Somatic Hypermutation In Stereotyped Subset 4 BCRs/mAbs of CLL Patients, Expressing IGHV4-34 gene, Edit Anti-DNA Reactivity

Abstract 2444

Extensive analysis of immunoglobulin gene sequences from chronic lymphocytic leukemia (CLL) patients reveals that ~30% carry stereotyped B-cell antigen receptors (BCRs) with highly homologous HCDR3s, over-represented in selected IGHV/IGHD/IGHJ rearrangements involving such IGHVs as 1–69, 4–34, and 3–21. This finding along with other structurally unique features of CLL BCRS suggests that antigens or superantigens or both may play an active role in the disease.

The mutational status of IGHV genes is one of the most powerful prognostic molecular markers in CLL. Patients with IGHV genes with less than 2% differences from the germ line (“unmutated”; U-CLL) have a more aggressive clinical course compared to those with more than 2% differences (“mutated”; M-CLL). In subsets of sequences with stereotyped HCDR3s certain amino acid changes appear to be “CLL-biased”. One of the most distinctive SHM patterns has been described in sequences expressing stereotyped IGHV4-34 BCRs belonging to Subset 4, that are also characterized by long HCDR3s enriched in basic residues (especially arginine).

The IGHV4-34 gene encodes antibodies which are autoreactive due to their recognition of the N-acetyllactosamine (NAL) antigenic determinant of the I/i blood group antigen, also expressed on a 220-kDa CD45 B cell-specific isoform; this recognition is mediated in a superantigenic fashion via a motif in HFR1 of the IGHV4-34 sequence. In addition, IGHV4-34 antibodies with HCDR3 sequence motifs enriched in basic (electropositive) amino acids have been shown to react against B cells and DNA, although HCDR3 electropositivity per se does not universally endow IGHV4-34 B cells with this property.

Along these lines, we wanted to investigate if the modifications of subset 4 IGHV4-34 by SHM in precursors of the CLL clones could have affected the binding properties if the respective IG receptors, especially against B cells and DNA. We therefore recombinantly expressed 4 mAbs with IGHV4-34 rearrangements, from our CLL patients sample library; mAbs CLL183, CLL240, CLL342 belong to stereotypic subset 4 and have mutated heavy chain immunoglobulin genes, whereas CLL141 expresses an unmutated gene without belonging to any described subset.

As previously reported, we tested reactivity of these mAbs with apoptotic and viable B and T cells. All 4 mAbs reacted significantly only with viable cells; the 3 mAbs belonging to subset 4 bound only viable B cells, with a preference for naïve B cells. We also tested the reactivity of these mAbs with single stranded DNA (ssDNA) and double stranded (dsDNA). We used common assays for identifying nuclear autoantigens in autoimmune disorders: 1) for binding total anti-nuclear antigens, human epithelial cell line HEp-2, and 2) for binding dsDNA, Crithidia Luciliae (CL). None of these tests had a strong positive result.

To identify a role played by SHM in editing the anti-DNA reactivity, we chose to revert to germline 9 amino acid positions mutated in the IGHV and 1 in the IGKV sequences of CLL240. We either made revertant antibodies retromutated at single positions or at all mutated positions (completely unmutated and germline-like). Enzyme-linked assay (ELISA) for the semi-quantitative detection of ssDNA and dsDNA showed that the complete revertant antibody regained the ability to strongly bind ssDNA and partially dsDNA. The light chain did not appear to play a crucial role in the reacquisition of the binding, because the recombinant mAb expressing only the IGKV in the germline sequence did not bind ssDNA or dsDNA, as did the original recombinant mAb. Some of the revertants also showed a stronger binding to normal human B cells, with a few of them acquiring a partial ability to also bind T cells. The role of the single revertants is being further analyzed.

Few IGHV4-34 CLL sequences are altered in the specific HFR1 motif responsible for NAL binding. This suggests that these IGHV4-34 expressing CLL cells could still bind and be stimulated for clonal expansion by NAL epitopes present in self and exogenous antigens. However, the findings of the present study also indicate that the distinctive modifications introduced by SHM in the stereotyped IGHV4-34 BCRs of subset 4 CLL likely represent a censoring mechanism for avoiding intense self-reactivity, mediated by the arginine-rich HCDR3s, and subsequent clonal deletion.