Production of homeobox A10 gene transgenic pigs by somatic cell nuclear transfer

XlAO Qian , ZHAO Chang-zhi, LlN Rui-yi Ll Guang-lei Ll Chang-chun WANG Hai-yan XU Jing XlE Sheng-song , YU Mei ZHAO Shu-hong

1 Key Laboratory of Agricultural Animal Genetics, Breeding, and Reproduction, Ministry of Education/The Cooperative Innovation Center for Sustainable Pig Production, Huazhong Agricultural University, Wuhan 430070, P.R.China

2 College of Animal Science and Technology, Institute of Tropical Agriculture and Forestry, Hainan University, Haikou 570228, P.R.China

Abstract Homeobox A10 (Hoxa10) gene is one of the most important candidate genes associated with the reproductive performance of humans and mice. Overexpression of Hoxa10 in mouse endometrium can increase litter size. Moreover, Hoxa10 plays a key role in regulating the embryo implantation of sows. This study aimed to generate transgenic pigs using Hoxa10 via somatic cell nuclear transfer (SCNT). We established seven Hoxa10-transgenic cell lines, and two of the cell lines were selected as nuclear donors for the transfer. A total of 1 270 cloned embryos were generated and transferred to f ive surrogate mothers (Landrace×Yorkshire). Eight cloned male piglets were produced including one with cryptorchidism. Six transgenic piglets grew up healthy and produced 56 offspring. Finally, we obtained six transgenic male pigs and 26 transgenic positive offspring that can be used to further study the regulatory mechanism of Hoxa10 on the reproductive performance of pigs.

Keywords: Hoxa10 gene, transgenic pig, somatic cell nuclear transfer, fetal f ibroblasts, oocyte

1. lntroduction

Litter size is one of the most important economic traits of pigs as it largely determines the production eff iciency of the pig industry. Improving litter size effectively using traditional breeding methods is extremely diff icult because litter size is determined by multiple factors, such as fertilization rate, ovulation rate, embryo implantation, uterine capacity and fetal survival, and its heritability is low (0.10-0.15) in a sex-limited manner (Johnson et al. 1999; Linville et al. 2001). With the development of biotechnology, transgenic technology is regarded as an effective way to solve the problem, because it can shorten the breeding time greatly and allow the production of genetically modif ied animals that can be used for basic research.

Previous studies showed that homeobox A10 (Hoxa10) gene is associated with the reproductive traits of humans and mice. The expression of Hoxa10 was detected in the fetal mouse uterus and adult human endometrium (Satokata et al. 1995; Taylor et al. 1997). Hoxa10 was also associated with endometrial receptivity (Taylor et al. 1999; Bagot et al. 2000) and embryo implantation (Taylor 2000). DNA methylation of Hoxa10 is related to eutopic and ectopic endometrium (Andersson et al. 2014). The decrease of Hoxa10 gene expression in murine uterus resulted in a signif icant decrease of implantation rate (Xu et al. 2014). Moreover, Hoxa10 regulates massive downstream target genes and Lnc RNAs related to the development of endometrium for embryo implantation (Zhu et al. 2013; Jiang et al. 2014; Lin et al. 2015). In pigs, the expressions of Hoxa10 mRNA and protein are upregulated when implantation begins in the adult porcine uterus (Blitek et al. 2010). Rare DNA sequence variations in this gene could cause developmental defects in the female internal genitalia (Ekici et al. 2013).

Although the reproduction-related function of Hoxa10 gene has already been verif ied in human and mice, it is not clear whether it can play similar role in pig reproductive traits. Additionally, somatic cell nuclear transfer (SCNT) has provided an alternative, eff icient pathway for the production of transgenic animals. Some genetically modif ied animals have been created via SCNT, such as DKK1 transgenic Tibet minipigs (Liu et al. 2017) and FBXO40 knockout pigs (Zou et al. 2018). In a previous study, an estrogen (E2)/progesterone (P4) response element in the promoter region of porcine Hoxa10 was identif ied and used to construct an expression vector pc DNA 3.1-Hoxa10 (Wu et al. 2013). We have established a transgenic mouse model with this expression vector and found that the expression level of Hoxa10 was signif icantly increased in transgenic mice compared with wild-type mice when they were treated with hormones (Lin et al. 2016). In this study, we hypothesized that the Hoxa10 transgenic pig model might be produced by SCNT. Therefore, the aim of this study was to generate the transgenic pigs expressing this hormone-inducible promoter construct by SCNT.

