Comprehensive LC-MS and MS/MS studies of N-glycans derived from biological samples



Journal Title

Journal ISSN

Volume Title



Glycosylation is a common type of protein modification and is related to various biological activities such as protein folding, antibody stability, cell-cell interaction, and communication. Glycans are also linked to the progression of different types of diseases and can be considered as potential biomarkers for diagnosis and prognosis. Hence, development of a reliable and high-throughput quantitative glycomic strategy is necessary for the understanding protein biofunctions, disease diagnosis and quality control of biopharmaceuticals. In recent decades, mass spectrometry (MS) has become one of the most powerful bioanalytical instrument because of its capability in identifying biomolecules and elucidating structures through tandem MS. In this study, efforts have been made to improve MS-based quantitative glycomic profiling and glycan structure interpretation methods. Previous glycan identification and quantitation were relying on high resolution MS (HRMS). However, HRMS is not ubiquitous in labs because of its high cost and large size. We developed a method that enable the use of triple quadrupole MS working in multiple reaction monitoring (MRM) modes for the identification and quantitation of glycans. A transition list was generated for different types of permethylated N-glycans. Other parameters like collision energy and number of transitions for each precursor ion were also optimized. Sensitive and reliable quantitation of glycans released from model glycoproteins, human blood serum, and cancer cell lines were achieved with these optimized instrument setup. The use of separation technique is always necessary for the analysis of glycans released from biological sample because of the high diversity in structures and large concentration dynamic range of glycans in a biological system. An efficient LC separation would improve quantitation reliability and provide more structural information of glycans. In our pervious analysis, permethylated glycans were always separated by reverse phase LC (RPLC) at ambient temperature. However, severe peak broadening was always observed for highly branched glycans, resulting in overlapped peak for different glycans and limited recognition of various glycan isomers. We addressed the issue of peak broadening during RPLC-MS analysis of permethylated glycan. We determined that separation under tempertures above ambient would reduce the peak width and increase the retention of glycans on the C18 column. In a series of column temperature test, we determined that 55 oC was an optimized temperature with a balance of separation efficiency and column lifetime. Glycans isomers released from ribonuclease B and fetuin were practically separated at high temperature RPLC-MS and matched the NMR data. Multi-linear regression results also demonstrated our hypothesis that permethylated glycan molecule has more rigid structure and hydrophobicity at a higher temperature. Although significant improvement in separating permethylated glycans was achieved in this study, there are still difficulties in resolving all glycan isomers in this platform. In order to acquire more glycan structural information, different LC columns should be investigated. Porous graphite carbon (PGC) column is a type of mix mode LC column that is specifically sensitive to isomers; it has been utilized for the separation of native glycans. However, the application of PGC column in separating permethylated glycan was not successful according to literature. Efficient separation was achieved using PGC column at temperatures above ambient because of the influence of temperature on the glycan 3D conformation. Isomers resulted from different fucose site, different galactose site and different monosaccharide linages were baseline resolved and identified in our optimized PGC-LC-MS/MS platform. Tandem MS in conjunction with isomeric separation capability allowed the identification of many glycan strcutures. Diagnostic ion at m/z 468.27 was proposed for identifying fucosylation site, which is an essential glycan structure possessing various biofunctions. Meanwhile, diagnostic ions for discriminating β 1,3 and β 1,4 linked galactose were also found by collision induced dissociation (CID) and higher energy collisional dissociation (HCD) MS/MS. More work needs to be done on LC-MS quantitation of glycans because of several factors that induced errors to MS-based quantitation. Fluctuation in ionization efficiency, MS response bias and poor linearity prompts the need for using internal standards for more reliable quantitation. iGlycoMab is a monoclonal IgG2 antibody that has stable isotope label (15N) in all N-acetylglucosamine of glycan resided on the heavy chain. Errors from both sample handling and LC-MS analysis can be normalized to this internal standard. However, there was degradation observed for 15N labeled internal standard when mixed with blood serum sample for sample preparation. We tested different sample preparation protocols and finally concluded that tryptic treatment or 90 oC thermal denaturation would suppress internal standard degradation. Smaller standard deviation and better quantitation linearity were obtained with the utilization of iGlycoMab internal standard. Multiplexed analysis in MS would normalize variations from MS instrument and reduces total analysis time for multiple samples. AminoxyTMT is a multiplexing reagent that specifically designed for quantitative glycomics. The reagents would interact with glycan reducing end at ambient temperature and generate different reporter ions in tandem MS analysis. The intensity of reporter ion can be used as a representative of the abundance of same glycan structure in various samples. The issue of low reporter ion yield showed up when we first analyzed aminoxyTMT labeled glycan in LC-MS/MS. Conducting MS3 on Y1 ion could improve the yield of reporter ions; however, this method would increase the instrument requirements. Here we utilized the formation of different adducts to modify labeled glycan fragmentation pattern. The multiplexing analysis of model glycans accurately matched expected ratio with a less than 10% bias. Optimized LC-MS condition was applied in the analysis of glycans released from different cancer cell lines and blood serum from patients with various esophageal diseases. Native glycans are not suitable for direct analysis because of the lack of a chromosphere or fluorophore for optical detection. Additionally, glycans do not exhibit low ionization efficiency for MS analysis. Numerous efforts have been made to derivatize glycans with different reagents in order to achieve better LC separation and higher sensitivity. A comprehensive assessment is necessary for providing guidance in selecting derivatization reagents and LC column for various analysis purpose. In this study, we have investigated derivatization strategies including 2AB, procainamide, aminoxyTMT, RapiFluor-MS labeling and permethylation. We compared separation efficiency of differently derivatized glycans on C18, HILIC, and PGC column using both analytical LC-MS and nanoLC-MS platforms. The quality of MS2 spectrum generated by different strategies was also assessed for glycan structure elucidation. RapiFluor-MS labeling provides fastest sample preparation because of the use of rapid digestion enzyme and quick labeling reaction. Meanwhile, RapiFluor-MS labeled glycan exhibited highest MS signal in MS analysis, which is over 100 times greater than 2-AB labeled and native glycans, about 3 times greater than procainamide labeled glycans and about 1.5 time of permethylated glycan intensity. For sialylated glycans, the advantage of permethylation showed up. Permethylated sialylated glycan not only exhibited significantly higher intensity than other derivatized glycans but also has a more accurate reflection of molar abundance among all glycan. Meanwhile, permethylation also stables labile glycan structure, more reliable quantitative analysis and structural analysis by this method. The abovementioned LC-MS quantitative glycomic profiling methods were effectively utilized in the analysis of blood serum of patients with liver diseases, rat brain tissues, and metastatic breast cancer cell lines. Glycan biomarker was discovered for these diseases with the help of glycan quantitation methods developed above.



Glycosylation, LC-MS, Permethylation, Biomarkers