Background: Despite spectacular advances in deciphering the molecular landscape of acute myeloid leukemia (AML), the clinical course of this disease has not improved dramatically for decades. Resistance to therapy and disease recurrence almost invariably occur, highlighting the need for a deeper understanding of these processes. Our general strategy takes into account decreased oxygen (O2) tension, a micro-environmental hallmark which remains understudied in the context of AML. While low O2 compartments such as bone marrow niches are well-recognized hosts of drug resistant AML cells, in vitro studies are still routinely performed in supra-physiologic ambient O2 concentration (21%), with potentially important consequences for clinical translatability. We hereby present an integrated molecular dissection of AML cells response to Cytarabine under O2-controlled conditions and exploit the information to develop novel synergistic/additive treatment combinations.
Methods: M14, THP1, OCIAML3 and primary cells (PC) derived from AML patients of different risk groups were incubated in culture medium under 21% O2 or 1% O2 in the presence or absence of Cytarabine (Cy, 1-7.5 µM), Rosuvastatin (Ro, 10-200µM) and/or Pravastatin (Pra, 10-200µM). After 48h, proliferation and apoptosis assays were performed using MTT assays, annexin V/PI staining and FACS analysis. Characterization of gene expression and expression levels of total and phosphorylated proteins were assessed using RNAseq and Reverse Phase Protein Arrays (RPPA), respectively. Gene Set Enrichment Analysis was performed following GSEA User Guide. Quantitative PCR reactions were performed using PowerUp SYBR Green PCR Master Mix (Applied Biosystems, USA) on 7900HT Real-Time PCR System. Intracellular cholesterol (ICC) levels were measured using WAKO Cholesterol assay kit.
Results: Treatment with Cy under 1% O2 induced a specific gene signature of metabolic stress responses, including the perturbation of cholesterol homeostasis. Detailed analysis of the cholesterol pathway via RNAseq showed pathway induction by hypoxia itself and severe depletion by Cy. Focused RT-PCR assays on the cholesterol pathway components HMGCS1, MSMO1, LSS and SQLE demonstrated significant induction under 1% compared to 21% O2, and significant downregulation in response to Cy. In line with our RNAseq and PCR data, RPPA showed that critical cholesterol synthesis pathways, such as ACC1, are downregulated in response to Cy. Surprisingly, we observed that ICC levels significantly increased after 48h in response to treatment with Cy (p<.01, n=4 [M14] and [THP1], 21% O2; p<.05, n=4 [M14] and [THP1], 1% O2). In addition, Cy therapy led to strong induction of CD36 mRNA expression (e.g. up to 11.7-fold at 21% O2, p<.05, up to 7.3-fold at 1% O2, p<.01, n=3 [M14]). In order to investigate the effects of cholesterol inhibition against AML cells, we treated several AML cell lines and PCs with Ro or Pra under 21% and 1% O2. We found that Ro exerts dose dependent antileukemic activity at 21% and 1% O2 and, at all doses tested, was significantly more effective than Pra, which was not effective against AML cells at 1% O2. The cytotoxic effects of Ro consisted of lowering ICC levels, induction of apoptosis and inhibition of AML cell growth. Further, Ro combined with Cy significantly enhanced the antileukemic effects of Cy alone and exhibited additive to synergistic activity in AML cell lines (CImax= 0.55, p<.05, n=3) and strong synergy in PCs (CImax= 0.22, n=10) derived from AML patients of different risk groups under 21% and 1% O2. Remarkably, Ro was not toxic against human CD34+ cord blood cells.
Treatment with Cytarabine induces a genetic signature of metabolic stress response, including the perturbation of the cholesterol biosynthesis pathway, in AML cells.
Depletion of intracellular cholesterol levels is counterbalanced by upregulation of CD36 expression in Cytarabine-treated AML cells under 21% and 1% O2 conditions.
Rosuvastatin exerts antileukemic activity via downregulation of intracellular cholesterol levels, induction of apoptosis and inhibition of AML cell growth.
Rosuvastatin displays greater antileukemic effects than pravastatin.
Rosuvastatin combined with Cytarabine acts synergistically against AML cells.
Conclusions: Further investigation of Rosuvastatin as a component of anti-AML therapy is warranted.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.