Neuronal Restrictive Silencing Factor Silencing Induces Human Amniotic Fluid-Derived Stem Cells Differentiation into Insulin-Producing Cells

Abstract

Islet cell replacement represents the most promising approach for the treatment of type I diabetes. However, it is limited by a shortage of pancreas donors. Here, we report that human amniotic fluid-derived stem cells (hAFSCs) can be induced to differentiate into functional insulin-producing cells by knocking down neuronal restrictive silencing factor (NRSF). In this study, lentiviral vectors were used to deliver small interference NRSF (siNRSF) RNA into hAFSCs. After infection with lentivirus containing siNRSF, hAFSCs were successfully induced to differentiate into insulin-producing cells. The differentiated siNRSF-hAFSCs expressed genes specific for islet cells, such as Pdx1, Hnf4α, Isl-1, Nkx6.1, Insulin, and Glut2. These cells also produced and released C-peptide in a glucose-responsive manner. These findings indicated that hAFSCs could be induced to differentiate into insulin-producing β-like cells by NRSF silencing.

Introduction

I slet cell transplantation for patients with type I diabetes is a promising approach for achieving insulin independence [1]. However, the extreme shortage of matched organ donors has limited the efficacy of this approach. This shortage has evoked a large-scale search for alternative sources of islet cells. Recent studies have shown that insulin-secreting cells can be generated from embryonic stem cells [2], induced pluripotent stem cells [3], or adult stem cells [4]. However, teratoma formation of embryonic stem cells, the scarcity of these cells, and the invasive procedures required to isolate adult stem cells have limited their use. As a result, human amniotic fluid-derived stem cells (hAFSCs) are attracting attention as a possible alternative therapy. hAFSCs are easily attained through routine amniocentesis, which is a common and safe screening tool used to study the human fetus. It has been reported that hAFSCs have strong proliferative capacity and can be differentiated into a wide range of cell types [5]. However, Trovato et al. reported that under the experimental conditions used in their study, the subcultured AFSCs did not differentiate into β cells [6]. The ability of hAFSCs to differentiate into functional islet β cells needs to be examined.

Neuronal restrictive silencing factor (NRSF), also known as repressor element-1 silencing transcription factor, is a transcriptional regulator that interacts with gene networks involved in the self-renewal of embryonic stem cells and neuronal development [7]. NRSF was first recognized for its ability to bind to a specific DNA sequence and block the transcription of several neuronal differentiation genes in nonneuronal cells or neural stem/progenitor cells [8]. Several studies have suggested that NRSF also negatively regulates the expression of neural genes in islet cells [9]. Recently, more NRSF target genes including key transcription factors such as Pax4, NeuroD1/Beta2, HNF4α, and Neurogenin3, which regulate islet cell development, have been reported [10,11]. These data suggest that NRSF may also function as an important regulator during islet cell differentiation. Further studies demonstrate that down-regulation of NRSF may be required for proper islet development, because overexpression of NRSF in islet cells results in fewer islet cells, reduced insulin production, and glucose intolerance [12]. Thus, in differentiated insulin-producing cells, NRSF expression should be avoided [13] to permit the full expression of its target genes that encode islet development-related and exocytosis-related proteins. Accordingly, we hypothesized that silencing NRSF expression would induce hAFSCs differentiation toward islet β cells.

In these studies, we constructed a lentiviral-mediated RNA interference vector that stably downregulated the expression of NRSF in hAFSCs. We found that NRSF-silenced hAFSCs differentiated into insulin-producing cells. These differentiated cells expressed islet cell-specific genes, such as Pdx1, Isl-1, Nkx6.1, Glut2, and Insulin. The function of these cells was confirmed by measuring C-peptide synthesis and release upon glucose challenge. These results suggest that hAFSCs could be induced to differentiate into insulin-producing β-like cells by NRSF silencing.

Materials and Methods

Isolation, culture, and phenotype analysis of hAFSCs

Amniotic fluid samples (3–5 mL) were obtained after informed consent from pregnant women undergoing amniocentesis (weeks of gestation 20 ± 2) at the Beijing Hospital of Integrated Traditional and Western Medicine. The samples were centrifuged at 1,600 rpm for 5 min, and the pellets were resuspended in low-glucose Dulbecco's modified Eagle's medium (DMEM; Invitrogen) containing 12% fetal bovine serum (FBS; Hyclone), 4 ng/mL basic fibroblast growth factor (bFGF; R&D Systems), and 1% penicillin/streptomycin at 37°C in 5% CO₂. Cells were cultured in 6-well plates for 2 weeks to form primary cell clones. The medium was not replaced during the first week. Then they were harvested using trypsin and plated into tissue culture plastic plates at a density of 6 × 10³ cells/cm².

