MS4A3: A New Player in Leukemic Stem Cell Survival in Chronic Myeloid Leukemia

Background: We have previously demonstrated that the transcriptional profile of diagnostic CD34⁺ cells from chronic phase chronic myeloid leukemia (CP-CML) patients exhibiting primary cytogenetic resistance to imatinib overlaps with that of patients with myeloid blast phase CML (BP-CML) (McWeeney et al. Blood 2010). These data suggest that primary resistance to tyrosine kinase inhibitors (TKIs) and advanced disease are biologically related. The hematopoietic cell cycle regulator, MS4A3, was identified as a principal component of the gene expression classifier predicting response to imatinib. Low MS4A3 correlated not only with primary TKI resistance, but also with shorter overall survival in CP-CML (n=35). Consistently, microarray (n=19 CP-CML; n=16 BP-CML), qRT-PCR (n=22 CP-CML; n=17 BP-CML), and immunoblot (n=3 CP-CML; n=3 BP-CML) analyses demonstrated that MS4A3 mRNA and protein levels are reduced in CD34⁺ progenitor cells from BP-CML versus CP-CML patients, with no difference between CP-CML and normal CD34⁺progenitors (n=3) (Eiring et al. ASH 2015 #14). These data suggest that MS4A3 may play a role in both primary TKI resistance and blastic transformation of CML.

Results: To assess the functional role of MS4A3 in CML and TKI response, we used ectopic MS4A3 expression and shRNA-mediated MS4A3 knockdown in CD34⁺ cells from BP-CML and CP-CML patients, respectively. Ectopic expression of MS4A3 in BP-CML CD34⁺ progenitors (n=5) markedly reduced colony formation in the presence and absence of imatinib, consistent with a tumor suppressor role for MS4A3 in CML. While re-expression of MS4A3 alone did not increase apoptosis compared to empty vector-expressing controls, imatinib-induced apoptosis in BP-CML CD34⁺ cells was increased by 62%, with no effect on normal CD34⁺ cord blood cells (n=2). Conversely, shRNA-mediated MS4A3 knockdown (shMS4A3) in CP-CML CD34⁺ cells (n=7) reduced the effects of imatinib in colony formation and apoptosis assays, with no effect on normal CD34⁺ progenitors (n=4). In contrast to a previous report (Donato JL, et al. J Clin Invest 2002), we detected no change in cell cycle status of CML or normal CD34⁺ cells upon MS4A3 ectopic expression or knockdown (n=3). Altogether, these data suggest that MS4A3 positively regulates patient survival and imatinib response in CML progenitor cells.

To evaluate MS4A3 in the leukemic stem cell compartment, we performed qRT-PCR on primary CP-CML cells (n=5) and observed that MS4A3 mRNA levels are 22-fold higher in committed CD34⁺38⁺ progenitors compared to more primitive CD34⁺38^- stem cells, suggesting a role for MS4A3 in differentiation. Consistently, qRT-PCR, immunoblot, and flow cytometry demonstrated that MS4A3 mRNA and protein were upregulated in CP-CML CD34⁺ cells upon G-CSF treatment (n=3). Flow cytometry also revealed that shMS4A3 in CP-CML CD34⁺ cells resulted in a reduction of CD11b⁺ cells by ~45% in the presence of G-CSF (n=3). To assess the function of MS4A3 in CML stem cells, we performed long-term culture-initiating cell (LTC-IC) assays and xenografts into NSG mice upon MS4A3 knockdown in CP-CML (n=3). shMS4A3 increased Ph⁺ LTC-IC colony formation in the absence, and even more so in the presence, of imatinib, with no effects on Ph^- LTC-ICs. Consistent with these data, shMS4A3 enhanced engraftment of CD34⁺CD45⁺GFP⁺ cells into the bone marrow of NSG recipient mice. Preliminary data in primary TKI-resistant and BP-CML CD34⁺ cells suggests regulation of this gene by promoter hypermethylation.

Conclusions: Altogether, these data suggest that MS4A3 plays a key role in imatinib response of 1) patients with primary TKI resistance, 2) patients with BP-CML, and 3) the CML stem cell compartment. Since the effects of MS4A3 in CML do not involve changes to the cell cycle, experiments are underway to identify the mechanism by which MS4A3 improves imatinib response and survival in CML.