基於質譜法之代謝組學整合性策略:從單細胞分析到疾病標誌物之研究

Abstract

代謝組學是指探討生物樣本中的代謝物（如體液、細胞和組織）以發現生物標記物。其中，質譜法其具有高靈敏度、選擇性和高掃描速度等特性，得以在化學層級上作為分析代謝物中一種合適的方法。然而，基於質譜法的代謝組學正在面臨一些挑戰，如樣品某些代謝物的含量較低或同時存在著不同極性和離子化性質的代謝物以及較低含量的代謝物導致鑑定上有困難。為了解決這些挑戰，本論文提出了使用奈米級微流液相層析結合多噴嘴發射器來增強儀器靈敏度，以及使用化學衍生化的方法來提高代謝物的離子化效率以及分離效果。這些方法可以提高低含量代謝物的檢測以及優化代謝組學的分析流程。 第一部份，本論文呈現出單細胞分析的多模式平台，此平台使用奈米級微流液相層析和多噴嘴發射器質譜儀以及單細胞選擇技術結合，以研究群體和亞群體內的代謝組學異質性。我們透過此分析平台，分別從單顆人類骨肉瘤細胞和膠質母細胞瘤細胞中鑑定出15和17種脂類。單細胞選擇技術還被用來分離具有DNA損傷反應和快速遷移的異常亞族群細胞並且透過分析平台進一步探討代謝組學差異。此多模式平台提供了一種有前途的單細胞代謝組學微量尺度之分析方法，為細胞異質性提供了有價值的洞察。 第二部份，本論文呈現出使用高性能化學同位素標記平台對非酒精性脂肪性肝病進行全面的代謝組學分析。其中非酒精性脂肪性肝病進一步分類為非酒精性脂肪肝和非酒精性脂肪性肝炎。我們的研究包括代謝組和總體基因體學數據的相關性分析。該研究鑑定出潛在的生物標記物，可以使用非侵入性之程序在血漿中準確地診斷非酒精性脂肪性肝病之亞型與健康對照組。同時，該研究還探討了非酒精性脂肪性肝病亞型患者的肝臟和尿液樣本，進一步透過代謝通路分析發現其相關代謝物可能會被調控以緩解非酒精性脂肪性肝病進展。此外，總體基因體學分析確定了在非酒精性脂肪肝中具有上下調控的特定細菌物種，暗示了腸道菌群失調和其疾病發展之間的聯繫。這些資訊可能有助於進一步研究非酒精性脂肪肝病的疾病機制和治療方法的發展。

Metabolomics refers to the study of metabolites in biological samples such as biofluids, cells, and tissues to discover biomarkers. Mass spectrometry (MS) is a powerful tool for analyzing metabolites at a chemical level due to its high sensitivity, selectivity, and fast scanning speed. However, MS-based metabolomics faces challenges including lower sample abundance, coexistence of metabolites with diverse polarities and ionization properties, and difficulty identifying low-abundance compounds. To address these challenges, the dissertation proposes using nanoflow liquid chromatography combined with multinozzle emitters to enhance the sensitivity of instrument, chemical derivatization to improve metabolite ionization and separation efficiency. These methods can improve the detection of low-abundance metabolites and overall quality of metabolomics results. First of all, the dissertation presents the multimodal platform for single-cell analysis that combines nanoflow liquid chromatography and multinozzle emitters MS with a functional single-cell selection (fSCS) pipeline to investigate metabolomics heterogeneity within the population and subpopulation. Using this platform, 15 and 17 lipids were identified from single human osteosarcoma and glioblastoma cells, respectively. The fSCS pipeline was also used to isolate and we analyzed the aberrant subpopulations of cells with DNA damage response and fast migration. This multimodal platform offers a promising approach for the nano-to-micro analysis of single-cell metabolomics, providing valuable insights into cell heterogeneity. Second, the dissertation presents the comprehensive metabolomic profile of non-alcoholic fatty liver disease (NAFLD) by analyzing four different specimens using high-performance chemical isotope labeling (CIL) platform. Our investigation includes correlation analyses between metagenomic and metabolomic data. We have identified potential biomarkers that can accurately diagnose the presence of nonalcoholic fatty liver (NAFL), nonalcoholic steatohepatitis (NASH), and healthy control status in plasma using a non-invasive procedure. Moreover, the results of the metabolomics pathway analysis may suggest that metabolites were regulated to alleviate the progression of NAFLD by inhibiting ROS and reducing lipid accumulation in the liver, as demonstrated by different specimens. Additionally, metagenomic analyses pinpoint specific bacterial species with up- or down-regulation in NAFLD, suggesting a link between gut microbiota dysbiosis and NAFLD development. Overall, these chapters offer significant insights into metabolomics'' potential applications and advancements.