association of novel polymorphisms of forkhead box l2 and ...the chinese dagu chicken is an...

Association of novel polymorphisms of forkhead box L2 and growthdifferentiation factor-9 genes with egg production traits in local Chinese

Dagu hens

N. Qin,∗ Q. Liu,∗ Y. Y. Zhang,∗ X. C. Fan,∗ X. X. Xu,† Z. C. Lv,∗ M. L. Wei,∗ Y. Jing,∗ F. Mu,∗and R. F. Xu∗1

∗Department of Animal Genetics, Breeding and Reproduction, College of Animal Science and Technology, JilinAgricultural University, Changchun 130118, Jilin, China; and †Department of Animal Genetics, Breeding andReproduction, College of Animal Science and Technology, Huazhong Agricultural University, Wuhan 130118,

Hubei, China

ABSTRACT Transcription factor forkhead box L2(FOXL2) and growth differentiation factor-9 (GDF9)genes have critical roles in the regulation of hen ovar-ian development. In the present study, these genes wereexplored as possible molecular markers associated withBW, hen-housed egg production, and egg weight inChinese Dagu hens. Samples were analyzed using thePCR–single strand conformation polymorphism (PCR-SSCP) technique followed by sequencing analysis, andtwo novel single nucleotide polymorphisms (SNPs) wereidentified within these candidate genes. Among them,an A/G transition at base position 238 in the cod-ing region of the FOXL2 gene and a G/T transver-sion at base position 1609 in exon 2 of the GDF9 genewere found to be polymorphic and named SNPs A238Gand G1609T, respectively. The SNP A238G (FOXL2)leads to a nonsynonymous substitution (isoleucine77-to-valine), and when the 360 Dagu hen samples were

divided into genotypes AA and AB, allele A was foundto be present at a higher frequency. Furthermore, theAA genotype correlated with significantly higher hen-housed egg production at 30, 43, 57, and 66 wk of ageand with a higher egg weight at 43 wk (P < 0.05).For the SNP G1609T (GDF9), the hens were typedinto TT and TC genotypes, with the T allele shownto be dominant. The TC genotype was also markedlycorrelated with higher hen-housed egg production anda higher egg weight (P < 0.05). Moreover, four haplo-types were reconstructed based on these two SNPs, withthe AATC haplotype found to be correlated with thehighest hen-housed egg production at 30 to 66 wk of ageand with higher egg weights at 43 wk (P < 0.05). Col-lectively, the two SNPs identified in this study mightbe used as possible genetic molecular markers to aidin the improvement of egg production traits in chickenbreeding.

Key words: Chinese Dagu hens, egg production traits, polymorphisms, FOXL2, GDF92015 Poultry Science 94:88–95

http://dx.doi.org/10.3382/ps/peu023

INTRODUCTION

Egg production is one of the most important eco-nomic traits in egg-laying poultry (Kim et al., 2004).Many of the modern chicken breeds have been gen-erated by conventional breeding methods (e.g., self-selection and family selection), with enhanced eggproduction traits a key focus. However, egg productionis inherited as a polygenic trait, with low to moderateheritability, making the level of genetic improvementsdifficult to estimate (Luo et al., 2007; Biscarini et al.,

C© 2015 Poultry Science Association Inc.Received July 25, 2014.Accepted September 12, 2014.1Corresponding author: College of Animal Science and Technology,

Jilin Agricultural University, Xincheng Avenue 2888, Changchun City130118, Jilin Province, China; E-mail: [email protected]; Tel: +86-0431-84556026; Fax: +86-0431-84556026

2010; Venturini et al., 2013). Previous studies have sug-gested a candidate gene approach as a cost-effectivemeans of investigating associations of gene polymor-phisms and quantitative trait loci responsible for vari-ations in traits of interest (Rothschild and Soller, 1997;Linville et al., 2001). The identification and utiliza-tion of potential candidate genes and associated geno-types can have significant economic implications andhas become increasingly important in chicken breedingprograms (Liu et al., 2010; Zhang et al., 2012). Moreand more animal breeding researchers have been at-tracted to identifying single nucleotide polymorphisms(SNPs) in candidate genes associated with egg pro-duction traits (Kim et al., 2004; Zhou et al., 2008; Xuet al., 2010a,b; Li et al., 2013). However, more candidategenes and SNPs are required for marker-assisted selec-tion in chicken breeding. Recently, the forkhead box L2(FOXL2) and growth differentiation factor-9 (GDF9)genes have been implicated in the regulation of ovarian

88

at Iowa State U

niversity on January 16, 2015http://ps.oxfordjournals.org/

Dow

nloaded from

mailto:[email protected];

http://ps.oxfordjournals.org/

ASSOCIATION OF FORKHEAD BOX L2 WITH EGG NUMBER IN HENS 89

follicular development and maturation and ovulation inchickens, with a potential effect on egg production inlaying hens (Govoroun et al., 2004; Hudson et al., 2005;Johnson et al., 2005; McDerment et al., 2012). FOXL2,a member of the forkhead (FH)/HNF-3-related familyof transcription factors, is the earliest known markerof ovarian differentiation and a key regulator of ovar-ian development and function (Schmidt et al., 2004;Uhlenhaut et al., 2009; Sridevi and Senthilkumaran,2011). In chickens, the FOXL2 gene contains asingle exon encoding a 305–amino acid protein(Govoroun et al., 2004). FOXL2 is expressed in gran-ulosa cells of developing follicles and in maturing andovulated oocytes of hen ovaries (Loffler et al., 2003;Govoroun et al., 2004) and likely plays a significantrole in granulosa cell differentiation and follicle devel-opment and maintenance, as formerly demonstrated inmammals (Schmidt et al., 2004; Pisarska et al., 2011).Furthermore, previous studies have shown FOXL2 toact as a transcriptional repressor of P450scc (CYP11A),P450aromatase (CYP19), and cyclin D2 (CCND2),which are markers of ovarian follicle proliferationand differentiation in chickens and mammals (Hudsonet al., 2005; Bentsi-Barnes et al., 2010; Kuo et al.,2012).

