【作者】 吴健；

【导师】 邓昌彦； 熊远著；

【作者基本信息】 华中农业大学 ， 动物遗传育种与繁殖， 2007， 博士

【摘要】 运用分子标记进行标记辅助选择，已经成为当今猪育种的一种趋势。而它的基础与前提是寻找对经济性状具有较大遗传贡献率的基因或标记。随着猪基因组研究的深入，猪的基因组信息资源已日渐丰富，特别是许多影响猪重要经济性状的数量性状位点（Quantitative Trait Locus，QTL）的成功定位，给我们提供了控制目标性状位点的大致染色体区域。进一步的研究应该从QTL的粗放位点到数量性状的基因，分离和克隆决定数量性状的基因，研究其结构和功能，最终将其应用于分子标记辅助选择，来加速猪的遗传改良。位置候选基因法综合有效的利用猪现有的基因组资源，直接从猪经济性状QTL区域内筛选候选基因进行研究，是目前最经济、快速的基因分离方法。本研究根据国内外报道的QTL定位结果、猪基因的定位信息、猪．人比较图谱以及基因生理、生化功能四方面信息，从猪4、12和17号染色体上生产性状QTL区域内选取5个基因作为拟分离的候选基因，目前，这些基因的序列、结构和功能在猪中尚未见报道。应用比较基因组学、生物信息学等方法分离鉴定了这些基因以及基因的不同转录产物，研究了5个基因的某些多态性位点与猪重要生产性状的相关性以及基因的时空表达情况，并对基因的启动子进行了基础研究。本研究具体结果如下：1．CA3（Carbonic anhydraseⅢ，碳酸酐酶3）基因：（1）本研究以人CA3基因序列作为探针，在NCBI的EST数据库中比对得到猪EST序列并进行拼接。根据拼接序列设计引物在猪cDNA中扩增，获得了大白、长白、梅山三个猪种CA3基因全长编码序列891bp，梅山品种序列已提交到GenBank，登录号为AY789514。序列分析发现，该cDNA序列具有783bp的开放阅读框，编码260个氨基酸组成的蛋白质，其编码区序列与人、鼠的相似性均为87％。（2）根据已获得的cDNA序列设计引物，在猪基因组DNA中扩增，获得三个猪种CA3基因基因组全长序列9589bp，该基因包含7个外显子和6个内含子，梅山品种序列提交到GenBank，登录号为DQ675018。（3）建立了该基因第1内含子363bp处C／T突变的PCR-NcoI-RFLP分型方法；在所检测的8个猪种中，国外猪种中均是B等位基因占优势，而在梅山猪中只检测到了AA基因型；在其它的地方猪种中，两种等位基因均存在；在139头大×梅F₂代资源群体中进行NcoI-RFLP标记与性状的关联分析，该位点基因型不同时，皮率、肥肉率、瘦肉率、瘦肥比率、板油重、花油重、眼肌高、眼肌面积、背膘厚、股二头肌色值和肌内水分差异显著或极显著。（4）不同猪种间多序列比对发现，在该基因内含子5中存在SJ160微卫星位点，内含子4中存在SJ158微卫星位点和一个串联重复为TC的新微卫星位点；在所检测的7个猪种中，SJ160微卫星位点发现了2个等位基因，SJ158微卫星位点和串联重复为TC的新微卫星位点均发现了3个等位基因。（5）在320头大×梅F₂代资源群体中进行微卫星标记与性状的关联分析，SJ160微卫星位点基因型不同时，胴体长、肩部背膘厚和臀部背膘厚性状差异显著或极显著；SJ158微卫星位点不同基因型在皮率、骨率、肥肉率、瘦肉率、屠宰率、板油重、内脂率、肋骨数、6-7腰椎间背膘厚、臀部背膘厚和肌内脂肪性状存在显著差异；串联重复为TC的新微卫星位点与皮率、屠宰率、内脂率、胴体长、臀部背膘厚、平均背膘厚和平均皮厚关联显著。（6）利用荧光定量PCR技术对CA3基因进行定量分析显示，该基因在大白猪和梅山猪肌肉组织中差异表达，在梅山猪中高表达；该基因在大白猪肌肉发育的三个阶段（1日龄、60日龄、120日龄）均有表达，而且表达量随日龄呈递增的趋势。2．PKLR（Pyruvate kinase，liver and red blood cell，肝脏丙酮酸激酶）基因：（1）利用人、鼠PKLR基因的保守序列设计引物，在猪基因组DNA中扩增，获得了大白、长白、梅山三个猪种PKLR基因部分基因组序列2367bp，包含该基因第3外显子到第9内含子序列。（2）建立了该基因第4外显子上G／A替换的PCR-HaeⅢ-RFLP和第9外显子上C／T替换的PCR-CfoI-RFLP分型方法。在所检测的5个猪种中，除了鄂西黑猪外，其它猪种HaeⅢ-RFLP均检测到三种基因型；在322头大×梅F₂代资源家系中进行关联分析，该位点基因型不同时，板油重、6-7腰椎间背膘厚、平均皮厚和眼肌高性状差异显著。在所检测的5个猪群中，CfoI-RFLP存在三种基因型，梅山猪中仅检测到BB基因型，其它猪群三种基因型均存在；在124头试验猪群（长白猪、大白猪、大白×长白杂种猪、长白×大白杂种猪）中进行性状关联分析，该位点仅对骨率、臀部背膘厚、肌内脂肪和肌内水分具有显著效应。3．AGL（Glycogen debranching enzyme，糖原脱支酶）基因：（1）利用SMART技术、染色体步移及RT-PCR方法，鉴定了大白、梅山、杜洛克三个猪种AGL基因4种不同的转录产物，其cDNA全长序列分别为5506bp、5550bp、5232bp和5402bp；它们均具有4599bp的相同开放阅读框和537bp的3’非翻译区序列，通过在5’非翻译区（外显子1和外显子2）的选择性剪接形成4种不同形式的mRNA。（2）建立了5’非翻译区C／T转换的PCR-HpyCH4V-RFLP分型方法。（3）通过染色体步移获得翻译起始位点上游1489bp的5’末端序列，序列分析发现该序列具备高等动物启动子的一般特征：TATA框、CAAT框、GC框。另外，该序列中还存在一个525bp的CpG岛区域和一些潜在的转录因子结合位点，如Smad3、Smad4、C／EBPb、MAZF、USF、SP1、AP-2、MEF2、MyoD和HNF-3b／foxa2等。4．ENO3（Muscle-specific enolase，肌肉特异性烯醇化酶）基因：（1）根据人、鼠保守序列进行RT-PCR扩增并结合SMART技术获得了大白、长白、梅山三个猪种ENO3基因的cDNA全长序列1437bp，大白品种序列已提交到GenBank，登录号为DQ355513。序列分析显示，该cDNA序列具有1305bp的开放阅读框，从第二外显子翻译，编码434个氨基酸组成的蛋白质，其编码区序列与人、鼠的相似性分别为90％和85％。（2）通过RACE-PCR以及直接在外显子1末端设计引物鉴定了该基因两种不同的转录产物：由选择性剪接，在外显子1的3’末端插入／缺失142bp的片段而形成两种不同形式的mRNA，其中具有完整外显子1序列的代表转录本1，而缺失142bp片段的短形式的mRNA代表转录本2。以最小自由能规则为依据的算法预测该基因5’非翻译区（UTR）RNA的二级结构显示，该基因两个转录本存在不同的二级结构，转录本1的5’UTR二级结构相对比较稳定，这可能是导致该转录本表达水平相对较低的原因。