Abstract
Abstract 4090
Leukemic stem cells (LSC) are believed to be responsible for disease maintenance and/or propagation. A hallmark example is chronic myelogenous leukemia (CML). CML is a lethal malignancy that results from hematopoietic stem cell (HSC) transformation by the oncogene BCR-ABL. BCR-ABL tyrosine kinase inhibitors (TKI) are highly effective drugs to induce disease remission in CML patients and prolong their survival, but do not eliminate the LSC population and are required to be maintained throughout the patient's life to prevent relapse. Therefore, there is a strong rationale to develop strategies to target residual LSC to enable cessation of TKI treatment without leukemia relapse. Growing evidence indicates that signals from the bone marrow (BM) microenvironment are crucial for the maintenance of LSC and it has been suggested that LSC and HSC may locate in different niches where distinct microenvironment generated signals arise. We have previously demonstrated that Connexin-43 (Cx43) regulates HSC activity by preventing chemotherapy-induced senescence/apoptosis by allowing stromal-mediated ROS scavenging (Taniguchi-Ishikawa E et al., PNAS 2012), and allows homing/engraftment of HSC through stromal mediated trans-stromal migration (Gonzalez-Nieto D et al., Blood 2012). Here, we analyzed whether the microenvironment Cx43 function is required for BCR-ABL LSC maintenance or BCR-ABL LSC are independent of microenvironment Cx43-mediated signals. For this purpose, we utilized binary transgenic mice which express p210-BCR-ABL driven by inducible (Tet-off) expression of Scl in HSC. An enriched fraction of LSC (1,000 Lin−/c-kit+/Sca1+ cells) from Dox-off Scl-tTA × TRE-BCR-ABL (SclTg) mice (CD45.2+) was competitively transplanted with 3×106 CD45.1+ BM cells as previously described by our group (Sengupta A et al., Blood 2010 & Blood 2012) into lethally-irradiated Col1α1-WT (OB/P WT) or Col1α1-Cx43f/f (OB/P Cx43Δ/Δ) mice. HSC/P BM chimera was analyzed at 18 weeks post-transplantation by measuring the BM content of CD45.2+ BM cells, CD45.2+ Lin−/c-kit+/Sca1+ (HSC/multipotential progenitors) and CD45.2+ Lin−/c-kit+/Sca1− (progenitors) BM cells. Chimera of equivalent CD45.1+ populations was used as an internal control of normal HSC engraftment. Interestingly, the content of CD45.2+ or CD45.1+ Lin−/c-kit+/Sca1+ was not significantly modified by the deficiency of Cx43 in the osteolineage microenvironment (average ∼ 0.50% of BM cellularity). However, the content of progenitors from SclTg mice was ∼85% decreased in OB/P Cx43Δ/Δ mice while the content of competitor progenitors was also decreased albeit at a lower level (∼30% reduction), indicating that the hematopoietic ability of LSC and HSC were significantly impaired by the deficiency of Cx43 in the osteolineage microenvironment. To determine whether this defect was exclusively dependent on osteolineage deficient BM stroma, we performed long-term BM cultures of Scl-tTA × TRE-BCR-ABL (or non-Tg control from littermates) LSK cells onto WT or Cx43-deficient irradiated stroma obtained from OB/P Cx43Δ/Δ mice (Gonzalez-Nieto D et al., Blood 2012). At the end of 4 weeks of culture, we observed an average reduction in the number of CFUs of 84% (vs 86% for Non-Tg control HSC) when LSC were cultured on OB/P Cx43-deficient stroma. Altogether, these data indicate that the osteolineage deficiency of Cx43 impairs the hematopoietic activity of both LSC and HSC. BCR-ABL activity in LSC does not confer independence of Cx43-mediated signals from the BM osteolineage microenvironment.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.