A characteristic N-glycopeptide signature associated with diabetic cognitive impairment identified in a longitudinal cohort study

Abstract

Background: Identifying a biomarker for the decline in cognitive function in patients with diabetes is important. Therefore, we aimed to identify the N-glycopeptides on plasma proteins associated with diabetic cognitive impairment in participants in a longitudinal study using N-glycoproteomics.

Methods: We used samples from the 3-year SONIC (Septuagenarians, Octogenarians, Nonagenarians Investigation with Centenarians) longitudinal cohort study of older Japanese people in the general population. First, we placed the participants with diabetes into two groups: those that did or did not have cognitive decline over a 6-year period. Next, their plasma protein profiles were compared between baseline and the 6-year time point using two-dimensional fluorescence difference gel electrophoresis. Finally, an N-glycoproteomic study of the focused proteins was performed using an enrichment technique and liquid chromatography-tandem mass spectrometry.

Results: Approximately 500 N-glycopeptides, derived from 18 proteins, were identified in each sample, from among which we identified the N-glycopeptides that were associated with diabetic cognitive impairment using multivariate analysis. We found that N-glycopeptides with sialylated tri- or tetra-antennary glycans on alpha-2-macroglobulin, clusterin, serum paraoxonase/arylesterase 1, and haptoglobin were less abundant, whereas 3-sialylated tri-antennary N-glycopeptides on serotransferrin were more abundant.

Conclusion: N-glycopeptides with sialylated multi-antennary glycans comprise a characteristic signature associated with diabetic cognitive impairment.

General significance: The characterized N-glycopeptides represent potential biomarker candidates for diabetic cognitive impairment.

Keywords: Cognitive impairment; Diabetes; Liquid chromatography-tandem mass spectrometry; Longitudinal cohort study; N-glycoproteomics; Proteomics.