Genotype and allele frequencies of polymorphisms in ABCG2, PPARGC1A and OLR1 genes in indigenous cattle breeds in Turkey

Abstract This study was carried out to determine polymorphisms of four genes in South Anatolian Red (SAR) and East Anatolian Red (EAR) indigenous cattle breeds in Turkey. Single nucleotide polymorphisms (SNPs) monitored in this study are Y581S in ATP binding cassette sub family G member 2 (ABCG2) gene, c.1892T>C and c.3359A>C in peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PPARGC1A) gene and g.8232C>A in oxidized low-density lipoprotein receptor 1 (OLR1) gene. The frequency of the ancestral allele A of the ABCG2 gene Y581S polymorphism was found to be very high (SAR: 0.63; EAR: 0.64) in both cattle breeds. The CC genotypes of PPARGC1A gene c.1892T>C (SAR: 0.65; EAR: 0.80) and OLR1 gene g.8232C>A polymorphisms (SAR: 0.82; EAR: 0.86), which are associated with high milk fat percentage, had higher frequencies than those of the other genotypes. In conclusion, we might suggest that the allele distribution of the ABCG2 gene Y581S polymorphism can be the evidence indicating autosomal gene flow from zebu cattle to SAR and EAR cattle breeds.


INTRODUCTION
ATP binding cassette sub family G member 2 (ABCG2) gene, which product is expressed in the mammary gland in cows, encodes a transporter protein that facilitates transport of medicines through the cell membrane by binding ATP. The level of expression signifi cantly increases during lactation compared to the dry period [1]. It has been suggested that this transporter protein plays a role in the secretion of xenobiotics and some micro-nutrients such as cholesterol and vitamin K3 into milk [2]. ABCG2 gene is located on chromosome 6 in cattle and is known to have important effects on the milk yield traits. A single nucleotide polymorphism (SNP) resulted from translocation of adenine/cytosine on the 14 th exon presents the missense mutation named Y581S as it leads to the replacement of the 581 th amino acid tyrosine with cysteine [3]. Another potential Quantitative Trait Locus (QTL) on the 6 th chromosome of cattle is Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1 Alpha (PPARGC1A) gene. This gene has been expressed in many organs that are metabolically active. It has been suggested that the product of this gene is associated with cellular energy metabolism, thermogenesis, adipogenesis, and gluconeogenesis [4,5]. Up to date, many SNPs have been determined for this gene [6][7][8]. However, only two of these have been suggested to have an effect on milk yield traits. The fi rst one is PPARGC1Ac.1892T>C that leads to replacement of threonine/cysteine on the 19 th position [6]. The second one is PPARGC1A-c.3359A>C located on 3' UTR region and it causes alanine/cisteine replacement at the 968 position [6]. Oxidized Low-density Lipoprotein Receptor 1 (OLR1) gene encodes surface receptors of vascular endothelial cells and contributes to the balance of low-density lipoproteins [9]. Oxidized low density lipoproteins (ox LDL) cause atherosclerosis and affect glucose and lipid metabolism in the mammary gland [7,10]. Naturally, the protein encoded by OLR1 gene affects these metabolisms [8]. OLR1 gene is located on the 5 th chromosome in cattle. It has been reported that one SNP located at 3'UTR region is associated with milk yield traits. This SNP resultes from the translocation of cytosine/timine and results in the replacement of cysteine/alanine at the 223th position in the fi nal protein encoded by this gene [7,8,10]. In the present study, genotype and allele frequencies of polymorphisms Y581S of ABCG2 gene, c.1892T>C and c.3359A>C of PPARGC1A gene, and g.8232C>A of OLR1 gene, which are proposed to have an infl uence on milk yield parameters, particularly on milk fat percentage, were estimated in South Anatolian Red (SAR) and East Anatolian Red (EAR) cattle breed in Turkey.

Animals and DNA isolation
SAR breed cattle were selected from the herds in South Anatolia whereas EAR cattle were selected from those located in Eastern Anatolia. In the selection of the animals, care was taken not to include animals that were parentally related so they were representative of their own breed characteristics. Blood samples were collected in sterile 2 ml tubes containing EDTA. Genomic DNAs were isolated using a standard ammonium acetate salt-out method [11]. The present study was approved by Istanbul University Local Committee on Animal Research Ethics (25.03.2010; Protocol Nr.:2010/51).

PCR-RFLP Analysis
The primers and annealing temperatures used to amplify target regions ABCG2 SNP Y581S, PPARGC1A SNPs c.1892T>C and c.3359A>C and OLR1 SNP g.8232C>A are given in Table 1. The PCR conditions used for all the regions were: initial denaturation at 94 0 C for 5 min, 1 min at 94 0 C, 1 min at the annealing temperature (Table 1), 1 min at 72 0 C for 35 cycles and a fi nal extension at 72 0 C for 10 min. Amplifi cation was performed in a volume of 25 μl containing 5 μl 10X PCR buffer, 100 μM dNTP, 10 pmol of each primer, 1.0 μM MgCl 2 , 2U Taq polimerase (Fermentas Life Sciences, Canada) and 50-100 ng genomic DNA. PCR amplifi ed SNPs Y581S, c.1892T>C, c.3359A>C and g.8232C>A were digested with enzymes PstI, BsuRI, NheI and PstI (Fermentas Life Sciences, Canada), respectively (10 U of each enzyme) and incubated at 37 0 C overnight. Electrophoretic separation of the digestion products was carried out with 1xTBE in 2% agarose gel for 30 min. under 120 V to differentiate the alleles A (292 bp, uncut) and C (268 and 24 bp); T (193 and 12 bp) and C (173, 20 and 12 bp); A (191 bp, uncut) and C (157 and 34 bp); A (143bp, uncut) and C (118 and 30 bp), respectively. The gels were subsequently stained with ethidium bromide and photographed on an UV transluminator.

