以粒線體DNA及雙限制酶切位點標定法探究臺灣小燕鷗的遺傳多樣性及族群結構

Abstract

近幾十年來，由於棲息地喪失和人為的干擾，包括小燕鷗 (Sternula albifrons) 在內的海鳥族群數量一直在下降。小燕鷗目前被列為臺灣的二級保育類鳥種 (也就是，珍貴稀有物種），是鷗科中唯一在台灣本島有繁殖地的物種。海鳥的族群遺傳多樣性及結構之研究有助於規劃保育策略及相關的物種的保護單位，然而，這種對海鳥，特別是台灣小燕鷗的族群遺傳之相關資訊仍缺乏。因此，本研究利用長度為861-bp的粒線體DNA控制區及使用雙限制酶切位點標定法（ddRAD）篩選出的5113個單核苷酸多態性（SNPs）為資料進行臺灣小燕鷗的族群遺傳研究。本研究於四個已知的小燕鷗繁殖地，分別爲臺灣西側的澎湖和彰化以及臺灣東側的宜蘭和花蓮，採集共59個幼鳥羽毛樣本。由於海鳥族群之分化可能受其不同族群遷徙路線的影響，而經臺灣遷徙的海鳥也可能沿西側與沿東側海岸通過。因此，本研究將樣本亦再依採集地分為兩個地理組（即臺灣西部和東部）以進一步探討臺灣小燕鷗於臺灣西側與東側間的族群是否有差異及評估其在區域上的差異程度。在進行族群遺傳相關的分析前，本研究也利用粒線體細胞色素c氧化酶 I （COI）基因序列確認野外採集樣本的物種及臺灣小燕鷗的亞種名稱定義。本研究意外發現其中一個從宜蘭採集的樣本的COI基因序列與美洲小燕鷗（S. antillarum）極度相似，並在後續分析中確認。美洲小燕鷗與小燕鷗有相近的親緣關係，但因美洲小燕鷗只分佈於北美和南美，故此推斷為該美洲小燕鷗的親鳥可能因遷徙時遭遇不好的氣候條件，迷失原遷徙路缐而意外跟隨小燕鷗族群來到臺灣進行繁殖。本研究中利用粒線體DNA控制區序列進行的族群遺傳分析與ddRAD序列的分析結果相互支持對應。粒線體DNA控制區序列的單型網狀圖（haplotype network）及ddRAD序列的集群分析 (clustering analyses) 均無法根據採集地以區分臺灣西側及東側的樣本為兩群。此外，利用兩種序列進行的AMOVA分析和計算的各群間的遺傳分化指數（pairwise ΦST/FST）顯示臺灣西側與東側的小燕鷗族群處於近乎零的基因分化情形，顯示出西側與東側的小燕鷗族群間有著高度的連通性。另外，本研究也將臺灣小燕鷗的粒線體DNA控制區序列與日本小燕鷗的相同基因序列合併並加以分析以比較臺灣和日本小燕鷗的族群遺傳多樣性的差異及探討兩者之間的族群連通性。單型網狀圖無法區分臺灣和日本的樣本為不同的地理區群集，但圖中三個高頻率出現的共同單倍型（common haplotype）均各別多數出現來自臺灣、日本冲繩及日本本島的樣本。AMOVA和各群間的遺傳分化指數也顯示臺灣、日本冲繩及日本本島的小燕鷗處於中至強度的族群分化情形。小燕鷗在臺灣及日本的族群間具有距離隔離（isolation by distance, IBD）現象，也就是族群間的遺傳距離會與地理距離成顯著正相關。本研究亦針對小燕鷗在臺灣及在日本的族群之族群遺傳分析結果進行比較及討論，並依據研究經驗及結果提出未來小燕鷗在臺灣可供參考的保育策略。

In recent decades, seabird populations including the Little Tern (Sternula albifrons) populations have been declining due to habitat loss and human disturbance. The Little Tern is currently listed as the class II species (rare and valuable species) under the Wildlife Conservation Act Republic of China (Taiwan) and it is the only species in the family Laridae who has breeding records in mainland Taiwan. Examining the genetic diversity and population structure of seabirds can help on managing the conservation units of the particular species. However, such kind of the population genetic analyses on seabirds, particularly the Little Tern in Taiwan remains poorly conducted. In this study, the Little Tern populations in Taiwan were examined based on two different types of data: the mitochondrial (mt) control region (D-loop) DNA sequences (861 bp) and double digest Restriction-site Associated DNA (ddRAD) sequencing data with 5113 single nucleotide polymorphisms (SNPs) generated. The feather samples were collected from 59 chicks of the Little Tern across four known breeding colonies located at western (Penghu and Zhanghua) and eastern (Yilan and Hualien) coasts of Taiwan. Since seabird populations can be shaped by their migratory routes and two main migrations along western and eastern coastal lines in Taiwan might be suspected, the studied populations were further grouped into two geographical groups (i.e., West and East of Taiwan) to determine the degree of population differentiation at regional scale. Before conducting the population genetic analyses, the mitochondrial Cytochrome c oxidase I (COI) gene (or DNA barcoding gene) sequence information of the samples were extracted and analysed to confirm the field identification and the subspecies status of the Little Tern in Taiwan. The phylogenetic analysis using COI sequences revealed that one individual collected from Yilan was highly matched to the Least Tern (Sternula antillarum), a congeneric species of the Little Tern which only breeds in the North and South America. The parents of this individual were believed to get lost due to bad weather while migrating to their breeding grounds in America and accidentally followed the Little Tern colonies to breed in Taiwan. As to the population genetic analyses, the results obtained based on D-loop sequences were consistent with the results obtained based on ddRAD sequencing data. The mtDNA haplotypes constructed based on D-loop sequences and the clustering analyses conducted based on ddRAD sequencing data did not cluster the samples into two geographical groups with respect to the West and East of Taiwan. Furthermore, the AMOVA analyses and pairwise ΦST /FST estimations based on both types of data revealed little to no population differentiation among populations and between regions. The findings of this study suggested a high population connectivity among the breeding colonies in Taiwan. Additionally, the obtained D-loop sequences of the Little Tern from Taiwan were compiled with those from Japan deposited in the NCBI GenBank to compare the genetic diversity and to examine the phylogeographic break that may shape the diversity of the Little Tern populations in eastern Asia. The resulting haplotype network did not clearly separate Taiwanese and Japanese populations but the three most common haplotypes were prevalent for mainland Japan, Okinawa and Taiwan samples, respectively. The Little Tern populations may be frequently connected but with some restrictions on their gene flow that caused moderate to great differentiation among the three populations, which further supported by the AMOVA analyses, pairwise ΦST estimations and the positive yet significant isolation by distance (IBD) pattern. The possible explanations leading to little evidence for genetic structure of the Little Tern populations in Taiwan and moderate to strong structure among Taiwan, Okinawa and mainland Japan populations were discussed. The concerns for the Little Tern protection in Taiwan were also discussed to recommend the conservation strategies of this species in Taiwan.