All the animal treatment processes and the protocols we used were approved by the Animal Care and Use Committee of Guangdong Province, China. The approval ID or permit numbers are SCXK (Guangdong) 2011-0029 and SYXK (Guangdong) 2011-0112.

To obtain the 5′-flanking region of this gene, pYIPF3 mRNA sequence (GenBank No. NM_001245980) was blasted in the high throughput genomic sequences database (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Then a porcine genomic DNA clone (GenBank No. CU928765.2) containing the gene was isolated and approximately 2,000 bp of 5′-flanking sequences was amplified with the primers YIPF-proF/R (Table 1).

Tissues were obtained from three Landrace barrows. Total RNA was extracted from the tissues using Trizol reagent (Invitrogen, Carlsbad, CA, USA). RNA quality was assessed by agarose gel electrophoresis and the ND-2000 Spectrophotometer (Thermo Scientific, Waltham, MA, USA) and was discarded if the 260/280 ratio was not between 1.8 and 2.1. The RNA was then transcribed into cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA). Real-time polymerase chain reaction (RT-PCR) was performed using Hieff qPCR SYBR Green Master Mix (YEASEN Bio, Shanghai, China) on a LightCycler 480 System (Roche, Basel, Switzerland). Beta-actin gene (β-actin) was used for endogenous control. The PCR was then performed in triplicate and the gene expression levels were quantified relative to the expression of endogenous control. The primers used in this study were list in Table 1.

To further illustrate the YIPF3 expression changes at transcription level after virus infection, a stimulation experiment of PK15 cells with polyinosine-polycytidylic acid (poly I:C) was performed. Poly I:C, an interferon inducer, is a synthetic mimetic of viral dsRNA and induces immune responses similar to a viral infection [13]. PK15 cells were plated in 6-well plates at an amount of 2.5×10⁵ cell/well. When cells reached about 80% confluence, the medium was replaced with fresh challenge media, with or without poly I:C (Sigma-Aldrich, St Louis, MO, USA) at a final concentration of 2 μg/mL and 10 μg/mL, respectively. Cells were harvested at 6, and 12 h. Triplicate cell samples were collected at each time point. Individual RNA from the three samples at each time point was measured in duplicate.

In order to gain knowledge about the subcellular localization of porcine YIPF3 protein, coding sequence of pYIPF3 gene was amplified using YIPF-C-F/R primers (Table 1) and subcloned into the EcoRI/XhoI site of the pEGFP-C1 vector (Clontech, Takara Bio Inc., Mountain View, CA, USA) to yield a mammalian expression plasmid pYIPF3-GFP. After sequencing verification, the plasmids were extracted using an Endo-free Plasmid Mini Kit (Omega Bio-Tek, Inc., Norcross, GA, USA). PK15 cells were seeded onto cover slips in 6-well plates. Transient transfections of pYIPF3-GFP and pEGFP-C1 (control) were performed with lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Twenty-four hours after transfection, the cells were fixed at 37°C for 15 min with 3.7% formaldehyde, and then incubated with 10 μM Hoechst 33342 for 10 min at room temperature. Finally, the images were obtained and analyzed using the Olympus FluoView FV1000 Confocal Microscope (Olympus, Tokyo, Japan).

To determine transcription start sites (TSS) of pYIPF3, a primer-extension RT-PCR method was utilized [14,15], which is a procedure that employs a same antisense primer located in exon 1, and various overlapping sense primers ranging from nt −27 to nt +51 (Table 1). The PCR products were subcloned into the XhoI/HindIII site of pGL3-Basic (Promega, USA). The plasmids were co-transferred into 293T cells in quadruplet with an internal control pRL-TK (Promega, USA). Empty pGL3-Basic vector with pRL-TK was also transfected in parallel as an external control. Twenty-four hours after transfection, the activities of firefly luciferase in pGL3 and Renilla luciferase in pRL-TK were measured on BioTek Synergy HT Multi-Mode Microplate Read (BioTek, Winooski, VT, USA).

According to the result, fragments with highest activity were selected to be analyzed using the PROMO web tool and TFSEARCH. Transcription factor binding sites with high scores and interesting characteristics were mutated using QuickChange Lightning Site-Directed Mutagenesis Kit (Stratagene, LaJolla, CA, USA) with three pairs of primers (Table 1), multisite-directed mutagenesis was also performed using multisite-directed mutagenesis kit (Stratagene, USA) with above primers. After sequencing verification, firefly luciferase and Renilla luciferase activities of these mutational vectors were measured as above.

Nuclear protein extracts were prepared from 293T cells using Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific, USA). The protein concentration of each nuclear extract was detected by Pierce BCA Protein Assay Kit (Thermo Scientific, USA). All probes with 22 nucleotides were labeled at 5′ end with Biotin (Table 1, Probes). In the competition experiments, unlabeled oligonucleotides or mutated oligonucleotides without Biotin were added in 50-fold excess. When testing the effect of antibody on the mobility complex, 5 μL of the antibody was added to the nuclear extracts prior to the labeled probes. The primary antibodies of ETS2 (sc-351 X, Santa Cruz Biotechnology, Santa Cruz, CA, USA), ZBTB4 (sc-514883, Santa Cruz Biotechnology, USA) were added to the binding reactions to perform super gel shift experiments. LightShift Chemiluminescent EMSA Kit (Thermo Scientific, USA) was used in the electrophoretic mobility shift assay (EMSA), and the protocol was performed according to the paper [16].