2. Materials and methods

All experimental protocols in this study were approved by the Ethics Committee in Huazhong Agricultural University, China.

2.1. Experimental design and treatments

The porcine fetal f ibroblasts used in this study were isolated from Yorkshire fetus which were obtained from a farm in Tianjing City of China. Completely randomized design was applied in this experiment. The cells were randomly divided into three groups in order to detect the proportions of G0/G1 phase of the cells. The cells in group 1 were cultured to 60-70% conf luence, the cells in group 2 were cultured to full conf luence and the cells in group 3 were treated with serum starvation for 3 d. Each group contained three replicate wells.

2.2. Establishment of Yorkshire fetal f ibroblast cell lines

Considering that the Hoxa10 is one of the reproductive trait candidate genes, primary cultured male Yorkshire fetal f ibroblasts were used as donor cells to generate genetically modif ied pigs in this study. Fetal f ibroblasts were isolated from 33-d old Yorkshire fetuses following the method described by Denning et al. (2001). The cells were grown in Dulbecco minimum essential medium (DMEM) (Gibco, Life Technologies, USA) with 10% fetal bovine serum (FBS) (Gibco) in 5% CO2and 100% humidity at 38.5°C. Sex determination was conducted based on the sex determining region of the Y-chromosome (SRY) (Lambert et al. 2000; Kondo et al. 2006). Two strains of male fetal f ibroblasts were selected and cultured until 8, 10 and 13 passages to evaluate the growth curves. The primers for sex determination are listed in Table 1 (SRY-F/R).

2.3. Establishment of transgenic cell colony

Hoxa10 gene expression vector (pcDNA3.1-Hoxa10) was constructed in a previous study (Appendix A) (Wu et al. 2013). At approximately 70-80% conf luence, cells in 6-well plates (Corning, Corning Inc., USA) were prepared for transfection. Liposome 2000 (6 μL) (Sigma-Aldrich, UK) was mixed with DMEM (250 μL), and the mixture was allowed to stand for 5 min. Thereafter, 250 μL DMEM containing linearized plasmid (2.4 μg) was added into the mixture, which was held constant at room temperature for 20 min. At 24 h post-transfection, the cells were selected with G418 (800 μg mL-1) for 8 to 10 d. Subsequently, the cells in the experimental group were cultured in the presence of G418 at the concentrations of 600, 400 and 200 μg mL-1for 2 d and maintained at the concentration of 200 μg mL-1. Under the selection of G418 for 15 to 20 d, the surviving cell colonies were formed, which were transferred using medical blood collection needle with a caliber of 200-300 μm into 48-well plates and cultured with G418 (200 μg mL-1) to expand the monoclonal cell lines.

2.4. ldentif ication of transgenic cells using PCR

The genomic DNA of G418-resistant monoclonal cell line was extracted by using a commercial kit (TIANamp Genomic DNA Kit, China) for positive detection. A 153-bp fragment extending from pc DNA3.1 expression vector into the promoter of Hoxa10 gene, a 294-bp fragment extending from the internal Hoxa10 gene into the pc DNA3.1 expression vector, and a 343-bp fragment of the pcDNA3.1 expression vector were amplif ied via PCR using primer sets KZQ-F/R, KDZ-F/R and ZT-F/R, respectively (the primer sequences and amplif ication conditions are listed in Table 1). The PCR products were sequenced to conf irm their identity.

