Abstract
Islet transplantation is a promising treatment for diabetes, but the shortage of donor islets limits its broad application. Induced pluripotent stem cells (iPSCs) provide an alternative source for generating insulin-producing cells; however, whether the somatic cell origin of human iPSCs influences pancreatic endocrine differentiation remains incompletely defined. In this study, we generated iPSCs from human pancreatic duct cells (HD-iPSCs) and compared their differentiation propensity and functional characteristics with human fibroblast-derived iPSCs (HF-iPSCs) under identical differentiation conditions. HD-iPSC-derived cells showed higher expression of pancreatic endocrine and β-cell-associated markers, including insulin, PDX1, and FOXA2, compared with HF-iPSC-derived cells. Flow cytometric analysis further confirmed a higher proportion of insulin-positive cells in differentiated HD-iPSC-derived cells. Functionally, HD-iPSC-derived cells exhibited greater glucose-stimulated C-peptide secretion than HF-iPSC-derived cells, although their secretory capacity remained lower than that of native human islets. Following transplantation into streptozotocin-induced diabetic mice, HD-iPSC-derived cells reduced blood glucose levels more effectively than HF-iPSC-derived cells, and insulin-positive grafts were detected in vivo. These findings suggest that human pancreatic duct cell-derived iPSCs have enhanced pancreatic endocrine differentiation potential compared with fibroblast-derived iPSCs. Although further maturation and optimization are required, pancreatic duct cells may represent a favorable somatic cell source for generating iPSC-derived insulin-producing cells for diabetes cell therapy.
Introduction
One of the major determinants of both type 1 and type 2 diabetes mellitus is loss of mass and/or function of the β-cells in the pancreas. Transplantation of an entire pancreas or pancreatic islets is an effective method to treat diabetes, however, there is severe shortage of available donor tissues. Most studies have demonstrated that the expansion capacity of islets in vitro is limited or not yet achieved such practical scalable expansion methods.1–4 Other sources of β-cells, such as stem cells, bone marrow cells, pancreatic acinar cells, pancreatic duct epithelial cells (duct cells), and α-cells were shown to be able to differentiate into insulin-producing cells.5–7 Among these, human pancreatic duct epithelial cells have the advantage that they can be harvested from pancreatic remnants after islet isolation without difficulty, which are to be dumped out. Recently, it has been reported that pancreatic duct cells themselves have the potential to differentiate into insulin-producing cells in adult mice and humans through endothelial-mesenchymal transition, suggesting the possibility of its hierarchical advantage that could give higher chances to be differentiated into β-cells. 8 Based on these findings, additional approaches were made to utilize the duct cells to pre-clinical stage such as in vitro pancreatic duct cell expansion to enhance its differentiation efficiency. 9 We have previously reported that human pancreatic duct cells could be transdifferentiated into insulin-producing cells by Activin A & Exendin-4 both in vitro & in vivo.10,11 Although human iPSCs can be generated from multiple somatic cell types, increasing evidence suggests that the cell type of origin may influence differentiation propensity, particularly at early passages, through residual epigenetic memory or lineage-associated transcriptional features.12–22 Therefore, the selection of the original somatic cell source may be important when iPSCs are intended for directed differentiation into a specific lineage. Human pancreatic duct cells are developmentally and anatomically related to the pancreatic epithelial lineage and can be obtained from pancreatic remnants after islet isolation. These characteristics raise the possibility that pancreatic duct cell-derived iPSCs may be more amenable to pancreatic endocrine differentiation than iPSCs derived from non-pancreatic somatic cells, such as fibroblasts.
In the present study, we generated human pancreatic duct cell-derived iPSCs (HD-iPSCs) and directly compared their pancreatic endocrine differentiation potential with that of human fibroblast-derived iPSCs (HF-iPSCs) under the same differentiation conditions. We further evaluated insulin-positive cell formation, β-cell-associated gene expression, glucose-stimulated C-peptide secretion, and the ability of differentiated cells to reduce blood glucose levels after transplantation into streptozotocin-induced diabetic mice. This study was designed as a comparative assessment of the influence of somatic cell origin on insulin-producing cell differentiation rather than as an optimization of a differentiation protocol.
