A Robust Strategy for Negative Selection of Cre-LoxP Recombination-Based Excision of Transgenes in Induced Pluripotent Stem Cells

Viral vectors remain the most efficient and popular in deriving induced pluripotent stem cells (iPSCs). For translation, it is important to silence or remove the reprogramming factors after induction of pluripotency. In this study, we design an excisable loxP-flanked lentiviral construct that a) includes all the reprogramming elements in a single lentiviral vector expressed by a strong EF-1α promoter; b) enables easy determination of lentiviral titer; c) enables transgene removal and cell enrichment using LoxP-site-specific Cre-recombinase excision and Herpes Simplex Virus-thymidine kinase/ganciclovir (HSV-tk/gan) negative selection; and d) allows for transgene excision in a colony format. A reprogramming efficiency comparable to that reported in the literature without boosting molecules can be consistently obtained. To further demonstrate the utility of this Cre-loxP/HSV-tk/gan strategy, we incorporate a non-viral therapeutic transgene (human blood coagulation Factor IX) in the iPSCs, whose expression can be controlled by a temporal pulse of Cre recombinase. The robustness of this platform enables the implementation of an efficacious and cost-effective protocol for iPSC generation and their subsequent transgenesis for downstream studies.


Introduction
Reprogramming of somatic cells into a pluripotent-stem-cell-like state by retrovirus-based expression of Yamanaka transcription factors (Oct3/4, Sox2, Klf4 and c-Myc) have opened up new vistas in using lineage-specific cells for disease modeling, drug screening, developmental biology studies and cell-based therapies [1][2][3][4][5][6][7][8][9][10][11]. Since then cellular reprogramming has been achieved by the delivery of transcription factors via lentivirus, adenovirus, plasmid, transposon, mRNA or protein [12][13][14][15][16][17]. Out of these methods the lentiviral approach is preferred because of its high efficiency of transgene insertion and pluripotency induction, except for translation [18]. The integrative nature of lentiviral transgenesis would require excision of the transgenes after reprogramming. It has been demonstrated that iPSCs with their transgenes excised resemble ES cell-lines more than the iPS cell-lines that have the reprogramming transgenes unexcised [19]. Moreover, expression of the reprogramming factors may be aberrantly activated in the iPSCs to interfere with differentiation. Therefore, investigators have looked at various ways to remove the transgenes after lentiviral reprogramming.
Cre recombinase-mediated excision of LoxP-flanked reprogramming cassette has been extensively reported in the literature. This approach is usually based, firstly, on a transient expression of a GFP-tagged-Cre expression cassette in the reprogrammed cells. Subsequently, GFP+ cells are sorted and screened for transgene excision [20]. This procedure assumes that almost all the GFP+ cells will have their transgenes removed by Cre recombinase. However, due to the frequent cell division in iPSCs leading to quick dilution of the Cre recombinase transcripts and the stochastic nature of Cre recombinase-mediated excision [21], it is unlikely that most of the iPSCs will have their reprogramming transgenes excised. This method therefore mandates an extensive and laborious screening procedure for isolating the iPSCs with excised transgene. Secondly, a florescent-protein tag is included in addition to the reprogramming factors in the construct to facilitate sorting [22]. This was done by replacing the cMyc gene with mCherry (a fluorescent protein) due to space constraint in a lentiviral vector. However, exclusion of c-Myc led to a loss of reprogramming efficiency and hindered maturation of the differentiated lineages [23]. Both the aforementioned approaches require fluorescence-activated cell sorting (FACS) that apart from being costly, is not often easily accessible to the investigators. Moreover, these methods necessitate the disruption of a colony to make a single cell suspension for FACS. This step, especially for human iPSC derivation, is undesirable because pluripotent cells in isolation experience selection pressure and hence are prone to growth-promoting chromosomal aberrations [24].
After reprogramming, often a need arises to further engineer the iPSCs for downstream applications. This may be transgenes to tag stem cells for lineage tracking [25,26] or to express genes for therapeutic and developmental studies [27][28][29][30]. Antibiotic selection is often used for the selection of transgene-integrated pluripotent stem cells. However, continued antibiotic expression can affect the expression of the transgene of interest. Many protocols have been reported to remove the selection cassette by using site-specific recombinase technology [31][32][33]. However these techniques face the same drawbacks mentioned earlier. In addition to the ability to express a transgene for downstream studies it is often necessary to have a temporo-spatial control over the expression. It can be achieved by expressing Cre recombinase under the control of a tissue specific promoter [34,35] and/or by a transient temporal pulse of Cre recombinase. Therefore, these considerations (post-transgenesis removal of selection cassette and temporo-spatial control) have prompted us to develop a vectorbased solution by utilizing Cre-loxP recombinase technology with subsequent HSV-tk/ganciclovir-based negative selection.