GDF9 is a secreted oocyte glycoprotein belonging tothe transforming growth factor beta (TGFβ) super-family (McGrath et al., 1995; Aaltonen et al., 1999;Chang et al., 2002; Su et al., 2004). Interestingly,GDF9 gene is expressed in the oocytes and granu-losa cells of ovarian follicles in chickens and is a cen-tral regulator of folliculogenesis and ovulation rate(Dong et al., 1996; Johnson et al., 2005). The chickenGDF9 gene has been cloned and found to be com-prised of two exons that encode a 454–amino acidprotein that plays a critical role in enhancing granu-losa cell proliferation in hen ovarian follicles (Interna-tional Chicken Genome Sequencing Consortium, 2004;Johnson et al., 2005). Eight GDF9 gene mutations as-sociated with increased ovulation rates in heterozygouscarriers and sterility in homozygous carriers have beenreported in Belclare and Cambridge sheep (Hanrahanet al., 2004). Correspondingly, FOXL2 and GDF9 se-quence mutations may greatly influence chicken ovar-ian development and egg-laying performance. Whilethe Chinese Dagu chicken is an important animal re-source, little characterization of FOXL2 and GDF9 ge-netic polymorphisms and possible correlations with eggperformance in indigenous Chinese breeds has beenperformed.

In this study, a PCR–single strand conformationpolymorphism (PCR-SSCP) approach and sequenc-ing analysis was used to examine FOXL2 and GDF9fragments for novel sequence polymorphisms. Associa-tions between the newly identified genotypes and theegg production traits were explored in local ChineseDagu hens. The aim of this work is to provide a poten-tial genetic molecular marker able to facilitate improvedegg production traits in chicken breeding.

MATERIALS AND METHODS

Birds and Trait Measurements

The Chinese Dagu chicken breed was provided byCollege of Animal Science and Technology of Jilin Agri-cultural University. The eggs used for hatching wererandomly selected from the Dagu chicken population,with 360 hens hatched and raised in layered batteriesunder the same rearing conditions, including free ac-cess to water and feed in accordance with the nutri-ent requirements of local Chinese Dagu hens (NY/T33-2004, China). Approaching 16 wk of age, hens werereared in individual cages under constantly maintainedconditions. All of the birds were exposed to a 16L:8Dphotoperiod, with lights on at 5:00 AM. After the startof laying, eggs were picked and recorded daily, with eggweights determined on one day per week. Following feedand water restrictions at 30 and 43 wk of age, BW wasrecorded and the individual laying performance calcu-lated. Egg production traits examined in this study in-cluded hen-housed egg production (egg-laying number)at 30, 43, 57, and 66 wk of age; and egg weight andBW at 30 and 43 wk of age. All animal experimentswere performed in accordance with laws of the People’sRepublic of China regarding animal protection.

DNA Samples and PCR Amplification

For each bird (360 birds were sampled in this work),about 1 mL blood was collected from the wing vein at300 d of age, and DNA was extracted using a standardphenol-chloroform method. To detect genomic DNApurity, 1% agarose gel electrophoresis and ultraviolet-spectrophotometer assay were performed, with finalconcentrations between 2 and 10 ng/μL detected.Primers were designed based on the forkhead box L2(FOXL2) (GenBank accession no. NM 001012612.1)and and growth differentiation factor-9 (GDF9) (Gen-Bank accession no. NC 006100.3, GeneID: 404533) genesequences in chickens. The primer pair (Table 1) usedto amplify the fragment for FOXL2 and GDF9 geneswas screened after examination.

PCR reactions were performed in a total volumeof 50 μL, including 20 μL 2X Taq Master Mix(Cwbio, Beijing, China), 100 nM of each primer, 25to 50 ng template DNA, and 22 μL RNase-free wa-ter. The PCR conditions included 94C for 2 min, fol-lowed by 35 cycles at 94C for 30 s each for denaturing,56C (FOXL2) 54C(GDF9) for 30 s for annealing (seeTable 1), 72C for 30 s for extension, and a final exten-sion at 72C for 2 min.

Cloning of PCR Products, Sequencing, andAlignment

PCR products were purified with the Wizard prepPCR purification system (Promega, Madison, WI),were cloned into the Promega pGEM-T easy vector

at Iowa State U


Dow

nloaded from


90 QIN ET AL.

Table 1. Primer information for the fragments of amplified chicken FOXL2 and GDF9 genes.