Materials and methods
Animals
All animal procedures were performed in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The animal experimental protocol was approved by the Institutional Animal Care and Use Committee of the Samsung Biomedical Research Institute (SBRI), Sungkyunkwan University School of Medicine (Permit No. 000756; approval date: 16 June 2017; approval period: 16 June 2017 to 15 June 2018). BALB/c-nude mice (Orientbio, Sungnam, Gyeonggi-do, Korea) were housed at a constant temperature (23.5 ± 2.0°C) and humidity (50 ± 5%) under a 12:12-h light/dark cycle.
Isolation of human islets
Human pancreata were procured from cadaveric donors (age 35–55 years) through the Korea Network for Organ Sharing (KONOS). Written informed consent for organ/tissue donation and research use was obtained through KONOS in accordance with institutional and national regulations. The use of human pancreatic islets and duct cells for research purposes was approved by the Institutional Review Board of Samsung Medical Center (IRB No. SMC 2005-11-013-007; approval date: 28 November 2005), and the utilization of these cells was restricted to research purposes. Pancreatic islets were isolated by organ distension using intraductal collagenase (Boehringer Mannheim, Ingelheim am Rhein, Germany) and digested using a modified automated method. 23 Islet purification was performed with a COBE 2991 processor (COBE, Lakewood, CA, USA) and a large-scale continuous density gradient 24 prepared from Ficoll-diatrizoate media using the pancreatic digests from cadavers (n=5). After islet isolation, purified islets were incubated for at least 24 to 48 hr at 37°C in culture medium until experiments.25–30
Culture of human pancreatic duct cells
After islet isolation for transplantation, residual cells were collected and plated in M199 media (Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Auckland, NZ) in culture dishes at 37°C in an incubator. 50 μM geneticin (Cellgro, Manassas, VA, USA) and 5 mM streptozotocin (STZ, Sigma, St. Louis, MO, USA) were added to inhibit fibroblast proliferation and to remove residual β-cells, respectively. The STZ and geneticin-containing M199 media (Gibco) were changed every day.
Transfection of reprogramming genes with lentivirus into human duct cells and iPSC culture
1x105 cells/well were infected with lentivirus encoding Klf-4, Sox-2, Oct-3/4 and c-Myc conjugated within 96 well plates. Each lentiviral vector was diluted at 1:100 ratio in human duct cell growth media for 4 hr. Cells were cultured for 1 week, and media was changed with human duct cell growth media and embryonic stem cell media at 1:1 ratio. Infection was tried twice as required. Embryonic stem cell media consisted of DMEM F-12 media (Gibco) with 20% knock-out serum (KSR, Gibco), 1% non-essential amino-acid (NEAA, Gibco), 1% L-glutamine (Gibco) and β-mercaptoethanol (Sigma). With this mixed media, cells were cultured for 1 week. Media was fully changed with embryonic stem cell culture media, and colonized cells were hand-picked and seeded on mouse embryonic fibroblasts (MEF, ATCC, Manassas, VA, USA). IPSCs derived from human skin fibroblast was purchased from WiCell, Wisconsin International Stem Cell Bank (Madison, WI, USA).
Reverse-transcription polymerase chain reaction (RT-PCR)
Human endocrine cell-specific markers & primers.
Human ES specific markers primers.
DNA methylation assay
Methylation primers (bisulfide sequencing).
Embryoid body formation and three germ layer differentiation
Cells were rinsed and collected with Human ES cell embryoid body (EB) formation medium (Chemicon, Tokyo, Japan) to remove all additional cells. Cells were resuspended in 4-5 ml of Human ES Cell EB Formation medium. Then, 1x104 cells/20 μl in embryoid body formation media were seeded by hanging-drop culture method on the cover of 100 mm dish. At the bottom of the dish, PBS was poured to inhibit dehydration. After 3 days, EB was gathered and plated on gelatin-coated or other adherent dishes at different stages for further differentiation. Differentiated cells were analyzed by immunocytochemistry to detect lineage specific markers.