In this study, we design an excisable loxP-flanked lentiviral construct that a) includes all the reprogramming elements in a single lentiviral vector expressed by a strong EF-1a promoter; b) enables easy determination of lentiviral titer; c) enables transgene removal and cell enrichment using LoxP-site-specific Cre-recombinase excision and Herpes Simplex Virus-thymidine kinase/ ganciclovir (HSV-tk/gan) negative selection; and d) allows for transgene excision in a colony format. To further demonstrate the utility of this Cre-loxP/HSV-tk/gan strategy in downstream application of iPSCs, we also show the incorporation of a nonviral therapeutic transgene (human blood coagulation Factor IX) in the iPSCs, whose expression can be controlled by a temporal pulse of Cre recombinase.

Vector design and construction
PlasmaDNA software (University of Helsinki) was utilized for the in silico cloning. Self-cleaving 2A-peptide-linked reprogramming factors: Myc, Klf-4, Oct-4, Sox-2 (MKOS) were cloned from the Addgene Plasmid 20866: pCAG2LMKOSimO. The MKOS fragment was cut out with EcoRI (NEB-New England Biolabs) and was subsequently blunted with DNA Polymerase I, Large (Klenow) Fragment (NEB). pWPXL (Addgene Plasmid 12257) vector was linearized with SpeI (NEB) and BamHI (NEB) which resulted in the removal of the GFP (Green fluorescence protein) fragment. The vector was blunted with DNA Polymerase I, Large (Klenow) Fragment. The fragments were then ligated with T4 DNA ligase (NEB). The ligation resulted in a plasmid which was labeled as pMKOS-WPXL. In pMKOS-WPXL the MKOS fragment was expressed by the strong EF1alpha promoter of pWPXL. In the next cloning step IRES (internal ribosome entry site)-HSV-tk (Herpes Simplex Virus-Thymidine Kinase) was PCR amplified from Addgene Plasmid 12243: pLOX-gfp-iresTK by utilizing the primers tcgagctcaagcttcgaatta and taaaggtaccgtcgagccaaa. Phusion high-fidelity polymerase (Finnzymes) was used for the amplification. The cycling parameters were: Initial denaturation: 98uC630 sec; Denaturation: 98uC65 sec, Extension: 72uC660 sec, for 35 cycles; Final extension: 72uC610 min; Hold: 4uC. The PCR product was phosphorylated with T4 Polynucleotide Kinase (NEB). It was then ligated to pMKOS-WPXL that was linearized with NdeI (NEB) and blunted with Klenow enzyme. The final form had a MKOS and a HSV-tk cassette with an internal ribosome entry site (IRES) in between them; both driven by a strong EF1 alpha promoter. The plasmid was named as pMKOS-TK-WPXL.
For demonstrating the utility of Cre-lox/HSV-tk/Gan technology in achieving removal of selection cassette and temporo-spatial control over transgene expression, the vector pCag-loxP-PGK-HSV-tk-BlastR-tpA-loxP-FIX was constructed. Gateway recom-bination cloning technology (Life technologies, Grand Island, NY) was utilized to generate the vector. The blood coagulation factor IX Gateway entry vector (pENTRY4-FIX) was constructed by inserting a codon optimized version of human factor IX cDNA (GeneArt, Regensburg, Germany) in the gateway entry vector pENTRY4. The Gateway destination vector: pCag-loxP-PGK-HSV-tk-BlastR-tpA-loxP-Dest was constructed by utilizing multiple steps of cloning.
Step 1. A CAG promoter (cytomegalovirus early enhancer element and chicken beta-actin promoter) was inserted into the destination vector pRosa26-DEST (Plasmid 21189 Addgene) just proximal to the 1 st loxP site. It was achieved by ligating the 3.2 kbps EcoRV-AscI fragment from pRosa26-DEST with the 11.7 kbps AscI-NruI fragment from plasmid Ai9 (Addgene Plasmid 22799).
