台灣地方種與栽培品種水稻之遺傳歧異度分析

Abstract

稻米是世界上最重要的糧食作物之一，提供人體熱量來源。稻為禾本科稻屬作物，亞洲野生稻在演化過程中逐漸產生秈、稉二亞種。經由環境篩選與人為馴化，淘汰落粒性、休眠性等影響種植或產量的性狀，馴化成早期地方栽培種。育種家利用地方種以現代育種流程，嚴格地篩選對偶基因育出栽培品種，是以栽培品種的基因歧異度相較地方種和野生稻狹隘。現今全球氣候暖化，乾旱、不穩定的降雨量等極端氣候出現的頻率增加，影響病蟲的生活習性，是以，氣候的變遷會影響植株生長、改變病蟲害發生的時間與種類，而減少稻穀產量。栽培品種的基因庫變異狹小，然而地方種之質資源豐富，含有許多抗生物與非生物對偶基因。以不同來源的種原作為雜交親本雖可改良作物性狀，但是否也能同時增加品種間的遺傳變異，仍有待親緣考量。 本試驗蒐集自台灣、日本、中國大陸等亞洲國家之83地方種、53栽培品種與12野生稻，以75個分子標幟搭配高解析度的毛細管電泳分析基因型，進行此148品系的次族群結構、親緣關係與基因歧異度等分析。有66個(88%)分子標幟之PIC (Polymorphic Information Content)超過0.5，具有高度的訊息含量，其中PIC超過0.9的分子標幟有7個，每個分子標幟偵測到26種以上的對偶基因，可應用於連鎖與圖分析，特殊專一性對偶基因可用於品種鑑定。以模式導向的structure族群結構，大致分成秈稻栽培品種、秈稻地方種、稉稻栽培品種、稉稻地方種與野生稻五個次族群，除了秈稻地方種與野生稻之間無明顯分化外，其餘次族群間之遺傳分化指數Fst > 0.15，皆具適當程度以上的分化，另外，野生稻的基因歧異度為0.74，大於秈型地方種的0.58。主座標分析(principle coordinates analysis)，主要分成秈、稉兩群，栽培品種間較地方種之遺傳距離集中，栽培品種間有相似的遺傳背景，地方種與野生稻的分布較鬆散，相似性低而變異性較高。以距離導向的鄰接法(neighbor-joining method)親緣圖分析，主要分秈、稉兩大群，在秈稉之下品系間存在明顯遺傳距離，可再細分成12群，栽培品種與地方種分別成群，栽培品種間親緣集中，與地方種間有著一定距離以上的親緣關係，與主座標分析結果類似。以structure軟體模擬出的秈、稉背景劃分地方種的亞種屬性，探討現今稻米在馴化過程以及台灣育種歧異度變遷。秈稻族群平均每個分子標幟位點可偵測到8.31種對偶基因，PIC數值0.58，基因歧異度為0.61；稉稻族群每個位點偵測到7.05種對偶基因，PIC數值為0.55，基因歧異度為0.59，由此可見秈稻的歧異度大於稉稻。秈稉兩族群的Fst為0.31，高於族群高度分化數值界線之0.25，秈、稉此兩族群的基因體大不相同。不論秈、稉稻，地方種的PIC與基因歧異度皆大於栽培品種，顯示栽培品種在育種過程中歧異度流失。以民國70年台灣水稻育種政策由產量轉為米質為界，劃分台灣栽培品種的育種之早、晚期，屬於早期的台灣秈稻栽培品種有7個，晚期有12個，其基因歧異度分別為0.52與0.51，兩期的Fst小於0.05，顯示在育種過程中，台灣秈稻栽培品種育種親本雖然相似但維持著歧異度；屬於早期的台灣稉稻栽培品種只有3個，晚期有25個，基因歧異度分別為0.29與0.42，兩期的Fst小於0.38，雖然晚期的稉稻栽培品種的歧異度高於早期，但不排除早期稉稻栽培品種的基因歧異度較晚期低是受限於品種數量的緣故。本試驗分析稻品系間的親緣關係、次族群結構與基因歧異度，合理且有效地評估遺傳資源，可針對不同的育種需求挑選適當的親本改良栽培品種增加遺傳變異。

