Analysis of Immunoglobulin Heavy Chain Variable Genes In Burkitt Lymphoma Uncovers An Antigen Role In the Pathogenesis of the Disease

Abstract 4135

The analysis of the immunoglobulin heavy chain variable (IGHV) genes in B-cell derived tumors can yield relevant pathogenic information; it contributes to the definition of the normal cell counterpart and in some neoplasias allowed the identification of subsets of stereotyped B cell receptors associated to different clinical/phenotypic features and outcome. According to the WHO 2008 Classification, Burkitt lymphomas (BL) derive from either germinal center (GC) or post GC B cells. This study was aimed at a comprehensive investigation of the IGH genes in BL; for that, 27 samples of BL were studied, including both pediatric (n=15) and adult (n=12) cases from the 3 clinical variants of the disease, namely endemic (n=1), sporadic (n=22) and immunodeficiency-associated (n=4).

All the samples analyzed harbored in frame IGHV-D-J rearrangements, the great majority (96.3%) being functionally productive. The comparison of the IGH genes usage in BL with the normal repertoire found in CD5 neg B cells detected no differences concerning IGHJ and IGHD usage. However, the usage of IGHV genes was significantly different from that of the normal B cell counterpart (p=0.21). Genes like IGHV3-21 and IGHV5-a were only found in BL samples, with a frequency of 11.1% and 3.7% respectively. Other genes were overrepresented in BL with respect to normal B cells, namely IGHV2-70 (7.4% vs 1.3%), IGHV3-30 (11.1% vs 5.2%), IGHV4-34 (7.4% vs 4.3%), IGHV4-39 (11.1% vs 5.2%) and IGHV4-59 (14.8% vs 9.1%). These results pointed to a biased used of certain IGHV genes by BL cells. No preferential V-D-J rearrangement was detected and only 3 samples had common HCDR3 features, although with different V-D-J rearrangements.

IGHV mutational load was calculated as the percentage of germline identity (GI). Most of the BL samples (70.4%) harbored mutated IGHV genes (<98% GI), whereas borderline/minimally mutated IGHV genes (98-99.9% GI) were detected in a 22.2% of the samples. Only 2 out of 27 cases (7.4%) presented unmutated IGHV genes (100% GI). Interestingly, the pattern of the aminoacid (aa) substitutions introduced by the SHM process revealed a precise targeting: 81.5% of the cases showed the same aa replacements indicating that common antigens must be implicated.

Intraclonal diversity (ID) was assessed by analysis of the subcloned nucleotide sequences. Only confirmed mutations, i.e.mutations observed more than once in the subclones from the BL case, were considered. Under this definition, 48.1% of BL cases had ongoing nucleotide mutations. Nevertheless, analysis of aa replacements introduced by unconfirmed nucleotide mutations showed that some aa substitutions were shared by other cases. Thus, considering that those unconfirmed mutations were confirmed by other case, these changes were acknowledged as real mutations, this raising the percentage of BL samples with ID to a 88.9%. Following this, it could be concluded that the majority of the cases presented ID, indicating that their normal counterpart is a centroblast experiencing the GC reaction.

It has been previously shown that the IGHV3 subgroup harbors the Staphilococcal protein A (SpA) binding motif. In this sense, we found that a 44.4% of the BL cases used a gene from this subgroup. Interestingly, the SpA binding site was preserved in all of the IGHV3 aa sequences analysed. SpA is the prototypic B-cell superantigen and the fact that its binding motif was present in the main IGHV gene used by BL cells in this series suggest that superantigens could play a biological role in this disease.

In conclusion, this work allowed the confirmation of the postulated normal counterpart for BL being a centroblast experiencing the GC reaction. More, the biased use of IGHV genes and the recurrent hypermutations that were detected suggest a role for common antigens in the selection of the clonogenic progenitors. Finally, the conserved SpA binding motif found in BL cases using IGHV3 genes indicates that superantigens may also be implicated.