行政院國家科學委員會專題研究計畫 期中進度報告
植物相剋物質作為天然除草劑的研究：台灣杜鵑之植物毒物
質之萃取、純化、鑑定及其植物毒性(1/3)
計畫類別： 個別型計畫 計畫編號： NSC91-2621-B-110-001-執行期間： 91 年 08 月 01 日至 92 年 07 月 31 日 執行單位： 國立中山大學生物科學系(所) 計畫主持人： 周昌弘 計畫參與人員： 陳英美 報告類型： 精簡報告 處理方式： 本計畫可公開查詢中
華
民
國 92 年 5 月 13 日
1
行政院國家科學委員會補助專題研究計畫期中
報告
※※※※※※※※※※※※※※※※※※※※※※※
※※※
※
※
※
※
※
※
※
※
※
※
※
※
※※※※※※※※※※※※※※※※※※※※※※※
※※※
計畫類別：█個別型計畫
□整合型計畫
植物相剋物質做為天然除草劑之研究：
血桐植物台灣杜鵑之植物相剋物質之毒性分析、分離、純化及
鑑定（21/3）
AllelopathicAllelopathic
substances as natur ally occur r ing
her bicide：allelopathic activity and isolation, pur ification and
identification of allelopathic substances in
1
計畫編號：NSC 8991－2311621－B－110－01801
執行期間： 8991 年 08 月 01 日至 902 年 07 月 31
日
計畫主持人：周昌弘
共同主持人：
計畫參與人員：曾梅慧
李宗徽、
郭悅雄陳英美
本成果報告包括以下應繳交之附件：
□赴國外出差或研習心得報告一份
□赴大陸地區出差或研習心得報告一份
□出席國際學術會議心得報告及發表之論文各一份
□國際合作研究計畫國外研究報告書一份
執行單位：國立中山大學生物科學系
中
華
民
國
九十二年
五月
七日
1
行政院國家科學委員會專題研究計畫期中報告
植物相剋物質作為天然除草劑之研究：
血桐植物台灣杜鵑之植物相剋物質之毒性分析、分離、純化及鑑定
（21/3）
Allelopathic substances as natur ally occur r ing her bicide：allelopathic
activity and isolation, pur ification and identification of allelopathic
substances in Macaranga tanariusRhododendron formosanum
（1/3）
計畫編號：NSC 91－2621－B－110－001NSC 89-2311-B-110-018 執行期限：8991 年 8 月 1 日至 902 年 7 月 31 日 計畫主持人：周昌弘 國立屏東科技大學熱帶農 業所， 國 立 中 山 大 學 生 物 科 學 系，中央研究院植物所 計畫參與人員：曾梅慧李宗徽 國立台灣大學植物所博 士班研究生 郭悅雄陳英美 國立台灣大學化學 系私立嘉南藥理科技大學醫藥化學系國立中山大學生物科學系
Pr eliminar y Repor t on
Allelopathic
Pr enylflavanones fr om
the Potential of
F allen Leaves of
Macaranga tanariusRhododendron formosanum
Chang-Hung Chou 1,2,2, Mei-HuimsTzong-HueiTsengLee32,Yih-Ming Chen3,and
1
1
Graduate Institute of Tropical Agricalture, National Pingtung University of Science
and Technology, Pingtung , Taiwan 912Department of Biological Sciences, National
Sun Yat-sen University, Kaoshiung, Taiwan 804
2
Institute of Botany, Academia Sinica, Taipei, Taiwan 115
32
Department of Biological Science, National Sun Yat-sen University, Kaoshiung,
Taiwan 804Department of Botany, National Taiwan University, Taipei, Taiwan 106
43
Department of Applied Chemistry, National Taiwan UniversityChia Nan University
一、摘要
Abstr act
A 台灣杜鵑為台灣特有種植物，海拔 分布較低，整體族群分布於海拔 600~2400 專公尺的山地，不過主要集中在 1000~1500 公尺處，喜溼度高的環境，長於稜脊或陡坡 處行成小面積純林為本省霧林帶中優勢而 常見的林種。過去，據本研究室對鴛鴦湖保 護區枝之台灣杜鵑林下植被的覆蓋度和植 種歧異度，以及台灣杜鵑葉的幹流水可抑制 種子發芽的初步生物試驗結果，推測台灣杜 鵑含有的次級代謝物具有高的植物相剋活 性。因此擬定進行三年的研究計劃畫已以萃 取、分離、純化和鑑定花、葉、莖及根部等 所含有的植物毒性物質，並進一步評估以此 等植物毒性物質作為除草劑的可行性。methanol extract of the fallen leaves of Macaranga tanarius (Euphorbiaceae) was suspended in H2O and then partitioned with
ethyl acetate. The ethyl acetate solubles were fractionated to yield two new prenylflavanones, tanariflavanones A (1) and B (2), and the known (-) –nymphaeol-C (3). 11 Their structures were elucidated primarily by NMR, circular dichroism and mass spectroscopic methods. This paper describes the structural elucidation of 1 and 2, and their inhibitory effect on radicle growth of germinating lettuce seeds at 200 ppm.
Intr oduction
The genus Macaranga is one of the largest genera of the Euphorbiaceae, with approximately 300 species. 1 Previously reported components from the species of this genus include diterpenoids, triterpenoids, steroids 2,3and hydrolyzable tannins4 from M. tanarius, a prenyl stilbene, vedelianin 5 and a
geranyl flavonol 6 from M. vedeliana, antibacterial prenylated flavanones 7 from M. pleiostemona, chromenoflavones 8 from M. indica, and cytotoxic geranyl stilbenes 9 from M. schweinfurthii. M. tanarius is a common tropical tree distributed from southern Asia to northern Australia. 10 The allelopathic potential of the fallen leaves of M. tanarius was indicated from field observations and bioassays, prompting us to study its chemical components.
Exper iment
二、目的植物相剋作用（allelopathy）是 Plant
Mater ial. The fallen leaves of M. tanarius were collected from the campus of National Taiwan University, Taiwan, in 1998. A voucher specimen (voucher no. 01566) has been deposited at the Herbarium of the Department of Mathematics and Science Education, Taipei Municipal Teachers College, Taipei, Taiwan.植物釋放的次級代 謝產物（secondary metabolite）已以抑制本 身或其鄰近的植物之種子發芽，根生長，植 株的發育及開花結果，此類植物相剋物質或 稱植物毒物質（phytotoxin），其中部分植物 毒物質在濃度較低時會促進植物的生長或 成為生長激素（growth hormone），故這些 對於生物具有活性的物質稱為生物相生相 剋物質（allelochemicals）。利用生物間的相 生相剋作用來作為生物防治的治本工作。使 用這些來自植物的相剋物質以抑制雜草的 生 長 視 為 天 然 植 物 殺 草 劑 （ naturally occurring herbicide）。 在台灣許多農作物因該作物的分泌 植物毒性或作物代謝物殘留於土壤中經分 解後產生之有機酸而抑制後繼作物的生長 而導致生產力降低。。有此可看出植物相剋 作用在植物生態和農業生態上扮演著重要 的角色。自工業革命以後，農田趨於用機械 耕作和大量施灑化學肥料的結果導致土壤 1
1 5 理化性狀改變及農藥的殘餘使農業生態失 去平衡，並使地下水資源的受到污染。因此 鼓勵農民盡量少用農藥及化學肥料外，若能 將植物的相剋物質取代合成的除草劑或農 藥，對當今及未來的農業將有很大的貢獻。 將使整個生態系不在因過度使用化學農藥 而失去平衡。 全世界杜鵑屬植物大約 800 餘種，台 灣所產則約有 16 種。台灣杜鵑為台灣特 有，分布於全島中海拔霧林帶稜線、上坡段 的地形位置，常生長於檜木林下，同時為本 省霧林帶中優勢而常見的植種。然而目前沒 有針對其所含的化學成分加以探討的研究 文獻。據本研究室近年來的野外觀察及其幹 流水的生物活性試驗的初步結果顯示，台灣 杜鵑具有高的植物相剋潛能，然未深入探討 其所具有的天然活性物質及評估該活性物 質被利用為天然除草劑的可行性，而本人計 畫主持人過去在植物相剋作用的研究已有 相當的基礎及成果，若配合國家農業生物技 術枝開發研究政策及響應環保的重要性等 投入此天然植物除草劑的研究，則部分成果 可直接應用於農業生產上，有助於經濟發展 和環境保護。 三、研究方法 （一）採樣
Extraction and Isolation. The
dried fallen leaves of M. tanarius were ground to powder, and 4 kg of the powder was extracted with MeOH (100 L) at room temperature (7 days x 2). The extract was evaporated in vacuo to yield a residue which was suspended in H2O (1 L),
and this was partitioned with ethyl acetate (1L x 3). The combined ethyl acetate layer afforded a black syrup