Step 2. A HSV-tk cassette (cut out with SalI and EcoRI from Invivogen plasmid pORF-HSV-tk) was inserted proximal to the IRES site in pVitro1-Blast (Invivogen) by utilizing the BamHI restriction endonuclease cut site.
Step 3. HSV-tk IRES BlastR cassette was cut out from the plasmid from step 2 (NcoI and EcoRI) to be cloned distal to the PGK (phosphoglycerate kinase) promoter in PGKdtabpA (Addgene plasmid 13440) by replacing the dta (diphtheria toxin) cassette (mobilized with restriction endonucleases NcoI and BclI).
Step 4. PGK-HSV-tk IRES BlastR cassette from the plasmid in step 3 (mobilized with NotI and EcoRI) was inserted into the plasmid PGKneotpAlox2 (Addgene plasmid 13444) by replacing the PGK neomycin resistance cassette (removed by HindIII and EcoRI digestion). This step allowed for the introduction of a triple polyA STOP cassette (tpA) distal to the selection cassette.

Lentivirus production, concentration and titration
Lentiviral particles were produced by transfecting HEK 293T cells with the plasmid pMKOS-TK-WPXL along with 2 nd generation packaging plasmids (Addgene plasmid 12260: psPAX2 and Addgene plasmid 12259: pMD2.G). One million HEK-293T cells were seeded on a 75 cm 2 surface area tissue culture flask. The HEK-293Ts were cultured with 10% fetal bovine serum (FBS) (Atlanta Biologicals) in DMEM-HG (GIBCO-11960) supplemented with L-glutamine, pyruvate and MEM-NEAA (GIBCO). Calcium chloride particle-based transfection method was utilized to deliver the plasmids once the HEKs were 70% confluent. 12.5 ml of fresh medium was added to the culture two hours before the transfection. The plasmid particles for transfection were produced by mixing 14 mg of MKOS-TK, 8 mg of PSPAX2, 4.3 mg of pMD2.G, 363 ml of TE 0.1X, 192 ml of water and 62.15 ml of CaCl2 2.5M. Then 627 ml of HBS2x was added to the mixture drop by drop under continuous vortexing. After 16 hours of incubation of the HEKs with the particles the medium was changed to fresh 10 ml of media supplemented with 4 mM (final concentration) of caffeine. Media was replaced three more times with an interval of 12 hours between the changes. The last media change was done without caffeine. 40 ml of the supernatant was collected and centrifuged at 250 g for 5 min to pallet out the cellular debris. The supernatant was then filtered with a .45 mm syringe filter. The filtered supernatant was concentrated up to 506 with Amicon Ultra 100 kDa filter (Millipore, cat. code UFC910008). The centrifugation was done at 2500G. The concentrated virus was then stored at 280uC for future use.
For determining the titer of the lentivirus, 25,000 primary mouse embryonic fibroblasts (PMEFs) were seeded in each well of a 6 well plate (BD falcon). Increasing quantities of the concentrated viral supernatant: 0 ml, 2 ml, 10 ml, 25 ml, 50 ml, 100 ml, were added to each well. Sequebrene was also added to the medium at a concentration of 8 mg/ml. Medium was replaced after 2 days with Ganciclovir added to the medium at a concentration of 4 mM. Ganciclovir-supplemented media was replaced every 2 days. The cell numbers were assessed after 5 days of Ganciclovir selection under a phase contrast microscope. After 7 days of selection the cells were fixed and stained with DAPI and observed under a fluorescence microscope.

Cell culture and iPSC production
Reprogramming was induced in mouse embryonic fibroblasts (MEFs) derived from a Myh-6 GFP transgenic mouse (alpha-Myosin heavy chain promoter driven green fluorescence protein). Transgenic MEFs isolated from mouse embryos (13 d.p.c.) were expanded in standard growth media conditions [36]. For the isolation of primary mouse embryonic fibroblasts, a single pregnant female mouse was euthanized by CO2 asphyxiation. Animal procedures were conducted according to the guidelines for the care and use of laboratory animals set and approved by the Institutional Animal Care & Use Committee (IACUC) of Duke University & Duke University Medical Center. Passage 2 MEFs were seeded at a density of 25,000/well of a 6-well plate coated with .1% gelatin. After 16 hours of culture the cells were transduced with 100 ul of concentrated and titrated MKOS-TK lentivirus. 24 hours later the transduced cells were passaged to a 10 cm dish coated with 0.1% gelatin. The medium was changed to iPSC medium 2 days after the passaging. iPSC media was produced by conditioning 10% FBS in DMEM-HG supplemented with L-glutamine, sodium pyruvate and MEM-NEAA (GIBCO) with STO-SNL cell line (Soriano ES Feeder cell line SNL 76/7 STO) for 24 hours. The STO-SNL cell line was purchased from Mutant Mouse Regional Resource Center (MMRRC; Catalogue # 015892-UCD-CELL). The conditioned medium was mixed with fresh 20% FBS in DMEM-HG medium at a ratio of 1:1. The final FBS concentration of the conditioned medium was 15%. By the 10 th day of reprogramming well defined colonies started appearing. On day 14 individual iPSC colonies were picked up and passaged on Mitomycin-C inactivated feeder layer of MEFs.