Genes Sequences of primers (5′-3′) Product length (position) Annealing temperature (C)

FOXL2 F:CCTACTCCTACGTGGCCCTGAT 150 bp 56R: GGTTGTGGCGGATGCTGTT (152–301 nt)

GDF9 F: ACTTTCACTCGGTGGATT 175 bp 54R: ATGCTGGGACATACTTGG (1472–1647 nt)

(Cwbio, Beijing, China) according to the methodspublished by Sambrook and Russell (2001), and theobtained fragments were commercially sequenced.Two independent PCR amplifications were performedfor each sampled bird, with sequences analyzed us-ing the basic local alignment search tool (BLAST)(http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastn&PAGE TYPE=BlastSearch&LINK LOC=blasthome) to confirm the expected chicken FOXL2 andGDF9 gene sequences. The obtained sequences werethen aligned using the DNAMAN software version 6.0to identify nucleotide substitutions.

Genotyping by PCR-SSCP andReconstruction of Haplotypes

To evaluate FOXL2 and GDF9 gene polymorphisms,the confirmed PCR products were further analyzedusing the single strand conformation polymorphism(SSCP) assay as previously described (Xu et al., 2005).Briefly, Every 10 μL PCR product was mixed with6 μL loading dye (0.025% bromophenol blue, 0.025%xylene cyanol, 98% deionized formamide, 2% glyc-erin, 10 mM ethylene diamine tetraacetic acid). De-natured at 99C for 10 min beforehand, the sampleswere cooled rapidly on ice and then loaded on 10%polymerized gels (acrylamide: bisacrylamide, 39:1) ofsize 16 by 18 cm. Electrophoresis was carried out at130 V for 7 h at room temperature in 1× Tris/BoricAcid/EDTA (TBE) buffer. After silver stain, the gelswere detected under upper white light of panel by agel photography system (GeneSnap from SynGene).The method of silver staining consisted of the follow-ing procedures: (1) gel syringing with doubly distilledwater for 1 min; (2) fixing with 10% ethanol for 5min; (3) discoloring with 1% nitric acid for 3 min;(4) syringing for 1 min; (5) staining in 0.2% silver ni-trate for 20 to 30 min; (6) syringing twice, each for1 min; (7) developing in 3% sodium carbonate (contain-ing 0.05% formaldehyde and 20 μg/mL hyposulphite);(8) terminating the development with 10% acetic acid;and (9) depositing in distilled water for detection. Af-ter gel photography, the images were collected and thebirds were typed based on the electrophoresis patternsof each sample. To avoid false positive or negative re-sults due to artificial manipulation in the experiment,each sample was confirmed by repeated amplificationsand detections. Haplotypes were reconstructed accord-ing to the genotyping data obtained from all 360 in-dividuals with the PHASE program (Stephens et al.,2001).

Polymorphism Evaluation

Genotype and allelic frequencies at each single nu-cleotide polymorphism (SNP) site were calculated,with each polymorphism evaluated for Hardy-Weinbergequilibrium using a Pearson goodness-of-fit chi-squaretest (df = 1). Gene homozygosity (Ho), heterozygos-ity (He), effective number of alleles (Ne), and poly-morphism information content (PIC) were statisticallyanalyzed using the POPGENE v. 1.32 software (Yehet al., 1997).

Marker-trait Association Analysis

Associations of single polymorphisms or haplotypeswith laying performance traits were analyzed using theGLM procedure in SPSS 18.0. The model was as below:

Yijk = m + Li + Gj + Fk + eijk

Where:

Yijk = Phenotypic value of the target trait, such asegg-laying numberμ = Population meanLi = Fixed effect of the lineGj = Fixed effect of the SNP genotype or haplotypeFk = Random effect of the familyeijk = Residuals

Type III sum of squares was used in each test. Valueswere considered significant at P < 0.05 and presentedas least square means ± standard errors (SE).

RESULTS

Analysis of the Nucleotide SequenceAmplified

A 150 bp PCR amplicon was obtained for fork-head box L2 (FOXL2), and a 175 bp fragment forgrowth differentiation factor-9 (GDF9) was clonedusing specifically designed primers, with the ob-tained fragments sequenced. Comparisons of the pre-dicted sequences with the corresponding GenBank se-quences (NM 001012612.1 for FOXL2 and NC 006100.3for GDF9) were performed using the Basic Lo-cal Alignment Search Tool (BLAST) software pro-vided by the National Center for BiotechnologyInformation (NCBI) server. Primer specificity was

at Iowa State U


Dow

nloaded from

http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastn&PAGEprotect unhbox voidb@x kern .06emvbox hrule width.3emTYPE=BlastSearch&LINKprotect unhbox voidb@x kern .06emvbox hrule width.3emLOC=blasthome





Figure 1. Amplification of the fragments of chicken FOXL2 andGDF9 genes. M, DL 2000 DNA marker (2,000; 1,000; 750; 500; 250; and100 bp, respectively.); Lanes 1–3, fragments of GDF9 gene amplified;Lanes 4–6, fragments of FOXL2 gene amplified.

confirmed, with PCR amplicon sizes corresponding tothe expected sequences of the candidate genes (Fig-ure 1, Table 1) and further confirmed by comparing thecloned PCR product sequences with the direct genomicPCR products from the same individual. For all of theexamined birds, no more than 2 allelic sequences wereobserved, suggesting that the primer pair specificallyamplified a single gene.

