【作者】 赵永国；

【导师】 卢长明；

【作者基本信息】 中国农业科学院 ， 遗传育种， 2009， 博士

【摘要】 亲本间具有一定的遗传差异是作物利用杂种优势的前提,而作为杂种优势利用最成功的作物之一,甘蓝型油菜遗传基础十分狭窄,且研究主要集中于高产、优质等少数目标性状,加剧了其遗传基础缩小的趋势,为甘蓝型油菜的可持续发展带来隐患。通过二倍体亲本种杂交、人工合成的甘蓝型油菜有着极为广泛的遗传变异,其与栽培油菜亲缘关系较远,如能引入当前油菜杂交育种,将对加快杂种优势利用与保持油菜可持续发展等具有重要意义。通过人工合成甘蓝型油菜与栽培油菜杂交,进行杂种优势利用,已经展现出较好潜力,但是缺乏深入系统的研究,NRFl(人工油菜与栽培油菜F1杂种,hybrid crossed by B.napus cultivars and resynthesized rapeseeds,NRF1)组合的杂种优势表现与利用可行性、产量性状的遗传特性、杂种产量的限制因子及内源激素与产量性状关系等方面的研究未见报道。本研究首次利用多学科交叉方式,从数量遗传学、细胞遗传学、激素生理学和分子生物学等多个角度,对NRF1杂种产量优势及其限制因子进行了剖析。以4个人工合成甘蓝型油菜与4个栽培油菜为亲本,按Grifings I法进行完全双列杂交,将56个F1组合分为RS×RS、NRFl(包括RS×BN和BN×RS两种组合)和BN×BN三种类型,系统研究了NRF1杂种产量性状等的杂种优势,讨论其在油菜育种中的利用可行性,并以Hayman方法研究NRF1产量性状的遗传特性,对其基因效应与亲本产量性状及与杂种优势的关系进行分析;通过SSR和SRAP分子标记研究人工合成甘蓝型油菜的遗传多样性,分析分子标记遗传距离与产量性状杂种优势的关系;测定不同类型油菜F1杂种及亲本的内源激素含量,研究内源激素含量及其优势变化规律,讨论其与产量性状的关系。同时,通过研究NRF1杂种的自交不亲和特性、花粉育性和染色体减数分裂行为特征,探讨人工合成甘蓝型油菜产量限制因子一每角粒数的主要机制。研究的具体结果如下:1、人工合成甘蓝型油菜杂种优势与产量性状遗传特性1.1所有56个F1组合单株产量的中亲优势为-1.53%～134.42%,平均为32.77%,除R7×R13外,其它55个组合均表现正向优势,3个产量组成因子的杂种优势以单株角果数最高,平均为24.49%,其次为每角粒数,平均为15.71%,千粒重优势最低,平均为1.22%。种子含油量具有一定的杂种优势,但优势强度低于单株产量,含油量中亲优势介于-8.47%～15.07%之间,平均为4.43%,其中44个组合表现正向优势,占总数的78.57%。1.2将所有F1组合分为RS×Rs、BN×BN和NR三种类型,NRF1杂种显示出比双亲更强的营养生长优势,在所有性状上均与双亲均值达到显著差异水平,与对照ZS10相比,在株高、分枝高、分枝长、分枝数、主花序长和单株角果数等农艺性状也普遍表现出较高的优势。1.3 32个NRF1组合的单株产量中,有7个表现正向对照优势,所占比例为21.88%,其中05Rl×XY15的单株产量为24.47g,在所有参试材料中居第二位,人工合成甘蓝型油菜杂种优势具有利用可行性。产量组成因子结果表明:NRF1组合的单株角果数(436.96)极显著高于RS×RS(416.19)和BN×BN组合(370.89);千粒重在3种类型组合中无显著差异;每角粒数则表现为BN×BN(17.35)>NR(11.58)>RS×RS(8.28),是NRF1杂种产量与产量杂种优势的主要限制因子。1.4 NRF1组合的产量与产量组成因子符合“加性-显性”模型,单株产量和每角粒数具有较高的狭义遗传力,其遗传表现为加性效应方差大于显性效应方差,单株角果数和千粒重则表现为显性效应方差大于加性效应方差。Wr+Vr与亲本的单株产量、单株角果数、千粒重和每角粒数相关系数分别为-0.86,-0.62,-0.41和0.47,表明显性基因增加前三个性状的表现值而减少每角粒数。所有56个F1组合中,单株产量最高的2个组合ZS10×XY15和05R1×XY15,其一般配合力和特殊配合力均为最高,在杂交育种时,应首先选择一般配合力好的亲本,同时注重选择特殊配合力高的组合。1.5分子标记遗传距离与单株产量及产量杂种优势的相关系数分别为-0.12和0.07,均未达到显著相关水平,通过分子标记遗传距离难以预测F1产量优势表现。2人工合成甘蓝型油菜遗传距离与遗传多样性2.1 SSR和SRAP两种标记在24份材料中共扩增出168条多态性带,其中18对SSR引物共检测到73条多态性条带,平均每对引物扩增4条带;19对SRAP引物共检测到95条多态性条带,平均每个引物扩增5条带,SRAP引物扩增多态性优于SSR标记。2.2 UPGMA聚类与主成分分析表明,人工合成甘蓝型油菜与栽培油菜在遗传相似系数O.54处分为单独的2类,两类材料间的遗传基础差异显著,且前者具有更为广泛的遗传多样性。3杂种内源激素变化规律及其与产量性状关系3.1激素含量在不同组合、不同时期差异很大,3个杂交组合中GA和iPA含量在不同时期总体变化趋势一致,即从苗期到蕾苔期逐渐升高,并在花期达到最高值,随后在15DAP角果中迅速降低。3.2所有时期的iPA含量与单株角果数、每角粒数和单株产量都为正相关关系,且三个杂交组合的iPA在所有时期均表现很强的正向优势,表明iPA对产量性状有重要作用,可能通过加快细胞分裂速度,使杂种产生更多的角果数和粒数,并产生较高的产量。3.3参试材料在花期花蕾中的4种内源激素含量均达到显著差异水平,表明内源激素含量主要由基因型控制,且GA和iPA在花蕾中含量最高,花期花蕾是内源激素调控的主要器官。4 NRF1杂种产量与产量优势主要限制因子(每角粒数)的可能机理4.1花期自交和蕾期自交结果表明,NRF1杂种存在着较强的自交不亲和性,不亲和特性的强弱与NRF1杂种结实性密切相关。其中以R2为人工品系亲本的NRF1杂种,其花交结角率、每花结籽数和每角结籽数分别为46.9%、4.3和9.2粒种子,高于其它RS为亲本时的NRF1组合,表明以R2为亲本的NRF1杂种具有相对较强的自交亲和特性,以此推断:以自交亲和性好的人工合成油菜为亲本时,可以显著提高NRF1杂种的结实性。完全双列杂交实验中,32个NRF1组合,只有4个组合的每角粒数表现正向对照优势,分别为:05R1×S72、05R1×HYl 5、05R1×XY15和05R4×xY15,其中的05R1跟R2是同一材料,这4个组合同时是所有32NRF1组合单株产量最高的4个组合。可见,提高NRF1杂种的自交亲和性,改善杂种结实性,可显著提高产量水平。4.2 NRF1杂种在细胞减数分裂时期出现异常,但在后期I及以后趋于正常;同时,虽然NRF1杂种的花粉育性(平均为80.7%)低于对照中双9号,但与花器官发育正常与否没有显著关系,表明NRF1杂种的细胞减数分裂异常和较低的花粉育性不是导致其结角率和结实性低的主要原因。更多还原