Analysis of the pYIPF3 cDNA sequence revealed that it encoded a protein of 347 residues with a highly conserved Yip1 domain in mammals. Further, there are five transmembrane domain distributed between residues Ile-149 and His-294 (data not shown) as predicted by TMHMM (http://www.cbs.dtu.dk/services/TMHMM/).

The expression pattern of pYIPF3 mRNA showed that it has highest expression level in lymph, followed by lung, spleen and liver. Lower expression levels were detected in backfat, kidney, stomach, small intestine, brain and heart, whereas almost no expression in longissimus dorsi and biceps femoris (Figure 1).

The YIPF3 fusion protein was found to distributed in Golgi apparatus and cytomembrane (Figure 2), which was consistent with its transmembrane domain predicted by TMHMM. Moreover, many documents about the location and function of yeast Yip1p and Yif1p also supported this result. Green fluorescence was detected throughout the control cells transfected with GFP vector alone (data now shown).

As described previously, hYIPF3 gene usually expresses in fetal spleen, adult bone marrow CD56⁺ NK and CD19⁺ B cells, which drove us to further investigate its response under mimic viral infections. Our results demonstrated that pYIPF3 mRNA in control cells was only occasionally detected. Transfection of poly (I:C) led to a significantly higher expression compared with the control. Moreover, the levels of mRNAs were markedly induced under administration of high concentration (10 μg/mL) poly I:C rather than low concentration (2 μg/mL) (Figure 3). In contrast to 6-hour poly (I:C) stimulation, the 12-hour stimulation provided higher mRNA expression levels. As a result, it seems that poly (I:C) triggered pYIPF3 expression in a time- and concentration-dependent manner.

Like human homologous genes, pYIPF3 has 9 exons, and the total length of the gene body is about 4,833 bp (Figure 4A). In which, the potential TSS was denoted as +1. Electrophoresis showed that all sense primers up to primer 6 yielded PCR products, indicating that these sequences were part of pYIPF3 mRNA, whereas the 7th sense primers failed to yield PCR products indicating that it was not part of pYIPF3 mRNA (Figure 4B). Based on the RT-PCR results, it was concluded that the TSS might distribute on regions nt −27 to nt −14, located between the 3′ ends of primers YIPF-TSS(6)F and YIPF-TSS(7)F. Agreeing with this, the “C” nucleotide of TSS (Figure 4C) predicted by the promoter predictor (http://www.fruitfly.org/seq_tools/promoter.html) was actually found in this region with highest score 0.98.

To find the important region required for transcription activity within the 5′-flanking sequence of pYIPF3 gene, the DNA clone extending from nt −1,789 to nt +74, was truncated into serial fragments and inserted into the pGL3-Basic vector (Figure 4C). As shown in Figure 5A, all constructs pGL3-1877 to pGL3-286, but not pGL3-130, presented high transcription activity. This result suggested that pGL3-130 lacks important activation elements mediating pYIPF3 transcription. The longer construct pGL3-286 displayed a highest transcription activity than the others, indicating that the core promoter is located from −198 to −42 (Figure 5A). Using the prediction tool of transcription factor binding site, we found that this fragment harbors three potential binding sites for multiple transcription factors including ETS2 (at −150 to −143 of antisense strand), ZBTB4 (at −130 to −123) and ZBTB14 (at −88 to −79 of antisense strand), which are all highly conserved among three species (Figure 5B). Deletion of ETS2 or ZBTB4 binding sites of pGL3-286 caused 39% or 52.5% drop of transcription activity, respectively, and almost abolishment of promoter activity was observed in the double-deletion plasmid pGL3-286-mut3 (Figure 5C). These experiments highlight the transcription enhance roles for both ETS2 and ZBTB4. On the contrary, mutation of ZBTB14 binding site resulted in prominently increase of transcription activity even when either of ETS2 and ZBTB4 binding sites was mutated simultaneously (Figure 5C).

Our study found that the three binding sites were crucial for promoter activity of pYIPF3. Thus, these binding sites were used as probes in EMSA. Obviously, the probes corresponding to the three cis-elements were combined by unknown proteins, because they all had a lag in the lane 2 compared to lane 1 (Figure 6). Compared with lanes 2, the bands of all lanes 3 in Figure 4 became less intense in terms of competition with unlabeled wild-type probe. It indicated that the binding of probes for proteins was specific. Consistent with expectations, the mutation probes did not weaken the bands (Figure 6A, 6B).

As there was no suitable antibody for ZBTB14, super shift assays were only performed for ETS2 and ZBTB4 potential binding sites. The results clearly showed that the bands were significantly lagged when the antibodies were added (lane 7 in Figure 6A and 6B), which indicated that the probes can be combined by ETS2 and ZBTB4 proteins, respectively. When immunoglobulin G antibody was added, there was no similar lag band, which further indicated that the binding was specific (lane 9 in Figure 6A, B).