Differentiation of iPSCs into insulin-producing cells
HD-iPSCs and HF-iPSCs were differentiated into insulin-producing cells using the same stepwise differentiation protocol modified from a previously described method, 31 allowing direct comparison of differentiation propensity between the two iPSC sources. The differentiation procedure consisted of four sequential phases over approximately 20–22 days, as summarized in Figure 5(a). In Phase I, cells were induced toward the endoderm lineage for 4 days in DMEM/F-12 medium supplemented with Activin A (50 ng/mL), Wortmannin (1 µg/mL), 0.2% BSA, 0.5% N2 supplement, and 0.5% B-27 supplement. In Phase II, cells were further directed toward pancreatic lineage specification for 4 days in DMEM/F-12 and IMDM mixed at a 1:1 ratio and supplemented with all-trans retinoic acid (2 µg/mL), FGF7 (200 ng/mL), Noggin (50 ng/mL), 0.5% BSA, 0.5% N2 supplement, 0.5% B-27 supplement, and 0.5% ITS. In Phase III, progenitor expansion was induced for 5 days in high-glucose DMEM supplemented with EGF (50 ng/mL), 0.5% BSA, 10% N2 supplement, and 0.5% ITS. In Phase IV, cells were further differentiated toward insulin-producing cells for 7–9 days in DMEM/F-12 medium supplemented with basic FGF (10 ng/mL), nicotinamide (10 mM), Exendin-4 (50 ng/mL), BMP-4 (10 ng/mL), and 1% ITS. At the end of differentiation, cells were analyzed for pancreatic endocrine marker expression, insulin-positive cell formation, and glucose-stimulated C-peptide secretion. The same differentiation schedule, culture conditions, and assay procedures were applied to both HD-iPSCs and HF-iPSCs to ensure direct comparability between the two iPSC sources.
Glucose-stimulated C-peptide secretion assay
Differentiated cells were assessed for glucose-responsive C-peptide secretion using a static glucose-stimulated secretion assay. Briefly, 1 × 10^5 differentiated cells per well were washed twice with Krebs-Ringer bicarbonate HEPES (KRBH) buffer and preincubated in KRBH buffer before glucose stimulation. Cells were then incubated in KRBH buffer containing low glucose (3.3 mM) or high glucose (16.7 mM) for 60 min. Supernatants were collected, and human C-peptide levels were measured using a human C-peptide ELISA kit (Mercodia, Uppsala, Sweden; #10-1172-01) according to the manufacturer’s instructions. C-peptide secretion was normalized to total protein content. Static GSIS was selected to compare endpoint glucose-responsive C-peptide secretion among differentiated human ductal cells, HF-iPSC-derived cells, HD-iPSC-derived cells, and native human islets under identical assay conditions. The same cell number, incubation volume, glucose concentrations, and stimulation time were applied to all differentiated cell groups to minimize assay-related variability.
Immunocytochemistry and histochemistry
Antibodies used in IHC, ICC and western blot.
Flow cytometric analysis of insulin-positive cells
To quantify the proportion of insulin-positive cells derived from iPSCs, differentiated cells were dissociated into single-cell suspensions using TrypLE Express (Thermo Fisher Scientific, Waltham, MA, USA) at 37°C for 5 min. The enzymatic reaction was quenched with culture medium, and cells were centrifuged at 300 × g for 5 min. Cell pellets were washed with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA) for 15 min at room temperature, and permeabilized with 0.1% Triton X-100 in PBS for 10 min at room temperature. After washing with staining buffer (PBS containing 1% bovine serum albumin), cells were blocked with 10% normal goat serum (Thermo Fisher Scientific) for 30 min at room temperature. Cells were incubated with a mouse anti-human insulin primary antibody (Abcam, Cambridge, UK; cat. no. ab7842; 1:100) overnight at 4°C. After washing, cells were incubated with an Alexa Fluor 488-conjugated goat anti-mouse IgG secondary antibody (Thermo Fisher Scientific; cat. no. A-11001; 1:500) for 30 min at room temperature in the dark. Unstained cells and an isotype-matched mouse IgG1 control (Abcam) were used to establish gating thresholds and define background fluorescence. Data acquisition was performed using a BD FACSCanto II flow cytometer (BD Biosciences, San Jose, CA, USA), and at least 10,000 events were recorded per sample. Debris and doublets were excluded based on forward- and side-scatter profiles. Flow cytometry data were exported and analyzed in R using flowCore and ggcyto packages. Insulin-positive cells were defined as events exceeding the isotype control-defined threshold, and the proportion of insulin-positive cells was calculated as a percentage of the gated single-cell population.
