【作者】 周俊；

【导师】 柳俊；

【作者基本信息】 华中农业大学 ， 遗传学， 2014， 博士

【摘要】 马铃薯(Solanum tuberosum L.)在地球上的栽培历史超过8000年,广泛种植于158个国家和地区,是世界上最重要的粮食作物和经济作物之一,在保障世界粮食安全中发挥着重要作用。由于无性繁殖,病毒性退化是影响马铃薯产量的重要问题。脱毒种薯的产生解决了这一问题,而试管薯是脱毒种薯生产的首要环节。不同基因型马铃薯的试管薯形成能力差异较大,同时由于受到栽培马铃薯同源四倍体遗传复杂性的限制,关于马铃薯块茎形成的遗传研究多借助于降倍后的二倍体或者近缘二倍体野生种。即便如此,目前有关马铃薯块茎形成的遗传报道也十分有限,而且多是基于田间块茎的形成情况。而栽培马铃薯为同源四倍体,且试管薯的形成与田间块茎形成在环境条件的控制方面还有一定差异,而关于栽培马铃薯试管块茎形成的遗传研究目前还是空白。本研究旨在四倍体水平上,构建马铃薯试管薯形成的遗传分离群体,在此基础上绘制遗传连锁图谱,并对试管薯形成相关QTL进行定位,以此分析马铃薯试管薯形成的遗传基础。主要研究结果如下：1.基于组织培养条件下的试管薯形成表型,对来自4个杂交组合的119个马铃薯栽培种基因型材料在两种光周期条件下进行鉴定,发现8h/d光照时间处理比16h/d光照时间处理更有利于试管薯的形成；不同组合对光周期的敏感性不同,同一组合中不同基因型间亦存在较大差异。采用经典遗传理论对组合E中96个基因型材料的表型分离进行分析,x~2检测显示,试管薯形成与否可能受一对主效基因控制,其在两种光照条件下表现出不同的遗传效应,在短日照条件下表现为累加效应,长日照条件下则表现为互补效应。2.从上述119个基因型材料中筛选得到极端表型材料E108(8h/d和16h/d光周期条件下均能形成试管薯)和E20(8h/d和16h/d光周期条件下均没有试管薯形成),构建了关于试管薯形成能力差异的F1分离群体MTI(237子代)。3.对MT I群体237子代进行试管薯形成能力表型鉴定,发现该四倍体群体在两种光周期条件下的表型均呈偏性分布。在8h/d光周期条件下,结薯的基因型个数与不结薯的基因型个数分别为200和37,其比率(200：37)符合5：1(p=0.6629)。而16h/d光周期条件下,不结薯的基因型数与结薯的基因型数分别为163和74。长日照条件下能形成试管薯的这74个基因型在短日照条件下也形成试管薯。4.利用来自128对AFLP引物组合、65对SSR引物和3对候选基因引物的711个标记,分别构建了双亲的遗传连锁图谱。父本图谱(E108)包含315个标记,全长948cM,由12个连锁群组成,所有12个连锁群都含有完整的4条同源染色体。依据其包含的64个SSR标记的定位信息,我们成功地将这12个连锁群与马铃薯12条染色体一一对应。母本图谱(E20)包含341个标记,由14个连锁群组成,全长1286cM,其中12个连锁群含有完整的4条同源染色体,其余2个连锁群分别覆盖到3条同源染色体。同样依据其包含的65个SSR标记的定位信息,我们将这14个连锁群与马铃薯染色体一一对应,其中第1和IV号染色体分别含有2个连锁群。5.基于单标记分析方法,共检测到10个标记具有显著遗传效应,其中8个标记只在短日照条件下(8h/d光周期)具有显著效应,3个标记只在长日照条件下(16h/d光周期)具有显著效应,1个标记在两种光周期条件下均具有显著效应。这10个标记中的9个分别被定位在父本E108的第V号染色体和母本E20的第V、IX和XII号染色体上6.利用双亲图谱进行区间做图(Interval Mapping),共定位到4个与马铃薯试管块茎形成相关的QTL。其中V号染色体上两个,分别为来自父本图谱的MT05和来自母本图谱的mt05;另外两个QTL均来自母本图谱,mt01-1位于I号染色体上,mt09位于IX号染色体。MT05符合加性效应,对表型变异的贡献率达到16.23%。mt01-1表现为双显性互补模型,mt05和mt09均符合单显性模型。mt05表现为正向遗传效应,mt09表现为负向遗传效应,mt01-1、mt05和mt09对表型变异的贡献率分别为6.60%、5.33%和4.81%。基于MT05的加性模型,父本E108产生的6种配子(Q12、Q13、Q14、Q23、Q24和Q34)中,显性基因型(Q-)与隐性基因型(qq)的比例为5：1,与MT I群体在短日照环境中试管薯形成与否的表型分布比例吻合,表明MT05位点中可能含有控制四倍体马铃薯试管薯形成的主效基因。7.通过MT05与mt05、mt09之间的互作分析,我们认为MT I群体中试管薯形成表型是受主效QTL效应控制,并存在微效调节位点mt09。主效QTL MT05可,能是控制个体的试管薯形成与否,而mt09则可能影响个体的试管薯形成比例。8.对3个块茎形成相关的候选基因(StSP6A, StCO和StCDF1)进行定位,结果显示,其中2个候选基因StSP6A和StCO在MT I群体中均不存在多态性,只有候选基因StCDFl被定位于父本图谱(E108)第V号染色体的短臂(16cM和18cM),且远离试管薯形成相关的主效QTL MT05的one-LOD置信区间(80-95cM)。上述结果表明,在MTI群体中定位到的与试管薯形成相关的QTLs与上述3个候选基因没有关联,即在该四倍体群体中定位的这些QTLs可能包含了新的与块茎形成相关的基因。9.将MT05区段锚定到马铃薯DM基因组序列上发现,该区段内包含216个蛋白质编码基因。将这216个基因与本实验室高通量测序结果中可能与光周期调控试管薯形成以及蔗糖浓度影响试管薯形成的基因进行比对,结果显示上述216个基因中有54个在不同蔗糖浓度影响试管薯形成的过程中差异表达,表明本研究定位到的主效QTLMT05可能包含了涉及到蔗糖调节途径影响试管薯形成的基因。本研究构建了近乎完整的马铃薯四倍体遗传连锁图谱,并首次定位了影响马铃薯试管块茎形成的QTL。基于表型分布、QTL的遗传效应模型以及QTL之间的互作分析,我们提出了该四倍体群体的试管薯形成是受到一对主效基因的调控,并存在微效基因的辅助修饰。而候选基因的定位结果显示本研究中定位到的QTLs可能包含了新的与块茎形成相关的基因。本研究的结果证实了前人报道的马铃薯块茎形成受到少数主效基因调控的遗传分析,揭示了马铃薯四倍体中可能存在新的块茎形成相关基因,提供了从试管块茎形成的角度寻找新的对于马铃薯块茎形成机理的认识和理解的可能性。更多还原

【Abstract】 Potato originated from the Andes before8000years has become one of the most important food crops and plays a vital role in global food security, feeding more than one billion people with an annual production of over300million tonnes in more than one hundred countries. Due to the vegetative propagation, the degeneration caused by viral diseases of the seed potato is the key factor to cause low yield and benefit of potatoes and it is also the most important constraint of the potato production in China. The production of virus-free seed potatoes based on in vitro tuberization has greatly alleviated this problem. So to