Table 1 Primer pairs used in the study

2.5. Somatic cell nuclear transfer

Cumulus-oocyte complexes (COCs) were aspirated from the ovaries and matured in vitro for 42-44 h following the protocol described by Lai and Prather (2003). The capacity of donor cells under different cell-culture conditions is related to the cell cycle phase (Dalman et al. 2010). Fetal f ibroblasts were distributed into three different culture conditions (60-70% conf luence, full conf luence and serum starvation for 3 d) to examine the cell-cycle using f low cytometry. Nuclear transfer was performed under the same conditions as previously described (Wu et al. 2013). After the donor nuclei were transferred into the enucleated oocytes, the reconstructed oocytes were activated to initiate subsequent development. The activation procedure involved 1.3 kV mm-1, 30 μs, 2 direct-current pulses (1-s interval), and the fusion rate was observed after 30 min. After activation, the reconstructed embryos were cultured for 12-48 h (1-cell to 4-cell stage) prior to transfer into recipient females. At the exact time of estrus, a large number (＞230) of 1- to 4-cell stage embryos with good shape was surgically transferred into the oviduct of surrogate mothers (Landrace×Yorkshire, LY). The embryo transfer process was completed within 5-36 h after the onset of estrus.

2.6. Detection of transgenic piglets

Genomic DNA was isolated from the tail biopsies of all eight transgenic piglets, one wild-type piglet and the F1 offspring using Tissue DNA Extraction Kit (TIANamp Genomic DNA Kit). The primers used for positive detection are listed in Table 1 (KZQ-F/R, KDZ-F/R, ZT-F/R). The PCR products were sequenced to conf irm their identity. Real-time PCR was used to determine the copy number of the Hoxa10 transgene. The linearized plasmid of pc DNA3.1-Hoxa10 was mixed with wild-type genomic DNA in 100, 101, 102, 103and 104Hoxa10 gene copies, and was used to establish the standard curve. Primers of ZT1-F/R (Table 1) were used to amplify the fragment of the plasmid (pc DNA3.1-Hoxa10). Primers of TFRC-F/R (Table 1) were used to amplify the reference gene transferrin receptor (TFRC). The copy number of TFRC was used as reference to quantify the copy number of the inserted foreign DNA in the genome. The insertion site was detected using Genome Walking Kit (TaKaRa, Japan) according to the manufacturer's instructions.

2.7. Statistical analyses

The data of the proportions of G1 phase of the cells in three treatment groups was subjected to a one-way analysis of variance (ANOVA). The experimental unit was the replicate well. Fisher's least signif icant difference (LSD) test was used to evaluate differences between means, and the signif icance threshold was set to be 0.05. The one-way ANOVA and LSD test were both implemented by R Software (Team 2013).

3. Results

3.1. Establishment of Yorkshire fetal f ibroblast cell line

A few cells were dissociated from the tissues after 5 d of tryptic digestion. At the 8 d, cell quantities reached to 80-90% conf luence. Finally, seven lines of Yorkshire fetal f ibroblasts were established, and no differences in morphology among the primary cell lines were observed. After sex determination, two lines of male f ibroblasts (No. 4 and No. 6) were selected to be cultured to passage 8, 10 and 13 (named as F8, F10 and F13, respectively) for plotting the growth curves. Results suggest that these two cell lines on F8, F10, and F13 passed through three stages (latent period, logarithmic growth phase, and plateau phase) and similar growth trends (Fig. 1), suggesting that the cells could be used as donor cells for transgenic manipulation before the 13th generation.

3.2. Establishment of transgenic cell colony with Hoxa10 gene

Under selection in G418 (800 μg mL-1) for 7-8 d, cells in the negative control group were completely dead, whereas G418-resistant cells were sparsely formed, and small colonies were generated in the experimental group. After G418 selection for several days, resistant cell colonies were subsequently transferred onto 48-well plates and cultured with G418 (200 μg mL-1) to expand the monoclonal cell lines. Finally, seven G418-resistant cell lines were established. PCR analysis of DNA from resistant cell lines was carried out to determine whether the recombinant plasmids were integrated into the genome. The amplif ied products were designed to contain a part of pc DNA3.1 expression vector and Hoxa10 was inserted to enhance the reliability. The amplif ied products were successfully obtained via amplif ication of the seven cell colonies with the three pairs of primers (Appendix B).

