【作者】 单文娟；

【导师】 马合木提·哈力克；

【作者基本信息】 新疆大学 ， 动物学， 2011， 博士

【摘要】 兔属物种由于适应性强,生存环境多样,其群体遗传学和系统发育研究对于揭示物种进化、探索物种与地理环境的适应性等热点问题具有重要意义。中国的兔属动物共有9种,新疆分布有三种:塔里木兔(Lepus yarkandensis Gunther 1875),草兔(Lepus capensis Linnaeus, 1758)和雪兔(Lepus timidus Linnaeus, 1758)。目前,针对这三个物种的群体遗传学以及系统发育研究有限。塔里木兔和雪兔为国家二级保护动物,其中塔里木兔为我国特有种,仅分布于新疆塔里木盆地周围分散的绿洲,栖息地片段化程度日益严重。本研究选用两个线粒体基因(mtDNA)标记:控制区(D-loop)和细胞色素b(Cytb)基因,以检测塔里木兔种群的遗传多样性水平、基因交流状况;分析地理隔离对该物种群体遗传结构的作用;评价第四纪气候变化对塔里木兔遗传多样性及群体历史的影响。本研究测定了来自塔里木盆地周围二十个采集地224个塔里木兔样本的D-loop区553bp序列和Cytb全长1140bp的序列。通过对所得基因的序列特征分析,结合Mantel tests分析和多层次分子变异分析(AMOVA),发现塔里木兔群体总体的mtDNA核苷酸多样度与其它哺乳动物的相比属于较低水平(D-loop 3.3%和Cytb 0.8%),各地理种群具有丰富的特有单倍型,种群间已经出现显著分化,这种遗传分化和单倍型的地理分布模式是由于地理隔离(河流、绿洲、沙漠、戈壁)及栖息地片段化(绿洲退缩及分散分布)而导致的种群间有限的基因流造成的。对这两个多态性遗传标记的单独及合并数据分别用邻接法(Neighbor- joining;NJ)和贝叶斯法(Bayesian inference)构建系统发育树,结果显示:224个塔里木兔样品分为五个进化枝,每个进化枝包含的样品分布区稍显混杂,然而,当将这20个采样地划分为东西南北四个地理种群时,西南部群体与东北部群体便呈现出明显的遗传分化,合并数据的中介网络图更清晰地反映了这种系统地理分化格局。种群动力学及遗传多样性数据显示,塔里木盆地西南部可能是塔里木兔在最大冰期时的避难所,分别发生在0.21 MYA、0.090 MYA及0.054 MYA的三次冰期后扩张事件导致塔里木兔现今的分布格局。值得注意的是,不是冰期,而是水源,才是塔里木兔退缩至避难所的主要推动力。从系统树及实际的地理隔离看,塔里木兔可能经历了从塔里木盆地西南部到北部再到东部的扩散过程。种群历史过程及冰期避难所的研究也为塔里木兔生物多样性保护重点区域的划分提供了依据。基于本文线粒体基因的研究结果,塔里木兔五个进化枝具有较高的支持率和较多的突变步数,各进化枝间的遗传距离(0.7%-2.3%)达到亚种水平,西南组群与东北组群间分歧显著,均提示塔里木兔可能存在亚种的分化。但是,这种亚种划分仅基于我们线粒体基因的研究结果,是否有效还需要结合核基因、形态学以及生态学方面的证据。新疆草兔(Lepus capensis)的群体遗传结构至今无系统的研究报道,亚种水平的分类也长期存在争议。本文测定了形态分类上的新疆草兔三个亚种共74个个体的mtDNA控制区553bp和Cytb全长1140bp的序列。新疆草兔核苷酸多样度为0.021±0.010;单倍型多样度为0.974±0.007。Mantel tests和AMOVA分析显示,新疆草兔除东部与中部种群间没有明显分化外,其余种群间已产生十分明显的分化,种群间的基因交流有限,可能是由于山脉戈壁等地理隔离而造成的。新疆草兔群体的中性检验和核苷酸错配分析显示多峰分布,这主要是由新疆草兔群体存在较为明显的遗传分化而致。基于新疆草兔mtDNA D-loop和Cytb的合并数据所构的NJ树和贝叶斯树以较高的支持率聚为三枝,每枝包含了来自相同或相邻地理区域的草兔样品,显示出了一定的地理分布格局。中介网络图也得到了相同的结果。本研究的结果支持形态分类上草兔西域亚种(L.c. lehmanni)的分类地位;但中亚亚种(L.c. centrasiaticus)被分为两个独立的进化枝,提示可能存在两个亚种;帕米尔亚种(L.c. pamirensis)可能已达到种的分化水平。目前关于新疆兔属物种间的系统发育关系还不清楚。本研究采用线粒体两个基因(D-loop和Cytb)以及一个核基因(MGF)标记来探讨新疆兔属物种间的系统发育关系。线粒体系统发育树的构建采用邻接法和贝叶斯法,核基因采用邻接法和最小进化法(Minimum Evolution; ME)。我们共测定了353个新疆野兔样本的线粒体控制区553bp和Cytb全长1140bp的序列;由于存在高比例的杂合子,通过分子克隆共得到了304个个体的453条核基因MGF序列(592bp)。从线粒体的系统发育树上可以看出,新疆的草兔和塔里木兔序列分为3个世系:其中两个世系分别代表塔里木兔与草兔各自特有的世系,另一个世系为新发现的世系。新世系包含了来自塔里木兔及草兔采集地的样品。三个世系中草兔最先分歧出来,塔里木兔与新世系形成姐妹群的关系。唯一的一条新疆雪兔线粒体序列包含在草兔的世系中。核基因MGF的分析结果表明,草兔和塔里木兔序列在核基因进化树上相互混杂并大致分为两大类群。雪兔的核基因单独形成一枝,并且最先分歧出来。线粒体与核基因的结果相矛盾说明新疆草兔、塔里木兔以及雪兔之间可能存在杂交,但是进一步的证据还需要来自形态方面如外形及头骨的考查。更多还原

【Abstract】 Hares (Lepus) distribute throughout China and occupy a wide variety of habitats, including deciduous, boreal and temperate rain forest, prairie and shrub-steppe. Study about hare species is very worthful to understand the species evolution and assess the relationships between species and environment. Of the nine Chinese hare species,three distribute in Xinjiang: Lepus yarkandensis Gunther 1875, Lepus timidus Linnaeus 1758 and Lepus capensis Linnaeus, 1758. To date, studies about molecular evolution of these three species are rather limited.Yarkand hare and mountain hare were listed in the Second Category of State Key Protected Wildlife List in 1988. The Yarkand hare is