發展個人化腦脊髓液蛋白體定量策略

Abstract

腦脊髓液為臨床上與中樞神經系統關係最密切的體液樣本，分析期成份能提供關於神經系統疾病與損傷的病理機制某些獨特的訊息。但是,腦脊髓液裡蛋白質的濃度差異過大、低濃度、高含量蛋白質的干擾, 都侷限了低含量的生物標靶蛋白之開發。為了要作更有效的個人化腦脊髓液的分析，我們結合以奈米探針為基礎的去高含量蛋白技術及同位素標記法(iTRAQ)或免標定定量法(Label-free quantitation) 開發一個具高靈敏性及高再現性的個人化腦脊髓液蛋白體分析方法。在腦脊髓液分析結果顯示，帶有albumin抗體的奈米粒子與其他市售去除高含量蛋白的試劑相比,奈米粒子可以更有效的達到去除腦脊髓液裡的高含量蛋白，並鑑定到1.4倍的蛋白質。在結合同位素標定定量法及免標定定量法的部分，兩種定量策略都提供相近的準確度(同位素標定法的平均值為0.03，免標定定量法為-0.066, log2 尺標)以及誤差範圍(同位素標定法標準偏差為0.27，免標定定量法為0.3, log2尺標)。 在論文的第二部分中，我們將這兩種定量策略應用在分析體顯性腦動脈血管病變合併皮質下腦梗塞及腦白質病變(CADASIL)病患與年齡相對應的正常人，及經過兩次手術治療的脊椎損傷病患，以了解蛋白質在腦脊髓液裡的表現量。在CADASIL 病患中，我們定量了311個蛋白質，並且有39個蛋白質在60%以上的病患中皆有異常表現量。在這些變異蛋白質中, Amyloid precursor protein, Apolipoprotein E, Angiotensinogen, Alpha-1-acid glycoprotein 1, and Alpha-1-acid glycoprotein 2 都曾被直接或間接地報導與CADASIL的致病基因，NOTCH3基因有關。更進一步的，我們也利用西方點墨法驗證Amyloid precursor protein在病人的腦脊髓液中Amyloid precursor protein的含量是下降的。 在脊椎損傷的部分，從7對脊椎損傷病患的脊髓液中，我們找到了233個蛋白質且其中221 可被定量。目前分析結果得到在7組脊椎損傷病患中，因經過FGF 治療方式所導致的不同表現量蛋白比例為19-45%。31個蛋白質在4個以上的經過二次手術的脊椎損傷病患中皆有異常表現量。在此結果中找到四種蛋白質Apolipoproteins, Tresferrin, Tubulin, Zinc finger proteins 曾被報導與脊椎損傷有關。 在此篇論文中，我們提出的腦脊髓液定量分析平台能有效地尋找針對CADASIL 疾病診斷及追蹤脊椎損傷治療效果的標靶蛋白質。

Cerebrospinal fluid (CSF) is an important specimen to accurately reflect pathological processes and provides an ideal window for insights into mechanisms and detection of biochemical changes, such as protein biomarker associated with neurodegenerative disorders and spinal cord injury. However, the wide dynamic range, low protein concentration, and presence of high abundant albumin in CSF pose challenges for comprehensive proteome identification. To facilitate the identification of disease biomarker candidates, we integrated nanoprobe-based albumin depletion technology with iTRAQ or label-free quantitation for quantitative analysis of personalized CSF proteome. The anti-albumin immobilized magnetic nanoparticles (anti-albumin@MNPs) showed 1.4-fold increase in the number of protein identification than commercially available kits to deplete the high abundant albumin in CSF. Combining with iTRAQ or Label-free quantitation, the two platforms provide similar accuracy (mean = 0.003 and -0.066 for iTRAQ and label-free quantitation, respectively in log2 scale) and reproducibility (standard deviation = 0.27 and 0.3 for iTRAQ and label-free quantitation, respectively in log2 scale) In the second part of thesis, we applied these two strategies to analyze the expression levels of CSF proteome for two diseases: (1) patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) in comparison with age-matched controls and (2) patients with spinal cord injury (SCI) after first and second FGF treatments. A total of 311 proteins were quantified and 39 proteins showed differential expression in more than 60% of CADASIL patients. Among them, amyloid precursor protein, Apolipoprotein E, angiotensinogen, Alpha-1-acid glycoprotein 1, and Alpha-1-acid glycoprotein 2 have been reported to be related to NOTCH3, the disease-causive mutant gene in CADASIL. Further validation using western blot analysis confirmed the down-regulation of amyloid precursor protein in CSF samples from CADASIL patients. For the CSF analysis for patients with SCI, 233 proteins were identified and 221 proteins were quantified in 7 paired SCI samples from patients after first and second FGF treatments. The preliminary analysis revealed that the FGF treatment induced differential expression of 19%-45% of quantified proteins for the 7 paired SCI patients. Among these proteins, 31 proteins showed common differential expression in more than 4 of SCI patients before and after operation. Four proteins, including Apolipoproteins, tresferrin, tubulin, Zinc finger proteins, had been reported to relate to spinal cord injury. Our method demonstrated the potential on the identification of potential biomarker for CADASIL disease and the treatment efficiency of SCI.