（3）根据已获得的cDNA序列设计引物，在猪基因组DNA中扩增，获得三个猪种ENO3基因基因组全长序列5376bp，包含12个外显子和11个内含子，大白品种序列已提交到GenBank，登录号为DQ676935。（4）建立了该基因第9内含子内405bp处T缺失突变的PCR-StuI-RFLP分型方法；在所检测的8个猪种中，均存在两个等位基因，国内猪种大部分均是A等位基因占优势，而长白、杜洛克猪中是B等位基因占优势。分别在132头试验猪群（长白猪、大白猪、大白×长白杂种猪、长白×大白杂种猪）和138头大白×梅山F₂代资源群体中进行StuI-RFLP标记与性状的关联分析，该位点基因型不同时，肥肉率、背膘厚、肌肉色值、大理石纹和肌内脂肪性状差异显著。（5）利用实时荧光定量PCR技术对ENO3基因的两个转录本进行时空表达分析显示：除了脾以外，该基因转录本1在其它的9个组织（肌肉、脂肪、心、肝、肺、肾、胃、小肠、子宫）中均有表达，在肝脏和肺中表达丰度最高，在肌肉和子宫中相对表达量最低；而转录本2在以上所有的组织中均有表达，在肌肉和心脏中表达量最高，在脂肪、肺和胃中表达量次之，在肝脏、脾、肾、小肠和子宫中均是少量表达。不同品种间表达分析显示，两个转录本在60日龄大白猪和梅山猪肌肉组织中均差异表达，转录本1在梅山猪中高表达，相反，转录本2在大白猪中高量表达，而在梅山猪中低表达。在大白猪三个不同发育阶段肌肉组织中，两个不同转录本均有表达，转录本1的表达量随日龄的增加有所降低；而转录本2在1日龄时表达量最高，2月龄略有下降，在4月龄时表达量又升高。在以上所有肌肉组织和其它组织样品中，除了肝脏和肾以外，转录本1的表达量均显著低于转录本2的表达量，尤其是在肌肉组织中，两个转录本的表达丰度相差最大。（6）通过基因组步移获得该基因转录起始位点上游1090bp的启动子序列，序列特征分析显示：该序列中存在TATA框、GC boxes、一个长度为388 bp的CpG岛区域和一些潜在的转录因子结合位点（MEF3、MEF2、AP-1、Sp-1、GATA-1、EGR1、USF、N-Myc、STAT、MyoD、AREB6、MAZ、MZF1）。（7）在该基因的启动子区构建6个缺失片段报告基因表达载体，瞬时转染猪肾细胞系，分析启动子活性，发现从-139bp到+108bp没有启动子活性，而其它5种缺失体都存在启动子活性；从-964bp到-794bp和-536bp到-355bp内，相对荧光素酶活性降低；而从-794bp到-536bp和-355bp到-260bp内，相对荧光素酶活性上升。这些结果说明在获得的启动子序列中，相对于转录起始位点来说，从-139bp到+108bp的片段不能启动基因的转录，而从-260bp到+108bp序列内已存在基因转录所必须的基本启动子序列，从-260bp到-139bp内存在启动基因转录所必须的元件；从-794 bp到-536bp和-355 bp到-260bp内存在正调控元件，而从-964 bp到-794bp和-536 bp到-355bp内存在负调控元件。5．FOXA2／HNF3B（Hepatic nuclear factor-3-beta，肝细胞核因子3B）基因：（1）以人FOXA2基因序列作为探针，在NCBI的EST数据库中比对得到猪EST序列并进行拼接。根据拼接序列设计引物，在猪cDNA中扩增，获得大白、梅山两个猪种FOXA2基因的完全编码序列1891bp，该序列具有1374 bp的开放阅读框，编码457个氨基酸组成的蛋白质。（2）根据已获得的cDNA序列设计引物，在猪基因组DNA中扩增，获得两个猪种FOXA2基因基因组全长序列3925bp，包含3个外显子和2个内含子序列，其整个基因结构与人FOXA2基因转录本2相似。（3）建立了该基因第二内含子55bp处C／G替换的PCR-AvaI-RFLP分型方法。（4）组织表达谱分析显示，该基因在肌肉、脂肪、心脏和胃中不表达，在肺中表达丰度最高，在睾丸、胚胎、肝脏、脾、肾和小肠中低量表达。更多还原

【Abstract】 Nowadays, it already becomes a tendency for the applications of marker-assistedselection （MAS） by molecular markers to the field of pig breeding. The basis of MASwas to seek the major gene and marker which contributed genetically much to theeconomic traits. With the development of pig genome research, the pig genomicinformation becomes abundant. Especially, QTL mapping of economic trait provides thechromosome fragment which contributed to the target traits. The following researchshould be from QTL coarse locus research to the variation at quantitative trait gene leveland identity the major genes for QTL and study their structure and function which will beused for marker-assisted selection to accelerate the progress of genetic improvement inpigs. Positional candidate cloning is an economic and fast method of gene isolation,which utilizes pig genomic information and seeks gene to study from economic trait QTL.Based on QTL mapping results, gene localization, pig and human comparative map andphysiological or biochemical functions of gene, five candidate genes for economic traitsQTL were selected from pig chromosome 4, 12 and 17. The sequences, structures andfunctions of these genes have not been reported in pig at present. These genes anddifferent transcripts of genes were isolated