Clonality and isotype usage spanning different PC longevity populations. (A) Examples of lineage trees created based on reconstructed BCR sequences for DGP- or TG2-specific clones containing different isotypes and longevities from an untreated (UCD2136) and treated (TCD1934) patient. IGHV3-23 and IGHV3-23D are grouped together as IGHV3-23(D). (B) Log₁₀ index-sorting values for CD19 and CD45 for highlighted TG2-specific (left) and DGP-specific (right) clone groups for two representative patients (CD2136 and CD2106). Negative values were converted to 1 before they were logarithmized. Only the largest clones are shown, with members of each clone in the same color. (C) IgA1:IgA2 ratio and IgM usage in controls for which a natural composition of PC longevities was sorted (left, n = 3). IgA1:IgA2 ratio (center) and percent IgM PCs (right) between PCs of different longevities in all controls (n = 5) are color coded for each individual. No IgA2 PCs were detected among the only nine long-lived PCs from CD2146, and no IgA1:IgA2 ratio was plotted for long-lived PCs for this patient. P values were calculated with a paired Student’s t test. (D) IgA subclass usage (left) and IgM usage (right) between disease-specific and other PCs in CeD. Each point represents IgA1:IgA2 ratio (left) or percent IgM PCs (right) in an individual based on reconstructed BCR sequences; data from the same individual are connected with a gray line. All four UCeD and three TCeD patients were included in the comparisons between TG2-specific PCs and PCs of unknown specificity; only patients from which DGP-specific PCs were sorted (two UCeD patients, two TCeD patients) were included in the comparisons including DGP-specific PCs. P values were calculated with a paired Student’s t test. *, P < 0.05; **, P < 0.01; ns, not significant.