(250 g) that was subsequently chromatographed over silica gel with a hexane/EtOAc gradient solvent system. Crude compounds 1, 2, and 3 were all eluted with 30 % EtOAc in hexane. Further purification by HPLC [Merk LichroCART 250-10 Cat. 1.50179 Lichrosorb Si 60 (7μm)] gave 1 (30 mg), 2 (7 mg), and 3 (500 mg) using 30 % Et OAc/hexane. 本研究的樣區除了鴛鴦湖保護區之 霧林帶外，還有中部山區如阿里山及南部中 央山脈的大武山南麓等多個樣區，分別採集 台灣杜鵑的花、葉、莖及根，攜回實驗室後 置於 35℃-40℃的烘箱中約經一個禮拜使之 烘乾，然後磨成粉保存於-20℃的冰箱中， 以備後續實驗之使用。 （二）植物毒物質之萃取、分離及純化 取 200 克台灣杜鵑乾粉，參考並修改 Lee at al.（2000）的方法，以 80﹪MeOH 萃 取，萃出物續以 Hexane 分層，濃縮下層之 含水和甲醇的部分，所得之粘稠濃縮液以 ethyl acetate 及水進行二次分層，各層濃縮 物進行植物毒生物分析，其中具有活性的濃 縮物，依序以填充 PVPP（Sigma U.S.A）的 層析管柱進行分離後，再經由薄層色層分析 （TLC）及紙色層分析（PC）分離後，， 最後使用高效液相層析管柱（Hyperprep HS C18,250 × 10 mm, Keystone Scientific, U.S.A）純化以獲得純質。 （三）植物毒物之生物分析 純 化 所 得 之 化 合 物 依 Chou and Muller（1972）及 Chou（1999b）的方法進 行種子發芽即幼根生長分析，以瞭解每種化 合物的毒性強度進而歸納出其抑制強度與 毒性大小的相關性，作為往後探討其殺草機 制的參考。所使用的生物分析採用種子為萵 苣（Lactuca sativa L.l.）的種子以測其種子 發 芽 及 幼 根 生 長 及 採 用 和 巴 拉 草 2
1 5 （Brachiaria mutica）的匍匐莖以測其側根 生長的比較。 滲透壓濃度的測定：分別配製台灣杜 鵑落葉水溶萃取液及標準試劑麥芽糖水溶 液為 0.1、0.2、0.3、0.4、0.5、1.0、2.0、3.0、 4.0、5.0 等的濃度百分比，使用 OSMOMAT 030（GONOTEC， Germany）測量其滲透 壓。 Spectroscopic Analysis. IR spectra were recorded on a
Perkin-Elmer 983G
Spectrophotometer. 1
H and 13
C NMR spectra were run on a Varian Unity Plus 400 spectrometer. EIMS and specific rotations were taken on a JEOL JMS-HX 300 mass spectrometer and a JASCO DIP -1000
digital polarimeter,
respectively. Extracts were chromatographed on silica gel (MERK 70-230, 230-400 mesh, ASTM).
Bioassays. A 200 ppm solution of
each pure isolated compound in MeOH was prepared for bioassay. The solutions were spread onto the silica TLC sheets (1 cm x 5 cm), allowing the MeOH to evaporate completely in a laminar flow hood. Prior to bioassay, the TLC sheets were placed in Petri dishes and moistened with distilled H2O, then
surrounded by wet sponge without contact. Lettuce (Lactuca sativa L. var. longifolia Lam.) seeds were imbibed in distilled H2O for 2 hr, then placed on
the TLC sheets containing the pure
compound or on untreated control sheets. The Petri dishes were sealed with M parafilm (American National Can), then placed in an incubator at 25 ℃ for 48 hours in dark.
四、結果
Results
The molecular formula (C30H36O7) of 1,
showing one more oxygen atom than 3, was obtained from its HREIMS and 13C NMR data. The IR spectrum of 1 indicated the presence of hydroxyl (3422 cm-1), conjugated carbonyl (1647 cm-1) and aromatic (1610 and 1500 cm-1) groups. Its UV spectrum exhibited maxima at 231 (sh) and 293 nm. The 1H NMR spectrum of 1 exhibited signals for a phenolic OH atδ12.35 (s, OH-5), which was a hydrogen strongly bonded to the 4-carbonyl group. An ABX system at δ5.48 (1H, dd, J = 12.8, 2.8 Hz), 2.73 (1H, dd, J = 17.2, 2.8 Hz) and 3.12 (1H, dd, J = 17.2, 12.8 Hz), revealed diagnostic for H-2 and H-3 of a flavanone nucleus.11 The appearance of three singlet methyl groups attached on olefin (δ 1.57, 1.64, 1.74), four multiplet methylene protons ( δ 2.06), two benzylic methylene protons (δ3.42, br d, J = 6.8 Hz ), two triplet vinyl protons (δ 5.02, 5.17) indicated the presence of a geranyl group. 11 The presence of a 2,2-dimethylchromane moiety with a secondary hydroxy group was revealed by signals atδ1.31 and 1.35 (3H each, s), 2.64 (1H, dd, J = 16.8, 5.2 Hz, Ha-1´´´), 2.87 (1H,
dd, J = 16.8, 5.2 Hz, Hb-1´´´), and 3.83 (1 H, t,
J = 5.2 Hz, H-2´´´). The geranyl group was assigned at C-2´ due to H2-1´´ and H-2 having
NOESY correlation. There was also a significant MS fragment at m/z 384 [M+ -124]. 11 The absence of a bathochromic shift in the UV spectrum of 1, after the addition of aluminum chloride, indicated a substituent at
1
5
C-6. 7 The methylene protons at δ2.64 and 2.87 were assigned to a benzylic methylene. The coupling constant between H-1´´´ and H-2´´´ clearly indicated that the hydroxyl group at C-2´´´ occupied an axial position. Therefore, 2-hydroxy-3,3-dimethylchromane must fuse on ring A. The ortho-coupled signals atδ6.80 and 6.96 (J = 8.4 Hz, H-5´, -6´) as well as H-6´ and H-3 having NOESY correlation suggested that C-1´, 2´, 3´, and 4´ positions in ring B were substituted. Therefore, two phenolic hydroxys (δ 5.56 and 5.89, exchangeable) were assigned to C-3´ and C-4´. Thus, compound 1 was deduced to be 2´-geranyl- 5,3´, 4´ -trihydroxy-6,7-(2,2-dimethyl-3-hydroxychro mano)flavanone. The absolute stereochemistry at C-2 of 1 was established as S based upon a positive extremum at 344 nm (θ= +13590) and a negative extremum at 291 nm (θ= -34010) in its circular dichroism spectrum.12
Compound 2 had the molecular formula C30H34O6 on the basis of HREIMS
and 13C NMR data. The IR spectrum of 2 showed bands attributed to hydroxyl (3374 cm-1), conjugated carbonyl (1641 cm-1) and aromatic (1606 and 1449 cm-1) groups. Its UV spectrum exhibited maxima at 227 (sh), 274, and 285 nm. Comparison of 13C NMR data of 2 with those of known (-)-nymphaeol-C (3) suggested that 2 possessed the same skeletal structure as 3. Closer inspection of 1H,13C, and DEPT NMR spectra of 2 revealed the presence of three methyls(δH 1.41, 1.55, and 1.65; δC 17.6,
25.6, and 26.1 ), two methylenes (δC 40.7
and 22.7 ), one oxygenated quaternary carbon (δC 79.0 ), and two double bonds (δC 118.9
and 130.9; 123.7 and 132.1 ). In the 1H NMR spectrum of 2, vicinally-coupled protons atδ 5.65 and 6.58 (J = 10.0 Hz) indicated that an oxygen atom at C-3´ must have cyclized onto C-3´´to form a pyran ring as shown in the structure 2. This arrangement was supported