Genotyping by PCR-SSCP andReconstruction of Haplotypes

Genotyping of the amplified FOXL2 and GDF9 tar-get fragments was conducted by PCR–single strandconformation polymorphism (PCR-SSCP) analysis.The typing results showed that there were 2 genotypes(AA and AB) in the coding region of the FOXL2 gene(Figure 2) and two genotypes (TT and TC) in theexon 2 of the GDF9 gene within the Dagu hen pop-ulation (Figure 3). Haplotype reconstruction was per-formed based on these genotype data, and 4 haplotypes(ABTC, ABTT, AATC, and AATT) were identifiedamong the 360 individual hens examined. The haplo-type present at the highest frequency was the AATThaplotype (0.42), with the AATC haplotype the nextmost frequent (0.26), followed by ABTT (0.21) andABTC (0.11).

Polymorphism of the Target Sequences

Polymorphic fragments were aligned based on thePCR-SSCP banding patterns of the FOXL2 and GDF9genes following sequencing of all of the PCR prod-ucts. In the FOXL2 fragment, an A/G transition atbase position 238 in the coding region was detected(Figure 2) and named single nucleotide polymorphisms(SNP) A238G. Furthermore, this noted A/G transitionleads to a nonsynonymous substitution, isoleucine77-to-valine. For this SNP, the birds sampled were typed aseither AA or AB genotypes by PCR-SSCP analysis. In

Figure 2. PCR-SSCP band patterns at the A238G site of thechicken FOXL2 gene fragment. (A) The capital letters (AA and AB)on the top indicate the different genotypes at the SNP locus by PCR-SSCP. (B) The A/G transition at base position 238 in the exon ofthe FOXL2 gene (GenBank accession no. NM 001012612.1) was foundby sequencing and alignment. The bird with homotype A238A wasclassified as genotype AA, and the heterotype of A238G was namedgenotype AB based on the band patterns of the PCR-SSCP and thesequencing information.

the GDF9 gene fragment, a G/T transversion at baseposition 1609 in exon 2 was identified and named SNPG1609T (Figure 3). This G/T transversion leads to asynonymous substitution, with birds typed as either TTor TC following PCR-SSCP analysis.

Allele and Genotype Frequencies

Genotypic and allelic frequencies for the FOXL2 andGDF9 genes are provided in Table 2. For the SNPA238G (FOXL2 gene), the frequency of allele A wasnotably higher than allele B, with the frequency of geno-type AA higher than genotype AB in this population.For the SNP G1609T (GDF9 gene), the frequency ofallele T was higher than allele C, with the frequency ofgenotype TT higher than genotype TC. The polymor-phisms at both sites, A238G of FOXL2 and G1609T ofGDF9, were evaluated and found to exhibit significantgenetic disequilibrium between the A and B alleles ofFOXL2 and the T and C alleles of GDF9 (P < 0.05).

As shown in Table 3, gene homozygosity (Ho) washigher than gene heterozygosity (He) for both the SNPA238G (P1) of FOXL2 and for the SNP G1609T (P2)of GDF9, with effective allele numbers of 1.363 (P1)and 1.417 (P2). The value of polymorphism information

at Iowa State U


Dow

nloaded from


92 QIN ET AL.

Figure 3. PCR-SSCP band patterns at the G1609T site of thechicken GDF9 gene fragment. (A) The capital letters (TT and TC)on the top indicate the different genotypes at the SNP locus by PCR-SSCP. (B) The G/T transversion at base position 1609 in exon 2 of theGDF9 gene (GenBank accession no. NC 006100.3) was detected by se-quencing and alignment. The bird with homotype G1609G was termedgenotype TT, and the heterotype of G1609T was named genotype TC.

content (PIC) for He in P2 (GDF9 gene) was higherthan that of P1 (FOXL2 gene), but the polymorphismwas not higher and varied only from moderate (0.251)to low (0.231).

Association of Genotypes with LayingPerformance in Dagu Hens

The SNP A238G (FOXL2) genotype AA was signif-icantly associated with higher hen-housed egg produc-tion at 30, 43, 57, and 66 wk of age and with egg weightat 43 wk (Table 4; P < 0.05). Interestingly, for theSNP G1609T (GDF9), genotype TC was markedly cor-related with higher hen-housed egg production at 30,43, 57, and 66 wk of age and with egg weight at 43 wk(Table 4; P < 0.05). However, no significant differencewas observed between AA and AB or between TC andTT regarding BW at 30 and 43 wk (P > 0.05) or eggweight at 30 wk (P > 0.05).

Among the 4 haplotypes, haplotype AATC wasfound to be correlated with the highest hen-housedegg production at 30 to 66 wk of age and highestegg weight at 43 wk (Table 5; P < 0.05), followedby haplotype ABTC (P < 0.05); the lowest levelswere associated with haplotype ABTT. While haplo-type AATC and ABTC had significant difference inBW at 43 wk (P < 0.05), no significant differenceswere noted between haplotype AATC and the otherhaplotypes regarding BW. Concerning egg weight at 30wk, there were no significant differences between the4 haplotypes.

DISCUSSION

The present work aimed to elucidate the relation-ships between polymorphisms of the FOXL2 and GDF9genes and egg production traits in Chinese Dagu hens.It is well known that egg production is inherited as apolygenic trait, with low to moderate heritability (Luoet al., 2007; Biscarini et al., 2010; Venturini et al., 2013),making genetic improvement of this trait based on esti-mated breeding values more costly and poorly effective.

