Figure Legends
Figure 1 Frequency of 2S albumin-binding B-cells
Frequency of 2S albumin-binding B-cells, double-positive for allergen-tetramer (PE and APC) staining, expressed as percentage of the total B-cell number acquired from the respective sample, indicated with the median. Each frequency is labelled with the respective patient number and the number of successfully amplified corresponding heavy and light V(D)J gene transcripts in relation to sorted B-cells. Statistical significance was proven with the Kruskal-Wallis test and corrected for multiple testing using the Dunn’s multiple comparison test (adjusted p-value); all = allergic, tol = tolerant, n-a = non-atopic
Figure 2 Specificity and functionality of heterologously expressed mAbs
A Concentration-dependent (1 to 10 µg/ml) binding of heterologously expressed mAbs (n = 42) to Ara h2 and 6 defined as OD values corrected for their binding to transferrin; filled dots indicate mAbs expressed as IgE and open dots mAbs expressed as IgG1; bold dash line is the cut-off for IgE mAbs and non-bold dash line the cut-off for IgG1 antibodies
B Binding of heterologously expressed mAbs to Ara h2 and 6 (n = 24) at 10 g/ml defined as OD values
corrected for their binding to transferrin; filled dots indicate mAbs expressed as IgE and open dots mAbs expressed as IgG1, bold dash line is the cut-off for IgE mAbs and non-bold dash line the cut-off for IgG1 mAbs; black = allergic, grey = tolerant, light grey = non-atopic
C Upregulation of CD63 upon loading of human basophils with mAb mixture and subsequent stimulation with Ara h2 and 6 in a concentration-dependent manner (5 to 1000 ng/ml); the dots indicate the mean of duplicate measurements and their range; Mix 1 = Pt-6-2, Mix 2 = Pt-6-2 + Pt-6-17, Mix 3 = Pt-6-2 + Pt-3-2 + Pt-6-11, Mix 4 = Pt-6-17, Pt-3-2, Pt-6-11, Mix 5 = Pt-1-32, Pt-6-11, Pt-6-6, Pt-6-8, Mix 6 = NA-1-10 + Pt-4-28 + Pt-4-34 + Pt-6-2, Mix 7 = Pt NA-1-10 + Pt-4-28 + Pt-4-34, Mix 8 = Pt-1-32 + Pt-6-15 + Pt-6-21, stripped basophils before loading, Rituximab expressed as IgE; Supplementary, Figure S4: comparison to stimulation with cow’s milk extract, dash line: threshold level set to 5%
Figure 3: Descriptive gene analysis and motif analysis of HCDR3 regions
Gene lineage analysis of successfully sequenced heavy chain gene transcripts defined as productive (allergic n = 151; tolerant n = 68; non-atopic, n = 51), irrespectively of successfully sequenced corresponding light chain gene transcript
A Isotype distribution split for allergic (all), tolerant (tol) and non-atopic (n-a) donors
B Numbers of non-silent mutations expressed with the median (95% CI) and split for allergic (all), tolerant (tol) and non-atopic (n-a) donors
C Clonotypes were defined as identical V gene, J gene and HCDR3 length and related clones were aligned using ClustalW2. The IgE mAb of Pt-10 was highly related to an IgG1 antibody of Pt-4.
D Tendency of non-redundant HCDR3 regions (allergic n = 88; tolerant n = 55; non-atopic, n = 48) to form double-bended structures on amino acid level expressed as Kidera factor 5 in allergic (all), tolerant (tol) and non-atopic (n-a) donors (median and 95% CI). Correction for multiple measurements (Bonferroni) resulted in no statistically significant difference. More physicochemical properties of the HCDR3 region are shown in the Supplementary, Figure S5 and 6.
E HCDR3 sequence motifs were analyzed by Levenshtein distances and hierarchical clustering (detailed description in the method section) and sequences with distances ≤ 5 were grouped into one motif. More detailed description of the motifs is presented in Table 3.