Neurofilament Light Chain under the Lens of Structural Mass Spectrometry

Abstract

Neurofilament light chain (NfL) is an early nonspecific biomarker in neurodegenerative diseases and traumatic brain injury, indicating axonal damage. This work describes the detailed structural characterization of a selected primary calibrator with the potential to be used in future reference measurement procedure (RMP) development for the accurate quantification of NfL. As a part of the described workflow, the sequence, higher-order structure as well as solvent accessibility, and hydrogen-bonding profile were assessed under three different conditions in KPBS, artificial cerebrospinal fluid, and artificial cerebrospinal fluid in the presence of human serum albumin. The results revealed that NfL is a structurally heterogeneous protein, eliciting a large conformational flexibility. Its structural ensemble changed when it was diluted with an aqueous buffer versus a surrogate matrix, artificial cerebrospinal fluid (aCSF), and/or aCSF with human serum albumin. Various regions of protection and deprotection in the protein head, central helical, and tail domains that experienced altered solvent accessibility and conformational changes caused by different solvent conditions were identified. Moreover, interfacial residues, which may play a role in a potential direct interaction between NfL and human serum albumin, emerged from hydrogen-deuterium exchange mass spectrometry (HDX-MS). These data pinpointed distinct regions of the protein that may participate in such an interaction. Overall, critical quality attributes of a potential primary calibrator for NfL measurements are provided. These findings will ultimately inform ongoing biochemical and clinical assay development procedures and manufacturing practices, giving careful consideration during sample handling and method development.

Keywords: human serum albumin; hydrogen−deuterium exchange; mass spectrometry; neurofilament light chain; primary calibrator; structural characterization.

Figure 1

Intact mass analysis of NfL acquired on a Thermo QExactive Plus mass spectrometer equipped with a Vanquish UHPLC system eluted on an ACE 3 C4-300 column. (A) Intact mass spectrum of NfL, the inset shows the chromatogram of NfL. (B) Deconvoluted mass of human recombinant NfL with two dominant species, one representing the full-length NfL without the first methionine residue and another containing a potassium adduct.

Figure 2

(A) Different structural domains of NfL are composed of the head (1–92), coil 1A (93–124), linker 1 (125–137), coil 1B (138–234), linker 1/2 (235–252), coil 2A (253–271), linker 2 (272–280), coil 2B (281–396), and the tail that spans subdomain A (397–443) and subdomain B (444–543). (B) Peptides identified are shown across the entire sequence of NfL. Liquid chromatography-tandem mass spectrometry (LC/MS-MS) peptide mapping of human recombinant NfL. Data were collected following tryptic digestion on a Thermo QExactive Plus orbitrap coupled to a Vanquish UHPLC system fitted with an Acquity BEH C18 column and analyzed in Peaks Studio v.7.5.

Figure 3

(A) MEMHDX logit plot of NfL in aCSF vs aCSF HSA considering all time points and replicates with blue denoting peptides undergoing protection and red denoting peptides displaying deprotection. (B) Predicted structure of NfL–HSA complex by AlphaFold 3 color-coded with the prediction confidence. (C) Statistically significant regions from HDX-MS data overlaid on the predicted complex of NfL with HSA, which were determined in MEMHDX (p-value 0.001) following the comparison of NfL in aCSF vs aCSF in the presence of HSA across all time points, considering peptides which fulfilled or exceeded at least one of the statistical thresholds. Blue denotes protection, red denotes deprotection, pale cyan represents regions that were not statistically significant, gray indicates residues that were not present in the sequence coverage, and HSA is shown in wheat color. Blue regions near the wheat-colored HSA are suggestive of a potential binding interface.

Figure 4

Pairwise correlation matrix of the relative fractional uptake values. Pearson correlation analysis was conducted to examine interactions among various regions of NfL. Initially, differential uptake values were computed considering all time points between two conditions and arranged peptides according to the NfL amino acid sequence. These peptides were then clustered, and the average relative uptake was computed for each group. Subsequently, pairwise correlations among columns representing the averages of 10 peptides were calculated, and estimated functional regions were mapped onto the correlation matrix. The color bar on the right shows the direction of the correlation, blue color indicates the positive correlation, and red color shows the negative correlation. The size of the squares is scaled to the magnitude of the correlation, whereby empty or faded squares indicate that no correlation was found between the compared segments. (A) Correlation among the regions of NfL diluted with KPBS vs diluted with aCSF. (B) Correlation among the regions of NfL diluted with KPBS vs NfL diluted with aCSF in the presence of HSA. (C) Correlation among the regions of NfL diluted with aCSF vs NfL diluted aCSF in the presence of HSA. The size of each square corresponds to the magnitude of the correlation it represents. Specific regions exhibiting strong positive and negative correlations are highlighted with black rounded boxes to illustrate examples of differential regional correlations as a visual guide.