Open Search of Peptide Glycation Products from Tandem Mass Spectra

Identification
of chemically modified peptides in mass spectrometry
(MS)-based glycation studies is a crucial yet challenging task. There
is a need to establish a mode for matching tandem mass spectrometry
(MS/MS) data, allowing for both known and unknown peptide glycation
modifications. We present an open search approach that uses classic
and modified peptide fragment ions. The latter are shifted by the
mass delta of the modification. Both provide key structural information
that can be used to assess the peptide core structure of the glycation
product. We also leverage redundant neutral losses from the modification
side chain, introducing a third ion class for matching referred to
as characteristic fragment ions. We demonstrate that peptide glycation
product MS/MS spectra contain multidimensional information and that
most often, more than half of the spectral information is ignored
if no attempt is made to use a multi-step matching algorithm. Compared
to regular and/or modified peptide ion matching, our triple-ion strategy
significantly increased the median interpretable fraction of the glycation
product MS/MS spectra. For reference, we apply our approach for Amadori
product characterization and identify all established diagnostic ions
automatically. We further show how this method effectively applies
the open search concept and allows for optimized elucidation of unknown
structures by presenting two hitherto undescribed peptide glycation
modifications with a delta mass of 102.0311 and 268.1768 Da. We characterize
their fragmentation signature by integration with isotopically labeled
glycation products, which provides high validity for non-targeted
structure identification.

在基于质谱（MS）的糖基化研究过程中，对化学修饰肽的识别是一项至关重要的任务，但同时也充满挑战。有必要建立一种匹配串联质谱（MS/MS）数据的方法，该方法能够识别已知及未知的肽糖基化修饰。本研究提出了一种开放搜索方法，该方法利用经典及经过修饰的肽碎片离子。后者通过修饰的质量变化（mass delta）进行偏移。这两种方法均提供了关键的结构信息，可用于评估糖基化产物的肽核心结构。此外，我们还利用修饰侧链的冗余中性损失，引入了一种称为特征碎片离子的第三类离子，以实现匹配。研究表明，肽糖基化产物的MS/MS光谱包含多维信息，且在未尝试使用多步匹配算法的情况下，通常超过一半的谱信息被忽视。与常规及/或修饰肽离子匹配相比，我们的三离子策略显著提高了糖基化产物MS/MS光谱的可解释中值分数。以阿马多里产物表征为例，我们应用该方法自动识别所有已确立的诊断离子。我们进一步展示了该方法如何有效地应用开放搜索概念，并通过展示两种此前未描述的、具有102.0311和268.1768 Da质量变化的肽糖基化修饰，实现了未知结构的优化阐明。通过将同位素标记的糖基化产物与其整合，我们对其碎片特征进行了表征，这为非靶向结构识别提供了高可靠性。