Phenotype analysis of hAFSCs was performed using flow cytometry. The cells were incubated with antibodies for CD29 (BioLegend), CD34 (R&D Systems), CD38 (BioLegend), CD41 (PharMingen), CD49 (Serotec), CD80 (PharMingen), CD90 (R&D Systems), CD105 (BioLegend), CD144 (eBioscience), HLA-DR (BioLegend), HLA-ABC (BioLegend), and SSEA-4 (R&D Systems) conjugated to either FITC or PE. Mouse IgG conjugated to FITC or PE was used as a negative control.

Immunofluorescence analysis

Cells were seeded in 48-well plates, fixed with phosphate-buffered saline containing 4% paraformaldehyde for 20 min at room temperature, and blocked with phosphate-buffered saline containing 10% normal donkey serum, 0.1% Triton X-100, and 1% bovine serum albumin for 30 min at room temperature. The cells were then incubated with primary antibody overnight at 4°C and further incubated with secondary antibody (FITC- or TRITC-conjugated goat anti-mouse or anti-rabbit IgG; TRITC-conjugated rabbit anti-goat IgG; R&D Systems). The primary antibodies used in these experiments were anti-Oct4 (mouse IgG; Abcam), anti-Nanog (rabbit IgG; R&D Systems), anti-SSEA-4 (mouse IgG; R&D Systems), anti-Pdx1 (goat IgG; R&D Systems), anti-Insulin (mouse IgG; R&D Systems), and anti-C-peptide (rabbit IgG; R&D Systems). Images were captured using an Olympus phase-contrast fluorescent microscope (Olympus IX-71).

Small interfering RNA vector production

Small interfering RNA (siRNA) vectors containing the NRSF sequence or a scrambled sequence were prepared as previously described [14]. Following our previous method, we constructed another siRNA vector that targets the human NRSF gene with the targeting sequence GTGTAATCTACAGTATCAC [15]. Primers (N19)TTCAAGAGA(19N)TTTTTTC and TCGAGAAAAAA(N19)TCTCTTGAA(19N)A were designed to allow cloning into the vector pSicoR, which also expresses enhanced green fluorescent protein (EGFP) under the control of the cytomegalovirus (CMV) promoter. Primers were annealed and cloned under the control of the U6 promoter in the vector pSicoR.

Lentivirus generation and infection

Lentiviruses were generated as previously described [16]. Briefly, 293FT cells were transfected with pLP1, pLP2, pLP/VSVG, pSicoR-GFP-small interference NRSF (siNRSF), or pSicoR-GFP-siControl (scrambled control) using Lipofectamine 2000. The supernatants were collected at 48–72 h after transfection and used to infect AFSCs. Lentiviruses containing the NRSF siRNA sequence or the scrambled sequence were termed Lv-siNRSF and Lv-siControl, respectively. hAFSCs infected with Lv-siNRSF were named siNRSF-hAFSCs, whereas hAFSCs infected with Lv-siControl were named siControl-hAFSCs.

Differentiation

hAFSCs infected with Lv-siNRSF or Lv-siControl were planted in 6-well plates at a density of 1.0 × 10⁴ cells/cm² for differentiation using a 2-stage protocol. In the first stage, cells were seeded in DMEM/F12 (Gibco) supplemented with 2% FBS, 2% B27 (Gibco), 10 ng/mL bFGF, and 10 ng/mL Activin-A (Pepro Tech) [17] for 7 days. In the second stage, the culture medium was changed to DMEM/F12 containing 2% FBS, 2% B27, 10 ng/mL Activin-A, 10 ng/mL betacellulin, and 10 mM nicotinamide for the next 7 days.

Reverse transcription–polymerase chain reaction

Total RNA was isolated using Trizol Reagent (Invitrogen) according to the manufacturer's protocol, and 500 ng of total RNA was reverse-transcribed into cDNA using M-MLV reverse transcriptase (TaKaRa) in a 10 μL reaction mixture. Polymerase chain reaction (PCR) was performed with rTaq polymerase (TaKaRa) following the manufacturer's protocol. The cycle conditions were as follows: 94°C for 5 min followed by 35 cycles (94°C denaturation for 50 s, 56°C–60°C annealing for 30 s, 72°C elongation for 40 s), with a final incubation at 72°C for 4 min. The primers are listed in Table 1.

