Activin A, a Potential Key Factor of the Malignant Bone Marrow Niche, Enhances B-Cell Precursor-Acute Lymphoblastic Leukemic Cell Migratory and Invasive Properties

The bone marrow (BM) represents a peculiar microenvironment characterized by a high concentration of growth factors and cytokines necessary for hematopoiesis, that make it a sanctuary for leukemic cell homing, survival and proliferation. B-cell precursor-Acute Lymphoblastic Leukemia (BCP-ALL) reprogram the BM stroma to create a leukemia-supporting and chemoprotective niche. Strategies to modulate the microenvironment could offer new approaches for anti-leukemia therapies. We identified ActivinA, a TGF-β family member, with a well-described promoting role in several solid malignancies, as a new potentially targetable leukemia-favoring factor.

ActivinA resulted overexpressed in the BM plasma of 108 BCP-ALL pediatric patients compared to 44 Healthy Donors (HDs), as evaluated by ELISA assay. Upon in vitro culture, primary BCP-ALL cells significantly increased ActivinA secretion by mesenchymal stromal cells (MSCs) derived from HD BM. Interestingly this effect was achieved both by cell-to-cell contact-mediated mechanisms (direct contact) and by soluble factor release (transwell-mediated co-culture). Interestingly, MSCs isolated from the BM of leukemic patients showed an intrinsic ability to secrete higher amounts of ActivinA (mean=983.6±362.9 pg/mL), compared to their normal counterpart (mean=218.5±31.99 pg/mL). In addition, we demonstrated that the pro-inflammatory cytokines IL-1β, IL-6 and TNF-α are increased in the leukemic BM and that they were able to synergize with leukemic blasts in inducing ActivinA release by MSCs (>100 fold increase compared to basal production).

Both type I and type II Activin receptors were found to be expressed by leukemic cells as demonstrated by flow cytometry analysis of ALK4, ACVR2A and ACVR2B receptors and western blot analysis of ALK2. Of note, ActivinA exposure could increase the mRNA expression levels of type I receptors ALK2 and ALK4, thus suggesting a possible self-reinforcing mechanism.

Gene expression analysis of ActivinA-treated BCP-ALL cells showed that this protein was able to significantly influence motility-associated molecular pathways. Accordingly, time lapse microscopy analyses revealed that ActivinA significantly increased random motility of leukemic cells (p<0.0001). In addition, ActivinA was able to almost double the migration of BCP-ALL primary cells in response to CXCL12, as demonstrated by in vitro chemotaxis assays. The specificity of the observed effect was demonstrated by using the ALK4 specific inhibitor SB431542.

CXCL12 reduction is one of the typical microenvironmental alterations occurring in the leukemic BM. In line with literature data, we showed 6-fold decrese of CXCL12 levels in the BM of BCP-ALL patients compared to HDs (p<0.0001). Dose-responses chemotaxis experiments revealed that ActivinA was able to sensitize leukemic cells to suboptimal CXCL12 concentrations. On the other site, ActivinA exerted an opposite effect on CD34⁺ cells isolated both from HD cord blood or BM. In detail, in HD-CD34⁺ cells ActivinA severely impaired CXCL12-induced migration (p<0.0001). This opposite effect could be explained by ActivinA ability to increase free cytosolic calcium only in leukemic cells, both basally and after addition of CXCL12 (flow cytometry analysis of Fluo-4 NW stained cells). Of note, calcium levels of HD-CD34⁺ cells resulted unaffected or even decreased (in 2 out of 3 experiments performed) by ActivinA treatment. Since calcium is critically involved in boosting cytoskeleton dynamics, we analysed, by flow cytometry, actin polymerization in phalloidin stained BCP-ALL cells. Interestingly, ActivinA treatment of BCP-ALL cells significantly increased the rate of conversion of globular into filamentous actin, which is a prerequisite of site-directed migration, as soon as CXCL12 was added. Moreover, ActivinA resulted a leukemia-promoting factor: protein treatment significantly increased the in vitro migration of BCP-ALL cells through Matrigel-coated transwells in response to CXCL12, thus stimulating leukemic cell invasiveness.

Overall, ActivinA resulted a key factor conferring a migratory advantage to leukemic cells over healthy hematopoiesis within the leukemic niche. Indeed, our in vitro findings provide the biological rationale for designing novel therapeutical approaches targeting the leukemia-stroma interplay.