Reprogramming cassette excision and verification
The excision of the MKOS-IRES-TK cassette was achieved by the transient transfection of a plasmid expressing a fusion protein of Cre recombinase and GFP (Addgene plasmid 13776: pCAG-Cre:GFP). Individual iPSC clones were passaged on a feeder layer of Mitomycin-C-inactivated PMEF (Millipore) in a 24-well TCPS plate. pCAG-Cre:GFP transfection was done by Stemfect MESC2 transfection reagent (Stemgent) following the manufacturers protocol. The iPSCs were cultured for the next 5 days with daily media change so as to allow for the excision of the MKOS-TK cassette. On the 6 th day Ganciclovir (Invivogen) selection was started at a final concentration of 4 mM. The excision of the construct in the surviving iPSC cells were probed by genomic PCR. Two sets of primers were used to probe the presence/ absence of the construct. Genomic Gapdh primers were used as a control for the PCR. Primer set 1 extended from Klf-4 to 2A (CAGGCGAGAAACCTTACCAC, AGACTTCCTCTGCCCTCTCC-202) and primer set 2 amplified a region extending from WPRE to HSV-TK (GGAG-GATTGGGGACAGCTT, CATAGCGTAAAAGGAGCAACA-466). The genomic DNA was extracted by DNAeasy Blood and Tissue kit (Qiagen). Genomic PCR was done with the terra PCR kit from Clontech. The cycling conditions were: 98uC for 2 min; 98uC for 10 sec, 60uC for 15  Generation of human blood coagulation factor IX secreting iPSCs The iPSCs were transfected with the pCag-loxP-PGK-HSV-tk-BlastR-tpA-loxP-FIX construct linearized with PacI and AscI (NEB). 10 ug of the construct was electroporated into 1 million iPSCs by using the AMAXA nucleofaction protocol A023 (Lonza). Selection of the successfully transfected cells were done by exposing the cells to 10 mg/ml of Blasticidin (Invivogen) for 4 days. Subsequently removal of the HSV-tk IRES Blast cassette was achieved by the transient transfection of CRE:GFP. Mouse antihuman FIX primary antibody (hematologic technologies, AHIX-5041) at a concentration of 20 ug/ml was used for immunocytochemistry. Alexa Fluor 488 Goat Anti-Mouse IgG (Life technologies, A-11001) was used as the secondary antibody. Forward primer: TCCATCGTGAACGAGAAGTG and reverse primer: TAGTTGTGGTGGGGGATGAT was used to detect FIX by RT-PCR. FIX chromogenic assay (Aniara, Biophen) was used to perform the functional test. 50,000 iPS cells seeded for 48 hours in a single well of a 6-well plate was used to condition the media for the chromogenic assay. 2 ml of 6 mg/ml of Vit.K-supplemented N2B27 medium was applied to the cells for 24 hours to condition the media. The conditioned media was then used for the chromogenic assay following the manufacturer's protocol.

Results
The lentiviral reprogramming construct had a strong EF-1alpha promoter which is known to resist silencing [37]. In addition, the promoter has an intron for enhancing the expression of the transcript by facilitating transcript splicing [38]. pWPXL also contains a post-transcriptional regulatory element of woodchuck hepatitis virus (WPRE) that increases transgene expression levels [39]. All the above measures were designed to enable higher expression levels with less silencing of the reprogramming factors, resulting in efficient pluripotency induction. The plasmid also had a loxP site in the 39 LTR (long terminal repeat) which was copied to the 59 LTR during the process of reverse transcription of the viral RNA, thereby flanking the reprogramming cassette. The four factors viz. c-Myc, Klf-4, Oct-3/4, Sox2 (MKOS) were stitched into one poly-cistronic construct by 2A self-cleaving peptide sequences [40,41]. The IRES-HSV-tk cassette was inserted distal to the MKOS cassette. This allowed for the expression of HSV-tk (a negative selection cassette) under the same Ef1-alpha promoter as the MKOS factors. A schematic displaying the various elements is shown in figure 1.