Table 1.

Description of Primers Used for Reverse Transcription–Polymerase Chain Reaction and Quantitative Reverse Transcription–Polymerase Chain Reaction Analyses

Gene	Forward primer (5′ to 3′)	Reverse primer (5′ to 3′)
Foxa2 ^a	TTCTTTCCCGTTTTCCTCCT	AGACGGTGTTGCAGAGACG
Hnf4a ^a	ACTACATCAACGACCGCCAGT	ATCTGCTCGATCATCTGCCAG
Isl-1 ^b	ATTTCCCTATGTGTTGGTTGCG	CGTTCTTGCTGAAGCCGATG
pdx1 ^a	AAGCTCTGATTGGTTGTTGA	CTGATGGAGTAACCCATTGA
Glut2 ^a	GAATAACATTGCCAAACGTC	AAGAACCATCAGCATGTC
Nkx6.1 ^b	ACACGAGACCCACTTTTTCCG	TGCTGGACTTGTGCTTCTTCAAC
Insulin ^a	GATGGAAAATCTCACAGAAGGG	AGGTCCCCTTCTAGAATTCTGG
Pax4 ^a	GGCTGTGTGAGCAAGATCCTAGGA	TTGCCAGGCAAAGAGGGCTGGAC
NRSF ^a	ATTGAAGTTGGCTTAGTG	TATGGGTAGATTCGTTGA
Foxa2	CTGAGCGAGATCTACCAGTGGA	CAGTCGTTGAAGGAGAGCGAGT
Hnf4a	ACTACATCAACGACCGCCAGT	ATCTGCTCGATCATCTGCCAG
pdx1	TGATACTGGATTGGCGTTGT	GCATCAATTTCACGGGATCT
Glut2	GCTACCGACAGCCTATTCTA	CAAGTCCCACTGACATGAAG
Insulin	GCCTTTGTGAACCAACACCTG	GTTGCAGTAGTTCTCCAGCTG
Pax4	AAGAAAGCAGCTTGCGTTGAC	GGGCGTGAGACAGAGGTATC
GAPDH ^b	GAGTCAACGGATTTGGTCGT	TTGATTTTGGAGGGATCTCG

Primer sets were used in quantitative RT-PCR.

Primer sets were used in both RT-PCR and quantitative RT-PCR.

RT-PCR, reverse transcription–polymerase chain reaction.

Real-time quantitative PCR (Q-PCR) analysis was performed on a Bio-Rad iQ5 system using the SYBR Green PCR Master Mix (TaKaRa), and all the experiments were repeated 3 times.

Western blot analysis

Cells were lysed in radio immunoprecipitation assay (RIPA) buffer containing a complete protease inhibitor cocktail tablet. Protein samples were separated in 8% polyacrylamide gels and transferred to polyvinylidene difluoride membranes. The membranes were incubated with primary antibodies including anti-NRSF (goat IgG; Santa Cruz) and anti-actin (mouse IgG; Santa Cruz) overnight at 4°C, and detection was performed with peroxidase-conjugated goat anti-rabbit or anti-mouse IgG using a chemiluminescence detection system.

Luciferase activity assays

Cells were cultured in normal medium or induction medium (DMEM/F12 medium supplemented with 2% FBS, 2% B27, 10 ng/mL bFGF, and 10 ng/mL Activin-A) for 5 days. We previously constructed the vector pcDNA3.1-HIP-Luc, which is a human insulin promoter-luciferase reporter [18]. This vector was electroporated into the cells along with pRL-CMV (Promega), a CMV promotor regulated Renilla luciferase control, to normalize the luciferase activity using a Neon Transfection System (Invitrogen) according to the manufacturer's protocol. Cells were harvested and lysed 48 h later, and luciferase activities were measured using the dual luciferase assay system (Promega) in a luminometer (Berthold).

C-peptide assays

After 14 days of induction, both siNRSF-hAFSCs and siControl-hAFSCs were analyzed for C-peptide content and secretion. Triplicate groups of induced cells (3 × 10⁵) were washed with KRB buffer and preincubated with Krebs-Ringer bicarbonate (KRB) buffer for 1 h followed by KRB buffer containing 5.5 or 23 mM glucose for 1 h at 37°C. The buffer was collected and stored at −20°C prior to testing. To analyze the levels of intracellular C-peptide, cells were immersed in an acidic alcohol solution (pure ethanol containing 10% glacial acetic acid) and sonicated. The amount of C-peptide was measured using an Ultrasensitive C-peptide ELISA kit (Mercodia) following the manufacturer's protocol. Protein concentration was determined using a bicinchoninic acid (BCA) protein assay system (Pierce). All values were determined against a standard curve prepared with human C-peptide.

