開發質譜法用於腸道微生物肽聚醣之結構鑑定以及Amuc_1100外膜蛋白之標靶定量

Abstract

細菌的細胞包膜是由一層或兩層細胞膜，加上一層肽聚醣(Peptidoglycan, PGN)所組成，是維持細胞完整性和形態的關鍵。位於細胞的最外層邊界，細胞包膜含有肽聚醣、膜蛋白和脂多醣等生物活性成分，這些成分是調節著宿主和微生物組相互作用的媒介，同時也是免疫相關疾病和代謝紊亂的潛在治療標靶。現代質譜及其相關技術的出現，其高靈敏度、提供詳細結構資訊的能力和可靠的定量性能，使其成為研究廣泛的微生物衍生化合物不可或缺的分析工具。 然而，開發用於深入研究的分析工具仍面臨挑戰。本論文展示了兩種創新的質譜法應用，闡釋了肽聚醣的結構特徵，以及對Amuc_1100膜蛋白進行相對定量分析。 第一部分著重在腸道細菌肽聚醣的結構特性分析。肽聚醣是一種由細胞壁胜肽(Muropeptide)組成的網狀聚合物，可作為微生物的保護屏障。研究人員透過肽聚醣辨識系統探索了宿主與微生物組的相互作用，並發現了調節宿主反應的關鍵細胞壁胜肽。然而，大部分常見的細胞壁胜肽表徵技術是勞動密集的，且需要手動分析解譜，這主要是源自複雜的交聯肽聚醣結構。每個物種都有獨特的部分修飾和橋間/橋內肽，進一步使得肽聚醣的結構分析變得複雜。在這項工作中，我們開發了一個高通量自動化細胞壁胜肽分析平台(High-throughput Automated Muropeptide Analysis, HAMA)，利用串聯質譜法(MS/MS)和電腦計算的二次質譜碎片匹配來全面鑑定細胞壁胜肽結構、量化其各自含量並推斷肽聚醣交聯類型。我們使用來自大腸桿菌和金黃色葡萄球菌的已知結構肽聚醣展示了HAMA平台的有效性，並將其應用擴展到常見腸道細菌，包括雙歧桿菌、擬桿菌、乳桿菌、腸球菌和阿克曼氏菌。我們的分析可透過HAMA平台準確地鑑定單聚體/多聚體的細胞壁胜肽，並明確地區分結構異構體。此外我們發現細胞剛性可能透過雙歧桿菌屬內肽間橋的長度或轉肽位點與肽聚醣結構的緊密性有關。HAMA平台展示了自動化、直觀和準確的肽聚醣組成分析，有望深入了解醣類合成後修飾、肽間橋的變化以及肽聚醣的交聯類型。 在第二部分中，我們開發了Amuc_1100外膜蛋白的標靶蛋白質體學檢測。Akkermansia muciiniphila是著名的黏蛋白降解細菌，因與人類宿主的代謝疾病密切相關，使其成為有前途的次世代益生菌。此合作計畫旨在透過評估Amuc_1100膜蛋白的豐度，從台灣A. muciiniphila分離株中找尋有前途的A. muciiniphila益生菌株，該蛋白先前已被證實對肥胖和糖尿病小鼠的代謝有所改善。我們採用散彈槍蛋白質體分析法(Shotgun proteomics)和標靶蛋白質體分析法(Targeted proteomics)分別對Amuc_1100膜蛋白進行定性和定量分析。利用胰蛋白酶肽的直接測量揭示了七株A. muciniphila分離菌株中Amuc_1100蛋白的相對含量，其中DSM 22959菌株表現出最高含量。這項觀察結果似乎與Toll樣受體2 (TLR2)細胞體外生物活性測定的結果一致。然而，除了Amuc_1100蛋白之外的其他成分也可能激活TLR2受體，從而導致測定結果的複雜性。總體而言，本研究提出穩健且潛在通用的方法來定量Amuc_1100膜蛋白，並提供分離菌株間Amuc_1100蛋白含量和氨基酸序列多樣性的見解。

The bacterial cell envelope, comprising one or two membranes supplemented with a layer of peptidoglycan (PGN), is pivotal for maintaining cell integrity and morphology. Situated at the outermost boundary of the cell, this envelope harbors bioactive components such as PGN, membrane proteins, and lipopolysaccharides, which mediate signal transduction in host-microbiome interactions and offer potential therapeutic targets for immune-related diseases and metabolic disorders. The emergence of modern mass spectrometry and its associated technologies has made it an indispensable analytical tool for investigating the broad range of microbial-derived compounds due to its high sensitivity, ability to provide detailed structural information, and reliable quantitative performance. However, challenges persist in the development of analytical tools for in-depth studies. This dissertation presents two innovative mass spectrometry applications that elucidate the structural characteristics of PGN and enable relative quantitative analysis of the Amuc_1100 membrane protein. The first section focuses on the structural characterization of gut bacterial PGNs. Peptidoglycan, a mesh-like polymer consisting of muropeptides, serves as a protective barrier for microorganisms. Researchers have explored host-microbiome interactions through PGN recognition systems and discovered key muropeptides that modulate host responses. However, most common characterization techniques for muropeptides are labor-intensive and involve manual analysis of mass spectra, primarily due to the complex cross-linked PGN structures. Each species has unique moiety modifications and inter-/intra-bridges, which further complicates the structural analysis of PGN. In this work, we developed a high-throughput automated muropeptide analysis (HAMA) platform, leveraging tandem mass spectrometry and in silico muropeptide MS/MS fragmentation matching to comprehensively identify muropeptide structures, quantify their abundance, and infer PGN cross-linking types. We demonstrated the effectiveness of the HAMA platform using well-characterized PGNs from E. coli and S. aureus and extended its application to common gut bacteria, including species of Bifidobacterium, Bacteroides, Lactobacillus, Enterococcus, and Akkermansia. Our analysis accurately identifies muropeptide mono-/multi-mers and unambiguously discriminates structural isomers via the HAMA platform. Furthermore, we found that the cell stiffness may be correlated to the compactness of the PGN structures through the length of interpeptide bridges or the site of transpeptidation within Bifidobacterium species. The HAMA framework exhibits an automated, intuitive, and accurate analysis of PGN compositions, promising insights into post-synthetic modifications of saccharides, variation in interpeptide bridges, and cross-linking types within bacterial PGNs. In the second section, we developed a targeted proteomic assay for the Amuc_1100 outer membrane protein. Akkermansia muciiniphila, a well-known mucin-degrading bacterium, is closely associated with metabolic diseases in the human host, making it a promising next-generation probiotic. This collaborative project aimed to identify promising probiotic strains of A. muciniphila from among Taiwanese A. muciniphila isolates by evaluating the abundance of the Amuc_1100 membrane protein, which has previously demonstrated metabolic improvements in obese and diabetic mice. We employed shotgun proteomics and targeted proteomics for the qualitative and quantitative analysis of the Amuc_1100 membrane protein, respectively. The direct measurement of tryptic peptides revealed the relative abundance of Amuc_1100 protein in seven A. muciniphila isolates, with the DSM 22959 strain exhibiting the highest abundance. This observation appeared to align with the result of in vitro bioactivity assay conducted on Toll-like receptor 2 (TLR2) cell lines. However, other components aside from the Amuc_1100 protein may also activate the TLR2 receptor, contributing to assay result complexity. Overall, this study introduces a robust and potentially universal approach for quantifying the Amuc_1100 membrane protein, offering insights into the within-species diversity of Amuc_1100 protein abundance and amino acid sequences among A. muciniphila isolates.