3.3. Preparation of transgenic embryo with Hoxa10

The porcine oocytes were obtained via aspiration of large (＞3.1 mm in diameter) follicles and were cultured to maturation. After 42 h from the start of maturation, cumulus cells were removed via manual pipetting in the presence of 0.03% hyaluronidase. The oocytes with the extrusion of the f irst polar body (metaphase II arrest, MII) were selected for enucleation. As for the donor cells, we detected the proportions of G0/G1 phase of the Yorkshire fetal f ibroblast cells in three groups. The proportion of G1-phase cells in group 1 was signif icantly lower (P＜0.01) than those in group 2 and group 3 for both F8 and F13 cells, while there was no signif icant difference in proportions of G1-phase cells between group 2 and group 3 for either F8 or F13 cells (Appendix C). Two strains of monoclonal cells with good growth performance were cultured to full conf luence as nuclear donor cells. The reconstructed embryos were cultured into 1-cell stage and 4-cell stage (Fig. 2) and then transferred into the tubal ampullae of estrus sows. Each sow were transplanted with not less than 230 embryos, and three sows produced eight live born piglets in total (Fig. 3). Among these piglets, six survived for more than 20 months (Table 2). The pregnancy rate was 80% based on the proportion of pregnant recipients to the total number of recipients; the abortion rate was 25% based on the proportion of abortion recipients to the total number of pregnant recipients (Table 2).

Fig. 1 Growth curve of Yorkshire fetal f ibroblasts. A, growth curve of No. 4 cell. B, growth curve of No. 6 cell. Blue line represents the cells in the 8th passage, red line represents the cells in the 10th generation, and green line represents the cells in the 13th generation. The number of cells were normalized using Log2. Each value indicates the mean±standard deviation of three replicate wells (n=3).

3.4. Detection of transgenic piglets with Hoxa10

PCR, real-time PCR and genome walking were performed to conf irm the integration of Hoxa10 in the cloned piglets. PCR analysis indicated that the seven (one piglet died early and the data was not shown) cloned piglets and 26 F1 transgenic offspring contained the fragments of both pc DNA3.1 expression vector and Hoxa10 (Appendix D). Real-time PCR based on SYBR Green I was used to calculate the copy numbers of the recombinant vector inserted into the genome DNA of the cloned piglets. The primers were designed to be located in the region of the pc DNA3.1 expression vector due to the endogenous expression of Hoxa10. The melting curve of primer ZT1 and primer TFRC presented a unimodal pattern, suggesting the reliability of the amplif ication results (Appendix E-a). We set copy numbers of 100, 101, 102, 103and 104as standards to generate absolute standard curves (Appendix E-c). The amplif ication curves of standard samples showed that the distances of the f ive samples were unanimous and the curves were sigmoid (Appendix E-b). The copy numbers of Hoxa10 gene were greater than 3 (Table 3). Genome walking was based on nested PCR with different primers and annealing temperatures to identify unknown f lanking sequences. The products of the 3rd PCR were sequenced and aligned with the sequences of the pig chromosomes. BLAST results suggested that Hoxa10 was integrated on SSC14 in the cloned pig from No. 2 cell, and on SSC4, SSC11, SSC15 and SSC18 in the cloned pigs from No. 3 cell (Table 4).

Fig. 2 Reconstructed oocytes after fusion of 0 h (A), 24 h (B), 48 h (C), and 6 d (D), respectively.

