【作者】 闫伟；

【导师】 罗玉柱； Huitong Zhou； Jon GH Hickford；

【作者基本信息】 甘肃农业大学 ， 动物遗传育种与繁殖， 2013， 博士

【摘要】 本研究采用候选基因策略，以FABP4、IGFBP3和SPP1基因为候选基因，应用PCR-SSCP技术检测了中国藏绵羊2个群体（欧拉羊和甘加羊）以及新西兰9个绵羊品种共11个群体的候选基因多态性。明确绵羊候选基因单核苷酸突变位点、鉴定潜在的单体型和基因重组事件、预测潜在非同义突变（cSNP）对蛋白结构和功能的影响。研究新西兰绵羊和中国藏绵羊群体候选基因遗传特征和差异。基于FABP4蛋白对动物脂肪沉积和生长的潜在作用，分析柯伯华斯绵羊多脂型和瘦肉型群体FABP4基因变异分布差异和罗姆尼羊FABP4变异对生长性状及脂肪（GR值）和胴体性状的影响，探讨FABP4基因遗传变异作为绵羊生长、脂肪和胴体性状分子标记的可能性。主要研究结果如下：1. FABP4、IGFBP3和SPP1基因多态性9个新西兰绵羊品种和2个藏绵羊群体的1469个样本中，FABP4基因外显子2-内含子区域和外显子3-内含子3区域分别检测出11种和10种不同的SSCP带型。外显子2-内含子2区域的SSCP带型为A₁A₁、B₁B₁、C₁C₁、D₁D₁、A₁B₁、B₁C₁、A₁C₁、A₁D₁、C₁D₁、B₁D₁和B₁E₁(A₁A₁、B₁B₁、C₁C₁和D₁D₁为纯合子，A₁B₁、B₁C₁、A₁C₁、A₁D₁、C₁D₁、B₁D₁和B₁E₁为杂合子。外显子3-内含子3区域SSCP带型为A₂A₂、B₂B₂、C₂C₂、D₂D₂、A₂B₂、A₂C₂、B₂C₂、C₂D₂、B₂D₂和A₂D₂（A₂A₂、B₂B₂、C₂C₂和D₂D₂为纯合子，A₂B₂、A₂C₂、B₂C₂、C₂D₂、B₂D₂和A₂D₂为杂合子）。9个新西兰绵羊品种和2个藏绵羊群体的384只个体中，IGFBP3和SPP1均为相同的SSCP带型。绵羊FABP4基因共检测出9条不同的变异序列，分别为A₁、B₁、C₁、D₁、E₁、A₂、 B₂、 C₂和D₂（GenBank accesson number: JX290313-JX290317和JX409931-JX409934）。鉴定8处单核苷酸变异位点，分别为c.246+33位置碱基C缺失、c.246+37位置A>G转换、c.246+46位置C>T转换、c.246+47位置G>A转换、c.317位置A>G转换、c.348+166位置T>C转换、c.348+298位置T>C转换和c.348+356位置C>T转换。c.317位置A>G转换造成氨基酸突变，即106位Lys（AAG）>Arg（AGG）。2.绵羊FABP4基因SNPs连锁不平衡、单体型和重组分析对483只绵羊两多态区域SNPs进行连锁不平衡分析。结果表明绵羊FABP4基因两多态区域SNPs间连锁不平衡程度较低（D′=0.548，r2=0.119），发生重组的概率较高且存在14种潜在的单体型。单体型分别为A₁A₂、A₁B₂、A₁C₂、A₁D₂、B₁A₂、B₁B₂、B₁C₂、C₁A₂、C₁B₂,、C₁C₂、D₁A₂、D₁B₂、D₁C₂和E₁A₂。绵羊FABP4基因两多态区域分别鉴定一条Chi-like序列（5′GTTGGTGA3′和5′GCTGGTGA3′）以及多条8碱基反向重复序列，表明其可能促进绵羊FABP4重组发生。5个不同家系后代观察到多种单体型，且利用定位绵羊不同染色体DQA₂、ADRB3和PRNP基因证实各单体型个体与父代遗传关系正确，表明绵羊FABP4基因曾经发生重组事件。，3.绵羊FABP4蛋白三级结构模拟及Lys106Arg氨基酸突变效应预测采用同源建模方法构建FABP4蛋白三级结构模型，其基本单元主要由2个α螺旋和10个β链组成。模型可靠性打分为0.73（0-1之间），表明预测模型结构可靠性较高。基于建模结果，预测106位Lys（AAG）>Arg（AGG）突变不直接影响绵羊FABP4蛋白结构和功能，但c.317A>G非同义突变可能存在变异连锁效应影响绵羊FABP4蛋白和功能。4.新西兰绵羊和中国藏绵羊群体FABP4基因遗传特征及差异10个绵羊群体共575只个体中，10个绵羊群体均检测出A₁、B₁和C₁。除无角陶赛特和甘加羊群体外，8个绵羊群体检测出D₁，而E₁仅在无角陶赛特和考力代群体检出。A₁频率为38.4%，为优势等位基因且A₁B₁为优势基因型。10个绵羊群体均检测出A₂、B₂和C₂，但考力代、美利奴、欧拉羊和甘伽羊群体未检测出D₂。B₂频率为48.3%，为优势等位基因且A₂B₂为优势基因型。新西兰绵羊FABP4基因两多态区域偏离Hardy-Weinberg平衡状态（P﹤0.01），而藏绵羊FABP4基因两多态区域处于Hardy-Weinberg平衡状态（P>0.05）。罗姆尼群体与派仑代群体基因型分布无差异（P>0.05），美利奴群体与考力代群体基因型分布有差异（P<0.05）。新西兰绵羊群体FABP4基因两个多态区域均为高杂合度，且多态信息含量（PIC）多为高度多态（PIC>0.5）。中国藏绵羊群体均为高纯合度，且多态信息含量（PIC）均为中度多态（0.25<PIC <0.5）。10个绵羊群体的聚类结果表明欧拉羊和甘伽羊、考力代和美利奴羊及罗姆尼和派仑代羊分别聚在一起。5新西兰柯泊华斯绵羊瘦肉型和多脂型群体FABP4基因变异分布差异在本研究瘦肉型和多脂型群体间，FABP4基因变异分布差异显著（P<0.001）。在外显子2-内含子2区域，多脂型绵羊群体观察到5种等位基因且A₁为优势等位基因（0.51），瘦肉型绵羊观察到3种等位基因且C₁为优势等位基因（0.89）。在外显子3-内含子3区域，多脂型绵羊观察到2种等位基因且B₂为优势等位基因（0.83），瘦肉型绵羊观察到3种等位基因且C₂为优势等位基因（0.59）。6.罗姆尼绵羊FABP4基因变异对生长性状的影响分别建立3种不同的效应模型评估等位基因（单体型）存在/缺失、等位基因拷贝数和基因型对4个生长性状（初生重、断尾重、断奶重和断奶前生长率）的影响。单体型A₁-A₂和C₁-C₂均与低初生重存在关联（P<0.1）。单体型C₁-A₂与高初生重存在关联（P<0.1）。单体型C₁-A₂与低断尾重存在关联（P<0.1）。单体型C₁-C₂与低断奶重存在关联（P<0.1）。基因型对断尾重、断奶重和断奶前生长率有显著影响（P<0.05）且基因型A₂C₂断尾重（12.831±0.273Kg）、断奶重（28.849±0.452Kg）和断奶前生长率（265.668±4.779g/d）均最低。7.罗姆尼绵羊FABP4基因变异对脂肪和胴体性状的影响分别建立3种不同的效应模型评估等位基因（单体型）存在/缺失，等位基因拷贝数和基因型对脂肪性状（GR值）和胴体性状（热胴体重、腿部肉产量、腰部肉产量、肩部肉产量、总产肉量、腿部肉比例、腰部肉比例和肩部肉比例）的影响。等位基因A₁与高GR值、低腿部肉产量、低腰部肉产量和低总肉产量存在关联（P<0.05/0.1）。等位基因A₂与低热胴体重存在关联（P<0.1）。等位基因B₁与高腿部肉比例和低肩部肉比例存在关联（P<0.1）。等位基因D₁与高肩部肉产量存在关联（P<0.1）。单体型A₁-A₂与低热胴体重、低腿部肉产量、低腰部肉产量、低总肉产量、低腰部肉比例和高肩部肉比例存在关联（P<0.05/0.1）。基因型对肩部肉产量有影响（P<0.1）且基因型A₁B₁肩部肉产量（16.296±0.133Kg）最低。更多还原

