Structure of human IgA1 and its O-glycans

(A) IgA1 has an extended hinge region that contains between three and six O-glycans attached to serine or threonine residue between position 225 to 236. (IgA1 with five O-glycans per hinge region is shown.)
(B) Glycosylation of IgA1 is mediated by stepwise co/post-translational modifications. First, N-acetylgalactosamine (GalNAc) is added to serine/threonine residue by activity of N-acetylgalactosaminyl-transferase (GalNAcT2; step 1). Next, a galactose moiety is added to GalNAc by core 1 beta 1,3-galactosyltransferase (C1GalT1) and core 1 beta3GalT-specific molecular chaperone (Cosmc; step 2). Sialic acid may then be added to the galactose moiety by alpha2,3 sialytransferase (ST2,3; step 3) or to the GalNAc moiety by alpha2,6 sialytransferase (ST2,6; step 4). Alternatively, sialic acid may be added to GalNAc by ST2,6 before the addition of galactose (step 2a). Notably, sialylated GalNAc (step 2a) cannot be subsequently galactosylated, whereas galactosylated GalNAc may be sialylated at either the GalNAc or galactose moiety, or both (step 3 and/or 4).
(C) These steps produce a combination of different O-glycoforms of varying degrees of galactosylation and sialylation. The relative proportion of poorly galactosylated IgA1 is increased in IgA nephropathy.