by the observation of a fragment at m/z 407 [M-CH2CH2CH=C(Me)2]+ in the EIMS of 2. 13
The 1H NMR spectrum [δ1.74, 1.80 (each 3H, s), 3.32 (2H, br d, J = 5.8 Hz, H-1´´´), 5.23 (H, br t, H-2´´´)] further showed the presence of one 1,1-dimethylallyl group. This group was placed at C-6, as 2 also showed no bathchromic shift on the addition of aluminum chloride. Compound 2 had the same absolute configuration (2 S) as 1 and 3 on the basis of the positive Cotton effect at 340 nm ( θ = +12390) and the negative Cotton effect at 287 nm (θ= -16570). The structure of 2 was thus assigned as shown.由
種子生物試驗的結果顯示，台灣杜鵑和桃花
心木對巴拉草有明顯的根長抑制作用（附圖
1 一）。於是，把台灣杜鵑的萃取液經由
Bbutanol 、Ddichloromethane、及 Ddiethly ether 分離四層後，再經由 PVPP 填充管柱 分離 Ddichloromethane 層可以收集濃縮後
得到 11 個樣品。把這些樣品再一一的作萵
苣種子發芽及幼根生長的生物試驗，其結果 唯 有 四 個 樣 品 ： CYM-1-21-1 、 CYM-1-24-3 、 CYM-1-25-3 、 CYM-1-25-4 等有顯著的地對根生長有抑制作用（附圖2 二）。 這四個樣品在經由 HPLC 的分離分 別可得：CYM-1-21-1 有兩個明顯的波峰， CYM-1-24-3 有 兩 個 明 顯 的 波 峰 ， CYM-1-25-3 有 一 個 顯 著 的 波 峰 ， CYM-1-25-4 有一個明顯的波峰（圖三）（附 圖 3）。 這些經由 HPLC 分離所得的波峰將 再正進一步的作生物試驗分析，及純化及結 構式之鑑定。。 五、討論 台灣杜鵑對巴拉草的明顯抑制可 3
1 5 由其對根的抑制百分比看出。在不同濃度施 加下，0.5﹪~5﹪皆具抑制效果，且抑制百 分比皆在 50﹪以上。高濃度 5﹪濃度下其抑 制效果為 100﹪的抑制，而低濃度的 0.5﹪ 濃度下也有 60﹪的抑制效果。由此項數據 得知確切符合野外觀察台灣杜鵑林相的抑 制現象。 以 2﹪HOAc 為展佈劑做 PC 紙色層 分析，上的其 Rf 值各為 CYM-1-21-1； 0.81-0.89 /0.92-0.99；CYM-1-24-3：0.71- 0.79/0.89-96 ； CYM-1-25-3 ： 0.91-0.98 ； CYM-1-25-4 ： 0.74-0.79/0.91-0.97 ， 而 比 較 Dd ichloro-methane 層萃取出來的四個可能含 有抑制化合 物 樣 品 中 以 CYM-1-21-1 Rf 值 為 0.92-0.99 具 最顯著的抑制效果，而其 Rf 植也是另 外三者 所顯示的範圍。據此推測有抑制作用相 關的 化合物為 flavonoid 類的二次代謝 物。 測量測透壓的結果顯示出台灣杜鵑葉 的萃取 液滲透壓不大，表示其抑制生長的原因可排 除掉滲透壓這點變數（圖四）。 本研究接下來刻進行各種將一一分析 並以鑑定出其具 有抑制效果的化合物。對於開發台灣杜鵑做 為天然的除草劑，，瞭解其最佳作用濃度是 相當重要的必 然的，。經過初步的實驗所得，在 0.4 ﹪水溶萃取液濃度下 下其 抑制效果達到最大，往後若提高濃度並 不會有 顯著 增強抑制的效果，有此可知該濃度為 最佳作 用的濃度。 Tanariflavanone A (1)：greenish oil;[α]24.6 D +26.8。(c 0.6, CHCl3); UV (MeOH) λmax (logε) 231 sh (4.34), 293 (4.29) nm; IR (KBr)νmax 3422, 1647, 1610, 1500, 1457 cm-1 ; 1 H NMR (400 MHz, CDCl3)δ1.31, 1.35, 1.57, 1.64, 1.74 (each 3H , s), 2.06 (4H, m, H-5´´,-6´´), 2.64 (1H, dd, J = 16.8, 5.2 Hz, Ha-1´´´), 2.73 (1H, dd, J = 17.2, 2.8 Hz, Heq-3), 2.87 (1H, dd, J = 16.8, 5.2 Hz, Hb-1´´´), 3.12 (1H, dd, J = 17.2, 12.8 Hz, Hax-3), 3.42 (2H, br d, J = 6.8 Hz, H-1´´), 3.83 (1H, t, J = 5.2 Hz, H -2´´´), 5.02 (1H, br t, J = 6.8 Hz, H-7´´), 5.17 (1H, br t, J = 6.6 Hz, H -2´´), 5.48 (1H, dd, J = 12 .8, 2.8 Hz, H-2), 5.94 (1H, s, H -8), 6.80, 6.96 (each 1H, d, J = 8.4 Hz, H -5´, -6´), 5.56, 5.89, 12.35 (each 1H, s, OH); 13 C NMR (100 MHz, CDCl3) δ196.4 (s, C-4), 161.9 (s, C-9), 161.5 (s, C-7), 160.8 (s, C-5), 144.8 (s, C-4´), 142.5 (s, C -3´), 138.9 (s, C-3´´), 132.2 (s, C-8´´), 128.3 (s, C-1´), 126.4 (s, C-2´), 123.7 (d, C -7´´), 121.3 (d, C-2´´), 118.9 (d, C-6´), 112.9 (d, C-5´), 102.7 (s, C-6), 99.9 (s, C-10), 96.2 (d, C-8), 78.6 (s, C -3´´´), 76.9 (d, C -2), 68.8 (d, C-2´´´), 42.4 (t, C -3), 39.6 (t, C -5´´), 26.3 (t, C-6´´), 25.7 (q, C -10´´), 25.4 (t, C-1´´), 25.0 (q, C-4´´´), 24.9 (t, C-1´´´), 22.0 (q, C-5´´´), 17.7 (q, C -9´´), 16.2 (q, C-4´´) ; EIMS m/z 508 [M]+ (33), 490 (10), 423 (50), 384 (29), 237 (100); HREIMS m/z
1
5
508.2446 ( calcd for C30H36O7, 508.2451).
Tanariflavanone B (2)：brownish oil;
[α]24.6
D +28.2。(c 0.5, CHCl3); IR νmax
3374, 1641, 1606, 1449, 1156 cm-1
; UV (MeOH) λmax (log ε ) 227 (4.83), 274
(4.47), 285 (4.49) nm; 1 H-NMR (400 MHz, CDCl3) δ1.41, 1.55, 1.65, 1.74, 1.80 (each 3H, s), 2.08 (4H, m, H -5´´, -6´´), 2.73 (1H, dd, J = 17.2, 2.8 Hz, Heq-3), 3.11 (1H, dd, J = 17.2, 13.2 Hz, Hax-3), 3.32 (2H, br d, J = 5.8 Hz, H -1´´´), 5.07 (H, br t, J = 5.7, H-7´´), 5.23 (H, br t, J = 5.8 Hz, H-2´´´), 5.48 (1H, dd, J = 13.2, 2.8 Hz, H-2), 5.65 (1H, d, J = 10.0 Hz , H -2´´), 5.97 (1H, s, H -8), 6.58 (1H, d, J = 10.0 Hz, H-1´´), 6.81 (1H, d, J = 8.4 Hz, H-5´), 6.89 (1H, d, J = 8.4 Hz, H-6´); 5.66, 6.32, 12.38 (each 1H, s, OH); 13 C NMR (100 MHz, CDCl3) δ 196.3 (s, C-4), 163.7 (s, C -7), 161.3 (s, C-9), 161.0 (s, C-5), 145.1 (s, C-4´), 139.7 (s, C-3´), 135.3 (s, C -3´´´), 132.1 (s, C-8´´), 130.9 (d, C-2´´), 124.8 (s, C-1´), 123.7 (d, C -7´´), 121.5 (d, C -2´´´), 118.9 (d, C-1´´), 118.9 (d, C -6´), 118.8 (s, C-2´), 114.5 (d, C-5´), 107.2 (s, C-6), 102.9 (s, C-10), 95.5 (d, C-8), 79.0 (s, C-3´´), 76.0 (d, C -2), 42.5 (t, C-3), 40.7 (t, C-5´´), 26.1 (q, C-4´´), 25.8 (q, C-4´´´), 25.6 (q, C-10´´), 22.7 (t, C-6´´), 21.1 (t, C-1´´´), 17.8 (q, C-5´´´), 17.6 (q, C-9´´) ; EIMS m/z 490 [M]+ (17), 407 (100), 219 (26). HREIMS m/z 490.2374 (Calcd for C30H34O6, 490.2346).
Phytotoxicity
of
Prenylflavanones
The tentative phytotoxic
activity of 1 and 2 was evaluated by determining their effect on the radicle elongation of germinating lettuce. At 200 ppm, 2 inhibited radicle length of lettuce up to 30% as compared to the distilled H2O control, while 1 caused
11% growth inhibition. These results suggest that compounds 1 and 2 play an allelopathic role in M. tanarius.
Acknowledgments
This work was supported by grants (NSC-88-2311-B-001-NSC-89-2311-B-001-026) to C. H. Chou awarded by the National Science Council of Republic of China.
Refer ence and Notes六五
、參考文獻Borner, H., 1960Webster, G. Ann. Missouri Bot. Garden 1994, 81, 33-144., Liberation of organic substances from higher plants and their role in soil richness problem. Bot.Rev（Lancaster）26: 393-424.
