Efficient Production of Fluorescent Transgenic Rats using the piggyBac Transposon

Rats with fluorescent markers are of great value for studies that trace lineage-specific development, particularly those assessing the differentiation potential of embryonic stem cells (ESCs). The piggyBac (PB) transposon is widely used for the efficient introduction of genetic modifications into genomes, and has already been successfully used to produce transgenic mice and rats. Here, we generated transgenic rats carrying either the desRed fluorescent protein (RFP) gene or the enhanced green fluorescent protein (eGFP) gene by injecting pronuclei with PB plasmids. We showed that the transgenic rats expressed the RFP or eGFP gene in many organs and had the capability to transmit the marker gene to the next generation through germline integration. In addition, rat embryonic stem cells (ESCs) carrying an RFP reporter gene can be derived from the blastocysts of the transgenic rats. Moreover, the RFP gene can be detected in chimeras derived from RFP ESCs via blastocyst injection. This work suggests that PB-mediated transgenesis is a powerful tool to generate transgenic rats expressing fluorescent proteins with high efficiency, and this technique can be used to derive rat ESCs expressing a reporter protein.

strong red or green fluorescence, as detected by a stereo fluorescence microscope (S165, Leica, Germany). The emission wavelength peaks were detected between 510 nm and 600 nm using in vivo imaging instruments (IVIS, PerkinElmer, USA), indicating that the marker genes had been integrated into the genomes by transposition (Fig. 1D, supplementary Fig. S1A,S1B).
We sacrificed one RFP rat to assess whether the PB transposon had integrated into different organs. All eight organs assessed (the stomach, heart, liver, lung, intestine, kidney, brain and spleen) showed strong red fluorescence ( Fig. 2A), indicating that the PB transposon had been introduced efficiently. A PCR analysis of the RFP gene in the eight organs further confirmed this result (Fig. 2B). These data showed that RFP and eGFP (see supplementary Fig. S2A) rats can be efficiently generated using genomic transposition.

Germline transmission of RFP and eGFP.
We assessed the gametes of the founder rats to determine whether the RFP and eGFP marker genes can be transmitted to the progeny through the germline. The data revealed that testes and germ cells from male rats were RFP-positive under a fluorescence microscope (Fig. 2C,D). However, mature spermatids without a cytoplasm were RFP-negative, indicating that RFP was located inside the cytoplasm (Fig. 2E). Furthermore, we performed a fluorescence-activated cell sorting (FACS) analysis to determine the percentage of RFP-positive cells in haploid germ cells (round spermatids and mature spermatids). In the RFP testes group, 34.4% of the haploid peak gated cells were RFP-positive, whereas no RFP-positive cells were observed in the wild type (WT) control haploid cells (see supplementary Fig. S2B). These data showed that male RFP rats can generate gametes with the RFP modification. In a parallel experiment, female RFP rats were assessed for the formation of RFP germ cells. The data showed that ovaries and germ cells at the mature MII oocyte and germinal vesicle (GV) stages (Fig. 2F,G and supplementary Fig. S2C) were RFP-positive, which implied that female RFP rats can also form gametes with the RFP modification by PB transposon. Furthermore, the RFP-positive rats grew to adulthood, and full-term pups (F1) carrying RFP expression were obtained after crosses with WT rats (Fig. 2H and supplementary Table S1). In this assay, the fluorescent eGFP gene is also an ideal genetic marker for rats (data not shown). In conclusion, the PB-introduced RFP or eGFP marker gene can be stably inherited by the next generation through germline transmission.
Integration of the PB transposon in the rat genome. We investigated the insertion sites of the PB transposon in the genomes of the RFP and eGFP rats. Among the 10 insertion sites, 5 were inserted into an intron and 5 were inserted into intergenic sites (Fig. 3A). The five genes with PB insertion in an intron were Macf1, Tyw1, Ptpn3, Dact1 and Rad51b, which are located on chromosomes 5, 12, 5, 6 and 6, respectively (Fig. 3B). Some of the rats had a single copy insertion, but many others had multiple copy insertions (Fig. 3C). We explored the genomes of 21 RFP rats and 24 eGFP rats by inverse PCR to evaluate the efficiency of the transgene transfer (see supplementary Fig. S3A). Our data showed that the PB transposon could efficiently deliver the exogenous genes into the rat genome (see supplementary Fig. S3B,S3C). A southern blot assay was performed to identify the number of PB transposon copies. Our data showed that the PB transposon could integrate into the genome at multiple sites (Fig. 3D), which is consistent with the inverse PCR result (see supplementary Fig. S3B). Furthermore, we explored the entire genome to test whether the PBase vector integrated. Although PBase integration occurred in both RFP-and eGFP-positive rats, some of the offspring were PBase-free, with stable integration of the marker genes ( Fig. 3E and supplementary Fig. S3D).

