Conceived and designed the experiments: MYC DK CHK KSK. Performed the experiments: MYC DK CHK HCK EY JIM SK JP YC. Analyzed the data: MYC DK CHK HCK JIM SK KSP KAL DYH YC RL KSK. Wrote the paper: MYC DK CHK KSK.
Current address: Department of Pediatrics, Severance Children's Hospital, Yonsei University, Seoul, Korea
Among the coauthors, Dr. Young Chung and Dr. Robert Lanza are employed by Stem Cell International Inc. However, this does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
Given the usefulness of rats as an experimental system, an efficient method for generating rat induced pluripotent stem (iPS) cells would provide researchers with a powerful tool for studying human physiology and disease. Here, we report direct reprogramming of rat neural precursor (NP) cells and rat embryonic fibroblasts (REF) into iPS cells by retroviral transduction using either three (Oct3/4, Sox2, and Klf4), four (Oct3/4, Sox2, Klf4, and c-Myc), or five (Oct3/4, Sox2, Klf4, c-Myc, and Nanog) genes.
iPS cells were generated from both NP and REF using only three (Oct3/4, Sox2, and Klf4) genes without c-Myc. Two factors were found to be critical for efficient derivation and maintenance of rat iPS cells: the use of rat instead of mouse feeders, and the use of small molecules specifically inhibiting mitogen-activated protein kinase and glycogen synthase kinase 3 pathways. In contrast, introduction of embryonic stem cell (ESC) extracts induced partial reprogramming, but failed to generate iPS cells. However, when combined with retroviral transduction, this method generated iPS cells with significantly higher efficiency. Morphology, gene expression, and epigenetic status confirmed that these rat iPS cells exhibited ESC-like properties, including the ability to differentiate into all three germ layers both in vitro and in teratomas. In particular, we found that these rat iPS cells could differentiate to midbrain-like dopamine neurons with a high efficiency.
Given the usefulness of rats as an experimental system, our optimized method would be useful for generating rat iPS cells from diverse tissues and provide researchers with a powerful tool for studying human physiology and disease.
The cloning of Dolly the Sheep over a decade ago demonstrated that adult somatic cells could reprogrammed back to a state of pluripotency
The rat animal model is one of most valuable models for the study of numerous human diseases as well as for therapeutics development. For instance, 6-OHDA lesioned rats is one of most popular animal model for Parkinson's disease (PD)
In this study, we sought to establish an efficient procedure to generate iPS cells from two different rat tissues, neural precursors (NPs) and rat embryonic fibroblast (REF), by introducing total extracts from ESCs and/or retroviral transduction of defined transcription factors. We found that introduction of ESC-extracts into rat NP cells failed to generate iPS cells inducing only partial reprogramming. However, rat iPS cells were successfully generated from both NPs and REF by retroviral transduction of reprogramming factors with or without c-Myc, and the efficiency was significantly improved when these two methods were combined. Notably, we established an optimal procedure to generate and maintain rat iPS cells by culturing the cells on REF instead of mouse embryonic fibroblast (MEF) as the feeder in the presence of mitogen-activated protein kinase kinase (MEK) and glycogen synthase kinase 3 (GSK3β) inhibitors (PD0325901 and CHIR99021, respectively). Rat iPS cells derived from our optimized procedure exhibited ESC-like properties by morphological, gene expression, epigenetic status, proliferation, and differentiation criteria. In particular, we show that these rat iPS cells can efficiently differentiate to multiple neuronal lineages including midbrain-like dopaminergic neurons which will serve as invaluable platform for bioassay and cell transplantation studies of PD.
We employed neural precursor (NP) cell culture from micro-dissected cortices from rat embryonic day 14 (day of conception = day 0). Time-pregnant Sprague-Dawley (SD) rats were purchased from Charles River Laboratories. INC. (Wilmington, MA). All animal procedures were performed in accordance with National Institute of Health guidelines and were approved by the Animal Care and Use Committee (IACUC) at McLean Hospital, Harvard Medical School.