Hui, W. H.; Ng, K. K.; Fukamiya, N.; Koreeda, M.; Koreeda, M.; Nakanishi, KChou, C .H.,1986, The role of allelopathy in subtropical aproecosystems in Taiwan. In: The Science of Allelopathy, Putnam, A. L. and C. S. Tang（eds.）, John Wiley and Sons, New York, Chichester, Brisbane, Toronto, Singapore. pp. 57-73.
Chou, C. H. 1999a. Allelopathy in biodiversity and sustainable agriculture. Crit. Rev. in Plant Sci. 18: 609-636.
1
5
Hui, W. H.; Li, M. M.; Ng, K. K. Phytochemistry 1975, 14, 816-817Chou, C. H., S. G. Chang, C. M. Cheng, Y. C. Wang, F. H. Hsu and W. H. Den,1989, The selective allelopathic interaction of a pasture-forest intercropping in Taiwan Ⅱ. Interaction between kikuyu grass and three hardwood plants. Plant and Soil .101: 207-215.
Lin, J. H.; Nonaka, G. I.; Nishioka, I. Chem. Pharm. Bull. 1990, 38, 1218-1223.Chou, C. H. and Y. T. Chung, 1974, The allelopathic potential of Miscanthus floridulus. Bot. Bull. Academia Sinica 15: 14-27.
Thoison, O.; Hnawia, E.; Guéritte-Voegelein, F.; Sévenet, T. Phytochemistry 1992, 31, 1439-1442.Chou, C. H. and Y. L. Kuo, 1986, Allelopathic research in subtropical vegetation in Taiwan Ⅲ . Allelopathic exclusion of understory species by Leucaena lencocephala. J. Chem. Ecol.
12: 1431-1448.
Hnawia, E.; Thoison, O.; Guéritte-Voegelein, F.; Bourret, D.; Sévenet, T.
Phytochemistry 1990, 29,
2367-2368.Chou, C. H. and C. H. Muller, 1972, Allelopathic mechanism of Arctostaphylos glandulosa var. zacaensis. Am. Midl. Nat. 88: 324-347.
Chou, C. H and L. L. Lee, 1992, Allelopathic substances and interactions of Delonix regix（Boj. Raf.） j. Chem. Ecol. 18: 2285-2303.
Schütz, B. A.; Wright, A. D.; Rali, T.; Sticher,
O. Phytochemistry 1995, 40,
1273-1277.Chou, C. H. and Y. F. Lee, 1991, Allelopathic dominance of Miscanthus transmorrisonensis in an alpine grassland community in Taiwan. J.
Chem. Ecol. 17: 2267-2281.
Sultana, S.; Ilyas, M. Phytochemistry 1986, 25, 953-954.Chou, C. H. and G. R. Waller （eds.）, 1989, Phytochemical Ecology. Institute of Botany, Academia Sinica Monograph Series No.9, Taipei.
Beutler, J. A.; Shoemaker, R. H.; Johnson, T.; Boyd, M. R. J. Nat. Prod. 1998, 61, 1509-1512.
Hsieh, C. F. “Flora of Taiwan”, Vol. III, Epoch Publish Co., Ltd., Taipei, 1977, p. 479.Chou, C. H and L. L. Lee, 1992, Allelopathic substances and interactions of Delonix regix（Boj. Raf.） j. Chem. Ecol. 18:2285-2303.
Yakushijin, K.; Shibayama, K.; Murata, H.; Furukawa, H. Heterocycles 1980, 14, 397-402.Chou, C. H., C. Y. Fu, S. Y. Li and Y. F. Wang, 1998, Allelopathic potential of Acacia confusa and related species in Taiwan. J. Chem. Ecol. 24（12）: 2131-2150.
Gaffield, W. Tetrahedron 1970, 26, 4093-4108.Chou, C. H. 1999a. Allelopathy in biodiversity and sustainable agriculture. Crit. Rev. in Plant Sci. 18:609-636.
Gliessman, S. R., 1991, Agroecology: Researching the Ecological Basis for Sustainable Agriculture. Springer-Verlag, N. Y. Berlin, Heidelberg, London, Paris, Tokyo, Hong Kong.
Huang, L.; Fullas, F.; McGivney, R. J.; Brown, D. M.; Wani, M. C.; Wall, M. E. J. Nat. Prod. 1996, 59, 290-292.Chou, C. H. 1999b. Methodologies for allelopathic research: from fields to laboratory. In F. A. Macias （ed.）Advances in Allelopathy . Vol.1. A science for the Future International Allelopathy Society, Servicio De Pullicaciones Universidad de
1
5
Cadiz, pp. 3-24.
Lee. T. H., Giu, F., G. R. Waller, Chou, C. H. 2000. Three new flavonol galloyglycoside from leaves of Acacia confusa. J. Nat. Prod. 63: 710-712.
Muller, C. H., 1966, The role of chemical inhibition（allelopathy）in vegetationed composition. Bull. Torrey Bot. Club. 93: 332-351.
Rice, E. L., 1984, Allelopathy. 2nd Ed. Academic Press, New York and London. Rice, E. L., 1995, Biological Control of
Weeds and Plant Diseases: Advances in Applied Allelopathy. University of Oklahoma Press: Norman and London.
Rizvi, S. J. and V. Rizvi （ eds. ） , 1992, Allelopathicy: Basic and Applied Aspects. Chapman& Hall, London.
Tseng, M. H., Chou, Y. M. and Kuo, Y. H. 2001. Allelopathic prenylflavonones from the fallen leaves of Macaranga tanarius. J. Nat. Prod.64（6）: 827-828.
0 20 40 60 80 100 120 0.71-0.79 0.81-0.89 0.89-0.96 0.91-0.97 0.92-0.99 以2%醋酸展佈後之紙色層圖上斑點之Rf值 抑制百分比
種子發芽
幼根生長
圖一：台灣杜鵑落葉之水溶萃取液對巴 拉草根生長的影響。 圖二：台灣杜鵑水容溶萃取液經分離純化後之天然物對萵苣種子發芽及 根生長之抑制百分比。純化之天然誤物以 2﹪HOACc 經紙色層 展圖二：台灣杜鵑水容萃取液經分離純化後之天然物對萵苣發芽 及根生長之抑制百分比。純化之天然誤以 2﹪HOAC 經紙色層展 附圖一 0 20 40 60 80 100 120 0.10 0.20 0.30 0.40 0.50 1.00 2.00 3.00 4.00 5.00
葉水溶萃取液,%濃度
根長抑制百分比
1 5 5 附圖二
CYM-1-24-3種子試驗
0 20 40 60 80 100 120 Control Rf=0.71-0.79 Control Rf=0.89-0.96 PC的Rf值 抑制百分比 種子生長抑制生長百分比 根長抑制百分比1 5 5 佈後之生物分析，不含萃取液之色層為控制組（control）。 CYM-1-21-1： CYM-1-24-3： CYM-1-25-3： CYM-1-25-4： 6
1 5 5 圖三：以高效能液相層析儀分析台灣杜鵑水溶萃取液之色層圖譜。每圖中之主 峰波（peak） 刻在純化與鑑定中。 7
1 5 5 0 20 40 60 80 100 120 140 0.01 0.20 0.30 0.40 0.50 1.00 2.00 3.00 4.00 5.00
樣品濃度百分比
滲透壓值
台灣杜鵑
麥芽糖
圖四：台灣杜鵑落葉之水溶萃取液及麥芽糖水溶液在不同百分1 5
5 比濃度下之滲透壓比較。
1 5
5
*Genetic var iability and phytlogeogr aphy of Miscanthus sinensis var . condensatus, an apomictic gr ass, based on the RAPD finger pr ints
Chang.- Hung. Chou1, Yu-Chung. C. Chiang1, 2 and Tzen-Yuh. Y. C Chiang3 1 Institute of Botany, Academia Sinica, Nankang, Taipei, Taiwan 115; 2Depar tment of Biology, National Taiwan Nor mal Univer sity, Taipei, Taiwan 105; 3Depar tment of Biology, National Cheng-Kung Univer sity, Tainan, Taiwan 700701
Received
3Cor r espondence3Author to whom all cor r espondence should be addr essed: T. Y. Chiang, Depar tment of Biology, National Cheng-Kung Univer sity, Tainan, Taiwan 700701. Tel.: +886-6-2757575 ext. 65525 (lab); Fax: +886-6-2742583,. eE-mail: [email protected]
Running titlte: Genetic var iabilityApomixis in M. sinensis var . condensatus
O O OH OH OH HO 3
1 5
5
Abstr act
DNA finger pr inting using r andom amplified polymor phic DNA (RAPD) mar ker s was employed to investigate the genetic var iation of Miscanthus sinensis var . condensatus, an apomictic gr ass, between within and among populations of Miscanthus sinensis var . condensatus, an apomictic gr ass distr ibuteding along the coasts of Taiwan and Ryukyu Islands. A total of 250 plants of thr ee populations of Taiwan, i.e. Southeast Coast, Or chid Islet and Gr een Islet, and two populations of the Ryukyu, i.e. Ishigaki and Amami-O-Shima Islets, wer e sampled. After using 40 r andom pr imer s, tThe amplified fr agmentspr oducts of some 40 r andom pr imer s showed monomor phic banding patter ns within all populations as well as among the thr ee populations examined fr om of Taiwan (South-Eeast Coast, Or chid Islet and Gr een Islet). LLow genetic var iation (, with only two polymor phic loci), existed acr oss over all populations. Never theless, but significant genetic differ entiation, was detected both between geogr aphical ar eas, i.e. populations fr om Taiwan and Ryukyu (Φct = 0.864), and between populations (Φst = 1.0) fromof Ishigaki and Amami-O-Shima Islets of Ryukyu. For compar isons, 12 individuals of Miscanthus sinensis var . glaber along a 7 km tr ansect wer e also sampled and analysed. In contr ast, a Hhigh level of var iation was exhibited in anthe concur r ently studiedthe outcr ossing taxon, M. sinensis var . glaber . The contr ast of genetic composition was lar gely In addition to the apomictic r epr oductioncor r elated with the differ ent systems of mating adopted in differ ent lineages. As well, low genetic var iation acr oss populations of M. sinensis var . condensatus might be due to high salinity acting , as a selective agent, of in the substr ate, as a selection agent, may be r esponsible in par t veon in par t for the low genetic var iation acr oss the geogr aphical r egions acr oss populations. With a cost of r educed genetic heter ogeneity, apomixis may have pr ovided a mechanism for avoiding the tr ansmission of endophytic fungi to the descendent. The likely Pphytlogeogr aphic patter n of M. sinensis var . condensatus , based onas r eflected by the RAPD data, appear ed to contr adict with an " isolation by distance" model, but r epr esentsflected a long isolation a
closer geological between Taiwan and Ryukyu ever since the mid-dle
Pleistocene.r elationship between Ishigaki and Amami-O-Shima Islets.