Titration of lentivirus was done following the generation and concentration of the virus. Increasing amounts (2 ml, 10 ml, 25 ml, 50 ml and 100 ml) of lentivirus were applied to 2.5610 4 PMEFs for the titration along with a no-virus control. On the 5 th day of ganciclovir selection it was observed that 50 ml of virus was killing most of the PMEFs (Figures 2 A-F inset). Doubling the amount of virus to 100 ml improved the selection only marginally. Therefore, we decided against trying out larger amounts of virus and settled on 100 ml for the reprogramming studies. Day 7 DAPI staining confirmed this approach (Figures 2 A-F (Figure 2 L), suggesting their pluripotent nature. Based on the morphology and staining of the colonies we performed all the subsequent experiments with the two best clones that will be henceforth denoted as clone-1 and clone-2.
In a first attempt to apply the cardiac differentiation protocol to the two clones we observed no beating foci or areas displaying green fluorescence (due to activation of the Myh6-GFP transgene) by day 9 of differentiation. To probe whether this lack of differentiation was caused by the continued expression of the reprogramming factors we stained the differentiated cells with OCT4 and SSEA-1 antibodies. We observed several OCT4+ areas (Figure 3 B), which consisted of cells that were rounded and small and had a prominent nucleus (Figure 3 A). Moreover, these areas also stained in a patchy fashion for SSEA-1 ( To prove that ganciclovir resistance of the clones arose from the excision of the transgene and not due to transgene silencing, we performed PCR on genomic DNA to amplify two different areas on the transgene with genomic Gapdh as a control. The ganciclovir-selected clones had successfully removed the transgene as evidenced by the lack of a band even after 40 cycles of amplification (Figure 3 J). RT-PCR showed that the expression levels of Oct4 and Nanog decreased in the day 9 transgene-excised differentiating cells when compared to the unexcised group. No HSV-tk and WPRE expression was observed in the excised clones on RT-PCR (Figure 3 L). SSEA-1, Oct4 and AP activity were all markedly decreased in the excised clones (Figures 3 M-P).
The excised iPSC clones demonstrated typical ES cell-like compact colony morphology composed of small rounded cells with a large prominent nucleus (Figures 4 A, B). The colonies also stained for embryonic stem cell and pluripotency markers viz. AP, SSEA-1, OCT-4, NANOG (Figures 4 C-F). To further confirm the iPSC status, semi-quantitative RT-PCR was performed with Oct4, Sox2, Klf4, Nanog, Gdf3 and Rex1 primers. Mouse ES D3 line was used as a positive control. Both excised clones were positive for all the markers (Figure 4 G). The clones were then differentiated into ectoderm (neuronal differentiation), endoderm (hepatic differentiation) and mesoderm (cardiac differentiation).
The cells stained for Alpha Feto protein (AFP) which is expressed in the hepatocytes during early development (Figures 5 A, B) and Tuj1 which is a neuronal marker (Figures 5 C, D). Robust cardiac differentiation was evidenced by the appearance of multiple green Myh6 GFP+ beating areas (Video S1 & S2) and positive staining with cardiac alpha-ACTININ by the 9 th day of differentiation  ( Figures 5 E, F). On RT-PCR the differentiated cells were positive for the markers of early cardio-vascular mesoderm (Nkx2.5 and Flk-1), endoderm (AFP and Hnf3b) and ectoderm (Pax6 and Wnt1) (Figure 5 G).
