Close this search box.

Forward programming of hiPSCs towards beta-like cells using Ngn3, Pdx1, and MafA – Scientific Reports

Generation of inducible lentiviral constructs

For the inducible expression of our markers of interest, two lentiviral constructs were generated. The first lentiviral construct, pLenti_EF1a-rtTA_BsdR was generated as follows: The sequence for EF1a-rtTA, flanked with attB1 and attB2 sites, was synthesized and subsequently shuttled into a pDONR221 backbone through ThermoFisher, yielding an entry vector. Gateway Cloning was then used between this construct and pLenti6.3_V5-DEST_promoterless_mcs_BsdR conferring Blasticidin resistance to generate pLenti_EF1a-rtTA_BsdR.

The second lentiviral construct, pLenti_TRE-NPM_HygR, was generated as follows: The markers of interest, NGN3 (GenBank ID 50,674), PDX1 (GenBank ID 3651), and MAFA (GenBank ID 389,692) were designed downstream of the TRE3G promoter. Multicistronic expression of the markers of interest was achieved using the self-cleaving 2A-peptides T2A and P2A. This sequence was synthesized through ThermoFisher within a pENTR221 backbone and recombined with pLenti6.3_V5-DEST_promoterless_mcs_HygR conferring HygromycinB resistance to generate pLenti_TRE-NPM_HygR. Final construct maps (Supp.Fig. 1a,c) and detailed cloning information are available upon request.

Lentivirus production and transduction

The lentiviruses were produced separately using human embryonic kidney 293 T cells. Cells were cultured in DMEM + GlutaMax (Gibco, 31,966–021), 10% FCS (Gibco, 10,270–106), 1% NEAA (Gibco, 11,140–035), 1% L-Glutamine (Gibco, 25,030–024) and 1% Pen/Strep (Gibco, 15,140,122). At 80% confluency, the cells were transfected overnight with the lentiviral constructs along with Lipofectamine 2000 Reagent (Invitrogen, 11,668–500) and Ready-to-use Lentiviral Packaging Plasmid Mix (Cellecta, CPCP2KA) in OptiMEM + GlutaMax (Gibco, 51,985–026), 5% FCS, and 1% Pen/Strep. Medium was changed the next day, and viruses were harvested each day for the next two days and stored at − 80 C. The virus was then centrifuged at 19,600 rpm at 4 °C for 80 min. Supernatant was removed and 100 µL of DPBS + 1% BSA was pipetted on top of the pellet. The tube was then sealed with parafilm and incubated at 4 °C overnight before the pellet was resuspended, and aliquoted. The lentiviral titre was determined using a p24 ELISA.

hiPSCs39 were maintained on Matrigel (Corning, 356,231)-coated dishes in mTesR + (StemCell Technologies, 100–1130) medium supplemented with 1% Pen/Strep, and 50.0 µg/mL Normocin (InvivoGen, ant-nr-1). hiPSC were transduced using the EF1a-rtTA_BsdR lentivirus, and successful recombinants were selected for using 1.0 µg/mL of Blasticidin S HCl (Gibco, A1113902) for five days, now referred to as rtTA-hiPSCs. Subsequently, rtTA-hiPSCs were transduced using the TRE-NPM_HygR lentivirus, and successful recombinants were selected for using 25.0 µg/mL of HygromycinB (Gibco, 10,687,010) for six days, now referred to as rtTA-NPM hiPSCs. The rtTA-NPM hiPSCs were maintained on Matrigel-coated dishes in mTesR + medium supplemented with 1.0 µg/mL Blasticidin, 25.0 µg/mL HygromycinB, 1% Pen/Strep, and 50.0 µg/mL Normocin.

