Establishment of an induced pluripotent cell line from Taiwan black silkie chick embryonic fibroblasts for replication-incompetent virus production

The objective of this study was to establish a versatile cell line for replication-incompetent virus production and inactivation with formaldehyde to generate a model of cell-based vaccine manufacturing process. To achieve this goal, we took advantage of the easily accessed chick embryonic fibroblasts. Nine-day old chick embryonic fibroblasts were obtained and subjected to be transduced with a set of lentivirus to develop a chick induced pluripotent stem (ciPS) cell line. Morphological features, positive periodic acid-Schiff staining as well as strong immunocytofluorescence of alkaline phosphatase, intestinal (ALPI) and POU class 5 homeobox 1 (POU5F1) proteins suggested that these chick embryonic fibroblasts have been transformed into ciPS cells. Further differentiation and immunocytofluorescence assays confirmed that this ciPS cell line possesses capacities and potentials to form embryoid bodies, differentiate into all three embryonic layers: ectoderm, mesoderm and endoderm with evidence of strongly positive and specific molecular markers. Immunoblot analysis next demonstrated that through recombinant DNA technology and the 2nd generation lentiviral transfer system, the goose hemagglutinin gene (H5) gene was packaged into the replication-incompetent virus and highly expressed in a bladder cancer-derived cell line, T24, after transduction. The titer of ciPS-generated replication-incompetent virus is comparable to that from the Phoenix-AMPHO cell line, which is a commercial and high productive retrovirus producer. Our study successfully established a ciPS cell line which is able to produce replication-incompetent virus, providing a new strategy for cell-based vaccine production after virus inactivation.

Cell-based have several advantages over egg-based vaccines, e.g., they (1) eliminate the need for embryonated chicken eggs from managed and biosecure flocks, (2) combine the automate upstream and downstream processes, (3) reduce the potential for contamination by viable and nonviable particulates, (4) eliminate the 4 to 6 months lead times for the organization of egg supplies, (5) have faster, high-volume start-up times for production, (6) have higher initial purity, (7) could supplement seasonal vaccine supplies when multiple strain changes are necessary, (8) would substantially increase global stockpiles of pandemic influenza vaccines 6 and (9) have no risk of egg-allergies/anaphylaxis 7 , which is mainly caused by hypersensitivity to 4 allergens found in the egg white: ovomucoid, ovalbumin, ovotransferrin and lysozyme 8 . In fact, FDA estimated that about 20% improved efficacy for the cell-based vaccine relative to the egg-based vaccines for the 2017-2018 season 9 .
Pluripotent stem cells, i.e. embryonic stem cells 10 or induced pluripotent stem cells (iPSC) 11 are able to differentiate into cells of all three embryonic layers: ectoderm, mesoderm and endoderm and to form chimera or teratoma 12 . In 2006, Yamanaka's group reported 4 crucial genes: POU domain, class 5, transcription factor 1 (Pou5f1, also known as Oct3/4), SRY (sex-determining region Y)-box 2 (Sox2), Kuppel-like factor 4 (gut) (Klf4) and Myc, myelocytomatosis oncogene (Myc), which were essential for pluripotency in mouse embryonic fibroblasts 11 . Through global gene analysis, the induced cells were confirmed to display similar transcriptional and epigenetic signature to that of embryonic stem cells 13,14 . So far, several reprogramming approaches including integrating and non-integrating methods have been applied to establish iPSCs 15 . While iPSC technology is expected to revolutionize regenerative medicine, disease modeling and drug discovery in the near future 16 , here we propose that iPSCs may serve as a tool for cell-based vaccine production (inactivated virus) with a recombinant goose influenza H5 gene.

Results
Transduction of chick embryonic fibroblasts. The morphology of chick embryonic fibroblasts isolated from a 9 days old chick embryo is shown in Fig. 1A. Morphologies of fibroblasts prior-(passage #3, culture for 19 days) and post-transduction for 3, 9, 12 and 21 days are also shown ( Fig. 1B-F). After transduction and subculture for 5-6 days (passage #1), the shape of cells gradually turned into epithelial-like pattern (data not shown). In day 12 (passage #2), aggregation of epithelial-like cells were found (Fig. 1E). Starting day 21, embryonic stem cell-like cells were formed (Fig. 1F). In days 28 to 30 (passage #4), most cells aggregated in masses. Figure 2 shows the morphologies of transduced cells in passage #8 (P #8), #12, #21 and #30. Cells were continually subcultured to passage #35 (~300 days) and some cell aliquots were transferred to liquid nitrogen storage. Further culture of the freezing and thawing cells showed good survival and growth rates (data not shown). Although not all fibroblasts were successfully transduced by the Set of Lentivirus (LIN28A, NANOG, SOX2, POU5F1, KLF4 and MYC) in the beginning, only few fibroblasts were found after subculture to passage #12 to #13, due to their differentiation and unable to proliferate anymore (Fig. 2). www.nature.com/scientificreports www.nature.com/scientificreports/ Expression of chick induced pluripotent stem cells-specific markers. Several markers were examined to evaluate whether transduced chick embryonic fibroblasts were transformed into chick induced pluripotent stem cells (ciPS) cells. Histological staining indicated that strong positive periodic acid-Schiff (PAS) (glycogen-specific) staining in cells after culture for 60 and 280 days, respectively (Fig. 3A,B). Immunocytofluorescence further signified solid alkaline phosphatase, intestinal (ALPI) (Fig. 3C) and POU5F1 (Fig. 3D) expression levels in cells after culture for 280 days. These results suggested that a ciPS cell line has been successfully established.