Statistical analysis
Genotype and allele frequencies of ABCG2 SNP Y581S, PPARGC1A SNPs c.1892T>C and c.3359A>C and OLR1 SNP g.8232C>A in SAR and EAR cattle were calculated by using PopGene 32 software [12]. A chi-square test was also performed to check Hardy-Weinberg equilibrium at each locus by the same program (Table 2).

RESULTS
The genotype and allele frequencies of ABCG2 gene Y581S polymorphism, PPARGC1A gene c.1892T>C and c.3359A>C polymorphisms, and OLR1 gene g.8232C>A polymorphism are given in Table 2

DISCUSSION
Considering the genes affecting milk yield and composition traits in cows, it has been suggested that one gene related to these traits exists in all autosomal chromosomes. The most important genes affecting the amount and percentage of milk fat are found on Bos taurus autosomal chromosomes (BTA) 5, 6, 9, 14, 20 and 26 [13]. In the present study, allele and genotype frequencies in the polymorphisms of ABCG2 and PPARGC1A genes that are located on chromosome 6 and OLR1 gene located on chromosome 5 were determined for SAR and EAR cattle reared in Anatolia. Ron [15]. In our study, ABCG2 A allele frequency in SAR and EAR cattle was found to be 0.63 and 0.64, respectively. These fi ndings support the results by Ron et al. (2006) [14]. Previously, genes frequencies close to Bos indicus breeds were reported for SAR and EAR [16]. In the light of previous studies carried out using data on mitocondrial DNA and Y and autosomal chromosomes [17], this can be explained by the occurrence of a zebu gene fl ow to the cattle in the near east and remaining of this limited with autosomal genes in Anatolian native breeds.

PPARGC1A gene
The relationship between c.1892T>C and c.3359A>C SNPs of PPARGC1A gene and milk yield traits was evaluated in many previous studies. However, such relation could not have been proven clearly. Weikard et al. (2005) suggested a relationship between low milk fat level in German-Holstein cattle and the frequency of C allele at position c.1892T>C [6]. On the other hand, this could not be confi rmed by Khatip et al. (2007) in a later study carried out on American Holstein cattle [7]. Alim et al. (2012) reported that milk protein level was higher in Chinese Holstein homozygous cows with TT genotype at the c.1892T>C locus [18]. On the other hand, Komisarrek et al. (2012) reported no relationship between T allele and milk yield traits [19]. Similarly, Kowalewska -Luczak et al. (2010) did not fi nd any relation between c.1892T>C and c.3359A>C polymorphisms of PPARGC1A gene and milk yield traits in Jersey cattle [20]. In the previous studies, the highest genotype frequency for CC genotype was found 0.68 in German Holstein [6], 0.53 in Polish Holstein-Friesian cattle [8], 0.56 in German Holstein-Friesian [21], 049 in Chinese Holstein cattle [18] while CT genotype was found at the highest frequency of 0.65 in the University of Wisconsin dairy herd [7], and 0.72 in Jersey breed [20]. Kowalewska-Luczak et al. (2010) found the frequency of PPARGC1A gene c.3359A>C polymorphism genotype AA to bethe highest in the Jersey breed cattle while they did not fi nd genotype CC [20]. In our study, CC genotype at the c.3359T>C locus, was found at very high levels as 0.65 and 0.80 in SAR and EAR, respectively while no TC heterozygote genotype was found in neither breed. Although our fi ndings are in agreement with those genotype frequencies reported by Weikard

OLR1 gene
Previous studies indicated that OLR1 genotypes have important effects on the amount and percentage of milk fat [7,10,21]. Khatip et al. (2006) reported that CC genotype and C allele signifi cantly increased the amount and percentage of milk fat in the Holstein breed cattle [10]. They suggested that 3'-UTR polymorphism of OLR1 gene plays an important role in the expression and translocation of the gene and OLR1 expression and synthesis of OLR1 was particularly higher in genotype CC. These fi ndings were supported by similar results reported by Khatip [21]. Finding higher frequencies of C allele in Bison bison, Swiss Brown, Jersey, and Italian-Swiss Brown cattle compared to that found in Holstein breed cattle was interpreted as higher milk fat percentage in those breeds than those seen in Holstein [10,21]. C allele frequencies of SAR and EAR breeds were found close to those of Bison bison, Swiss Brown, Jersey, and Italian-Swiss Brown cattle. Thus, we can suggest that this similarity is because of higher milk fat percentage of SAR and EAR.
In conclusion, in respect to Y581S polymorphism of ABCG2 gene, SAR and EAR cattle were found to be closer to Bos indicus breeds. This fi nding is in agreement with the results of the previous study that indicated zebu gene fl ow to these two breeds along with autosomal gene fi ndings. In addition, in terms of genotypes and alleles of SNPs c.1892T>C of PPARGC1A gene and g.8232C>A of OLR1 gene, these two breeds were found close to other cattle breeds with higher milk fat percentage. These fi ndings are parallel to the fact explaining the higher milk fat percentage in these two native breeds.

ACKNOWLEDGEMENT
This study was supported by the Research Fund of Istanbul University, Project no: 6627.