Inactivation Of BANK1 By a Novel IGH-Associated Translocation and 5’ Hypermethylation In B-Cell Lymphomas

A novel IGH-associated reciprocal translocation, t(4;14)(q24;q32), was identified, along with trisomy 9, in 20 of 20 metaphases by conventional karyotyping in a case of malignant gastric post-transplant lymphoproliferative disorder (PTLD). Cloning of the translocation site by inverse PCR identified BANK1 (B-cell scaffold protein with ankyrin repeats 1), a B-cell-specific adaptor protein with putative functions in B-cell receptor and CD40 signaling, as a novel IGH translocation partner. The breakpoints were located at the Sα region of IGH and intron 1 of BANK1. The translocation juxtaposed the two genes in opposite orientations, and surprisingly, resulted in transcriptional inactivation of BANK1 as a result of dissociation of the major BANK1 promoter. While BANK1 isoforms were expressed in all tonsillar B-cells, with lower levels (∼ 5 fold) in the germinal centers (GC) compared to naïve and memory B-cells, transcription from the major promoter in the tumor was absent and transcription from the minor promoter was reduced 50% relative to GC B-cells, suggesting that the non-translocated BANK1 allele was also inactivated. The total BANK1 expression was very low (∼10% of normal GC B cells) and crytic promoter activation was not identified. Several genes (PPP3CA, MIR1255A, FLJ20021 and SLC39A8), located 180 to 440 kb away from BANK1, were analyzed for mRNA expression; there is no significant activation in any of these genes, further supporting that BANK1is indeed the target gene affected by the translocation.

Interphase FISH using break-apart BANK1 probes confirmed breakpoint in the index case but did not identify translocations in additional 15 PTLDs and 68 diffuse large B-cell lymphomas (DLBCL), implying that BANK1 translocation may be a rare event. To determine if BANK1 inactivation may occur in B-cell lymphomas by other mechanisms, 23 B-cell lymphoma cell lines, including 8 Burkitt lymphoma (BL), 9 diffuse large B cell lymphoma (DLBCL), 3 primary effusion lymphoma (PEL), and 3 classical Hodgkin lymphoma (cHL) were bisulfite sequenced to assess the methylation status of 37 CpG dinucleotides in a 436 base-pair region at the 5’ end of BANK1, which extends across exon 1 into the 5’ portion of intron 1. High level of methylation (>60% methylation on average among all CpGs) was seen in all 3 cHL and 2 of 3 PEL cell lines. Regional methylation was seen in 3 of 8 BL lines and 1 of 3 PEL lines. No hypermemethylation was identified in the DLBCL lines or in normal tonsils. Hypermethylation was associated with almost complete silencing of BANK1 transcription. In the DLBCL lines and BL lines without BANK1 hypermethylation, BANK1mRNA expressions were variable, ranging from <5% to 130% of GCB cells.

To confirm that BANK1 hypermethylation is present in primary lymphoma cases, methylation status of 17 of the 37 CpGs were assessed in 23 cHL cases using en bloc formalin-fixed, paraffin-embedded materials and also laser-capture micro-issected Hodgkin/Reed-Sternberg (HRS) cells. There was evidence of BANK1 hypermethylation in the tumor cells in 9 of 23 cHL. Tumor cell specificity of BANK1 hypermethylation was further confirmed in 4 cHL cases using micro-dissected HRS cells. HRS cells were negative for BANK1 in 28 of 29 cHL cases examined by immunohistochemistry, suggesting that other mechanisms other than DNA methylation may be responsible for silencing BANK1expression.

To investigate whether BANK1 has biological effects on B-cells related to lymphoma development, exogenous BANK1 was re-introduced to BC3, a PEL cell line showing marked BANK1 hypermethylation with absence of BANK1 expression. We established a stable doxycycline-inducible BC3 cell line expressing BANK1. Inhibition of cell growth was observed 2 to 3 days after doxycyline induction, and the number of viable cells with transfected BANK1 was only 25% compared to BC3 cells carry vehicle alone at day 6. An analysis of 5-bromo-2’ deoxyuridine (BrdU) incorporation after 48 hours of doxycline induction revealed that the fraction of cells in S-phase was reduced by 50% in the BANK1 transfectants, suggesting that BANK1has a negative effect on cell proliferation in these B cells.

In summary, we have identified a novel IGH translocation partner and provide an example of an unusual consequence (gene inactivation) of IGH-associated translocation. We provide for the first time evidence of a potential role of BANK1 down-regulation in the development of B-cell lymphomas.