Keywor ds: apomixis, genetic var iability, Miscanthus sinensis var . condensatus, phytlogeogr aphy, population s differ entiation, RAPD, system of mating.
1 5
1 5
1 5
5
Intr oduction
Apomixis, a clonal r epr oduction via seed and independence independent of fer tilization, occur s thr oughout the plants, fr om algae, fer ns (Richar ds 1986Schneller et al. 1998), to angiosper ms (Richar ds 1986; Asker & J er king 1992; Koltunow et al. 1995Schneller et al. 1998). Recur r ent apomixis has been is well documented in species of mor e than 40 angiosper m families (Richar ds 1986; Asker and J er king 1992). , such as the Poaceae (Or tiz et al. 1997), Rosaceae (Holm & Ghatnekar 1996; Ur-Rahman et al. 1997), Euphor biaceae (Nassar et al. 1998), Compositae Aster aceae (Koltunow et al. 1995), and Plumbaginaceae (Palacios & Gonzales-Candelas 1997). Devoid of genetic r ecombination, apomictic pr ocesses, occur r ing in the ovule, usually r esult in genetic unifor mity in pr ogeny (Ur -Rahamn et al. 1997), within populations (Holm &and Ghatnekar
1996), or even acr oss populations within species (Palacios &and
Gonzales-Candelas 1997).
Miscanthus sinensis Ander ss. var . condensatus (Hack.) Makino, a taxon of the (Poaceae), is gr owinggr ows inat the littor al habitats, and r anges with a wide r ange in distr ibution fr om Hokkaido (J apan) to Botel Tobago (Or chid Island, Taiwan) and some Pacific Islands (Hsu, 1978; Koyama 1987). Miscanthus Ander ss. is typical of many dominant gr asses ofin easter n Asia, being wind-pollinated and usually monoecious. Vegetative pr opagation via r hizomes is the most common means by which patches ar e able to expand their size. In addition to their clonal natur e, apomixis with ster ile pollens has been well documented in M. sinensis var . condensatus (Adati 1958). In contr ast, viable pollens, which isar e capable of being car r ied for a widelong distancer angeover a long distance, and nor mal fer tilization occur in most other Miscanthus species and var ieties. In Seedthese taxaMiscanthus the seed disper sal, however , is constr ained by the seed weights and canould r each up to only five meter s (Chou &and Lee 1991). A pr evious RAPD investigation of M. flor idulus (Labill) War b., another dominant lowland species at lowlands, r evealed significant genetic differ entiation and var iability between among populations (Chiang 1993). Among the many biological factor s affecting the populationthe evolution of populations, the mating system of mating acts one of the most impor tant plays a cr itical r ole in r oles that deter mininge the levels of heter ogeneity and heter ozygosity levels in populations. Apomixis is believed to be a for ce that r educes the heter ogeneity and moves the genetic composition towar ds fixationed (Weising et al.et al. 1995).
1 5
5
pr imer s is fr equently applied to differ entiating differ entiate among populations (Lynch &and Milligan 1994; van Heusden &and Bachmann 1992), to infer r ing the systematic r elationships (Adams &and Demeke 1993), to detecting the intr ogr ession (Waugh et al. 1992), and to distinguishing the taxa with differ ent r epr oductive str ategies (e.g., Mazzucato et al. 1995; Or tiz et al. 1997). Compar ed to the allozyme and RFLP techniques, RAPD finger pr inting usually pr ovides detects gr eater var iation either within or between populations (Bostock et al. 1993; Liu &and Fur nier 1993; Caetano-Anolles and Gr esshoff 1994; Heun et al.et al. 1994; Peakall et al.et al. 1995; Per os et al.et al. 1996). Among the biological factor s affecting the population evolution, system of mating acts one of the most impor tant r oles that deter mine the levels of heter ogeneity and heter ozygosity in populations. Apomixis is believed to be a for ce that r educes the heter ogeneity and moves the genetic composition towar d fixed (Weising et al. 1995). .
The pr imar y objective of this study is to deter mine the genetic var iability of the apomictic extent in M. sinensis var . condensatus using RAPD mar ker s. With an expectation of low genetic var iability in this gr ass poaceous taxon, this techniqueRAPD finger pr inting is an ideal tool for investigating the low levels of genetic var iation and differ entiation among isolated island populations. In or der to compar e the effects of differ ent br eeding systems of mating, the genetic diver sity of M. sinensis var . glaber (Nakai) Li, a closely r elated conspecific and endemic taxon of Taiwan (Chou et al. 1999) with outcr ossing sexual r epr oduction, was also sur veyed.
Using RAPD techniques, we also investigated the genetic differ entiation between populations of M. sinensis var . condensatus of fr om Taiwan and the Ryukyu Islands, wher e four other var ieties gr ow.. AGeogr aphically, although Ishigaki Islet belongs to the Ryukyu Ar chipelago, it is closer to Taiwan (with by some 100 km) separ ation r ather than it is to Amami-O-Shima Islet, which is in the centr al Ryukyu and about 400 km nor th of Ishigaki Islet (Fig. 1). With wind pollination, pollen disper sal is usually constr ained by the gr avity. An " isolation by distance" model (i.e., the cor r elation between degr ee of genetic differ entiation and geogr aphical distance) would be expected in populations of M. sinensis var . condensatus. In other wor ds, the population of Amami-O-Shima Islet would be most diver ged fr om other populations.
1 5
5
Mater ials and Methods Population samples
Thr ee populations of Miscanthus sinensis var . condensatus of in Taiwan, i.e., the South--east Coast and two offshor e islets (Or chid Islet and Gr een Islets) , and two populations of the Ryukyu, i.e. ( Ishigaki and Amami-O-Shima Islets), of Miscanthus sinensis var . condensatus (Fig. 1) wer e sampled. These populations occur r ed mostly along the seashor e. In each population, seven to 13 subpopulations wer e sampledincluded, fr om which five individuals wer e r andomlyhaphazar dly collected at thr ee-meter inter vals.
In or der to elucidate the effects of the mating systems, i.e.,. apomixis vs. outcr ossing, five subpopulations with 12 individuals (16.5K, 17.8K, 20.6K, 21.8K, and 22.9K of the mileage mar ker s at Highway 18) with 12 individuals altogether of the sexually-r epr oducing M. sinensis var . glaber wer e sampled along a 6 km tr ansect at Chiayi County of Taiwan (Fig. 1), which is the most souther ly and dominant population on the island. . Fr esh shoots of both taxaall plants wer e r insed with the tap water and wer e dr ied using silica gel in the field using silica gel. Upon r etur ning Back to the labor ator y, the specimens wer e stor ed at -70°C until they wer e pr ocessed.
DNA extr action and RAPD sur vey
Total DNA was extr acted following the CTAB pr ocedur es of Doyle &and Doyle (1987). We used the polymer ase chain r eaction (PCR) with ar bitr ar y pr imer s for obtaining RAPDs accor ding to the method of Williams et al. (1990). We used 5 ng of DNA template, 0.3 µM primer (Operon Technologies, Alameda, California), 1 unit of Taq polymer ase (Pr omega, Madison, USA), 2.5 uL of PCR buffer , 2.5 mM MgCl2, and 100 µM of each dNTP for each 25µL PCR reaction.
