Baf250a Is a Regulator of HSC Populations.

Abstract 701

Using Meis1 as a bait to screen the cDNA library from Hoxa9+Meis1 leukemia we identified Baf250a, a component of the chromatin remodeling Swi/Snf complex, as a Meis1-interacting protein. In mammalian cells, co-immunoprecipitation experiments confirmed this interaction and showed that only a minor fraction of cellular Meis1 interacts with Baf250a, suggesting a restriction of Meis1-Baf250a( Swi/Snf ) interactions to a specific cellular context. To reveal the role of Baf250a action in normal and leukemic HSC behavior we generated mouse strains expressing Baf250a protein lacking amino acids encoded by exons 2 and 3. Baf250a^+/Δ mice were born in expected proportions and exhibited no hematological abnormalities. Intercrosses of Baf250a^+/Δ (C57Bl/6J) adults yielded normal proportions of wild-type(WT), Baf250a^+/Δ and Baf250a^Δ/Δ day E14.5dpc embryos, but no viable Baf250a^{Δ /Δ} newborn mice. These, however, could be obtained in a mixed C57Bl/J6 x Sve129 background, suggesting existance of a mouse strain-specific modifier(s) of Baf250a activity. Examination of E14.5 dpc fetal liver (FL) cell populations obtained from Baf250a^{Δ /Δ} and WT littermates showed that cellularity and clonogenic progenitor content of Baf250a^Δ/Δ FL was comparable to the WT controls, and that erythoid, megakaryocytic and myeloid cell lineages developed normally in the presence of Baf250a^Δ. Transplantation experiments in conditions of limit dilution revealed that Baf250a^Δ/Δ FL comprised 8-12-fold higher HSC numbers than WT controls. After transplantation into adult WT recipients the Baf250a^{Δ /Δ} and WT HSCs exhibited comparable proliferation potential (output of mature cells per individual transplanted HSC), suggesting that of Baf250a^Δ conferred no repopulation advantage to mutant HSC, but rather enhanced the ability of fetal liver microenvironment to support expansion of HSC populations. To explore this possibility we next examined the ability of Baf250a^{Δ /Δ} stromal cell cultures to support the in vitro self-renewal of stem/progenitor cell populations identifiable as late cobblestone area forming cells (CAFCday28), and found that Baf250a^Δ/Δ stromal cell layers increased the numbers of cobblestone areas 4-6-fold compared to controls, and increased the probability of CAFCday28 self renewal divisions (n=24, p<0.05) compared to controls as determined by their ability to form secondary late cobblestone areas. These observations thus suggested that Baf250a acts as a negative regulator of fetal HSC populations. However, we also found that Baf250a^Δ/Δ genetic background accelerated the onset of Hoxa9 (n=7, mean latency 140 and 90 days for WT and Baf250a^Δ/Δ, respectively, p<0.05) and NUP98-Hoxa9 + Meis1 (n=8, mean latency 125 and 49 days for WT and Baf250a^Δ/Δ, respectively, p<0.005) induced AML, implying that Baf250a may also act as an intrinsic regulator of primitive hemopoietic cells. Together, results of our experiments identify a novel mechanism that limits the expansion FL HSC populations regulated by the Baf250a, and suggest a tumor suppressor function for this protein in Hoxa9-induced leukemogenesis.