Verification and Analysis of Sheep Tail Type-Associated PDGF-D Gene Polymorphisms

Simple Summary PDGF-D can be considered a candidate gene for selection for sheep tail type. This study investigated genetic variation of the PDGF-D gene in sheep with different tail types verified at a cellular level and revealed the molecular mechanism of PDGF-D in sheep tail fat deposition. We detected a total of two SNPs among 533 sheep. g.4122606 C > G site was significantly correlated with tail length, and g.3852134 C > T site was significantly correlated with tail width. In addition, overexpression of PDGF-D in sheep preadipocytes can promote adipogenic differentiation. The PDGF-D gene may participate in sheep tail fat deposition and could be used for molecular marker-assisted selection of sheep tail type. Abstract The aim of this study was to examine the correlation between the platelet-derived growth factor-D (PDGF-D) gene and sheep tail type character and explore the potential underlying mechanism. A total of 533 sheep were included in this study. Polymorphic sites were examined by Pool-seq, and individual genotype identification and correlation analysis between tail type data were conducted using the matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (MALDI-TOF-MS) method. JASPART website was used to predict transcription factor binding sites in the promoter region with and without PDGF-D gene mutation. The effect of PDGF-D on adipogenic differentiation of sheep preadipocytes was investigated. Two single nucleotide polymorphism sites were identified: g.4122606 C > G site was significantly correlated with tail length, and g.3852134 C > T site was significantly correlated with tail width. g.3852134 C > T was located in the promoter region. Six transcription factor binding sites were eliminated after promoter mutation, and three new transcription factor binding sites appeared. Expression levels of peroxisome proliferator-activated receptor gamma (PPARγ) and lipoproteinlipase (LPL) were significantly up-regulated upon PDGF-D overexpression. Oil red O staining showed increased small and large oil drops in the PDGF-D overexpression group. Together these results indicate the PDGF-D gene is an important gene controlling sheep tail shape and regulating sheep tail fat deposition to a certain degree.


Population Genetic Analysis of Polymorphisms in the PDGF-D Gene
Microsoft Excel 2013 (Microsoft Inc., Redmond, WA, USA) was used to calculate the allele frequencies, polymorphic information content (PIC), heterozygosity (He), and effective number of allele (Ne). The Hardy-Weinberg equilibrium was tested for each site through the chi-square test. Association analysis between SNP genotypes and/or haplotypes and three tail traits were conducted by SAS 9.2 software, based on the following animal model: y i = µ + G i + b i + e i , in which y i was the trait measured in individual, µ was the overall mean, G i was a fixed effect corresponding to the genotype of polymorphisms, b i was the breed effect, and e i was a random residual effect.

Bioinformatics Analysis
The promoter and coding sequences of the PDGF-D gene were obtained from NCBI (https://www. ncbi.nlm.nih.gov/). The structures of transcriptional factor (TF) binding sites before and after mutations in promoter regions were also evaluated using JASPAR (http://jaspar.binf.ku.dk/cgi-bin/jaspar_db.pl).

Cell Culture and Transfection
Sheep preadipocytes were isolated from the tail fat of a 70-day-old fetus in Hu sheep as described by Cai et al. [17]. The cells were seeded in 6-well plates overnight and cultured in DEME/F12 (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, Hyclone, Logan, UT, USA) and 2% penicillin/streptomycin at 37 • C in a humidified 5% CO 2 incubator. The following day, cells were cultured in a new fresh medium containing the PDGF-D-overexpression lentiviral vector (Genechem, Shanghai, China) for 12 h. Preadipocytes were then cultured in new differentiation medium containing 10% FBS, 1 µM dexamethasone (Macklin, Shanghai, China), 0.5 mM isobutylmethylxanthine (Macklin), and 10 mg/mL insulin (Macklin) for 2 days, followed by 10 mg/mL insulin alone for 2 days. Virus without PDGF-D overexpression served as negative control. The date that cells were cultured with differentiation medium was set as the first day (1 d).

Quantitative Real-Time PCR Analysis
Cells were collected from the PDGF-D group and negative control group at various times (0, 1, 3, 5, and 7 d). Total RNA was extracted from cells using the TRIzol (Invitrogen, Carlsbad, CA, USA) method, and cDNA was synthesized from total RNA. qRT-PCR was performed using the TransStart Green qPCR SuperMix (TransStart Green, Beijing, China) and 480 II LightCycler instrument (Roche, Basel, Switzerland), and samples were analyzed in triplicate. The 2 −∆∆Ct method was used to calculate the relative expression of target genes, and ACTB (beta-actin) served as the reference gene. qRT-PCR primer information is listed in Table S2.