and identified with comparative genomics andbioinformatics approaches, etc. Association studies of the gene polymorphisms withproduction traits were performed and temporal and spatial expression analyses of thesegenes were investigated, as well as the general study on the promoter of the genes. Themain results are as follows:1. CA3 （Carbonic anhydraseⅢ） gene:（1） A number of pig ESTs were initially identified using the cDNA sequence ofhuman CA3 by running a BLASTN search against the GenBank ’EST-others’ databases.These ESTs were retrieved and then assembled into one contig. From this contig, primerpairs were designed and yielded overlapping PCR products. They produced a completecoding sequence of 891bp for pig CA3 of LargeWhite, Landrace and Meishan breeds,which includes 783bp of coding sequence and encodes a 260 amino acid protein andshares 87% identity with human and mouse cDNAs. The cDNA sequence of Meishanbreed has been deposited GenBank under accession number AY789514. （2） According tothe obtained cDNA sequence of porcine CA3 gene, primer pairs were designed to amplifythe DNA templates from three pig breeds. A 9589-bp genomic DNA sequence coveringthe entire coding region of porcine CA3 was amplified. The porcine CA3 gene iscomposed of 7 exons and 6 introns. The nucleotide sequence of Meishan breed has been deposited GenBank under accession number DQ675018. （3） A T/C substitution atposition 363 within intron 1 could be detected as a NcoI PCR-RFLP （restriction fragmentlength polymorphism）. Allele frequencies for NcoI PCR-RFLP were studied in unrelatedpigs from eight breeds and allele B was predominant in the Western pig breeds andgenotype AA was fixed in Meishan pigs. There existed two alleles in other Chineseindigenous pig populations. Association analysis in 139 F₂ animals showed significantassociations between NcoI-RFLP genotypes and SP, FMP, LMP, RLE LFW, CFW, LEH,LEA, BFT, MCV2 and WM. （4） Comparative sequencing of three pig breeds revealedthat microsatellite SJ160 was identified within intron 5 and SJ158 and a novelmicrosatellite marker which included a tandem repeat of （TC）_n were identified withinintron 4. We also detected the allele number and frequencies of the three loci in 7 pigbreeds. For SJ160 microsatellite marker, only two alleles were detected. For SJ158microsatellite marker and the novel microsatellite marker which included a tandemrepeat of （TC）_n, only three alleles were detected. （5） The association analysis betweenthe CA3 three microsatellite polymorphisms and product traits in 320 F₂ offspring（YorkshirexMeishan） were performed. Statistically significant associations with CL,BFT1 and BFT3 were found at the SJ160 locus. At the SJ158 locus, statisticallysignificant associations with SP, BP, FMP, LMP, DP, LFW, IFR, RNS, BFT4, BFT3 andIMF were found. At the novel microsatellite （TC）_n locus, statistically significantassociations with SP, DP, IFR, CL, BFT3, ABF and AST were found, but no