Embryonic cortices were dissected from rat embryos and mechanically dissociated in Ca2+/Mg2+-free Hank's balanced salt solution (CMF-HBSS). Cells were plated at 8000 cells/cm2 on 10 cm tissue culture dishes pre-coated with poly-L-ornithine (PLO; 15 µg/ml) at 37°C two hours followed by fibronectin (FN; 1 µg/ml) overnight. NPs were allowed to proliferate in the presence of 20 ng/ml basic fibroblast growth factor (bFGF; R&D Systems, Minneapolis, MN) in serum-free medium (N2) for 4–6 days
Induced pluripotent stem (iPS) cells were generated and maintained in ES medium, Dulbecco's modified Minimal Essential Medium (DMEM, Invitrogen, Carlsbad, CA), supplemented with 2mM L-glutamine (Invitrogen, Carlsbad, CA). 1mM β-mercaptoethanol, 1x non-essential amino acids (NEAA; Invitrogen, Carlsbad, CA), 15% fetal bovine serum (FBS, Sigma-Aldrich, St. Louis, MO), 100 U/ml penicillin, 100 µg/ml streptomycin (Invitrogen) and 2000 U/ml leukemia inhibitory factor (LIF; Chemicon, Termecula, CA) supplemented with signal inhibitors, CHIR99021 (3 µM; Axon Medchem, Groningen, Netherland) and PD0325901 (0.5 µM; Axon Medchem). IPS cells were maintained on feeder layers of mitomycin C (10 µg/ml media, Sigma-Aldrich)-treated REF cells. For picking and passaging, rat iPS cells were washed once with ES medium and then mechanically picked (until passage 10) or incubated with 1 mg/ml collagenase type IV (Stem cell Technology, INC., Vancouver, Canada) for 10 min. An appropriate volume of the medium was added, and the contents were transferred to a new dish on REF feeder cells. The split ratio was 1∶1 (until passage 5) and after routinely 1∶3. For feeder-free culture of iPS cells, the plate was coated with gelatin (Stem cell Technology, INC.).
Mouse ESCs (J1) were propagated in vitro using feeder-free conditions without signal inhibitor supplementation. ESC-extracts were prepared when cultures reached 70–80% confluence. To prepare ESC-extracts, cells were washed with PBS once followed by one wash with cell lysis buffer (100 mM HEPES, pH 8.2, 50 mM NaCl, 5 mM MgCl2, 1 mM dithiothreitol, and protease inhibitors), followed by sedimentation at 400 g, suspension in 1 volume of cold cell lysis buffer, and incubated for 30–45 min on ice. Cells were sonicated. The supernatant was aliquoted, frozen and stored at −80°C. Lysate from 10 million J1-ESCs was used to generate 100 µl of extract. Control NP or pluripotent factors infected NP cells were washed in cold PBS and in cold CMF-HBSS. Cells were suspended in aliquots of 100,000 cells/100 µl of HBSS, and centrifuged at 2,500 g for 5 min at 4°C. Sedimented cells were suspended in HBSS, and streptolysin-O (SLO; Sigma) was added to a final concentration of 400 ng/ml. Permeabilization was assessed by monitoring uptake of FITC-labeled F(ab')2-antibodies from a separate sample 24 h after resealing and replating the cells. After permeabilization, cells were suspended at 1000 cells/µl in 100 µl of ESCs extract containing an ATP-regenerating system (1 mM ATP, 10 mM creatine phosphate, and 25 µg/ml creatine kinase), 100 µM GTP (Sigma-Aldrich), and 1 mM each nucleotide triphosphate (NTP). The tube containing cells was incubated horizontally for 1 h at 37°C in a CO2-incubator with occasional agitation. After dissociation, cells were plated into gelatin-coated plates for clone formation with N2/ES media (1∶1 volume mixture) containing bFGF (1 ng/ml). To reseal plasma membranes, we add 2 mM CaCl2 to the culture media.
