Generation of Transgenic Rats through Induced Pluripotent Stem Cells*

Background: Rat induced pluripotent stem cells (riPSCs) failed to produce transgenic rats. Results: We found that an optimized induction medium improved the efficiency of iPSC generation from rat somatic cells. The riPSCs could successfully generate transgenic rats. Conclusion: We could generate high quality riPSCs that could be used to produce transgenic rats. Significance: RiPSCs can be used as a novel tool in genetic and genomic studies of the rat. The rat is an important animal model for human disease research. Using inhibitors of glycogen synthase kinase 3 and MAPK signaling pathways, rat embryonic stem cells and rat induced pluripotent stem cells (riPSCs) have been derived. However, unlike rat embryonic stem cells, germ line competent riPSCs have only been derived from Wistar rats at low efficiency. Here, we found that an optimized induction medium containing knock-out serum replacement and vitamin C improved the rate and efficiency of riPSCs generation from Dark Agouti rat fibroblasts and Sertoli cells. riPSCs maintained an undifferentiated status for >30 passages and could differentiate into various cells types including germ cells when injected into rat blastocysts. Moreover, transgenic riPSCs could be generated through the PiggyBac transposon, which could be used to generate transgenic rats through germ line transmission. riPSCs can be used as a novel tool in genetic and genomic studies of the rat.


The rat is an important animal model for human disease research. Using inhibitors of glycogen synthase kinase 3 and MAPK signaling pathways, rat embryonic stem cells and rat induced pluripotent stem cells (riPSCs) have been derived. However, unlike rat embryonic stem cells, germ line competent riPSCs have only been derived from Wistar rats at low efficiency.
Here, we found that an optimized induction medium containing knock-out serum replacement and vitamin C improved the rate and efficiency of riPSCs generation from Dark Agouti rat fibroblasts and Sertoli cells. riPSCs maintained an undifferentiated status for >30 passages and could differentiate into various cells types including germ cells when injected into rat blastocysts. Moreover, transgenic riPSCs could be generated through the PiggyBac transposon, which could be used to generate transgenic rats through germ line transmission. riPSCs can be used as a novel tool in genetic and genomic studies of the rat.
Somatic cells can be reprogrammed into pluripotent stem cells by forced expression of the four Yamanaka factors (Oct4, Sox2, Klf4 and c-Myc) (1). Induced pluripotent stem cells (iPSCs) 4 can differentiate into multiple kinds of functional somatic cells and generate all-iPSC mice by tetraploid complementation (2). They are a promising resource for cell replacement therapy with lower or no immune rejection after transplantation (3,4). iPSC technology can also be successfully used for research of specific disease models (5)(6)(7)(8). However, the efficiency of iPSC generation is extremely low and accompanied by wide variation of the quality of iPSC colonies, which may hinder iPSC applications.
The rat is used as an important model of human diseases, especially those of the neural and cardiac systems, because of similar physiological features (9). In addition, the efficiency of small organ transplantation in mice is extremely low, which hinders regenerative medicine research in mice. When using rat iPSCs (riPSCs) to generate large organs, testing the function of the transplanted organ in the rat will be easier and more consistent. However, unlike in the mouse that can be used to generate transgenic animals with embryonic stem (ES) cells and iPSCs, the use of pluripotent stem cells to generate transgenic rats has only been successful using rat ES cells. riPSCs can be generated from the somatic cells of Wistar, Dark Agouti (DA), Fischer 344 (F344) and Brown Norway rats, which differentiate into the three germ layers in teratomas and chimeric rats (10,11). However, only Wistar riPSCs can contribute to the germ line of chimeric rats (12). Whether live rats with germ line transmission can be obtained from iPSCs of strains other than the Wistar rat is unclear. Moreover, similar to rat ES cells, whether riPSCs can be used to generate transgenic rats is unknown (13,14).
Chemicals and induction medium have been reported to improve mouse and human iPSC generation (15)(16)(17), but few studies have focused on riPSCs. Previous reports have shown that knock-out serum replacement (KOSR) and vitamin C (Vc) can improve mouse iPSC generation. On the other hand, whether combined use of KOSR and Vc can facilitate riPSC generation is unclear. Here, we optimized an induction system that combined KOSR and Vc to improve riPSC generation from both DA rat fibroblasts and Sertoli cells. The riPSCs expressed pluripotency markers and could differentiate into various cell types including germ cells. We also generated transgenic rats using these riPSCs, which provides a new approach to generate rat disease models.