【Abstract】 It is generally believed that higher genetic distance between their respective parents results in higher heterosis. Although Brassica napus is one of the most successful crops in heterosis utilization worldwide, its genetic basis is narrow when compared with its diploid parent Brassica rapa and Brassica olecerea. The gene pool of elite rapeseed breeding material has been further eroded by an emphasis on specific traits like high yield, double low quality etc, and the genetic variability in this important oilseed crop is restricted with regard to many valuable characters for breeding purposes. The resynthesized Brassica napus crossed by its diploid parents is a useful strategy for broadening the genetic basis of breeding material, and large genetic distance to rapeseed cultivars make it possible for hybrid rapeseed breeding. So far, few thorough studies have been made on the agronomic and genetic performance of hybrids between resynthesized and natural rapeseed(NRF1).In this study, many technologies with quantative genetics, cytogenetics, physiology in hormonal level and molecular biology were first used to analysize seed yield traits of NRF1i in detail. An 8×8 complete diallel experiment using four B. napus cultivars (BN) and four resynthesized rapeseed lines (RS) as parents was designed to study heterosis and genetic inheritance for yield components. SSR and SRAP techniques were employed to assess genetic diversity in 24 materials, including 16 resynthesized rapeseed and 8 cultivars. Relationship between genetic distance and F1 performance was analyzed. Endogenous hormones were tested in three F1 hybrids and their parents to determine heterosis and their relationship to yield components. Self-incompatibility, pollen fertility and meiosis of NRF1 were analysed to detemine the main factor resulting in low seed per pod in NRF1. The main results are as follows:1. Heterosis and genetic inheritance of NRF1 combinations1.1 All the 56 hybrids had positive mid-parent heterosis (MPH) in seed yield per plant with a mean of 32.77%, except Rl×R7. As for yield components, pods per plant had highest heterosis with a mean of 24.49%, followed with seeds per pod, with a mean of 15.71%, heterosis for 1000 seed weight is low, with a mean of 1.22%. Significant heterosis for seed oil content was also observed, but it was less than that of yield per plant. Among 56 hybrids, the mid-parent heterosis for seed oil content ranged from -8.547% to 15.07% with an average of 4.43%. About 78.6% combinations (44/56) had positive heterosis for seed oil content.1.2 The 56 hybrids were divided into three types: RS×RS, BN×BN and NRF1. Compared with the both parents, NRF1 showed better agronomic characters than mid-parent values in all traits. The NRF1 were also taller in plant height, longer in primary branches and main inflorescence, more in branches and increased in pod number per plant when compared with cultivar check (ZS10).1.3 There were seven NRF) which showed positive check heterosis in seed yield per plant, accounted for 21.88%, among which 05R1×XY15 had the second highest seed yield per plant in all the 56 hybrids, indicating that NRF1 had great potential in increasing seed yield. The NRF1 had much more pods per plant (436.96) than BNxBN ( 370.89) types, while had lower seeds per pod. Because 1000 seed weight is normal in NRF1, seeds per pod is the limiting factor to increasing plant yield.1.4 Diallel analysis showed that both additive and dominant effects were significant for yield components, additive gene effects were larger than non-additive gene effects in yield per plant and seed per pod, while non-additive gene effects were larger than additive gene effects in pods per plant and 1000-seed weight. Dominant genes had positive effects on yield per plant, pods per plant and 1000-seed weight and