Characterization of induced tissue-specific stem cells from pancreas by a synthetic self-replicative RNA

Induced pluripotent stem (iPS) cells have significant implications for overcoming most of the ethical issues associated with embryonic stem (ES) cells. Furthermore, our recent study demonstrated the generation of induced tissue-specific stem (iTS) cells by transient overexpression of the reprogramming factors using a plasmid combined with tissue-specific selection. In this study, we were able to generate RNA-based iTS cells that utilize a single, synthetic, self-replicating VEE-RF RNA replicon expressing four reprogramming factors (OCT4, KLF4, SOX2, and GLIS1). A single VEE-RF RNA transfection into mouse pancreatic tissue resulted in efficient generation of iTS cells from pancreas (iTS-P cells) with genetic markers of endoderm and pancreatic progenitors and differentiation into insulin-producing cells more efficiently than ES cells. Subcutaneous transplantation of iTS-P cells into immunodeficient mice resulted in no teratoma formation. Bisulfite genomic sequencing demonstrated that the promoters of Oct4 and Nanog remained partially methylated in iTS-P cells. We compared the global gene-expression profiles of ES cells, iTS-P cells, and pancreatic islets. Microarray analyses confirmed that the iTS-P cells were similar but not identical to ES cells compared with islets. These data suggest that iTS-P cells are cells that inherit numerous components of epigenetic memory from pancreas cells and acquire self-renewal potential. The generation of iTS cells may have important implications for the clinical application of stem cells.


Characterization of iTS-P cells by RNA vector.
We applied a stepwise differentiation protocol to evaluate the six clones (Fig. 2a) with a higher Pdx1 mRNA expression than other clones that were to be differentiated into insulin-producing cells 21,22 . Since iTS-P cells express endodermal cell markers, we included stages 4 and 5 of the induction protocol in the stepwise differentiation protocol. All six clones induced the expression of insulin-2 mRNA (Fig. 2b). A reverse transcription-polymerase chain reaction (RT-PCR) analysis of isolated RNAs showed that the iTS-P23 cells expressing the highest insulin-2 mRNA after treatment of induction medium had lost the VEE-RF-RNA replicon by passage 8 (Fig. 2c). The iTS-P23 cells continued to divide actively beyond passage 30 without changes in their morphology or growth activity (Fig. 2d).

Gene expression of iTS-P cells.
To investigate the gene expression in iTS-P23 cells, an RT-PCR analysis of ES cell marker genes was performed. The expression of the pluripotency markers Nanog, Sox2, and Oct3/4 in iTS-P23 cells was significantly lower than that in ES cells (Fig. 3a). We next investigated the gene expression patterns of endodermal markers. Cells differentiated from ES cells (generated by a stepwise differentiation protocol that relies on intermediates thought to be similar to the cell populations present in the developing embryo; Stage 1-3) 21,22 were used as a positive control. The expression of endodermal marker genes (forkhead box protein a2; Foxa2, hepatocyte nuclear factor 1β, 4α, 6; Hnf1β, 4α, 6) was detected in iTS-P23 cells (Fig. 3b), which was similar to the patterns detected in our mouse pancreatic stem cell line HN#13 and in the iTS-P cells we previously reported 13,14,23 , but not ES cells. We next investigated the gene expression patterns of pancreatic/endocrine progenitor (PP) markers. Cells isolated from E10-14 mouse embryos were used as a positive control. The expression of pancreatic/endocrine progenitor marker genes (pancreas transcription factor 1 subunit alpha; Ptf1A, Neurogenin-3; Ngn3, Sox9) was detected in PP cells. The expression of Sox9 gene was detected in iTS-P23 cells, whereas the expression of Ptf1a, Ngn3, or Nkx6.1 genes were not strongly detected in the cells (Fig. 3c). We next investigated the gene expression patterns of pancreatic markers. Pancreatic islets were used as a positive control. The expression of Pdx1 gene was detected in iTS-P23 cells, whereas the expression of Ptf1a, NeuroD, or Nkx6.1 genes were not strongly detected in the cells (Fig. 3d).

