Modulation Of The Glycolytic Metabolism In Acute Myeloid Leukemia Cells

Glycolysis is the central axis of cellular metabolism. The cancer cell bioenergetic status heavily relies on high glycolytic rates, even in aerobic conditions, thus sustaining the expensive processes of cell growth and proliferation. Growing evidences show that signaling aberrations - especially those involving PI3K/Akt/mTOR, HIF1a, Ras/Raf/MEK/ERK - are strictly connected to the establishment of a pro-glycolytic metabolism, through a multi-level crosstalk between proteins and metabolites that contribute to the acquisition of an energetic background granting a proliferative advantage. Here we investigated the glycolytic rate of resting and activated normal peripheral blood lymphocytes (NPBLs) and of acute myeloid leukemia (AML) cell lines. In an attempt to modulate the cellular metabolism for therapeutic intervention, we tested the following compounds that directly interfere with major metabolic or signaling pathways: dichloroacetate (DCA), a glycolysis inhibitor; aminooxyacetate (AOA), a glutaminolysis inhibitor; ST1326 (kindly given by Sigma-Tau), a fatty acid oxidation (FAO) inhibitor; and the MEK inhibitor PD0325901 (Selleck Chemicals). The cytotoxic drug effects were evaluated on two human leukemia cell lines, U937 and OCI-AML3, characterized by PI3K/Akt/mTOR and Ras/Raf/MEK/ERK hyperactivation, respectively. Cell counts, apoptosis (AnnexinV), glucose and lactate levels (GEM4000, Instrumentation Laboratory, UK) were measured. The glucose consumption rate (GCR) and lactate production rate (LPR) were calculated according to Li et al. (Biotechol. Appl. Biochem., 2005, 42, 73-80). Resting NPBLs were characterized by a very low glycolytic rate, according to their quiescent state, while cultured phytohemagglutinin-activated NPBLs displayed a remarkable increase in glycolytic rate: the GCR calculated over 72 hours showed a 25 fold-increase, while LPR had a of 10 fold-increase. Acute myeloid cell lines showed a high glucose catabolism: at 24h the U937 cell line, compared to activated NPBLs, had a 6.7 fold higher GCR, while the OCI-AML3 cell line showed a 4-fold increase. DCA exposure showed at 24h no detectable effect on GCR, LPR and apoptosis at concentrations ranging from 0.01 to 0.5mM on the U937 cell line. Apoptosis effects were detected only at higher concentrations: AnnexinV positive cells increased from 4.3 ± 1.5 (control) to 62.8 ± 16.4 (5mM) and 88.1 ± 16.8 (10mM). Exposure to AOA (24h at 1000µM) slightly increased GCR (1.23-fold) and LPR (1.22-fold) on U937 cells, followed by apoptotic effects at 72h: from 6.24 ± 4.2 (control) to 10.4 ± 0.8 at 100µM to 83.5 ± 0.7 at 1000µM. The FAO inhibitor ST1326 (10µM at 24h) induced a 3-fold increase of GCR in the U937 cell line. Apoptotic effects were seen in the U937 cells at 72h, from 5.0 ± 2.3 (control) to 35.9 ± 5.7 at 5µM to 64.1 ± 20.5 at 10µM. Conversely, GCR, LPR and apoptosis did not change on the OCI-AML3 line following ST1326 exposure. The MEK inhibitor PD0325901 caused a reduction of GCR and LPR on OCI-AML3 cells (6-fold GCR decrease, 2-fold LPR decrease at 100nM); apoptosis at 72h ranged from 6.3 ± 1.1 (control) to 15.5 ± 3.9 at 10nM to 45.3 ± 0.1 at 100nM. The U937 cells proved resistant to this compound, showing no metabolic perturbation and absence of apoptotic effects. In summary, this study indicates that exploiting the metabolism as a target for therapeutic intervention appears to be a promising new strategy. In fact, the inhibition of glycolysis by blocking either the activity of the enzymes that directly participate to the metabolic pathway or key components of cell signaling has proven to be effective in inducing apoptosis in AML cells. Interestingly, the opposing response to the various compounds observed in the two AML models may likely reflect their divergent signaling network, prompting further studies to evaluate the correlation between aberrant signal transduction pathways and peculiar metabolic profiles.