working solution (1:100 dilution of the stock in PBS) for 30 min at 37 °C. After that, cells were washed two times with PBS and examined under a microscope.

Quantitative real time PCR

Total RNA was isolated with a NucleoSpinTriPrep kit (Macherey–Nagel, Düren, Germany) according to manufacturer’s protocol. Concentration of nucleic acid samples was measured spectrophotometrically (NanoPhotometer, Implen, Germany) and 250 ng of RNA was reverse-transcribed using random hexamer primers and M-MuLV reverse transcriptase according to manufacturer’s specifications (New England Biolabs). 50 ng of cDNA template along with 200 nM of primer and 6 μL of SYBR Select Master Mix (Life Technologies) was used in reaction. Primer sequences are listed in Additional file 3: Table S3. Quantitative real-time PCR was performed on StepOnePlus™ Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). In order to confirm the specificity of amplification, the melting curve was analysed in each case using StepOne™ Software v2.2.2 (Applied Biosystems). Normalized relative expression level was calculated utilizing the method described previously by Pfaffl [21] based on sample’s average C_T value and PCR efficiency.

Differentiation of iPSC into insulin producing cells

The protocol of iPSC differentiation into insulin producing cells was based on the modified procedure described by Kunisada et al. [13]. Definitive endoderm was obtained using the following method. iPS cells with genetically introduced transgenes were seeded onto a 6-well plate covered with Geltrex or Vitronectin. Cells were maintained in Essential 8 medium for the next 4 days (with daily medium exchange) until they reached 70–80% of confluence. Then, the culture medium was changed to RPMI-1640 supplemented with 100 ng/mL of Activin A, 3 μM of CHIR99021 and 2% foetal bovine serum. iPS cells were maintained in abovementioned medium for 24 h to initiate the differentiation towards endodermal lineage. Afterwards, cells were incubated in RPMI-1640 basal medium supplemented with 100 ng/mL of Activin A and 2% of FBS for the next 2 days.

Definitive endoderm obtained in such manner was further differentiated in iMEM medium supplemented with 10 μM SB431542, 1 μM dorsomorphin, 2 μM retinoic acid and 1% of B27 or defined NS21 supplements (R&D Systems, Minneapolis, MN, USA). The cells were cultured in this medium for 7 days with medium exchange every other day. After that, cells were incubated in iMEM containing 10 μM forskolin, 10 μM dexamethasone, 10 μM nicotinamide, 5 μM Alk 5 inhibitor II and 1% of medium supplement (B27 or NS21). In order to induce TET promoter-controlled transgene expression, 10 ng/mL of doxycycline was added do the IPC differentiation media either through stage III or through stages III and IV.

ELISA assays

Cells grown in culture conditions inductive for pancreatic islets specification were preincubated in L15 medium without insulin and supplemented with 2.5 mM glucose in 37 °C for 4 h. Supernatants were collected, filtered through 0.45 µm filter and used for determination of insulin secretion. For ELISA experiments, wells of 96-well plate (Human Insulin ELISA kit, Sigma-Aldrich) coated with antibodies against human insulin were incubated with 100 µL of analysed supernatants along with insulin concentration standards run in parallels, for 2.5 h with gentle shaking. The whole assay was performed at room temperature. After the incubation, solutions were discarded and wells were rinsed three times with 1× Wash buffer. Afterwards, 100 µL of the detection buffer containing biotinylated antibodies was applied to each well and incubated for 1 h. After the washing step (three times with 1× Wash buffer), bound antibodies were detected with streptavidin-conjugated horse-radish peroxidase (100 µL of the solution with 45 min of incubation). The reaction was visualised by incubation with 100 µL of colorimetric TMB substrate (30 min, development was ended with 50 µL of Stop solution. Absorbance was determined spectrophotometrically at 450 nm on NanoPhotometer (Implen).

Statistical analyses

The reprogramming assays, xCELLigence, qPCR and ELISA experiments were carried out at least three times and presented as the average values ± SEM (standard error of the mean). Difference between samples was compared by the two-tailed Student’s t test and was considered significant at p < 0.05.

Reprogramming of somatic cells to pluripotent state in defined culture conditions

Generation of pluripotent stem cells in conditions that ensure reproducible induction, expansion and their controllable directed differentiation is an indispensable requirement for their therapeutic application. Clinically-relevant iPS cells and differentiated derivatives need to be obtained and maintained in fully defined culture conditions. For this purpose, we aimed to optimize somatic cells culture as well as the step of reprogramming vectors delivery. Different cell culture media were prepared according to compositions shown in Additional files 4 and 5: Tables S4 and S5, and evaluated for their ability to sustain growth of the initial somatic cells. In addition, as control, we tested medium containing foetal bovine serum and commercially available growth media. The impact of medium composition was tested using xCELLigence system that monitors cell proliferation. As shown in Fig. 2a, HFF-1 fibroblasts showed the highest proliferation rate in medium F#3 that contained recombinant albumin, transferrin, EGF, FGF2, PDGF-AB as well as hydrocortisone, sodium selenite and lipid concentrate. For renal epithelial cells, defined medium Ep#5 (containing albumin, transferrin, insulin, EGF, FGF2, PDGF-AB, epinephrine, triiodo-l-thyronine, sodium selenite and chemically defined lipids) provided the highest proliferation rate. However, it needs to be noted that serum-containing media outperformed all abovementioned media in terms of their effect on cell survival and growth.

