Kyoto probe-1 reveals phenotypic differences between mouse ES cells and iTS-P cells

Kyoto probe 1 (KP-1) rapidly distinguishes between human ES/iPS (hES/iPS) cells and their differentiated cells. Recently, we generated induced tissue-specific stem cells from pancreas (iTS-P cells) using reprogramming factors and tissue-specific selection. The iTS-P cells have self-renewal potential, and subcutaneously transplanting them into immunodeficient mice did not generate teratomas. In this study, we applied KP-1 to analyze mouse ES (mES) cells and mouse iTS-P (miTS-P) cells. KP-1 completely stained mES cells in colonies, but only miTS-P cells at the edge of a colony. This difference was caused by cell type-specific expression of different ABC transporters. These finding suggest that KP-1 will be useful for distinguishing between iPS and iTS-P cells.


Quantitative reverse transcription-polymerase chain reaction (RT-qPCR) analysis of selected genes encoding ABC transporters in mES and miTS-P cells.
We determined the levels of mRNAs encoding the ABC transporters Abcb1, Abcg2, Abcc5, Abcc10, Abcc12, and Abca2 in mES and miTS-P cells. Abcb1 and Abcg2 were expressed at significantly higher levels by mES cells compared with those expressed by miTS-P cells (Fig. 4A). These data suggest that Abcb1 and Abcg2 were unlikely to mediate the efflux of KP-1 from mES cells. In contrast, the levels of Abcc5, Abcc10, Abcc12, and Abca2 mRNAs in iTS-P cells were significantly higher compared with those in mES cells (Fig. 4B). These data suggest that at least one of these ABC transporters mediated the efflux of KP-1 from miTS-P cells.
Next, when we used siRNAs specific for ABC transporters to understand their role in KP-1 selectivity (Supplemental Fig. 2), we found that KP-1 stained miTS-P cells after treatment with Abcc10-siRNA (Fig. 5C) or Abcc 12 siRNA (Fig. 5D). In contrast, the central cells of miTS-P colonies were weakly stained by KP-1 after treatment with Abcc5-siRNA (Fig. 5E), and central cells of miTS-P colonies were not stained by KP-1 after treatment with Abca2-siRNA (Fig. 5F) or control-siRNA (Fig. 5G).
Immunohistochemical analysis detected Abcc10 and Abcc12 expression by the central cells of miTS-P colonies (Fig. 6A,B). In contrast, Abcc5 was expressed in an irregularly mottled pattern by miTS-P colonies (Fig. 6C), and the border cells of miTS-P colonies expressed Abca2 (Fig. 6D). These data suggest that Abcc10 and Abcc12 participated in mediating the efflux of KP-1 from miTS-P cells. KP-1 treatment of mES, miTS-P, and mixtures of miPS/miTS-P cells. (A) mES cells were added (2 × 10 5 cells per well) to a six-well plate and cultured for 5 days, treated with 2 µM KP-1 for 3 h at 37 °C, rinsed with PBS, and images were acquired using a fluorescence microscope. Scale bar = 200 µm. (B) miTS-P cells were added (2 × 10 5 cells per well) to a six-well plate and cultured for 5 days, treated with 2 µM KP-1 for 3 h at 37 °C, rinsed with PBS. Scale bar = 200 µm. (C) A mixture of miPS/miTS-P cells was added (2 × 10 5 cells per well) to a six-well plate and cultured for 5 days, treated with 2 µM KP-1 for 3 h at 37 °C, and rinsed with PBS. Scale bar = 200 µm. arrow; miPS cells, arrow head; miTS-P cells (D) Time-course of miTS-P cells treated with KP-1. miTS-P cells were added (2 × 10 5 cells/well) to a six well plate. miTS-P cells were cultured from days 1 to 3 and 5 and then treated with 2 µM KP-1 for 3 h at 37 °C Cells were rinsed with PBS, and. Scale bar = 200 µm. www.nature.com/scientificreports/

