整合性醣蛋白質體技術分析具不同轉移能力的癌細胞株

Abstract

目前，研究複雜的樣品如細胞或血清蛋白質體的醣基化，主要有兩種互補性方 式，一種是基於親和力或化學方式來純化醣蛋白，再藉由酵素去除醣質並以蛋白質 體學的方法針對去醣化的醣胜肽進行分析；另一種方式則是著重在以醣質體分析法 鑑定被酵素切下來的醣質結構。然而，這兩種方法不管單獨或合併使用，都無法解 決特異位點醣化圖譜 (site-specific glycosylation pattern) 這個問題。因此，改善醣 基化研究的策略是非常必要的，包括在複雜樣品中，針對個別的醣化位點的不同醣 質種類做相關的定量分析。 在本篇論文中，提供一個整合了蛋白質序列、醣化位點和醣質組成的分析方式， 並將之應用在具不同轉移能力細胞株的醣基化比較分析。首先，我們應用過濾式樣 品製備法 (Filter Aided Sample Preparation) 去除介面活性劑 (detergent) 並藉由 凝集素 (lectin) 純化醣胜肽。所純化的醣胜肽再以高精準度的質譜儀 LTQ Orbitrap Velos 的HCD MS / MS 斷裂模式分析完整的醣胜肽及鑑定其胜肽序列；最後再 經 由標靶性 CID MS3 確認醣胜肽的蛋白質序列。應用這種策略，我們鑑定出 101 條 獨特的醣化胜肽，這些醣化胜肽分別來自於5 個目標蛋白的 13 個醣化位點：EGFR, integrin α5, integrin β1, LAMP-1 和LAMP-2。關於特異位點醣化圖譜方面，我們找 到 EGFR 位點特異性呈現的醣質及 integrin β1 具有特定位點的 core-fucosylation。 同時，我們應用 Progenesis LC-MS 軟體針對每一個個別醣化位點的不同醣質 種類在 MS 層次做定量分析，應用 Progenesis LC-MS 方法，在具備高轉移能力 的 A431-III 細胞株共發現22個表現上升和10個表現下降的醣化胜肽。藉由本篇論 文顯示，在醣胜肽的層次去探討醣質特異性的改變有其好處並且對於特異位點醣化 圖譜的研究有其重要性。

In general, there are two major approaches to study the glycosylation of complex sample such as cell lysate or serum. One is based on affinity or chemical capture to enrich the glycoproteins followed by enzymatic removal of their glycans and proteomic analysis of the deglycosylated glycopeptides. The complementary glycomics-based approach focuses on the analysis of released glycan pools. However, both methods, either alone or in combination, fail to address the issue of site-specific glycosylation pattern. Consequently, improved strategies for glycosylation studies including the quantitation of different glycan forms associated with individual glyco-sites that occur in complex samples, is much needed. In this thesis, I have developed an experimental workflow in which analyses of protein sequence, glycosylation sites and glycan compositions are integrated, for a comparative analysis of the glycosylation patterns of cell lines with different metastasis ability. We have adapted the ‘‘filter aided sample preparation’’ (FASP)-based method developed by Wiśniewski et al in which detergent are removed and glycopeptides are enriched by binding to lectins. The glycopeptides analysis is coupled with high-precision mass spectrometry for measurement of intact glycopeptides and identification of their peptide cores by MS/MS (LTQ-Orbitrap Velos, HCD mode). The peptide sequences are then confirmed by targeted CID MS3. Using this approach, we identified 101 unique glycopeptides mapping to 13 glycosylation sites from 5 target proteins : EGFR, integrin α5, integrin β1, LAMP-1 and LAMP-2. For the site-specific glycosylation, we find the site-specific representative glycans of EGFR and site-depentant core-fucosylation of integrin β1. In parallel, each glyco-forms of individual glyco-sites are quantitated by Progenesis LC-MS software at the MS level. The Progenesis LC–MS approach revealed 22 up-regulation and 10 down-regulation glycopeptides in higher metastasis ability A431-III cells. The strategy involving the investigation of glycan specific changes at the glycopeptide level has many advantages and and our findings underline the importance of protein site-specific glycosylation.