negative effects on seeds per pod (r=-0.86, -0.62, -0.41 and 0.47, respectively). The two highest yield hybrids ZS10×XY15 and 05R1×XY15 showed best general combing ability (GCA) and special combing ability (SCA), it is requisite to select parents with high GCA and combinations with high SCA to obtain high seed yield F1s.1.5 There was no significant correlations between genetic distance and F1 yield per plant, it is difficult to predict F1 heterosis by molecular markers.2 Genetic distance and genetic diversity in resynthesized Brassica napus2.1 With SSR and SRAP molecular markers, total 168 polymorphic bands were amplified in the 24 materials, among which 73 bands were obtained from 18 pair SSRs, with a mean of 4 bands per SSR marker, 95 bands were obtained from 19 pair SRAPs, with a mean of 5 bands per SRAP marker, the polymorphism in SRAP is better than in SSR.2.2 With UPGMA (Unweighted Pair Group Mathematics Average) clustering and PCA (Principal Component Analysis), the resynthesized Brassica napus and rapeseed cultivars were significantly divided into two groups, indicating that they had different genetic basis and the former had more genetic diversity.3 Endogenous hormones in hybrids and their relationship to seed yield3.1 Endogenous hormones changed dramatically in different stages and different stages, the trends of variation were similar in three F1 hybrids for GA and iPA, which reached their highest levels at flowering stage with a mean of 876.26 ng/g FW and 1084.04ng/g FW, respectively.3.2 Except for 1000 seed weight, the other yield traits showed positive relationship to iPA in all stages, and high heterosis of iPA among three F1 combinations were detected, indicating there might be important roles of iPA to yield traits. The high level of iPA in hybrids make it possible to produce more meristem and then more pods per plant, and got higher seed yield in hybrids finally.3.3 The materials showed significant differences for four endogenous hormones in buds at flowering stage, indicating genetic variation of hormones in this stage; GA and iPA contents reached their highest level in buds at flowering stage, implying that this stage was the main stage of synthesis and metabolism of hormones.4 Possible mechanisms resulted in the low seed per pod in the NRF1 hybrids4.1 The results of bud self-pollination and flower self-pollination indicated that there was obvious self-incompatibility in NRF1, which was significantly correlated to seed per pod. The NRF1 from R2 as resynthesized rapeseed parent showed best pods per bud, seed per bud and seed per pod, indicating inferior self-incompatibility in NRF1 when R2 as resynthesized rapeseed parent. The results were in accordance with the former study in the second chapter: four NR hybrids 05Rl×S72, 05RlxHY15, 05Rl×XY15 and 05R4×XY15 showed positive check heterosis in seed per pod, 05R1 and R2 here was the same material. The four combinations were the best ones of all 32 NRF1 in seed yield per plant. 4.2 Meiosis of NRF1 hybrids was somewhat abnormal and might be responsible for the lower pollen fertility. However, no significant abnormality was found in first and second anaphases and neither in tetrad stage. No positive correlation was found between flower organ abnormality and pollen fertility. That meant the above two factors might be responsible for the lower seed per pod, but not the main factors.更多还原