Discussion
Here we show that KP-1 distinguished between mES and miTS-P cells. Specifically, KP-1 stained mES cells to the same extent as hES/hiPS cells, while in striking contrast, KP-1 only stained border cells of miTS-P colonies. Our present findings extend our previous findings demonstrating that iPS cells are distinguished from iTS-P cells through their differential expression of Pdx1 25,28 . Compared with a study of human ES cells 18 , our study showed that mouse ES cells differentially express ABC transporters. Further, the mES and miTS-P cells exhibited different morphologies, with the mouse ES cells forming clusters while the miTS-P cells formed cobble-stonelike colonies, which formed a monolayer. Accordingly, we can exclude the possibility that KP-1 was unable to penetrate through larger colonies to reach the iTS-P cells. ABCB1 and ABCG2 mediate the efflux of KP-1 from human cells, and the expression of each transporter is repressed in hES cells 36 . In contrast, our microarray and RT-qPCR data show that mES cells expressed Abcb1 and Abcg2, and previous studies show that ABCG2 is expressed at low levels in hES cells, while Abcg2 is expressed high-levels in mES cells 37 Further, ABCG2 is expressed at high levels by hES cells 38 . Despite these conflicting data regarding the expression of ABCG2 in hES cells, we conclude from our present data and those of others [36][37][38] that Abcb1 and Abcg2 are unlikely to cause the efflux of KP-1 from mouse cells. Further, our present microarray and RT-qPCR data lead us to conclude that Abcc5, Abcc10, Abcc12, and Abca2 may mediate the efflux of KP-1 from mouse cells. www.nature.com/scientificreports/ Our present studies using the Abcc5 inhibitor sildenafil 33,34 , Abcc10 inhibitor CsA 18,35 , and siRNA of each ABC transporter show that Abcc10 and Abcc12 participate in the mechanism of efflux of KP-1 from miTS-P cells. The lipophilic anion transporter Abcc10 23,39-42 mediates the transport of glucuronate conjugates such as E 2 17βG and GSH conjugates such as LTC 4 41 . Further, Abcc10 may possess a bipartite substrate binding pocket that interacts with anionic and lipophilic ligands. The transport of E 2 17βG is competitively inhibited by organic anions such as LTC 4 , glycolithocholate 3-sulfate, and MK-571, as well as by lipophilic agents such as CsA 41 . Abcc12 was identified using a functional genomics approach and bioinformatics analysis 43,44 . The predicted amino acid sequence of Abcc12 exhibits the highest degree of similarity with Abcc5. Although the function of Abcc12 is unknown, it may differ from that of other family members 43,44 .
Our present findings that KP-1 stains only the border cells of colonies formed by miTS-P cells suggest that the ABC transporters were differentially expressed in central cells vs border cells of the colony. However, KP-1 stained most miTS-P cells on day 1 after cell passage, suggesting that the KP-1-negative central cells reversibly incorporated KP-1-positive cells after cell passage. The morphology of miTS-P colonies was cobblestone-like, and the central cells of the colony were contact-inhibited. Thus, the inhibition of cell division upon contacting a neighboring cell may prevent miTS-P cells from forming tumors. In contrast, border cells of the miTS-P colony and miTS-P cells on day 1 after cell passage were not contact-inhibited and continued to divide. Thus, contact inhibition may explain the ability of KP-1 to stain miTS-P cells.
In summary, we present compelling evidence that KP-1 distinguishes between mES cells and miTS-P cells, which serves as a useful tool to distinguish iPS and iTS-P cells that are generated using the same method. As iPS cells and iTS-P cells are generated by the same reprograming factors, KP-1 is useful for distinguishing between iPS cells and iTS-P cells. Our data further reveal differences between pluripotent stem cells and tissue-specific stem cells as well as the differences in expression of ABC transporters by human and mouse ES cells. KP-1 is therefore useful for selecting stem cells and for readily predicting the expression of patterns of ABC transporters.