Table 2. Genotypic and allelic frequency at the SNP locus of FOXL2 andGDF9 genes in the Dagu chicken population.

No. of Genotype AlleleSNP Genotype chickens frequency Allele frequency χ2

A238G (FOXL2) AA 246 0.683 A 0.842 12.74∗AB 114 0.317 B 0.158

G1609T (GDF9) TT 231 0.642 T 0.821 17.15∗TC 129 0.358 C 0.179

∗P < 0.05 was accepted to be statistically significant when the data were analyzedusing a Pearson goodness-of-fit chi-square test (df = 1).

Table 3. Polymorphism information analysis of chicken FOXL2 andGDF9 genes in the Chinese local Dagu chicken population.

Gene Gene Effective PolymorphismSNP homozygosity heterozygosity allele information

(Ho) (He) number content(Ne) (PIC)

A238G (FOXL2) 0.733 0.267 1.363 0.231G1609T (GDF9) 0.706 0.294 1.417 0.251

at Iowa State U


Dow

nloaded from



Table 4. Association of the polymorphism in chicken FOXL2 and GDF9 genes with egg production trait in the local Daguhen population.

Genotypes

FOXL2 GDF9

Egg production trait AB (114) (X ± SE) AA (246) (X ± SE) TC (129) (X ± SE) TT (231) (X ± SE)

BW at 30 wk (kg) 2.83 ± 0.63 2.73 ± 0.03 2.84 ± 0.05 2.75 ± 0.36BW at 43 wk (kg) 3.11 ± 0.08 2.99 ± 0.06 2.97 ± 0.08 3.07 ± 0.06Hen-housed egg production at 30 wk (no.) 18.13 ± 2.07a 26.59 ± 1.75b 30.23 ± 2.37a 20.06 ± 1.64b

Hen-housed egg production at 43 wk (no.) 74.95 ± 5.18a 98.77 ± 3.81b 109.88 ± 4.44a 80.34 ± 3.95b



Egg weight at 30 wk (g) 55.93 ± 1.06 56.97 ± 0.65 56.95 ± 0.98 56.50 ± 0.69Egg weight at 43 wk (g) 57.97 ± 0.83a 59.87 ± 0.49b 60.61 ± 0.82a 58.66 ± 0.47b

a,bMeans within a row for each gene lacking a common superscript differ (P < 0.05). The no. represents the total number of eggs atthe corresponding age. Numbers in parentheses indicate the number of individual hens in each group (n = 360).

Table 5. Association of different haplotype polymorphisms in chicken FOXL2 and GDF9 genes with egg production trait in thelocal Dagu hen population

Haplotype

Egg production trait ABTC (39) (X ± SE) ABTT (75) (X ± SE) AATC (93) (X ± SE) AATT (153) (X ± SE)

BW at 30 wk (kg) 2.99 ± 0.12a 2.84 ± 0.06ab 2.79 ± 0.04ab 2.70 ± 0.04b

BW at 43 wk (kg) 3.16 ± 0.19a 3.10 ± 0.09ab 2.86 ± 0.09 b 2.98 ± 0.08ab

Hen-housed egg production at 30 wk (no.) 24.17 ± 4.18a 14.56 ± 2.26b 30.57 ± 2.69a 23.74 ± 2.00a

Hen-housed egg production at 43 wk (no.) 90.17±9.25bd 66.90 ± 6.09a 115.79 ± 4.02c 87.62 ± 4.50bd

Hen-housed egg production at 57 wk (no.) 129.17 ± 12.80bd 95.04 ± 9.09a 135.94 ± 6.73c 125.42 ± 6.81bd

Hen-housed egg production at 66 wk (no.) 146.83 ± 14.70bd 100.40 ± 9.59a 181.06 ± 5.77c 138.68 ± 7.70bd

Egg weight at 30 wk (g) 55.70 ± 2.92 55.59 ± 0.94 57.08 ± 0.71 57.01 ± 0.85Egg weight at 43 wk (g) 59.79 ± 2.17abc 57.75 ± 0.74a 60.88 ± 0.72b 58.71 ± 0.60ac

a,b,c,dMeans within a row for each gene lacking a common superscript differ (P < 0.05). The no. represents the total number of eggs at thecorresponding age. Numbers in parentheses indicate the number of individual hens in the group (n = 360).

However, genetic factors are destined to play a pivotalrole in promoting egg production traits to further thiseconomically important resource. Involved in this pro-cess are not only members of the glycoprotein hormonefamily of gonadotropins, such as follicle-stimulatinghormone (FSH) and luteinizing hormone (LH) but alsoa wide variety of local intraovarian factors that playcritical roles in regulating normal follicular develop-ment and oocyte maturation. These processes are me-diated by cellular and tissue-level signal transduction,including transcription factors such as forkhead box L2(FOXL2) and members of the transforming growthfactor beta (TGFβ) superfamily, including growthdifferentiation factor-9 (GDF9) (Dong et al., 1996;Duffin et al., 2009; Kim et al., 2013; Nonis et al., 2013).Polymorphisms associated with egg production–relatedhormones, growth factors, and sex hormones such asfollicle-stimulating hormone beta subunit (FSHb), LH,prolactin (PRL), growth hormone (GH) and cytosolicphosphoenolpyruvate carboxykinase (PEPCK) havebeen intensively studied in chickens (Feng et al., 1997;Parsanejad et al., 2003; Cui et al., 2006; Onagbesanet al., 2006). However, little is known regarding poly-morphisms in chicken FOXL2 and GDF9 genes. In anattempt to identify novel DNA markers associated withegg production traits in chickens, we examined poly-

morphisms in FOXL2 and GDF9 and evaluated theirassociations with egg production traits.