【Abstract】 In this study, polymorphisms of the FABP4, IGFBP3and SPP1genes in1853sheepsourced from ten breeds （NZ Romney, Poll Dorset, Corriedal, Merino, Perendale, Dorper,Dorset Down, Suffolk, Coopworth, Ola and Ganjia） in New Zealand and China werestudied using PCR-SSCP technology. The aim was to indentify polymorphic sits,haplotypes and recombination event in candidate genes, as well as to assess the effect ofthe putative cSNP on structure and function of the protein. The diversity of geneticcharacter at FABP4locus among various population were also analysed. Given that theFABP4protein plays potential role in fat deposition and growth in the livestock, the fatand lean sheep from coopworth breed were used to analyse the association of FABP4variations with the fat/lean phenotype. To search for molecular markers, the association ofproduction traits for the fat trait （GR value）, growth traits （birth weight, tailing weight,weaning weight and pre-weaning growth rate） and carcass traits （hot weight, leg yield,loin yield, shoulder yield, total yield, proportion leg yield, proportion loin yield andproportion shoulder yield） were also analysed in Romney sheep. The main results show asfollows:1. Polymorphism of ovine FABP4, IGFBP3and SPP1A total of1469samples were typed in two polymorphic regions of FABP4gene. In theexon2-intron2region, eleven different SSCP patterns were observed （A₁A₁，B₁B₁，C₁C₁，D1D1，A1B1，B1C1，A1C1，A1D1，C1D1，B1D1and B1E1）, SSCP patterns （A1A1，B1B1，C1C1，D1D1） represent homozygous and SSCP patterns （A1B1，B1C1，A1C1， A₁D1，C1D1，B1D₁and B₁E₁） represent heterozygous. In the exon3-intron3region, ten elevendifferent SSCP patterns were observed （A₂A₂，B₂B₂，C₂C₂，D2D₂，A2B₂，A2C₂，B₂C₂，C₂D₂，B₂D₂and A₂D₂）, SSCP patterns （A₂A₂，B₂B₂，C₂C₂，D₂D₂） represent homozygousand SSCP patterns （A2B2，A2C2，B2C2，C2D2，B2D2and A2D2） represent heterozygous.In384samples, no different SSCP patterns were observed in IGFBP3and SPP1generespectively.In1469sheep, different nine sequence variants （A1，B1，C1，D1，E1，A2，B2，C2，D₂） were identified（Genbank accesson number: JX290313-JX290317; JX409931-JX409934）. Eight polymorphic sites including c.246+33C deletion, c.246+37A>Gsubstitution, c.246+46C>T substitution, c.246+47G>A substitution, c.317A>Gsubstitution, c.348+166T>C substitution, c.348+298T>C substitution, c.348+356C>Tsubstitution were identified. c.317A>G substitution results in the amino acid change:Lys（AAG）106Arg（AGG）.2. The