We used the same platform for a proof-of-concept study to insert a therapeutic transgene (human blood coagulation factor IX) in the iPSCs albeit through a non-viral electroporation-based method. Selection of the cells for transgene integration was done with blasticidin antibiotic. Cre recombinase was used to excise the antibiotic cassette followed by HSV-tk/ganciclovir negative selection of the cells. Figure 6 A is a schematic diagram for the vector pCag-lox-PGK-HSV-tk-BlastR-tpA-lox-FIX pre-and postexcision. In the unexcised form, HSV-tk and blasticidin resistance protein are expressed from the PGK promoter. The tpA (triple polyA) STOP signal prevents any leaky expression of FIX before Cre-mediated excision of the selection cassette by acting as a strong transcription termination signal. Factor IX immunostaining (Figures 6 B, C, D (Figure 6 A). The excision event also led to the removal of the tpA signal, thereby triggering CAG promoter-driven expression of the FIX gene. Figures 6 E, F and G show the expression of FIX on ICC which is punctate in appearance and is distributed around the nucleus. This agrees with the fact that inside the cell FIX is concentrated in the endoplasmic reticulum On an average 33.9 ng of active factor IX was secreted by the excised-iPSCs (seeding density of 50,000 cells/well of a 6-well plate) in 2 ml of media over 24 hours.

Discussion
In this study we have described the utility of a technology platform (Cre-Lox site-specific transgene excision and subsequent negative selection with HSV-tk/ganciclovir) for iPSC research in two distinct ways.
Initially, the strategy was implemented in a reprogramming construct to ensure better functionality of the iPSCs. A polycistronic reprogramming construct (four reprogramming factors stitched together with the self-cleaving 2A factors) was put alongside a HSV-tk negative selection cassette in a lentiviral format. A strong EF1-alpha promoter with an intron along with WPR element was used to increase the reprogramming-transgene expression. We consistently obtained reprogramming efficiencies between 0.4 and 0.5%, which is better than what was reported in the literature with a similar kind of construct and without the addition of reprogramming boosting molecules [42]. The proviral size of the construct was approximately 10.7 kb. Reports of utilizing lentiviral vectors to package and deliver .10 kb of transgenic elements are rare in the literature [43], because of a significant decrease in lentiviral titer with such large transgenes. However, our apprehension was allayed when we were able to maintain a workable titer with our construct. We have also demonstrated the utilization of HSV-tk/ganciclovir negative selection for a simple bright-field-microscopy-based lentiviral titration. This avoids costly and laborious methods for lentiviral vector titration such as ELISA or quantitative RT-PCR. We also demonstrated that reprogramming-transgene excision is necessary as the strong EF1 alpha promoter does not get inactivated in the differentiating iPSC derivatives. This was evidenced by the large Oct4+, SSEA-1+, AP+ areas on staining and also by the expression of HSV-tk and WPRE on RT-PCR of the differentiating cells. Upon administration of ganciclovir to the differentiating derivatives, most of the cells were killed. This indicated that HSV-tk was still being expressed in most of the differentiating cells. We then showed that a transient transfection of the iPSCs with a Cre:GFP plasmid and subsequent selection with ganciclovir led to the emergence of new GFP2 colonies which were transgene-free. Massive cell death ensued after ganciclovir selection even though most of the colonies were expressing Cre recombinase. This might be attributed to Cre recombinase expression not translating into transgene excision in most of the cells, and the fact that HSV-tk expression might lead to the death of a neighboring HSV-tk negative cell on exposure to ganciclovir by a phenomenon called ''bystander effect'' [44]. The iPSCs that emerged after ganciclovir selection maintained their pluripotency as evidenced by the positive staining for pluripotent stem cell markers: AP, Oct4, SSEA-1, Nanog and RT-PCR data. The iPSCs differentiated into all the major lineages and also expressed ectoderm (Pax6, wnt1) endoderm (Hnf3B, AFP) and mesoderm (Flk1 and Nkx2-5) markers on RT-PCR, thereby bolstering the claim that the expression of HSV-tk and subsequent ganciclovir treatment along with the loss of reprogramming transgene had no effect on the maintenance of pluripotency. Moreover the cells that differentiated from the transgene-excised iPSCs showed a decrease in the expression of pluripotency markers. This may account for the appearance of robust Myh6-GFP + beating areas by day-9 of differentiation whereas the differentiating iPSCs from their unexcised counterparts showed delayed beating which appeared on day-11.