Methods
Mice and cell culture. The University of the Ryukyus review committee approved experiments using mice. C57/BL6 mice (24-weeks-old) (CREA) provided the source of primary pancreatic tissue. Mouse pancreata were digested in 2 ml of cold M199 medium containing 2 mg/ml collagenase (Roche Boehringer Mannheim). The digested tissues were incubated in Dulbecco's modified Eagle's medium (DMEM; Invitrogen) containing 15% fetal bovine serum (FBS; BIO-WEST). NOD/scid mice (8-weeks-old) (CREA) were used to study teratoma formation. Mouse ES cells (ATCC) and iTS-P cells were maintained on feeder layers of mitomycin C-treated STO www.nature.com/scientificreports/ cells cultured in complete ES cell media containing 15% FBS (Millipore) as previously described [25][26][27][28] . ES cells were passaged every 4 days, and iTS-P cells were passaged every 4 to 5 days.

Generation of iPS and iTS-P cells through replicon transfection. Generation of iTS-P cells was
conducted as previously described 28   www.nature.com/scientificreports/ medium. Advanced DMEM containing 200 ng/mL B18R protein was supplied every day until iTS-P colonies were generated.
KP-1 treatment. mES (RIKEN BRC Cell Bank, Tsukuba, Japan), miTS-P, and a mixture of miPS/miTS-P cells (2 × 10 5 cells per well) were added to a six-well plate with mouse STO feeder cells. mES and miTS-P cells were similarly prepared without feeder cells. After cultured cells were incubated with 2 μM KP-1 (Goryo Kayaku Co. Ltd., Sapporo, Japan) for 3 h at 37 °C, cells were rinsed with PBS, and a fluorescence microscope was used to acquire images.

Flow cytometric analysis of mES and miTS-P cells.
After treating mES and miTS-P cells with KP-1, washing twice with ice-cold PBS, and dissociating them using 0.25% trypsin-EDTA, cells were counted using a Novocyte Flow Cytometer (ACEA Biosciences, Inc., San Diego, CA, USA) according to the manufacturer's instructions.
Microarray analysis. Microarray analysis was conducted as previously described 28 . Total RNA from ES or iTS-P cells was labeled with biotin. Samples were hybridized using a GeneChip 3´ IVT PLUS Reagent Kit (Affymetrix, Tokyo, Japan) and a CeneChip Hybridization, Wash, and Stain Kit (Affymetrix, Tokyo, Japan) according to the manufacturer's protocol. Arrays were scanned using a GeneChip Scanner 3000 7G (Affymetrix, Tokyo, Japan), and were analyzed using Transcriptome Analysis Console 4.0 software (Thermo Fisher Scientific, Waltham, MA, USA).
RT-qPCR analysis. RT-qPCR analysis was conducted as previously described 28  ABC transporter inhibitors. After 5 days of passage, miTS-P were treated with 2 µM KP-1 for 3 h. Cells were rinsed with PBS before microscopy. The cells were then treated with 5 µM cyclosporin A (Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) (an Abcc10 inhibitor) or 50 µM sildenafil (Selleckchem.com, USA) (an Abcc5 inhibitor) for 2 h, retreated with 2 µM KP-1 for 3 h and rinsed with PBS.
RNA interference. After 5 days of passage, miTS-P cells were treated with 2 µM KP-1 for 3 h at 37 °C. Cells were rinsed with PBS before microscopy. The miTS-P cells were then transfected with 30 pmol siRNA of Abcc10, Abcc12, Abcc5, or Abca2 (Thermo Fisher Scientific, CA, USA) and cultured for 24 h. The miTS-P cells were then retreated with 1 µM KP-1 for 2 h at 37 °C and rinsed with PBS.

Statistical analysis.
Data are expressed as the mean ± SE. Two groups were compared using the Student t test. Differences between groups were considered significant if p < 0.05. All methods were performed in accordance with the relevant guidelines and regulations.

Data availability
All datasets are available from the corresponding author upon reasonable request.