Doxycycline induction towards the pancreatic lineage

To begin the induction towards the pancreatic lineage, rtTA-NPM hiPSCs were cultured in medium supporting β-cell differentiation14 with and without 1.0 µg/mL doxycycline hyclate (Sigma Aldrich, D9891-1G). The medium recipe is as follows: MCDB131 (Cellgro, 15–100-CV), 3.6 mg/mL D-( +)-Glucose (Sigma, G7528), 1.75 mg/mL NaHCO3 (Sigma, S5761), 20.0 mg/mL FAF-BSA (Sigma, A9576), 1% Glutamax (LifeTech, 35,050–061), 0.5% ITS-X (Invitrogen, 51,500,056), 0.25 mM Vitamin C (Sigma, A4544), 100 nM Retinoic Acid (Sigma, R2625), 1.0 µM T3 (EMD Millipore, 64,245), 20.0 ng/mL Betacellulin (Peprotech, 100–50), 10.0 µM Y27632 (Abcam, ab120129), 0.25 µM Sant1 (Sigma S4572), 10.0 µg/mL Heparin (Sigma H3149), 10.0 µM Alk5 II (Axxora ALX-270–445), 1.0 µM XXI (EMD Millipore 565,790), 1% Pen/Strep, and 50.0 µg/mL Normocin.

During the induction, cells were cultured for five days as a monolayer with daily medium changes. At the end of the five days, 3D spheroids were generated by creating a single cell suspension of the cultures and seeding 2000 cells / well in a 96-well u-bottom plate (Greiner Bio-One, 650,101). The spheroids were cultured for five more days without further medium changes for a total culture period of ten days. All results presented are from at least three separate inductions.

RNA extraction and RT-qPCR

The Qiagen RNeasy Mini Kit (Qiagen, 74,101) was used for RNA extraction of samples and DNase (Promega, M6101) treatment was performed according to manufacturer’s protocol. cDNA synthesis was performed using M-MuLV Reverse Transcriptase (New England Biolabs, M0253S), and qPCR was run using the Universal PCR Master Mix (Applied Biosystems, 4,304,449). The list of primers used can be found in Supplementary Table 1. Ready-to-use mixtures of primers and probes (Gene Expression Assays, Thermo Fisher, 4,331,182) were used unless otherwise specified. Gene expression was normalized to the housekeeper gene GAPDH. Relative mRNA levels (2^ΔΔCt) of the transduced rtTA-NPM hiPSCs were determined in relation to the un-transduced hiPSCs which was represented with a value of 1. Fold change of mRNA levels (ΔΔCt) of the cells cultured with and without DOX were determined in relation to that of the rtTA-NPM hiPSCs at the start with a timepoint 0 (TP0).

Ethical statement for the study of human tissue

Studies with human fetal tissue and human adult donor islets were conducted with approval from the Ethics Committee at the Medical Faculty of the Eberhard Karls University Tübingen and at the University Hospital in Tübingen in accordance with the ICH-GCP guidelines (IRB #406/2011BO1 and #290/2016BO1). Human islets for research were provided by the Alberta Diabetes Institute IsletCore at the University of Alberta in Edmonton ( Islet isolation was approved by the Human Research Ethics Board at the University of Alberta (Pro00013094). All donors’ families gave informed consent for the use of pancreatic tissue in research.

Human donor islet culture and static GSIS

Healthy human donor islets were purchased from the Alberta Diabetes Institute IsletCore (Alberta, Canada). Individual donor islets were handpicked into wells of 96well U-bottom non-adherent well plates (Greiner Bio-One, 650,101) and cultured for three days to recover from shipping conditions in CMRL1066 without Glutamine (Gibco, 21,530–027), 0.5% FAF-BSA (Sigma Aldrich, A9576), 1 g/L glucose (Gibco, A24940-01), 4 mM Glutamax (Gibco, 35,050–061), 1% Pen/Strep (Gibco, 15,070–063), and 100 µg/mL Normocin (InvivoGen, ant-nr-1).