Since previously used transfection reagent, FuGENE 6 (Promega, Madison, WI, USA) has undisclosed composition, we sought the chemically defined alternative. Recently, copolymer of low-weight polyethylenimine (PEI) and Pluronic polymers has been reported as a low-toxic and efficient DNA delivery agent in established cell lines and primary cells [20]. As the L64-PEI (PCM-04) polymer was not commercially available we conducted conjugation of PEI (M_W = 1200 Da) to PEG-PPG-PEG block polymer (pluronic L64, M_W = 2900 Da) and evaluated its potential as a plasmid delivery carrier. HeLa cells were transfected with pCXB-EmGFP plasmid expressing optimized GFP protein under control of CAGGS promoter using synthesised L64-PEI reagent at 1:3 plasmid to polymer weight ratio. Transfection efficiency was monitored by flow cytometry and epifluorescence microscopy. As shown in Fig. 3b, the number of GFP-positive cells reached 12.8% for cells transfected with L64-PEI, whereas FuGENE 6 transfection reagent provided ~4 times higher efficiency.

In order to better understand the composition of the minimal constituents necessary for the acquisition of pluripotent state by reprogrammed somatic cells, we address the question of the applicability of different protein components of extracellular matrix on the efficiency of pluripotent cells induction. For this purpose, cell culture vessels were coated with murine proteins secreted by Engelbreth-Holm-Swarm (EHS) sarcoma cells (Geltrex, 45 μg/cm²), human recombinant Vitronectin (1 μg/cm²), porcine gelatin (1 μg/cm²), human recombinant Laminin-511 (1 μg/cm²), human recombinant Laminin-521 (1 μg/cm²), human fibronectin (0.5 μg/cm²), human recombinant Collagen I (5 μg/cm²), human Collagen IV (5 μg/cm²) and fractions of laminins isolated from EHS tumour (1 μg/cm²). Induced pluripotent stem cells were derived from adult exfoliated renal epithelial cells and HFF-1 neonatal fibroblasts. These cells were seeded on protein-coated 6-well plates at density 20,000 cells per well (for renal epithelial cells) or 80,000 cells per well (in case of fibroblasts) and transfected with oriP/EBNA-1 episomes as described previously [17]. Reprogramming experiments were carried out for 3–4 weeks till the emergence of compact, tightly packed ESC-like colonies which were fixed with paraformaldehyde and evaluated for the alkaline phosphatase activity. The exemplary phosphatase stainings along with the iPSC colony forming efficiencies in fibroblasts and epithelial cells are presented in Fig. 3. The highest efficiencies of reprogramming of fibroblasts were obtained in case of Laminin-511 (0.08%) and Laminin-521 (0.07%), whereas for human renal epithelial cells the most efficient reprogramming was achieved with Geltrex matrix (1.1%), but also with human, recombinant type I Collagen (0.8%). Altogether, our results confirmed that pluripotency can be induced with application of defined and xeno-free protein components.

Finally, to evaluate polymer utility in derivation of induced pluripotent cells on defined ECM constituents, we plated HFF-1 fibroblasts and urinary epithelial cells on culture vessels coated with recombinant human Collagen I and Laminin-511, respectively. One day later, cells were transfected with 2 μg of episomal plasmids using L64-PEI reagent. The transfection was repeated 24 h later and cells were maintained in TeSR-E7 medium for next 2 weeks. After that, the culture medium was replaced with Essential 8 medium and cells were retained for next 2–3 weeks until the emergence of ESC-like, tightly-packed, clustered colonies. At the end of the experiment, culture plates were fixed with PFA and stained for alkaline phosphatase activity (Additional file 6: Figure S1). The reprogramming efficiencies reached 0.01% for fibroblasts and 0.08% for renal epithelial cells.

Taken together, our results clearly demonstrate that induced pluripotent stem cells can be generated from human somatic cells using fully defined, xeno-free protocols.

