E2A-HLF Fusion Transcription Factor Mediating Aberrant Expression of LMO2 in B-Precursor ALL with t(17;19).

LMO2 is a transcription factor critical for hematopoiesis, and its ectopic expression was reported in T-ALL having translocations such as t(11;14) in which the TCRα/δ locus is fused to LMO2. Moreover, aberrant activation of LMO2 via integration of the retroviral vector was reported in T-cell malignancies developed in the patients who received IL2Rγc gene therapy for X-SCID. Consistently, transgenic mice of LMO2 under the control of thymocyte-specific promoter developed T-cell lymphoma. Although these demonstrate that LMO2 is oncogenic in T-lymphoid cells, significance of LMO2 in B-precursor ALL remains totally unclarified. Gene expression of LMO2 is regulated by distal and proximal promoters: the former generates full-length transcript and the later generates transcript lacking exons 1–3, both of which encode the same open reading frame. Recently, the binding site of PAR transcription factors was identified as a critical region for distal promoter activity of LMO2 in hematopoietic cells. A family of PAR transcription factors belongs to the bZIP factor and includes HLF, TEF and DBP. Since E2A-HLF fusion derived from t(17;19) in B-precursor ALL has bZIP domain of HLF and transactivation domain of E2A, it is feasible to postulate that E2A-HLF might induce aberrant expression of LMO2 through its distal promoter. To confirm this, we first analyzed the expression of LMO2 transcripts derived from both distal and proximal promoters in t(17;19)-ALL cell lines by real time RT-PCR with the primers for exon 4/5. GAPDH expression was used as an internal control. Of note, all of 4 t(17;19)-ALL cell lines expressed LMO2 transcripts at an equivalent level to those in T-ALL cell lines with t(11;14). Moreover, the level of LMO2 transcripts in t(17;19)-ALL cell lines was significantly higher than that in 28 other B-precursor ALL cell lines including Ph1-ALL (n=6), t(1;19)-ALL (n=7), and MLL⁺-ALL (n=9). Similar results were obtained by real-time RT-PCR using primers for exon 1/2, which specifically detect the LMO2 transcripts derived from the distal promoter. We next analyzed LMO2 protein expression by Western blotting using α-tubulin expression as an internal control. LMO2 was expressed in all of 4 t(17;19)-ALL cell lines as high as that in T-ALL cell lines with t(11;14), whereas it was almost undetectable in 16 out of 28 other B-precursor ALL cell lines, indicating that overexpression of LMO2 in t(17;19)-ALL is a unique property among B-precursor ALLs. Next, to directly test the possibility that E2A-HLF induces LMO2 expression, we transfected E2A-HLF into the B-precursor ALL cell line 697, whose LMO2 expression level is very low, using Zn-inducible vector. When E2A-HLF expression was induced by Zn, LMO2 transcripts as well as LMO2 proteins were immediately upregulated. The induction of LMO2 expression by E2A-HLF was dependent on the DNA-binding and transactivation activity of E2A-HLF, since the mutants lacking either the DNA-binding or transactivation domains failed to induce LMO2 expression. Further, direct binding of E2A-HLF to the PAR binding site in the distal promoter was demonstrated in EMSA. These observations not only indicate that E2A-HLF directly induces aberrant expression of LMO2 in t(17;19)-ALL but also suggest that overexpression of LMO2 might involve in the leukemogenetic potential of E2A-HLF.