Polymorphism of the prolactin gene and its association with egg production traits in native Chinese ducks

Cui Wang, Zhenhua Liang, Wenhua Yu, Yanping Feng, Xiuli Peng, Yanzhang Gong, Shijun Li 1 Key Lab of Agricultural Animal Genetics, Breeding and Reproduction, Huazhong Agricultural University, Wuhan, Hubei 430070, PR China 2 Hubei Key Laboratory of Animal Embryo and Molecular Breeding, Institute of Animal Husbandry and Veterinary, Hubei Academy of Agricultural Sciences, Wuhan, Hubei 430064, PR China


Introduction
Prolactin (PRL) is a single-chain polypeptide hormone that belongs to the growth hormone gene family and is synthesized mainly in the anterior pituitary of all vertebrates.In mammals, it regulates diverse biological functions including promoting the development of breast gland, inducing lactation, maintenance of pregnancy and impelling the development of embryos (Byrnes & Bridges, 2005;Bonomo et al., 2007;Lu et al., 2010).In avian species, PRL is a crucial hormone in induction and maintenance of incubation behaviour and regulation of the follicular development (Sharp et al., 1988;Reddy et al., 2002).
Since the chicken PRL gene was cloned and sequenced (Watahiki et al., 1989), most studies focused on detecting polymorphisms in this gene.Recently, a large number of SNPs have been reported in the chicken PRL gene.Cui et al. (2006) obtained six SNPs (C-2402T, C-2161G, T-2101G, C-2062G, T-2054A and G-2040A) and a 24-bp indel (insertion-deletion) from direct sequencing and association analysis showed that the 24-bp indel was associated with egg production and chicken broodiness traits.Three mutations screened (C-1607T, C-5749T and T-5821C) by Liu et al. (2007) showed that there is a correlation between different haplotypes and egg production.In geese, three SNPs (A-401G, G-268A and T-266A) in the 5'-proximal region of PRL gene have been screened and statistical analysis suggested that these polymorphisms have the potential to be utilized in molecular breeding for egg production (Jiang et al., 2009).All the above studies displayed that PRL is an important candidate gene on egg production.
As an important agriculture poultry species, duck egg has become a source of protein in human diet, but the egg performance of some native duck breeds remains to be improved.In this study, we used six native duck breeds as materials to screen the polymorphisms of five exons and their flanking sequences of duck PRL gene, then analyzed the correlation between the polymorphisms and the reproductive traits in an F 2 resource population of white Liancheng x white Kaiya.

Material & Methods
Blood samples from five native duck breeds were provided by the Institute of Animal Husbandry and Veterinary, Hubei Academy of Agricultural Sciences, namely the Shanma (n = 37), Shaoxing (n = 43), Jinyun (n = 37), Jingjiang (n = 86) and Youxian (n = 46).These five populations were raised outdoors and production records of every duck were not available.Samples from the F 2 resource population of white Liancheng X white Kaiya (n = 350) come from Hankou Jingwu Industry Garden Ltd. and were reared in cages in an open-sided house and subjected to conventional breeding management conditions.The traits recording included the body weight at the age of first egg, the age of the first egg and egg production of each individual duck.The egg qualities were measured at 295 -300 days, and the measurements included egg weight, Haugh unit, egg index, percentage of yolk, percentage of albumen and shell strength (Zhang et al., 2005).
Genomic DNA was extracted using the phenol-chloroform-isoamyl alcohol method (Sambrook et al., 1998) and the amount of genomic DNA was evaluated by Spectrophotometer ND-1000 (Nano-Drop, USA), adjusted to a density of 50 -300 ng/ L with TE buffer and then stored at -20 ºC for PCR amplifications.
Based on the complementary DNA sequence of the duck (Anas platyrhynchos) (GenBank Accession no: AB158611), five pairs of primers were designed using Oligo 7 primer analysis software to amplify the duck PRL gene (Rychlik, 2007), covering the full length of five exons and their flanking region sequences.Primer information is given in Table 1.The PCR reaction was performed in a 15 L final volume, containing 50 -300 ng of genomic DNA, 1×PCR buffer, 0.5 M of each primer, 25 M of dNTPs, 2.0 mM MgCl 2 and 0.2 units Taq DNA Polymerase (TakaRa, Dalian, China).Thermal cycling conditions were 5 min at 94 °C followed by 35 cycles of 30 s at 94 °C, 30 s at annealing temperature, 30 s at 72 °C, and a common final extension at 72 °C.
Aliquots of 5 L of the PCR products were mixed with 10 L of the denaturing solution (95% formamide, 25 mM EDTA, 0.025% xylene-cyanole and 0.025% bromophenol blue), heated for 10 min at 98 °C and chilled on ice rapidly.Denatured DNA was subjected to 10% PAGE (polyacrylamide gel electrophoresis) analysis, which was run with 1×TBE buffer for 16 20 h at 4 °C under a constant voltage (120 150V).The gel was stained with 0.1% silver nitrate and visualized with 2% NaOH solution (Zhang et al., 2007).
The PCR products of two samples represented different PCR-SSCP genotypes were followed by purifying, transformation, cloning prior to sequencing.The sequences were aligned by DNASTAR package (DNASTAR Inc., Madison, WI, USA), restriction endonuclease sites were detected by using Primer premier 5.0 software (Lalitha, 2000).A total of 599 samples from the six duck breeds were subjected to genotyping analysis using the PCR-RFLP technique.The PCR reaction was performed as described above.3 L of the PCR products were digested overnight with 5 units of Xba1/Pst1 (TOYOBO, Osaka, Japan) at 37 °C, and the digested products were visualized on 1.5% agarose gels stained with ethidium bromide and visualized with ultraviolet light to record the genotype of each sample.
The frequency of genotypes and P-values for Hardy-Weinberg equilibrium test were estimated and calculated by the procedure FREQ in SAS system 9.1.3(SAS Institute, Cary, N.C., USA) for 2 .Using the general linear model (GLM) procedures, associations between the SNP and egg traits were analyzed according to the following model: where Y is the observed values of egg traits; µ is the population mean; S and G, are the fixed effects of sire family and genotype, respectively; e is the random error.The CORR procedure was applied to estimate phenotypic correlations among the egg traits.Values are considered significant at P <0.05 and are presented as least square means ± standard error.

