Identification of a Specific Pseudo attP Site for Phage PhiC31 Integrase in Bovine Genome

1. Background
      Advanced biotechnology based on genetic engineering techniques open a window of opportunity in transgenic animals and livestock improvement. One of the main aspects in the creation of transgenic animals is gene targeting technology, which allows inserting the exogenous genes in the identified sites in mammalian genome. Phic31 integrase system has been developed for site specific gene integration and used as a tool for gene targeting in mammalian genome modification (1). This system mediated by a member of serine-catalyzed enzymes encoded by phiC31 phage. This phase initiates its life cycle by integrating its whole genome into specific sites named attB in Streptomyces bacteria genome (2). This recombination accrued when phage DNA sequence named attP annealed to attB site in bacterial genome. These sites have often ~30 base pairs (bp) in length (2). In nature, phiC31 integrase pairs a phage attP site with an attB site in the host Streptomyces genome (2). Interestingly, Groth et al., (2000) showed that these sites existed in mammalian genome with a diverse homology to native attB and attP sites and could be recognized successfully by phiC31 integrase system (1). These sites located in different regions of mammalian chromosome and named pseudo attP sites (1).  PhiC31 integrase system as a procedure in genome modification has two unique features. The system is transcriptionally active and leads to a long-term expression of integrated gene. Moreover, phiC31 integrase has limited target site in the host genome that helps site specific recombination (3, 4). PhiC31 integrase has proven to work well in a wide range of species, including plants and mammalian cells (1).
 Several studies reported recognized pseudo attp sites with diverse homology to wild type of attp and attB sequence in human, mouse, rat and bovine genome (3, 5-6). Due to the attP length , the frequency for existence of this site is very low in mammalian genome (8, 9). These sites usually were located in repetitive elements, such as L1, HERV-L repeats and integration sites (6). A few studies have shown the application of phiC31 integrase system with the aim of creating transgenic cattle by somatic cell nuclear transfer (7, 10). In these studies, four pseudo attP sites found on chromosomes 4, 10, 19 and 28 in bovine genome.
2. Objectives
New pseudo attP site, namely BF5 on bovine fibroblast chromosome, was found that was distinct from earlier reports.  For this, an inverse PCR strategy on stably EGFP transformed bovine cell line was implemented. Further direct sequencing and a BLAST search indicated that BF5 site was located on chromosome 5 and exhibited 39% identity to wild type of attP site. BF5 site could be considered for gene targeting towards producing transgenic bovine lines. 

3. Materials and Methods
Unless otherwise indicated, all chemicals and media were obtained from Sigma Chemical Company (St. Louis, MO, USA) and Gibco-BRL (BRL, Grand Island, NY, USA).

3.1. Plasmids
Plasmids pCMVInt, pBCPB+ and PBD2 (1, 11) were kindly obtained from Prof. M. P. Calos (Stanford University, USA) (Figure 1). Briefly, pCMVint encodes phiC31 integrase cDNA and pBCPB+ contains attP and attB sites, which are flanked at both sides of LacZ encoding sequence specific for intra-molecular assay. Furthermore, PBD2 encompasses coding regions for EGFP cDNA and one attB site. All plasmids were kept in DH5α strain of E. coli (Invitrogen, USA). Plasmids were purified using QIAprep Spin Miniprep Kit (Qiagen, Germany).

3.2. Primary Cell Culture
Primary bovine fetal fibroblast cell culture was established by isolating the cell from a 65 day old fetus derived from a natural mating. The skin of the fetus was washed extensively in Ca2+ and Mg2+ free phosphate buffer solution (PBS-) containing 1% (v/v) of a cocktail of penicillin-streptomycin and amphotericin B. The skin was cut into 2-3 mm2 pieces. The explants were cultured in Dulbecco’s modified Eagle medium F-12 (DMEM/F-12) containing 10% (v/v) fetal calf serum (FCS), 1% (v/v) penicillin-streptomycin and amphotericin B at 37°C in a humidified atmosphere of 5% CO2 until reaching 90-95% confluency. Cells were passaged twice, and were frozen in liquid nitrogen. Cells were thawed and passaged prior to transfection.