Oil Red O Staining
Oil red O dye was produced by combining saturated oil red O original solution (Biotopped, Beijing, China) and distilled water in a ratio of 3:2 and filtering the mixture. Cells differentiated for 7 days were washed twice with PBS, fixed with 4% paraformaldehyde for 20 min, and then rinsed with distilled water. Oil red O dye was added, and cells were incubated for 10 min. Stained cells were then rinsed with distilled water 2-3 times and observed and photographed under a microscope.

Statistical Analysis
All data reported are expressed as mean ± SE. Student's t-test was carried out using SPSS software (SPSS 17.0, Chicago, IL, USA) for statistical analysis of the data. A p value of < 0.05 was considered to be statistically significant.

SNP Detection and Genotyping
The sequencing results revealed two SNPs in the PDGF-D gene ( Figure 1): g.4122606 C > G and g.3852134 C > T. Both SNPs were genotyped and classified into three genotypes in the experimental population consisting of 208 Hu sheep, 171 Tibetan sheep, and 154 Dorper-Hu hybrid sheep ( Figure 1).
Basel, Switzerland), and samples were analyzed in triplicate. The 2 −ΔΔCt method was used to calculate the relative expression of target genes, and ACTB (beta-actin) served as the reference gene. qRT-PCR primer information is listed in Table S2.

Oil Red O Staining
Oil red O dye was produced by combining saturated oil red O original solution (Biotopped, Beijing, China) and distilled water in a ratio of 3:2 and filtering the mixture. Cells differentiated for 7 days were washed twice with PBS, fixed with 4% paraformaldehyde for 20 min, and then rinsed with distilled water. Oil red O dye was added, and cells were incubated for 10 min. Stained cells were then rinsed with distilled water 2-3 times and observed and photographed under a microscope.

Statistical Analysis
All data reported are expressed as mean ± SE. Student's t-test was carried out using SPSS software (SPSS 17.0, Chicago, IL, USA) for statistical analysis of the data. A p value of < 0.05 was considered to be statistically significant.

SNP Detection and Genotyping
The sequencing results revealed two SNPs in the PDGF-D gene ( Figure 1

Genetic Parameters Calculation
The genetic parameters calculation results are shown in Table 1. All sites were in Hardy-Weinberg status (p > 0.05). g.4122606 C > G was in moderate PIC status (0.25 < PIC < 0.5). The He and Ne of g.4122606 C > G were 0.47-0.5 and 1.88-2.0, respectively. The g.4122606 C > G site was

Genetic Parameters Calculation
The genetic parameters calculation results are shown in Table 1. All sites were in Hardy-Weinberg status (p > 0.05). g.4122606 C > G was in moderate PIC status (0.25 < PIC < 0.5). The He and Ne of g.4122606 C > G were 0.47-0.5 and 1.88-2.0, respectively. The g.4122606 C > G site was highly variable. In addition, the alleles of g.4122606 C > G were evenly distributed. However, g.3852134 C > T showed low PIC status (PIC < 0.25), and the He and Ne were lower than those of g.4122606 C > G. These data showed that g.3852134 C > T was in a less variable status, and the alleles were unevenly distributed.

Correlation Analysis between PDGF-D Gene Polymorphism and Tail Type in Sheep
The association analysis results revealed that g.4122606 C > G was significantly related with tail length (p < 0.05, Table 2). The tail length of sheep with the GG genotype was significantly shorter than CC and CG genotype carriers. No differences were observed with tail width and circumference. However, g.3852134 C > T was significantly related with the tail width (p < 0.05, Table 2). Regarding g.4122606 C > G, the CC genotype carriers had the smallest tail length, width, and circumference. Particularly, the tail width of sheep with CC genotype was significantly narrower than TT genotypes (p < 0.05). However, the tail length and circumference showed no differences between genotypes. Note: The size of the sheep's tail in adulthood (tail length, tail width, and tail circumference) is basically determined, and the difference within the same variety is small. Therefore, we finally merged the three breeds of sheep into one large group for correlation analysis. In the same column, values with different lower-case letters are significantly different (p < 0.05).