EXPERIMENTAL PROCEDURES
Animals-DA, Sprague-Dawley (SD) and F344 rats and CF-1 mice were purchased from Vital River Laboratories, Beijing, China, and maintained in a specific pathogen-free environment. All experiments were performed in accordance with the Beijing Animal Protection Laws of China.
Infection of REFs-The lentiviral vector 4F2A (Addgene) was used to reprogram REFs. At 48 h after transfection of 293 cells, supernatants containing virus particles were harvested, filtered, and subjected to ultracentrifugation for concentration.
REFs were plated at 2 ϫ 10 5 cells/35-mm culture dish at 1 day prior to infection. doxycycline (2 g/ml) and Polybrene (4 g/ml) were added on the first day of infection (day 0), and rLIF was added to the medium on day 1. Infected REFs were seeded onto a feeder layer in N2B27 medium containing doxycycline, rLIF, Vc, and 10% KOSR on day 2. On day 3.5, 2i was added to the medium, and ES cell-like colonies appeared from day 4.5. Colonies with a typical dome shape were picked up at day 5.5-6.5 and expanded in N2B27 medium containing 2i and rLIF.
Flow Cytometry-An Alexa Fluor 488-conjugated anti-SSEA-1 monoclonal antibody (Millipore) was used according to the manufacturer's instruction. At day 6 postinfection, SSEA-1-positive cell populations were analyzed by flow cytometry.
Genomic PCR, RT-PCR, and Quantitative Real-time PCR (Q-PCR)-Genomic DNA and total RNA were extracted by a MicroElute Genomic DNA Kit (Omega) and TRIzol reagent (Invitrogen), respectively, according to the manufacturers' instructions. cDNA was synthesized from 2 g of total RNA using Super Script III RT (Invitrogen) and an oligo(dT) primer (Promega). Q-PCR analysis was performed using SYBR Green Real-time PCR Master Mix (Toyobo) in triplicate. Gene expression levels were normalized to expression of the housekeeping gene ␤-actin. PCR primers are listed in supplemental Table S1.
Bisulfite Sequencing-Genomic DNA (200 ng) from each sample was treated with a Methylamp DNA modification sample kit (Epigentek) to convert cytosine to uracil according to the manufacturer's instructions. The promoter and enhancer regions of Oct4 were amplified by PCR using EX TaqHS, cloned into the PMD-18-T vector, and sequenced with M13 reverse and forward primers. PCR primers are listed in supplemental Table S1.
Teratoma Formation-For teratoma formation, 2 ϫ 10 6 riPSCs cells were subcutaneously injected into severe combined immune deficiency (SCID) mice. Teratomas were collected after approximately 4 weeks and fixed in 4% paraformaldehyde for paraffin embedding and hematoxylin and eosin staining following standard procedures.
Blastocyst Injection-Ten to 15 riPSCs cells were injected into blastocysts collected from the uterus of 4.5 days postcoitum female F344 rats followed by transfer into the uterine horns of E3.5 pseudopregnant female SD rats. Chimeric rats and germ line-transmitted rats were identified by coat color. Red fluorescence protein (RFP) was detected by an IVIS Spectrum system (PerkinElmer Life Sciences).
Simple Sequence Length Polymorphism (SSLP) Analysis-SSLP analysis was used to determine rat strains. DNA samples were extracted from riPSCs and the tails of littermates and chimeric rats. Tail DNA from DA, F344, and SD rats was used for the controls. The primer sequences are described in The National Bio Resource Project for the Rat in Japan.
Electroporation-Approximately 1 ϫ 10 6 riPSCs were electroporated with 4.5 g of PiggyBac vector (PiggyBac plasmid with an EF1␣ promoter-driven DsRed-IRES-neo r cassette) and 1.5 g of PBase vector (PiggyBac integrase) according to the manufacturer's instructions. Electroporated cells were cultured in N2B27 medium (without antibiotics) containing 5% KOSR and 200 g/ml G418. Neomycin-resistant colonies exhibiting RFP fluorescence were picked up and transferred to 4-well plates after 7 days of selection.
