藉由代謝體學和化學計量法評估禾本科作物之活性成分

Abstract

主要糧食與飼料作物為禾本科植物，其植物次級代謝物除參與生長與防禦的調控外，亦具多樣的生理活性，為保健食品的重要來源。本研究以高解析質譜為核心分析平台，結合代謝體學與化學計量學策略，採用非目標式、猜測式與目標式分析流程，以偵測不同類型的次級代謝物。藉此克服化學結構多樣性與現有資料庫覆蓋範圍有限，所造成次級代謝物準確註釋與系統性分析之困難。結果顯示禾本科作物中次級代謝物酚酸、類黃酮、花青素、酚醯胺及皂苷為主要類別；線上資料庫初步篩選同時偵測到醛類、胺基酸衍生物、脂肪酸、有機酸、生物鹼、香豆素、二萜、吲哚及核苷酸等多種代謝物。非目標式代謝體分析結果顯示，狼尾草的特徵酚酸為綠原酸類化合物；其類黃酮衍生物除具有O-醣苷鍵結外，亦具有相當比例的C-醣苷鍵結結構。進一步透過自建資料庫比對，共註釋出10種花青素，其中7種為首次於狼尾草中發現。在12種穀物樣品中共鑑定出53種酚醯胺化合物，包括aliphatic型、aromatic型、agmatine型/二聚體及燕麥胺類，其中含有17對順反異構物。利用分子網絡分析模式首次於小米中揭示九種酚醯胺醣苷結合態。發芽後的穀物幼苗中，酚醯胺主要在根部大量蓄積，且Z型異構體比例超過40%，本研究亦首次對該類異構體進行相對定量分析。本研究所建立的分析模式、化學結構資訊與資料庫，成功應用於禾本科植物多類次級代謝物的註釋與比較；可為其他禾本科作物研究提供植物化合物層面的參考；亦為天然物研究、功能性食品開發及較少研究之植物物種的化學特徵解析，提供可行的策略與參考框架。

Major cereal and forage crops are Poaceae plants. Their secondary metabolites not only participate in regulating growth and defense but also exhibit diverse bioactivities, serving as important sources of functional ingredients. In this study, the high-resolution mass spectrometry platform integrated with metabolomics and chemometric approaches was employed as the core analytical system. Untargeted, suspect, and targeted workflows were adopted to detect various classes of secondary metabolites, thereby overcoming the challenges in accurate annotation and systematic analysis arising from structural diversity and the limited coverage of existing databases. The results revealed that phenolic acids, flavonoids, anthocyanins, phenolamides, and saponins constitute the major groups of secondary metabolites in Poaceae crops. Preliminary screening using online databases additionally detected a wide range of metabolites, including aldehydes, amino acid derivatives, fatty acids, organic acids, alkaloids, coumarins, diterpenes, indoles, and nucleotides. Non-targeted metabolomic analysis showed that chlorogenic acids were the characteristic phenolic acids in Napier grass, while its flavonoid derivatives contained not only O-glycosidic linkages but also a considerable proportion of C-glycosidic structures. Further comparison with an in-house database enabled the annotation of 10 anthocyanins, seven of those were characterized in Napier grass for the first time. Across 12 cereal samples, a total of 53 phenolamides were putatively identified, including aliphatic, aromatic, agmatine-derived/dimeric, and avenanthramide types, comprising 17 E/Z-isomeric pairs. Molecular networking analysis further revealed nine glycosylated phenolamides in millet for the first time. In germinated cereal seedlings, phenolamides were predominantly accumulated in roots, with Z-isomers accounting for more than 40%. This study also represents the first relative quantification of these isomers. The analytical workflows, together with the established structural information and databases, were successfully applied to the annotation and comparison of multiple classes of secondary metabolites in Poaceae crops. The analytical workflows, along with the established structural information and databases, were successfully applied to the annotation and comparison of multiple classes of secondary metabolites in Poaceae crops. These results provide compound-level references for future studies on other Poaceae crops and offer feasible strategies and frameworks for phytochemical research, functional foods development, and the structural characterization of less-studied plant species.