Sensitive and Quantitative MS-bases Glycomic Mapping Platform

糖链已经成为生物学研究中一个有趣但具有挑战性的领域。多聚糖在体内的作用 许多重要的生物学领域，例如但不限于：免疫系统、细胞发育、细胞 分化/黏附、宿主-病原体相互作用、蛋白质信号和蛋白质稳定性。异常 糖基化与多种疾病有关，包括癌症、囊性纤维化和骨关节炎。 糖组学/糖蛋白组学研究旨在量化和表征糖链结构(包括连接和 位置异构体)、蛋白质结合部位和蛋白质的特性。大约50%的哺乳动物 蛋白质是糖基化的，但与非糖基化的蛋白质相比，其丰度相当低。此外， 许多多聚糖可以占据同一蛋白质上的同一多聚糖连接位置；也就是说，同一蛋白质池可以 将几种不同类型的多糖连接到同一部位，每一种都可能具有不同的功能或 特定的活动。蛋白质多聚糖根据其氨基酸结合部位分为两类： N-糖链的天冬酰胺和O-糖链的苏氨酸、丝氨酸和酪氨酸。 一种已被成功用于研究细胞中N-糖链的策略是释放或分离N-糖链 从具有PNGase F酶的蛋白质中提取多聚糖，并研究其全局多聚糖组成的样品。一个 这种方法的缺点是，所谓的天然多糖具有较低的电离效率，这使得它们的 用质谱仪进行分析相当困难；然而，这个敏感性问题可以通过过甲基化来克服。 多聚糖。葡聚糖有许多异构体，这使得LC-MS/MS很难准确分析它们，如果 同分异构体不能拆分。这项提议证明了我们能够分离全甲基化的多聚糖 在质谱分析前用加热的PGC色谱柱进行异构体(目标1)，导致极 准确表征和定量生物样品中的葡聚糖异构体的灵敏分析方法。尽管 异构体多聚糖的分离已有报道，以前的研究只解决了天然和还原末端 标记的糖链结构。由于过甲基化多聚糖表现出至少两个电离效率 比上述结构高几个数量级，提高灵敏度的重要性， 用于检测生理浓度下的结构，这伴随着异构体的分离 全甲基化多糖(目标1)怎么夸大都不为过。 为了克服LC-MS样品之间电离效率的差异，我们成功地 具有各种稳定同位素组合的过甲基化多聚糖实现前所未有的定量多聚糖 对来自细胞培养实验、生物液和生物组织的样本进行比较。 通过实施我们的多层次同位素标记策略(代谢性15N标记，18O减少 末端标记和多路复用全甲基化)，可以增加潜在多路传输样本的数量 生物液、组织样本和细胞培养从之前的最高8个增加到16个和32个 样本(目标2)。生物来源的高通量异构体多聚糖的表征 样品可以通过结合AIMS 1和2中描述的方法来获得。当使用PNGase F时 广泛地从蛋白质中释放N-葡聚糖，对于O-葡聚糖不存在或尚未发现这样的酶。 我们开发了一种快速方法，RAIDR(快速氢氧化铵)，异丁酸O-葡聚糖 脱糖反应)，用于选择性地释放O-糖链；RAIDR离开蛋白质和N-糖链 毫发无损，允许进行兼容的下游分析(目标3)。除了改进LC-MS分析 方法，我们还建议在MALDI中添加石墨烯纳米片和碳纳米颗粒 用于增强样品制备、清理和提高两种原生物质的电离效率的基质 和全甲基化多聚糖(目标4)。基于质谱学的实验可以产生大量的 手动分析起来很麻烦的数据。有许多著名的蛋白质组学软件包 可用的，但很少能全面分析血糖数据集的。我们已经开发了多糖来 分析糖链数据集并打算扩展其功能(目标5)以处理糖链异构体(目标1)， 多重过甲基化多糖(目标2)、O-葡聚糖(目标3)和用MALDI-MS分析的多糖(目标4)。 所提出的方法和算法的发展将帮助我们和合作者更好地理解 食道发育过程中葡聚糖异构体的属性及生物医学意义， 乳腺癌和肝癌。我们也期待这里提出的分析工具和算法将是有益的 致其他对了解其他系统中葡聚糖异构体生物学属性感兴趣的科学家 从中受益。

Glycomics has emerged as an interesting yet challenging area of research in biology. Glycans function in numerous important biological areas such as, but not limited to: the immune system, cell development, cell differentiation/adhesion, host-pathogen interactions, protein signaling, and protein stabilization. Abnormal glycosylation has been associated with several diseases including cancer, cystic fibrosis, and osteoarthritis. Glycomics/glycoproteomics studies aim to quantify and characterize glycan structures (including linkage and positional isomers), protein attachment sites, and the protein’s identity. Approximately 50% of mammalian proteins are glycosylated but their abundance is rather low compared to non-glycosylated proteins. Furthermore, numerous glycans can occupy the same glycan attachment site on a protein; that is the same protein pool can have several different types of glycans attached to the same site, each with a potentially different function or a particular activity. Protein glycans are divided into two classes based on their amino acid attachment sites: asparagine for N-glycans and threonine, serine, and tyrosine for O-glycans. A strategy that has been successfully employed to investigate N-glycans in cells is to release or separate the glycans from proteins with the enzyme PNGase F, and study the global glycan composition of a sample. A drawback to this approach is that, so called, native glycans possess low ionization efficiencies which make their analysis by mass spectrometry quite difficult; however, this sensitivity issue can be overcome by permethylating glycans. Glycans have many isomers which can make their accurate analysis by LC-MS/MS difficult if the isomers