Formation of definitive endoderm in the absence of animal-derived components

iPSC were differentiated into definitive endoderm cells using activators of TGFβ and Wnt signalling pathways. Cells were incubated for 24 h in RPMI1640 basal medium containing 100 ng/mL of Activin A and 3 μM CHIR99021. Next, cells were maintained in culture for 2 days in RPMI1640 medium with Activin A. In addition, these media were supplemented with serum or its replacements to extend lifespan of cells. Four culture media were prepared according to compositions shown in Additional file 7: Table S6. Formulated differentiating media contained 2% of bovine (DE1) or human serum (DE2) or in case of DE3 and DE4 media contained a mixture of various defined additives, deprived from the serum. The influence of these media on the process of definitive endoderm formation was studied by qPCR measurements and immunocytochemical stainings. Levels of mRNA encoding protein markers characteristic for endoderm (SOX17, FOXA2, CXCR4) were analysed and are presented in Fig. 4a, along with raw data included in Additional file 8: Table S7. The highest relative expression of marker genes was observed for FBS-containing medium DE1 and defined medium DE4, that included recombinant albumin, insulin, transferrin as well as ascorbic acid and chemically defined lipids.

Presence of definitive endoderm markers was evaluated on protein level as well, using antibodies against E-cadherin, CXCR4 and SOX17. Results of immunocytochemical analysis are shown in Fig. 4b. For all tested media we detected cells with cytoplasmic presence of CXCR4, SOX17 localized in cell nuclei, as well as clusters of cells positive for E-cadherin.

Collectively, these results indicate that in terms of the efficiency and endodermal markers expression, the definitive endoderm can be formed without animal-derived products at level comparable to culture conditions where serum was included.

Insulin secretion is enhanced by PDX1 and NKX6.1 induction during IPC progenitors’ maturation

In this study, two strategies for generation of insulin producing cells were coupled. First method involves forced expression of specific transcription factors in differentiating cells. The second approach includes differentiation of initial stem cells by activation or inhibition of certain signalling pathways by small chemical molecules and protein ligands.

To analyse the impact of Pdx1 and Nkx6.1 transcription factors on the process of derivation of insulin producing cells, we generated doxycycline-regulated pluripotent cell lines. For this purpose, iPS cells were transduced with lentiviral vector carrying Tet-On 3G transactivation gene. After the negative selection with G418, these cells were again transduced with lentiviral particles carrying PDX1-VP16 and NKX6.1 coding sequences under the control of the TRE3G promoter. Following selection with puromycin antibiotic, positive colonies with stably integrated constructs were collected and evaluated for inducible expression by doxycycline. Induced and control samples were analysed by immunocytochemical stainings as shown in Fig. 5a.

Established regulated cell lines were differentiated into definitive endoderm, and further to pancreatic progenitors through the 1 week of incubation in iMEM medium with addition of 1% of B27 or NS21 supplements, 2 μM retinoic acid, 10 μM SB431542, and 1 μM dorsomorphin. For maturation of insulin producing cells, the culture medium was changed to iMEM basal medium supplemented with 1% of NS21, 10 μM forskolin, 10 μM dexamethasone, 10 μM nicotinamide and 5 μM Alk 5 inhibitor II. Cells were maintained in these conditions for 7 days. Expression of PDX1-VP16 and NKX6.1 transgenes or combination of both was induced by including doxycycline in medium for pancreatic progenitors and/or in medium for pancreatic islets specification as shown on diagram in Fig. 1. To evaluate the ability of obtained cells to secrete insulin, IPC cells were incubate in L15 medium supplemented with 2.5 mM glucose. The secretion of insulin hormone was measured with ELISA assay, normalized for total protein concentration and presented in Fig. 5b. The presence of insulin and C-peptide in established IPC was evaluated by immunocytochemical analysis and shown in Fig. 5c.

In order to further determine the identity of differentiated cells, we performed immunocytochemical analysis using antibodies characteristic for β-cells and pancreatic lineage. iPSC lines expressing PDX1-VP16, NKX6.1 and combination of both factors from the doxycycline-regulated promoter were differentiated into insulin producing cells. Expression of transgenes was induced either on the stage of generation of IPC progenitors or during maturation of insulin producing cells. Results of experiments shown in Fig. 6 revealed in all studied samples the presence of cell clusters positive for MafA, Pax6, SLC30A8 and Tyrosine hydroxylase. With use of Somatostatin-specific antibodies we could detect single hormone expressing cells among differentiated cell populations. In addition to SLC30A8 protein detection in ICC assay, we performed stainings with dithizone, a zinc-chelating agent, which indicated the presence of functional zinc transporter ZnT8 in obtained insulin producing cells.

Altogether, generated cells were shown to secrete insulin into the culture medium and displayed molecular markers of pancreatic cells. However, the highest insulin concentration was detected in case of cells with transcription factors induction at stage IV (maturation of Insulin Producing Cells). Furthermore, cells expressing both PDX1-VP16 and NKX6.1 secreted higher level of the hormone, regardless the induction time point.