MSACL 2024 Abstract

INTRODUCTION: Leakage of Cerebrospinal Fluid (CSF) can result from trauma, lacerations, and surgery, and if untreated can result in potentially life-threatening meningitis. CSF leaks can be diagnosed through the detection of β2-Transferrin (β2-Tf) from nasal secretions via gel electrophoresis. β2-Tf is a uniquely glycosylated proteoform of Transferrin found primarily in CSF, which can be distinguished from the most abundant Transferrin proteoform (β1-Tf) by its altered electrophoretic mobility. However, gel-based detection of β2-Tf requires hours of intensive labor and has low specificity due to the numerous alternative glycoforms of Transferrin. Alternative, higher-throughput assays are thus desirable for improved patient care.

Recently, our group applied High-Resolution Mass Spectrometry (HR-MS) coupled to Microprobe In-Emitter Elution (MPIE) to discriminate Transferrin isoforms. This method successfully resolved β2- from β1-Tf in nasal secretions derived from patient samples with suspected CSF leakage, but has relatively low sample-to-sample throughput. Herein, we describe a novel approach utilizing Liquid Chromatography Mass Spectrometry (LC-MS) to rapidly profile Transferrin glycoforms from patient nasal secretion samples. Furthermore, we explore solutions to address significant blood contamination in nasal secretions, which may otherwise interfere with this approach. Furthermore, we explore solutions to address significant blood contamination in some nasal secretions, which may otherwise interfere with this approach.

METHODS: Our approach consists of two primary steps: 1) Sample processing and high abundance protein depletion via commercial filtration and depletion columns. 2) LC-MS analysis of the resulting extract to resolve Transferrin glycoforms by retention time and m/z. This approach retains the high specificity of the previously demonstrated MPIE-MS method, but takes advantage of existing high-throughput liquid chromatography platforms to reduce the sample-to-sample analysis time. Secretion samples were analyzed using a Vanquish HPLC system coupled to a Q-Exactive Plus Mass Spectrometer (ThermoFisher). Broadband mass spectra were collected (FWHM resolution of 17,500 at 200 m/z), and the deconvoluted average masses were used to identify Transferrin glycoforms.

RESULTS: Our approach was benchmarked using remnant nasal secretion samples from patients with known CSF leakage. Examining 9 samples , the resulting ESI spectra revealed peaks at 79,554 Da and 78,008 Da, consistent with β1-Tf (Pred. MW (Average): 79,554.71 Da) and β2-Tf (Pred. MW (Average): 78,008.35 Da) within ~10 ppm mass accuracy. The 79.5 kD peak was observed in all samples, while the 78 kD peak was observed in 7 of 9 samples. Samples with missing β2-Tf were marked by significant blood contamination. Genetic variants (2/9 samples) resulted in mass shifts for both β1- and β2-Tf variants, but Transferrin glycoforms identifiable due to the preserved mass shift between the β1- and β2-Tf peaks (~1546.4 Da). An optimized LC-MS gradient enables rapid (~10 minute) sample-to-sample injection times.

Samples with blood contamination represent an additional challenge for identification of β2-Tf, owing to ion suppression from blood-derived β1-Tf. We have developed enrichment and depletion methods to increase the sensitivity for β2-Tf in blood contaminated samples, and anticipate that these approaches will reduce or eliminate this challenge.

NOVELTY: Our approach enables high-throughput, MS-based measurement of Transferrin glycoforms (i.e. β1- and β2-Tf) from clinical nasal secretion samples for the first time.