RFP-positive ESCs derived from the PB-integrated rats.
We next performed rat ESC derivation experiments using RFP blastocysts harvested from the PB-integrated DA strain rats. RFP was stably expressed in both E4.5 blastocysts and derivative outgrowths after 5 days of culture in N2B27 medium supplemented with "2i" (PD0325901 and CHIR99021) and leukemia inhibitory factor (LIF) 20,21 (Fig. 4A). Standard rat ESC cell lines that expressed RFP were established and passaged in vitro for long periods. The expression of alkaline phosphatase (Fig. 4B) and pluripotent marker genes (Fig. 4C) indicated that the PB insertion did not affect the core pathways regulating pluripotency. We assessed whether RFP would be silenced during the ESCs culture procedure using FACS analysis. The results indicated that the percentage of RFP-positive cells at passage five was 98.7% (Fig. 4D) and was 97.8% after passage 12, with no subsequent significant decrease (Fig. 4E). Hence, PB integration can stably introduce RFP genes into the genome of the derivative ESCs, and the expression remained constant.
We produced chimeras by injecting blastocysts to examine the differentiation potential of these RFP-positive rat ESCs (see supplementary Fig. S4A). In two independent donor cell lines, 39 offspring were derived from 86 reconstructed and transferred embryos (Table S2). Ten chimeric rats were generated, with the contributions distinguished by coat color (Fig. 4F) and fluorescence detection (Fig. 4G). In summary, genomically integrated, In this study, we developed a simple and efficient method to generate transgenic rats that carry fluorescent marker genes. Our results showed that RFP and eGFP rats can be produced by microinjection of PB vectors carrying RFP or eGFP genes into the pronuclei of zygotes. We further showed that the PB system and fluorescence genes can synergistically form a powerful tool for genetic research in mammals.

Discussion
Sleeping beauty was the first DNA-based transposon system used for genomic engineering in mammalian cells. However, because PB was proven to efficiently deliver genes in mice in 2005 22 , it has been widely used in gene modification. Without requiring DNA synthesis, the PB transposon can excise itself by forming a hairpin structure 2 , which provides a seamless excision with no "footprint". Recently, the combination of CRISPR/Cas9 and the PB transposase system was used to produce a model to track and transform neocortical progenitors and provided a new strategy to study genetic function 23,24 . The protocol of co-injecting PB and PBase-encoding mRNA into pronuclei was specifically developed to improve the efficiency of producing transgenic animals and prevent re-transposition events, and can obtain average transformation frequencies of 80% 25 . Cytoplasmic microinjections of hyPBase mRNA and pPB-CAG-TagRFP DNA showed that 94.4% of blastocysts were TagRFP-positive 26 . However, it is a leap from the cellular level to an individual animal, and the injection of the PBase mRNA has great potential as an easier and highly effective method to generate transgenic animals. In this study, PBase did not show active state in the generated eGFP/RFP transgenic rats, which might relate to powerful reproductive ability of rodents. But application of this technique in larger farm animals warrants more investigations. To date, the PB system has proven itself as a versatile genetic tool for various applications, such as mutagenesis and transgenesis. Compared with sleeping beauty and other conventional viral vectors, such as retrovirus AAV and adenovirus, the PB system has a larger cargo size and a protein domain fusion that can flexibly modify the transposase to achieve site-directed integration 2 .
The PB transposon is preferentially biased toward transcription units, particularly in regions that contain genes. We found that out of 10 randomly selected rats, five had insertion sites in introns and five had insertion sites in intergenic regions. This system produced fluorescent rats with no deficiencies. In addition, some rats had a single copy insertion, whereas others had multiple insertions. As PB is prone to integrate near or within coding units, we have not clearly determined the reason why the rats did not carry insertion sites in exons or promoter regions.
Our study provided a new method to generate genetically modified rats expressing a fluorescent protein in different organs. RFP pups were produced by mating RFP rats with WT rats, which indicated that the fluorescent