Transplantation into the STZ-diabetic mice
Diabetes was induced in male BALB/c-nude mice at 6 weeks of age, by a single intraperitoneal administration of 200 mg/kg body weight of STZ dissolved in 0.05 mmol/L citrate buffer (pH 4.5). After verification of hyperglycemia for 1 week, the left kidneys of the mice were exposed through a small incision on the flank under inhalational isoflurane anesthesia. A small channel was made under the renal subcapsular space with a 26-gauge needle. 1x106 differentiated HD-iPSCs and HF-iPSCs were transplanted into the subcapsular spaces of the kidney. Fasting blood glucose levels were determined using the Accu-Check Active System (Roche, Mannheim, Germany) every other day for 40 days. 32
Statistical analyses
Data were expressed as means ± SEM. Mann–Whitney U test was performed to compare the differences between two independent groups. One-way ANOVA with post hoc analyses was used to compare the differences between several groups. And a two-way analysis of variance or t-test was performed using Prism software 5.0 (GraphPad Software, Inc., San Diego, CA, USA). P values <0.05 were considered statistically significant.
Results
Generation of iPSCs from human pancreatic duct cells
IPSCs were generated according to the scheme in Figure 1(a). Human duct cells were transfected with lentiviral vectors encoding Klf-4, Oct-4, Sox-2 and c-Myc conjugated with GFP, and GFP-positive cells were gathered into colony at this step (Figure 1(b)). 1 week later, colonized cells were transferred onto mouse embryonic fibroblast (MEF) (Figure 1(c)). Rapid expansion of iPSCs from human duct cells (HD-iPSCs) was observed. These cells strongly expressed Klf-4, Oct-4, Sox-2 and c-Myc (Figure 2(a)). They also expressed stem cell markers such as endogenous phosphatase and NANOG, SSEA-4 and TRA-1-60 (Figure 2(b)). ES specific marker expressions from the HD-iPSCs were also significantly increased in HD-iPSCs compared with MEF and human pancreatic duct cells (Figure 3(a) and (b)). The expression levels of these markers were similar to HF-iPSCs (Figure 3(b)). NANOG and OCT4 were unmethylated compared with control group (Figure 3(c)). Karyotyping of HD-iPSCs showed normal diploid karyotypes. Prolonged passages (over 20 passages) did not induce any abnormalities in chromosomes of HD-iPSCs (Figure 3(d)). Generation of iPSCs from human pancreatic duct cells (HD-iPSCs). (a) Schematic overview of HD-iPSC generation. After treatment with geneticin and streptozotocin, cells were transduced with lentiviral vectors encoding Klf4, Oct4, Sox2, and c-Myc, each conjugated with GFP. (b) GFP-positive cells appeared after transduction and formed colonies by day 20 (x200). (c) Colonies were expanded on a mouse embryonic fibroblast (MEF) feeder layer (x100). HD-iPSCs express pluripotency markers and exhibit iPSC-like morphology. (a) Expression of KLF4, OCT4, SOX2, and c-MYC after reprogramming. Hand-picked colonies successfully expressed the transduced genes. (×200) (b) Representative morphology of HD-iPSC lines 15 and 16. Endogenous alkaline phosphatase (AP) activity was detected. (×100, ×200) Pluripotency markers NANOG, SSEA-4, and TRA-1-60 were confirmed by immunostaining. (×100). HD-iPSCs exhibit ESC-like genetic and epigenetic characteristics. (a) HD-iPSCs expressed ESC marker genes and the four reprogramming factors (x100). (b) Relative gene expression profiles. HD-iPSCs showed significantly higher ESC-specific gene expression compared with MEFs and human duct cells, and levels comparable to human fibroblast-derived iPSCs (HF-iPSCs). (c) DNA methylation analysis of OCT4 and NANOG promoters. Methylation levels were presented relative to human duct cells. (d) Representative karyotype of HD-iPSCs showing normal diploid chromosome patterns.