understand the genetic basis of in vitro tuberization is not only an important aspect of theoretical research, but also critical to improve potato yield and quality. But as an autotetraploid species, the complexity of tetrasomic inheritance and the lack of pure lines increase the difficulty of genetic analysis of the inherited characteristics of cultivated potatoes (Solanum tuberosum L.) and the genetic analysis of in vitro tuberization is hard and lacking.To understand the complex genetic basis of in vitro tuberization of the cultivated potato, we constructed a tetraploid population (F1) of237genotypes segregating on in vitro tuberization, developed linkage maps of this population (F1) and mapped QTLs for the percent of in vitro tuberized plantlets (%IVT). The main results include:1. Based on the performance of in vitro tuberization in two different photoperiods, we evaluated the phenotype of119genotypes from4crosses, and we found that the8h/d photoperiod promoted the in vitro tuberization compared with16h/d; different crosses showed different sensitivity to the different photoperiods. The results of the analysis of the segregation ratio between tuberized and nontuberized genotypes suggest that the ability to tuberize in these cresses is controlled by a pair of major genes, and they have different interaction modes in different photoperiods, additive in8h/d photoperiod and complementary in16h/d photoperiod. 2. From these119genotypes, we selected E108(tuberized well and quickly both in8h/d and16h/d photoperiod) and E20(nontuberize in either photoperiod) to generate the segregation F1population MT I which consisted of237individuals.3. Used the same method, we evaluated the phenotype of this tetraploid potato population MT I. The distribution of the phenotypic data in the population MT I was skewed both in8h/d photoperiod and16h/d photoperiod. In8h/d photoperiod, the segregation ratio of the tuberized genotypes to the nontuberized genotypes in the population was200:37which was consistent with a5:1ratio (p=0.6629). The ratio of the well tuberized (%IVT>20%) genotypes to the poorly tuberized (%IVT<20%) genotypes was117:120which was consistent with a1:1ratio (p=0.8415). In16h/d photoperiod, a large part of the population had not tuberized and the segregation ratio of the tuberized genotypes to the nontuberized genotypes was74:163. All of the74genotypes which tuberized in16h/d photoperiod also tuberized in8h/d photoperiod.4. Based on the711useful markers resulted from primers of128AFLP、65SSR and3candidate genes, we constructed the genetic linkage maps of two parents. The paternal map for E108(well tuberized) consisted of315markers, covered a total length of948cM and included12linkage groups, all of which contained all four homologous chromosomes. The maternal map for E20(nontuberized) consisted of341markers, covered a total length of1286cM and included14linkage groups,12of which contained all four homologous chromosomes. All12chromosomes of potato were tagged using the104SSR markers.5. The results of the marker-trait association analysis showed that8and3markers were significantly (P<0.01) associated with the phenotypic data in8h/d and16h/d photoperiod and one of them was the identical marker. There were both positive and negative alleles affecting in vitro tuberization under each photoperiod and the same positive allele was detected both in8h/d and16h/d photoperiod. The total10markers associated with the phenotypic data in8h/d or16h/d photoperiod were located on chromosome V of E108and chromosome V, IX and XII of E20. 