Introduction: Acute lymphoblastic leukemia (ALL) is the most common form of childhood malignancy, accounting for 25% of all childhood cancers. Although great strides have been made in the treatment of childhood leukemia, close to 20% of patients will have resistant disease eventually leading to death. In approximately 40% of relapses, the central nervous system (CNS) is involved, alone or in combination with other sites, thus representing a major clinical concern. We recently demonstrated that ActivinA, a pleiotropic cytokine that belongs to the TGF-β superfamily, plays a crucial role within the leukemic niche. In addition, we showed by using a xenograft mouse model of human B-Cell Precursor Acute Lymphoblastic Leukemia (BCP-ALL), the ability of ActivinA to enhance both bone marrow (BM) engraftment and metastatic potential into extra-medullary sites of leukemic cells. Interestingly, an increased leukemic burden was observed in the CNS of mice receiving ActivinA-treated cells (Portale et al., 2019). In this study, we attempted to elucidate how leukemic NALM-6 cell line pre-treated or not with ActivinA modifies the CNS microenvironment by using an in vitro model of organotypic cortical brain slices. Methods:200-µm-thick brain slices from C57BL/6J or CX3CR1+/GFP (that express GFP in microglial cells) postnatal (P1-3) mice were obtained using a vibratome and placed onto a cell culture insert. After 7 days of recovery from cut, slices were microinjected with NALM-6 cells expressing iRFP670 pre-treated or not with ActivinA (50ng/mL for 24h). Proliferation of NALM-6 was quantified as the area of iRFP670 fluorescent signal in the brain slices up to 72h after injection. Brain slice cell death was measured by propidium iodide incorporation assay. Microglia activation was evaluated in CX3CR1+/GFP slices by longitudinally quantifying GFP signal up to 7 days post injection. In addition, microglia morphology was evaluated by shape descriptor parameters at 3 and 7days post-injection. Lastly, qRT-PCR assay to evaluate the modulation of microglial pro-and anti-inflammatory genes, was conducted. Results: We observed an exponential increase of iRFP670+ NALM-6 cells in the brain tissue up to 72h post-injection. Of note, pre-treatment with ActivinA significantly increase NALM-6 proliferation at 48 hours (p<0,05; n=7) and 72 hours (p<0,05; n=7) compared to not treated cells. ActivinA pre-treatment of NALM-6 cells was also associated with an higher mortality of brain slices compared to not stimulated NALM-6 cells at 24 hours (p<0,001; n=6 independent experiments) and 72 hours (p<0,05; n=6 independent experiments) after microinjection. By using CX3CR1+/GFP slices, we observed an increased microglia activation in ActivinA-pretreated NALM-6 microinjected slices compared to not treated at 5 (p<0,05; n=7) and 7 days (p<0.01; n=7) post-injection. Microglia morphological analysis showed that at 3 days post-injection, ActivinA pre-treated NALM-6 cells induced a slight increase of round-shaped microglia compared to CTRL (p<0,05; n=3) and to untreated NALM-6 injected slices (p<0,05; n=3), indicative of microglia reactivity to leukemic cells. However, at 7 days post-injection microglia from ActivinA pre-treated NALM-6 injected slices, showed an increase of cell area and perimeters (p<0,05; n=3), a reduced circularity and solidity (p<0,01; n=3) and a slight increase of aspect ratio (p<0,05; n=3) compared to CTRL slices, indicative of a reduced phagocytic activity of microglial cells. These modifications were associated with significant changes in gene expression, characterized by a down-regulation of microglial pro-inflammatory genes (iNOS, PAI-1; p<0.001; n=7) and an up-regulation of the anti-inflammatory ones (Arginase-1, CD206; p<0.01; n=7). Conclusions: We established an in vitro model to study leukemia cell interactions with brain tissue. Our data indicate that ActivinA pre-treatment promote higher NALM-6 proliferation in organotypic cortical brain slices compared to unstimulated leukemic cells. ActivinA pre-treated NALM-6 induce microglial modifications, characterized by a decrease reactivity and an induction of anti-inflammatory phenotype, favoring leukemic cell survival and proliferation. Our model can aid in answering important questions about brain changes after leukemic cell infiltration and could speed up the process of drug screening for leukemic patients.
Read full abstract