A NUP98-HOXD13 Fusion Gene Impairs Differentiation of T and B Lymphocytes.

Chromosomal translocations leading to NUP98 gene fusions are associated with a wide range of hematologic malignancies, including acute myelogenous leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL), and myelodysplastic syndrome (MDS). The NUP98-HOXD13 (NHD13) gene fusion was first identified in a patient with a MDS that progressed to AML. Recently, we used an NHD13 fusion gene to develop a mouse model of MDS that recapitulates all of the key findings of the human disease, including ineffective hematopoiesis leading to peripheral blood cytopenias, dysplasia, and progression to AML. In addition to the features noted above, we observed that the NHD13 mice were lymphopenic, and 10–30% of these mice, depending on the strain background, progressed to T-ALL. These findings prompted us to define the lymphocyte development of NHD13 mice. CBCs obtained from clinically healthy NHD13 mice showed lymphopenia (2.21 vs. 8.72 K/μL, p<0.01) compared with wild type (WT) littermate controls. This lymphopenia was due to a decrease in both T and B cells, as FACS analysis of peripheral blood (PB) from NHD13 mice revealed a marked decrease in CD4 single positive (SP) and B220+/IgM+ cells (p<0.05) compared to WT controls; similar findings were observed in the spleen (p<0.01). The percentage of CD8 SP cells was not different between the NHD13 and WT mice in either the PB or spleen. To investigate the cause for the B-cell lymphopenia, we determined the Hardy fractions of bone marrow B cells. Although the pro-B cell (B220+/CD43+, Hardy fractions A–C) populations showed no difference between NHD13 and WT BM, the NHD13 BM showed decreased pre-B and B cell (B220+/CD43−, Hardy fractions D–F) populations, suggesting impaired differentiation at the pro-B to pre-B stage. Thymi from NHD13 mice (n=7, median age=7 months) showed grossly decreased size and decreased total number of thymocytes (1.75×107 vs. 9.21×107, p<0.01). The CD4/CD8 DN population was markedly increased (p<0.001) and the CD4/CD8 DP population markedly decreased (p<0.001) in NHD13 compared to WT mice. In addition, there was a variable increase in the DN1 and DN2 population, as well as a decrease in the DN3 and DN4 population in thymi from the NHD13 thymus, suggesting a partial block at the DN2 to DN3 transition. To determine clonality of the NHD13 thymocyte population, we used degenerate RT-PCR to identify clonal TCRβ gene rearrangements. As expected, the WT thymi showed polyclonal TCRβ gene rearrangements. However, 5 of 6 NHD13 thymus samples showed clonal DJ rearrangements, with over half of the TCRβ rearrangements in the thymus showing an identical D-J junction, but distinct V-D junctions. This finding suggested that there was a massive clonal expansion of DN2 cells that had undergone a DJ rearrangement, but not completed a VDJ rearrangement, further supporting the contention of a partial block at the DN2 to DN3 transition. Interestingly, analysis of TCRβ gene rearrangements in NHD13 spleens showed no evidence of cells with clonal DJ rearrangements, suggesting that the thymocytes with clonal DJ rearrangements did not mature and migrate from the thymus. Finally, we noted that although there is a marked increase in DN thymocytes from the NHD13 mice, the T-ALL that developed in these mice were typically DP or CD4 SP, suggesting that a rare cell that “escapes” the T-cell differentiation block is susceptible to leukemic transformation. Taken together, these findings demonstrate that the NHD13 transgene inhibits lymphoid as well as myeloid and erythroid differentiation, and is oncogenic in lymphoid as well as myeloid cells.