Differentiation of iTS-P cells into insulin-producing cells.
To address the possibility that iTS-P cells were prone to skewed differentiation into insulin-producing cells, the expression of insulin-1 and insulin-2 genes and proteins was analyzed in spheroids derived from iTS-P cells and compared to embryonic bodies (EBs) in ES cells (Fig. 4a). The spheroids from iTS-P23 cells differentiated into insulin-producing cells (Fig. 4b) more efficiently than EBs from ES cells (Fig. 4c). The insulin-positive cells were C-peptide-positive (Fig. 4b), thus excluding the possibility of the uptake of insulin from the media; 5.56 ± 0.68% of the differentiated cells were insulin/C-peptide-positive (n = 10), while the percentage of insulin/C-peptide-positive cells differentiated from ES cells was <1% (n = 10). This skewed differentiation manifests as their increased capacity for spontaneous differentiation into insulin-producing cells. The differentiated iTS-P cells expressed Pdx1 and Nkx6.1 proteins (Fig. 4b) and Nkx6.1 and MafA mRNA (Fig. 4c). To evaluate whether the differentiated cells had glucose sensitivity, the differentiated cells from the iTS-P cells were exposed to low (2.8 mM) or high (20 mM) concentrations of glucose. The amount of insulin released by the cells was approximately 10-20 fold higher than that released by an ES-derived population at both glucose concentrations, although the amount of insulin was lower than that of islets (Fig. 4d).
We applied a direct differentiation method (stepwise differentiation protocol 21,22 ) to drive the generation of insulin-expressing cells. Using the method, iTS-P23 cells efficiently differentiated into insulin-producing cells (insulin/Cpeptide-positive cells; 6.87 ± 2.15%, n = 8; Fig. 4e). The differentiated iTS-P cells expressed insulin-1, -2 and MafA mRNA (Fig. 4f). The amount of insulin released by the differentiated iTS-P cells was higher than that released by an ES-derived population at both glucose concentrations (Fig. 4g).
Bisulfite genomic sequencing of the promoter regions of Oct3/4, Nanog, Insulin-1, and Insulin-2 in iTS-P and ES cells. Bisulfite genomic sequencing demonstrated that the promoters of Oct3/4 and Nanog remained methylated in iTS-P cells but were demethylated in ES cells. In contrast, the promoters of insulin-1 and insulin-2 were methylated in both iTS-P and ES cells, although the rate of methylation was higher in ES cells than in iTS-P cells (Fig. 5a). These results demonstrate that methylation of the promoters in iTS-P cells is not similar to that in ES cells. We also evaluated the promoters of Pdx1 and Nkx6.1. The promoter of Pdx1 was demethylated and that of Nkx6.1 was methylated in iTS-P cells. On the other hand, the promoters of both Pdx1 and Nkx6.1 The colony numbers of iTS-P and iPS cells. The synthetic self-replicative RNA was transfected into pancreatic tissue from C57/BL6 mouse at 24 weeks of age, and the number of colonies was counted after 23-45 days. (c) Teratoma formation/tumorigenicity assay. A total of 1 × 10 6 to 1 × 10 7 iPS/iTS cells were inoculated into each humerus and thigh of NOD/scid mice. (d) The quantitative RT-PCR analysis of Pdx1 genes, markers of pancreatic stem/progenitor cells, in iTS-P cells. A total of 28 iTS-P clones were evaluated for their expression of Pdx1 by quantitative RT-PCR. #5, 7, 10 clones were iPS cells. The data are expressed as the Pdx1-to-Gapdh ratio, with the ratio of the #41 clone arbitrarily set at 1 (n = 5). Error bars represent the standard error. were demethylated in pancreatic islets and were methylated in acinar cells. In ductal cells, the promoter of Pdx1 was demethylated and that of Nkx6.1 was methylated in iTS-P cells (Fig. 5b). These results demonstrate that methylation pattern of Pdx1 and Nkx6.1 promoters in iTS-P cells is similar to that in ductal cells.