Results

In vitro growth and characterization of hAFSCs

hAFSCs were isolated from adhering cell clones containing cells with 2 different morphologies: one set of cells had a fibroblastoid, spindle-shaped morphology, whereas the other had an epithelioid, polygonal morphology (Fig. 1A). With passaging, the epithelioid cells rapidly disappeared from culture. By passage 3, the hAFSCs showed a homogenous fibroblast-like morphology (Fig. 1B).

FIG. 1.

In vitro growth of hAFSCs. (A) Initial adhering hAFSCs; black arrowheads indicate hAFSCs with a spindle-shaped morphology, and white arrowheads indicate cells with an epithelioid morphology. (B) Appearance of hAFSCs at the 3rd passage. (C) Phenotype of hAFSCs. hAFSCs were positive for SSEA-4, CD29, CD90, CD105, and HLA-ABC, but negative for CD49, CD34, CD38, CD41, CD80, CD144, and HLA-DR. The dotted curves represent negative controls. (D) Immunofluorescence analysis for ES markers such as SSEA-4, Nanog, and Oct4 in hAFSCs. DAPI was used for nuclear staining. Scale bar = 100 μm. hAFSCs, human amniotic fluid-derived stem cells. Color images available online at www.liebertonline.com/scd

To characterize the hAFSCs, we tested the subcultured hAFSCs for stem cell markers and lineage-specific antigens. Flow cytometry analysis revealed that these cells expressed high levels of SSEA-4, CD29, CD90, CD105, and HLA-ABC but CD49, CD34, CD38, CD41, CD80, CD144, and HLA-DR were not detected (Fig. 1C). Immunofluorescence staining revealed that the majority of these cells were positive for SSEA-4, Nanog, and Oct4 (Fig. 1D).

The expression of NRSF is downregulated in hAFSCs by RNAi

The infection efficiency of the lentivirus in hAFSCs was determined using fluorescence microscopy. We found that almost all of the hAFSCs were infected with either Lv-siNRSF or Lv-siControl (Fig. 2A). To determine the efficiency of lentivirus-mediated RNA interference in hAFSCs, quantitative real-time PCR and western blotting were used to detect the expression level of NRSF in siNRSF-hAFSCs and siControl-hAFSCs at 3 days after infection. The results showed that Lv-siNRSF downregulated NRSF mRNA levels by 71% (Fig. 2B). Using western blotting, we found that there was weak NRSF protein expression in the siNRSF-hAFSCs (Fig. 2C). These results suggested that Lv-siNRSF was highly efficient in silencing the function of the NRSF gene. Using Q-PCR, we found that NRSF silencing caused a strong upregulation of Pax4 and Insulin transcripts (Fig. 3A). To further assess the effect of NRSF silencing on the expression of the Insulin gene, a luciferase reporter system was used to detect the transcriptional activity of the Insulin gene. The results showed that under normal culture conditions the luciferase activity of the insulin promoter in siNRSF-hAFSCs was significantly greater when compared with siControl-hAFSCs (Fig. 3B). When these cells were subjected to induction conditions, the activity of the insulin promoter in induced siNRSF-hAFSCs was increased ∼1.4 times when compared with uninduced siNRSF-hAFSCs. In contrast, the activity of the insulin promoter in siControl-hAFSCs was very low and no change was detected between uninduced and induced culture conditions. These results indicate that siNRSF-hAFSCs can further differentiate into insulin-producing cells under suitable induction conditions.

FIG. 2.

Downregulation of NRSF expression in hAFSCs by NRSF silencing. (A) The morphology of siNRSF-hAFSCs observed in a optical microscope (a) and a fluorescence microscope (b) showing that almost all hAFSCs were infected by lentivirus. Scale bar = 100 μm. (B) Real-time PCR was performed to detect the expression levels of the NRSF gene. The expression of NRSF in siNRSF-hAFSCs was downregulated by almost 71% at the mRNA level, and similar results were found by western blots of the protein level (C). NRSF, neuronal restrictive silencing factor; siNRSF, small interference NRSF; PCR, polymerase chain reaction. **p < 0.01. Color images available online at www.liebertonline.com/scd

FIG. 3.

