Fig 1.
ARG2 is specifically induced by hypoxia in chronic phase CML progenitors.
(A) Expression of ARG2 in cord blood (CB, n = 4, pooled), chronic phase CML (CP, n = 3) and blast crisis (BC, n = 3) CD34+ progenitors under normoxia (21% O2) or hypoxia (0.5% O2) for 96 hours. ARG2 transcript and protein levels were determined by RT-PCR (A) and western blot (B) respectively. (C) ARG2 upregulation persists upon BCR-ABL inhibition. CP CML CD34+ progenitors were treated with imatinib (1μM) at 21% or 0.5% O2 for 96 hours, then harvested and cell lysates probed for ARG2 protein by western blot. (D) Arginase activity is increased by hypoxia. CP CML CD34+ progenitors (n = 2) were incubated under 21% or 0.5% O2 for 96 hours, lysed and arginase activity quantified using a colorimetric assay to detect urea production. Right panel shows the representative image of the colorimetric assay. (E) Hypoxia induces expression of ARG2 in CML cell lines. K562 and KCL22 cells were grown at 21%, 2% or 1% O2 for 48 hours and the level of ARG2 protein determined by western blot. (F) ARG2 is the dominant arginase in CML. K562 cells were transfected with control, ARG1 or ARG2 siRNA and incubated under 21% or 0.5% O2 for 48 hours, before being subjected to arginase activity quantification (average of 4 experiments).
Fig 2.
ARG2 is expressed in subsets of hematologic malignancies and is regulated by HIF1-α and HIF2-α.
(A) Expression of ARG2 was determined by western blot in a panel of acute myeloid leukemia (AML), acute lymphoid leukemia (ALL) and multiple myeloma (MM) cell lines under 21% or 0.5% O2. (B) HIFs regulate ARG2 transcription. Luciferase reporters containing the ARG2 promoter were transfected into HL60 cells, and incubated under 21% or 1% O2 for 48 hours before assaying for luciferase activity (average of 3 experiments). (C) Knockdown of HIF1−α, HIF2−α, or ARNT (HIF1−β) attenuates hypoxic induction of ARG2. K562 cells were first transfected with control, HIF1−α, HIF2−α or ARNT siRNA, incubated at 21% or 0.5% O2 for 48 hours, following which ARG2 transcript and protein levels were measured by RT−PCR (average of 3 experiments) and western blot (D) respectively. Knockdown of HIF1−α or HIF2−α attenuates hypoxic induction of ARG2 in HL60 cells. HL60 cells were transduced with shRNA-expressing vectors targeting Luc (control), HIF1−α, or HIF2−α. The transduced cells were treated with 150 μM CoCl2 for 48 hours and were lysed for (E) western blotting or for (F) arginase activity quantification (average of 4 experiments).
Fig 3.
Anti-leukemic activity of the arginase inhibitor nor−NOHA in ARG2-expressing cells.
(A) Schematic of cellular arginine metabolism. ARG2 converts arginine into ornithine and urea, a reaction which can be inhibited by the arginase inhibitor nor−NOHA. (B) Anti-leukemic activity of nor−NOHA. K562 cells were treated with vehicle (DMSO) or increasing concentrations of nor–NOHA (0.1, 0.5 or 1mM) under 21% or 1.5% O2 for 72 hours, and cell viability determined by Annexin V/ 7−AAD staining. (C) Effect of arginase inhibition on cellular amino acid composition. K562 cells were treated with 500 μM nor–NOHA under 21% or 1% O2 for 48 hours. The metabolites were extracted and subjected to quantification by LC−MS (average of 3 experiments). (D) Ornithine does not rescue cell death induced by nor–NOHA. Experiments were as in (B), and ornithine (1mM) was added at the beginning of incubation (average of 3 experiments). (E) Cell lines expressing and upregulating ARG2 under hypoxia are more sensitive to nor−NOHA-induced cell death than those without ARG2 upregulation. Leukemia and myeloma cell lines were grown under 21% or 1.5% O2 for 48 hours, and ARG2 expression determined by western blot on cell lysates. (F) The indicated cell lines were grown at 21% or 1.5% O2 for 72 hours, and the effect of nor−NOHA on cell viability was analysed by Annexin V/ 7−AAD staining (average of 4 experiments).
Fig 4.
Treatment with nor−NOHA overcomes hypoxia-mediated TKI resistance in CML cells.
(A) K562 or KCL22 cells were treated with imatinib (0.5μM) and/or nor−NOHA (1mM) under 21% or 1.5% O2 for 72 hours. Cell viability and cell numbers were determined by Annexin V/ 7−AAD staining and flow cytometry (average of 3 experiments). (B) Primary CP CML (n = 6) or normal cord blood (CB; n = 3) CD34+ cells were treated with imatinib (0.5μM) and/or nor−NOHA (1mM) at 21% (normoxia; N) or 1.5% O2 (hypoxia; H) for 96 hours. The cells were then harvested and plated for colony formation assays (CFA).
Fig 5.
Depletion of ARG2 does not induce cell death or compromise the effects of nor−NOHA.
(A) K562 cells were transfected with control, ARG1 or ARG2 siRNA and incubated at 21% or 1% O2 for 48 hours, with or without imatinib (1μM). Levels of ARG2, cleaved PARP and cleaved caspase 3 were detected by western blot on cell lysates. K562 and HL60 cells were transduced with shRNA expressing vector targeting Luc (control) or ARG2, and transduced cells were grown at 21% or 1.5% O2 for 48 hours (B; western) or 72 hours (C; Annexin V/ 7−AAD staining V) before harvesting for the respective assays. (D) Schematic showing the genomic sequence for a portion of the first exon of ARG2. The targeting sequence of two independent sgRNAs are boxed, with the PAM sequence in bold. The translational initiation site (+1) is marked. (E) Characterization of the ARG2-KO clones generated by CRISPR/Cas9 nuclease. The control (Crtl) or ARG2 targeting sgRNA-transduced K562 cells were subjected to puromycin selection to generate single cell clones. The clones were incubated at 21% or 1% O2 for 48 hours, and ARG2 expression was detected by western blot. (F) Arginase activity of the knockout clones was quantified as described for Fig 1D (average of 3 experiments). (G) The control and knockout clones were treated with nor−NOHA (1mM) at 21% or 1.5% O2 for 72 hours, and cell viability was determined by Annexin V/ 7−AAD staining (average of 3 experiments).
Fig 6.
Genetic ablation of ARG2 and nor−NOHA treatment produce distinct effects on cellular respiration.
(A) Measurement of oxygen consumption rate (OCR), extracellular acidification rate (ECAR), and photon production rate (PPR) by the Seahorse XF−24 Analyser. The control (WT) or ARG2−KO K562 cells (two clones each) were treated with 1mM nor−NOHA for 24 hours before harvesting for analysis. Sequential injection of oligomycin (Oligo), carbonyl cyanide−4−(trifluoromethoxy) phenylhydrazone (FCCP), and rotenone & antimycin A (R/A) are as indicated. In each experiment, measurements were performed in quintuplicate for each cell line and condition. A representative graph of three independent experiments is shown. (B) Quantification of the maximal and basal OCR, ECAR and PPR levels. (C) Model depicting the relationship between HIFs, ARG2, and the cellular effects of nor−NOHA.