Microarray data. We compared the global gene-expression profiles of ES cells, iTS-P cells, and pancreatic islets using microarrays (Fig. 6). A total of 45,037 genes were evaluated, and 28,934 genes (64.2%) were positive in ES cells, iTS-P cells, and/or pancreatic islets, while 16,103 genes (35.8%) were negative in these cells. A total of 1,760 genes (3.9%), including Oct3/4, Sox2, and Nanog, were positive in only ES cells; 1,794 genes (4.0%), including L-Myc, were positive in only iTS-P cells; 2,365 genes (5, 3%), including insulin-1 and insulin-2, were positive  in only pancreatic islets; 5,043 genes (11.2%), including Dlk1 and N-Myc, were positive in ES cells and iTS-P cells; and 1,206 genes (2.7%), including Pdx1, Hnf1β, and Hnf4α, were positive in iTS-P cells and pancreatic islets. The analysis revealed that iTS-P cells are clustered more closely with ES cells than pancreatic islets, although iTS-P cells are markedly different from ES cells. A total of 16,061 genes (35.7%), including Gapdh and β-actin, were positive in all 3 cell types.

Discussion
Cell replacement therapies, such as transplantation of purified islets, have emerged as promising alternatives to whole-organ transplantation for the treatment of patients with type 1 diabetes 24,25 . These observations are proof-of-concept and have intensified interest in treating diabetes not only by cell transplantation but also by stem cells. The regeneration of β cells from stem and progenitor cells is an attractive method of restoring the islet cell mass. Protocols for the in vitro differentiation of ES/iPS cells based on normal developmental processes have generated β-like cells that produce high levels of insulin 21,22,26 , albeit at low efficiency and without full responsiveness to extracellular levels of glucose. Although pancreatic stem/progenitor cells have been identified 23,27-32 , pancreatic "progenitor" cells have limited self-renewal capacity, and it is extremely difficult to isolate human pancreatic "stem" cells with self-renewal capacity 33 . Therefore, the generation of iTS-P cells using iPS-cell technology may create several possibilities for the development of new treatments for diabetes. The iTS-P cells were able to differentiate into insulin-producing cells more efficiently than ES cells. Furthermore, the iTS-P cells do not form teratomas. ES/iPS cells carry a risk of teratoma formation, even after transplantation of differentiated cells derived from ES/iPS cells, due to possible contamination with undifferentiated cells. This is one of the advantages of iTS-P cells over ES/iPS cells in terms of potential clinical use. Bisulfite  genomic sequencing in this study clearly demonstrated that the promoters of Oct3/4 and Nanog remained methylated in iTS-P cells, while the promoters were demethylated in ES cells. Moreover, quantitative RT-PCR showed that there were few expressions of Oct3/4 or Nanog. These results demonstrate that methylation of the promoters in iTS-P cells is not similar to that in ES cells; therefore, iTS-P cells are unlikely to have pluripotency or teratoma formation.
The global gene-expression profiles of ES cells, iTS-P cells, and pancreatic islets using microarrays showed that iTS-P cells were markedly different from iPS cells and pancreatic islets. Of the 45,037 total genes evaluated, 11.2% were positive in both ES cells and iTS-P cells, while 2.7% were positive in both iTS-P cells and pancreatic islets, showing that iTS-P cells were more closely related to ES cells than pancreatic islets. Interestingly, L-Myc was positive in only iTS-P cells, while c-Myc and N-Myc were positive in both ES cells and iTS-P cells. The Myc family of transcription factors comprises c-Myc, N-Myc, and L-Myc and has been implicated in the generation of a variety of human tumors. It has been reported that L-myc knockout mice develop normally 33 , embryos lacking c-myc die before E10.5 due to hematopoietic and placental defects 34,35 , and N-myc-deficient embryos die before E11.5 due to neuroectodermal and heart defects 36 . Myc activity is essential for efficient cellular reprogramming 37 and has complex roles in various stem and progenitor cell types 38 . A recent study showed that Myc controls the biosynthetic machinery of stem cells without affecting their potency 39

. Differences in the L-Myc expression between ES cells and iTS-P cells may be an important factor influencing differences in the characteristics of ES cells and iTS-P cells.
