以液相層析四極柱質譜儀開發三酸甘油脂之分析方法並分析2型肺泡上皮細胞及高三酸甘油脂血症患者血漿

Abstract

三酸甘油脂 (Triglyceride, TG) 做為人體內脂肪運輸與能量儲存的主要分子。TG大於 150 mg/dL是目前診斷高三酸甘油脂血症 (Hypertriglyceridemia, HTG) 的標準，且被認為有較高風險罹患急性胰臟炎 (Acute pancreatitis, AP)、二型糖尿病 (Type 2 diabetes mellitus, T2DM) 與心血管疾病 (Cardiovascular disease, CVD)。在臨床上，三酸甘油脂僅被用以當作診斷標準，且因支鏈脂肪酸 (Fatty acid, FA) 所產生的複雜結構導致分析困難而使不同支鏈組成所造成的潛在影響尚未明瞭。本研究致力於開發以質譜為主要分析工具之偽標的分析平台以促進三酸甘油脂之研究。 於第一章節中，我們開發使用液相層析串聯質譜系統的三酸甘油脂分析流程與平台。我們最佳化樣品配置流程以完整萃取樣品中的三酸甘油脂並降低基質效應 (Metric effect)；同時最佳化液相層析參數以獲得良好分離效果。全圖譜掃描 (Full scan mode) 被用以取得各式樣品間不同的三酸甘油脂種類列表，並藉由子離子掃描 (Product ion scan, PIS) 以獲得個別分子中的支鏈組成。這些子母離子可組合成多反應監測 (Multiple reaction monitoring, MRM) 離子對 (Transition)。層析結果顯示甲醇 (Methanol, MeOH)與異丙醇 (Isopropyl alcohol, IPA)的混合溶液可完整溶解樣品且Kinetex® 碳18管柱(2.6µm; 50 x 2.1mm) 提供良好層析圖譜。在現有的層析條件下，所有三酸甘油脂可依其碳當量 (Equivalent carbon number, ECN) 序列析出。本實驗挑選兩種樣品進行平台測試，分別為剪除脂肪酸合成蛋白 (Fatty acid synthase, FASN) 基因之鼠肺泡上皮細胞2型 (Murine type 2 alveolar epithelial cells, AEC2) 細胞株－MLE-12細胞與高三酸甘油脂血症患者血漿，上述兩者分別代表不同種類之樣品。在兩種不同的樣品底下我們偵測到不同的三酸甘油脂種類列表與不同的支鏈組成，發現對於三酸甘油脂組成分析的研究，針對樣品先進行TG種類篩選的重要性。 第二章節中，我們分析頑固性與非頑固性高三酸甘油脂血症患者血漿，並觀察到兩者血漿呈現特殊差異趨勢。其中我們將嚴重三酸甘油脂血症患者依其服用降血脂藥後之反應分成頑固性與非頑固性患者 (頑固性患者服藥後體內三酸甘油脂濃度持續高於150 mg/dL)。四種與脂蛋白酶和脂蛋白代謝相關基因：LPL、LMF1、ApoA1、ApoA5，亦被收錄以評測基因對頑固性症狀之影響。 140份血漿樣品中，80份來自頑固性患者、20份來自非頑固性患者、40份來自健康對照組。其血漿中三酸甘油脂含量分別為：784.4 ± 818.2 mg/dL、116.8 ± 28.1 mg/dL、和 84.9 ± 38.5 mg/dL。偏最小平方判別分析 (Partial least squares-discriminant analysis, PLS-DA) 顯示頑固性高三酸甘油脂血症患者有較低比例的高不飽和度三酸甘油脂。在經過用藥紀錄的校正後，17 個三酸甘油脂中的支鏈在三酸甘油脂血症患者血漿中有顯著性的較低比例，且其中有多個omega-6 類型的脂肪酸。再者，若將TG 19:1_34:2 (OR: 0.971; 95% CI: 0.94-0.99) 和TG 18:2_38:5 (OR: 0.998; 95% CI: 0.99-1.00) 兩個三酸甘油脂佐以用藥紀錄可有效區分高三酸甘油脂血症中的頑固性症狀患者，其曲線下面積 (Area under the receiver operating characteristic, AUROC) 可從0.826上升到0.944。 本研究開發一個高效的偽標的分析平台並藉由兩種樣品模型展示平台之應用性。我們同時發現頑固性高三酸甘油脂血症之患者體內有較低比例含多不飽和支鏈的三酸甘油脂。這個發現有助於在臨床鑑定潛在的頑固性症狀患者。然而，僅作為一個探索性實驗 (Pilot test)，本現象仍須更多的關注與研究以鑑定頑固性症狀背後不同三酸甘油脂組成之影響。

