| 摘要: | 可逆性腦血管收縮症候群 (Reversible cerebral vasoconstriction syndromes, RCVS)是一種中樞神經系統疾病，有著可逆性及多發性腦動脈收縮、雷擊式頭痛和較少見的多發性神經缺陷等症狀，但目前對此疾病的診斷以及蛋白質分子機制還未全盤了解，所以我們計畫分析腦脊髓液中蛋白質的組成和表現量，並找到可以用於診斷和治療的新型蛋白質生物標記，在之前的研究中，我們使用了數據依賴型採集模式質譜 (data dependent acquisition mass spectrometry, DDA-MS) 分析了40個腦脊髓液樣品，最後發現了18個生物標記候選蛋白。
我們使用了最新的且再現性更高的方法非數據依賴型採集模式質譜 (data independent acquisition mass spectrometry, DIA-MS)來分析腦脊髓液樣品，首先我們要先建立DIA-MS方法，我們混合了16個腦脊髓液物樣品並使用層析技術，構建了含有958個蛋白質的圖普庫，其中含有前驅離子荷質比、碎片離子荷質比、碎片離子相對強度和相對滯留時間（iRT）等信息，未使用層析技術所建立的圖譜庫只有475個蛋白質相比之下使用層析技術所見的圖普庫提供了更高的蛋白質鑑定覆蓋率，再和擁有1647蛋白質個的公開腦脊髓液圖譜庫相比，我們的圖譜庫在測試的樣品中可以提供744個蛋白質鑑定結果，而來自公開腦脊髓液圖譜庫則鑑定到612個蛋白質，是因為我們自己建立的脊髓液圖譜庫對於自己的腦脊髓液樣品更相近，我們還使用20個合成肽，針對18個候選蛋白進一步增加圖譜庫的資訊來幫助我們更完整地鑑定到18個候選蛋白，在測試的樣品中更進一步針對18個候選蛋白增加了5個肽和1個蛋白的鑑定。
我們應用了新開發的DIA-MS方法定量比較12個控制和13個病人的腦脊髓液蛋白質組，根據DIA-MS數據我們利用直接建立圖譜庫的策略將48種蛋白質增加到RCVS圖譜庫，再用這個圖譜庫和DIA-MS數據進行比對鑑定，並通過積分碎片離子圖譜組合峰面積進行定量，最後在25個腦脊髓液樣品中鑑定到762個蛋白質，和之前利用數據依賴性採集質譜（DDA-MS）方法分析40份腦脊髓液樣品的結果相比，多鑑定到了333個蛋白質。比對人類蛋白質組圖譜，DIA-MS方法提供更多鑑定數以及更多腦源性蛋白質。進一步定量顯示在雙尾t檢驗後，病人和控制個體之間117個蛋白質差異表達（p <0.05）。使用IPA和DAVID對117個差異表達蛋白進行的生物富集分析富集了與炎症相關的功能，此外鑑定了18個候選蛋白中的13個，鑑定到的這13個候選蛋白中有3個包括PRNP，THBS4和CA1和之前的結果也是一樣的也是差異表達蛋白。 PRNP在先前和目前的結果中具有增加的趨勢，並且與朊病毒疾病，對氧化應激的反應和腦源性蛋白質相關。這三種蛋白質可以作為進一步研究的相容性候選部分。
;Reversible cerebral vasoconstriction syndromes (RCVS) are a group of central nerves system disorders that show reversible multifocal cerebral arteries narrowing, thunderclap headache and less commonly focal neurologic deficits. Due to limited knowledge about RCVS, the molecule mechanism are still unknown and the diagnosis tools remain to be developed. Revealing the cerebrospinal fluid (CSF) protein compositions and expression may provide an opportunity to advance our knowledge about the role of proteins in neurological disorders and to identify novel protein biomarkers for diagnosis and treatment purposes. In our previous study, we have analyzed 40 CSF sample with data dependent acquisition mass spectrometry (DDA-MS) method and revealed 18 candidate proteins.
In this thesis, we devolved a data independent acquisition mass spectrometry (DIA-MS) method for analyzing the CSF proteome. In the first part, we have generated a RCVS proteome mass spectral library by reversed phase stagetip fractionation strategies using DDA method together with LC-MS/MS analysis. Using CSF mixture sample (N=16), a library with precursor m/z, fragment m/z, fragment relative intensity, and relative retention time (iRT) information from 958 proteins were constructed, which provide higher proteome coverage compare to 475 proteins identified without fractionation method. Compared to the public CSF spectral library which had 1647 protein, our library can provide 744 proteins identification result which is more than 612 proteins from public CSF library, because of the own library is more specific to own CSF samples. We further optimized the DIA approach for quantitative proteomics by synthetic peptides and ALD4, and 5 peptides and 1 proteins were quantified additionally. The developed DIA-MS method was applied to quantitatively compare the CSF proteome of 12 non-RCVS controls and 13 RCVS patients. All the DIA-MS data were processed by generated DIA based spectral library, spectral library based identification and quantified by integrating the peak area of fragment alignment. Totally, we increased 48 proteins into the RCVS library and identified 762 proteins in 25 CSF samples, which revealed more RCVS proteome composition of 333 proteins from previous results of 40 CSF samples by data dependent acquisition mass spectrometry (DDA-MS) method. The DIA-MS method provides more identification and also more brain originated proteins based on human proteome map.
Further quantitation show that 117 proteins are differentially expressed between the RCVS and control individuals after two tailed t test (p<0.05). The biological enrichment analysis using IPA and DAVID of 117 differentially expressed proteins (DEPs) enrich functions related to inflammation, immune response, prothrombin, leukocyte and superoxide radicals degradation. 13 of 18 candidate proteins which are selected from previous results of 40 CSF samples are identified, and 3 of 13 identified candidate proteins including PRNP, THBS4 and CA1 are also DEPs. PRNP has the increase trend in previous and currently results and is associated with prion disease, response to oxidative stress and brain originated protein. These three proteins can be the confidence candidate portions for further investigation. |
