Association of PPARγ2 polymorphisms with carcass and meat quality traits in a Pietrain x Jinhua F2 population

The PPARγ2 gene is a key regulator of both proliferation and preadipocyte differentiation in mammals. Herein its genotype and allele frequencies were analyzed using PCR-SSCP in eight pig breeds (N = 416). Two kinds of polymorphisms of the PPARγ2 gene were detected, including a previously reported shift SNP A177G (Met59Val) in exon 1 and a novel silent mutation G876A in exon 5. The results revealed that European pig breeds carry a higher allele A frequency at the A177G locus and a fixed GG genotype at the G876A locus. Allele A at the G876A locus was only found in Jinhua pigs. The association between haplotype (A177G/G876A) and carcass and meat quality traits was analyzed in a Pietrain x Jinhua F2 population (N = 248). The PPARγ2 gene was found to be significantly associated with backfat thickness at the shoulder (p < 0.05), 6–7th ribs (p < 0.01), last rib (p < 0.01), gluteus medius (p <0.05) and ham weight (p < 0.01). Significant effects of different haplotypes on ham weight and backfat thickness at the 6–7th ribs, last rib, and gluteus medius were also observed.


Introduction
Genetic effects on porcine meat quality have historically been widely recognized and marker-assisted selection has been used by animal breeders to improve the quality and performance of livestock (Van der Steen et al., 2005). Reducing body fat, increasing lean percentage, and improving meat quality are critical issues for the pig breeding industry. Thus, there is great interest in having comprehensive molecular marker information for improvement of domestic swine.
The peroxisome proliferator-activated receptor-g (PPARg) is a member of the nuclear hormone receptor superfamily. It exists as two distinct isoforms (g1 and g2) produced from a single gene sequence by alternative splicing, resulting in amino acid differences in the N-termini (Fajas et al., 1997;Omi et al., 2005). PPARg1 is expressed in a variety of tissues at a relatively low level, while PPARg2 is exclusively expressed in adipose tissue and at ten times higher transcriptional activity (Tontonoz et al., 1994a;Werman et al., 1997). PPARg2 is known as a key regulator of adipogenesis involved in the regulation of fatty acid uptake and storage . It participates in the transcriptional activation of several adipogenic transcription factors and lipogenic genes (Wu et al., 1999;Pasceri et al., 2000;Rosen et al., 2002;Arimura et al., 2004), and directs the differentiation of preadipocytes to adipocytes. Interestingly, overexpression of PPARg2 in fibroblast cell lines using retroviral vectors efficiently converts them to adipocytes (Tontonoz et al., 1994b). Previous research also supported this view, indicating a dual role for PPARg2 in the regulation of both proliferation and preadipocyte differentiation (Fajas et al., 2002;Samulin et al., 2008). Therefore, PPARg2 appeared to be a promising candidate gene for improving carcass and meat quality traits. However, very little work has yet been conducted to investigate the role of the porcine PPARg2 gene in carcass and meat quality. Emnett et al. (2000) found a signicant association between the PPARg2 Met59Val polymorphism, meat quality and backfat thickness in Berkshire, Duroc, Hampshire and Landrace swine breeds, while Grindflek et al. (2004) found a significant association between the Met59Val/A220G/ G324A haplotype and ham weight in a Norwegian pig population, but they did not denote a major effect on backfat thickness.
The current study was designed to detect novel single nucleotide polymorphisms (SNPs) of the porcine PPARg2 Genetics andMolecular Biology, 34, 1, 56-61 (2011) Copyright © 2011, Sociedade Brasileira de Genética. Printed in Brazil www.sbg.org.br gene using polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) and DNA sequencing, and to further investigate the relationship between these SNPs and carcass and meat quality in a Pietrain x Jinhua F2 crossbred population.