NRSF knockdown enhanced the expression of Insulin and Pax4 transcripts. (A) Quantitative PCR was used to detect the expression levels of Pax4 and Insulin genes in siNRSF-hAFSCs and siControl-hAFSCs. (B) Insulin promoter activities were detected in different cells. The human Insulin promoter-luciferase reporter was electroporated into the uninduced siControl-hAFSCs (a), uninduced siNRSF-hAFSCs (c), induced siControl-hAFSCs (b), and induced siNRSF-hAFSCs (d) along with pRL-CMV using a Neon Transfection System. The activities of the luciferase reporters in uninduced siControl-hAFSCs (a) were assigned an activity value of 1.0, and the activities of reporters in other cells were expressed as n-fold relative to the activity of (a). Each experiment was performed at least 3 times in triplicate. Values represent the means ± SD of 3 independent experiments. *P < 0.05, **P < 0.01. SD, standard deviation.

Differentiation of siNRSF-hAFSCs into insulin-producing cells

To test whether NRSF silencing can promote hAFSCs to differentiate into pancreatic endocrine cells, the expression levels of several genes specific to endoderm or islet cells were detected by reverse transcription (RT)–PCR and Q-PCR. By RT-PCR, we found that the hAFSCs or uninduced siControl-hAFSCs expressed low levels of Foxa2, Hnf4α, and Pax4, but did not express the marker genes of islet endocrine cells (Fig. 4A). Q-PCR was further performed to compare the expression levels of these genes in induced cells at different stages. NRSF knockdown enhanced the expression of Hnf4α, Pax4, Pdx1, Nkx6.1, Glut2, and Insulin transcripts (Fig. 4B, D0). The differentiation medium further enhanced the expression of these genes, especially Foxa2, Pax4, Isl-1, and Glut2 transcripts, in the siNRSF-hAFSCs (Fig. 4B, D7 and D14). To determine the presence of insulin-positive cells, we performed immunofluorescence staining to detect the protein expression of Pdx-1, Insulin, and C-peptide in the induced cells. The results showed that differentiated siNRSF-hAFSCs expressed Pdx-1, Insulin, and C-peptide (Fig. 5B). However, the induced siControl-hAFSCs did not express these markers (Fig. 5A). These siControl-hAFSCs still expressed the stem cell-related markers such as SSEA-4, Oct4, and Nanog (Supplementary Fig. S1A; Supplementary Data are available online at www.liebertonline.com/scd). In addition, the induced siNRSF-hAFSCs showed no expression of these stem cell markers (Supplementary Fig. S1B), which also suggest that the induced siNRSF-hAFSCs are in a state of differentiation.

FIG. 4.

Detections of endodermal or islet cell-related genes in the uninduced or induced siControl-hAFSCs and siNRSF-hAFSCs. (A) Reverse transcription–PCR was used to detect the gene expression of Foxa2, Hnf4α, Pax4, Pdx1, Isl-1, Nkx6.1, Insulin, and Glut2 in hAFSCs and uninduced siControl-hAFSCs. (B) Quantitative PCR analysis of the expression levels of endodermal or pancreatic lineage genes in the uninduced siNRSF-hAFSCs or siControl-hAFSCs (D0), or induced cells at day 7 and day 14 (D7, D14). All mRNA expression levels were normalized to the house keeping gene GAPDH expression. The value of uninduced siControl-hAFSCs (D0) is arbitrarily set as 1. Data are presented as mean ± SD from 3 independent experiments. *P < 0.05, **P < 0.01. Foxa2, forkhead box a2; Hnf4α, hepatocyte nuclear factor 4α; Pax4, paired box gene 4; Nkx6.1, Nk6 transcription factor related locus 1; Isl-1, islet-1; Pdx-1, pancreatic and duodenal homeobox 1; Glut2, Glucose transporter 2.

FIG. 5.

Immunofluorescence analysis for Pdx1, Insulin, and C-peptide in induced siControl-hAFSCs (A) and induced siNRSF-hAFSCs (B) at day 14. DAPI was used for nuclear staining. Scale bar = 100 μm. Color images available online at www.liebertonline.com/scd