significantconclusion can be made on other traits. （6） Expression analysis of CA3 gene wasperformed by Real-time PCR. The porcine CA3 gene was differentially expressed in theskeletal muscle between Yorkshire and Meishan breeds. The expression level of CA3 washigher in the skeletal muscle of Meishan than Yorkshire. During the three stages （1 day,60 day and 120 day） of skeletal muscle development in Yorkshire pigs, we observed thatthe CA3 mRNA expression was up-regulated with age.2. PKLR （Pyruvate kinase, liver and red blood cell） gene:（1） According to the high similar region between human and mouse PKLR cDNAsequences, primer pairs were designed to amplify the DNA templates from three pigbreeds. The partial DNA fragments of PKLR gene of LargeWhite, Landrace and Meishanpig breeds were isolated and the 2376bp nucleotide sequences of the fragments thatinclude the sequences from exon 3 to intron 9 were obtained with cloning and sequencing.（2） A PCR-HaeIII-RFLP developed to detect G/A substitution in exon 4 and PCR-CfoI-RFLP to detect C/T substitution in exon 9. For PCR-HaeⅢ-RFLP, unrelatedindividuals from 5 different pig breeds and the 322 F₂ animals in Large White×Meishan resource family were genotyped for population variation investigations and associationanalysis. Results showed that there were three genotypes in all pig breeds with theexception of Exi pigs. This locus was significantly associated with LFW, BFT4, AST andLEH. For PCR-CfoI-RFLP, unrelated individuals from 5 pig breeds and the 124 pigs（Landrace pigs, LargeWhite pigs, LargeWhite×Landrace and Landrace×LargeWhite pigs）were genotyped for population variation investigations and association analysis. Thereexisted three genotypes and genotype BB was fixed in Meishan pigs. Statisticallysignificant associations with BP, BFT3, IMF and WM were found.3. AGL （Glycogen debranching enzyme） gene:（1） The four transcripts of AGL gene of LargeWhite, Meishan and Duroc breedswere isolated by RACE-PCR, genome walking and RT-PCR. The cDNA sequences offour transcripts were 5506bp, 5550bp, 5232bp and 5402bp, respectively. The four formsof mRNA by differential splicing in the 5’-untranslated sequence have same codingsequence of 4599bp and 3’-untranslated sequence of 537bp. （2） A C/T substitution in 5’UTR could be detected as a HpyCH4V PCR-RFLP. （3） A fragment of 1489bp upstreamstart code was obtained by genome walking. Sequencing analysis showed the presence ofcanonical TATA, CAAT and GC boxes. Moreover, a CpG Island of 525bp and manypotential binding sites for transcription factors were identified, including Smad3, Smad4,C/EBPb, MAZF, USF, SP1, AP-2, MEF2, MyoD and HNF-3b/foxa2.4. ENO3 （Muscle-specific enolase） gene:（1） According to the high similar region between human and mouse ENO3 cDNAsequences, primer pairs were designed to amplify the cDNA templates from three pigbreeds by RT-PCR and RACE-PCR. They produced a full-length cDNA sequence of1437bp for pig ENO3 of LargeWhite, Landrace and