Prognostic Value and Functional Consequences of Calcium Signalling in Acute Myeloid Leukaemia

Abstract

Ca 2+ is a ubiquitous second messenger that regulates a large number of cellular functions by acting on various proteins and signaling pathways. Abnormal expression/function of protein involved in Ca 2+ signaling have been associated with oncogenesis. However, its implication in the mechanisms that drive normal hematopoietic cells to leukemic transformation remain largely unknown. In order to identify actors of Ca 2+ signaling of potential prognosis value in acute myeloid leukemia (AML), , we performed a systematic screening on databases and identified expression of STIM2 as negatively correlated with disease free and overall survival while ORAI2 expression presented the opposite pattern. Interestingly, we noticed that ORAI2 and STIM2 expression differed depending on FAB subtypes. Indeed the 50% of patients in the upper quartile of STIM2 expression presented a myelo-monocytic differentiation while monocytic differentiation was associated with a lower ORAI2 expression. On a series of 407 AML patients from the ALFA 07-02 protocol, we observed that higher STIM2 was found in intermediate and high ELN17 while a lower ORAI2 level was seen in ELN17 high-risk patients. We quantified STIM2 and ORAI2 expression in two AML cell lines with monocytic potential, THP1 and OCI-AML3. When cells were driven into monocytic differentiation, we observed at RNA and protein level a decreased ORAI2 expression while STIM2 increased significantly. We then focused our studies on STIM2. To decipher its function in monocytic cell lines, we chose a shRNA-mediated gene knockdown approach. We first measured the capacitive Ca 2+ entry after STIM2 knockdown and found it increased, due to a predominant downregulation of the 2.1 STIM2 isoform, which has a negative effect on Ca 2+ entry, while 2.2 STIM2 isoform was less impacted. We then performed a transcriptomic analysis using the Nanostring technology, and identified a STIM2 signature characterized by dysregulation of genes involved in apoptosis, cell cycle and chromatin modification. Phenotypically, STIM2 knockdown drastically decreased cell proliferation and viability, through activation of the mitochondrial pathway of apoptosis and blockage of cell cycle at the G2/M phase. We observed that STIM2 knockdown led to a genomic stress with DNA double strand breaks as shown by an increased p-ϒH2AX, followed by activation of the p53 pathway, decreased expression of cell cycle regulators such as CDK1-cyclin B1 and the CDC25c phosphatase, and a decreased apoptosis threshold with a low antiapoptotic/proapoptotic protein ratio. Inhibition of p53 using Pifithrin-α reverted apoptosis and cell cycle blockage whereas p-ϒH2AX induction remained high. Moreover, STIM2 knockdown led to p-ϒH2AX activation but neither apoptosis nor cell cycle blockage in the p53-deficient cell line HL60, highlighting the fact that the genomic stress occurred downstream p53 induction in cells after STIM2 knockdown. At last, using in vitro monocytic differentiation of CD34 + cells, we observed a higher STIM2 expression at RNA level in differentiated cells. Acquisition of monocytic markers was faster after STIM2 forced expression, while STIM2 knockdown impaired CD14 expression and led to a p-ϒH2AX, p53 induction, apoptosis and cell cycle blockage in agreement with data observed in leukemic cell lines. In summary, we describe here STIM2, the “forgotten” member of SOCE, as a new actor in the proliferation, survival and differentiation of human normal and malignant monocytic cells. Considering (i) the association of a high STIM2 expression with an adverse prognosis in AML and (ii) the association of STIM2 with AML harboring monocytic/myelomonocytic differentiation, STIM2 may represent an interesting protein to target in these types of leukemia in the future