Balanced Wnt5a-Mediated Signaling Is Necessary for Normal Proliferation of Primitive Hematopoietic Cells.

Abstract 2533

Poster Board II-510

Multiple members of the Wnt family of ligands have been implicated in the regulation of self-renewal and proliferation of hematopoietic stem cells (HSCs). Previously, we have observed that ex vivo expansion of HSCs in the presence of recombinant murine Wnt5a (rmWnt5a) resulted in increased hematopoietic repopulation. Based on these data, we hypothesized that Wnt5a is necessary for normal function of HSCs and hematopoietic progenitors (HPCs). Since Wnt5a deficiency (Wnt5a^−/−) is perinatal lethal in vivo, we tested this hypothesis using in vitro Dexter stroma cultures established using whole bone marrow. To determine the ability of Wnt5a to support hematopoiesis in the context of the adult hematopoietic microenvironment, we cultured lineage-negative (lin⁻) HPCs on irradiated bone marrow stroma in the presence of 5 μg/ml Wnt5a-neutralizing antibody (Wnt5a-Ab). After two weeks, we observed that hematopoietic cells cultured on untreated stroma contained 4.8-fold more myeloid CFU (33.1 ± 12.3 CFU/10⁴ cells) than cells cultured on Wnt5a-Ab stroma (6.9 ± 0.7 CFU; n = 3, p < .01). A similar difference was observed after 4 weeks (control: 16.0 ± 7.2 CFU/10⁴ cells; Wnt5a-Ab: 1.3 ± 2.3 CFU; n = 3; p = .03). In the converse experiment, lin⁻ HPCs were cultured on stroma in the presence of Wnt5a conditioned medium (Wnt5a-CM). We observed after two weeks that hematopoietic cells cultured on stroma with control-CM contained 4.4-fold more myeloid CFU (29.8 ± 13.5 CFU/10⁴ cells) than cells cultured on Wnt5a-CM stroma (6.8 ± 2.2 CFU; n = 4, p = .02). Together, these data indicate that Wnt5a-mediated signaling must be balanced in order for normal hematopoiesis to occur. To determine if the effects of Wnt5a required the presence of developmental stage-specific factors, we established stroma cultures from fetal spleens harvested from E17.5 Wnt5a^−/− mice and littermate controls and seeded them with lin⁻ HPCs. We observed that hematopoietic cells cultured on control fetal spleen stroma contained 12.5-fold more myeloid CFU (72.6 ± 21.6 CFU/10⁴ cells) than cells cultured on Wnt5a^−/− fetal spleen stroma (5.8 ± 5.8; n = 6, p < .001) after two weeks. These data indicate that the effect of Wnt5a on hematopoietic cells is independent of the developmental stage of the surrounding microenvironment. To determine if the effects of Wnt5a were due to regulation of proliferation or differentiation of hematopoietic cells, we cultured bone marrow cells in cytokine-supplemented methylcellulose in the presence of 300 ng/ml rmWnt5a. We observed a 5.5-fold decrease in the number of myeloid CFU formed in cultures with rmWnt5a (14.1 ± 3.8/10⁴ cells) compared to control (77.6 ± 5.1 CFU; n = 3; p < .001), suggesting that Wnt5a could be regulating both processes. In the converse experiment, we cultured bone marrow cells for 4 days in cytokine-supplemented serum-free media with the same dose of rmWnt5a after which equal numbers of cells were plated in rmWnt5a-free methylcellulose. We did not observe any difference in CFU frequency between control (19.3 ± 4.2 CFU/10⁴ cells) and rmWnt5a (24.0 ± 2.6 CFU; n = 3) cultures, indicating that treatment with rmWnt5a inhibited hematopoietic proliferation but not differentiation. To identify the mechanism by which Wnt5a regulates HSC and HPC proliferation, we analyzed potential Wnt5a-mediated signaling pathways. We observed that Wnt5a induced intracellular Ca²⁺ (iCa²⁺) flux in HSCs (defined as lin⁻, Sca-1^HI, c-kit^HI; LSK). Previous studies have shown that Wnt5a-mediated induction of iCa²⁺ can result in activation of the NFAT family of transcription factors. Since NFATc1 promotes quiescence of hair follicle stem cells and is expressed in HSCs, we hypothesized that the effects of Wnt5a required activation of NFAT family members. We cultured bone marrow cells with Wnt5a-CM in the presence of cyclosporine A (CsA), which inhibits activation of NFAT factors. In agreement with our earlier findings, we observed that culturing bone marrow cells in Wnt5a-CM increased the percentage of quiescent (defined as Ki-67-) LSKCD34⁻ HSCs (79.7 ± 3.3%) compared to control-CM (55.0 ± 1.6%; n = 3; p < .001). This increase was inhibited by CsA (69.4 ± 2.6%; n = 3; p = .01 compared to Wnt5a-CM alone). In conclusion, our data point to a role for Wnt5a in regulating HSC and HPC proliferation and that this function may require the activation of NFAT transcription factors.