节点文献

毛竹的遗传结构及群体演化

Population Structure and Phylogenetic Relationships of Phyllostachys Edulis (Poaceae) in China

【作者】 蒋维昕

【导师】 丁雨龙;

【作者基本信息】 南京林业大学 , 植物学, 2013, 博士

【摘要】 毛竹(Phyllostachys edulis (Carrière) J. Houz.),为禾本科(Poaceae)竹亚科(Bambusoideae)刚竹属(Phyllostachys)植物,是我国分布最广面积最大的重要经济和生态竹种。为了有效评估、利用和保护毛竹遗传资源,本研究利用20对微卫星荧光标记(SSR)对中国现有分布区内有代表性的34个毛竹居群进行遗传多样性及群体结构的深入研究,进而探讨毛竹现有分布格局与可能的演化特点。主要研究结果如下:(1)微卫星多态性。本研究利用从NCBI中获得的毛竹cDNA序列开发适用于毛竹群体遗传研究的20对多态微卫星(SSR)位点。所有SSR位点在毛竹物种水平上多态百分率(Pp)为100%,在804个个体中共检测到169个等位基因,每个位点的等位基因数变化范围为4~15,平均为8.5个;PIC值在0.034~0.663之间,平均值为0.318;SSR位点的等位基因片段长度范围在12~51bp之间,75%的位点遵循任意等位基因突变模型(IAM),20%和5%的位点分别按逐步突变模型(SSM)和SSR位点侧翼区可能发生插入或删除事件。(2)遗传多样性和遗传分化。毛竹拥有中等水平的遗传多样性和群体间遗传分化,34个毛竹居群的平均等位基因丰富度和Nei’s基因多样度分别为2.31和0.320,群体内的遗传变异程度与地理位置无显著相关性,难以发现不同居群间的系统地理关系;毛竹物种水平上基因多样度和基因分化系数分别为0.376和0.162,这与AMOVA分析得出15.45%的遗传变异存在于居群间结论一致。尽管毛竹的大部分遗传变异(84.55%)存在于居群内,但相比于大部分林木群体(Fst <0.15),毛竹居群间的遗传分化程度仍然较高;毛竹群体大多数位点显著偏离哈迪温伯格平衡(P <0.05),总群体水平和单个群体水平固定指数分别为-0.192和-0.440,表明无论在总体水平还是群体内个体间,毛竹群体表现为杂合体过量现象,这体现了杂合子具有更强的生活力,更能适应选择压力,同时在毛竹这一以营养繁殖为主且世代周期长的物种中,杂合子优势更能稳定的保存下来;Mantel检验表明,群体间的遗传距离与地理距离呈显著正相关(r=0.247/0.215, P <0.05),比较符合距离分离模式,根据Fst估算毛竹的基因流为1.203,表明群体间的基因交流较少,尽管人为移栽可能会增加群体间的基因流,但仅在一定距离范围内(就近移栽),没有显著影响整个毛竹的遗传变异格局。(3)克隆多样性。与有性兼营养繁殖的物种相比,毛竹的克隆多样性程度中等偏低,这可能与毛竹的独特生物学特性有关,即以营养繁殖力强,开花周期长,自然结实率低。在34个毛竹居群804个体中共检测出261个多位点基因型(其中77个出现频率≥2),克隆分布范围最广的分布于14个居群中,克隆株数最大为107;毛竹物种水平上克隆尺度为3.08,群体水平上基因型比率(PD)为0.38,平均克隆尺度(Nc)为3.1±1.39,Simple多样性指数(D)为0.74±0.19,基因型分布的均匀度(E)为0.49±0.45。(4)毛竹的聚类分析。基于贝叶斯聚类结果,将34个采样群体按遗传组成分为两大群组,大群组(C1)包括绝大多数毛竹居群,小群组(C2)主要包括陕西周至、湖南怀化、广东仁化及贵州黎平群体,中南-华南地区为两个基因库的混合区域,该结果与基于遗传距离的聚类分析结果结果一致,组间分化程度极高,Fst=0.253;根据组成分分析(PCoA)、邻接法(NJ)以及K=3时的Structure聚类结果,大群组内毛竹居群分为两个亚组,其一主要包括来自云贵川交界、湖北、安徽霍山和福建共17个居群,其二主要包括广西、江西、浙江、江苏和安徽广德共13个居群,中南的群体则分属于两个亚组,尽管亚组间不存在地理区域划分,但大多数同一地理来源的居群属于同一亚组内。(5)部分毛竹居群传播历史的验证。毛竹有着非常悠久的栽培历史,SSR分析揭示出毛竹各栽培群体的来源不同,部分遗传多样性低的群体来源比较单一。陕西周至和山东崂山为南竹北移时期引种栽培群体,其中,陕西周至遗传多样性非常高,历史资料记载其引种来源为湖南、湖北等地,通过荧光SSR分析证实,可能来自湖南怀化、广东仁化和湖北咸宁等;根据聚类分析结果,也证实了山东崂山居群来源于浙江、江西和福建;根据资料记载,云贵川的毛竹为早期引种栽培,贵州赤水的毛竹是在清乾隆三十四年(公元1769年)从福建引种栽培,通过本研究进一步证实,云贵川交界的毛竹可能来源于福建武夷山,并经过长时间的发展分化及对环境的适应,四川兴文和贵州赤水居群内存在有一定的特有等位基因。(6)保护策略。对于遗传变异较丰富的居群,如湖南怀化、贵州黎平和广东仁化,应给予重视和优先保护,防止深度破坏该物种;对于特有基因丰富度高的居群,如四川兴文,以及分布在各个群体的特异种质要加大力度进行异地保存和利用。