4. Discussion

Fetal f ibroblasts are ideal nucleus donors in somatic cell nuclear transfer because they are capable of extensive proliferation and are dispersed in connective tissues throughout the fetal body (Lai and Prather 2003). Meanwhile, the status of donor cells before transplantation has certain inf luence on the development of cloned embryos. Nevertheless, very few reports exist on the number of passages wherein the donor cells could be cultured. In this study, we measured the growth curves of two lines of Yorkshire fetal f ibroblasts on the passages of 8, 10, and 13. Results showed that all the cells had experienced the latent period, the logarithmic phase, and the platform period, and different cell lines showed the similar growth trend, suggesting that the Yorkshire fetal f ibroblasts before the 13th generation could be used as nuclear donor cells.

Fig. 3 Transgenic piglets.

Table 2 In vitro development of cloned embryos reconstructed with donor cells and production of cloned pigs by nuclear transfer

Table 3 Copy number of transgenic pigs detected by real-time PCR

Table 4 Blast results between unknown sequences and pig chromosome sequences

Previous studies have shown that donor cells used for nuclear transfer in G0/G1 phase have a high blastocyst rate (Campbell et al. 1993; Linville et al. 2001; Li et al. 2006). Goto et al. (2013) compared the inf luence of G0-phase (G0-SCNT group) and early G1-phase (eG1-SCNT group) donor cells on the developmental ability of reconstructed embryos and found that the blastocyst rates of the two groups were similar, but the overall production eff iciency of the cloned offspring in eG1-SCNT groups (12.7%) was higher than that in G0-SCNT groups (3%) (P＜0.05). To isolate the donor cells out of the cell cycle, we designed three treatment groups to detect the quantities of the G1-phase cells. We found that the quantity of G1-phase cells was not signif icantly different between full conf luence and serum starvation 3 d, but the cells in the former group were of more vitality. Therefore, we chose the G1-phase cells in full conf luence as the donor cells for SCNT.

The quantity of mature oocytes also affects the development of cloned embryos. In this study, the used oocytes were derived from follicles. More oocytes from medium (3-5 mm) and large follicles (＞5 mm) formed pronuclei and developed into blastocyst compared with oocytes from small follicles (＜3 mm) (Marchal et al. 2002). Therefore, we collected oocytes from large follicles (3.1- 8.0 mm in diameter) and cultured them to maturation, resulting in the embryo fusion rate reaching up to 90%. The rate of blastocyst development was also ideal (about 20%).

In general, developmental aberrancies, such as abortion, perinatal mortality, and morbidity rate, are usually likely to occur in transgenic animals than in normal animals (Mc Creath et al. 2000; Onishi et al. 2000). Consistently, though a total of 1 270 embryos were transferred into the oviducts of f ive LY sows in this study, only three surrogate mothers were successful pregnancy to delivery and produced eight cloned piglets. One surrogate mother returned to estrus and one aborted after embryo transfer. In addition, two piglets died during the inchoate breeding, and only six transgenic piglets grew healthily and produced 56 offspring. Physiological and biochemical detection results suggest that the offspring were healthy (data not shown).

Southern blot or real-time PCR is usually used to detect the copy number of the transgene fragment. However, Southern blot hybridization is time- and labor-consuming. Therefore, many researchers adopted real-time PCR in detecting transgene insertion, which is faster and more precise than other methods. Previous studies have demonstrated that the method of real-time PCR assay can produce consistent results as Southern blot in quantitatively determining the copy numbers of transgenes (Ingham et al. 2001; Ponchel et al. 2003), indicating that the data from real-time PCR assay are reliable in calculating transgene insertion copy number. To determine the transgene integration site in the present study, we employed genome walking, which was tested to be reliable (Hasegawa et al. 2013) instead of traditional methods, such as f luorescence in situ hybridization (FISH).

5. Conclusion

We produced six Haxa10 transgenic male pigs and 26 transgenic positive offsprings via SCNT. These transgenic models could be used for the further study on the function of Hoxa10 related to the litter size of pigs.

Appendicesassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm

Acknowledgements

This study was supported by the National Major Special Project on New Varieties Cultivation for Transgenic Organisms, China (2014ZX08006-005 and 2014ZX0800950B).