Fc-Glycosylation in Human IgG1 and IgG3 Is Similar for Both Total and Anti-Red-Blood Cell Anti-K Antibodies

Abstract

After albumin, immunoglobulin G (IgG) are the most abundant proteins in human serum, with IgG1 and IgG3 being the most abundant subclasses directed against protein antigens. The quality of the IgG-Fc-glycosylation has important functional consequences, which have been found to be skewed toward low fucosylation in some antigen-specific immune responses. This increases the affinity to IgG1-Fc-receptor (FcγR)IIIa/b and thereby directly affects downstream effector functions and disease severity. To date, antigen-specific IgG-glycosylation have not been analyzed for IgG3. Here, we analyzed 30 pregnant women with anti-K alloantibodies from a prospective screening cohort and compared the type of Fc-tail glycosylation of total serum- and antigen-specific IgG1 and IgG3 using mass spectrometry. Total serum IgG1 and IgG3 Fc-glycoprofiles were highly similar. Fc glycosylation of antigen-specific IgG varied greatly between individuals, but correlated significantly with each other for IgG1 and IgG3, except for bisection. However, although the magnitude of changes in fucosylation and galactosylation were similar for both subclasses, this was not the case for sialylation levels, which were significantly higher for both total and anti-K IgG3. We found that the combination of relative IgG1 and IgG3 Fc-glycosylation levels did not improve the prediction of anti-K mediated disease over IgG1 alone. In conclusion, Fc-glycosylation profiles of serum- and antigen-specific IgG1 and IgG3 are highly similar.

Keywords: Fc glycosylation; IgG1; IgG1 Fc; IgG3; IgG3 Fc; antibodies; mass spectrometry.

Figure 1

Examples of IgG1 and IgG3 Fc N-glycan structures showing skewing of anti-K Fc-glycosylation compared to total immunoglobulin G (IgG). (A) A schematic structure of the diantennary glycan found in human IgG. Blue square, N-acetylglucosamine; red triangle, fucose; green circle, mannose; yellow gray circle, galactose; purple diamond, N-acetylneuraminic (sialic) acid. (B,C) Mass spectrometric analysis of Fc glycopeptides encompassing N297 of IgG1 (B) and IgG3 (C). Spectra from a representative patient are shown, with mass/charge ratio (m/z) of the glycopeptides on the x-axis and relative abundance on the y-axis. Asterisks indicate non-glycopeptide signals.

Figure 2

IgG1 and IgG3 antibodies show skewing of anti-K Fc-glycosylation compared to total immunoglobulin G (IgG). Glycosylation features found in alloantibodies against the Kell antigen. Total (x-axis) vs. antibody specific (y-axis) IgG1 (A) and IgG3 (B) shows a decreased anti-K fucosylation, galactosylation, and sialylation and increased bisection compared to total IgG. Statistical analysis was done using a paired t-test.

Figure 3

Correlation between IgG1 and IgG3 Fc-glycosylation for total and anti-K specific immunoglobulin G (IgG). Total (A) and anti-K (B) IgG1 (x-axis) vs. IgG3 (y-axis) Fc-glycosylation shows a significant correlation between fucosylation, bisecting GlcNAc, galactosylation, and sialylation except for anti-K bisecting GlcNAc. The dotted line indicates the line of identical IgG1 and IgG3 glycosylation and illustrates the similarity of total IgG1 and IgG3 glycosylation, except for sialylation which seems higher in IgG3. Statistical analysis was done using a linear regression analysis.

Figure 4

High galactosylation correlates with low Hb level. (A) Anti-K IgG1, (B) IgG3 glycosylation, and (C) the combination of IgG1 and IgG3 Fc-glycosylation and antibody level (y-axis) correlated to Hb level (x-axis). Galactosylation shows a significant correlation with Hb level for (A) anti-K IgG1 and (C) the combination of IgG1 and IgG3 Fc-glycosylation and antibody level. Statistical analysis was done using Spearman rank correlation.

Figure 5

Low bisection and high galactosylation correlate with worse clinical outcome. (A) Anti-K IgG1, (B) IgG3 glycosylation, and (C) the combination of IgG1 and IgG3 Fc-glycosylation and antibody level (y-axis) correlated to disease severity (x-axis). Bisection and galactosylation show a significant correlation with disease severity. There is no correlation between fucosylation or sialylation and disease severity. Statistical analysis was done using one-way ANOVA with Bonferroni correction for multiple comparison testing.