In the current study, two novel single nucleotide poly-morphisms (SNPs) in chicken FOXL2 and GDF9 frag-ments were identified, with the SNP A238G (FOXL2)found to significantly correlate with a heightened hen-housed egg production at 30, 43, 57, and 66 wk ofage and with increased egg weight at 43 wk (P <0.05). More interestingly, the SNP G1609T (GDF9)was significantly correlated with heightened hen-housedegg production at 30, 43, 57, and 66 wk of age andwith increased egg weight at 43 wk (P < 0.05). Whilethese two candidate genes have been previously impli-cated in follicular development (Govoroun et al., 2004;Johnson et al., 2005), the spatiotemporal localization ofthe FOXL2 transcript was found to show an expressionpattern very similar to that of the GDF9 gene in theoocytes and granulosa cells within the variously sizedfollicles examined (unpublished data by us). Further-more, FOXL2 gene expression appears to temporallyand spatially correlate with GDF9 gene expression inhen ovary development (Johnson et al., 2005), initiallysuggesting that the two candidates are coordinately in-volved in the regulation of chicken egg productivity. Inthe present study, 4 haplotypes (ABTC, ABTT, AATC,and AATT) were detected, and the association analysis

at Iowa State U


Dow

nloaded from


94 QIN ET AL.

of haplotypes showed that FOXL2 and GDF9 polymor-phisms are significantly associated with egg productiontraits in Dagu chickens. Additionally, these results in-dicate that the 2 SNPs identified in FOXL2 and GDF9might serve as possible molecular markers to aid in theimprovement of egg production traits in chicken breed-ing.

Additionally, this study demonstrated that the fre-quency of allele A was predominantly higher than alleleB and that the frequency of genotype AA was higherthan genotype AB for the SNP A238G (FOXL2). Forthe SNP G1609T (GDF9), the frequency of allele Twas significantly higher than that of allele C, and thefrequency of genotype TT was higher than that of geno-type TC. Moreover, the gene homozygosity (Ho) washigher than gene heterozygosity (He) for both SNPsA238G and G1609T. Both of this, SNPs were found tobe under genetic disequilibrium possibly due to alleleA (FOXL2) or allele T (GDF9) being naturally one ofthe predominant alleles during genetic evolution, thusbeing more conserved and more common than otheralleles in this population. Furthermore, it is possiblethat allele A or T may be tightly linked with eitheran advantageous allele or with an artificially selectedeconomically favorable trait, such as higher hen-housedegg production and egg weight. Hence, the homozygoteswith genotype AA or TT were either promoted undernatural selection pressures to be better adapted, or theywere artificially selected for favorable agricultural at-tributes. In fact, the egg production traits of local Dagupopulations have been improved with the aim of en-hancing early sexual maturity, egg-laying numbers, andegg weight traits for the past six generations. In thisprocess of Dagu bird breeding, the A or T alleles maybe coincidently linked with one or more of the selectedbreeding traits, thus presenting a possible explanationfor higher allelic frequencies. Additionally, it cannot beignored that the number of birds examined in each pop-ulation was not enough to demonstrate the true event,and an extreme allele frequency was estimated as a re-sult. Collectively, the present data strongly support theconclusion that the two novel FOXL2 and GDF9 poly-morphisms can be associated with hen-housed egg pro-duction and egg weight, thus acting as potential geneticmarkers for egg productivity in Chinese Dagu chickenbreeding.

ACKNOWLEDGMENTS

This work was supported by the National High Tech-nology Research and Development Program of China(863 Program) (No. 2011AA100305), the National Nat-ural Science Funds (No. 31272431), the China Agri-culture Research System (No. CARS-42), and the KeyProject of the Education Department of Jilin Province(No. 2013-47). We thank LetPub (www.letpub.com) forits linguistic assistance during the preparation of thismanuscript.

REFERENCES

Aaltonen, J., M. P. Laitinen, K. Vuojolainen, R. Jaatinen, N. Horelli-Kuitunen, L. Seppa, H. Louhio, T. Tuuri, J. Sjoberg, R. Butzow,O. Hovatta, L. Dale, and O. Ritvos. 1999. Human growth dif-ferentiation factor 9 (GDF9) and its novel homolog GDF9B areexpressed in oocytes during early folliculogenesis. J. Clin. En-docrinol. Metab. 84:2744–2750.

Bentsi-Barnes, I. K., F. T. Kuo, G. M. Barlow, and M. D. Pisarska.2010. Human forkhead L2 represses key genes in granulosa celldifferentiation including aromatase, P450scc, and cyclin D2. Fer-til. Steril. 94:353–356.

Biscarini, F., H. Bovenhuis, E. D. Ellen, S. Addo, and J. A. vanArendonk. 2010. Estimation of heritability and breeding valuesfor early egg production in laying hens from pooled data. Poult.Sci. 89:1842–1849.