The expression of Pdx1 in clones #6, 12, 17, 20, 23, and 41 was higher than that in the other clones (Fig. 1d) and the expression of insulin-2 after their differentiation differed among the clones (Fig. 2b). Our previous study also showed that the differentiation ability of iTS-P cells into insulin-producing cells depends on the individual clone 13 . Since Yamanaka's group showed the generation of germline-competent iPS cells via the selection for the expression of Nanog instead of Fbx15, the method of clone selection may be one of key factors for generating high quality iTS-P cells.
In conclusion, we generated iTS cells from mouse pancreas cells using a single, synthetic, self-replicating RNA vector expressing four reprogramming factors in order to avoid potential integration problems. The characterization of iTS-P cells derived from the RNA vector by methylation and microarray analyses clearly differed from that of ES cells or pancreatic islets. The advantages of iTS cells over iPS cells are (1) easier generation, (2) more efficient differentiation, and (3) no teratoma formation. Another group recently showed the induction of expandable tissue-specific stem/progenitor cells through the transient expression of YAP/TAZ 40 . The technology to generate iTS cells by reprogramming factors and tissue-specific selection may also be useful for the generation of other tissue-specific stem cells.

Research Design and Methods
Mice and cell culture. All mouse studies were approved by the review committee of University of the Ryukyus. The 24-week-old C57/BL6 mice (Charles River Laboratories Japan, Inc., Kanagawa, Japan) were used for primary pancreatic tissue preparations. Mouse pancreas was digested with 2 ml cold M199 medium containing 2 mg/ml collagenase (Roche Diagnostics Corporation, Indianapolis, IN, USA). The digested tissues were cultured in Dulbecco's modified Eagle's medium (DMEM; GIBCO-Invitrogen, Carlsbad, CA, USA) with 10-20% fetal bovine serum (FBS; GIBCO-Invitrogen). Eight-week-old nude or NOD/scid mice (CREA) were used for teratoma formation studies.
Mouse ES cells (ATCC, Manassas, VA, USA), iPS cells, and iTS cells were maintained in complete ES cell media with 15% FBS (Merck Millipore, Tokyo, Japan) on feeder layers of mitomycin C-treated STO or SNB cells, as described previously 1,41 . ES cells were passaged every three days, and iTS-P cells were passaged every five days.

Generation of iPS and iTS-P cells by replicon transfection.
iPS and iTS-P cells were generated as described previously 12 using a Simpicom TM RNA Reprogramming Kit (Millipore). Pancreatic cells were seeded onto a T25 plate on day 0 and cultured to 90-100% confluency on day 1. To minimize the IFN response, cells were treated with 1 mL Advanced DMEM containing 0.2 µg of B18R protein 2 h before transfection. A total of 1 µg RNA mixture (0.5 µg VEE-OKS-iG/0.5 µg B18R mRNA) was transfected with Lipofectamine 2000. After 3 h, the transfection medium was changed to Advanced DMEM containing 200 ng/mL of B18R protein. On day 11, Advanced DMEM was replaced with ES culture medium. Puromycin (0.8 µg/ml) was added from days 2 to 10. Cells were passaged onto STO feeder cells on day 11 and cultured in ES culture medium. Advanced DMEM containing 200 ng/mL of B18R protein was supplied every day until iPS or iTS-P colonies were generated.