Immunofluorescence Staining of the Genital Ridge-Genital ridges from E15.5 fetuses were fixed in 4% paraformaldehyde overnight and embedded in paraffin. Fixed sections were permeabilized with 0.5% Triton X-100 in PBS at room temperature for 30 min and then incubated with 2% BSA/PBS (w/v) for 1 h at room temperature for blocking. Sections were incubated with a primary rabbit anti-DDX4/mouse vasa homolog antibody (1:500; Abcam) overnight at 4°C and then an Alexa Fluor 488conjugated donkey anti-rabbit IgG (1:400; Jackson Laboratory) for 1 h at room temperature. Hoechst 33342 was used for nuclear staining. Sections were observed under a confocal laser scanning microscope.

RESULTS
KOSR and Vc Improve the Generation of riPSCs-To test the combinatorial effects of KOSR and Vc on riPSC induction, DA REFs were transfected with the four Yamanaka factors to generate riPSCs (Fig. 1, A and B). At 36 h postinfection, REFs with marked morphological changes were subcultured onto dishes containing mouse embryonic fibroblast feeder cells and cul-tured with riPSC induction medium supplemented with or without KOSR and Vc (KOSR-Vc group versus control group). ES cell-like colonies appeared at day 4.5 postinfection, and colonies with a typical stem cell morphology were picked up at 5.5 days in the KOSR-Vc group (Fig. 1C). AP staining was performed to detect the positive number of primary riPSC colonies. The results showed that the number of AP-positive colo- nies in the KOSR-Vc group was almost 12-fold (459 versus 42) higher than that in the control without KOSR and Vc at day 5.5, and 8-fold higher (988 versus 124) than that in the control at day 7.5 postinfection (Fig. 1, D and E). We performed live staining of SSEA-1 to determine the percentage of SSEA-1-positive cells among induced cells by flow cytometry. The percentage of SSEA-1-positive cells in the KOSR-Vc group was approximately 6-fold more than that in the control group at day 5.5 postinfection (3.45% versus 0.55%) (Fig. 1F). In addition, pluripotency markers SSEA-1 and Esrrb detected by immunofluorescence showed strong expression in the KOSR-Vc group, whereas weak staining was observed in the control group at day 5.5 postinfection (Fig. 1G). Q-PCR analysis of Esrrb and Nanog expression showed 4-and 6-fold higher expression in the KOSR-Vc group than that in the control group at day 5.5, respectively (Fig. 1, H and I). These data suggest that KOSR and Vc improves riPSC generation. To date, this result is the most rapid reprogramming of riPSCs with very high efficiency.  RT-PCR analysis of gene expression in undifferentiated rESC/riPSC and embryoid bodies formed from rESC (DA5-3) and riPSCs (riPS-1), respectively. C, teratoma formation of riPSCs. All three germ layers were detected by hematoxylin and eosin staining. D, chimeric rats derived by injecting DA (Agouticoated color) riPSC into F344 (white-coated color) blastocysts. Agouti-coated color indicates the derivation from DA riPSC. E, SSLP analysis of the background of chimeric rats from riPS-1. Primer sequences were cited from The National Bio Resource Project for the Rat in Japan. F, two germ line offspring (a male and a female) produced by a chimeric rat (male, from riPS-1 cells) mated with a wild-type white SD rat (female). G, genomic PCR analysis of transgenic Yamanaka factors in germ line transmission rats. DNA from the plasmid, riPSCs (riPS-1), the chimeric rat, and SD rat were used as the controls. H, litter of pups with different coat colors produced from germ line transmission rats (female, also shown in F). I, genotyping of the pups in H for the Agouti gene. The Agouti gene (A/A) was present in DA rats, but not in SD rats.