Material and Methods
Animal and data collection The association between the SNPs and recorded traits were analyzed in a Pietrain x Jinhua F2 crossbred swine population. The Pietrain x Jinhua resource population was constructed from purebred Pietrain sires and Jinhua dams, including 6 F1 boars, 20 F1 sows and 248 F2 progeny. The 248 PJF2 offspring were raised under standard conditions at The Experimental Pig Farm of Zhejiang University and slaughtered at a commercial abattoir. In addition, DNA samples of eight pig breeds were collected, consisting of 96 Jinhua, 36 Jiaxing Black, 54 Bihu, 66 Shengxian Hua, 37 Landrace, 67 Yorkshire, 30 Duroc and 30 Pietrain pigs (Table 1). The sampled pigs were from sources that were unrelated throughout the last three generations.
PJF2 animals were slaughtered (following electric shock) at 219 days of age (SD = 31.4 days, live weight 81.3 ± 10.58 kg). After slaughter, backfat thickness (BFT) was measured on the left side at four locations (shoulder, 6-7 th ribs, last rib, and gluteus medius). At 45 min postmortem, muscle pH and temperature measurements were taken on the longissimus dorsi (LD) muscle using a Thermo Scientific Orion pH meter (Model 230A Plus, USA) and a digital thermometer (Ama-digit 14 TH 35°C -500°C Amarell electronic, Kreuzwertheim, Germany), respectively. Loin eye area (LEA) of the LD muscle was traced on an acetate film between the 13/14 th ribs and quantitated by planimetry. Samples of the longissimus dorsi muscle were taken at the 13 th rib for the determination of water-holding capacity (WHC), intramuscular fat (IMF), protein (IMP) and water (IMW) content. WHC was determined with the filter paper press method (Wierbicki and Deatherage, 1958), while intramuscular fat, protein and water content were determined using a Meat Analyzer (Antaris II FT-NIR analyzer, Thermo Electric company, USA). Additionally, the ham was cut from the last lumbar vertebra and the feet were removed.
PCR amplifications were carried out in 25 mL reaction volumes containing at the following final concentrations: 50 ng of template DNA, 400 mM of dNTPs (Sangon, China), 0.25 mM of each primer and 1 unit (U) of Taq polymerase (TaKaRa, Japan). The PCR protocol consisted of an initial denaturation at 94°C for 5 min, followed by 36 cycles of denaturation at 94°C for 30 s, annealing at 55°C for PPARg2-P1 or 51°C for PPARg2-P2 for 30 s, and extension at 72°C for 30 s, with a final extension at 72°C for 10 min. The porcine PPARg2 alleles were detected using PCR-SSCP. A 5 mL aliquot of each amplicon was diluted in denaturing solution (98% formamide, 10 mM EDTA, 0.025% bromophenol blue, 0.025% xylene-cyanol) denatured at 95°C for 5 min, rapidly cooled on ice and resolved in acrylamide:bisacrylamide (29:1) (Bio-Rad) gels at 400 V for 4 h at 4°C, in 1 x TBE buffer (89 mM Tris base, 89 mM boric acid, 2 mM EDTA, pH 8.0). The gels were silver-stained according to the method of Bassam et al. (1991).

Sequence analysis
Nucleotide sequence translation and comparisons were carried out using the DNAMAN software. The BLAST algorithm of NCBI was used to search homologous sequences in the NCBI GenBank database.

Statistical analysis
Linkage disequilibria between all pairs of biallelic loci were calculated using measures of Lewontin's D' and r 2 (Pritchard and Przeworski, 2001) in the PJF2 population. Haplotypes were determined by the Linkage algorithm of the HaploView software (V 2.01). The genotype and allele distributions in purebreds were tested for Hardy-Weinberg equilibrium using the c 2 test.
For studying the association between the PPARg2 gene haplotypes and recorded traits in the PJF2 crossbred animals, the following model was used: where Y ijklm is the observation of the trait; m is the population mean; S i is the fixed effect of sex; F j is the fixed effect of the father; G k is the fixed effect of the k th haplotype; l l is the random effect of litter; b is the regression coefficient of the slaughter weight; X ijklm is the slaughter weight as covariate, and e ijklm is the random residual error. The least square means method of the GLM (General Linear Model) procedure (SPSS 16.0) was used for association analysis between the PPARg2 haplotypes and recorded traits.

