開發醣蛋白水解及純化之方法

Optimization of Glycoprotein Digestion and Sequential Enrichment for Identification of Glycoproteome

蘇莞之(Wan-Chih Su)

指導教授 ： 陳玉如

國立臺灣大學/理學院/化學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.03046

摘要

近年來質譜儀已經被大量使用於醣蛋白質體學的研究，然而，由於醣基的多樣性、含醣的胜肽含量低、以及游離效率低鑑定醣基修飾的位置及醣的結構仍面對嚴峻的挑戰，因此，在目前的分析流程中，例如：醣蛋白水解及純化方法的效率，對於後續以質譜為基礎的醣蛋白分析具有很重要的角色。在此論文中，我們使用六種標準醣蛋白，分別為：辣根過氧化物酶、牛胎球蛋白B、牛胎球蛋白A、雞卵白蛋白、牛乳鐵蛋白及人類運鐵蛋白，針對酵素水解及醣胜肽純化方法做最佳化條件，以改善醣蛋白質體的鑑定結果。
在論文的第一部分，我們使用上述六種單標準醣蛋白測試醣蛋白變性及水解的條件。在變性方法測試中，我們使用四種破壞蛋白質結構的方法((1)攝氏95度加熱、(2) 50% 醇類分子2,2,2-三氟乙醇、(3) 6M 鹽類分子尿素及(4) 6M 酸性鹽類分子鹽酸胍)測試對醣蛋白水解的效率。在這四種方法中，50% 2,2,2-三氟乙醇變性方法能夠在質譜偵測得到最多數量的醣胜肽(為其他三種方法的1.2到1.4倍)及最高的醣胜肽訊號(為其他三種方法的1.5到32.4倍)。在水解條件的測試中，我們以[二硫蘇糖醇]：[碘乙酰胺]：[第二次二硫蘇糖醇]不同濃度(mM)比例 (5：25：25 (條件A) 、5：45：45 (條件B)、10：25：25 (條件C)、10：45：45 (條件D))測試，在這四種條件中，條件A (5:25:25)可以得到最多的醣胜肽鑑定數目。因此，在之後的醣蛋白混和樣品中，我們將使用50% 2,2,2-三氟乙醇結合5 mM的二硫蘇糖醇、25 mM碘乙酰胺及25 mM二硫蘇糖醇的蛋白質水解方法。
在論文的第二部分，我們期望發展結合非專一性的親水交互型液態色層分析 (Hydrophilic interaction liquid chromatography) 及針對辨識唾液酸的二氧化鈦兩種純化醣胜肽方法，並利用StageTips純化平台以提供效能更好的純化結果。先以二氧化鈦而後連接親水液態層析的序列純化方法(二氧化鈦─親水液態層析)應用於標準醣蛋白混和樣品的結果中，相較於單一以二氧化鈦純化的20條醣胜肽或親水液態層析方法的39條醣胜肽方法，我們可以得到更多的47條完整醣胜肽。最後，為了驗證我們的序列純化方法之應用性，我們分析了人類海拉細胞 (HeLa cell) 膜蛋白，以二氧化鈦─親水液態層析做序列純化的三重複實驗結果，我們可以鑑定到398.7 ± 19.5條醣胜肽，在親水液態層析─二氧化鈦序列純化的三重複實驗結果，則可以鑑定到373.7 ± 37.9條醣胜肽，相較於單一親水液態層析及二氧化鈦方法鑑定到的302.7 ± 41.5及309.7 ± 29.2條醣胜肽多了1.2到1.4倍。總結而言，不論是在標準醣蛋白混和樣品及較複雜的生物樣品中，我們發展的二氧化鈦─親水液態層析序列純化方法都能夠成功提升醣蛋白鑑定率。
因此，此醣蛋白變性及水解方法結合醣胜肽序列純化策略提供一個高效能的醣蛋白樣品處理及純化策略，可提升醣蛋白鑑定率且保留完整醣基，期望未來這個研究平台能夠促進glycoproteome的深入分析結果。