Discussion
In this study, we found that replication-incompetent virus embracing recombinant goose influenza H5 gene could be produced in a silkie chick embryonic fibroblast-derived ciPS cell line. The titer of H5-containing replication-incompetent virus generated from ciPS cells was similar to which was produced from Phoenix-AMPHO cells. These results suggest that this ciPS cell line is competent for virus production. Phoenix-AMPHO is a cell line originated from HEK293 for the generation of helper-free and amphotropic retroviruses. Phoenix-AMPHO cells are highly transfectable with either calcium phosphate-mediated transfection or lipid-based transfection protocol, i.e., up to 50% or higher of cells can be transiently transfected. This cell line is capable of carrying episomes for long-term stable production of retrovirus 17,18 . www.nature.com/scientificreports www.nature.com/scientificreports/ Although we used fibroblasts which were derived from fertilized eggs to establish ciPS, in the long-term culture system up to 280 days, all potential allergens in the egg white have been removed during medium changes every 2-3 days. Similar to the process of Flucelvax ® production (inactivated vaccine) 19 , we inactivated H5-embracing replication-incompetent virus with formaldehyde. The INF level in culture medium from inactivated replication-incompetent virus-transduced T24 cells was significantly downregulated compared to non-inactivated replication-incompetent virus. Interferons (INFs) are a primary defense against pathogens due to their strong antiviral activities. There are three groups of INFs: type I, type II and type III, based on their genetic, structural and functional characteristics and receptors on cell surfaces. The type I INFs are the major group and comprise INFα, INFβ and INFε proteins, etc. The response of type I INFs to infections by immunodeficiency virus, hepatitis viruses, and influenza viruses have been well studied 20 . Significant downregulation of the INFα level in culture medium from the transduction of inactivated replication-incompetent viruses compared to activated ones, suggested that an effective inactivation was achieved, similar to the concept of an attenuated vaccine.
We provided evidence including the cellular morphology, positive staining of PAS, ALPI and POU5F, the capacity of formation of embryoid bodies, positive staining of specific markers for three embryonic germ layers in embryoid bodies as well as high replication-incompetent virus titer after transfection of the ciPS cells. All of the above suggested the potential pluripotent of this ciPS cell line and its efficiency for virus production. Among those genes [Eukaryotic translation elongation factor 1 alpha 1(EF1A)-driven lin-28 homolog A (LIN28A), Nanog homeobox (NANOG), SOX2, POU5F1, KLF4 and bHLH transcription factor (MYC)] we introduced into fibroblasts to establish this ciPS cell line, MYC is a well-known oncogene. Indeed, immunoblot analysis showed that the MYC protein level was higher in ciPS compared to that in fibroblasts (data not shown). Using bioinformatics tools, Zhang et al. identified 593 iPSC consensus genes. Conspicuously, of 593 genes, 209 were also expressed in human cancer cell lines and/or tissues. Moreover, 5 oncogenes were overexpressed in the iPSCs and one oncogene RAB25 (a member of the RAS oncogene family) was expressed in the iPSC-differentiated cells, signifying that these iPSC consensus genes implicated in tumorigenesis 21 . Accordingly, new strategies to downregulate the expression of oncogenic genes in iPSCs to prevent cell transformation are imperative.
To ensure the establishment of a ciPS cell line successfully, in addition to the human homologs of original mouse genes (Oct4, Sox2, Klf4 and Myc) which were reported by Takahashi and Yamanaka (2006), we moreover introduced human LIN28A and NANOG genes into chick embryonic fibroblasts to establish the ciPS cell line. LIN28A is a RNA binding protein and expressed in processing bodies, where it targets mRNA and microRNA regulation. LIN28A inhibits the expression of several microRNAs including LET-7 which regulates cell proliferation and differentiation 22,23 . LIN28A promotes the efficacy of iPSC generation in a cell cycle-dependent pathway 24 . www.nature.com/scientificreports www.nature.com/scientificreports/ Also, LIN28A-mediated post-transcriptional regulation of POU5F1 expression in human embryonic stem cells 25 . Together with NANOG, LIN28A was able to replace KLF4 and MYC to generate iPSCs from human normal lung cells 26 . The NANOG gene encodes a DNA binding homeobox transcription factor which involved in embryonic stem cells (ESCs) proliferation, renewal and pluripotency. In differentiating cells, NANOG can block ESCs differentiation 27,28 and also autorepress its own expression in a POU5F1/SOX2-independent manner 29 . Therefore, in order to avoid the expression of oncogenes such as MYC, another cell line without introduction of any oncogene is under construction. Indeed, a similar approach using the same Set of Lentivirus was earlier applied to establish a chicken-induced pluripotent cell line, BA3, to support replication and growth of Newcastle disease virus LaSota vaccine strain 30 , reinforcing the potential of ciPS cells as a tool to produce cell-based vaccine. Here, we further showed that a combination of ciPS cells and recombinant DNA technology can avoid handling danger viruses in the vaccine production process.
We used the viral integrative gene delivery method to establish the ciPS cell line, which is an efficient strategy through the transgene insertions in the chromosomal DNA. However, this viral delivery system has disadvantages involving genome integration, insertional mutagenesis and lack of silencing (leaky expression) 31 32 . In the future, iPSCs may provide an improved cell-based vaccine production, if the above mentioned methods are feasible.
In conclusion, we established a chick induced pluripotent cell line which could be used to produce high-titer replication-incompetent virus. The titer increased ~10 folds after purification and concentration. Inactivation of the replication-incompetent virus significantly downregulated the INFα level in cell culture medium after www.nature.com/scientificreports www.nature.com/scientificreports/ transduction, suggesting its potential for cell-based vaccine production. Specific molecular markers and the formation of embryoid body confirmed the pluripotency of this ciPS cell line.