PCR amplification took place in a MJ Ther mal Cycler (PTC 100) ther mal cycler (Per kin Elmer ) using the following pr ofilepr ogr ammed for : an initial 3 min denatur ation at 94°C, 35 cycles of 15s 20s at 94°C , 20s 30s at 36°C and 60s 90s at 72°C, followed by a 10-min final extension at 72°C. Reactions were car r ied out in 25 µl volumes with 0.5 ng of template DNA, 2.5 µl of 10× reaction
buffer , 2.5 µl of MgCl2 buffer, 2.5 µl of dNTP (8mM), 0.1 µM primer, and 0.7
unit of Taq polymer ase (Pr omega). For ty 10-base oligonucleotide pr imer s wer e scr eened selected fr om the Oper on (Alameda, CA) K (K1- K20) and N (N1- N20) ser ies. RAPD pr oducts wer e electr ophor esed on 2% Nusieve 3:1 3:1 agarose gels (FMC Bio Pr oducts, Rockland, Maine, USA), stained by ethidium br omide, and photogr aphed with Polar oid type 667 film. A size maker of the Bio 100 DNA Ladder (PROtech, Taipei) was used.
1 5
5
Data analysis
A RAPD data matr ix based on the pr esence and absence of loci was pr epar ed by with the aid of AMOVA-PREP (ver sion 1.01; Miller 1998). We used AMOVA vs. ver sion 1.55 (Excoffier et al. 1992) to deduce the significance of geogr aphical divisions both betweenamong populations and between among r egions. The statistics of molecular var iants
Φ
CT (among r egions),Φ
ST(among populations), and
Φ
SC (among populations within r egion) wer eestimated. The significance of theses F-statistic analogues was evaluated by 1,000 r andom per mutations of sequences among populations. The UPGMA dendr ogr am of populations was dr awn based on pair wise dissimilar ities using softwar e TFPGA (ver sion 1.3; Miller 1997).
1 5
5
Results
After the RAPD sur vey vVar ious banding patter ns wer e r evealed with when by differ ent pr imer s wer e appliedused, such as a single band of 600 base pair s (bp) was amplified by with the pr imer K2 and multiple bands r anging fr om 300 to 1,300 base pair s wer e amplifiedbp by with the pr imer K6. Most banding patter ns appear ed to bewer e monomor phic, both between and within populations ofin M. sinensis var . condensatus. In total, 1,034 monomor phic bands wer e identified. Among Of the 40 pr imer s usedscr eened, only two yielded polymor phic bandings
patter ns between individuals or populations. A 600 bp DNA fr agment
amplified with the pr imer K9 occur r ed only in the population of Amami-O-Shima Islet population (lanes 13-15)and, while another a 275 bp DNA fr agment was shar ed by all populations of fr om both Taiwan and the Ryukyu (Fig. 2). Amplified DNA fr agments with the pr imer K12 differ entiated the Taiwan populations fr om those of the Ryukyu (lanes 10-15, Fig. 3). Her e, s' : a 650 bp DNA occur r ed exclusively in the populations of fr om Ryukyu Islands while another 425 bp fr agment wer e was r estr icted to Taiwan' s populations (Fig. 3). No genetic var iation was detected between among populations of in Taiwan. And RAPD fr agements wer e monomor phic at the intr a-population level. Hier ar chical analyses of RAPD finger pr ints under anunder AMOVA indicated that the pr opor tion of molecular var iance was attr ibutable to the differ ence among r egions (
Φ
CT = 0.864; P< 0.001), among populations (Φ
ST = 1.00; P< 0.001), and among populations within the Ryukyu r egions (Φ
SC = 1.00; P<0.001) in M. sinensis var . condensatus. The AMOVA analysis of RAPD data indicated a patter n of genetic var iation with 86.4% of the differ ences distr ibuted between geogr aphic r egions, and 100.0% of the differ ences distr ibuted among populations. A UPGMA tr ees (Fig. 4) based on the pair wise similar ities of populations indicated a close r elationship similar ity between Ishigaki and Amami-O-Shima Islets, which together was then r elated to the gr oup of and of the thr ee Taiwan populations.In contr ast to the low genetic var iation in exhibited by M. sinensis var . condensatus. , of the 40 scr eened pr imer s, nine (i.e., N1, N2, N5, N6, N7, N8, N9, N10, and N13) of 40 pr imer s , yielded polymor phic fr agment bandings patter ns inin sur veying M. sinensis var . glaber . A close r elationship between individuals 3, 4, 6, 9 and 11 was indicated by RAPDs with the pr imer N8, an. And individuals 1, 5, 10 and 12 shar ed identical fr agments (Fig. 5).
1 5
5
Discussion
To explain the low levels of detected genetic var iation in M. sinensis var . condensatus, a couple of possible biological r easons possibilities should be consider ed. The sensitivity and the r esolution power on resolving genetic var iation of RAPD finger pr inting ar e the pr emises for fur ther discussion. Accor ding to the compar isons made between RAPD and other molecular typing systems (Bostock et al. 1993; Caetano-Anolles & Gr esshoff 1994), RAPD finger pr inting appear s to be one of the most power ful tools for unveiling the genetic differ ences not only between populations but also among individuals or clones. RAPD is ther efor e useful for r evealing the existing var iation. In this study, high genetic var iation r evealed by the amplification with nine oligo pr imer s in M. sinensis var . glaber as well as in a pr evious investigation on of M. flor idulus (Chiang 1993) all suggested the competence of the RAPD finger pr inting.
Appar ently, in M. sinensis var . condensatus Llow genetic var iability in M. sinensis var . condensatus is lar gely cor r elated with its apomixis, in this type of br eeding systema r epr oductive pr ocess fr equently found in plants, via which the pr obability of genetic r ecombination is dr amatically r educed (Naumova 1993; Avise 1994). It was shown that within the obligateobligateor y agamosper mous populations of dandelionsTar axacum, all individuals wer e genetically identical at 15 isozyme loci (Hughes &and Richar ds 1988). Obser vations ofn Nor th Amer ican Rubus species of Nor th Amer ica based on an M13 finger pr inting sur vey (Nybom &and Schaal 1990) contr asted the levels of genetic var iation between apomictic and sexually r epr oducing species and species with r epr oduction with gamete fusion. Var iation between populations usually can be usually detected in the species with facultative apomictic speciesxis, even if the apomixis is pr esuper ior ly dominant over the r egular sexual r epr oduction. Thus, Ggenetic homogeneity among theacr oss populations ofr om Taiwan, which ar e separ ated fr om each other by somea 30 km wide channel, as well as within the Ryukyu populations,, suggestsed an obligateobligator ye apomixis be mor e likely occur s in M. sinensis var . condensatus, although mor e cytological evidence is r equir ed.
Another possible biological factor likely r esponsible for the low genetic var iation in M. sinensis var . condensatus is the clonal pr opagation, wher ebyvia which, without genetic tr ansmission, plants mer ely expand their clump size or number r ather than incr easing their genetic diver sity. Clonal r epr oductionnatur e has been discover ed r epor ted in the Miscanthus sinensis complex, such as in M. sinensis var . tr ansmor r isonensis (Hayata) Lee, one of the dominant species of the
1 5
5
gr asslands at in high mountains of Taiwan (Chou et al. 1991), as well as in M. sinensis var . condensatus (Chou et al. 1999). But, However , clonal pr opagation alone does not necessar ily r esult in genetic homogeneity acr oss populations. Noticeaably, gGenetic var iation within populations of many clonal plants has been detected, even within a limited geogr aphical r ange, based on thusinge RAPDs analyses (cf. Ayr es &and Ryan 1997; Scheeper s et al. 1997; Sydes &and Peakall 1998; Ayr es &and Ryan 1999; Tyson et al. 1998; Bushakr a et al. 1999; Esselman et al. 1999).