The ability of the induced cells to produce insulin can be confirmed by detecting either intracellular insulin or C-peptide content. We chose to detect the intracellular C-peptide content because of the controversy surrounding insulin uptake from media supplements by cells [19]. Intracellular C-peptide is the byproduct of de novo insulin synthesis. ELISA results showed that the content of intracellular C-peptide was ∼1.24 ng/mg in the differentiated siNRSF-hAFSCs. In contrast, there was an extremely low or even undetectble level of C-peptide in the induced siControl-hAFSCs (Fig. 6A). To further analyze whether insulin or C-peptide could be secreted into the medium, cells were incubated with KRB buffer containing either low or high concentrations of glucose. By ELISA, we found that the differentiated siNRSF-hAFSCs could release C-peptide in response to glucose stimulation. C-peptide release in the high-glucose medium was nearly 3 times greater than the amount released in the low-glucose medium (Fig. 6B). However, the amount of released C-peptide was almost undetectable in induced siControl-hAFSCs, and no notable changes were detected in response to different glucose concentrations. These results suggest that siNRSF-hAFSCs indeed differentiate into functional insulin-secreting cells.

FIG. 6.

(A) Intracellular C-peptide content was detected in induced cells. The cells were cultured in a 2-stage inducing medium for 14 days and then immersed in an acidic alcohol solution. Cell extracts were tested using an Ultrasensitive C-peptide ELISA kit. (B) Glucose-stimulated C-peptide release from the induced cells. The induced cells were incubated with KRB buffer containing 5.5 or 23 mM glucose for 1 h at 37°C. The buffer was collected for C-peptide measurement. Values represent the means ± SD of 3 independent experiments. **P < 0.01.

Discussion

In this study, we explored the differentiation of hAFSCs into insulin-producing cells. We obtained hAFSCs from routine amniocenteses and confirmed that hAFSCs derived from the developing fetus have the phenotype characteristics of pluripotent embryonic stem cells and adult stem cells. By RT-PCR, we found that hAFSCs express the anterior definitive endoderm marker Foxa2 and the primitive gut tube marker Hnf4α, reflecting their ability to differentiate toward the endodermal lineage. However, the potential of hAFSCs to differentiate into islet cells has not been tested, possibly because the mechanisms responsible for islet cell differentiation from stem cells remain elusive. Therefore, we examined the possibility of generating insulin-producing cells from hAFSCs through the downregulation of NRSF production.

NRSF is a member of the Kruppel-type zinc finger transcription factor family. This repressor is known to prevent the transcription of neuronal genes in most cell types other than islet β cells and neurons, which are essentially devoid of the repressor element-1 silencing transcription factor protein [13]. Several studies suggest that genes targeted by NRSF, such as Complexin I and Connexin 36 [12,20], play important roles in islet cells function. Moreover, it has been suggested that some islet development-related genes are negatively regulated by NRSF. Thus, the absence or low expression of NRSF may be essential for the development of healthy pancreatic islet β cells. Here, we used RNAi-based technology to silence NRSF gene expression in hAFSCs. The downregulation of NRSF expression in hAFSCs may trigger a derepression of NRSF-controlled genes and promote hAFSCs to differentiate toward insulin-producing cells. Previously, we found that NRSF could repress human insulin gene expression by binding to NRSE in its promoter region [21]. Using a luciferase reporter system, we found that NRSF silencing increased the transcriptional activity of the insulin gene in hAFSCs under normal culture conditions. To promote cellular differentiation, we added soluble factors into the medium, including FGF2 and Activin A [17]. These cytokines have been reported to be involved in mediating the suppression of sonic hedgehog signaling in the posterior foregut, which is required for initiation of pancreas gene expression [22]. When the factors were added into the induction medium, we found that the activity of the insulin promoter in induced siNRSF-hAFSCs was increased compared with that in uninduced siNRSF-hAFSCs. These data suggest that NRSF plays a dominate role in regulating insulin promoter activity and that the inducing medium is subordinate but necessary for enhancing the differentiation of siNRSF-hAFSCs into insulin-producing cells. Further studies evaluating which extracellular signals regulate NRSF gene expression may help to direct the differentiation of stem cells into islet cells.