HD-iPSCs differentiated into three germ layers
HD-iPSCs were induced to spontaneously differentiate into embryoid body (EB). After 3 days, these EBs were gathered and plated onto culture dishes for additional analyses (Figure 4(a)). We analyzed the expression of endodermal (α-Fetoprotein), mesodermal (Smooth Muscle Actin and Cardiac Troponin), and ectodermal (MAP2, NESTIN) lineages in differentiated cultures. The embryoid bodies from HD-iPSCs were differentiated into cells of all three germ layers (Figure 4(b)). After 60 days, HD-iPSCs formed teratomas containing various tissues of three germ layers (Figure 4(c)). Neural cord, glandular epithelium, cartilage, follicle, mucin-secreting glands, smooth muscle, adipose tissue, and gut-like epithelium were detected (Figure 4(d)). HD-iPSCs form embryoid bodies and teratomas. (a) Embryoid body (EB) formation after 5 days. EBs aggregated into spherical structures (left) and were transferred to gelatin-coated dishes (right). (b) EBs expressed markers of all three germ layers: endoderm, mesoderm, and ectoderm. (c, d) Sixty days after transplantation, HD-iPSC-derived teratomas contained diverse tissue types, as confirmed by H&E staining.
HD-iPSCs differentiated into insulin-producing cells
We applied a previously described stepwise differentiation protocol,
31
with modifications, to differentiate HD-iPSCs into insulin-producing cells over approximately 20–22 days (Figure 5(a)). At the pancreatic progenitor cell expansion stage, rapid cell proliferation was observed. Cell colonies were expanded more in the last stage of differentiation (Figure 5(b)). By treatment of Activin A and Wortmannin, the specific marker genes of definitive endoderm were expressed in differentiated HD-iPSCs. The expression peaks of SOX17 and FOXA2 were detected in phase 1. At the same time, the expression of OCT3/4 that had a higher expression level was rapidly decreased throughout the period of differentiation induction. NANOG level also showed same pattern with OCT3/4. The expression of SOX17, the marker of definitive endoderm, was high in phase 1. In contrast, the expressions of β-cell-specific markers, such as Insulin, NeuroD, NKX6.1, GLUT2, NGN3, PDX-1, PAX4 and 6 were gradually increased, and maximized in phase 4 (Figure 5(c)). These results were confirmed again with immunostaining, as well. SOX17 and FOXA2, markers of definitive endoderm, were strongly expressed at this stage. In addition, NGN3 was also detected at this phase (Figure 6(a)). In the differentiation phase 2, we observed the co-expression of PDX-1 with FOXA2 and HNF-1β, which is consistent with the observation that the four transcription factors were expressed at the pancreatic progenitor stage in other studies. These data suggest that HD-iPSCs were successfully induced to differentiate into pancreatic progenitors. Treatment with EGF in phase 3 could increase the number of PDX-1-positive cells that were differentiated from HD-iPSCs. In phase 3, PDX-1-positive cells were merged with FOXA2, SOX9 and HNF-1β -positive cells (Figure 6(b)). To further characterize differentiation into insulin-producing cells in phase IV, we examined the expression of pancreatic endocrine hormones and β-cell-associated markers by immunocytochemistry. Insulin- and C-peptide-positive cells were abundantly detected, whereas Glucagon-positive cells represented a smaller population. In addition, few Amylase- or Somatostatin-positive cells were detected, and NKX6.1 expression was observed in a subset of insulin-positive cells (Figure 6(b)). To more quantitatively assess insulin-positive cell formation, we performed anti-insulin staining followed by flow cytometric analysis. Insulin-positive gates were defined using unstained and isotype-matched controls. Flow cytometry showed that HD-iPSC-derived cells contained a higher proportion of insulin-positive cells than HF-iPSC-derived cells (36.1% vs. 24.1%, respectively; Figure 6(c)). These results support the enhanced pancreatic endocrine differentiation propensity of HD-iPSCs compared with HF-iPSCs under the same differentiation conditions. Stepwise differentiation of HD-iPSCs into