Changes in the glycan composition of glycoproteins in different biofluids are a dynamic metabolic indicator of health and immune status. In recent years, especially for blood-derived fluids, large cohorts of patient samples have been analyzed for total N-glycan content or for N-glycan compositions on immunoglobulin G. Most current analysis methods require significant sample processing, derivatization and purification steps prior to detection. Combined with the amount of time required for these steps, clinical diagnostic applications for glycans are limited. Using workflows adapted from MALDI imaging mass spectrometry of N-glycans in tissues, we have evaluated different approaches to developing rapid approaches for analyzing the total N-glycan composition of clinical biofluids like serum and plasma.
The primary objective of this work was to develop a N-glycan profiling method that can analyze clinical biofluids in a rapid and high-throughput manner utilizing mass spectrometry imaging.
Patient de-identified pooled serum or plasma samples from healthy donors were used for method development and evaluation. Donors with breast cancer or benign conditions were used to demonstrate the clinical utility. Small aliquots of unprocessed samples (1 µL) were mixed in different solvent formulations and spotted directly on amine-reactive hydrogel-coated slides. Multiple fixation, drying and washing methodologies were used to de-salt and promote adhesion of the sample to the slides. The samples were incubated in a humidity chamber for an hour to allow binding to the slide. A molecular coating of recombinant PNGase F was sprayed onto the slides using an HTX TM Sprayer. After a two-hour incubation, CHCA matrix was applied and native N-glycan profiles were acquired for each spot by MALDI FTICR MS in positive ion mode. Data were visualized and analyzed using SCiLS Lab software. Similar methods were used for profiling on a target ground steel plate.
The goal was to develop an N-glycan profiling analysis method for biofluids that meet the criteria of having high sensitivity, requiring minimal sample processing and can be done in less than 8 hours. An initial emphasis was put on serum samples obtained from donors with breast cancer, as there is need for a biomarker to supplement mammography screening. To ensure optimal detection of the N-glycans by MALDI-FTICR MS, a key step was to remove salts and other soluble interfering compounds present at high concentrations in the blood samples. For wash solutions, different mixtures of de-lipidating organic solvents and aqueous solvents were tested. Using an optimized combination of fixative, drying and aqueous solvent washing steps slides were processed for digestion with PNGase F using established tissue imaging workflows. Depending on the sample and preparation conditions, 75 or more N-glycan species in the mass range m/z = 1100-3300 can be detected from a single serum spot (a 1 ul equivalent). Processing times as low as 6 hours are possible, and further optimization can likely decrease this time. The reproducibility of the method was established by processing serum and plasma standards on three successive days and comparing intensities of the most abundant N-glycans and the N-glycan structural classes. Utilizing the method on a clinical serum cohort of breast cancer patients displayed the potential for distinguishing disease and disease-free patients by serum N-glycan analysis. Adaptions of the method using a steel MALDI target plate and Endoglycosidase F3 were established.
The method allows large numbers of samples to be reproducibly analyzed with minimal processing time and sample manipulation requirements, with the long-range goal to develop the approach for routine application to clinical laboratory assays.