For the static GSIS assays, donor islets were incubated for one hour in β-Krebs with 2 mM glucose. Donor islets were then incubated in β-Krebs buffer with either 2 mM glucose or 20 mM glucose for another hour, following which the supernatant was collected and stored at -20 °C. The donor islets were lysed with acid ethanol overnight at 4 °C, collected, and stored at -20 °C for insulin content analysis. The levels of insulin secreted by the donor islets and the insulin content within the donor islets were analyzed using the human insulin ELISA kit (Mercodia, 10–1132-01). The GSIS index was calculated by dividing each samples’ insulin secretion during 20 mM glucose by 2 mM glucose.

Immunofluorescence (IF) staining and quantification

Following the five-day monolayer culture, cells were prepared for IF staining of the markers of interest. Cells were washed twice with PBS, fixed in 4% PFA for 20 min at room temperature (RT), permeabilized with 1% Triton-X for 30 min at RT, blocked with Goat Block for 30 min at RT, and incubated with the primary antibody overnight at 4 °C (see Supplementary Table 2 for list of antibodies used). Cells were washed and first incubated with the secondary antibody for 30 min at RT in the dark, and then with 2 µg/ml of 4′,6-Diamidin-2-phenylindole (DAPI) solution for 10 min at RT in the dark. Samples were mounted with Prolong Gold Antifade Mountant (Thermo Fisher Scientific, P36930) and imaged using the laser scanning microscope 780 (Carl Zeiss GmbH, Jena, Germany).

The 10-day spheroids, adult donor islets (from the McDonald Laboratory, Alberta, Canada), and 10-week-old fetal tissues (from the University of Tübingen, Tübingen, Germany) were prepared for IF staining, as previously described40. Briefly, samples were washed with PBS, fixed in 4% PFA, and embedded in paraffin with a Shandon Citadel 1000 (Thermo Fisher Scientific, Waltham, MA, USA). Samples were sectioned into 3 µm sections (Microtome RM2145, Leica, Nussloch, Germany), and deparaffinized with xylene, graded ethanol (100−50%), and VE-water. Antigen retrieval was performed using both Tris–EDTA (pH 9.0) and Citrate (pH 6.0) buffers. Primary antibody incubation was performed overnight at 4 °C, followed by secondary antibody incubation for 30 min at RT in the dark, with either DAPI solution for 10 min or DRAQ5 (ThermoFisher, 62,251) for 30 min at RT in the dark for nuclear staining. Samples were treated with the Vector® TrueVIEW® Autofluorescence Quenching Kit (Vector Labs, SP-8400) for 1 min at RT before mounting with Prolong Gold Antifade Mountant. Images were obtained using the laser scanning microscope 780. Positive nuclear staining of the markers of interest was manually counted and normalized per DAPI or DRAQ5 nuclear count. Gray value intensity (GVI) was determined per channel for each image, subtracted by the value for the negative control, and then normalized per DAPI or DRAQ5 nuclear count.

Glucose-stimulated insulin secretion (GSIS) assays

GSIS assays were performed in β-Krebs buffer (recipe can be found in Supplementary Table 3) supplemented with glucose. Cells were washed with PBS and synchronized in β-Krebs buffer with 2 mM glucose (low glucose condition) for one hour in the incubator. Following the synchronization step, cells were either incubated in β-Krebs buffer with either 2 mM glucose or 20 mM glucose (high glucose condition) for one hour in the incubator for the static GSIS assays. For dynamic GSIS assays, the spheroids were incubated in β-Krebs buffer with either 2 mM glucose for an hour followed by buffer with 20 mM glucose for one hour, and then again in buffer with 2 mM glucose for an hour in the incubator. Supernatant was then collected at the end of each incubation and stored at -20 °C overnight before performing either an insulin ELISA (Mercodia, 10–1132-01) or C-peptide ELISA (Mercodia, 10–1141-01). Cells were then trypsinized and counted for normalization of insulin secretion per cell count or per spheroid size.


Data presented is collected from at least three separate induction experiments where the data is presented as mean ± standard deviation (s.d.). Outliers were identified with Grubb’s test (p ≤ 0.05). Unpaired t-tests were used to analyze statistical differences between groups.