insulin-producing cells. (a) Schematic representation of the four-phase differentiation protocol used to generate insulin-producing cells. The protocol consisted of endoderm induction, pancreatic specialization, progenitor expansion, and maturation phases over approximately 20–22 days. Key culture media, supplements, and growth factors used in each phase are indicated. (b) Morphological changes during differentiation of HD-iPSCs into insulin-producing cells. Cell populations changed dynamically during the sequential differentiation phases. (c) Changes in the expression of pluripotency, endodermal, pancreatic progenitor, and β-cell-associated genes during differentiation. Pluripotency markers decreased during differentiation, whereas pancreatic endocrine lineage-associated markers increased progressively. Differentiated HD-iPSCs express pancreatic endocrine and β-cell-associated markers. (a) Immunocytochemical analysis of pancreatic lineage markers during differentiation. Definitive endoderm and pancreatic progenitor-associated markers, including SOX17, FOXA2, NGN3, SOX9, PDX1, and HNF1B, were detected during differentiation. Scale bars are indicated in each panel. (b) Immunocytochemical analysis of pancreatic endocrine and exocrine markers at the final stage of differentiation. Insulin- and C-peptide-positive cells were detected, whereas Glucagon-, Amylase-, and Somatostatin-positive cells represented smaller populations. NKX6.1 expression was detected in a subset of insulin-positive cells. Scale bars are indicated in each panel. (c) Flow cytometric analysis of insulin-positive cells in HF-iPSC- and HD-iPSC-derived cells. Unstained and isotype-matched controls were used to define the insulin-positive gate. Representative flow cytometric histograms are shown in panel C. (d) Quantification of insulin-positive cells determined by flow cytometry. HD-iPSC-derived cells showed a higher proportion of insulin-positive cells than HF-iPSC-derived cells (36.1% vs. 24.1%). (e) Relative proportions of insulin-, C-peptide, and glucagon-positive cells in differentiated HD-iPSC-derived cells.
To evaluate the glucose responsiveness of HD-iPSCs in vivo, differentiated HD-iPSCs and HF-iPSCs were transplanted into streptozotocin (STZ)-induced diabetic nude mice (n = 5 per group). Fasting blood glucose levels gradually decreased beginning 8 days after transplantation and remained reduced until removal of the grafts at day 40 (Figure 7(a)). Notably, HD-iPSC transplantation resulted in a significantly earlier reduction in blood glucose levels compared with HF-iPSCs (Figure 7(a)). To verify engraftment and insulin production in vivo, kidney sections containing transplanted cells were analyzed by immunofluorescence staining for insulin. Strong insulin-positive signals were detected in grafts derived from both HD-iPSCs and HF-iPSCs, whereas no insulin staining was observed in the STZ-only control group (Figure 7(b)). Functional superiority of differentiated HD-iPSCs compared with HF-iPSCs. (a) Blood glucose levels following transplantation of differentiated iPSCs into STZ-induced diabetic mice. Blood glucose levels increased again after graft removal at day 40 (#p < 0.05 HD-iPSC group vs. HF-iPSC group). (b) Immunofluorescence staining for insulin in kidney grafts after transplantation. Strong insulin-positive signals were observed in grafts derived from HD-iPSCs and HF-iPSCs, whereas no staining was detected in the STZ-only group (×400). (c) Glucose-stimulated C-peptide secretion at low (3.3 mM) and high (16.7 mM) glucose concentrations. Human islets were isolated from five independent donors (n = 5). (*p < 0.005 vs. 3.3 mM glucose; **p < 0.05 vs. differentiated duct cells and HF-iPSCs at 16.7 mM glucose; §p < 0.05 vs. 3.3 mM glucose in duct cells and HF-iPSCs; †p = 0.0146 vs. 16.7 mM glucose of HD-iPSCs). (d) Relative expression of Insulin and PDX-1 after differentiation. (*p < 0.05 vs. differentiated ductal cells; **p < 0.05 vs. HD-iPSCs; †p < 0.05 vs. HF-iPSCs). (e) Differentiated HD-iPSC-derived cells showed higher expression levels of insulin, PDX1, and FOXA2 compared with HF-iPSC-derived cells (*p < 0.01).