Our findings demonstrate that there is a profound change in gene expression in the induced siNRSF-hAFSCs. The differentiated siNRSF-hAFSCs expressed crucial β-cell transcription factors and functional marker genes, which could reflect the emergence of a β-cell phenotype. The homeodomain transcription factor Pdx-1, which is also known as Idx-1/Ipf-1, is an essential regulator of pancreatic development and islet function [23]. We confirmed the expression of Pdx-1 protein in the differentiated siNRSF-hAFSCs. We believe that the putative β-cells in our induced system produce insulin de novo because insulin gene expression was detected and the cells stained positive for insulin and C-peptide, a peptide made from pro-insulin, which is a marker of de novo insulin synthesis. We found that C-peptide could be secreted into the media by differentiated siNRSF-hAFSCs. In addition, these cells released C-peptide when stimulated with glucose in a manner comparable to that of adult human islets, possibly via activation of exocytosis-related genes. In contrast, the induced siControl-hAFSCs secreted little C-peptide and showed no glucose responsiveness. These data suggest that NRSF silencing is essential for hAFSC differentiation into insulin-secreting cells. The level of intracellular C-peptide in the differentiated siNRSF-hAFSCs was similar to that in insulin-producing cells from human embryonic stem cells or human neural progenitor cells [24,25]. The content and the released amount of C-peptide from these cells were much lower than the levels from adult human islets [24 –27]. Zhang's report showed that the C-peptide secretion amount of the adult human islets at 27.5 or 2.5 mM glucose concentration was 41.2 versus 16.1 ng/mg protein [27]. In our experiments, the amount of C-peptide released by the differentiated siNRSF-hAFSCs at 23 mM glucose concentration was ∼1 ng/mg protein, which was ∼40-fold lower than that of adult islets. The differentiated insulin-producing cells still expressed Foxa2 mRNA, indicating that these cells were not yet mature. There are 2 explanations for these results. First, the NRSF silencing efficiency is not 100%. As such, a complete knockout of NRSF may produce different biological effects. Second, culture conditions can affect the differentiation of NRSF knockdown cells. Culturing these cells on matrigel [28] or as a suspension may affect their growth, as can the addition of soluble factors.

In summary, our study clearly demonstrates that NRSF silencing may promote hAFSCs to differentiate into insulin-producing β-like cells, and these differentiated cells are able to secrete C-peptide in response to glucose stimulation. These results support us to further evaluate the use of hAFSCs as a resource for cell replacement therapy in diabetes.

Footnotes

Acknowledgments

This work was supported by the National High Technology Research and Development Program of China (no. 2006AA02A107), the National Basic Research Program of China (no. 2010CB945500), and the Beijing Municipal Natural Science Foundation of China (no. 5102036).

Author Disclosure Statement

No competing financial interests exist.

References

Shapiro

, Lakey

, Ryan

, Korbutt

, Toth

, Warnock

, Kneteman

, Rajotte

. 2000. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med, 343:230–238.

Kroon

, Martinson

, Kadoya

, Bang

, Kelly

, Eliazer

, Young

, Richardson

, Smart

, Cunningham

, Agulnick

, D'Amour

, Carpenter

, Baetge

. 2008. Pancreatic endoderm derived from human embryonic stem cells generates glucose-responsive insulin-secreting cells in vivo. Nat Biotechnol, 26:443–452.

Tateishi

, He

, Taranova

, Liang

, D'Alessio

, Zhang

. 2008. Generation of insulin-secreting islet-like clusters from human skin fibroblasts. J Biol Chem, 283:31601–31607.

von Mach

, Hengstler

, Brulport

, Eberhardt

, Schormann

, Hermes

, Prawitt

, Zabel

, Grosche

, Reichenbach

, Müller

, Weilemann

, Zulewski

. 2004. In vitro cultured islet-derived progenitor cells of human origin express human albumin in severe combined immunodeficiency mouse liver in vivo. Stem Cells, 22:1134–1141.

De Coppi

, Bartsch

Jr. , Siddiqui

, Xu

, Santos

, Perin

, Mostoslavsky

, Serre

, Snyder

, Yoo

, Furth

, Soker

, Atala

. 2007. Isolation of amniotic stem cell lines with potential for therapy. Nat Biotechnol, 25:100–106.

Trovato

, De Fazio

, Annunziata

, Sdei

, Favaro

, Ponti

, Marozio

, Ghigo

, Benedetto

, Granata

. 2009. Pluripotent stem cells isolated from human amniotic fluid and differentiation into pancreatic beta-cells. J Endocrinol Invest, 32:873–876.

Hermanson

. 2008. Stem cells have different needs for REST. PLoS Biol, 6:e271.

Schoenherr

, Anderson

. 1995. The neuron-restrictive silencer factor (NRSF): a coordinate repressor of multiple neuron-specific genes. Science, 267:1360–1363.

Atouf

, Czernichow

, Scharfmann

. 1997. Expression of neuronal traits in pancreatic beta cells. Implication of neuron-restrictive silencing factor/repressor element silencing transcription factor, a neuron-restrictive silencer. J Biol Chem, 272:1929–1934.

10.

Kemp

, Lin

, Habener

. 2003. Regulation of Pax4 paired homeodomain gene by neuron-restrictive silencer factor. J Biol Chem, 278:35057–35062.

