The results of this work contribute to the understanding of antibody-mediated allo- and autoimmunity towards blood cells. As such they also contribute to the development of new diagnostic assays to predict the severity of alloantibody-derived blood diseases like hemolytic disease of the fetus or newborn (HDFN) and fetal or neonatal alloimmune thrombocytopenia (FNAIT). These diseases can occur if there is a blood group incompatibility between a pregnant woman and the fetus. When the fetus has inherited an incompatible paternal blood group, maternal IgG antibodies can be formed, for example towards fetal red blood cells (RBC) or towards platelets, leading to HDFN or FNAIT, respectively.
The foundation of this work started because the concentration of antibodies found in these disease does not correlate well with disease severity – so bad in fact that this is by many investigators not considered very useful for diagnosis. This is especially true for FNAIT, where platelets and endothelial of the veins cells are destroyed, leading to thrombocytopenia and hence skin-, organ- or intracranial bleedings, which can lead to fetal death. This bad correlation results in diagnosis only being made after an affected pregnancy has occurred, with all the associated morbidity. Indeed, this is a good indicator that following pregnancies will also be affected. To prevent this, we took a close look at factors that may explain this discrepancy, allowing for an accurate diagnosis at treatment in the first potentially affected pregnancy.
The fragment crystallisable (Fc) region of IgG antibodies interacts with cell surface receptors (Fc receptors (FcγR)) that are found on effector leukocytes and can activate the immune system. At amino acid position 297, IgG contains a highly conserved N-linked glycosylation site. This glycan consists of an invariant structure, containing four N-acetylglucosamines (GlcNAc) and three mannoses. In addition, a bisecting GlcNAc (bisection), core fucose and one or two galactoses and N-acetyl neuraminic (sialic) acids can be attached. The interaction between antibodies and effector leukocytes via FcyRs is influenced by the conformation and exact composition of the IgG Fc-glycan, but exactly how has remained unknown.

When we therefore analyzed the N-linked glycans of human platelet antigen (HPA)-1a specific IgG1 with mass spectrometry in large series of FNAIT cases (n=166) including longitudinal samples (n=26), we found that Fc-glycosylation patterns are stable over time. Only fucosylation decreased significantly after the first pregnancy in many of the affected cases we investigated. We found that anti-HPA-1a fucosylation combined with galactosylation and antibody level correlated with disease severity, suggesting that these parameters may be feasible markers in screening for severe cases of FNAIT.
Similar to FNAIT, antibody formation against fetal RBC in HDFN leads to RBC breakdown and hence, fetal anemia, jaundice, intracranial bleeding or even fetal death. Anti-RhD, anti-Rhc, anti-K and anti-RhE are accountable for most cases with severe HDFN. In the Netherlands, all pregnant women are tested for red cell antibodies during the first trimester of pregnancy. However, there is no strict correlation between the antibody level present and the clinical outcome of the fetus, arguing that other factors might play a role, for example, Fc-glycosylation. We purified and analysed IgG1 Fc-glycosylation profiles of pregnant women in relation to pregnancy outcome. Similar to anti-HPA-1a antibodies, anti-RhD and anti-Kell can have decreased fucosylation of antigen-specific IgG1 compared to total serum IgG1, which correlated with clinical outcome. In addition, we found that high galactosylation and sialylation of anti-Rhc correlates with more severe disease. What is more, we discovered that immunization during pregnancy is not required for this response, as similar findings were found after immunization by blood transfusion. In conclusion, these studies imply that Fc-glycosylation may be used to predict HDFN disease severity and to identify patients needing treatment.
We also investigated whether the aberrant Fc-glycosylation seen for RBC alloantibodies are also found in RBC autoantibodies. It is known that in many autoimmune diseases serum IgG1 galactosylation and sialylation is lowered, but Fc-glycosylation of RBC-specific autoantibodies has never been thoroughly analyzed. Therefore, we analyzed RBC autoantibody and serum IgG1 Fc-glycosylation in relation to occurrence of hemolysis in 103 cases. Analysis of anti-RBC Fc-glycoprofiles suggested that lower bisection and higher galactosylation associate with lower hemoglobin levels. However, in contrast to alloantibodies against RBCs, RBC-bound IgG1 Fc-fucosylation was not changed compared to total IgG1. We did observe that total IgG1 purified from plasma of patients with RBC-bound antibodies showed significantly decreased galactosylation and sialylation levels compared to healthy controls, similar to what previously has been shown for other autoimmune diseases.
The precise mechanism why the immune responses against alloantigens triggers a potent non-fucosylated IgG response is unknown. Since we see this response against platelets and red blood cells in alloimmune, but not autoimmune responses, and because it are the B cells that produce the IgG antibodies, including the glycans, this suggests that the antigen-specific B cells in alloimmune responses are triggered in a unique way. This can be either directly or after receiving instruction from (a) secondary immune cell(s). This work has led to several testable hypotheses that we will explore in continuation of the groundwork laid down in this project.
In conclusion, in this LSBR-funded work we found differences between allo- and autoantibody Fc-glycosylation, with low fucosylation only being present in alloantibodies and correlating with clinical outcome. This discrepancy could be explained by the immune system reacting differently in allo- and autoimmune diseases, where alloantigens are recognized as foreign membrane antigens, inducing strong non-fucosylated immune responses with vital clinical consequences. Other important glycan changes were identified, in particular galactosylation, also found in autoimmune anemias that seem to contribute to elevated disease states. This knowledge has now culminated in a new monitoring trial (HIP-Study), which one day may be implemented nationally to identify pregnancies at risk and thus allowing for treatment before symptoms occur.

Thesis 2017: IgG-Fc-glycosylation in immune-mediated cytopenias, Myrthe Sonneveld