Results
Twelve SNPs were detected (Table 2), the C-381A site and C-5961T site were recognized by XbaI and Pst1 PCR-RFLP, respectively.The C-5961T mutation occurred in the functional domain of the PRL amino acid sequence and the resultant conversion of cysteine to arginine affects the existence of a heparin binding site (L-R-R-D-S-H-K).
The C-381A and C-5961T were selected to employ polymorphism analysis by using Xba1 and Pst1, PCR-RFLP, respectively.As shown in Figure 1, PCR-RFLP using Xba1 in PCR products being digested into three genotypes, GG (417 bp), TT (417 bp/354 bp) and TG (417 bp/354 bp/63 bp).Results of the chi-square fitness test indicated that Shaoxing and Jingjiang ducks were in Hardy-Weinberg equilibrium at this site; whereas for other detected populations, there were significant differences between the distribution theory and practical distribution of genotype in this site.Furthermore, these genotyping results displayed that allele G was in dominance in Shangma, Shaoxing, Jinyun, Jingjiang and Youma duck breeds, while allele T was in dominance in the F 2 stocks (Table 3).Note: 2 £ 0.05(2) £ 5.991¤ 2 £ 0.01(2) £ 9.210; ** indicates highly significant differences at P <0.01.whereas Jiangjiang and Youma ducks and the F 2 resource population were in Hardy-Weinberg disequilibrium (P <0.01).Moreover, these genotyping results indicated that allele C was in dominance in all detected duck breeds (Table 4).Association analysis between the two Xba1 polymorphisms and duck reproductive traits demonstrated that the shell strength of CC genotype was superior to those with other two genotypes (P = 0.076), while there was no significant association with any of the other reproductive traits (P >0.05) (data not shown).
For the Pst1 polymorphism, association analysis revealed significant association with egg production and egg weight (P 0.05) in the F 2 resource population.The ducks of genotype CC possessed higher egg production (P = 0.032) and egg weight (P = 0.053) than those of the CT genotype (Table 5).The TT genotype was not detected in this population.

Discussion
As a crucial hormone in the activation and maintenance of broodiness, PRL performs an important role in reproductive performance.Since the sequence of the duck PRL gene has been reported, the polymorphisms of this gene have been drawing attention (Kansaku et al., 2005).In the domestic breed of Gaoyou duck, a T-1326C mutation in intron1 was associated with egg weight at the age of 30 weeks and the proportion of double-yolk (Li et al., 2009).In Muscovy duck, two SNPs (T-3777C and A-3785G) were detected in exon 4, and found that there were significant differences between non-broodiness muscovies and broodiness muscovies, muscovies and bgducks in genotype frequencies (Wu et al., 2008).In this study, we did not detect the above polymorphism sites, although 12 novel SNPs were detected in six native duck breeds.The results revealed that the duck PRL gene is rich in polymorphisms in these duck populations.
There is a heparin binding site located in a region that is rich in basic amino acid residues, which can bind heparin and affect the activity of protein (Seno et al., 1990).This heparin binding site has also been found in the functional region of the PRL amino acid sequence in broiler chickens, dwarf chicken and turkeys (Hanks et al., 1989;Wong et al., 1991;Zhou et al., 2001).In this study, we found that the C-5961T mutation in the functional region of the PRL amino acid sequence causes an amino acid change (cysteine to arginine) which affects the existence of this heparin binding site (L-R-R-D-S-H-K).
Comparing the genotype and allele frequencies of the Xba1 site in six duck stocks, we found that the G allele was in dominance in Shangma, Shaoxing, Jinyun, Jingjiang and Youma duck breeds.However, allele T was in dominance in the F 2 resource population.This different distribution of genotypes in different duck populations may be ascribed to the different genetic background of these populations.At the Pst1 site, two genotypes were detected and the individuals of CC genotype possessed higher egg production and egg weight than those of CT genotype.Therefore, the CC genotype could be an advantageous genotype on egg production and egg weight of ducks.The fact that there were relatively fewer CT genotype ducks may be due to the high intensity of artificial selection and the disadvantaged genotype could have been eliminated gradually in the long-term breeding process.
In summary, we detected 12 novel SNPs on duck PRL gene, and analyzed the genotype distribution on six native duck breeds.The association with egg production and egg weight inspires us to study the role of duck PRL on the mechanism of egg formation and the follicular growing process in future research.

Figure 1
Figure 1 PCR-Xba1-RFLP results of duck PRL gene.Three different genotypes (GG, TT, TG) are shown at the top (M: Marker1, a lane representing a fragment of 63 bp not shown).

Figure 2
Figure 2 PCR-Pst1-RFLP results of duck PRL gene.Three different genotypes (CC, TT, CT) are shown at the top (M: DL2000 plus).

Table 1
The primer sequences and annealing temperatures of PRL gene

Table 2
The mutations on the PRL gene

Table 3
Distribution of Xba1 genotypic and allelic frequencies between different breeds £ 9.210; ** indicates highly significant differences at P <0.01; * indicates significant differences at P <0.05.

Table 4
Distribution of Pst1 genotypic and allelic frequencies between different breeds

Table 5
Statistical analysis of Pst1 genotypes and reproductive traits