The cDNA encoding hOct4, hSox2, hKlf4, hNanog and hc-Myc (Open Biosystems) were subcloned into the pCL retroviral expression vector
For viral transduction, NP or REFs cultured
Total RNA was purified with Trizol reagent (Invitrogen), five micrograms of total RNA were used for reverse transcription reaction with SuperScript II (Invitrogen) and oligo-dT primer, according to the manufacturer's instructions. PCR reaction conditions were optimized to determine the linear amplification range. Amplification products were identified by size. Primer sequences were: glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
(
Alkaline phosphatase (AP) staining was performed using the Alkaline phosphatase staining kit II (Vector Vector Laboratories, Burlingame, CA).
For immunocytochemistry, cells were fixed with 4% paraformaldehyde for 20 min at room temperature. After washing with PBS, cells were treated with PBS containing 10% normal goat serum and 0.1% Triton X-100 for 35 min at room temperature. Primary antibodies included SSEA1 (monoclonal, 1∶100, Developmental Studies Hybridoma Bank, Iowa, IA), Nanog (polyclonal, 1∶300, Abcam, Cambridge, MA), Rex1 (polyclonal, 1∶200, Abcam), smooth muscle actin (SMA; monoclonal, 1∶400, Dako, Glostrop, Denmark), anti-βIII tubulin (Tuj1; monoclonal, 1∶500, Covance, Richmond, CA), GFAP (polyclonal, 1∶500, DAKO), Sox17 (monoclonal, 1∶200, SantaCruz Biotech.), tyrosine hydroxylase (TH; rabbit or sheep polyclonal,1∶1,000, Pel-Freez, Rogers, AR), serotonin (5-HT; polyclonal,1∶2000, Sigma), GABA (polyclonal, 1∶700, Sigma), choline acetyl transferase (ChAT; polyclonal, 1∶700, Sigma), nestin (monoclonal, 1∶1,000, BD Sciences, Franklin Lakes, NJ), Ki67 (monoclonal, 1∶500, Novocastra laboratories Ltd., United Kingdom), Ptx3 (polyclonal, 1∶200, Zymed), Engrailed-1 (En-1; monoclonal, 1∶100, Developmental Studies Hybridoma Bank), collagen type I (monoclonal, 1∶100, Developmental Studies Hybridoma Bank), Fibronectin (monoclonal, 1∶5, Developmental Studies Hybridoma Bank). For detection of primary antibodies, fluorescence-labeled (Alexa fluor 488 or 568; Molecular Probes, Eugene, OR) secondary antibodies were used according to the specifications of the manufacturer. Cells were mounted in Vectashield containing 4′,6-diamidino-2-phenylindole (DAPI; Vector Lab.) and analyzed under a fluorescent microscope.
Cells were harvested by trypsinization and transferred to bacterial culture dishes in ES medium without LIF. Total RNA derived from EB on day 4 was used for RT-PCR analysis. After 5 days, aggregated cells were plated onto tissue culture dishes and incubated for another 8∼10 days with serum-free ITSFn medium
Rat NP-iPS cells (#2 and #4) and RES-iPS cells (#3 and #4) were suspended in DMEM containing 10% FBS. Nude mice were anesthetized and the cell suspension was injected subcutaneously into the kidney capsule. Four to six weeks after the injection, tumors were surgically dissected from the mice. Samples were weighed, fixed in PBS containing 4% formaldehyde, and embedded in paraffin. Sections were stained with hematoxylin and eosin.
Genomic DNA from cells was performed with the DNeasy Tissue Kit (Qiagen, Valencia, CA). Bisulfite treatment was done using the EpiTect Kit (Qiagen) following the manufacturer's instruction. Bisulfite treated DNA was amplified using primers designed for methylation PCRs (
Standard G-band chromosome analysis was performed by Cell Line Genetics (Madison, WI).