analysis of the linkage disequilibrium, haplotypes and recombination ofFABP4geneThe linkage disequilibrium of SNPs spanning both regions was analysed in483sheep.Theresults indicated that the relationship of SNPs in two regions was not disequilibriumcompletely （D′=0.548,r2=0.119）. Fourteen different haplotypes were identified includingA₁A₂, A₁B₂, A₁C₂, A₁D₂, B₁A₂, B₁B₂, B₁C₂, C₁A₂, C₁B₂, C₁C₂, D₁A₂, D₁B₂, D₁C₂,E₁A₂.Two chi-like sequences in both polymorphic regions were detected respectivelyincluding5′GTTGGTGA3′and5′GCTGGTGA3′, as well as several reverse repeatsequences comprising of eight bases. This results indicated that above specific sequencesmay facilitate recombination in the ovine FABP4gene. Different haplotypes wereobserved in the progeny from five sires, suggesting that recombination occurred in theFABP4gene. To ensure the truth, DQA₂, ADRB3and PRNP gene located at differentchromosomes were used to confirm the genetic relationship between the sire and theirprogeny.3. The reconstruction of three-dimensional model of the FABP4protein and theeffect of Lys106Arg change on the structure and founctionThe homology modeling method was used to construct the predicted structuresuccessfully, which was composed of two α helixes and ten β chains. The score of modelparameter was0.73（0-1）, indicating a high quality for the predicted model. It is predictedthat the Lys106Arg change will not affect the structure and function of the FABP4proteinobviously, but it has the possibility for c.317A>G substitution linking with other variationsto the FABP4protein indirectly.4. The difference of genetic characters at FABP4locus among variouspopulationsIn five hundred and seventy five sheep from ten populations, the allele A₁, B₁and C₁were common. The allele D₁was observed in eight populations except the Poll Dorset andGanjia, the allele E₁was only observed in the Poll Dorset and Corriedal. The allele A₁with the frequency of38.4%, and genotype A₁B₁was the most common. The allele A₂, B₂and C₂were observed in ten populations and the allele B₂with the frequency of48.3%and genotype A₂B₂was the most common, but the allele D₂was not observed inCorriedale, Merino, Ola and Ganjia.2test indicated that variations were deviated to the Hardy-Weinberg equilibrium, but variations were in the Hardy-Weinberg equilibrium inTibetan sheep （P>0.05）.There was not a difference for genotype between the Romney andPerendale population （P>0.05）, but there was a difference for genotype between theCorriedale and Merino （P<0.05）.At two polymorphic locus of FABP4, the highheterozygosity and PIC （PIC>0.5） were observed in NZ populations, but in two bitetpopulations the high homozygosity and moderate PIC （0.25<PIC <0.5） were observed.Cluster analysis in ten populations showed that Ola