To further assess functional insulin secretion capacity, glucose-stimulated C-peptide secretion was measured under low-glucose (3.3 mM) and high-glucose (16.7 mM) conditions. Differentiated HD-iPSC-derived cells exhibited significantly greater glucose-stimulated C-peptide secretion than HF-iPSC-derived cells (HD-iPSCs: 6.54 ± 0.38 ng/mg protein vs. HF-iPSCs: 5.25 ± 0.64 ng/mg protein, p < 0.05). However, the C-peptide secretion capacity of HD-iPSC-derived cells remained lower than that of native human islets (Figure 7(c)).
We next compared the expression of representative β-cell-associated genes after differentiation. HD-iPSC-derived cells showed higher expression levels of insulin and PDX1 compared with HF-iPSC-derived cells and differentiated ductal cells, although expression remained lower than that in native human islets (Figure 7(d)). In addition, differentiated HD-iPSC-derived cells showed 1.81-fold, 1.90-fold, and 2.6-fold higher expression of insulin, PDX1, and FOXA2, respectively, compared with HF-iPSC-derived cells (Figure 7(e)).
Discussion
In the present study, we generated iPSCs from human pancreatic duct cells and compared their pancreatic endocrine differentiation potential with that of fibroblast-derived iPSCs under identical differentiation conditions. HD-iPSC-derived cells showed higher expression of pancreatic endocrine and β-cell-associated markers, a greater proportion of insulin-positive cells by flow cytometry, increased glucose-stimulated C-peptide secretion, and improved blood glucose-lowering effects after transplantation compared with HF-iPSC-derived cells. These findings support the concept that human pancreatic duct cells may represent a favorable somatic cell source for generating iPSCs with enhanced propensity toward insulin-producing cell differentiation.
The primary aim of this study was not to optimize a differentiation protocol, but to directly compare HD-iPSCs and HF-iPSCs under the same experimental conditions. We therefore used an established stepwise differentiation protocol for both iPSC sources to minimize protocol-dependent variability and to allow direct comparison of differentiation propensity. Substantial progress has been made in the directed differentiation of human pluripotent stem cells into pancreatic progenitors and stem cell-derived β-like cells. 31–47Key developmental signals, including Activin/Nodal, retinoic acid, FGF, BMP, TGF-β, and Hedgehog pathways, have been used in stage-specific combinations to guide definitive endoderm, pancreatic progenitor, endocrine precursor, and β-like cell maturation. Compared with more recent differentiation strategies, the protocol used in the present study may not fully optimize β-cell maturation. Nevertheless, because the same protocol was applied to both HD-iPSCs and HF-iPSCs, the observed differences are informative for comparing the relative differentiation propensity of the two iPSC sources.
The enhanced differentiation propensity of HD-iPSCs may be related, at least in part, to the pancreatic origin of the parental cells. Previous studies have shown that donor cell type can influence the epigenome and differentiation potential of iPSCs, particularly at earlier passages, and that residual epigenetic memory may bias iPSCs toward lineages related to their cell type of origin.48–51 In the present study, HD-iPSC-derived cells showed higher expression of insulin, PDX1, and FOXA2 than HF-iPSC-derived cells after differentiation. These findings are consistent with the possibility that pancreatic duct cell-derived iPSCs may retain lineage-associated features that facilitate pancreatic endocrine differentiation. However, because HD-iPSCs and HF-iPSCs were derived from different donor sources, we cannot exclude the contribution of donor-dependent genetic or epigenetic differences. Future studies using multiple donor-matched iPSC lines will be required to distinguish the effects of cell type of origin from donor-specific variability.
