Abstract

RBM15 is the 5′ fusion partner in RBM15-MKL1 (aka OTT1-MAL), a putative oncoprotein in non-Down syndrome infants and children with acute megakaryoblastic leukemia (FAB-M7) containing t(1;22). RBM15 belongs to the “spen” family, which is characterized by the presence of three RNA recognition motifs and a spen paralog and ortholog C-terminal (SPOC) domain. RBM15-homologous Drosophila proteins are involved in regulation of a variety of signaling cascades including MAPK, Wnt, Notch, cyclin E and Hox pathways, but the normal functions of mammalian RBM15 remain largely uncharacterized. We determined that Rbm15 is highly expressed in hematopoietic stem cells (HSCs) as well as T-lineage cells. To study Rbm15 specifically in the regulation of HSC function, we generated Rbm15 conditional knockout mice using the Cre-LoxP system to overcome embryonic death observed with a conventional knockout. Using an inducible Mx1-Cre transgenic line, we conditionally deleted Rbm15 (deletion efficiency ∼96–100%) in HSCs. Both the percentage and absolute number of long-term HSCs (Lin-Sca1+ckit+/Flk2-) were increased in Rbm15-deleted (Rbm15lx/lx;Mx1-Cre+) mice (0.23 ± 0.02% of total nucleated marrow cells [TNMC], 2.3-fold higher) compared to matched littermate controls (Rbm15lx/lx;Mx1-Cre-) (0.10 ± 0.01% TNMC) (P<0.0001, n=18 mice per group). By contrast, total white blood cells (WBCs) were significantly decreased in the peripheral blood (PB) of Rbm15-deleted animals compared to controls (Rbm15-deleted: 5.33 ± 0.40 × 103/uL, Rbm15-intact: 10.26 ± 0.49 × 103/uL; P <0.0001, n=31 per group). Functional analysis of Rbm15-deleted HSCs by competitive repopulation showed these cells to be markedly impaired in their reconstitution of hematopoiesis in lethally-irradiated recipient mice, with only 9.73 ± 2.32% donor-derived cells in the PB of transplanted animals compared to a 47.52 ± 7.26% contribution by donor cells from littermate controls (P=0.00015; n=10 mice per group; 1:1 ratio of donor:wild-type competitor marrow). The serial transplantation ability of Rbm15-deleted HSCs was also severely decreased, with a decline in their engraftment and contribution to the blood of recipient mice noted beginning with the 3rd round of transplantation, culminating in essentially complete failure to engraft in the 4th round (wild-type donor-derived PB cells in 4th round ∼60.95% vs. ∼2.71% Rbm15-deleted donor-derived PB cells; P=0.00003). These defects in HSC function may be due in part to altered HSC cell cycle status in the absence of Rbm15. For example, in representative experiments using Ki-67/Hoechst 33342 and BrdU/7AAD staining, 24.1% and 19.3% of Rbm15-deleted HSCs were found to be in G1 and G2/M phases, respectively - a marked increase compared to Rbm15-intact controls (8.3% and 4.3%, respectively); furthermore, Rbm15-deleted HSCs exhibited significantly less BrdU incorporation (6.8%) compared to control HSCs (27.1%) in in vivo labeling studies. Consistent with these altered cell cycle parameters, hematopoiesis in Rbm15-deleted mice recovered significantly more slowly than controls following 5-FU exposure, with only half the number of total nucleated marrow cells at day 9 and half the number of mature PB WBCs at day 12 following 5-FU as compared to Rbm15-intact littermates (p=0.003, n=3 per group). Collectively, these data demonstrate a critical role for Rbm15 in maintaining HSC integrity and suggest a yet-to-be fully elucidated function for Rbm15 in modulating HSC cell cycle kinetics.

Author notes

Disclosure: No relevant conflicts of interest to declare.