Transcriptional Factor Binding Sites Prediction
The mutation g.3852134 C > T occurred in the promoter region of PDGF-D. The promoter sequence obtained from NCBI was used for predicting TF binding sites. Multiple TF binding sites were present in the promoter region, such as those for insulinoma-associated protein 1 (INSM1), homebox (HOXA5), and double homeobox 4 (DUX4) ( Table 3). In addition, several TF binding sites were absent after mutation, such as those for hepatocyte nuclear factor 4γ (HNF4G) and human intestinal trefoil factor (HItf) ( Table 3). Some new TF binding sites also appeared as a result of mutation, such as those for CCAAT/enhancer binding protein A (CEBPA) and Jun (Table 3).

Lentiviral Overexpression Efficiency Assay
We next used a lentivirus overexpressing PDGF-D. At 72 h after transfection, approximately 70% of cells expressed the GFP reporter gene ( Figure S1A,B), indicating that PDGF-D had been successfully integrated into chromosomes. qRT-PCR results showed that the PDGF-D gene was expressed at significantly higher levels in infected cells than levels in the negative controls at every time point ( Figure S1C).

Influence of Overexpressed PDGF-D on the Expression of Adipose Differentiation-Related Marker Genes
We next detected the expression level of PPARγ (peroxisome proliferator-activated receptor gamma) and LPL (lipoproteinlipase) genes, which are marker genes in adipogenesis (Figure 2). Overexpression of PDGF-D resulted in up-regulated expression level of these marker genes. The relative mRNA expression level of PPARγ was significantly higher than that of the negative control at 1, 5, and 7 d (p < 0.01), as well as at 3 d (p < 0.05). The relative mRNA expression level of LPL was significantly higher than that of the negative control at 0, 5, and 7 d (p < 0.01), as well as 1 d (p < 0.05). No significant difference was observed in the expression level of LPL at 3 d.

Lentiviral Overexpression Efficiency Assay
We next used a lentivirus overexpressing PDGF-D. At 72 h after transfection, approximately 70% of cells expressed the GFP reporter gene ( Figure S1A,B), indicating that PDGF-D had been successfully integrated into chromosomes. qRT-PCR results showed that the PDGF-D gene was expressed at significantly higher levels in infected cells than levels in the negative controls at every time point ( Figure S1C).

Influence of Overexpressed PDGF-D on the Expression of Adipose Differentiation-Related Marker Genes
We next detected the expression level of PPARγ (peroxisome proliferator-activated receptor gamma) and LPL (lipoproteinlipase) genes, which are marker genes in adipogenesis (Figure 2). Overexpression of PDGF-D resulted in up-regulated expression level of these marker genes. The relative mRNA expression level of PPARγ was significantly higher than that of the negative control at 1, 5, and 7 d (p < 0.01), as well as at 3 d (p < 0.05). The relative mRNA expression level of LPL was significantly higher than that of the negative control at 0, 5, and 7 d (p < 0.01), as well as 1 d (p < 0.05). No significant difference was observed in the expression level of LPL at 3 d.

Oil Red O Staining
We performed oil red O staining on sheep preadipocytes on day 7 after induction of differentiation. As shown in Figure 3, many little lipid drops were stained in red, and the lipid ring was observed. The number of lipid drops in the PDGF-D overexpression group was higher than that in the negative control. These results suggested that PDGF-D could promote the formation of lipid drops in adipocytes in vitro.

Oil Red O Staining
We performed oil red O staining on sheep preadipocytes on day 7 after induction of differentiation. As shown in Figure 3, many little lipid drops were stained in red, and the lipid ring was observed. The number of lipid drops in the PDGF-D overexpression group was higher than that in the negative control. These results suggested that PDGF-D could promote the formation of lipid drops in adipocytes in vitro.

Lentiviral Overexpression Efficiency Assay
We next used a lentivirus overexpressing PDGF-D. At 72 h after transfection, approximately 70% of cells expressed the GFP reporter gene ( Figure S1A,B), indicating that PDGF-D had been successfully integrated into chromosomes. qRT-PCR results showed that the PDGF-D gene was expressed at significantly higher levels in infected cells than levels in the negative controls at every time point ( Figure S1C).

Influence of Overexpressed PDGF-D on the Expression of Adipose Differentiation-Related Marker Genes
We next detected the expression level of PPARγ (peroxisome proliferator-activated receptor gamma) and LPL (lipoproteinlipase) genes, which are marker genes in adipogenesis ( Figure 2). Overexpression of PDGF-D resulted in up-regulated expression level of these marker genes. The relative mRNA expression level of PPARγ was significantly higher than that of the negative control at 1, 5, and 7 d (p < 0.01), as well as at 3 d (p < 0.05). The relative mRNA expression level of LPL was significantly higher than that of the negative control at 0, 5, and 7 d (p < 0.01), as well as 1 d (p < 0.05). No significant difference was observed in the expression level of LPL at 3 d.