Chang, H., C. W. Brown, and M. M. Matzuk. 2002. Genetic analysisof the mammalian transforming growth factor-beta superfamily.Endocr. Rev. 23:787–823.

Cui, J. X., H. L. Du, Y. Liang, X. M. Deng, N. Li, and X. Q. Zhang.2006. Association of polymorphisms in the promoter region ofchicken prolactin with egg production. Poult. Sci. 85:26–31

Dong, J., D. F. Albertini, K. Nishimori, T. R. Kumar, N. Lu, andM. M. Matzuk. 1996. Growth differentiation factor-9 is requiredduring early ovarian folliculogenesis. Nature 383:531–535.

Duffin, K., R. A. Bayne, A. J. Childs, C. Collins, and R. A. Anderson.2009. The forkhead transcription factor FOXL2 is expressed insomatic cells of the human ovary prior to follicle formation. Mol.Hum. Reprod. 15:771–777.

Feng, X. P., U. Kuhnlein, S. E. Aggrey, J. S. Gavora, and D. Zad-worny. 1997. Trait association of genetic markers in the growthhormone and the growth hormone receptor gene in a WhiteLeghorn strain. Poult. Sci. 76:1770–1775.

Govoroun, M. S., M. Pannetier, E. Pailhoux, J. Cocquet, J. P.Brillard, I. Couty, F. Batellier, and C. Cotinot. 2004. Isolationof chicken homolog of the FOXL2 gene and comparison of itsexpression patterns with those of aromatase during ovarian de-velopment. Dev. Dyn. 231:859–870.

Hanrahan, J. P., S. M. Gregan, P. Mulsant, M. Mullen, G. H. Davis,R. Powell, and S. M. Galloway. 2004. Mutations in the genes foroocyte-derived growth factors GDF9 and BMP15 are associatedwith both increased ovulation rate and sterility in Cambridge andBelclare sheep (Ovis aries). Biol. Reprod. 70:900–909.

Hudson, Q. J., C. A. Smith, and A. H. Sinclair. 2005. Aromataseinhibition reduces expression of FOXL2 in the embryonic chickenovary. Dev. Dyn. 233:1052–1055.

International Chicken Genome Sequencing Consortium. 2004. Se-quence and comparative analysis of the chicken genome provideunique perspectives on vertebrate evolution. Nature 432:695–716.

Johnson, P. A., M. J. Dickens, T. R. Kent, and J. R. Giles. 2005.Expression and function of growth differentiation factor-9 inan oviparous species, Gallus domesticus. Biol. Reprod. 72:1095–1100.

Kim, D., O. Ocon-Grove, and A. L. Johnson. 2013. Bone morpho-genetic protein 4 supports the initial differentiation of hen (Gallusgallus) granulosa cells. Biol. Reprod. 88:161. doi: 10.1095/biolre-prod.113.109694.

Kim, M. H., D. S. Seo, and Y. Ko. 2004. Relationship between eggproductivity and insulin-like growth factor-I genotypes in Koreannative Ogol chickens. Poult. Sci. 83:1203–1208.

Kuo, F. T., K. Fan, I. Bentsi-Barnes, G. M. Barlow, and M. D.Pisarska. 2012. Mouse forkhead L2 maintains repression of FSH-dependent genes in the granulosa cell. Reproduction 144:485–494.

Li, H. F., J. T. Shu, Y. F. Du, Y. J. Shan, K. W. Chen, X. Y. Zhang,W. Han, and W. J. Xu. 2013. Analysis of the genetic effects ofprolactin gene polymorphisms on chicken egg production. Mol.Biol. Rep. 40:289–294.

Linville, R. C., D. Pomp, R. K. Johnson, and M. F. Rothschild. 2001.Candidate gene analysis for loci affecting litter size and ovulationrate in swine. J. Anim. Sci. 79:60–67.

Liu, W. J., D. X. Sun, Y. Yu, G. Li, S. Q. Tang, Y. Zhang, Y.C. Wang, and Y. Zhang. 2010. Association of Janus kinase 2polymorphisms with growth and reproduction traits in chickens.Poult. Sci. 89:2573–2579.

at Iowa State U


Dow

nloaded from

http://www.letpub.com



Loffler, K. A., D. Zarkower, and P. Koopman. 2003. Etiology of ovar-ian failure in blepharophimosis ptosis epicanthus inversus syn-drome: FOXL2 is a conserved, early-acting gene in vertebrateovarian development. Endocrinology 144:3237–3243.

Luo, P. T., R. Q. Yang, and N. Yang. 2007. Estimation of geneticparameters for cumulative egg numbers in a broiler dam line byusing a random regression model. Poult. Sci. 86:30–36.

McGrath, S. A., A. F. Esquela, and S. J. Lee. 1995. Oocyte-specificexpression of growth/differentiation factor-9. Mol. Endocrinol.9:131–136.

McDerment, N. A., P. W. Wilson, D. Waddington, I. C. Dunn, and P.M. Hocking. 2012. Identification of novel candidate genes for folli-cle selection in the broiler breeder ovary. BMC Genomics 13:494.doi: 10.1186/1471-2164-13-494.

