IGHV Gene Replacement in B-Cell Chronic Lymphocytic Leukemia (B-CLL) Occurs at a Frequency Similar to That in Normal B Cells and May Augment Clonal Expansion by Permitting Autogenic/Microbial Clonal Stimulation.

B-CLL cells frequently express antigen receptors with varying degrees of self-reactivity. In this context, it is likely that the precursors of leukemic clones have attempted to reconfigure their receptors to eliminate or reduce affinity for autoantigen. IGHV replacement is a form of receptor reconfiguration that retains the existing IGHD/J rearrangement and swaps the originally rearranged IGHV gene for another upstream IGHV gene, exploiting cryptic heptamer-like sequences in the coding regions of most human IGHV genes. In this study, we explored the extent of IGHV replacement in 583 IGHVDJ rearrangements from patients with B-CLL and compared these findings to normal B cells. B-CLL IGH sequences used IGHV1/2/3/4/5/6/7 subgroup genes in 144/33/255/122/14/9/6 cases, respectively. IGHV3 and IGHV4 genes were used predominantly in mutated (M) sequences (136/255 and 75/122 cases respectively); in contrast, 83% (120/144) of IGHV1 genes were used in unmutated (UM) sequences. IGHV replacement in B-CLL was examined using a search method, also applied in the normal repertoire, looking for the presence of pentameric footprints, which are short stretches of 3′ nucleotides from the original IGHV that remain in the IGHVD joint after each round of IGHV replacement. This method uses a set of 33 pentamers deriving from the 3′ends of human IGHV genes carrying cryptic heptamers to identify potential IGHV replacement products. After alignment with sequences in the IGHVD junctions in B-CLL, only exact matches were accepted. At least one pentameric footprint consistent with IGHV replacement was detected in the IGHVD junction of 34 (19 M/15 UM) of 583 B-CLL cases (5.8%). Twenty-seven pentamers were detected in this cohort of sequences. B-CLL rearrangements with evidence of IGHV replacement footprints had longer HCDR3 regions (median,19 amino acids) compared to the series as a whole (median,17 amino acids). The most frequent IGHV genes that probably replaced a previously rearranged IGHV gene were IGHV1-69 (6 cases/5UM), IGHV3-23 (6 cases/1UM), IGHV3-7 (4 cases/1UM), IGHV3-33 (3 cases/2UM). In 30/34 cases (88%), the amino acids encoded by these footprint sequences in the HCDR3 where positively or negatively charged, a finding associated with autoreactivity. These residues likely contribute to antigen specificity and affinity and can be important for B-CLL cells, which are antigen-experienced, activated cells. In conclusion, IGHV replacement in CLL appears to occur at a frequency (5.8%), which is not inherently different from that reported in the normal B-cell repertoire (5%). Although IGHV replacement is ostensibly initiated to modify autoreactivity, its products still often resemble autoantibodies (increased HCDR3 length, aminoacid composition and charge). However since some of these HCDR3 properties also exist in anti-microbial mAbs, their retention after secondary rearrangements may be a protective as well as a potentially harmful mechanism. Such a selection for microbial reactivity supports the notion that B-CLL derives from cells expressing mAb with shared reactivity between autoantigens and microbes. In support of this possibility is the finding that a UM-IGHV1-69 recombinant mAb with evidence of IGHV replacement reacted against 5 different bacterial strains.