and Ganjia, Romney and Perendle, aswell as Corriedale and Romney were clustered respectively.5. The difference of FABP4variations in fat and lean line from Coopworth breedin NZThere was a difference （P<0.001） for sequence variants in both regions between twolines. In the eoxn2-intron2, only the allele B₁, C₁and D₁were observed and C₁was themost common （a frequency of89%） in the lean line, whereas in the fat line, five sequencevariants were observed and sequence A₁with a frequency of51%was the most common.In the exon3-region3, only the sequence A₂, B₂and C₂were observed and C₂was themost common （a frequency of59%） in the lean line. However, the C₂was not observedand B₂with a frequency of83%was the most common in the fat line..8. Effects of FABP4variations on growth trait in Romney sheepThree different models were used to assess the association of sequence and haplotypepresent/absent, copy number, genotype with growth traits （birth weight, tailing weight,weaning weight and pre-weaning growth rate）.The presence of haplotype A₁-A₂and C₁-C₂were trending to be associated with thedecreased birth weight （P<0.1）, but haplotype C₁-A₂were trending to be associated withthe increased birth weight （P<0.1）. The presence of haplotype C₁-A₂was trending to beassociated with the decreased tailing weight （P<0.1）. The presence of haplotype C₁-C₂was trending to be associated with the decreased weaning weight （P<0.1）. Genotype had asignificant effect on tailing weight （P<0.05）, the lambs with genotypep A₂C₂have thelowest weaning weight （28.849±0.452Kg）, tailing weight（12.831±0.273Kg） andpre-weaning growth rate （265.668±4.779g/d）.9. Effects of FABP4variations on fat and carcass traits in Romney sheepThree different models were used to assess the association of sequence and haplotypepresent/absent, copy number, genotype with growth traits （hot weight, leg yield, loin yield,shoulder yield, total yield, proportion leg yield, proportion loin yield and proportion shoulder yield）.The presence of A₁was trending to be associated with the increased GR value, legyield, loin yield and total yield （P<0.05/0.1）. The presence of A₂was trending to beassociatied with the increased hot weight （P<0.1）.The presence of A₁-A₂was trending tobe associated with the decreases hot weight （P<0.1）. The presence of B₁was trending tobe associated with an increase in proportion leg yield（P<0.1）. The presence of D₁wastrending to be associated with the increases shoulder yield （P<0.1）. The presence of A₁-A₂was associated with a decrease in leg yield, hot weight, loin yield, total yield, proportionloin yield and an increase in proportion shoulder yield （P<0.05/0.1）. Genotype had aneffect on shoulder yield （P<0.1）, lambs possessing the genotype A₁B₁have the lowestshoulder yield （16.296±0.133Kg）.更多还原