Detection of single nucleotide polymorphisms
Two fragments, PPARg2-P1 (204 bp) and PPARg2-P2 (350 bp) were amplified from genomic DNA. These were then assayed by the SSCP technique and products with unique patterns were sequenced to detect any single base substitution. One substitution located in exon 1 at position 177 (c.+177 A > G, GenBank accession No. NM_214379) causes an amino acid change from methionine to valine (Met59Val). This SNP has already been described in previous studies (Emnett et al., 2000;Grindflek et al., 2004). The primer PPARg2-P2 detected a novel polymorphism in exon 5 at position 876 (c.+876 G > A, Gen-Bank accession No. NM_214379) of the porcine PPARg2 gene. This SNP represents a silent mutation.

Genotype frequencies in different pig populations
The combined effects of the exon 1 and exon 5 substitutions were estimated as haplotype substitution effects. Genotype and allele frequencies of the SNPs A177G and G876A within pure breeds are shown in Table 1.
At A177G, allele A had a high frequency (> 75%) in European pig breeds, especially fixed in Pietrain swine, and a relatively lower frequency in Chinese native breeds (< 67%). In Jinhua, Jiaxing Black, Landrace and Yorkshire breeds, the alleles were at Hardy-Weinberg equilibrium. Interestingly, at G876A, allele A was found only in Jinhua pigs, whereas allele G was fixed in all other breeds. Therefore, allele A appears to be Jinhua-specific, and, consequently, the breeds were not at Hardy-Weinberg equilibrium with respect to this marker.

Haplotype analysis
Four haplotypes, H1 (A-G), H2 (A-A), H3 (G-G), and H4 (G-A), were found in PJF2 animals and had their frequencies and linkage disequilibria estimated (Table 2). Of the four types, the A-G haplotype (H1) was found to represent the majority. Linkage analysis indicated that there were certain linkage disequilibria between the two SNPs A177G and G876A (D' = 0.504).

Relationship between PPARg2 gene haplotypes and recorded traits
The results of the association analysis of PPARg2 gene haplotypes and recorded carcass and meat traits within the PJF2 population are shown in Table 3. PPARg2 gene haplotypes were found to be significantly associated with BFT at the shoulder and gluteus medius (p = 0.029 and 0.042, respectively), with significant effects observed on ham weight and BFT at the 6-7 th ribs and last rib (p = 0.000, 0.003, and 0.003, respectively). Pigs with the H1H3 genotype had lower ham weight compared to the other genotypes (p < 0.01). H2H3 genotype pigs had higher BFT at the 6-7 th ribs, last rib and gluteus medius (p < 0.01), while for BFT at the shoulder, the H2H3 genotype had relatively higher thickness, just slightly lower than that of the H1H3 genotype.