Construction of the vectors. Basic PB (PB533-A) and transpose vectors were purchased from SBI System
Biosciences and then modified. The desRed (RFP) or enhanced green fluorescent protein (eGFP) genes were inserted into the PB533A-1 vector at the EcoRI and BamHI sites.
Pronuclear microinjection. The PB plasmid (30 ng/μ l) carrying the RFP or eGFP genes and the PBase vector (10 ng/μ l) were co-injected into pronuclei of zygotes harvested from 0.5 dpc (days post-coitus) SD and DA strain rats to produce transgenic embryos. The reconstructed embryos were transferred to the oviducts of pseudo-pregnant female rats.
Inverse PCR and genotyping. The schematic of the inverse PCR was illustrated in Fig. S3A. Briefly, the genomic DNA from each sample was extracted using the MicroElute Genomic DNA kit (Omega, USA) and digested with BstYI for 16 hours at 37 °C. The BstYI enzyme was then inactivated at 80 °C for 30 min. The ligation reaction conditions were 4 °C for 16 hours after the direct addition of T4 DNA Ligase (NEB, USA) into the inactivated digestion reaction. The PCR experiments (primers are listed in the supplementary information; see supplementary Table S3.) were performed under the following conditions: 95 °C for 5 min, followed by 25 cycles of 95 °C for 30 sec, 58 °C for 30 sec and 72 °C for 2 min, and 72 °C for 10 min for terminal replication (first round); and 95 °C for 5 min, followed by 32 cycles of 95 °C for 30 sec, 60 °C for 30 sec, and 72 °C for 90 sec, and 72 °C for 10 min for terminal replication (second round). The products of the inverse PCR were cloned into the pMD ™ 18-T vector (TaKaRa, China) and then sequenced by Sanger sequencing with the M13F primer. The sequence files were analyzed and aligned using the BLAST tool (NCBI, USA) on the official website.
The fluorescent cassettes in the transgenic rats were amplified in a PCR assay under the following conditions: 95 °C for 5 min, 32 cycles of 95 °C for 30 sec, 59 °C for 30 sec, and 72 °C for 1 min, followed by 72 °C for 10 min.
A pair of primers (supplementary Table S3) was designed to amplify the PBase DNA in the genome, under the following conditions: 95 °C for 5 min, 32 cycles of 95 °C for 30 sec, 60 °C for 30 sec, and 72 °C for 1 min, followed by 72 °C for 10 min.
Southern blot. Genomic DNA from the transgenic rats was digested with BamHI and EcoRV at 37 °C for 16 hours, and then separated on 0.8% agarose gels prior to Southern analysis. The probe was synthesized using the Prime-a-Gene Labeling kit (Promega, lot:0000182475), which used the Neo cassette as the template, and the probe was labeled with alpha-P32dATP.
ESC derivation. Blastocysts labeled with RFP or GFP were seeded on mitomycin-C-treated mouse embryonic fibroblasts (MEFs) and cultured in the previously reported standard "2i" rat ESC medium, which was N2B27 supplemented with "2i" (PD0325901 and CHIR99021) and leukemia inhibitory factor (LIF) 27 . Outgrowths were picked manually and trypsinized into single cells with 0.05% trypsin-EDTA. The cells from each outgrowth (one cell line) were cultured in a new well of the plate in rat ESC medium on feeder cells. The medium was changed daily for each cell line, and the cell lines were passaged every other day.
AP staining and karyotype analysis. AP staining was performed according to the standard manufacturer's instructions using the alkaline phosphatase kit. The results were observed under an inverted microscope (DMi-8, Leica, Germany). Karyotype analysis was performed according to standard methods 28 .
Chimera production. Rat blastocysts were collected from 4.5 dpc F344 strain female rats and injected with 10 to 12 RFP-or GFP-labeled ESCs. The reconstructed embryos were cultured in mR1ECM (246 mOsM) 29 in a 37 °C incubator with 5% CO 2 for 30 min. Images of the reconstructed blastocysts were captured on an inverted microscope (DMi-8, Leica, Germany). The chimeric embryos were transferred to the uteri of pseudo-pregnant female SD rats. The chimeric rats were identified by coat color or expression of the fluorescent proteins.