Characterization of riPSCs-We picked up a total of nine colonies in three repeated experiments at day 5.5 postinfection and established seven stable riPSC lines (seven of nine, 77.8%) that were expanded for Ͼ30 passages. Two riPSC lines, riPS-1 and riPS-2, were randomly selected to perform subsequent experiments. We also generated four cell lines from eight riPSC colonies (riPS-c1, riPS-c2, riPS-c3, and riPS-c4; four of eight, 50%) at day 8.5 postinfection from the control group without KOSR and Vc. A rat ES cell line (DA5-3), which had previously shown germ line contribution, and REFs were chosen as positive and negative controls, respectively. RT-PCR analysis showed that the four Yamanaka factors were silenced in the two riPSC lines, whereas endogenous Oct4, Sox2, Klf4, and c-Myc were reactivated ( Fig. 2A). Karyotyping showed that the riPSCs had a normal karyotype of 42 (Fig. 2B). The riPSCs were APpositive (Fig. 2C). RT-PCR and Q-PCR analyses showed that key marker genes of pluripotency (Oct4, Sox2, Nanog, and Rex1) were highly expressed in riPSCs (Fig. 2, D and E). Western blotting and immunofluorescence confirmed a similar expression pattern of pluripotency markers (Fig. 2, F and G). Bisulfite sequencing results showed that the DNA methylation status of the enhancer and promoter regions of Oct4 were hypomethylated in riPSCs and the rat ES cell control, whereas they were hypermethylated in REFs (Fig. 2H). Taken together, riPSCs exhibited the characteristics of pluripotent stem cells at the molecular level.
Generation of riPSCs from Adult Somatic Cells-We further tested the feasibility of reprogramming adult somatic cells. Tail tip fibroblasts from 20-week-old DA rats were reprogrammed to riPSCs (Fig. 3A). ES cell-like colonies with morphologically typical features appeared under the KOSR-Vc condition at day 5.5 (Fig. 3B). Karyotyping, AP staining, and immunofluorescence showed that these riPSCs had a normal karyotype and were characterized as pluripotent stem cells (Fig. 3, C-E).
Passage 3 Sertoli cells, an interstitial cell type in a terminally differentiated state, were also isolated and transfected with the four Yamanaka factors (Fig. 3F). Morphologically ES cell-like colonies were picked up and expanded at day 5.5 postinfection (Fig. 3G). Karyotype analysis, AP staining, and immunofluorescence indicated that the reprogramming process was com-pleted in these iPSCs (Fig. 3, H-J). Thus, the KOSR-Vc combination robustly generates riPSCs from adult somatic cells.
Pluripotency of riPSCs-To determine the differentiation capacity of riPSCs, we allowed riPSCs to differentiate as embryoid bodies and evaluated the expression of differentiation marker genes by RT-PCR (Fig. 4A). We detected the expression of marker genes of the three germ layers, such as Nestin and Pax6 (ectoderm), Gata4 and Sox17 (primitive endoderm), AFP (endoderm), and Flk1 (mesoderm), whereas only slight residual expression of Oct4 similar to that in the DA ES cell control was detected at day 8 of differentiation (Fig. 4B).
Next, we examined the differentiation capacity of riPSCs in vivo. riPSCs (2 ϫ 10 6 ) were subcutaneously injected into SCID mice and assessed after 4 weeks. Histological examination showed that teratomas contained derivatives of the three germ layers (Fig. 4C). These results indicated that riPSCs could differentiate into the three germ layers.
We also performed a more stringent pluripotency assay by examining chimeric rats. riPSCs from the riPS1-6 cell line (DA background and Agouti coat color) were injected into F344 blastocysts (white coat color) and then transferred to SD pseudopregnant females to generate chimeras. A total of 72 chimeras were obtained among 132 live pups born ( Table 1). Most of the chimeric rats had a high percentage of riPSCs contribution as determined by coat color chimerism (Ͼ85%, Fig. 4D). The SSLP assay confirmed the riPSC contribution (Fig. 4E). As shown in Table 1, four riPSC lines (riPS-c1, riPS-c2, riPS-c3, and riPS-c4) from the control group were also injected into F344 blastocysts under the same condition, and 14 chimeras were obtained among 47 live pups born.

Generation of chimeras and germ line-competent offspring from riPSCs
riPS-1-6 were riPSC lines generated with the optimized induction system using KOSR and Vc. riPS c1-c4 were riPSC lines generated with the traditional induction system without KOSR and Vc. ( Fig. 4G) was confirmed by genomic PCR and showed that riPSCs contributed to the germ line. Moreover, germ line transmission rats (DA ϫ SD background) produced healthy progenies (Fig. 4, H and I). In summary, these data showed that the riPSCs were pluripotent stem cells, especially the iPSCs induced with KOSR and Vc, which showed a robust potential to contribute to the germ line.