3.3. Transfection and Colony Selection
Approximately 2×105 fetal bovine fibroblast cells were cultured in 6-well tissue culture test plate (Orange Scientific, Switzerland) containing DMEM/F-12 enriched with 10% (v/v) FCS until reaching to 60-70% of confluency. At this stage, cells were co-transfected by either 1 mg pBCPB+ or 1 mg PDB2 with 3 mg pCMVInt (1:3 ratio) using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer’s instruction. The media were changed 6 h post transfection and cells were incubated for one more day. Cells were divided into ten 100 mm culture dishes (Falcon, USA) and cultured for 2 days. Selection was performed with 400 mg.ml-1 of G418 for three weeks. Finally, G418 resistant colonies were isolated, cultured and were expanded for cryopreservation using standard procedures.

3.4. Fluorescence Microscopic Observation and PCR Screening for EGFP Integration
Cryopreserved transfected colonies were thawed and cultured in DMEM/F-12 supplemented with 10% FCS (v/v) and G418. Antibiotic resistant colonies (4) with pronounced growth were cultured on coverslips for microscopic analysis. In order to assess the expression of EGFP in G418 resistant colonies, cells were washed with PBS- and fixed with 4% (w/v) paraformaldehyde for 20 min. After extensive wash with PBS-, cells were stained with Hoechst 33342 (bis Benzimide H33342 trihydrochloride). Samples of antibiotic resistant cells were visualized under a fluorescent microscope (Olympus, Japan) and digital images were taken with camera (DP-70, Japan). To ensure for genomic integration of EGFP, PCR was performed using EGFP-F and EGFP-R primers (Table 1). Approximately 103 cells of each colony were snap frozen in liquid nitrogen and thawed in 37°C water bath and used in PCR. The PCR steps were carried out as follows: 94°C for 5 min as an initial denaturation step, followed by 38 cycles at 94°C for 30 s, 60°C for 45 s, and 72°C for 45 s. Final extension of 72°C for 5 min was performed. PCR products were subjected to 1% (w/v) agarose (CinnaGen, Iran) in TBE buffer (90 mM Tris-HCl, 64.6 mM boric acid, and 2.5 mM EDTA at pH 8.3). 

3.5. Analysis of PhiC31integrase Functionality in Bovine Fibroblast Cells
In order to test the phiC31 integrase activity in bovine fibroblast cells, pBCPB+ plasmid was used as an intra-molecular assay vector as it contains both attP and attB sites.  In case of phiC31 integrase functionality, two att sites recombine and generate one attL site on pBCPB+ plasmid. The recombined site could be recognized by a PCR amplicon with a length of 400 bp (1) (Figure 2A). To do this, bovine fibroblast cells were co-transfected with 3 mg pCMVInt and 1 mg pBCPB+ using lipofectamine 2000 (Invitrogen, USA). Simultaneously, cells were transfected with either 3 mg pBCPB+ or pCMVInt as control. For plasmid removal from the cell surface, the cells were treated with 50 U/ml DNaseI, one day post transfection. After 72 hours of transfection, cells were harvested and extra-chromosomal plasmids were extracted using QIAprep Spin Miniprep Kit (Qiagen, Germany) (12). As a positive control, pBCPB+ was recombined in the presence of in vitro translation product of phiC31 integrase as described previously (13). PCR was performed on the extracted plasmids using attF and attR primers (Table 1) as previously described (1).  
  
3.6. Identification of Pseudo attP Sites in Bovine Genome
In order to find possible new pseudo attP sites recognizable by phiC31 integrase, an IPCR approach was implemented as described previously (7). Stably transformed bovine fibroblast cells with pCMVInt and PDB2 were harvested and genomic DNA was extracted by DNeasy Blood & Tissue Kit (Qiagen, Germany). Genomic DNA (5 mg) from each sample was digested with a couple of compatible enzymes, BglII & BamHI, (Fermentas, Lithuania). These enzymes recognize two different sites, but produce similar cohesive ends. Moreover, both enzymes cut at least one site in PDB2. The digested fragments were extracted with phenol/chloroform and precipitated with ethanol. It was important to use low amounts of DNA for appropriate self-circulation of digested DNA in the ligation reaction for efficient inverse PCR. Thus, various amounts of  DNA (0.5 to 5 ng) were prepared and used for ligation using DNA Ligation Kit (TaKaRa, Japan) as described in manufacturer’s protocol.  Semi-nested PCR was performed across the left junction of assumed recombination site according to our earlier work (14). The circulated DNA was used as a template for the first round of PCR utilizing EGFP-F and attB-F3 primers (Table 1).  The first round PCR product (1 ml) with no dilution was used as template for the second round PCR; using attB-F3 and EGFP-F (nested) primers (Table 1). The obtained bands from IPCR were purified and ligated into T-vector (InsTAcloneTM PCR Cloning Kit, Thermo, USA), and sequenced via Fazapajuh Co. (Iran). To determine genomic location of pseudo attP, the obtained sequences were analyzed by BLAST and BLAT (www.genome.ucsc.edu/cgi-bin/hgBlat) search engines against bovine genome in databases at UCSC.