Dir ectional natur al selection that leads most loci in populations towar d fixationed can not be simply r uled out simply fr om the evolutionpossible biological factor s causing low level of genetic diver sity of M. sinensis var . condensatus. High salinity in the substr ate, as athe selectiveon agent, may have homogenisedhomogenized the genetic composition both within and between populations. For example, Llow var iation of the allozymes and RAPD finger pr intsing in mangr ove plants explains illustr ates the effects possibility oof an likely intense envir onmental selection (cf. Chen 1995, Chen 199???; SSun et al. 1998Huang and Chen 1999). Appar ently With no doubt, the evolution of mating systems of mating is also lar gely r egulated by the natur al selection. In gener aAs a r ule of thumb, the the r educed genetic diver sity r esulting fr om apomixis wouldmight har m the health of populations or make species endanger ed (cf. Palacios &and Gonzales-Candelas 1997). But aApomixis can be favour edfavor ed only when a tr ade-off between genetic stability and the benefits of asexual r epr oduction exists, which fur ther explains why facultative apomixis of in plants in natur e is mostlyis widespr ead in angiosper ms facultative(Cox 1988). Recent investigations r evealed that apomixis was favour ed with the benefits the of avoidance ofing the tr ansmission of systematic pathogens in cr ops (Nassar et al. 1998). Fungal infection occur s fr equently in gr asses (Adhikar i 1990; Li &and Sutton 1995; Bills 1996), including Miscanthus (Nakase et al. 1987). Inter estingly, our r ecent study based on the amplification of nr DNA ITS sequencesr egion using univer sal pr imer s found that most taxa of the Miscanthus sinenesis complex wer e infected by the endophytic fungi, except for M. sinenesis var . condensatus (Chiang et al. 19992000). Apomixis bypassing female meiosis and syngamy pr ovides an oppor tunity for expelling the pathogens, which appear ed to have long- existed in the lineage of Miscanthus, fr om the pr ogeny individuals and the whole populations.
In contr ast to the apomixis found in M. sinensis var . condensatus, the r epr oductive str ategy on of M. sinensis var . glaber utiliszes differ s.ent str ategies.Her e, The high levels of genetic var iation (along a 7 6 km tr ansect)
1 5
5
suggest wer eas highly cor r elated with the pr edominant outcr ossing in its r epr oduction. HThe high genetic heter ogeneity within populations may also may be associated with the self-incompatibility, that wasa featur e found in a few Miscanthus species (Gr eef &and Deuter 1993). It is notable that Aafter diver ging fr om atheir common ancestor , closely r elated taxabiological entities (such as populations, var ieties and species) may adopt differ ent r epr oductive str ategies in or der to complete their life cycles and ther eby obtain populations' ensur e their sur vival (cf. Cox 1988). In this case, the evolution of the apomixis in M. sinensis var . condensatus r epr esents a shift fr om a hypothetically the ancestr al state of (, i.e., outcr ossing), which occur s in most other Miscanthus taxa, to a novel modethe der ived state involving apomixis.
A UPGMA tr ee of RAPD data r eveals the phenotypic differ ences among populations. However , the dendr ogr am infer r ed (Fig. 4) The r econstr ucted phylogeny obviously somewhat contr adictsed thewith a hypothetical model of " isolation by distance," , in which a close r elationship between populations fr om Ishigaki Islet and Taiwan, based on geogr aphical distancepr oximity, was expected. On the contr ar y, populations of Ishigaki and Amami-O-Shima Islets appear ed to shar e a mor e common r ecent common ancestor ar e mor e similar . The phytlogeogr aphic patter n with showing significant genetic differ entiation at the r egional level, based on the hier ar chical Φ statistics, at regional level better ther efor e r eflectsed the geological histor y between Ryukyu and Taiwan. That is, the genetic differ entiation between among populations, with a lar ger pr opor tion, was, to a gr eat degr ee, has r esulted fr om the histor ical (geological) events instead of the r ecur r ent r ather than gene flow. Accor ding to Kizaki &and Oshir o' s (1977) hypothesis (also cf. Ota 1998), the Ryukyu AAr chipelago and Taiwan wer e wer e onceat one time once linked dur ing the ear ly Pleistocene, when a land br idge extended eastwar d fr om mainland Asia via thr ough Taiwan to the Ryukyu Islands. Meanwhile tThe ancestr al populations of many ter r estr ial plants and animals, including M. sinensis var . condensatus, may have wer e likely to disper se migr ated via thise land br idge into the above ar eas along the land br idge. Subsequently, with the r ise of the sea levels dur ing the mid-dle Pleistocene, let lower land br idge connection becaome submer ged, which causedcr eated the geogr aphical isolation between Taiwan and the Ryukyu ar chipelagospopulations and fur ther subdivision within each r egion.
, which was then separ ated into isolated islets dur ing the middle Pleistocene. The RAPD evidence of ofthe an amplified 650 bp DNA fr agment obtained usingwith the pr imer K12 (Fig. 3) indicated suppor tsed the hypothesis of a r ecent shar ed
1 5
5
Amami-O-Shima Islets, both of which ar e now 400 km apar t. In addition, another 425 bp fr agment (with the pr imer K12), which exists exclusively existed in Taiwan populations, suggestsed a possible long per iod of isolation between Ryukyu and Taiwan, which thus allowed mutations occur r ed and accumulated in each lineage (population) independently.. Accor dingly, aSubsequently a unique-occur r ingn autapomor phic DNA locus of 600 bp in populations fr om Amami-O-Shima Islet, amplified with the pr imer K9 (Fig. 2), must have evolved in the population of after the separ ation of Amami-O-Shima IslandIslet fr om Ishigaki.. Likewise, a 425 bp fr agment, amplified with the pr imer K12, exclusively fr om Taiwan populations suggested a long histor y of isolation between Ryukyu and Taiwan, allowing the mutations accumulated in the lineages independently.
Compar ed In contr ast to the long isolation between Taiwan and Ryukyu populations, in ter ms of both geology and geogr aphy, between Taiwan and Ryukyu, the thr ee populations of Taiwan, which ar e about 30 km apar t, appear to have a closer geological r elationship and a mor e r ecent geogr aphical separ ationwer elikely separ ated mor e r ecently. This shor t isolation thus may explains why . No genetic var iation and and differ entiation among the populations indicated r ecent geological events, which pr ovided shor t time span and let the accumulation of genetic var iation within each population mor e
unlikely among these populations wer ewascould har dly bedetected.
However ,Alter nativelyOn the other hand, the possibility of a r ecent disper sal fr om Taiwan to the islets can not be simply r uled out either . The Ffounder effect can usually easily causes the loss of genetic diver sity (Har tl &and Clar k 1997), which was demonstr ated based onin a RAPD investigation of a Canadian r ed pine (Pinus r esinosa; Mosseler et al.et al. 1992), so the loss of genetic diver sity in apomictic taxa is even mor e pr obable, let alone species with apomictic r epr oduction.
In conclusion, apomixis seems to have aacted has played as a dominant r ole in deter mining the low levels of genetic var iability in Miscanthus sinensis var . condensatus. Apomictic r epr oduction without female meiosis and syngamy pr obably pr ovides an oppor tunity for getting r id of thepur ging infecting fungi fr om the pr ogeny. Differ ences individuals, which may haves been long favour ed ever since the for mation of this taxon. Highly significant differ entiation between populations of geogr aphical ar eas (i.e. Taiwan andvs. Ryukyu) is ar e likely cor r elated associated with their geological histor y of long separ ation. Ssince the defor mation of the land br idge the two ar eas have separ ated. Although the Ishigaki Islet is closer to Taiwan geogr aphically, Ishigaki and
1 5
5
Amami-O-Shima Islets shar ed a common histor y based on geological evidences. The phylogeogr aphic patter n , which is not consistent with an " isolation by distance" model, r eflected the migr ator y r outes of populations based on the RAPD finger pr ints r evealed possible migr ator y r outes of populations of M. sinensis var . condensatus.
Acknowledgements
This study is suppor ted by the National Science Council (NSC) of Taiwan gr antedwas suppor ted by gr ants fr om by the National Science Council (NSC) of Taiwan gr ants to C. H. Chou （(NSC88-2311-B001-026-B27）and T. Y. Chiang (（NSC-8889-2311-B0061-002???-B27）???）. We ar e gr ateful to thank Pr of. Shigeyuki Baba and Ms. Nozomi Oshir o, and the faculty of Agr icultur e, Univer sity of the RyukyusRyukyu, and his assistant secr etar y,for the assistance in the field or ganized our fieldtr ip to collect Miscanthus samples fr om Ryukyus. We also thank Pr of. Huang Shong, the Ffaculty of Science Ccollege, National Taiwan Nor mal Univer sity, for the financial suppor t for ofto the field tr ip to Y. C. Chiang tr avel to collect Miscanthus samples fr om RyukyusRyukyu by his per sonal funds.
1 5
5
Refer ences
Adams, R. P., and Demeke, T. (1993.) Systematic r elationships in J uniper us based on r andom amplified polymor phic DNAs (RAPDs). Taxon, 42:, 553-571. Adati, S. (1958.) Studies on the genus Miscanthus with special r efer ence to the J apanese species suitable for br eeding pur pose as foddedfodder cr ops. Bulletin Bull. of Fac.ulty of Agr on.omy of Mie Uni.ver sity, 17:, 1-112.
Adhikar i, R. S. (1990.) Phylloplane mycoflor a of thr ee gr asses: a mycosociological study. Tr op.ical Ecol.ogy, 31:, 64-68.
Asker , S. E., and J er ling, L. (1992.) Apomixis in plants. CRC Pr ess, London. Avise, J . C. (1994.) Molecular Mar ker s, Natur al Histor y and Evolution. Chapman &and Hall, New Yor k.
Ayr es, D. R., and Ryan, F. J . (1997.) The clonal and population str uctur e of a r ar e endemic plant, Wyethia r eticulata (Aster aceae): allozyme and RAPD analysis. Mol.ecular Ecol.ogy, 6:, 761-772.
Ayr es, D. R., and Ryan, F. J . (1999.) Genetic diver sity and str uctur e of the nar r ow endemic Wyethia r eticulata and its congener W. bolander i (Aster aceae) using RAPD and allozyme techniques. Amer.ican J . our nal of Bot.any, 86:, 344-353.
Bills, G. F. (1996.) Isolation and analysis of endophytic fungal communities fr om woody plants. In: Endophytic FFungi in GGr asses and WWoody PPlants, SSystematics, EEcology, and EEvolution (eEdited byds S. C. Redlin SC,, and L. M. Car r is. LM), pp. 31-66. APS Pr ess, ST. Paul. pp. 31-66.
Bostock, A., Khattak, M. N., Matthews, R., and Bur nie, J . (1993.) Compar ison of PCR finger pr inting, by r andom amplification of polymor phic DNA, with other molecular typing methods for Candida albicans. J . J our nal of Gen.er al Micr obiol.logy, 139, : 2179-2184.
Bushakr a, J . M., Hodges, S. A., Cooper , J . B., and Kaska, D. D. (1999.) The extent of clonality and genetic diver sity in the Santa Cr uz Island ir onwood, Lyonothamnus flor ibundus. Mol.ecular Ecol.ogy, 8:, 471-475.
Caetano-Anoles, G., and Gr esshoff, P. M. (1994.) DNA amplification finger pr inting: a gener al tool with applications in br eeding, identification and phylogenetic analysis of plants. In : Molecular Ecology and Evolution: Appr oaches and Applications. Edited by(eds. B. Schier water , B, B. Str eit B, G. P. Wagner GP, and R. DeSalle R),. pp. 17-31. Bir khauser Ver lag Basel, Switzer land. pp. 17-31.
Chen, Y. S. 1995. Population genetics of Kandelia candel of Taiwan based on allozyme data. Master Thesis, Depar tment of Biology, National Taiwan Nor mal Univer sity, Taipei.
1 5
5
Chiang, Y. C. (1993.) The biochemical toler ance of Miscanthus flor idulus populations to lead. Master Thesis, Faculty of Sciences, Depar tment of Botany, National Taiwan Univer sity, Taipei.
Chiang, Y. C., CH Chou, C. H., TYand Chiaaing, T. Y. (19992000.) Detection of leaf-associated fungi in Miscanthus foliage based on PCR and nucleotide sequence of the inter nal tr anscr ibed spacer (ITS) r egion of nuclear r ibosomal DNA. Bot.anical Bull.etin of Acad.emia Sin.ica (in pr ess)
Chou, C. H., and Lee, S. C. (1991.) An ecological study on Miscanthus in Yangmingshan National Par k. Yangmingshan National Par k, Taipei.
Chou, C. H., Lee, Y. F., Chiu, C. Y., Wang, Y. C., and Hsu, F. H. (1991.) Population study of Miscanthus IV. Gr owth per for mance of M. flor idulus and M. tr ansmor r isonensis and their acclimation to temper atur es and water str ess. Botanical Bot. Bulletin of. Academia Acad. SinicaSin., 32:, 57-96.
Chou, C. H., S Huang, S., SH Chen, S. H., Kuoh, C. S., Chiang, T. Y., and Chiang, Y. C. (1999. ) Ecology and evolution of Miscanthus of Taiwan. Natl.tioanl Sci.ence Coun.ncil Monthly., 27:, 1158-1169.
Cox, P. A. 1988. Monomor phic and dimor phic sexual str ategies: a modular appr oach. In Plant r epr oductive ecology: patter ns and str ategies. Edited by J . L. Doust, L. L. Doust. Oxfor d Univer sity Pr ess, New Yor k. pp. 80-97.
Doyle, J . J ., and Doyle, J . L. (1987.) A r apid DNA isolation pr ocedur e for small quantities of fr esh leaf tissue. Phytochem.istr y Bull.etin, 19:, 11-15.
Esselman, E. J ., J ianqiang, L., Cr awfor d, D. J ., Windus, J . L., and Wolfe, A. D. (1999.) Clonal diver sity in the r ar e Calamagr ostis por ter i ssp. insper ata (Poaceae): compar ative r esults for allozymes and r andom amplified polymor phic DNA (RAPD) and inter simple sequence r epeat (ISSR) mar ker s. Mol.ecular Ecol.ogy, 8:, 443-451.
Excoffier , L., Smouse, P. E., and Quattr o, J . M. (1992.) Analysis of molecular var iance infer r ed fr om metr ic distances among DNA haplotypes: application to human mitochondr ial DNA r estr iction data. Genetics, 131:, 479-491.
Gr eef, J . M., and Deuter , M. (1993.) Syntaxonomy of Miscanthus × giganteus GREEF Gr eef et DeuEU. Angewand. dte der Bot.anik, 67:, 87-90.
Har tle, D. L., and Clar k, A. G. 1（1997.） Pr inciples of population genetics. 3r d edition. Sinauer Associates, Inc. Sunder land, Massachusetts.
Heun, M., Mur phy, J . P., and Phillips, T. D. (1994.) A compar ison of RAPD and isozyme analyses for deter mining the genetic r elationships among Avena ster ilis L. accessions. Theor .r etical and Appl.ied Genet.etics, 87:, 6689-696.
Holm, S., and Ghatnekar , L. (1996.) Apomixis and sexuality in hexaploid Potentilla ar gentea. Her editas, 125:, 53-60.
1 5
5
Hsu, C. C. (1978.) Gr amineae (Poaceae). In: Flor a of Taiwan. Edited by (eds. H. L. Li HL, T. S. LKiu TS, T. Koyama, T, C. E. DeVol CE),. Vol. V. , pp. 372-783. Epoch, Taipei. pp. 372-783.
Huang, S., and Chen, Y. C. 1999. Patter ns of genetic var iation of Kandelia candel among populations along coasts of south China Sea. In Pr oceedings of biodiver sity of wetlands in Taiwan and adjacent ar eas. Edited by T. Y. Chiang, and T. W. Hsu. The Council of Agr icultur e, Taipei. Pp. 51-57.
Hughes, J ., and Richar ds, A. J . (1988.) The genetic str uctur e of populations of sexual and asexual Tar axacum (dandelions)... Her edity, 60:, 161-171.
Kizaki, K., and I. Oshir o. I (1977.) Paleogeogr aphy of the Ryukyu Islands. Mar .ine Sci.ence Month.hly, 9:, 542-549.
Koltunow AM, Bicknell RA, Chaudhur y AM (1995) Apomixis: molecular str ategies for the gener ation of genetically identical seeds without fer tilization. Plant Physiology, 108, 1345-1352.
Koyama, T. (1987.) Gr asses of J apan and Its Neighbor ing Regions. An Identification Manual. The New Yor k Botanical Gar den, New Yor k.
Li, B. C., and Sutton, J . C. (1995.) Evaluation of leaf-associated micr oor ganisms for biocontr ol of tan spot in wheat foliage. Fitopatol.ogia Br asil.eir a, 20:, 543-552.
Liu, Z., and Fur nier , G. R. (1993.) Compar ison of allozyme, RFLP, and RAPD mar ker s for r evealing genetic var iation within and between tr embling aspen and bigtooth sapen. Theor.etical and Appl.ied Genet.etics, 87:, 97-105.
Lynch, M., and Milligan, B. G. (1994.) Analysis of population genetic str uctur e with RAPD mar ker s. Mol.ecular Ecol.ogy, 3:, 91-99.
Mazzucato, A., Bar caccia, G., Pezzotti, M., and Falcinelli, M. (1995.) Biochemical and molecular mar ker s for investigating the mode of r epr oduction in the facultative apomict Poa pr atensis L. Sexual Sex. Pl.ant Repr od.uction, 8:, 133-138.
Miller , M. P. (1997.) TFPGA: tools for population genetic analyses. A windows pr ogr am for analysis of allozyme data and molecular population genetic data. Ver sion 1.3. Depar tment of Biological Sciences, Nor ther n Ar izona Univer sity, Flagstaff, AZ.
Miller , M. P. (1998.) AMOVA-PREP 1.01: a pr ogr am for the pr epar ation of the AMOVA input files fr om dominant-mar ker r aw data. Depar tment of Biological Sciences, Nor ther n Ar izona Univer sity, Flagstaff, AZ.
Mosseler , A., Egger , K. N., and Hughes, G. A. (1992.) Low levels of genetic diver sity in r ed pine confir med by r andom amplified polymor phic DNA maker s. Can.adian J . our nal of For .est Res.ear ches, 22:, 1332-1337.