We then extended the utility of the technology to generate iPSCs expressing a model therapeutic transgene (human blood Figure 5. Immunocytochemistry and RT-PCR for differentiation markers in the differentiating transgene-excised iPSCs. The transgene-excised clones were differentiated into three germ layers viz. endoderm, ectoderm and mesoderm. Phase contrast images were taken for cells that we derived by implementing the endoderm differentiation protocol (B) and ectoderm differentiation protocol (D). Merged fluorescent images of endodermal marker: alpha feto protein (AFP) with DAPI (A) and ectodermal marker: TUJ1 (a neuronal marker) with DAPI (C) were acquired. For mesodermal differentiation cardiac differentiation protocol was implemented. Quite a few GFP+ (driven by Myh6 promoter) beating areas were observed (E) which also stained positive for a-ACTININ on ICC (F). Semi-quantitative RT-PCR was also done for differentiation markers: Nkx2.5 and Flk1 for mesoderm, AFP and Hnf3b for endoderm, Pax6 and Wnt1 for ectoderm. Gapdh was used as the PCR control (G). Both the iPSC lines pre-(1, 2) and post-excision (1Ex, 2Ex) were probed. Scale bar = 100 mm. doi:10.1371/journal.pone.0064342.g005 coagulation factor IX, FIX). To the best of our knowledge, this is the first work to demonstrate the generation of FIX-expressing iPSCs. We also showed that the non-viral construct can be used to trigger expression of the FIX transgene at a time of our choice (temporal expression control) by transient Cre recombinase transfection, while at the same time removing the antibiotic resistance cassette. This proof-of-concept study with FIX as a model can be translated to any other transgene. Application of gateway technology allows for the easy shuttling of any transgene into the destination vector to generate the working construct. The non-viral nature of the construct may allow for its use in translational studies. Moreover, the construct when placed between two homology arms can be used for targeted transgene insertion. We also showed that in this study the FIX transgene expression was triggered by transient Cre recombinase expression while the cells were still in the pluripotent state. We envisage that the robustness of this construct will also permit the triggering of the target gene in differentiating iPSCs. This ability becomes critical if a tight temporo-spatial control over the transgene expression is desired.
The methods described in this article will not only help in implementing simple, cost-effective and efficient transgenesis in iPSCs but also ensure the subsequent Cre recombinase-mediated transgene removal for full-fledged downstream functionality. Figure S1 Gateway recombination cloning for generating FIX expression vector. The gateway destination vector pCag-lox-PGK-HSV-tk-BlastR-tpA-lox-Dest was recombined with the FIX gateway entry vector (pENTRY4-FIX) by LR clonase enzyme to form the final expression vector pCag-loxP-PGK-HSV-tk-BlastR-tpA-loxP-FIX. (TIF) Table S1 RT-PCR primers for iPSCs and their differentiated derivatives.

(DOCX)
Video S1 Phase-contrast video microscopy of beating areas. (MP4) Figure 6. Temporal induction of transgenic FIX expression in iPSCs utilizing the Cre-loxP/HSV-tk/gan selection strategy. Schematic diagram for the vector pCag-lox-PGK-HSV-tk-BlastR-tpA-lox-FIX pre-and post-excision (A). In the unexcised form of the vector HSV-tk and blasticidin resistance protein is expressed from the PGK promoter. On Cre recombinase-mediated excision of the loxP flanked cassette (bottom half of the schematic) the CAG promoter (cytomegalovirus (CMV) enhancer fused to the chicken beta-actin promoter) starts expressing FIX as the tpA stop sequence is lost after the excision event. Both pre-and post-excision iPSC colonies were immunostained with FIX antibody and the nuclear stain DAPI. Phase contrast image of a pre-(B) and post-excision (E) iPSC colony merged with DAPI. Factor IX immunostaining for a pre-excision iPSC colony show only background staining (C) while that of a post-excision iPSC colony show punctate staining (F). FIX staining of a pre-excision (D) and postexcision (G) iPSC colony merged with DAPI staining image. RT-PCR for FIX and HSV-tk was done for pre-and post-excision states (H). HSV-tk was expressed only in unexcised state and FIX was expressed only in excised state. Gapdh was used as the house keeping PCR control. Mouse ES cell D3 line (MESC) was used as the negative control for the experiment. Chromogenic assay to measure the functional activity of the FIX released in the media was done for both the unexcised and excised iPSCs (I). Unconditioned media was used as a negative control. The absorbance of the chromogenic assay product at 405 nm for the media conditioned by excised iPSCs was found to be highly significant (p#.001, n = 3). doi:10.1371/journal.pone.0064342.g006 Video S2 Fluorescence video microscopy of Myh6 GFP+ beating areas. (MP4)