| 摘要: | 在多種蛋白質轉譯後修飾中，N-連接醣基化和磷酸化在生物功能上扮演著相當重要的功能，像是細胞訊息傳導、蛋白質的穩定性和移動。然而，由於固有的修飾異質性和質譜游離效率的相互抑制，因此發展醣基化和磷酸化胜肽純化的方法是必須的。雖然已有不少純化方法針對這兩種修飾，但多數方法都是針對兩種轉譯後修飾個別分析，近來，磁性奈米粒由於其能快速分離的能力和表面易於進行修飾而被廣泛地運用。
在此論文工作中，基於親水性作用力、活化的胺基和兩性離子官能基被修飾在磁性奈米粒子(由Elmer Austria Jr.合成並提供)的表面上用以純化出磷酸化和醣基化胜肽。藉由實驗條件的最佳化後，包含溶液組成、pH值、奈米粒子使用量和有機相比例，在醣基化胜肽的純化中，0.1%甲酸/95%乙腈(調整至pH7)作為活化緩衝液，而1%磷酸/80%乙腈和0.1%氨水/80%乙腈則作為洗滌緩衝液，最後由2%氨水/50%乙腈作為析出緩衝液。另外在磷酸化胜肽的純化方面，0.1%甲酸/95%乙腈(調整至pH 3) 作為活化緩衝液，而6%乙酸/85%乙腈
則作為洗滌緩衝液，最後由5%氨水/50%乙腈作為析出緩衝液。在磁性奈米粒子用量方面，醣基化/磷酸化胜肽的純化皆是以2微克胜肽比上100微克磁性奈米粒子的用量為最佳化條件。此外，藉由配合調整酸鹼值的策略，可在為α-和 β-酪蛋白(α-, β-casein)(磷酸化胜肽)以及辣根過氧化物酶(醣基化胜肽)1:1的混和胜肽中，在使用基質輔助雷射脫附游離飛行質譜儀的分析下，有效地在pH值3的環境下萃取分離出10條磷酸化胜肽以及在pH 7的環境下得到9條醣基化胜肽，分離效率分別為96.3% 和81.1%。
為增加樣品複雜度，在樣品中添加了牛血清白蛋白。在α-和 β-酪蛋白、辣根過氧化物酶和牛血清白蛋白1:1:1:1的混合胜肽樣品中，8條磷酸化胜肽和7條醣基化胜肽能分別在不同的沖提順序下被鑑定到。當提高了牛血清白蛋白的比例時，仍然有8條磷酸化胜肽被鑑定，但卻分別只有4條和5條醣基化胜肽在1:1:1:5和1:1:1:10的比例中被鑑定到。
接著將方法應用在PC9非小型細胞肺癌細胞(non-small-cell lung cancer cell)中，藉由直接純化醣基化胜肽的方式，鑑定到306條醣胜肽和35種醣型，此外，在去醣基化的作用下，1068條去醯胺基的胜肽被鑑定到。另一方面，246條磷酸化胜肽(231條單一磷酸化,15條多重磷酸化)則在直接純化磷酸化胜肽被偵測到。在起始200微克的PC9膜胜肽，經過除鹽和順序性純化策略下，246條磷酸化胜肽(209條單一磷酸化, 27條多重磷酸化),432條去醯胺基的胜肽以及243條醣胜肽(24種不同的醣型)皆在液相層析串聯三合一式質譜儀的分析下被鑑定。
這些結果都顯示，此種兩性親水作用磁性奈米粒子在針對大範圍的醣基化和磷酸化胜肽的純化和鑑定上，都具有相當高的發展潛力。
;Among various protein post-translational modification (PTM), N-linked glycosylation and phosphorylation are among the most ubiquitous PTMs and play important roles in many biological functions, such as signal transduction, protein stability, and mobility. However, due to inherent heterogeneity of modification site and structure and ion suppression effect from free peptides during mass spectrometry analysis, it is necessary to develop enrichment strategy prior to mass spectrometry (MS) analysis. Though there are many enrichment methods developed for these two PTMs, both glycopeptides and phosphopeptides require different enrichment approaches.
Recently, magnetic nanoparticles (MNPs) have been widely used in biomedical applications mainly because of its capability for surface functionalization and fast separation by magnetic extraction. In this thesis, based on the hydrophilic interaction, active NH2- and zwitterionic functional groups were anchored (provided by Elmer Austria Jr.) on the surface of the magnetic nanoparticles (amine/zwitterionic-hydrophilic interaction based magnetic nanoparticle, NH2/ZIC-HILIC MNP) for simultaneous enrichment of glycopeptides and phosphopeptides. Four main parts were optimized in this work, including buffer composition, pH value, organic composition and particle amount. In glycopeptide enrichment, 0.1% formic acid (FA) with 95% acetonitrile (ACN) (adjusted to pH 7) were used as incubation buffer; 1% phosphoric acid with 80% ACN and 0.1% ammonium hydroxide (NH4OH) with 80% ACN were applied as washing buffer and 2% NH4OH/50% ACN for elution buffer respectively.
On the other hand, 0.1% FA/95% ACN (adjusted to pH 3) were used as incubation buffer. 6% acetic acid (AA) with 85% ACN and 5% NH4OH/50% ACN were applied as washing buffer and elution buffer respectively. 2μg peptides to 100μg nanoparticles was the optimized usage ratio for both glycopeptide and phosphopeptide enrichment.
By pH modulation strategy, effective separation of both glycopeptides in pH=7 and phosphopeptides in pH=3 can be achieved from a 1:1:1 mixture of α- and β-casein (phosphoproteins) and HRP (glycoprotein), 10 phosphopeptides in phosphopeptide elute fraction and 9 glycopeptides in glycopeptide elute fraction were identified in MALDI-TOF analysis with 96.3% and 81.1% separation efficiency for phosphopeptides and glycopeptides respectively.
To increase the complexity of the sample, bovine serum albumin (BSA) protein was added into the sample. In ratio of HRP : α-casein : β-casein : BSA= 1:1:1:1, 8 phosphopeptides still can be detected and dominate in the phosphopeptide elute fraction; in the glycopeptide elute fraction, though there are still 7 glycopeptides can be detected, When BSA ratio go higher up to 1:1:1:10, no significant influence in phosphopeptide fraction, although still 8 phosphopeptides were still detected, however, there are only 4 and 5 glycopeptides were detected in ratio 1:1:1:5 and 1:1:1:10 respectively.
Further application to the proteome level was demonstrated in PC9 cell line, a non-small cell lung cancer (NSCLC) cell line. By direct glycopeptide enrichment, 306 unique intact glycopeptides and 35 unique glycan structure were identified. With deglycosylation treatment by PNGase F, 1068 deamidated peptides with consensus motif NxS/T/V/C (x belongs to any amino acids but not proline) were detected. On the other hand, 246 phosphopeptides (231 mono-phosphorylated and 15 multi-phosphorylated) were identified in direct phosphopeptides. With initial 200μg membrane peptide from PC9 cell and followed by desalting and sequential enrichment strategy, 236 phosphopeptides (209 mono-phosphorylated and 27 multi-phosphorylated) were detected in phosphopeptide elute fraction. 243 unique intact glycopeptides (24 unique glycan structure) and 432 deamidated peptides with glycosylation motif (after deglycosylation) were identified in glycopeptide elute fraction combined with LC-MS/MS analysis using orbitrap fusion lumos mass spectrometry. These results further demonstrate that NH2/ZIC HILIC MNP can be a highly-potential method for characterization of both phosphopeptides and glycopeptides. |