Triglyceride (TG) is essential in transporting dietary fat and storing energy. Hypertriglyceridemia (HTG) is a metabolic disease characterized by fasting TG levels above 150 mg/dL, and it is believed to be related to various diseases such as acute pancreatitis (AP), type 2 diabetes mellitus (T2DM), and cardiovascular diseases (CVD). Conventionally, TGs are only regarded as a whole in clinical diagnosis; the underlying role of different TG compositions remains unclear. Nonetheless, the structural complexity of three fatty acids (FA) chains makes TGs analysis challenging. The study aims to develop an LC-MS based pseudo-target TG analytical platform to facilitate TG analysis. In the first chapter, the study aimed to develop a TG analytical platform using LC-MS systems. The sample preparation protocol was optimized to thoroughly extract the sample and alleviate the metric effect, and LC parameters such as analytical column, gradient, and flow rate were optimized to provide good separation. Targeted TG species lists were acquired by full scan mode, while the product ion scan (PIS) was used to identify fatty acyl chains. The results were transformed into multiple reaction monitoring (MRM) transitions. The LC chromatogram suggested that the mixed solvent of methanol (MeOH) and isopropyl alcohol (IPA) could completely dissolve the extracted samples, and Kinetex® C18 Column (2.6µm; 50 x 2.1mm) provided better separation efficacy. Moreover, the optimized gradient profile could provide favorable linear correlations between equivalent carbon number (ECN) and retention time. Two demonstration samples, murine type 2 alveolar epithelial cells (AEC2) line, MLE-12, with Fasn gene deletion and hypertriglyceridemia (HTG) plasma, were chosen to display the capability of the platform. The finding of different TG species lists strengthened the importance of TG target list acquisition. The second chapter investigated TG compositions of severe HTG patients with and without refractory. The refractory was diagnosed as the patients had TG levels above 150 mg/dL after medical treatment. Four coding regions related to lipoprotein lipase (LPL) and apolipoprotein functionality, including LPL, LMF1, ApoA1, and ApoA5 genes, were sequenced in HTG patients to evaluate the genetic effects. Of 140 enrolled subjects, there were 80 refractory HTG (rHTG) patients, 20 non-refractory HTG (nrHTG) patients, and 40 normal controls. The TG levels of three groups are 784.4 ± 818.2 mg/dL, 116.8 ± 28.1 mg/dL, and 84.9 ± 38.5 mg/dL, respectively. The PLS-DA plot revealed that the rHTG patients had a lower proportion of TG with higher unsaturated degrees. In addition, 17 TG compositions showed a significantly lower proportion in rHTG after adjusting drug use, and the omega-6 FAs were commonly listed within these TG compositions. After incorporating TG biomarkers of TG 19:1_34:2 (OR: 0.971; 95% CI: 0.94-0.99) and TG 18:2_38:5 (OR: 0.998; 95% CI: 0.99-1.00) with drug use to identify patients with rHTG, the area under the receiver operating characteristic (AUROC) increased from 0.826 to 0.944. This study developed an effective TG analytical platform with the demonstration of two model samples. We also presented that rHTG patients processed a lower proportion of TG compositions with poly-unsaturated FA. The results may help differentiate the potential refractory severe HTG patients in clinical practice. Nevertheless, as a pilot study, more studies are needed to elucidate the mechanism of different TG compositions with refractory.