cannot be resolved. This proposal demonstrates that we are able to separate permethylated glycan isomerss with a heated PGC column before mass spectrometry analysis (Aim 1), resulting in an extremely sensitive assay to accurately characterize and quantitate glycan isomers in biological samples. Although the separation of isomeric glycans has been previously reported, prior studies only resolved native and reducing end labeled glycan structures. Owing to the fact that permethylated glycans exhibit ionization efficiencies at least two orders of magnitude higher than the aforementioned structures, the importance of the increase in sensitivity, for the detection of structures at physiological concentrations, that accompanies isomeric separation of permethylated glycans (Aim 1) cannot be overstated. To overcome the variation in ionization efficiency between LC-MS samples, we have successfully permethylated glycans with various stable isotope combinations to achieve unprecedented quantitative glycan comparisons across samples derived from cell culture experiments, biological fluids, and biological tissues. Through the implementation of our multi-level isotopic labeling strategies (metabolic 15N labeling, 18O reducing end labeling and multiplex permethylation), the number of potential multiplexed samples can be increased from eight, the previous maximum, to 16 and 32 for biological fluid and tissue samples and cell culture samples, respectively (Aim 2). High throughput isomeric characterization glycans derived from biological samples can be attained by combining the methods described in Aims 1 and 2. While PNGase F is used extensively to release N-glycans from proteins, no such enzyme exists or has been discovered for O-glycans. We have developed a rapid method, RAIDR (Rapid Ammonium hydroxide Isobutyric acid O-glycan Deglycosylation Reaction), for selectively releasing O-glycans; RAIDR leaves the protein and N-glycans unscathed which allows for compatible downstream analyses (Aim 3). In addition to improving LC-MS analytical methods, we are also proposing the addition of graphene nanosheets and carbon nanoparticles to MALDI matrices for enhanced sample preparation, cleanup, and an increase in the ionization efficiencies of both native and permethylated glycans (Aim 4). Mass spectrometry based experiments can generate a tremendous amount of data that is cumbersome to analyze manually. There are numerous well-known proteomic software packages available but few that can comprehensively analyze glycomic datasets. We have developed MultiGlycan to analyze glycomic datasets and intend to expand its functionalities (Aim 5) to handle glycan isomers (Aim 1), multiplexed permethylated glycans (Aim 2), O-glycans (Aim 3), and glycans analyzed with MALDI-MS (Aim 4). The development of the proposed methods and algorithms will help us and collaborators to better understand the attributes and biomedical significance of glycan isomers in the development and progression of esophagus, breast, and liver cancer. We are also expecting the analytical tools and algorithms proposed here to be beneficial to other scientists who are interested in understanding the biological attributes of glycan isomers in other systems to benefit from.