endemic to China and restricted to the scattered oases around the Taklamakan Desert in Tarim Basin. Unfortunately, in recent years, habitat fragmentation of Yarkand hare is becoming serious. In this study, two mitochondrial DNA (mtDNA) markers, control region (D-loop) and cytochrome b (Cytb), were used to examine population genetic diversity and assess the impact of geological isolation on phylogeographical structure of Yarkand hare. Also, Quaternary climatic oscillations may have affected the ecosystems and, consequently, the distributions and genetic structuring of the Yarkand hare. In this study, we obtained 553bp D-loop and 1140bp Cytb full sequences from 224 Yarkand hares. Nucleotide diversity value of Lepus yarkandensis is relatively low (D-loop 3.3%, Cytb 0.8%) compared with that of other reported mammals. Mantel tests and molecular variance analysis (AMOVA) suggested that population divergence was significant because a large number of private haplotypes were identified in different populations of Yarkand hare. This differentiation indicated gene flow among populations was limited, which might result from geographical isolation (rivers, oases, deserts) and habitat fragmentation. On genealogical tree, haplotypes of Yarkand hare were divided into 5 deeply divergent clades with high bootstrap support. When the samples were combined into four geographical groups, phylogeographic differentiation between the southwest group and the northeast group was observed and this structure was supported by pairwise FST values and Median-joining network (MJN). Populations from the northern and eastern Tarim Basin shared a similar history, as did those from the western and southern regions. Moreover, Demographical analysis and genetic diversity estimation indicated that the western and southern regions might have served as glacial refugia for the Yarkand hare during Quaternary climatic oscillations. The distribution of the Yarkand hare, especially in the northern and eastern parts, probably represented 3 postglacial colonization events, dated to 0.21, 0.090 and 0.054MYA, which corresponded to known interglacial periods. Given the relatively complete geographic isolation between the eastern and southern populations, the Yarkand hare likely dispersed during postglacial periods from the southwest to the north, and then onward to the east. Significantly, availability of water, rather than displacement by glaciers, was the dominant driving force for the retreat of the Yarkand hare to refugia. This differed from the dynamic mechanism for refugia occupation in Europe, North America, and the Qinghai-Tibetan Plateau in China. The demographical and historical patterns have important implications for conservation. In addition, significant sequence divergence (0.7%-2.3%) among five clades might suggest that subspecies differentiation based on mtDNA existed in Yarkand hare. However, whether the division of subspecies valid or not will require additional evidence from morphology and behavior.To date, the