Effects of N-glycans on the structure of human IgA2

Abstract

The transition of IgA antibodies into clinical development is crucial because they have the potential to create a new class of therapeutics with superior pathogen neutralization, cancer cell killing, and immunomodulation capacity compared to IgG. However, the biological role of IgA glycans in these processes needs to be better understood. This study provides a detailed biochemical, biophysical, and structural characterization of recombinant monomeric human IgA2, which varies in the amount/locations of attached glycans. Monomeric IgA2 antibodies were produced by removing the N-linked glycans in the CH1 and CH2 domains. The impact of glycans on oligomer formation, thermal stability, and receptor binding was evaluated. In addition, we performed a structural analysis of recombinant IgA2 in solution using Small Angle X-Ray Scattering (SAXS) to examine the effect of glycans on protein structure and flexibility. Our results indicate that the absence of glycans in the Fc tail region leads to higher-order aggregates. SAXS, combined with atomistic modeling, showed that the lack of glycans in the CH2 domain results in increased flexibility between the Fab and Fc domains and a different distribution of open and closed conformations in solution. When binding with the Fcα-receptor, the dissociation constant remains unaltered in the absence of glycans in the CH1 or CH2 domain, compared to the fully glycosylated protein. These results provide insights into N-glycans' function on IgA2, which could have important implications for developing more effective IgA-based therapeutics in the future.

Keywords: IgA antibodies; N-linked glycan; SAXS; flexibility; protein assembly; protein stability.

FIGURE 1

Amino acid sequence alignment and depiction of IgA2 (m2) variants. (A) Alignment of protein sequences of constant regions for IgA2 (m2) H4-WT and the generated mutants. N-linked glycosylation sites are highlighted in yellow, and exchanged amino acid residues are highlighted in red. The penultimate cysteine is highlighted in pink, and the alanine in the C3QA variant is shown in white. (B) Schematic illustration of the assembled IgA2 (m2) variants with N-glycosylation sites indicated by the specific sequon.

FIGURE 2

Purification and analysis of the IgA (m2) assembly. (A) Overlay of analytical size-exclusion chromatograms of affinity-purified monomeric IgA2 (m2) from small-scale transient transfections. Monomer (M), dimer (D), and polymer (P) peaks are indicated. No JC was co-expressed. (B) Overlay of normalized analytical size-exclusion chromatograms of affinity-purified dimeric IgA2 (m2) from small-scale transient transfections. JC was co-expressed to increase the amounts of dimeric IgA2 (m2). (C) SDS-PAGE under non-reducing conditions of purified monomeric IgA2 (m2) glycosylation mutants followed by Coomassie Brilliant Blue staining. Herceptin IgA2 (m2) (HER) is included for comparison.

FIGURE 3

Site-specific analysis of N-linked glycans on the different IgA2 (m2) variants. The visual representation highlights the more abundant species while grouping those making up less than 4% into an “other” category for clarity. The N-glycosylation sites are presented in order from N- to C-terminus.

FIGURE 4

Thermostability and secondary structure analysis of the different IgA2 (m2) variants. (A) Melting curves of IgA2 (m2) variants obtained by DSF. (B) The thermal unfolding transition was monitored using IgA2 intrinsic tryptophan fluorescence by nanoDSF. (C) CD spectra of IgA2 (m2) variants. Representative images of single measurements are shown. (D) Summary of the melting temperatures measured by DSF, nanoDSF, and CD. All experiments were repeated three times (n = 3).

FIGURE 5

Binding affinity of IgA2 (m2) glycosylation variants to the FcαRI receptor. All experiments were repeated three times, and mean values ± SD are shown. Significance levels are shown according to a Student’s t-test; ns, not significant, “*” p < 0.05.

FIGURE 6

Experimental SAXS profiles. (A) Table of SAXS and MALS experimental parameters. (B) P(r) functions calculated for the experimental SAXS curves for all tested glycoforms (colored as indicated). The P(r) functions are normalized at the maxima. The P(r) shoulder at r ∼ 80 Å indicates the Fab-Fc separation described within the atomic model of IgA2. The P(r) peak at 40 Å corresponds to the average size across Fc or Fab regions. (C) Normalized Kratky plot.

FIGURE 7

The flexibility of IgA2 (m2) glycosylation variants. (A) Residuals (experiment/model) for the fits of the two-state models (black, blue, and red) are presented alongside the optimal single model in grey. These residuals collectively indicate the necessity of employing the two-state model to fit the experimental SAXS curves satisfactorily. (B) The panel displays experimental SAXS profiles of the IgA2 glycoforms (scatter) with the corresponding theoretical SAXS profiles derived from their respective two-state atomistic models (line). (C) Histograms of the R _g distributions of the top 300 selected multistate models are shown for the two-state model. (D) Representation of the open and closed two-state models for C2Q. The glycan moiety is colored blue. (E) Modeling parameters R _g and w of IgA2 (m2) variants. Results from the best-fitting multistate models are shown.