Transgenic riPSCs Produce Transgenic Rats through Germ Line Transmission-To explore the feasibility of using riPSCs to generate transgenic rats, RFP and neo r genes driven by the EF1␣ promoter through PiggyBac transposons were electroporated into two riPSC lines (riPS-1 and riPS-2) to generate RFP transgenic riPSCs (Fig. 5A). After G418 selection for 1 week, two surviving clones expressing RFP were randomly picked up for expansion (riPS-tg-1 and riPStg-2; Fig. 5B).
Next, we injected diploid blastocysts with transgenic riPSCs. RFP-positive chimeric fetuses were harvested at E15.5, in which RFP-positive cells had widely contributed to multiple organs and tissues including the genital ridges (Fig. 5, C-E). Further analysis by immunofluorescence staining revealed mouse vasa homolog-and RFP-double positive cells within the germ line of the fetus, demonstrating the germ line contribution of RFPpositive riPSCs (Fig. 5F). Thirty-four full-term chimeric pups were obtained at day 21.5 (Fig. 5G). All chimeras exhibited an Agouti white inlaid coat color. Five adult chimeric rats from riPS-tg-1 (three males and two females) were mated with SD rats (white coat color) to test for germ line transmission. Three offspring with the Agouti color and RFP expression were obtained, indicating the competency of germ line contribution of transgenic riPSCs (Fig. 5H). We next performed genomic PCR analysis with primers designed to detect the EF1␣ promoter plus neo r and RFP genes. The data showed that germ line transmission rats inherited the transgene originating from transgenic riPSCs (Fig. 5I).
We further tested whether somatic cells from transgenic rats could be induced to pluripotent stem cells by directly adding doxycycline. Ear fibroblasts from a 1-month-old transgenic rat were isolated (Fig. 5J), and doxycycline was added daily for reprogramming. ES cell-like colonies with RFP expression emerged by day 5 (Fig. 5K). Karyotype analysis showed a normal number of chromosomes (2N ϭ 42) (Fig. 5L). AP staining was performed to demonstrate the pluripotent state of riPSCs (Fig.  5M). Taken together, these data show that transgenic riPSCs can produce transgenic rats through germ line transmission.

DISCUSSION
In this study, we showed that KOSR combined with Vc robustly improves the generation of riPSCs. The riPSCs possessed all the key features of rat ES cells, such as expression of pluripotency markers and contribution to multiple tissues including the germ line. Through germ line transmission, transgenic riPSCs produced transgenic rats.
The combination of KOSR and VC increased the rate and efficiency of riPSC generation. We have shown previously that a high concentration of KOSR instead of FBS can be used for efficient and rapid generation of mouse iPSCs (2,17). Moreover, Vc is an antioxidant that increases the proliferation of fibroblasts in the intermediate phase of reprogramming by alleviating cell senescence through a reduction of p53 levels and maintenance of intact DNA repair machinery (16). Using a combination of KOSR and Vc, we rapidly generated riPSCs at high efficiency from both fibroblasts and Sertoli cells. However, the detailed mechanism needs further investigation in the future.
Using the optimized induction system, we efficiently generated iPSCs from DA rat somatic cells. In addition, the riPSCs contributed to the germ line after blastocyst injection. Previous reports have shown that germ line-competent riPSCs can only be derived from Wistar background somatic cells with a recessive coat color. The present study provides the alternative of DA riPSCs for use in research. Compared with control iPSCs, the iPSCs generated with KOSR and Vc contributed more efficiently to the germ line, indicating that KOSR and Vc also improve the quality of riPSCs by reducing the variation of colonies. Using this robust reprogramming system, we can conveniently derive riPSCs with other backgrounds. This is the first report demonstrating that riPSCs can produce transgenic rats. This approach will be useful to study the mechanisms of diseases in rat models. However, the virus integration may have some negative effects. Generation of integration-free riPSCs from rat somatic cells and subsequent production of genetically modified rats with these riPSCs still require further study. This novel approach may improve the application of rat disease models and regenerative medicine in the future.