6. We utilized the IM routine of TetraploidMap for almost all linkage groups identified. Four QTLs were identified using the phenotypic data in8h/d photoperiod, but none was detected in16h/d photoperiod. A major QTL (MT05) with additive effect was detected on chromosome V of E108which explained16.23%of the variation for%IVT, and three minor QTLs (mt01-1, mt05and mt09) displaying duplex dominant and simplex dominant effects were located on chromosome I, V and IX of E20which explained6.60%,5.33%and4.81%of the variation for%IVT, respectively. Based on the additive model of MT05, the segregation ratio of the gametic genotypes (Q-:qq=5:1) matched the ratio of the tuberized genotypes to the nontuberized genotypes in the population suggesting that the segregation of in vitro tuberization in this population is controlled by a major-effect gene or genes.7. Based on the interaction analysis of the alleles between the major QTL (MT05) and minor QTLs (mt05and mt09), we found that there was interaction (may be epistasis) between mt09and allele3in MT05, but the effect of mt05was not significant and could be ignored. Although, the present study could not elucidate the minor QTL effect as to whether it is additive, dominant, or interactive, the major QTL effect with minor modifiers of in vitro tuberization were clearly confirmed in this population. Furthermore, we could conclude that the major QTL may control individuals of the progeny to tuberize or not, and the minor modifiers could influence individuals to tuberize well or poorly.8. We also tested the segregation of three candidate genes (StSP6A, StCO and StCDFl) in our population. There was no polymorphism in the candidate gene loci StSP6A and StCO. Only candidate gene StCDFl was mapped on the north arm of chromosome V of E108and the location of StCDFl in our population was near the position reported in the potato genome sequence superscaffold and pseudomolecule information, but out of the one-LOD support interval of MT05. The mapping results of these important candidate genes indicated that the QTL causal genes detected in our study are new. 9. The QTL support interval of MT05was matched the segment on potato (DM) genome through e-PCR, which harbored216genes. Compared these216genes with genes that were differentially regulated by day length or sucrose content identified in E26, we found that54genes of them were identical to the genes which differentially regulated by sucrose content. The result suggests that MT05causal genes may be involed in the sucrose controlled in vitro tuberization.In this study, we developed the almost complete linkage maps of a tetraploid population, identified a major QTL on chromosome V affecting in vitro tuberization, suggested a major-effect gene with minor modifiers model controlling this trait and found that the QTLs identified here correspond to new tuberization genes. Our work provides additional confirmation of previous researches which suggested that few-gene models are responsible for the tuberization process of cultivated potatoes, and we also provide the possibility that there are new genes involved in the tuberization process in tetraploid populations.更多还原