Aberrant IgA1 glycosylation is inherited in familial and sporadic IgA nephropathy

Abstract

IgA nephropathy (IgAN) is a complex trait determined by genetic and environmental factors. Most IgAN patients exhibit a characteristic undergalactosylation of the O-glycans of the IgA1 hinge region, which promotes formation and glomerular deposition of immune complexes. It is not known whether this aberrant glycosylation is the result of an acquired or inherited defect, or whether the presence of aberrant IgA1 glycoforms alone can produce IgAN. A newly validated lectin enzyme-linked immunosorbent assay (ELISA) was used to determine the serum level of galactose-deficient IgA1 (Gd-IgA1) in a cohort of 89 IgAN patients and 266 of their relatives. High Gd-IgA1 levels (> or =95th percentile for controls) were observed in all 5 available patients with familial IgAN, in 21 of 45 (47%) of their at-risk relatives (assuming autosomal dominant inheritance), and in only 1 of 19 (5%) of unrelated individuals who married into the family. This provides evidence that abnormal IgA1 glycosylation is an inherited rather than acquired trait. Similarly, Gd-IgA1 levels were high in 65 of 84 (78%) patients with sporadic IgAN and in 50 of 202 (25%) blood relatives. Heritability of Gd-IgA1 was estimated at 0.54 (P = 0.0001), and segregation analysis suggested the presence of a major dominant gene on a polygenic background. Because most relatives with abnormal IgA1 glycoforms were asymptomatic, additional cofactors must be required for IgAN to develop. The fact that abnormal IgA1 glycosylation clusters in most but not all families suggests that measuring Gd-IgA1 may help distinguish patients with different pathogenic mechanisms of disease.

Figure 1.

Structure of K1904 (A) K3041 (B). Symbols are described in the figure. Arrows identify the patients with biopsy-proven IgAN (2 deceased). (C) Distribution of Gd-IgA1 levels among IgAN patients, relatives at risk under an autosomal dominant model, marry-in relatives, and unrelated controls. Mean values for each group is indicated by solid black bars. The dashed horizontal lines indicate the 95th and 99th percentile cutoffs derived from controls. Gd-IgA1 values above the 95th percentile are shown in red. The P values for comparisons between the different groups are shown.

Figure 2.

(A) Distribution of Gd-IgA1 levels in sporadic IgAN patients (n = 84), their blood relatives (n = 202), and unrelated controls (n = 141). The dashed horizontal lines indicate the 95th and 99th percentile cutoffs for controls. Values above the 95th percentile are shown in red. (B) Distribution of Gd-IgA1 levels in relatives, after stratification based on Gd-IgA1 levels in the index case. The P values for comparisons between the different groups are shown.