關鍵字

醣蛋白水解 ； 醣胜肽序列純化 ； 醣蛋白質體學

並列摘要

The utility of mass spectrometry (MS) for identification of post-translational modifications of proteins has boosted glycoproteomics research in recent years. However, site-specific delineation on the glycan occupancy and structure is still a challenging task due to the diversity of glycan structures, low abundance and low ionization response of glycopeptides in tryptic peptide mixtures. Nevertheless, several procedures in the shotgun-based glycoproteomic analysis platform, such as glycoprotein digestion and enrichment methods, are crucial for subsequent mass spectrometry-based glycoproteomic identification. In this study, we attempted to optimize digestion and enrichment methods to improve glycoprotein identification with six standard glycoproteins, namely horseradish peroxidase (HRP), bovine fetuin-B (Fet), chicken ovalbumin (OVA), bovine asialofetuin (FetA), bovine lactotransferrin (LF) and human transferrin (TF) and HeLa cell membrane proteins.
In the first part of the thesis, we used these six single standard glycoprotein to evaluate glycoprotein denaturing methods and pre-digestion conditions. Of the denaturing methods, four protein denaturing methods, namely, heating (95℃), alcohol(50 % 2,2,2-Trifluoroethanol, TFE), neutral material (6M urea) and salt (6M guanidine-HCl) disruption, were applied to six standard glycoproteins. Among the four methods, the use of TFE (50%) denaturation provided the most glycopeptides identified (from 1.2 to 1.4 folds) and highest glycopeptide signals (from 1.5 to 32.4 folds) in the mass spectra. Of the pre-digestion conditions tested, various concentrations of ([DTT]: [IAM]: [2nd DTT], in mM) in the ratio of 5:25:25 (condition A), 5:45:45 (condition B), 10: 25:25 (condition C), and 10:45:45 (condition D) were compared on six standard glycoproteins. The condition (A) was shown to be the most appropriate pre-digestion condition for glycopeptides identification. Hence, for the standard glycoprotein mixtures, the 50% TFE denaturing method followed by pre-digestion condition consisting of 5 mM DTT, 25 mM IAM, and 25 mM 2nd DTT was employed to further enrich the glycopeptides.
In the second part of the thesis, we aim to develop a sequential glycopeptide enrichment method by hyphened purification with non-specific HILIC (Hydrophilic interaction liquid chromatography) and sialic acid targeted TiO2 StageTips methods. On the result of six standard glycoprotein mixture, sequential enrichment by TiO2-HILIC could enrich more glycopeptides (47) compared to separate single HILIC (39 glycopeptides) and TiO2 (20 glycopeptides) enrichment method. Finally, the efficacies of the sequential enrichment approaches were evaluated using HeLa cell membrane proteins. Sequential HILIC-TiO2 (373.3 ± 37.9) and TiO2-HILIC (398.7 ± 19.5) both also efficiently enriched more glycopeptides compared to separate single HILIC (302.7 ± 41.5) and TiO2 (309.7 ± 29.2) in HeLa membrane proteins. By TiO2-HILIC, the coverage of N-glycoproteome analysis was increase 1.2 to 1.4-folds, hereby increasing the efficiency of enriching more intact glycopeptides from cell. Thus, in both standard glycoprotein mixtures and HeLa cell membrane proteins, glycoproteome identification could be expanded by our sequential glycopeptide enrichment approach.
In conclusion, thus study provided an integrated protocol with optimized denaturation and the pre-digestion conditions and sequential glycopeptides enrichment method to increase glycopeptides identification coverage while preserve intact glycan and peptide structures which were valuable information for glycoproteome research. In the future, it is expected that this rational pipeline can be utilize to benefit in-depth analysis of glycoproteome.

並列關鍵字

Glycoprotein digestion ； glycopeptide sequential enrichment ； glycoproteome

參考文獻

[1] D.K. Hsu, H.-Y. Chen, F.-T. Liu, Galectin-3 regulates T-cell functions, Immunol Rev, 230 (2009) 114-127.

[2] G. Gupta, S. Sinha, N. Mitra, A. Surolia, Probing into the role of conserved N-glycosylation sites in the Tyrosinase glycoprotein family, Glycoconj J, 26 (2009) 691-695.

[3] B.B.H. Wulff, A. Chakrabarti, D.A. Jones, Recognitional Specificity and Evolution in the Tomato–Cladosporium fulvum Pathosystem, Mol Plant Microbe Interact, 22 (2009) 1191-1202.

[4] D.W.C. Danni L Meany, Aberrant glycosylation associated with enzymes as cancer biomarkers, Clin Proteomics, 8 (2011) 7.

[6] H.H. Freeze, E.A. Eklund, B.G. Ng, M.C. Patterson, Neurology of inherited glycosylation disorders, Lancet Neurol, 11 (2012) 453-466.