Rice is one of the most important cereal crops in the world which supplies the 23% calorie for human. Asian cultivated species, Oryza sativa, is classified in the family Poaceae and genus Oryza and includes two sub-species, indica and japonica. The distinct characteristics of indica and japonica could be observed, which were gradually diversified during the evolution of rice under natural selection and domestication. Some traits related to the yield or growth habitat of the rice, such as shattering and dormancy, would be selected against. Wild rice would be domesticated into landraces through these processes. Landraces were selected strictly by breeders in modern breeding process in order to establish cultivars for specific breeding goals. As the result, the genetic diversity of the modern cultivars is relatively smaller than those of landrace and wild rice. Nowadays, climate changes because of global warming increase the frequencies of extreme weather, such as drought and floods. This would also lead to the change of pathogens’ and insects’ living habit, which influence the growth and yield of modern cultivars because of their narrow gene pool. Landraces enriched genetic resources could possess resistance and tolerance to abiotic and biotic stresses. Landraces are useful germplasm to improve genetic diversity of modern cultivars. It would be better to account phylogenetic relationship before employ of landraces to breeding program. We used 75 markers subjected by high resolution capillary electrophoresis to genotype 83 landraces, 53 modern cultivars, and 12 wild rice which were collected from Taiwan, Japan, China and other Asian countries. Sixty six (88%) markers were displayed high polymorphism with PIC (Polymorphic Information Content) larger than 0.5, and seven markers were detected at least 26 alleles with estimated PIC larger than 0.9. All the highly polymorphic markers uncovered in this study could be applied to linkage analysis, and the alleles specific to varieties could be used for variety identification. In order to infer population structure, software structure conducted by model-based clustering method could be separated 148 accessions into 5 subpopulations, including indica cultivars, indica landraces, japonica cultivars, japonica landraces, and wild rice. There were great genetic differentiation among subpopulations (Fst>0.15), except indica landraces and wild rice. In spite of the fact that mentioned above, the gene diversity of wild rice (0.74) is still higher than of indica landraces (0.58). According to PCoA (Principle Coordinate Analysis), all accessions were grouped to two groups corresponding to indica and japonica. The genetic distances were closer among cultivars than among landraces, indicating that cultivars had more similar genetic background. The distribution of landraces and wild rice were sparse, indicating little similarity and high variation. The 148 accessions were separated into two distinct groups, indica and japonica, by Neighbor-joining phylogenetic tree, which is a distance-based clustering method. There are variations between accessions in both groups, which could be subdivided into 12 groups at least. Furthermore, modern cultivars and landraces can be subdivided into different groups which could imply a certain genetic distance. The average alleles number of every marker, PIC and genetic diversity among indica, was 8.31, 0.58, 0.61, respectively. Also in japonicas there were 7.05, 0.55, and 0.59, respec of indica tively. The genetic diversity of indica was expected to higher than of japonica. The Fst between indica and japonica was 0.31, which was higher than the threshold 0.25, implying the genomes of indica are highly differentiated from japonica. In the year of 1981, the government of Taiwan changed the breeding objective from yield to rice quality, which separated the cultivars in Taiwan into two stages, early stage and late stage. The genetic diversity of 7 indica cultivars in early stage and 12 in late stage were 0.52 and 0.51, respectively. Fst of indica cultivars between two stages was less 0.05, indicating no difference genetic diversity of parents used in indica breeding programs. On the other hand, the genetic diversity of japonica cultivars in early stage (n=3) and late stage (n=25) were 0.29 and 0.42, respectively. Fst of japonica cultivars was 0.38 between early and late stage. It indicated that there is a variation among the breeding parents of japonica cultivars during the breeding process. However, the large difference in sample size leading to bias estimation of genetic diversity could not be ruled out. This study suitably estimates the rice genetic resources by analyzing the phylogenetic relationship, subpopulation structure, and genetic diversity among 148 accessions. It could provide suggestions to improve cultivars in different requests and increase the genetic variation simultaneously.