Separation of Glycolipids/Sphingolipids from Glycerophospholipids on TiO2 Coating in Aprotic Solvent for Rapid Comprehensive Lipidomic Analysis with Liquid Microjunction Surface Sampling-Mass Spectrometry
Separation of Glycolipids/Sphingolipids from Glycerophospholipids on TiO2 Coating in Aprotic Solvent for Rapid Comprehensive Lipidomic Analysis with Liquid Microjunction Surface Sampling-Mass Spectrometry
复制标题
在非质子溶剂中从 TiO2 涂层上的甘油磷脂中分离糖脂/鞘脂，通过液体微结表面采样-质谱进行快速综合脂质组学分析
DOI：
10.1021/acs.analchem.0c01870
复制
发表时间：
2020-08-18
影响因子：
7.4
通讯作者：
Zhang, Zhimin
中科院分区：
文献类型：
--
作者：
Huang, Zehui;Wu, Qian;Zhang, Zhimin
在通过直接质谱（MS）进行的脂质组学分析中，具有高电离性的高丰度脂质（如甘油磷脂）会对具有差电离性和低丰度的脂质（如糖脂、鞘脂或甘油酯）造成离子抑制，这在很大程度上限制了脂质组学的检测覆盖率。在这项工作中，二氧化钛为基础的液体微交界面采样（LMJSS）与MS耦合用于分离生物样品中的甘油酯，磷脂和糖脂/鞘脂和快速分析的脂质在不同类别的高lipidome覆盖。我们发现，在非水非质子溶剂中，具有糖基或鞘氨醇基团的脂质可以选择性地与具有磷酸基团的脂质分离（选择性> 10）后，通过调节溶剂组成共富集在TiO 2上。因此，开发了一种选择性多步提取方法，通过将生物样品加载在中性非质子溶剂中的TiO2载玻片上，并通过LMJSS探针从TiO2载玻片上依次洗脱纯乙腈中的甘油酯、6%氨水-94%乙腈（v/v）中的甘油磷脂和5%甲酸-95%甲醇（v/v）中的糖脂/鞘脂。来自TiO 2载玻片的每个样品由LMJSS直接递送至MS进行分析。三个解吸步骤的总检测时间将控制在3 min内。使用脂质标准品（包括基质效应）评价每种脂质类别的方法性能（107 - 128%），RSD（0.4 - 16%），线性（0.98 - 0.99）、检测限（5 - 3000 ng/mL）、吸附平衡常数（10（2）-10（4））和吸附容量（1 - 38 μ g/mm（2））。最后，将该方法应用于血浆和脑组织的脂质组学分析，并与直接输注质谱法进行比较。结果表明，与直接输注质谱法相比，新方法在血浆和脑提取物中鉴定出的鞘脂/糖脂和甘油磷脂/甘油酯分别增加了（2 - 5）倍和40 - 50倍。采用标准加入校正法对检测到的脂类进行定量，并与传统的LC-MS方法进行了绝对定量结果的验证（相关系数> 0.98，相关线斜率= 0.87-1.05）。
In lipidomic analysis by direct mass spectrometry (MS), high abundance lipids with high ionizability (such as glycerophospholipids) would cause ion suppression to lipids with poor ionizability and low abundance (such as glycolipids, sphingolipids, or glycerides), which largely limits the detection coverage for lipidomics. In this work, TiO2-based liquid micro-junction surface sampling (LMJSS) coupled with MS was used for separation of glycerides, phospholipids and glycolipids/sphingolipids in biological samples and rapid analysis of lipids in different classes with high lipidome coverage. We found that, in nonaqueous aprotic solvents, lipids with a glycosyl or sphingosine group could be selectively separated from lipids with a phosphate group (selectivity >10) after being coenriched on TiO2 by tuning the solvent composition. Accordingly, a selective multistep extraction method was developed by loading the biosamples on TiO2 slides in neutral aprotic solvent, and sequentially eluting glycerides in pure acetonitrile, glycerophospholipids in 6% ammonia-94% acetonitrile (v/v) and glycolipids/sphingolipids in 5% formic acid-95% methanol (v/v) by LMJSS probe from TiO2 slide. Each eluate from TiO2 slide was directly delivered by LMJSS to MS for analysis. The total detection time with three desorption steps would be controlled in 3 min. The method performance for each lipid class was evaluated using lipid standards, including matrix effects (107-128%), RSDs (0.4-16%), linearity (0.98-0.99), detection limits (5-3000 ng/mL), the adsorption equilibrium constants (10(2)-10(4)) and adsorption capacity (1-38 mu g/mm(2)) of TiO2 coated slides to lipids. Finally, the TiO2-based-LMJSS-MS method was applied to lipidomic analysis for blood plasma and brain tissue, and compared with direct infusion MS. Results showed that (2-5)-fold more sphingolipids/glycolipids and 40-50 more glycerophospholipids/glycerides were identified in both plasma and brain extract by the new method comparing with direct infusion MS method. Detected lipids were quantified with standard addition calibration method, and the absolute quantitation results measured by TiO2-based-LMJSS-MS were verified with that by the traditional LC-MS method (correlation coefficient >0.98, slope of correlation line = 0.87-1.05).