58
PPARg2 SNPs and pig meat traits

Discussion
The Jinhua and Pietrain breeds are significantly different in carcass and meat quality phenotypes. As a Chinese domesticated breed, the Jinhua pig has a reputation for its thin skin, fine bones and tender meat, but has higher BFT and IMF content than other breeds. In contrast, the Pietrain type, which has a relatively lower BFT and IMF content, is famous for its high quality and high proportion of lean meat vs. fat content, and has been used worldwide as a sire breed. Observed differences in carcass and meat quality phenotypes are probably related to protein accretion and lipid synthesis (Renaudeau and Mourot, 2007). In this regard, it has been suggested that certain genes may play important physiological roles in fat-related pathways (Agarwal and Garg, 2006). Therefore we hypothesized that fat-related genes that are at considerable linkage disequilibrium may produce distinct phenotypes in a Pietrain x Jinhua F2 (PJF2) population. In this study, we found that genotype and allele frequencies for two analyzed polymorphisms differed between the Chinese native pig and the European pig, indicating that the two types of breeds had been subject to different genetic selection regimes. Lai et al. (2008) demonstrated that the region of the PPARg2 promoter comprising nucleotide base pair positions -457 to +129 is essential for the CEBPD (CCAAT/Enhancer binding protein D) response which activates PPARg2 transcription. They furthermore suggested that the A177G SNP might affect binding efficiency and thus the gene expression level at this site. Although the G876A SNP is a silent mutation, there is substantial evidence indicating that silent mutations may affect mRNA secondary structure and stability (Shen et al., 1999;Duan et al., 2003), mRNA decay rates, and ultimately lead to disease (Capon et al., 2004).
The porcine PPARg2 gene was mapped to chromosome 13 (Grindflek et al., 2000) and found to be close to several quantitative trait loci (QTL) and candidate genes for birth weight, daily weight gain and BFT (Andersson et al., 1994;Yu et al., 1995;Evans et al., 2003). Since it plays a key role in proliferation and preadipocyte differentiation, PPARg2 is an excellent target for studying growth and fat deposition traits. Experiments with homologous PPARg2deficient mice showed that genetic ablation of PPARg2 decreased adipose tissue expandability and the capacity for buffering toxic lipids in peripheral organs (Medina-Gomez et al., 2007). In humans, several SNPs within the PPARg2 locus (Pro12Ala, C1431T, C2821T, A2819G) have been identified in different populations, and their alleles demonstrated to be associated with obesity, type 2 diabetes, increased insulin sensitivity and decreased receptor activity (Deeb et al., 1998;González Sánchez et al., 2002;Meirhaeghe and Amouyel, 2004;Costa et al., 2009).
A total of five mutations have previously been identified in the porcine PPARg2 gene promoter and coding regions, these being A177G (M59V), G78A, A42G, A220G and G324A (Emnett et al., 2000;Grindflek et al., 2004). In our study we identified a novel Jinhua-specific SNP (A876G), as well as the previously reported A177G (M59V) polymorphism, and investigated haplotype effects in a PJF2 population.
Haplotypes can be used to correlate a specific phenotype with a specific gene in a small sample population (Drysdale et al., 2000) and can provide more information on the complex relationship between DNA variation and phenotype than single SNPs may afford (Stephens et al.,Chen et al. 59  2001). Thus, we used haplotype information to evaluate the relationship between PPARg2 polymorphisms and candidate traits in swine. Our results showed significant effects on ham weight (p < 0.01), consistent with previous results obtained by Grindflek et al. (2004). We also found in the PJF2 population significant differences between the different haplotypes with respect to BFT at the 6-7 th ribs (p < 0.01), the last rib (p < 0.01) and the gluteus medius (p < 0.01). No significant association was found between haplotypes and IMF. Grant et al. (2008) first reported that exposure to the PPARg agonist troglitazone (TRO) resulted in a 6.4-fold increase in the percentage of differentiated subcutaneous preadipocytes compared to intramuscular preadipocytes, and they conjectured that inherent differences in the capacity for adipose differentiation may exist between bovine subcutaneous and intramuscular preadipocytes. Thus, PPARg2 gene regulation is thought to favor subcutaneous fat deposition over intramuscular fat content.
The association observed in our study may arise from linkage between the PPARg2 gene and specific polymorphisms that are in linkage disequilibrium and may be located in another region of the PPARg2 gene, or by interaction with other quantitative trait loci. Furthermore, the linkage disequilibrium between A177G and G876A in the PJF2 population was unstable (r 2 = 0.249). Consequently, additional studies are needed to better assess the real impact of the effects of the PPARg2 gene on carcass and meat quality traits.