3.7. PCR Screening for Site Specific Integration
For screening of site-specific recombination junctions (attL), conventional PCR was performed on cell lysates produced via freeze-thaw using BoFib-R1and attB-F3 primers. Furthermore, to determine whether pDB2 vector was integrated into specific sites of phiC31, PCR was carried out using attRR and attB-F3 primers. In the case of genomic random integration, attB arm is intact and the PCR generated a 145 bp fragment. PCR generated a 167 bp band due to disruption of the attB sequence in the genome of target cells. PCR was carried out as follows: 94°C for 5 min as initial step of denaturation, followed by 35 cycles of 94°C for 30 s; 55°C for 30 s; and 72°C for 20 s, and a final extension period of 72°C for 5 min. Amplification of a 167 bp PCR product was an indicator of site specific recombination by phiC31 integrase.  

4. Results
To examine the phiC31 integrase activity in vivo, pBCPB+, intra-molecular assay vector, was co-transfected with pCMVInt encoding phiC31 integrase in the primary bovine fibroblasts.  PCR on transfected cell lysate extracts amplified a 400-bp PCR product (Figure 2A). This result indicated the phiC31 integrase proceeded appropriate site specific recombination on pBCPB+ in bovine fibroblast cytoplasm. As a positive control, in vitro product of phiC31 integrase activity was used according to our previous report (13; Figure 2B, Lane 5). The 400 bp band was evident in co-transfected lysate cells indicating phiC31integrase activity (Figure 2B, Lane 1). Following verification of phiC31 activity in an intra-molecular assay in a bovine fibroblast cell line, the study was conducted by using cells which stably were co-transformed with  pCMVInt and pDB2 encoding EGFP. During which (21 d), the cells were incubated with G418. IPCR was performed on isolated whole genomic DNA from resistant cell lines. Amongst several bands which were amplified in this procedure, two were subsequently extracted from the gel, cloned in T-vector and sequenced.  BLAST of one amplified IPCR sequence showed 39% identity in a 28 bp-length with wild type attP (Figure 3). Interestingly, the identity was evident in cutting site of phiC31integrase (dinucleotide: TT) as seen in identified pseudo attP (Figure 3). BLAST analysis also indicated that the identified pseudo attP site was located on chromosome 5 (NCBI accession No. NW 001495037.1) and namedBF5, accordingly (Figure 4). The results of sequencing showed that the phiC31 integrase, integrated PDB2 plasmid into the host genome in a site specific manner without any deletion or insertion in BF5 site. Furthermore, BLAT analysis showed that BF5 site was among long interspersed repeat elements (LINE) in a region between 53025424-53025451 (Figure 3).
 After identification of BF5 site in bovine genome, mixed resistant cells were cultured again in presence of G418. Resistant colonies (4) with adequate proliferation rate were picked for EGFP screening (cell lines A, B, C & D). PCR reaction on genomic DNA of all selected colonies amplified a specific 723 bp fragment of EGFP (Figure 5B, upper panel). Data indicated that only one cell line out of four clones demonstrated considerable EGFP expression as observed under a fluorescent microscope (Figure 5A, Line A). While the rest of colonies showed a faint level of fluorescence (Figure 5A, Line B) or no detectable EGFP expression (Figure 5A, Lines C and D). Moreover, PCR screening on BF5 was set using a primer pair (attB-F3 and BoFib-R1; Table 1) amplifying left recombinant junction sequence (attL), which annealed to attB arm and a partial sequence of bovine chromosome 5, respectively (Figure 5C). A 167 bp PCR product was generated due to site specific recombination on BF5 site as observed only in cell line A (Figure 5B, lower panel, Line A). In order to determine whether random integration has been occurred, we screened the genomic presence of  intact attB arms using attB-F3 and attRR primers (Table 1) in a PCR generating 145 bp product (Figure 5C). As observed, PCR on cell lines A and D did not yield the expected product, while random insertion was confirmed to occur in cell lines B and C (Figure 5B, middle panel).