Meishan breeds which includes1305bp of coding sequence and encodes a 434 amino acid protein and shares 90% and85% identity with human and mouse cDNAs, respectively. The cDNA sequence ofLargeWhite breed has been deposited GenBank under accession number DQ355513. （2）Two different transcripts of ENO3 gene were identified by RACE-PCR and designingprimers on 3’ end of exon 1. Differential splicing in the 5’-untranslated sequencegenerates two forms of mRNA that differ from one another in the presence or absence of a142 nucleotide fragment in 3’ end of exon 1. The transcript 1 represents the142-nucleotide insert present in the long form of ENO3 mRNA and the transcript 2represents the short form of ENO3 mRNA. RNA secondary structures of 5’UTR of ENO3two transcripts were predicted by the smallest free energy method. There are differentRNA secondary structures for two transcripts. RNA secondary structure of transcript 1 is more stable than transcript 2. It may be result in the low expression level of transcript 1.（3） According to the obtained cDNA sequence of porcine ENO3 gene, primer pairs weredesigned to amplify the DNA templates from LargeWhite, Landrace and Meishan breeds.A 5376-bp whole genomic DNA sequence covering the entire coding region of porcineEN03 was amplified. The porcine ENO3 gene is composed of 12 exons and 11 introns.The nucleotide sequence of LargeWhite breed has been deposited GenBank underaccession number DQ676935. （4） By comparing the sequences of three pig breeds, a Tdeletion mutation at position 405 in intron 9 was detected as a StuI PCR-RFLP showedallele frequency differences among 8 pig breeds. Allele A was predominant in the Chineseindigenous breeds. Allele B was predominant in the Landrace and Duroc pig breeds. Theassociation analysis between StuI-RFLP and product traits in 132 pigs （Landrace pigs,LargeWhite pigs, LargeWhite×Landrace and Landrace×LargeWhite pigs） and 138 F₂offspring （Yorkshire×Meishan） were performed. This locus was significantly associatedwith FMP, BFT, MCV, MM1, MM2 and IMF. （5） Using Real-time PCR, we carried outthe temporal and spatial expression analyses of ENO3 two transcripts in different tissues,different pig breeds and different development stages. The transcript 1 was highlyexpressed in liver and lung and only a very low level of expression was detected inmuscle, fat, heart, kidney, stomach, uterus and small intestine. The transcript 1 mRNAlevel was undetectable in spleen. The porcine ENO3 transcript 2 expression levels washigh in skeletal muscle and heart, with medium levels in fat, lung and stomach, with lowlevels in liver, spleen, kidney, small intestine and uterus. An expression pattern was alsoanalyzed in the skeletal muscle of both Western Yorkshire and Chinese Meishan pigs atday-60. The two transcripts were differentially expressed in the skeletal muscle betweenYorkshire and Meishan. The expression level of transcript 2 was higher in the skeletalmuscle of Yorkshire than