Statement.
All experiments and methods were performed in accordance with relevant guideline and regulations. The animal use protocol listed below has been reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) at Livestock Research Institute, Council of Agriculture, Taiwan, and experiments (and relevant protocols) were performed/used in accordance with the relevant guidelines and regulations of LRIIACUC, Protocol # 106-3.
Isolation and culture of chicken embryonic fibroblasts. All procedures were performed in sterile conditions. Fertilized eggs from a black silkie hen were incubated for 9 days and collected. Chicken embryos were washed with Dulbecco's phosphate-buffered saline (DPBS) to eliminate the yolk and blood. The embryo was then transferred to a 10-cm Petri dish containing 10 mL DPBS to remove the head, internal organs and chopped into small pieces in a 3-mL DPBS containing 0.25% trypsin and 0.02% EDTA. In order to obtain homogenized cells, the tissue was next subjected to pass through a 18 gauge 1-inch needle with a 5-mL syringe several time and incubated in a 5% CO 2 incubator for 5 min. Exactly 10 mL of culture medium [Dulbecco's Modified Eagle's Medium (DMEM, Invitrogen) containing 10% fetal bovine serum (FBS, Invitrogen, Carlsbad, CA, USA) and 1% penicillin/streptomycin (Sigma-Aldrich, St. Louis, MO, USA) was added to the tissue sample, incubated at room temperature for 5 min and subjected to be centrifuged (2000 rpm, 5 min, 5702 R, Eppendorf, Hamburg, Germany). The supernatant was removed and 10 mL medium was added, subjected to be centrifuged again. The supernatant was removed and cells were seeded at a density of 1 × 10 6 /10-cm Petri dish and incubated in an incubator with 37 °C and 5% CO 2 . Medium was renewed when cells were confluent (~70-80%) every 2 to 3 days.
Reprogramming of the chicken embryonic fibroblasts. The Set of Lentivirus (LIN28A, NANOG, SOX2, POU5F1, KLF4 and MYC] (#LV01006L, creative biogene, Shirley, NY, USA) was used to transduce cells. Each virus tube contained 10 8 transducing units/mL. Briefly, cells were subcultured to the 2 nd -3 rd passage and seeded on 3-cm Petri dishes at a density of 6 × 10 5 overnight. Virus particles embracing the above 6 genes (3 μL/ each) were mixed with 3 mL culture medium and added onto the cells. Cells were subsequently cultured in a 37 °C, 5% CO 2 incubator for 24 h. The medium was thereafter removed and replaced with medium for chicken embryonic stem cells [DMEM with human leukemia inhibitory factor (10 units/mL, Sigma-Aldrich), human stem cell factor (5 ng/mL, Sigma-Aldrich), human basic fibroblast growth factor (10 ng/mL, Sigma-Aldrich), human insulin like growth factor 1 (10 ng/mL, Sigma-Aldrich) and human interleukin 11 (0.04 ng/mL, Sigma-Aldrich)] in the following culture and subculture. After 10 passages, mTeSR ™ 1 medium containing 20% fetal bovine serum (FBS, STEMCELL ™ TECHNOLOGIES, Taipei, Taiwan) was used to maintain the putative ciPS cell line in Corning ® Matrigel ® hESC-Qualified Matrix (#354277, Corning, NY, USA)-coated 4-well dishes. Cells were easily detached after rinse with fresh medium in the absence of trypsin and subcultured every 2-3 days.