Based on previous studies showing that fusion of ESCs with somatic cells induce nuclear reprogramming
Next, we tested if retroviral transduction of the five (Oct4, Sox2, Klf4, c-Myc, and Nanog: OSKMN) and/or four factors (OSKM) could reprogram rat NPs and fibroblasts to generate iPS cells. Monitoring infection efficiency with the GFP-expressing pCL-retroviral vector, revealed that the majority of rat NPs (>95%) could be transduced (data not shown). The time schedule for rat iPS cell induction is summarized in
(a) Schematic time schedule of rat-iPS generation. (b) Clone formation after pluripotent factors induction with or without ESC-extract treatment. Shown are alkaline phosphatase (AP)-positive clone numbers (black columns) from the total numbers of colony (white columns) at 20 days after transduction. (Each column from n = 14 of 6 independent experiments, error bars indicate S.E.; *
We next tested whether retroviral transduction of 3 factors (Oct4, Sox2, and Klf4: OSK) without c-Myc can also generate ESC-like colonies. Although the efficiency was lower than 5 and 4 factors transduction, AP-positive ESC-like colonies could be generated by 3 factors without c-Myc (
Notably, these NP or REF-derived ESC-like colonies lost ESC-like morphology and AP activity when maintained under conventional murine ESC culture condition (
Using our established procedure, we further tested if combination of retroviral transduction and ESC extract treatment can improve the efficiency of reprogramming. Toward this end, NPs were permeabilized and treated with ESC-extracts six days post retroviral transduction (
For further analyses, we selected 11 iPS-like clones derived from NP (5 by OSKMN, 3 by OSKM, and 3 by OSK) and 9 clones derived from REF (3 by OSKMN, 2 by OSKM, and 4 by OSK)(
(a) Representative high magnified image of rat iPS cells grown on feeder. (b) Both rNP-iPS and REF-iPS cells exhibited strong alkaline phosphatase (AP) activity and were homogeneously labeled with antibodies against SSEA1 (green) and Nanog (red). (c) No karyotypic abnormalities were observed in REF-iPS #2. (d) iPS clones derived from rat neural precursor (rNP-iPS #1∼#4) and fibroblast (REF-iPS #1∼#4) express ESC markers. RT-PCR analysis of ES marker genes, Oct4, Nanog, ECAT1, ESG1, FGF4 and REX1. Rat neural precursor cell (rNP) and rat embryonic fibroblast cells (REF) were used as negative control. GAPDH was used as a loading control. (e) DNA methylation status upstream of Nanog and Oct4 in rat PC12, rat NP, and rNP-iPS clone (#2) using sodium bisulfite sequencing. The top panel indicates the CpG dinucleotide position of the Nanog and OCT4 promoter regions and the numbers show positions of CpGs relative to the translation start site. Each PCR product was subcloned and subjected to nucleotide sequencing analysis. Eight representative sequenced clones are depicted by open (unmethylated) and filled (methylated) circles for each CpG site.
The above gene expression of ESC markers strongly suggests that epigenetic reprogramming has occurred in these rNP-iPS and REF-iPS cells. To address this possibility, we next investigated the epigenetic status of the Oct4 and Nanog gene promoters in rat iPS cells (
To analyze the differentiation potential of these rat iPS cells in vitro, we used floating culture to induce EBs. After suspension culture, ball-shaped EB structures were generated from all rat-iPS clones. We attached these EB-like structures to tissue culture plate and maintained them for 8 to 10 days in ITSFn media
(a) Embryoid body (EB) mediated in vitro differentiation of rat iPS clones. Upper panel show phase contrast images of differentiated cells at 10 days after EB attachment. Definitive endoderm-like (left), contracting muscle-like (middle) and neuronal-like (right) cells are shown. Immunocytochemical analysis for the three germ layer differentiation was performed 10 days after EB attachment (Lower images). Sox17 (green, endodemal; left), smooth muscle actin (SMA, red, mesodermal; middle) and Tuj1 (green, ectodermal; right). Nuclei were stained with DAPI (blue). (b) Hematoxylin and eosin staining of teratoma derived from rFC-iPS cells (#2 and #3). Cells were transplanted into the kidney capsule of three SCID mice. A tumor developed from one injection site. Images are from a teratoma containing intestinal epithelium, respiratory epithelium (both endodermal); cartilage, muscle (both mesodermal); neural tube, epidermis (both ectodermal).
