Comprehensive Analytical Strategy for Biomarker Identification based on Liquid Chromatography Coupled to Mass Spectrometry and New Candidate Confirmation Tools
Comprehensive Analytical Strategy for Biomarker Identification based on Liquid Chromatography Coupled to Mass Spectrometry and New Candidate Confirmation Tools
复制标题
DOI：
10.1021/ac901087t
复制
发表时间：
2009-09-15
影响因子：
7.4
通讯作者：
Hopfgartner, Gerard
中科院分区：
文献类型：
--
作者：
Mohamed, Rayane;Varesio, Emmanuel;Hopfgartner, Gerard
介绍了一种用于生物流体中低质量生物标志物分子的LC-MS（/MS）综合分析平台。通过优化反相色谱柱和亲水相互作用色谱柱的色谱条件，建立了两种互补保留机制。LC分离是通过电喷雾电离在正极模式下耦合到质谱。这种策略使我们能够正确地保留和分离疏水和极性分析物。为此目的，用可能存在于生物流体中的38种特征明确的化合物的混合物生成了人工模型研究样品。这组化合物被用作标准的水溶液混合物或以不同浓度水平加标到尿液中，以研究LCMS（/MS）平台检测生物样品变化的能力。采用主成分分析法对数据进行无监督分析，然后进行主成分变量分组，寻找相关变量。该工具允许我们区分三个主要组，其变量属于(a)背景离子（存在于所有类型的样品中），(b)将尿液样品与水标准样品和空白样品区分开来的离子，(c)与加标化合物相关的离子。对这些基团的解释使我们能够识别和消除同位素、加合物、碎片等，并生成一个简化的m/z候选列表。然后将该列表提交给原型MZSearcher软件工具，该工具同时搜索从代谢组学数据库（例如，KEGG， HMDB等）中提取的潜在代谢物列表，以提出生物标记候选物。这些候选物的结构确认是通过离线收集馏分，然后进行纳米电喷雾输注，为光谱数据库查询提供高质量的MS/MS数据。
A comprehensive analytical LC-MS(/MS) platform for low weight biomarkers molecule in biological fluids is described. Two complementary retention mechanisms were used in HPLC by optimizing the chromatographic conditions for a reversed-phase column and a hydrophilic interaction chromatography column. LC separation was coupled to mass spectrometry by using an electrospray ionization operating in positive polarity mode. This strategy enables us to correctly retain and separate hydrophobic as well as polar analytes. For that purpose artificial model study samples were generated with a mixture of 38 well characterized compounds likely to be present in biofluids. The set of compounds was used as a standard aqueous mixture or was spiked into urine at different concentration levels to investigate the capability of the LCMS(/MS) platform to detect variations across biological samples. Unsupervised data analysis by principal component analysis was performed and followed by principal component variable grouping to find correlated variables. This tool allows us to distinguish three main groups whose variables belong to (a) background ions (found in all type of samples), (b) ions distinguishing urine samples from aqueous standard and blank samples, (c) ions related to the spiked compounds. Interpretation of these groups allows us to identify and eliminate isotopes, adducts, fragments, etc. and to generate a reduced list of m/z candidates. This list is then submitted to the prototype MZSearcher software tool which simultaneously searches several lists of potential metabolites extracted from metdbolomics databases (e.g., KEGG, HMDB, etc) to propose biomarker candidates. Structural confirmation of these candidates was done off-line by fraction collection followed by nanoelectrospray infusion to provide high quality MS/MS data for spectral database queries.