5. Discussion
Zinc-finger nucleases (ZFNs), transcription activator like effector nucleases (TALENs), as well as clustered regulatory interspaced short palindromic repeat (CRISPR)/Cas-based RNA-guided are newly  introduced systems which have been successfully used for gene targeting via homologous recombination technology (20, 21). Recombination stimulated by these technologies increases recombination efficiency, although off-targeting effects, introduction of mutations near the target site, as well as the required high technical support level, are several potential drawbacks of this system (22). These techniques could be replaced or at least combined in many stem cell research areas with the simpler and less technology demanding phiC31 integrase system. The phiC31 integrase system provides a relatively quick, simple, economical and reliable method for stem cell genome modification.
 PhiC31 integrase system is mainly used in gene therapy, and in creation of transgenic animals (5, 7 and 15). The most efficient feature of this system is site-specific recombination in host genome (16), which causes the long-term expression of exogenous transgene with low copy number in host genome (3).  In mammalian genome, the insertion sites, named pseudo attP, are subjective by minimal position effect of adjacent sequences. Additionally the random integration usually leads to gene silencing (17). In this paper, we have reported a new intrinsic pseudo attP site on chromosome 5 of cow, named BF5. BF5 with reasonable exogenous gene expression is located at an intergenic region, between two microRNAs, namely MIR763 and MIR2429, BF5  surrounding sequences were AT-rich (63%), suggesting the site to be appropriate for gene expression (10), similar to earlier reports  Furthermore, BF5 is noted to be among repeated elements similar to what was explained by Chalberg and coworkers. (6). It was shown that phiC31 integrase preferably integrates the DNA fragments in repetitive elements such as L1 and HERV-L repeats (18, 19).
 Several reports indicated that most integration events by phiC31 integrase in the mammalian genome result small deletion or insertions of fragments from the vector or the genome (3, 6). The results of attL sequencing in the present study showed that integration of BF5 reacted exactly in TT core without any deletion or insertion (Figure 4). The percentage of identity in core region of 28 bp between BF5 and attP was 39% (Figure 3), which was in the adequate range to be recognized by phiC31 integrase (9, 10). Among bovine fibroblast cell clones transfected by EGFP and phiC31 integrase encoding vectors, a cell line (LineA) was detected with an integration of exogenous DNA at BF5 site. In this cell line, a persistent expressed EGFP was observed more than three weeks under treatment with G418 (Figure 5).
Primer pairs specific for BF4, BF10, BpsF1 and BpsM1 pseudo attP sites amplifications were used (10).  However, no product was yielded (data was not shown). Accordingly, it can be concluded that the integration in cell Line D may be occurred in a randomized manner, causing a disruption into attB arm either via double strand break or insertional mutagenesis. To address this hypothesis, more experiments are needed to be performed.
 In conclusion, we found a new pseudo attP in bovine genome, which recognized by phiC31 integrase at the precision of integration and could be a proper site for expression of exogenous transgene in bovine fibroblast cells. Moreover, these findings could be used in future animal transgenic studies in the aim of creation of transgenic bovine.
 
Acknowledgements
    The authors send their gratitude to Prof Michelle Calos (Stanford University, USA) for his kind gifts: the vectors carrying phiC31integrase coding sequence and expression recombinant vector transferring attB sites for efficient integration.

Funding/Support
    This study was supported fully by a grant in aid of research from Royan Institute.

Conflict of interest
     None of the authors has any conflicts of interest to disclose and all authors support submission to this journal.

Implication for health policy/practice/research/   medical education
     This project provides feasible application of PhiC31 integrase for generation of transgenic animals.