11.

Johnson

, Mortazavi

, Myers

, Wold

. 2007. Genome-wide mapping of in vivo protein-DNA interactions. Science, 316:1497–1502.

12.

Martin

, Allagnat

, Chaffard

, Caille

, Fukuda

, Regazzi

, Abderrahmani

, Waeber

, Meda

, Maechler

, Haefliger

. 2008. Functional significance of repressor element 1 silencing transcription factor (REST) target genes in pancreatic beta cells. Diabetologia, 51:1429–1439.

13.

Abderrahmani

, Niederhauser

, Plaisance

, Haefliger

, Regazzi

, Waeber

. 2004. Neuronal traits are required for glucose-induced insulin secretion. FEBS Lett, 565:133–138.

14.

, Liu

, Yang

, Lv

, Chen

, Bai

, Nan

, Wang

, Pei

. 2008. Regulatory role of neuron-restrictive silencing factor in the specific expression of cocaine- and amphetamine-regulated transcript gene. J Neurochem, 106:1314–1324.

15.

Greco

, Smirnov

, Murthy

, Rameshwar

. 2007. Synergy between the RE-1 silencer of transcription and NFkappaB in the repression of the neurotransmitter gene TAC1 in human mesenchymal stem cells. J Biol Chem, 282:30039–30050.

16.

Yang

, Li

, Lv

, Zhang

, Chen

, Bai

, Nan

, Yue

, Pei

. 2008. NRSF silencing induces neuronal differentiation of human mesenchymal stem cells. Exp Cell Res, 314:2257–2265.

17.

Shi

, Hou

, Tang

, Jiang

, Wang

, Ding

, Deng

. 2005. Inducing embryonic stem cells to differentiate into pancreatic beta cells by a novel three-step approach with activin A and all-trans retinoic acid. Stem Cells, 23:656–662.

18.

Yang

, Li

, Liu

, Zhang

, Chen

, Lv

, Bai

, Nan

, Pei

. 2008. Construction and function study of NRSF lentiviral RNA interference vector. Prog Biochem Biophys, 35:151–158.

19.

Paek

, Morgan

, Lysaght

. 2005. Sequestration and synthesis: the source of insulin in cell clusters differentiated from murine embryonic stem cells. Stem Cells, 23:862–867.

20.

Martin

, Tawadros

, Meylan

, Abderrahmani

, Condorelli

, Waeber

, Haefliger

. 2003. Critical role of the transcriptional repressor neuron-restrictive silencer factor in the specific control of connexin36 in insulin-producing cell lines. J Biol Chem, 278:53082–53089.

21.

Liu

, Li

, Yang

, Yuan

, Zhang

, Q

, Lv

, Zhang

, Sun

, Chen

, Bai

, Nan

, Wang

, Pei

. 2007. NRSF plays a regulatory role in human insulin gene transcription. Prog Biochem Biophys, 34:952–959.

22.

Hebrok

, Kim

, Melton

. 1998. Notochord repression of endodermal Sonic hedgehog permits pancreas development. Genes Dev, 12:1705–1713.

23.

Stoffers

, Thomas

, Habener

. 1997. Homeodomain protein IDX-1: a master regulator of pancreas development and insulin gene expression. Trends Endocrinol Metab, 8:145–151.

24.

Jiang

, Shi

, Zhao

, Chen

, Yong

, Zhang

, Qing

, Sun

, Zhang

, Ding

, Li

, Deng

. 2007. In vitro derivation of functional insulin-producing cells from human embryonic stem cells. Cell Res, 17:333–344.

25.

Hori

, Gu

, Xie

, Kim

. 2005. Differentiation of insulin-producing cells from human neural progenitor cells. PLoS Med, 2:e103.

26.

Brandhorst

, Brandhorst

, Hesse

, Ambrosius

, Brendel

, Kawakami

, Bretzel

. 2003. Successful human islet isolation utilizing recombinant collagenase. Diabetes, 52:1143–1146.

27.

Zhang

, Jiang

, Liu

, Sui

, Yin

, Chen

, Shi

, Deng

. 2009. Highly efficient differentiation of human ES cells and iPS cells into mature pancreatic insulin-producing cells. Cell Res, 19:429–438.

28.

Lechner

, Nolan

, Blacken

, Habener

. 2005. Redifferentiation of insulin-secreting cells after in vitro expansion of adult human pancreatic islet tissue. Biochem Biophys Res Commun, 327:581–588.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.31 MB