To test in vivo pluripotency, we transplanted two rNP-iPS clones (#2 and #4) and two REF-iPS clones (#3 and #4) into the kidney capsule of nude mice. In all four cases, teratoma formation was observed in the kidney capsule at four to six weeks after injection (
We examined the differentiation potential of rNP-iPS and REF-iPS cells to specialized neural lineage cell types including dopaminergic neuronal fate, based on the five stage in vitro differentiation method
(a) Scheme of the rat iPS differentiation method used to induce dopaminergic neurons from rat iPS cells. Bottom pictures (i∼v) show representative image of each stage. (b) The five-stage neuronal differentiation method induced diverse neural differentiation in rat iPS cells. Six days after withdrawal of the mitogen bFGF at day 21 of the differentiation procedure, expression of nestin (b, left, red), Tuj1 (b, left, green; b, right, red) and GFAP (b, right, green) confirmed the neural identity of the REF-iPS derived neural differentiation. (c) Diverse subtypic neurons are expressed during rat iPS derived neuronal induction. Serotonergic (5-HT, red, left), GABAnergic (GABA, red, middle) or cholinergic (ChAT, red, right)/Tuj1 (green)-positive neurons were induced. Images were taken from REF-iPS#2 derived neuronal cells at day 21. (d) Dopaminergic neuronal differentiation from rat iPS#2-derived neuronal induction at day 21. Representative images of TH (red)/Tuj1 (green)-positive neurons derived from rat iPS cells (Inset, DAPI nuclear staining of the same field). The bottom table shows dopaminergic neuronal differentiation efficiency of rat fibroblast derived iPS cells (REF-iPS #2; from nine independent experiments, mean ± S.E). Indicated are the percent of neuronal (Tuj1+) cells per total (DAPI) cells or dopaminergic (TH+) cells per neurons (Tuj1+) at neuronal induction day 21.
The rat animal model is a critical experimental system for disease modeling, physiological, pharmacological, and behavioral studies of human
Salient features of this study are as follows. First, our optimized procedure generated multiple rat iPS clones from both NP and embryonic fibroblasts with similar efficiencies without genetic selection suggesting that it may be generally applicable to diverse rat primary tissues. Second, as shown in recent studies to derive rat ESCs
Based on previous studies showing that somatic cell fusion with ESCs induced epigenetic reprogramming of somatic chromosomes
Taken together, our results demonstrate that AP-positive colonies can be efficiently generated from NP and REF cells (approximately 0.2%) and approximately 20% of them became iPS-like colonies with the ESC morphology. Using our optimized procedure, we established 20 rat iPS clones (11 from NP and 9 from REF) and extensively characterized 4 NP-derived and 4 REF-derived clones in this study. Among them, three iPS clones (rNP-iPS#4, REF-iPS#3 and #4) were generated without c-Myc. All eight clones exhibited the molecular and cellular properties of fully reprogrammed iPS cells such as ESC-like morphology, cell proliferation, endogenous gene expression patterns, epigenetic status, in vitro and in vivo pluripotency. These clones were stably maintained for at least 25 passages with normal karyotype. Thus, our results strongly suggest that ESC-like iPS cells can be derived from genetically unmodified rat tissues including NPcells and fibroblasts. We previously presented some results of this study
Neural precursor (NP) cells isolated from rat E14 cortices are homologous cell populations and are actively mitotic. NP cells were plated at 10,000 cells/cm2 and expanded 3 days with bFGF (20 ng/ml) prior to immunocytochemical analyses. Left, a representative image of cells that are mostly positive for nestin (NP marker; red). Only a minor fraction of cells (typically <2%) are positive for TuJ1 (neuronal marker; green). Right, nestin-positive NP cells (green) were double labeled with the proliferation marker, Ki67 (red).