Oil Red O Staining
We performed oil red O staining on sheep preadipocytes on day 7 after induction of differentiation. As shown in Figure 3, many little lipid drops were stained in red, and the lipid ring was observed. The number of lipid drops in the PDGF-D overexpression group was higher than that in the negative control. These results suggested that PDGF-D could promote the formation of lipid drops in adipocytes in vitro.

Discussion
Many studies have suggested that PDGF-D may be associated with tail type [11][12][13][14][15][16]. However, few studies have conducted functional verification of PDGF-D in controlling the formation of sheep tails. In our research, we identified two SNPs in the PDGF-D gene that were related to tail traits. The g.4122606 C > G mutation was associated with tail length, and this mutation occurs in an intron region. While introns are not expressed as protein, they play a crucial role in transcriptional regulation [18], such as by encoding microRNA to regulate target genes [19], performing as promoters or enhancers, and participating in alternative splicing [20][21][22][23][24]. The g.4122606 C > G locus showed moderate polymorphism, relatively high heterozygosity, and a large degree of genetic variation, which could bring more selection effects. The g.3852134 C > T locus has lower polymorphism, lower heterozygosity, and less genetic variation, which may be related to the highly selective breeding of the experimental population during breeding. Compared with the g.4122606 C > G locus, the number of effective alleles at the g.3852134 C > T locus is smaller, indicating the uneven distribution of the g.3852134 C > T locus in the test population may be related to the selection and matching system of the test population.
TFs play significant roles in regulating gene expression. The mutation g.3852134 C > T located in the promoter region was associated with tail width. Previous studies showed that mutations in promoter regions may change TF binding sites [25][26][27]. Our results showed that some new TF binding sites appeared in the promoter harboring mutations, such as sites for C/EBPα and Jun. C/EBPα functions to trigger differentiation of preadipocytes into mature adipocytes [28,29]. Some studies showed the slight increase in C/EBPα expression after the decrease of C/EBPβ and C/EBPδ before the expression of adipocyte-specific genes [30,31]. C/EBP families participate in the early stage in adipogenesis. As a pleiotropic transcriptional activator, C/EBPα transactivates promoters from numerous adipocyte genes [4]. In addition, some reports showed that forced expression of C/EBPα in 3T3-L1 preadipocytes stimulated adipogenesis without any hormonal induction [2,3,5]. In contrast, blocking expression of C/EBPα inhibited adipogenesis [32]. Furthermore, some studies showed that C/EBPα is a key regulator related with insulin sensitivity. These observations demonstrated that C/EBPα is required for preadipocyte differentiation. Jun and Fos form a dipolymer named AP-1 (activator protein-1), which is an important positive regulator in adipogenesis.
Previous studies indicated that the PDGF family can promote preadipocyte proliferation and inhibit preadipocyte differentiation [33,34]. Another study showed that PDGF-BB can promote the adipogenic differentiation of fibroblasts [35]. These results indicate that the PDGF family participates in fat metabolism. In addition, the expression levels of adipose differentiation marker genes (PPARγ and LPL) significantly increased after the overexpression of PDGF-D in sheep preadipocytes, indicating PDGF-D plays a certain role in the tail fat deposition process of the sheep. Oil red O revealed increased accumulation of big and small lipid droplets after the overexpression of PDGF-D compared with the control group, which further proved the important role of PDGF-D in the tail fat deposition process of t sheep.
PDGF-D promotes mitosis, proliferation and division of vascular smooth muscle cells, and formation of new blood vessels [36]. The sheep tail is rich in blood capillaries. Angiogenesis and organofaction are two closely related processes in that the former is the premise of the latter. Oxygen and nutriments can be delivered to tissue after angiopoiesis. Whether PDGF-D participates in angiopoiesis and then adipogenesis or if it directly controls fat development should be examined in further studies. In addition, studies have shown that PDGF-D is related to sheep body size in various environments, and PDGF-D may also directly regulate sheep tail size [37].

Conclusions
The PDGF-D gene shows polymorphisms in sheep. g.4122606 C > G was significantly correlated with tail length and g.3852134 C > T was significantly correlated with tail width. In addition, Animals 2020, 10, 89 9 of 11 overexpressed PDGF-D in sheep preadipocytes can promote adipogenic differentiation. This result can be applied to molecular marker-assisted selection of sheep tail type.

Conflicts of Interest:
The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript: