FBXO11, a Regulator of BCL6 Stability, Is Recurrently Mutated in Burkitt Lymphoma

INTRODUCTION: We recently described inactivating mutations of the FBXO11 gene in Diffuse Large B Cell Lymphoma (DLBCL). One major function of FBXO11 is to regulate BCL6 stability as well as other targets such as SNAIL. BCL6 acts as an oncogene in several human B-cell lymphomas and its expression levels can be increased by several mechanisms including chromosomal translocations, point mutations in the promoter region and reduced degradation by inactivation of FBXO11. Thus, FBXO11 acts as an oncosuppressor in DLBCL by promoting the accumulation of BCL6. However, a clear and complete picture of the distribution of FBXO11 mutations in different lymphoma subtypes is missing, and the in vivo functions of FBXO11 in normal tissue and lymphoma development are completely lacking. In the present work, we investigated the frequency and distribution of FBXO11 mutations in an extended panel of human BCL6 positive lymphoma and we functionally validated novel FBXO11 mutations for their ability to induce BCL6 degradation.

METHODS: We sequenced the entire FBXO11 coding sequence by classical Sanger sequencing in 100 cases of Follicular Lymphoma (FL), 36 cases of Burkitt Lymphoma (BL), 8 BL cell lines and 8 Anaplastic Large cell lymphoma (ALCL) cell lines, which are typically BCL6-positive lymphomas. Moreover, we sequenced 50 cases of Marginal Zone B cell Lymphoma (MZL), which show variable expression of BCL6. To investigate whether the FBXO11 mutations interfere with FBXO11 activity we generated complementary DNA constructs containing these mutations. Wild-type FBXO11 or FBXO11 mutants were expressed in HEK-293T cells with constructs expressing BCL6 or SNAIL. Twenty-four hours after transfection, HEK-293T cells were treated with cycloheximide and harvested at different time points to check substrates expression and degradation by immunoblotting.

RESULTS: BL carried the highest frequency of FBXO11 mutations (12/44 cases analyzed, 27%), whereas FL and MZL were very rarely affected by FBXO11 mutations (1/100 FL and 1/50 MZL). The analysis identified several mutations, either located in the coding sequence mostly in the CASH (functional) domains or located in intronic sequences controlling exon splicing. Sequence changes included missense mutations (n=8), deletions (n=2), a frameshift deletion introducing a premature stop codon (n=1) and nucleotide substitutions at consensus splice donor sites (n=2) and acceptor site (n=1). Remarkably, these last mutations represent the first report of splice site mutations affecting the FBXO11 gene. By cDNA amplification and sequencing, we demonstrated skipping of entire exons: the splice site mutations located at the donor site of exon 15 and 19 lead to the loss of the exon 15 and 19 respectively, while the splice site mutation affecting the acceptor site of exon 16 results in the deletion of the adjacent exon 16. Two of the point mutations (K631N and Y122C), affecting one case of FL and MZL respectively, were also found in the previous panel of DLBCLs (Duan et al, Nature 2012). Both cases consisted of a low grade component adjacent to an area of high grade transformation. By microdissection, we separately studied the two components of each of these cases to demonstrate that the mutation was present only in the high grade component. Finally, to validate all the mutations found in BL, we generated mutant constructs corresponding to each of them. By this approach, we showed that all the mutations identified are loss of function variants because they impaired the FBXO11 ability to promote BCL6 and SNAIL degradation.

CONCLUSIONS: We identified several FBXO11 novel mutations in a large panel of BCL6-positive human lymphomas. All the mutations identified were inactivating the FBXO11 ability to induce BCL6 and SNAIL degradation. In contrast, mutations of FBXO11 are rare in low grade FL and MZL and, when present, are associated with high grade transformation. Together with our previous findings, this study showed that FBXO11 is mostly mutated in aggressive lymphomas such as DLBCL and BL, thus suggesting that FBXO11 mutations could contribute to the de-novo onset or transformation into high grade lymphoma.