(0.22 MB TIF)
Analysis of clones obtained from cortical NP cells by treatment with ESC-extracts. (a) Representative morphology of clones that are obtained at 2 weeks after treatment with ESC-extracts (right) and untreated NP cells (left). (b) RT-PCR analysis of mRNAs isolated from clones that are generated by treatment with ESC extracts and untreated controls. Some of these clones (approximately 20%) expressed Nanog, but none of them expressed Oct4. (c,d) Differentiation potential of these partially reprogrammed clones. NP cells without ESC treatment became almost completely astrogenic and only differentiated to astrocytes (c, left). In contrast, most clones obtained from ESC-extracts treated NPs, differentiated to astrocytes and neurons with comparable efficiencies (c, right). Differentiated cells were analyzed by immunostaining with antibodies against the neuronal marker, Tuj1 (green) and astrocytic marker, GFAP (red). Quantitative analyses of the differentiation potential of clones obtained from treatment with ESC-extracts (d). Results are presented as the mean ± SEM of % GFAP+ and TuJ1+ cells in the total cell population. (n = 15 from three independent experiments, *P<0.001).
(0.38 MB TIF)
Efficiencies of colony formation by feeder variants. Rat NP-derived ESC-like colony formation is influenced by feeder variants, i.e., mouse embryonic fibroblast (MEF) vs. rat embryonic fibroblast (REF) feeders after treatment with the five, four or three factors. Twenty-days post retroviral transduction analysis showing alkaline phosphatase (AP)-positive clones, on MEF (white) and REF (black) feeder (*P<0.001).
(0.05 MB TIF)
(a) Efficiencies of colony formation after treatment with the five, four or three factors from fibroblasts (REF) and neural precursor cells (NP). Twenty-days post retroviral transduction analysis showing alkaline phosphatase (AP)-positive clones, in black and the total numbers of colonies in white. (Each column from n = 12 (FC) or 14 (NSC) of 6 independent experiments, error bars indicate S.E.) (b) Efficiencies of AP-positive clones (black bar) and SSEA1 and AP double positive clones (patterned bar). These clones are derived from 50,000 NPs by treatment with 4 factors (O,S,K,M) or 3 factors (O,S,K) at 20 days post transduction.
(0.06 MB TIF)
Karyotypic analysis of rat neural precursor cells derived iPS clone #4. No karyotypic abnormalities were observed.
(0.09 MB TIF)
(a) Representative image of REF-iPS cells exhibited strong Rex1 (red; middle) activity. (b) Semi-quantitive RT-PCR analysis of endogenous (endo-) and transgenic (trans-) retroviral Oct4, Klf4 and Sox2 expressions in rat-iPS clones derived from rat neural precursor (rNP-iPS #1, 2 and 4) and fibroblast (REF-iPS #1, 3 and 4). All lines were at passage 10∼14. Expression of endogenous ES marker gene, Rex1, was used as control.
(0.17 MB TIF)
In-vitro and in-vivo differentiation of rNP-iPS clones (#1∼#4). (a) RT-PCR analysis of embryoid bodies (EBs) for three germ layer differentiation markers, endoderm (Foxa2), mesoderm (Brachyury) and ectoderm (βIII-tubulin, Tuj1). (b) Immunocytochemical analysis for differentiation to the three germ layer was performed 10 days after EB attachment. Sox17 (green, endodemal; left), desmine (green, mesodermal; middle), and GFAP (green, ectodermal; right). Nuclei were stained with DAPI (blue). (c) Teratoma derived from rNP-iPS cells. Hematoxylin and eosin staining of teratoma derived from rNP-iPS cells (#2 and #5). Cells were transplanted into kidney capsule of three SCID mice. A tumor developed from one injection site. Each image shows formed teratoma (up/left), cornea-like epithelium (endodermal; down/left), adipose tissue (mesodermal; up/middle), muscle tissue (mesodermal; down/middle), epidermis (ectodermal; up/right) and pigmented retinal epithelium (ectodermal; down/right).
(0.64 MB TIF)