Nonis, D., K. J. McTavish, and S. Shimasaki. 2013. Essential butdifferential role of FOXL2wt and FOXL2C134W in GDF-9 stimu-lation of follistatin transcription in co-operation with Smad3 inthe human granulosa cell line COV434. Mol. Cell Endocrinol.372:42–48.

Onagbesan, O. M., S. Metayer, K. Tona, J. Williams, E. De-cuypere, and V. Bruggeman. 2006. Effects of genotype and feedallowance on plasma luteinizing hormones, follicle-stimulatinghormones, progesterone, estradiol levels, follicle differentiation,and egg production rates of broiler breeder hens. Poult. Sci. 85:1245–1258.

Parsanejad, R., A. Torkamanzehi, D. Zadworny, and U. Kuhn-lein. 2003. Alleles of cytosolic phosphoenolpyruvatecarboxyki-nase (PEPCK): Trait association and interaction with mitochon-drial PEPCK in a strain of White Leghorn chickens. Poult. Sci.82:1708–1715.

Pisarska, M. D., G. Barlow, and F. T. Kuo. 2011. Minireview: Rolesof the forkhead transcription factor FOXL2 in granulosa cell bi-ology and pathology. Endocrinol. 152:1199–1208.

Rothschild, M. F., and M. Soller. 1997. Candidate gene analysis todetect genes controlling traits of economic importance in domesticlivestock. Probe. 8:13–20.

Sambrook, J., and D. W. Russell. 2001. Molecular Cloning: A Lab-oratory Manual. 3rd ed. Cold Spring Harbor Laboratory Press,Long Island, NY.

Schmidt, D., C. E. Ovitt, K. Anlag, S. Fehsenfeld, L. Gredsted, A.C. Treier, and M. Treier. 2004. The murine winged-helix tran-scription factor Foxl2 is required for granulosa cell differentiationand ovary maintenance. Development 131:933–942.

Sridevi, P., and B. Senthilkumaran. 2011. Cloning and differentialexpression of FOXL2 during ovarian development and recrudes-cence of the catfish, Clariasgariepinus. Gen. Comp. Endocrinol.174:259–268.

Stephens, M., N. J. Smith, and P. Donnelly. 2001. A new statisticalmethod for haplotype reconstruction from population data. Am.J. Hum. Genet. 68:978–989.

Su, Y. Q., X. Wu, M. J. O’Brien, F. L. Pendola, J. N. Denegre, M. M.Matzuk, and J. J. Eppig. 2004. Synergistic roles of BMP15 andGDF9 in the development and function of the oocyte-cumuluscell complex in mice: Genetic evidence for an oocyte-granulosacell regulatory loop. Dev. Biol. 276:64–73.

Uhlenhaut, N. H., S. Jakob, K. Anlag, T. Eisenberger, R. Sekido,J. Kress, A. C. Treier, C. Klugmann, C. Klasen, N. I. Holter,D. Riethmacher, G. Schutz, A. J. Cooney, R. Lovell-Badge, andM. Treier. 2009. Somatic sex reprogramming of adult ovaries totestes by FOXL2 ablation. Cell 139:1130–1142.

Venturini, G. C., R. P. Savegnago, B. N. Nunes, M. C. Ledur, G. S.Schmidt, L. El Faro, and D. P. Munari. 2013. Genetic parametersand principal component analysis for egg production from WhiteLeghorn hens. Poult. Sci. 92:2283–2289.

Xu, H., X. Shen, M. Zhou, M. Fang, H. Zeng, Q. Nie, and X. Zhang.2010a. The genetic effects of the dopamine D1 receptor gene onchicken egg production and broodiness traits. BMC Genet. 11:17.

Xu, H. P., X. Shen, M. Zhou, C. L. Luo, L. Kang, Y. Liang, H. Zeng,Q. H. Nie, D. X. Zhang, and X. Q. Zhang. 2010b. The dopamineD2 receptor gene polymorphisms associated with chicken brood-iness. Poult. Sci. 89:428–438.

Xu, R. F., K. Li, G. H. Chen, Y. Z. Qiang-Ba, H. Xu, D. L. Mo, C. C.Li, B. Fan, and B. Liu. 2005. Investigation of Genetic Variationof B-G Gene with PCR-SSCP and RFLP in Chinese IndigenousChickens. Ag. Sci. in China 4:101–105.

Yeh, F. C., R. C. Yang, T. B. J. Boyle, Z. H. Ye, and J. X. Mao. 1997.POPGENE, the user-friendly shareware for population geneticanalysis. Molecular Biology and Biotechnology Centre, Universityof Alberta, Edmonton, Alberta, Canada.

Zhang, L., D. Y. Li, Y. P. Liu, Y. Wang, X. L. Zhao, and Q. Zhu.2012. Genetic effect of the prolactin receptor gene on egg produc-tion traits in chickens. Genet. Mol. Res. 11:4307–4315.

Zhou, M., M. Lei, Y. Rao, Q. Nie, H. Zeng, M. Xia, F. Liang, D.Zhang, and X. Zhang. 2008. Polymorphisms of vasoactive intesti-nal peptide receptor-1 gene and their genetic effects on broodinessin chickens. Poult. Sci. 87:893–903.

at Iowa State U


Dow

nloaded from


association of novel polymorphisms of forkhead box l2 and ...the chinese dagu chicken is an...

Documents