Yang Xinyi, Du Zhuokun, Jiao Yuxin, Chen Sijie, Ning Juanjuan, Hong Haotian, Fu Bin, Liu Jiayu, Zhang Wanjun, Qin Weijie
School of Basic Medical Science, Anhui Medical University, Hefei 230032, China.
National Center for Protein Sciences Beijing, State Key Laboratory of Medical Proteomics, Beijing Institute of Lifeomics, Beijing Proteome Research Center, Beijing 102206, China.
Anal Chem. 2025 Jul 29;97(29):15855-15863. doi: 10.1021/acs.analchem.5c02002. Epub 2025 Jul 17.
Plasma represents a highly valuable clinical sample for protein biomarker discovery, offering a comprehensive source of physiological and pathological information. N-glycosylation plays key roles in various biological processes and enhances the sensitivity of plasma protein biomarkers for disease diagnosis. Consequently, large-scale characterization of the plasma proteome and N-glycosylation patterns by mass spectrometry (MS) is crucial for identifying biomarkers but remains highly challenging due to three major difficulties. First, plasma protein detection is limited as high-abundance proteins dominate the majority of the MS scans. Second, while plasma proteome coverage can be improved by constructing large-scale empirical spectral libraries and coupling them with DIA-MS, the labor- and time-consuming nature of experimental library generation imposes constraints on its wide adoption in clinical studies. Third, the low concentration and poor ionization of N-glycopeptides make their signals more susceptible to suppression in MS analysis. To address these issues, we developed an integrated workflow applying magnetic graphene-oxide (mGO) nanomaterial enrichment and an in silico predicted spectral library for low-abundance plasma proteome identification, along with tandem enrichment using hydrophilic interaction liquid chromatography (HILIC) for sensitive plasma N-glycoproteome profiling. In this way, 4538 plasma proteins were obtained in a single DIA-MS analysis using a QE-HF mass spectrometer, 10 times more than those obtained from direct analysis of neat plasma. Further HILIC enrichment of the mGO products enabled the identification of 7986 intact N-glycopeptides from 626 proteins with concentrations as low as the nanogram per liter range. Notably, 58.34% of these N-glycopeptides were undetectable by direct HILIC enrichment from neat plasma, highlighting the advantage of applying tandem enrichment.
血浆是蛋白质生物标志物发现中极具价值的临床样本，可提供生理和病理信息的全面来源。N-糖基化在各种生物过程中发挥关键作用，并提高血浆蛋白生物标志物用于疾病诊断的敏感性。因此，通过质谱（MS）对血浆蛋白质组和N-糖基化模式进行大规模表征对于识别生物标志物至关重要，但由于三个主要困难，这仍然极具挑战性。首先，血浆蛋白检测受到限制，因为高丰度蛋白主导了大多数MS扫描。其次，虽然通过构建大规模经验光谱库并将其与DIA-MS耦合可以提高血浆蛋白质组覆盖率，但实验库生成的 labor-和耗时性质限制了其在临床研究中的广泛应用。第三，N-糖肽的低浓度和差的电离使得它们的信号在MS分析中更容易受到抑制。为了解决这些问题，我们开发了一种集成工作流程，应用磁性氧化石墨烯（mGO）纳米材料富集和用于低丰度血浆蛋白质组识别的计算机预测光谱库，以及使用亲水相互作用液相色谱（HILIC）进行串联富集以进行灵敏的血浆N-糖蛋白质组分析。通过这种方式，使用QE-HF质谱仪在单次DIA-MS分析中获得了4538种血浆蛋白，比直接分析纯血浆获得的蛋白多10倍。mGO产物的进一步HILIC富集使得能够从626种蛋白质中鉴定出7986种完整的N-糖肽，浓度低至纳克每升范围。值得注意的是，直接从纯血浆中进行HILIC富集无法检测到这些N-糖肽中的58.34%，突出了应用串联富集的优势。
