Abstract
Background
Acute lymphoblastic leukemia (ALL) cells create a leukemic niche that protects malignant cells from the effects of cytostatic agents and immune cells by altering their bone marrow microenvironment. This malignant process can be counteracted by impairing the homing of leukemic cells towards the bone marrow. Hematopoietic cells express the chemokine receptor CXCR4 and migrate towards its ligand CXCL12, which is actively produced by MSCs in the bone marrow. Therefore clinical trials have been initiated using the CXCR4 antagonist AMD3100 (Plerixafor) during leukemia treatment. However, these trials, as well as priming of AML in more than 4000 patients using a CXCR4 dependent mechanism, have not resulted in improved overall survival rates. This suggests that CXCR4 inhibition is not sufficient to disrupt leukemic niches.
Objectives
In this study we investigated how leukemic cells regulate the chemoattractive properties of their microenvironment.
Results
Here we show, using an ex vivo niche model with primary MSCs, that B-cell precursor ALL (BCP-ALL) cells affect their healthy microenvironment without altering CXCL12 secretion. Using a transwell migration assay we studied the chemoattractive properties and chemokine secretion patterns of several cell types and co-cultures. We confirmed that BCP-ALL cells migrate towards a CXCL12 gradient produced by primary MSCs (11-fold more migrated cells compared to background, p < 0.001). Inhibition of CXCR4 by AMD3100 reduced migration towards MSCs by 80% (p < 0.01). BCP-ALL cells migrated even more towards co-cultures of BCP-ALL cells and primary MSCs (24-fold more migrated cells compared to background, p < 0.001). Strikingly, this ex vivo leukemic niche did not produce higher levels of CXCL12 compared to MSC mono-cultures. Moreover, the induced migration towards MSC-ALL co-cultures could not be inhibited by AMD3100 treatment, indicating that BCP-ALL cells enhance the chemoattractive properties of their microenvironment in a CXCL12-independent manner.
In contrast to BCP-ALL cells, the migration of CD34+ progenitor cells towards co-cultures of BCP-ALL cells and MSCs was significantly reduced (0.8-fold compared to migration towards MSCs, p < 0.05). Similar results were observed when we studied the migratory behavior of MSCs. MSCs actively migrated towards BCP-ALL cells (1.7 fold compared to background, p < 0.001), while migration of MSCs was significantly reduced towards MSC-ALL co-cultures (0.4-fold compared to migration towards BCP-ALL, p < 0.001).
To find candidate factors influencing this process, we quantified the secreted levels of 64 cytokines in co-cultures of patient-derived BCP-ALL cells and MSCs. We observed leukemia-driven cytokine secretion patterns that were not influenced by the source of primary MSCs. In contrast to unaltered levels of CXCL12, we observed significant inductions of MCP-1/CCL2 and MDC/CCL22 (CCR4-ligands), IL8 and GRO-1 (CXCR1/2-ligands) and IP10/CXCL10 (CXCR3-ligands).
Conclusion
Our data indicate that leukemic cells alter the chemoattractive properties of their microenvironment, resulting in the secretion of multiple chemokines into the leukemic niche. This leukemic niche is highly potent in attracting BCP-ALL cells and repels the influx of healthy hematopoietic cells and MSCs using a CXCL12-independent mechanism. Furthermore, our results identify candidate factors that might be valuable future therapeutic targets.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal