Dde I Polymorphism in Coding Region of Goat POU 1 F 1 Gene and Its Association with Production Traits

POU1F1 is a positive regulator for GH, PRL and TSHβ and its mutations associate with production traits in ruminant animals. We described a DdeI PCR-RFLP method for detecting a silent allele in the goat POU1F1 gene: TCT (241Ser)>TCG (241Ser). Frequencies of D1 allele varied from 0.600 to 1.000 in Chinese 801 goats. Significant associations of DdeI polymorphism with production traits were found in milk yield (*p<0.05), litter size (*p<0.05) and one-year-old weight (*p<0.05) between different genotypes. Individuals with genotype D1D1 had a superior performances when compared to those with genotype D1D2 (*p<0.05). Hence, the POU1F1 gene was suggested to the potential candidate gene for superior milk performance, reproduction trait and weight trait. Genotype D1D1, characterized by a DdeI PCR-RFLP detection, was recommended to geneticists and breeders as a molecular marker for better performance in the goat industry. (


INTRODUCTION
The goat industry is an important part among the socalled big domestic animals sector in China.The estimated size is more than 157,361,000 mainly reared in the northern China, which belonged to more than twenty native breeds (e.g.dairy, meat and wool breeds).Dairy-goat farming is significant to the economics of the western China with the characteristics of under-development and poverty.Goat's meat from young or adult animals has been consumed throughout the country for recent ten years.Moreover, the wool of goat is used in many ways, e.g. for the wrapping of the dead, the making of clothes (Boyazoglu et al., 2005).Large exports of wool and its related products give great chance for rural and western families to improve the economic situation (Dubeuf et al., 2004).Therefore, the further improvement and increase of the quantity and quality in goat dairy, meat and wool will better contribute to the Chinese society, particularly in economy, nutrition, tradition and religion.This issue can be resolved by culturing more and better goat breeds.However, it is difficult to culture excellent goat breeds by the traditional genetic and breeding method.So, many breeders mainly focus on DNA markers for animal selection and breeding through marker-assisted selection (MAS).
As a member of the POU-domain family gene, POU1F1 is a positive regulator for growth hormone (GH) (Zhou et al., 2005), prolactin (PRL) (Li et al., 2006) and thyroidstimulating hormone β (TSHβ) by binding to target DNA promoters as a dimer in mammalian animals (Jacobson et al., 1997).POU1F1 mutations associated with Snell dwarf (dw) and Jackson dwarf (dw-J) in mice and dwarfish in human (Li et al., 1990;Pfaffle et al., 1992;Revnau et al., 2004).Moreover, polymorphisms of POU1F1 gene associated with important production traits in cattle (Renaville et al., 1997a;Renaville et al., 1997b;Zhao et al., 2004) and in pig (Yu et al., 1995;Stancekov et al., 1999;Sun et al., 2002).Recently polymorphisms of sheep POU1F1 gene were firstly reported (Bastos et al., 2006).Few polymorphisms of goat POU1F1 gene and their associations with production traits had been described.Therefore, it was an interesting and important work to study polymorphisms in POU1F1 gene and their associations with production traits in goat, which could provide useful genetic markers for animal selection and breeding through marker-assisted selection (MAS).In this paper, we reported the identification of DdeI polymorphism at coding region of goat POU1F1 gene and evaluated its effects on production traits.

PCR conditions
Based upon the sequences of sheep POU1F1 gene (AJ549207) and bovine POU1F1 gene (Zhao et al., 2004), a pair of primers was designed to amplify the goat POU1F1 gene (Table 1).The 25 µl volume contained 50 ng genomic DNA, 0.5 µM of each primer, 1×Buffer (including 1.5 mM MgCl 2 ), 200 µM dNTPs and 0.625 units of Taq DNA polymerase (MBI).The cycling protocol was 4 min at 95°C, 35 cycles of denaturing at 94°C for 45 s, annealing at 54.5°C for 45 s, extending at 72°C for 1 min, with a final extension at 72°C for 10 min.

Genotyping of DdeI POU1F1 allele by PCR-RFLP
Aliquots of 20 µl PCR products of POU1F1 gene were digested with 10 U DdeI at 37°C for 5 h.The digested products were detected by 12.0% PAGE electrophoresis and stained with 0.1% silver nitrate (Zhou et al., 2006).

Statistical analysis
The chi-square test was used to analyze the Hardy-Weinberg equilibrium, which was performed by SPSS software (version 13.0).Statistical analysis was performed on the basis of records of production traits in 216 dairy goats and 452 IMWC goats, respectively.All analyses were done in two steps, first using a full animal model and then using a reduced animal model.The full animal model included fixed effects of marker genotype, birth year, season of birth (spring vs. fall), age of dam, sire, farm, sex, breed and random effects of measurements and animal.The reduced model was used in the final analysis.(1) Repeated measurements of the milk yield of 216 dairy goat (74 Sa, 80 LS, 62 GZ) were analyzed by the use of the statistical software SPSS (version 13.0) with the mixed linear model.
(2) The adjusted Linear Model I with fixed effects was used to analyze the relationships between genotypes and weight traits and litter size in 668 goats (74 Sa, 62 GZ, 80 LS and 452 IMWC).linear model I: Y ijklm = µ+S i +D ij +A k +G l + (SG) il +E ijklm , where Y ijklm was the trait measured on each of the ijklm th animal, µ was the overall population mean, S i was the fixed effect associated with the i th sire, D ij was the fixed effect associated with j th dam with sire i, A k was fixed effect due to the k th age, G l was the fixed effect associated with l th genotype (POU1F1/genotype D 1 D 1 and D 1 D 2 ), (SG) il was interaction between the i th sire and the l th genotype and E jiklm was the random error.(3) The adjusted linear model with fixed effects was used to analyze the relationship between genotypes and lana traits in 452 IMWC goats.Lack associated of farm, sex, and season of birth (spring vs. fall) with variability of traits indicated that these factors were not into linear model in IMWC

Synonymous codon bias analysis
According to the high homology among bovine, sheep and goat POU1F1 gene, the exon 6 of goat POU1F1 gene was analyzed by www.ncbi.nlm.nih.gov/Blastn and www.ebi.ac.uk/clustalw online software.Then, ORF finder was used to identify the amino acid sequence of exon 6 of goat POU1F1 gene.According to the online software (www.kazusa.or.jp/codon/countcodon.html), the codon frequencies of exon 6 of goat POU1F1 gene was calculated.Finally, the codon usage frequency was analyzed by the following formula (Kurland et al., 1991;Eyre-Walker et al., 1996;Lavner et al., 2005;Esley et al., 2006;Liu et al., 2006): F = m×k/n, among them, "F" meant codon usage frequency, " "m" meant the number of synonymous codon for specific amino acid, "n" meant the number of specific amino acid in analyzed sequences, "k" meant the number of usage for specific synonymous codon.
The establishment of relationships between genotype D 1 D 1 and D 1 D 2 and production traits was attempted (Table 3).Significant statistical results were founded in milk yield (*p<0.05), litter size (*p<0.05)and one-year-old weight (*p<0.05) between genotypes.

DISCUSSION
Goat, sheep and bovine POU1F1 gene locate in 1q21-22 of chromosomes (Woollard et al., 2000).Chormosome 1 q in ruminant is highly conserved at the gene order and cytogenetics levels.Interval mapping to detect QTL revealed significant effects on milk and protein yield associated with chromosome 1 in the region of bovine POU1F1 (Renaville et al., 1997).Moreover, the POU1F1 regulates expression of GH, PRL, TSHβ gene and itself (Sun et al., 2002).Hence, POU1F1 gene was considered to have effects on production and will benefit for the goat industry, whose DNA markers will contribute to animal selection and breeding through marker-assisted selection (MAS).
We were aware of few research related to the polymorphisms of goat POU1F1 gene and their association with production traits.In this paper, four mutations were revealed in the exon 6 of goat POU1F1 gene by PCR-SSCP and DNA sequencing method.One DQ826397:g.102T>G mutation identified a silent allele (p.S241S) and formed the DdeI polymorphism.The frequencies of POU1F1-D 1 allele varied from 0.600 to 1.000.Interestingly, we observed that D 2 D 2 genotype which was not detected in genotype analysis.We presumed that the absence of genotype D 2 D 2 associated with "major codon bias".According to the previous papers, the frequencies with which individual synonymous codons were used to code their cognate amino acids was quite variable from genome to genome and within genomes, from gene to gene.One particularly well documented codon bias was that associated with highly expressed genes in bacteria as well as in yeast; this was the so-called major codon bias (Kurland et al., 1991;Lavner et al., 2005;Esley et al., 2006;Liu et al., 2006).As the complete CDS sequence of goat POU1F1 gene was not available, the exon 6 region was used to analyze the codon bias.The analysis of codon usage frequency (F) revealed that different synonymous codons for Serine showed codon bias phenomenon.If F (codon)> 2.0, this codon was regarded as "high frequency codon"; while the codon was regarded as "low frequency codon" or "rare codon" if F (codon) = 0.000.Hence, "UCU" was regarded as "high frequency codon", while the "UCG" was called for "rare codon".From Figure 1, the genotype D 1 D 1 linked to "T" mutation and complied with the high frequency codon (UCU) which was called for major codon, while the genotype D 2 D 2 linked to "G" mutation and complied with the low frequency codon (UCG) which was called for rare codon.Major codon bias (UCU vs. UCG) was not an arrangement for regulating POU1F1 gene expression.Instead, the similar data suggested that this codon bias, which was correlated with a corresponding bias of tRNA abundance, was a global arrangement for optimizing the growth efficiency of cells (Kurland et al., 1991;Eyre-Walker et al., 1996;Coghlan et al., 2000;Archetti et al., 2004;Lavner et al., 2005;Esley et al., 2006;Liu et al., 2006).Moreover, we presumed that the rare tRNA abundance for UCG may greatly decrease the translation speed from mRNA to animo acid, and seriously restricted the synthesis efficiency of protein and others, then resulted in the absence of D 2 D 2 individual.
The statistical results revealed significant relationships between some traits and genotypes (*p<0.05).The individuals with genotype D 1 D 1 had better performance (e.g.milk yield, litter size and weight) than those of the individuals with genotype D 1 D 2 .Although this silent mutation didn't change amino acid sequence, it possibly resulted in the change of Serine synonymous codon usage frequency.The DQ826397:g.102T>G mutation changed the frequency of synonymous codon from 2.400 ("UCU", high frequency codon) to 0.000 ("UCG", rare codon).So, we presumed that the codon bias associated with expressed level of POU1F1.The geontype D 1 D 2 with rare codon (UCG) possibly associated the less expression level of POU1F1 which regulates expression level of GH, PRL and TSHβ gene, thus the genotype D 1 D 2 showed junior performance.This presume complied with the previous descriptions (Kurland et al., 1991;Eyre-Walker et al., 1996;Coghlan et al., 2000;Archetti et al., 2004;Lavner et al., 2005;Esley et al., 2006;Liu et al., 2006).Hence, we revealed that DdeI polymorphism of coding region of POU1F1 gene associated with milk performance, reproduction traits and weight traits, which preliminarily implied that POU1F1 gene has positive effects on them.
In this study, genotype D 1 D 2 of POU1F1 locus characterized by a DdeI PCR-RFLP detection was suggested to be molecular marker for junior milk yield, lambs and weight, as well as genotype D 1 D 1 for superior performances.

Figure 1 .Figure 2 .
Figure 1.The structure of goat POU1F1 gene and SNPs location for exon 6 and the DdeI enzyme site.CTNAG: the enzyme site of DdeI endonuclease;

Table 1 .
The primer sequences and their information of goat POU1F1 gene populations.The least square means estimates (LSM) with standard errors and multiple range tests for two POU1F1 genotypes and production traits were used.

Table 2 .
Genotype distribution and allelic frequencies at goat POU1F1 locus

Table 3 .
Associations of DdeI PCR-RFLP with production traits at goat POU1F1 locus Genotypes of DdeI PCR-RFLP genotyping a, b Means of traits with different superscripts were significantly different (LSD test, * p<0.05).

Table 4 .
Codon frequencies of exon 6 of goat POU1F1 gene

Table 5 .
The codon usage frequencies for Serine in the exon 6 of POU1F1 gene