Although HD-iPSC-derived cells showed improved β-cell-associated marker expression and glucose-responsive C-peptide secretion compared with HF-iPSC-derived cells, their functional capacity remained lower than that of native human islets. This finding is consistent with previous reports showing that insulin-producing cells differentiated from human pluripotent stem cells often exhibit incomplete functional maturation compared with adult human β-cells.40,45,47,52–54 In this regard, HD-iPSC-derived cells should be interpreted as insulin-producing β-like cells rather than fully mature human β-cells. Additional optimization of differentiation conditions, maturation culture, three-dimensional aggregation, and transplantation strategies will be required to further improve their functional maturation and therapeutic potential.
A limitation of the present study is that glucose-stimulated C-peptide secretion was evaluated using a static incubation assay. This approach allowed direct endpoint comparison among differentiated human ductal cells, HF-iPSC-derived cells, HD-iPSC-derived cells, and native human islets under identical conditions. However, dynamic perifusion analysis would provide more detailed information regarding biphasic secretion kinetics and potential diffusion-related effects, particularly in clustered or aggregate cultures. 55 Therefore, future studies using dynamic perifusion assays will be required to more precisely evaluate the functional maturation of HD-iPSC-derived insulin-producing cells.
Another limitation is that the present study used an established differentiation protocol rather than a more recently optimized human pluripotent stem cell-derived β-cell protocol.46,47 This design was chosen to compare the differentiation propensity of HD-iPSCs and HF-iPSCs under identical conditions, not to establish a maximally efficient differentiation method. Future studies applying updated differentiation protocols and larger numbers of independent iPSC lines will be important to validate whether pancreatic duct cell-derived iPSCs consistently show enhanced pancreatic endocrine differentiation potential.
In summary, HD-iPSCs generated from human pancreatic duct cells showed enhanced differentiation into insulin-producing cells compared with HF-iPSCs, as demonstrated by higher β-cell-associated gene expression, increased insulin-positive cell proportion, improved glucose-stimulated C-peptide secretion, and greater blood glucose-lowering effects after transplantation. Although the differentiated cells remained functionally immature compared with native human islets, these findings suggest that pancreatic duct cell-derived iPSCs may provide a useful platform for generating insulin-producing cells and for further optimization of β-cell replacement strategies.
Conclusion
Human pancreatic duct cell-derived iPSCs were successfully generated and differentiated into insulin-producing cells under the same conditions used for fibroblast-derived iPSCs. Compared with HF-iPSC-derived cells, HD-iPSC-derived cells showed higher pancreatic endocrine marker expression, a greater proportion of insulin-positive cells, enhanced glucose-stimulated C-peptide secretion, and improved blood glucose reduction after transplantation into STZ-induced diabetic mice. However, their functional capacity remained lower than that of native human islets, indicating that further maturation and protocol optimization are required. These findings suggest that human pancreatic duct cells may represent a favorable somatic cell source for generating iPSC-derived insulin-producing cells for diabetes cell therapy.
Footnotes
Ethical considerations
The use of human pancreatic islets and duct cells for research purposes was approved by the Institutional Review Board of Samsung Medical Center (IRB No. SMC 2005-11-013-007; approval date: 28 November 2005). Human pancreatic tissues were obtained from cadaveric donors through the Korea Network for Organ Sharing (KONOS) in accordance with institutional and national regulations. All animal experiments were approved by the Institutional Animal Care and Use Committee of the Samsung Biomedical Research Institute (SBRI), Sungkyunkwan University School of Medicine (Permit No. 000756; approval date: 16 June 2017; approval period: 16 June 2017 to 15 June 2018).
Consent to participate
Written informed consent for organ/tissue donation and research use of cadaveric donor pancreatic tissues was obtained through the Korea Network for Organ Sharing (KONOS) in accordance with institutional and national regulations.
Author contributions
S.B.L. and M.K.L. conceived and designed the study. S.B.L. acquired the data. S.B.L. and M.K.L. analyzed and interpreted the data. S.B.L. drafted the manuscript, and M.K.L. critically revised it. Both authors read and approved the final manuscript.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a grant from the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (Grant No. A084065).
Declaration of conflicting interests
The authors declare no competing interests.
Data Availability Statement
The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.