Quantitative PCR/RT-PCR. Total RNA was extracted from cells using an RNeasy Mini Kit (Qiagen, Tokyo, Japan). After quantifying the RNA by spectrophotometry, 2.5 µg of RNA was heated at 85 °C for 3 minutes and then reverse-transcribed into cDNA in a 25-µl solution containing 200 units of Superscript II RNase H-RT (Invitrogen), 50 ng random hexamers (Invitrogen), 160 µmol/l dNTP, and 10 nmol/l dithiothreitol. The reaction consisted of 10 minutes at 25 °C, 60 minutes at 42 °C, and 10 minutes at 95 °C. Polymerization reactions were performed in a Perkin-Elmer 9700 Thermocycler with 3 µl cDNA (20 ng DNA equivalents), 160 µmol/l cold dNTPs, 10 pmol appropriate oligonucleotide primers, 1.5 mmol/l MgCl 2 , and 5 units AmpliTaq Gold DNA polymerase genes-to-Gapdh ratio, with the ratio of the differentiated ES cells arbitrarily set at 1 (n = 5). Error bars represent the standard error. (g) The insulin release assay. Differentiated iTS-P and ES cells were stimulated with 2.8 and 20 mM D-glucose, and the amount of insulin released into the culture supernatant was analyzed by ELISA. Error bars represent the standard error. *p < 0.05.
The quantification of the mRNA was carried using the TaqMan real-time PCR system according to the manufacturer's instructions (Applied Biosystems, Foster City, CA, USA). PCR was performed for 40 cycles, including 2 minutes at 50 °C and 10 minutes at 95 °C as initial steps. In each cycle, denaturation was performed for 15 seconds at 95 °C, and annealing/extension was performed for 1 minute at 60 °C. PCR was carried out in 20 µl of solution using cDNAs synthesized from 1.11 ng of total RNA. For each sample, the expression of mRNA was normalized by dividing by the Gapdh expression level. Primers for mouse insulin-1, insulin-2, Oct3/4, sex-determining region Y-box 2 (Sox2), Nanog, Foxa2, Sox17, hepatocyte nuclear factor 1β (Hnf1β), Hnf4α, Hnf6, Ptf1a, Ngn3, Sox9, Nkx6.1, the pancreatic and duodenal homeobox factor-1 (Pdx1), NeuroD, and Gapdh are commercially available (Assays-on-Demand Gene Expression Products; Applied Biosystems).
Cell induction and differentiation. Directed differentiation into insulin-producing cells was conducted as described previously 21,22 , with minor modifications. ES cells (passage 45), iTS-P cells (passage 45) were used in The differentiation of ES/iTS cells into insulin-producing cells was also conducted by EB/spheroid formation. To initiate EB/spheroid formation, a semi-confluent 10-cm plate of ES or iTS cells was harvested using trypsin, and cell clumps were resuspended in ES cell medium without LIF, allowed to aggregate, and transferred to one well of a nonadherent six-well plate. EBs/spheroids were allowed to undergo spontaneous differentiation for seven days in suspension, after which they were collected and taken for RNA and protein analyses.
Teratoma formation/tumorigenicity assay. A total of 1 × 10 6 to 1 × 10 7 iPS/iTS cells were inoculated into each humerus and thigh of NOD/scid mice. As a positive control, we transplanted 1 × 10 6 ES cells into 1 thigh of the NOD/scid mice.
Immunostaining. The cells were fixed with 4% paraformaldehyde in PBS buffer. After blocking with 20% AquaBlock (EastCoast Bio, North Berwick, ME, USA) for 30 min at room temperature, the cells were incubated overnight at 4 °C with a guinea pig anti-insulin antibody (1:100; Abcam, Tokyo, Japan), rabbit anti-C-peptide antibody   Table 1. Amplified products were cloned using Mighty TA-cloning kit (TAKARA BIO INC, Shiga, Japan). Ten randomly selected clones were sequenced with the M13 forward and M13 reverse primers for each gene.
Microarray. The total RNA from ES cells, iTS-P cells, or islets was labeled with biotin. Samples were hybridized to the GeneChip 3′IVT PLUS Reagent Kit (Affymetrix, Tokyo, Japan) and CeneChip Hybridization, Wash and Stain Kit (Affymetrix) according to the manufacturer's protocol. Arrays were scanned with the GeneChip Scanner 3000 7G (Affymetrix). Data were analyzed using the Affymetrix GeneChip Command Console software program (Affymetrix).
Statistical analyses. The data are expressed as the means ± standard error. Two groups were compared using Student's t-test. P values of <0.05 were considered to indicate statistical significance.
All methods were performed in accordance with the relevant guidelines and regulations.