Meishan; in contrast, the transcript 1 was highly expressed inthe skeletal muscle of Meishan. During the three stages of skeletal muscle development inYorkshire pigs, we observed that the ENO3 transcript 1 mRNA expression was downregulated and the expression level of transcript 2 were shown to be at their highest levelsin skeletal muscle of postnatal 1 day and decreased day-60 and then was up-regulatedwith age. The expression level of the transcript 1 was lower in all of skeletal muscle andother tissue samples than transcript 2, with the exception of liver and kidney. （6） Weproduced a consensus promoter sequence of 1090bp up-stream to the transcriptioninitiation site by genome walking. Sequencing analysis showed the presence of canonicalTATA, GC boxes, CpG Island of 388bp and many binding sites for transcription factors,including MEF3, MEF2, AP-1, Sp-1, GATA-1, EGR2, USF, N-Myc, STAT, MyoD, AREB6, MAZ and MZF1. （7） Progressive deletion constructs in the pig ENO3 promoterregion were created by PCR. For the deletion analysis, ENO3 promoter constructs weretransfected into IBRS-2 pig kidney cells and were analyzed their activities. Constructp（-139, +108） showed no luciferase activity. Deletions from -964bp to -794bp and from-536bp to -355bp decreased ENO3 promoter activity and deletions from -794bp to -536bpand from -355bp to -260bp increased the activity. These results indicate the constructp（-139, +108） has no promoter activity but the 5’ flanking region including up to -260relative to +108 contains all the elements necessary to achieve basal promoter activity.Moreover, important positive regulatory elements reside between -794bp and -536bp andbetween -355bp and -260bp, whereas negative regulatory elements lie between -964bpand -794bp and between -536bp and -355bp.5. FOXA2/HNF3B （Hepatic nuclear factor-3-beta） gene:（1） Based on the BLAST result of comparing the nucleotides of human FOXA2 geneagainst pig EST database in NCBI, a set of high matched pig ESTs were assembled intocontig. From this contig, primer pairs were designed and yielded overlapping PCRproducts. They produced a complete coding sequence of 1891bp for pig FOXA2 ofLargeWhite and Meishan breeds which includes 1374bp of coding sequence and encodesa 457 amino acid protein. （2） According to the obtained cDNA sequence of porcineFOXA2 gene, primer pairs were designed to amplify the DNA templates fromLargeWhite and Meishan breeds. A 3925-bp whole genomic DNA sequence covering theentire coding region of porcine ENO3 was amplified. The porcine FOXA2 gene iscomposed of 3 exons and 2 introns and its genomic structure shares the higher identitywith human FOXA2 transcript variant 2. （3） A C/G substitution at position 55 in intron 2could be detected as an AvaI PCR-RFLP. （4） FOXA2 transcript levels were highest inlung, with lower levels in testicle, embryo, liver, spleen, kidney and small intestine,while levels in muscle, fat, heart and stomach were barely detected.更多还原