Mixed-phase weak anion-exchange/reversed-phase LC-MS/MS for analysis of nucleotide sugars in human fibroblasts

Abstract

Nucleotide sugars (NS) fulfil important roles in all living organisms and in humans, related defects result in severe clinical syndromes. NS can be seen as the "activated" sugars used for biosynthesis of a wide range of glycoconjugates and serve as substrates themselves for the synthesis of other nucleotide sugars. NS analysis is complicated by the presence of multiple stereoisomers without diagnostic transition ions, therefore requiring separation by liquid chromatography. In this paper, we explored weak anion-exchange/reversed-phase chromatography on a hybrid column for the separation of 17 nucleotide sugars that can occur in humans. A robust and reproducible method was established with intra- and inter-day coefficients of variation below 10% and a linear range spanning three orders of magnitude. Application to patient fibroblasts with genetic defects in mannose-1-phosphate guanylyltransferase beta, CDP-L-ribitol pyrophosphorylase A, and UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase showed abnormal levels of guanosine-5'-diphosphate-α-D-mannose (GDP-Man), cytidine-5'-diphosphate-L-ribitol (CDP-ribitol), and cytidine-5'-monophosphate-N-acetyl-β-D-neuraminic acid (CMP-Neu5Ac), respectively, in consonance with expectations based on the diagnosis. In conclusion, a novel, semi-quantitative method was established for the analysis of nucleotide sugars that can be applied to diagnose several genetic glycosylation disorders in fibroblasts and beyond.

Keywords: Congenital disorders of glycosylation; LC–MS/MS; Mixed-mode chromatography; Nucleotide sugars; Sugar metabolism.

Fig. 1

Generalized fragmentation scheme of nucleotide sugars in CID. 1, deprotonated nucleotide sugar [M-H]⁻; 2, fragment containing the sugar moiety bound to a phosphate anion; 3, nucleotide diphosphate anion fragment; 4, nucleotide dehydroxy-diphosphate fragment; 5, nucleotide monophosphate fragment; 6, dehydroxy-diphosphate ribose anion fragment; 7, metaphosphate anion/PO₃⁻ fragment (79 m/z); 8, dihydrogen phosphate anion/PO₄H₂⁻ fragment (97 m/z); 9, dehydroxy-pyrophosphate anion fragment (158.9 m/z). R marks the hydroxyl group that is exchanged for hydrogen in dTDP-sugars

Fig. 2

Transition ion ratio of all 17 nucleotide sugars included in the method (circles, normalized against maximum intensity transition). CV values for TIR are indicated by background coloring; highest intensity transition CV value is 0% due to the nature of normalization. Crossed out transitions are not monitored. Fragment numbers correspond to those used in Fig. 1. A facet grid bar graph ilustration for ease of data comprehension is shown in Supplementary Figure S2

Fig. 3

MRM chromatogram of a nucleotide standard mix with 200 nM concentration (CDP-ribitol was included from data of the control A cell line). All transitions are summed depending on precursor m/z. 1, CDP-ribitol; 2, CMP-Neu5Ac; 3, UDP-Man; 4, UDP-Gal; 5, UDP-Glc; 6, UDP-Ara; 7, UDP-Xyl; 8, UDP-GalNAc; 9, UDP-GlcNAc; 10, GDP-Man; 11, guanosine-5′-diphosphate-α-d-glucose (GDP-Glc); 12, GDP-Fuc; 13, 2′deoxy-thymidine-5′-diphosphate-α-d-glucose (dTDP-Glc); 14, ADP-Glc; 15, ADP-Rib; 16, dTDP-Rha; 17, uridine-5′-diphosphate-α-d-glucuronic acid (UDP-GlcA)

Fig. 4

Intra-day and inter-day relative residuals of intensity plotted against the analyte retention time. Residuals of retention time were transformed to match compounds intra-day retention time. Confidence ellipses (α = 0.95) were plotted for both intra-day and inter-day values

Fig. 5

Changes in disease relevant nucleotide sugars normalized against total peak area of all nucleotide sugars. a The difference in CDP-ribitol level in CRPPA-deficient patients A and B compared to control cell line A. b The decrease in GDP-Man level in GMPPB-deficient patients C and D compared to control cell line A. c The increase in CMP-Neu5Ac level in French type sialuria patient E compared to control cell line B. Error bars represent the 95% confidence intervals. ****p < 0.0001 (adjusted p-value; one-way ANOVA). Full normalized nucleotide sugar profiles of patient and control cell lines can be found in Supplementary Figures S5 to S7