Characterization of the ciPS cells.
To characterize whether the reprogrammed fibroblast had the features of pluripotent stem cells, PAS, ALPI and POU5F1 were detected by histochemistry and immunocytofluorescence, respectively. For PAS staining, cells were washed with PBS, fixed with 10% formaldehyde in ethanol for 5 min, washed with PBS twice, added with 1 mL periodic acid (#395B, Sigma-Aldrich), incubated for 5 min, washed with PBS for two more times, mixed with Schiff reagent (Sigma-Aldrich), incubated at room temperature for 10 min, washed with PBS twice and observed using a Axiovert 40 CFL microscope (Carl Zeiss, Oberkochen, Germany).

Formation of embryoid body, differentiation and determination of embryonic germ layers.
To form embryoid body, ciPS cells were subjected to suspending culture by hanging drops in a bacteriological Petri dish. ciPS cells were collected and cultured in 20 μL of mTeSR ™ 1 medium containing 20% FBS on the lid of 100-mm sterile Petri dish. Cells were cultured for 7 days (5% CO 2 ) and the medium was renewed every other day. At the end of culture, embryoid bodies were transferred to a 0.1% of gelatin-coated 4-well plate in the same medium for spontaneous differentiation. After culture for 14 days, formed embryoid bodies were subjected to immunocytofluorescence by probing anti-NEFL) (1:100, #AB9568, Millipore), anti-NPPA (1:100, #AB1970; Millipore) and anti-pan cytokeratin (1:100, #C2562, Sigma-Aldrich) primary antibodies, followed by hybridization with the secondary antibodies as appropriate (see also the above section). www.nature.com/scientificreports www.nature.com/scientificreports/ ATCC, Manassas, VA, USA) were seeded in a 6-well plate containing DMEM medium (Hyclone, San Angelo, TX, USA) with 1% penicillin/streptomycin (Corning ® ) and 10% FBS overnight and subjected to be transfected as a positive control. To generate the replication-incompetent virus embracing the goose influenza H5 gene, PolyJet ™ (15 μL, #SL100688, SignaGen ® Laboratories, Gaithersburg, MD, USA) was used to transfect the plasmid mixture [pH5-LAS2w.Ppuro (2.5 μg), psPAX2 (2.25 μg), PMD2.G (0.25 μg) and the medium was renewed after 16 h incubation in both cell lines.

Purification, tittering and concentration of replication-incompetent virus particles.
Media were collected at 48 h and 64 h post-transfection from both Phoenix-AMPHO and ciPS cells, mixed individually from the same cell line, filtered (0.22 μm) to remove any cell debris and aliquots of 1 mL were stored at −80 °C. The titer was measured using a QuickTiter ™ Lentivirus Quantitation Kit (Cell Biolab, Inc. San Diego, CA, USA) before and after concentration. Briefly, a standard curve was generated with 200, 100, 50, 25, 12.5, 6.25, 3.125, 1.5625 and 0 μg/mL (1:2 serial dilution) of the lentivirus RNA standard in a 96-well plate followed by reading with a GLOMAX ® -Multi Detection System (Promega, Fitchburg, WI, USA) using a 480/520 nm filter set. Exactly 100 μL of medium containing replication-incompetent virus particles was subjected to nucleic acid digestion, virus capture, protein denaturation, viral genome release and quantitation based on the user manual. Lenti-X concentrator (#TR30025, OriGene) was further utilized to concentrate replication-incompetent viral particles. Briefly, 1.5 mL Lentivirus Concentrator (5×) was added into 6 mL lentiviral supernatant, mixed by gentle inversion and incubated at 4 °C overnight. Thereafter, the reaction was centrifuged at 3,500 ×g for 25 min at 4 °C and the supernatant was removed carefully. The virus was resuspended in cold, sterile PBS (100 μL) by gently pipetting up and down and stored at −80 °C for further experiments.

Data availability
The data are shown in the main manuscript and available to readers.