genetic structure and genetic diversity of Lepus capensis in Xinjiang has not been systematically studied at the molecular level, and its subspecies taxonomic status has been under debate for years. According to traditional morphology, there are three subspecies of Lepus capensis distributed in Xinjiang: L.c. centrasiaticus, L.c. lehmanni and L.c. pamirensis. In this study, we determined 553bp D-loop and 1140bp Cytb sequences of 74 cape hares from Xinjiang Province. Nucleotide diversity of Cape hare in Xinjiang is 0.021±0.010 and haplotype diversity is 0.974±0.007. Mantel test and AMOVA analysis revealed a high level of differentiation among populations except central and eastern populations, suggesting that mountains and deserts may make an effective barrier against gene flow. Phylogenetic tree based on mtDNA grouped 74 Cape hare sequences into three clades with high bootstrap support. Each clade included samples from the same area or neighboring areas, which indicate a significant phylogeographic division pattern in Cape hare. Results from median-joining network analysis are in accordance with the phylogenetic analysis. Neutrality tests was insignificant and mismatch distrbution analysis showed a multimodal distribution which was caused by significant differentiation among populations. Our data supported the subspecies status of L. c. lehmanni. The fact that haplotypes of L. c. centrasiaticus were grouped into two distinct clades suggests that this traditional subspecies should be considered as two subspecies. In addition, L. c. pamirensis shows a significantly higher sequence divergence compared to other subspecies, and the difference even reached the level of species.The phylogenetic relationships among hare species in Xinjiang remain unclear. Here, we sequenced two mtDNA markers (D-loop and Cytb) in 353 samples and one nuclear fragment (MGF) in 304 samples to explore the evolutionary relationships of three hare species in Xinjiang. Gene tree based on mtDNA data clustered all haplotypes into three major clades with strong bootstrap support. One L. capensis clade included the majority of L. capensis haplotypes and one L. yarkandensis clade included the majority of L. yarkandensis haplotypes. However, the third clade contained haplotypes both from L. capensis and L. yarkandensis. This is a new mtDNA lineage. The only one L. timidus haplotype was placed into L. capensis clade. Of the three clades, the L. capensis clade branched first; The L. yarkandensis clade and the new lineage formed sister relationship. On nuclear gene tree, haplotypes of L. capensis and L. yarkandensis mixed together and grouped into two major groups. Haplotype of the only one L. timidus specimen formed a separate clade and branched first. The discrepancy between mtDNA and nuclear DNA analysis suggested that introgressive hybridization might occur among L. capensis, L. yarkandensis and L. timidus in Xinjiang. But evidence from morphology, especially skull characters, is essential to confirm this hypothesis.更多还原