Fig 2: SARS-CoV-2-specific IgA2 levels correlate with ecDNA in severely diseased patients. (a,b) Correlation of levels of IgG, IgA1 and IgA2 directed against S1+S2 protein of SARS-CoV-2 with CRP in the plasma of SARS-CoV-2-infected patients with (a) severe disease (N = 10) or (b) moderate disease (N = 31). Filled squares represent patients who died. (c) Correlation of levels of IgG, IgA1 and IgA2 directed against S1+S2 protein of SARS-CoV-2 in the plasma of SARS-CoV-2-infected subjects with severe disease (N = 10) with the amount of ecDNA. Filled squares represent patients who died. Significance was tested with Spearman’s correlation coefficient. p < 0.05 was considered significant. Data are presented as scatter plots.

Fig 3: SARS-CoV-2-specific IgA2 levels correlate with disease severity. (a) Levels of IgG, IgA1 and IgA2 directed against S1+S2 protein of SARS-CoV-2 in the plasma of healthy control subjects (N = 15) as well as SARS-CoV-2-infected patients with no or mild disease symptoms (N = 34), moderate symptoms (N = 31) or severe symptoms requiring intensive care treatment (N = 17). Dotted lines represent mean + standard deviation (SD) of the healthy control group. (b) Levels of IgG, IgA1 and IgA2 directed against S1+S2 protein of SARS-CoV-2 in the plasma of COVID-19 patients with severe disease who recovered (N = 13) or died (N = 4). Dotted lines represent mean + SD of the healthy control group from (a). Significances were tested with the Kruskal–Wallis test followed by Dunn’s multiple comparison test for all groups vs. the control group (a) and a two-sided Mann–Whitney U test (b). * p < 0.05; ** p < 0.01; *** p < 0.001. Data are presented as scatter plots with mean ± s.e.m.

Fig 4: IgA1 and IgA2 are differentially glycosylated.a Schematic overview of glycosylation sites in IgA1 and IgA2. Bold letters indicate glycosylation sites present in both IgA subclasses. b Representative lectin blots of IgA1 and IgA2 isolated from sera of healthy donors using antibodies against IgA or lectins against the core structure of N-glycans (=lens culinaris agglutinin), terminal a2,6-linked sialic acid (=sambuccus nigra agglutinin), and terminal galactose (=erythrina christagalli lectin). c Quantification of the lectin blots; n = 6 donors. d–f Mass spectrometric quantification of sialyation, galactosylation, bisection, fucosylation, and the presence of noncomplex structures for the glycosylation sites N144/N131 d, N340/N327 e, N41 f, and N205 g; n = 12 donors. Significances were tested with paired two-sided Student’s t-test c–g or paired one-way ANOVA followed by Bonferroni correction for selected pairs of columns d–g. *p < 0,05; **p < 0,01; and ***p < 0,001. Data are presented as scatter plots with mean ± s.e.m. c or box plots with medians and inter-quartile ranges + whiskers ranging from min to max d–g. Source data are provided as a Source Data file.

Fig 5: A high IgA2 percentage in ACPA correlates with higher RA disease activity.a Amount of serum IgA, IgA1, and IgA2 as well as IgA2 percentage in patients with RA (n = 48) and healthy controls (n = 32). b IgA2 percentage of total IgA compared to ACPA in the serum of ACPA-positive RA patients (n = 38). c Correlation of the IgA1 and IgA2 amount, and IgA2 percentage in total serum IgA and IgA-ACPA with the disease activity score (DAS) 28 score of ACPA positive RA patients (n = 38). Significances were tested with two-sided Student’s t-test using Welch’s correction a, paired two-sided Student’s t-test b, or Pearson c. *p < 0,05; **p < 0,01; and ***p < 0,001. Data are presented as scatter plots with mean ± s.e.m. Source data are provided as a Source Data file.