P0356QUANTITATIVE ASSESSMENT OF GALNAC NUMBER IN THE HINGE REGION OF IGA1 IN IGA NEPHROPATHY WITH CRESCENTS FORMING

Abstract

Abstract Background and Aims Aberrant IgA1 O-linked glycosylation of IgA1 hinge region (HR) has been shown playing an important role in the pathogenesis of IgA nephropathy (IgAN). Serum levels of galactose-deficient IgA1 (Gd-IgA1) was associated with the development and progression in IgAN, however variations in the composition of IgA1 HR glycoforms are unknown. In this study we aim to quantitative assessment of GalNAc number in the hinge region of IgA1 in IgAN with crescents forming using mass spectrometry. Method Plasma polymeric IgA1 (pIgA1) was purified from a discovery cohort including crescentic IgAN (n=11) non-crescentic IgAN (n=10) and healthy controls (HC, n=10) and a validation cohort (crescentic IgAN, n=10; non-crescentic IgAN, n=10 and healthy controls, n=11). After denaturation, reduction, alkylation and trypsin digestion, liquid chromatography-high-resolution mass spectrometry (LC-MS) was used to analyse the IgA1 HR glycans. The intensity of the identified IgA1 O-glycopeptide was calculated and expressed as the relative abundance for each glycopeptide. The molecular weight (MW) of intact O-glycopeptide was calculated as the formula following: MW = HR + x GalNAc + y Gal + z NeuAc + H+ (x, y and z represent the number of GalNAc, galactose and NeuAc which bind to one HR in IgA1 O-linked glycopeptides respectively) Glycopeptide content ratio = (glycopeptide peak intensity/ total glycopeptide intensity) ×100%. Results In the discovery cohort population, the level of Gd-IgA1 was highest in patients with crescentic IgAN, middle in patients with non-crescentic IgAN and lowest in healthy controls (347.69±57.89 U/ml vs 336.32±38.44 U/ml vs 330.14±33.22 U/ml), albeit that didn’t reach the statistical significance. There were significantly difference in GalNAc number of IgA1 HR among patients with IgAN and healthy controls. Overall the numbers of GalNAc bound to one HR was much lower in the patients than healthy controls. As shown in the Figure 1, the proportion of GalNAc 3, defined as O-glycans that were bound to one HR at 3 sites, were highest in patients with crescentic IgAN, then the non-crescentic IgAN and lowest in the healthy controls (9.92%±3.37% vs 6.65%±1.53% vs 4.05%±1.24%; p-value for the trend<0.001). Similar results were observed in GalNAc 4 (45.91%±4.75% vs 41.13%±2.95% vs 40.98%±2.95%; P=0.004. However regarding the GalNAc 5 and GalNAc 6, crescentic IgAN was lowest, then non-crescentic IgAN and highest in the healthy control (GalNAc 5: 45.17%±5.46% vs 46.90%±2.78% vs 48.05%±3.02%, P=0.001 GalNAc 6:1.62%±1.60% vs 3.95%±1.92% vs 5.39%±2.38%; P<0.001). These results were consistent in the validation cohort (Figure 1). Conclusion GalNAc numbers in IgA1 HR was lower in patients with IgAN especially in crescentic IgAN. These results suggest that the O-glycans of IgA1 were associated with the severe phenotype in IgA nephropathy.