【Abstract】 Moso bamboo, Phyllostachys edulis (Carrière) J. Houz.(Poaceae), native to China, is themost economically and ecologically important bamboo species. Since the economic interests andthe strong clonality, it has been widely cultivated in southern China, which inevitably reduces thenatural stands and leads to gene lost of this species. An accurate assessment of genetic diversityand structure within and between populations is crucial in order to design and implementappropriate conservation strategies and utilization of biodiversity in natural and domesticatedspecies. Previous RAPD studies revealed a low genetic diversity in this bamboo, but detailedinsights into the population genetics are largely missing. In this study, twenty polymorphicmicrosatellite (SSR) markers, developed from cDNA sequences of moso bamboo (Phyllostachysedulis) and fluorescently labeled, were used to evaluate the genetic diversity, putativephylogenetic relationships and population structure of Ph. edulis including34representativepopulations (804individuals) from across the geographic range of the plant in China. The mainresearch results and conclusions were as follows,(1) SSR PolymorphismFrom Phyllostachys edulis cDNA sequences available in NCBI, twenty polymorphicmicrosatellite loci were successfully developed to assess the population genetics of Ph. edulis.All twenty loci were polymorphic and revealed a total of169alleles across all34populations.The number of detected alleles (Na) varied from4~15, with an average of8.5alleles per locus.Polymorphism information content (PIC) was0.034~0.663(average,0.318) which shows thatall loci are informative. Range of allele size (R) at SRR loci were found to vary from12~51bp.75%of loci variation followed infinite allele mutation (IAM) which creates a new allele, and20%and5%were resulted from stepwise mutation model (SSM) with increase or decrease byone repeat type and from insertion-deletion event in the flanking regions, respectively.(2) Genetic diversity, differentiation and gene flowPh. edulis possessed middling a middle level genetic diversity and population differentiation.Gene diversity (H) and differentiation (Gst) at the species level were0.376and0.162respectively,in keeping with the majority of the genetic diversity occurring within populations (84.55%)according to the AMOVA results. Despite all this, compared with other forest trees, the geneticdifferentiation in Ph. edulis is till high. Rich gene diversity was detected in HN5(0.538), GD1(0.524) and GZ2(0.465), and mean allelic richness and gene diversity within population were2.31and0.32. There was no significant correlation between genetic variation level and thepopulation locations (latitude and longitude), through which it’s difficult to exposit thephylogenetic relationships of Ph. edulis. Most loci in the34populations significantly deviatedfrom HWE, and inbreed coefficient at individual population and total population level were-0.440and-0.192due to heterozygotes excess in Ph. edulis. It revealed that heterozygotes have stronger vitality than homozygotes under selective pressure, besides, the heterozygote advantagecould be fixed in this species with long generation period. An overall estimated number of geneflow (Nm=1.203) may indicate human activities promoted gene flow among populations ofmoso bamboo, which characterized by particularly extensive vegetative reproduction.Furthermore, a significant positive correlation of the genetic distances with the geographicdistances between populations was found (r=0.247/0.215, P <0.05). This ruled a simpleisolation by distance model and implied that migrations from human occurred in a limited regionand did not significantly changed the whole genetic structure of Ph. edulis.(3) Clonal diversityAll34Ph. edulis populations were multiclonal. The vasted clone distributed across14populations and the maximum clone size was107ramets indicating the strong clonality.Relatively medium to low clonal variation, compared with other sexual vs. clonal plants, wasfound in this bamboo species, which possibly resulted from its long flowering interval. A total of261multigenotypes (genets) were identified by genotyping804individuals (ramets) and theclone size at species level is3.08. Within populations, the proportion of distinguishablegenotypes (PD) and Simpson index (D) were0.38and0.74, respectively.(4) Population structureStructure analysis exhibited that there were two gene pools among populations, althoughsome individuals had mixed ancestry. Main cluster (C1) included the majority of populationswith high mean posterior probability (0.95) and a few populations were assigned to another (C2).While hierarchy AMOVA indicated a strong genetic differentiation (Fst=0.253) between the twoclusters. Bayesian model-based structure analysis revealed the presence of two clusters thatbasically consistent with the clustering results based on genetic distance. By neighbor-joining(NJ), principal coordinate analysis (PCoA) and Structure analysis when K=3, two subgroupswere detected within the main cluster (C1). They consisted of17and13populations respectivelyand did not correlate with geographic regions. However, We found that masses of populations ina limited region belonged to the same subgroups.(5) Partial populations relationshipsMoso bamboo has a very long cultivation history in China. SSR analysis revealed that somecultivation populations had different introduced sources. For two populations (SX and SD)introduced from different sources in the1960~70s, SX showed the similarity with HN5, GD1and GZ2, while SD was similar to populations from ZJ and JX. The higher genetic diversity inSX population, linked to the document, indicated that it might be introduced from variouslocations (including HN5) for many times. According to historical materials, Ph. edulispopulations on the border of Yunnan, Guizhou and Sichuan province, in south-western China,might be introduced and cultivated long long time ago. Once there was no moso bamboo inChishui of Guizhou province. GZ1population was introduced initially from Fujian province andfounded in the year A.D.1769. Together with the geography distribution and SSR analysis, weconcluded that populations in south-western China might originate from FJ5population in south-eastern China.(6) Conservation strategiesThe genetic structure in this study suggested some ideas for the protection and managementof Ph. edulis populations. Some populations in HN5, GD1and GZ2should have the first priorityin preserving of the habitats. SC2population with higher private alleles and other distinctivegenetic resources distributed in any populations should have the highest preservation in ex situconservation. For some cultivated and highly disturbed populations, the low genetic diversityrequires careful consideration.

节点文献中: