建立褐根病接種系統及探討葉片代謝體分析運用於樹木健康偵測之潛力

Abstract

樹木褐根病由有害木層孔菌 Pyrrhoderma noxium (≡Phellinus noxius) 所引起，為熱帶及亞熱帶地區常見樹木病害，罹病樹木往往在地上部出現病徵或病兆之前即因大量根系腐朽而風倒，造成潛在公安疑慮，故如何早期偵測以採取適當防治策略十分重要。回顧前人研究，雖已有寡核苷酸微晶片 (oligonucleotide microchip)、聚合酶連鎖反應 (polymerase chain reaction, PCR) 或恆溫環狀擴增法 (loop-mediated isothermal amplification, LAMP) 可被應用於褐根病偵測上，然褐根病菌於樹木根系並非均勻分布而採樣不易，且採樣造成的傷口有可能增加感染機會及弱化樹體。由於罹患褐根病的樹木，樹冠常見兩種典型病徵：葉片下垂並褐化的急性立枯及葉片黃化、小葉化的慢性萎凋，且前人研究發現由樹幹基部接種Ophiostoma novo-ulmi之英格蘭榆 (Ulmus procera)，其枝條中防禦相關基因之表現量顯著上升，故推估罹病樹木可能產生系統性防禦相關的次級代謝物及植物荷爾蒙，而可於葉片進行早期偵測。本研究首先以田間呈現急性萎凋病徵之枇杷及慢性萎凋病徵之榕樹為對象，建立以P. noxium麥粒均勻包覆根部之接種系統。枇杷苗於接種 P. noxium 後 21-35 天，地上部出現萎凋病徵，並可於根系再次分離到 P. noxium；相對地，榕樹苗於接種 P. noxium 後未出現地上部病徵，且僅能於少數植株的主根下半部及殘根中分離出 P. noxium。而後運用高效液相層析質譜法 (ultra-performance liquid chromatography-mass spectrometry, UPLC-MS) 偵測兩樹種在生物及非生物逆境下，葉片內特定目標 (targeted) 之抗性反應相關荷爾蒙水楊酸 (salicylic acid) 含量，以及具顯著變化的非特定目標 (untargeted) 代謝物訊號。生物性逆境組包括以 P. noxium麥粒接種源、無菌麥粒處理及對照組，樣本為接種前 (第0天)、接種後第 7、21、35 天之葉片；非生物性逆境組包括乾旱、淹水及對照組，樣本為處理前 (第0天)、接種後第 7、14、21 天之葉片。實驗結果顯示，枇杷及榕樹葉片水楊酸濃度於生物及非生物性逆境下，皆與對照組無顯著差異，顯示該等逆境所誘導之訊息傳導途徑與水楊酸無關。以變方分析法 (analysis of variance, ANOVA) 篩選在生物/非生物逆境與對照組具2倍以上或0.5倍以下顯著差異之非特定目標訊號，分別於枇杷生物性逆境篩選出 6 個、榕樹生物性逆境 5 個、枇杷非生物性逆境 9 個、榕樹非生物性逆境 7 個訊號。進一步與線上代謝體資料庫比對後，發現生物及非生物性逆境下具顯著差異之代謝物不同，僅枇杷葉片有一可能為相同的代謝物 chlorogenic acid；而所有比對到的代謝物除離層素 (abscisic acid, ABA) 之外，皆僅於單一時間點被偵測到與對照組具顯著差異。其中苯丙素類 (phenylpropanoids) 代謝物於枇杷及榕樹生物性逆境、枇杷非生物性逆境下皆有顯著增加，尤其枇杷於非生物性逆境處理 7 天後，尚未出現病徵時該類代謝物已明顯高於對照組，顯示其作為樹木健康指標之潛力。離層素有機會作為榕樹乾旱逆境的指標，於處理後的三個採樣時間點皆可偵測到其明顯高於對照組。黃酮類 (flavonoids) 代謝物則於接種 P. noxium 但尚未萎凋的枇杷與對照組有顯著差異。本研究為首度跳脫傳統根部採樣及核酸檢測法之嘗試，後續須進行盆栽及田間樹木之重複實驗，以驗證所篩選出的訊號是否與本次實驗有相同的變化趨勢，並以標準品核對或結構解析方式確認代謝物之種類，期許為樹木根部健康偵測提供新的參考方案。

Brown root rot caused by Pyrrhoderma noxium (≡Phellinus noxius) is a common tree disease in tropical and subtropical regions. Since underground root infection by P. noxium does not always result in obvious aboveground symptoms, the disease is often detected after the trees with severe root rot are toppled over by strong winds, which causes potential threat to public safety. Therefore, early detection followed by effective control measures is crucial. In previous studies, oligonucleotide microchip, polymerase chain reaction (PCR) and loop-mediated isothermal amplification (LAMP) have been applied for the detection of brown root rot disease. However, root sampling is not easy because the infection sites are not evenly distributed and the sampling itself causes wounds, which may lead to wound infection and weakening of the tree. Trees suffering from brown root rot disease show two typical types of above-ground symptoms, quick decline (with droopy and brown leaves) and slow wilting (with small and yellowing leaves). Previous study revealed that after inoculation with Ophiostoma novo-ulmi from basal stem, Ulmus procera showed significant enhanced expression of defense-related genes in its branches. These indicate that systemic acquired defense-related secondary metabolites or hormones in the leaves may be detectable at early stage of infection. In this study, we first set up a pot assay system, based on root inoculation with P. noxium–colonized grains using loquat and banyan cuttings, representing the trees showing quick decline and slow-wilting symptoms, respectively. Loquat cuttings showed aboveground symptoms 21-35 days after P. noxium inoculation, and P. noxium could be reisolated from the root system. Banyan cuttings, by contrast, showed no symptoms and P. noxium could only be reisolated from the lower part of the tap root or root residues of a few plants. Next, we used high-performance liquid chromatography-mass spectrometry (HPLC-MS) techniques to detect the targeted defense hormone salicylic acid and untargeted metabolites in the leaves of loquat and banyan under biotic and abiotic stresses. In the biotic stress trial, samples were collected from 0 (prior to inoculation), 7, 21, and 35 days after inoculation with P. noxium-colonized grains, sterilized grains, and control; in the abiotic stress trial, samples were collected from 0 (prior to treatment), 7, 14, and 21 days after drought treatment, flooding treatment, and control. In loquat and banyan, no significant differences between biotic/abiotic stresses and control were detected for salicylic acid, suggesting that salicylic acid did not appear to be involved in the defense signaling of P. noxium infection, drought, and flooding. The untargeted compounds showing significant differences between biotic/abiotic stresses and control (fold changes ≥ 2 or ≤ 0.5) were selected by analysis of variance (ANOVA). We detected 6, 5, 9, and 7 significantly different compounds from loquat under biotic stress, banyan under biotic stress, loquat under abiotic stresses, and banyan under abiotic stresses. The compounds were then predicted by using online metabolomics databases. The significantly different metabolites detected from biotic and abiotic stress treatments were different, with chlorogenic acid in loquat leaves as the only exception. All the predicted compounds (except abscisic acid, ABA) showed significant differences at a single time point. Among the predicted compounds, phenylpropanoids could be indicators of plant health: they were significantly induced in loquat/banyan under biotic stress and loquat under abiotic stresses (in particular, significantly higher level was detected in asymptomatic loquat at as early as 7 days after abiotic stress treatment). ABA can be an indicator of drought stress for banyan: it showed higher levels at all the three time points. Significant differences were also found for flavonoids in P. noxium-inoculated loquat before wilting. As a prioneer study aimed to find alternatives to conventional root sampling and DNA detection methods, the experiments need to be repeated to validate the time-course changes of candidate compounds in the plants grown in the pots and field. The identities of potentially known or novel compounds will be confirmed by using a commercial standard or by conducting structural analysis. The study will provide new insights into early detection of root health in trees.