Direct binding of arsenic trioxide to AMPK and generation of inhibitory effects on acute myeloid leukemia precursors

Abstract

Arsenic trioxide (As2O3) exhibits potent antineoplastic effects and is used extensively in clinical oncology for the treatment of a subset of patients with acute myeloid leukemia (AML). Although As2O3 is known to regulate activation of several signaling cascades, the key events, accounting for its antileukemic properties, remain to be defined. We provide evidence that arsenic can directly bind to cysteine 299 in AMPKα and inhibit its activity. This inhibition of AMPK by arsenic is required in part for its cytotoxic effects on primitive leukemic progenitors from patients with AML, while concomitant treatment with an AMPK activator antagonizes in vivo the arsenic-induced antileukemic effects in a xenograft AML mouse model. A consequence of AMPK inhibition is activation of the mTOR pathway as a negative regulatory feedback loop. However, when AMPK expression is lost, arsenic-dependent activation of the kinase RSK downstream of MAPK activity compensates the generation of regulatory feedback signals through phosphorylation of downstream mTOR targets. Thus, therapeutic regimens with As2O3 will need to include inhibitors of both the mTOR and RSK pathways in combination to prevent engagement of negative feedback loops and maximize antineoplastic responses.

Figure 1

Effects of As₂O₃ on AMPK phosphorylation in AML cells. U937 cells were treated with As₂O₃ (2 μM) for the indicated times. Equal amounts of total cell lysates were subjected to SDS-PAGE followed by transfer to PVDF membranes. Immunoblotting with the indicated antibodies against AMPKα (A), ACC (B) and LKB1 (C) and anti-HSP90 are shown. The immunoblots with antibodies against the phosphorylated forms of the proteins or against the total proteins were from lysates from the same experiments analyzed in parallel by SDS-PAGE.

Figure 2

Arsenic binds directly to AMPK and inhibits its kinase activity. A, An AMPK kinase assay was performed with AMPKα1/β1/γ1 recombinant protein using an ADP-Glo Kinase Assay. Drugs were used at a final concentration of 3 μM. * p <0.05, **** p<0.0001 using an unpaired t-test. B, Recombinant AMPK protein was treated for 30 min at 30^° C with As-Biotin (20 μM) to assess in vitro binding of arsenic to AMPK. For in cell binding experiments U937 cells were treated with As-Biotin(20 μM) for 2 hrs and subsequently lysed. Arsenic-biotin complexes were incubated with streptavidin beads overnight and then washed and eluted from the beads with 2 x Laemelli sample buffer. Samples were then subjected to SDS-PAGE, followed by immunoblotting for AMPKα. C, Mass spectroscopy was performed on recombinant AMPK protein and As-Biotin complexes. Shown is the AMPKα protein sequence annotated with the modifications found. C = cysteine unmodified O= oxidation and A= cysteine with As-Biotin. Highlighted in gray is the peptide coverage. D, HA tagged AMPKα WT and AMPKα C299,304A mutant proteins were over-expressed in HEK293 cells. Cells were treated with 20 μM As-Biotin for 2 hrs and subsequently lysed. Arsenic-protein complexes were isolated by incubation with streptavidin beads overnight and then washed and eluted from the beads with 2 x Laemelli sample buffer. Samples were then subjected to SDS-page, followed by western blotting for HA tag.

Figure 3

Loss of AMPK affects arsenic - dependent signals and induction of cytotoxicity. A, AMPK 1/2 WT and knockout MEFs were serum starved overnight and then treated with As₂O₃ (5 μM) for the indicated times. Total cell lysates were then resolved by SDS-PAGE and immunoblotted with the indicated antibodies. The immunoblots with antibodies against the phosphorylated forms of the proteins or against the total proteins were from lysates from the same experiments analyzed in parallel by SDS-PAGE. B, AMPK 1/2 WT and knockout MEFs were serum starved overnight and then treated with As₂O₃ (5 μM) for the indicated times. Total cell lysates were then resolved by SDS-PAGE and immunoblotted with the indicated antibodies. The immunoblots with antibodies against the phosphorylated forms of the proteins or against the total proteins were from lysates from the same experiments analyzed in parallel by SDS-PAGE. The blots shown are the same lysates shown in panel A but were run in a different gel. C, AMPKα 1/2 WT and knockout MEFs were plated in 96 well plates and treated with varying concentrations of As₂O₃, the mTOR inhibitor PP242 and As₂O₃+PP242 for 4 days. Viability was assessed using a WST-1 assay. Data are expressed as a percentage of vehicle control treated cells. Shown are means and standard errors of three independent experiments. D, AMPKα 1/2 WT and knockout MEFs were treated with varying concentrations of As₂O₃, or the RSK inhibitor BI-D1870 or the combination of As₂O₃+BI-D1870 for 4 days and viability was measured by a WST-1 assay. Data are expressed as a percentage of vehicle control treated cells. Shown are the means and standard errors of three independent experiments.

Figure 4

Knockdown of AMPKα1 antagonizes the anti-leukemic effects of arsenic. A, (Left) U937 cells were transfected with AMPKα1 siRNA and then plated in methylcellulose in the presence or absence of As₂O₃ (1μM) and leukemic progenitor colony formation (CFU-L) was assessed in clonogenic assays in methylcellulose. Data are expressed as a percentage of the untreated control siRNA transfected cells. Shown are the means + SE of six independent experiments. ** p < 0.01 using a paired t-test. (Right) Immunblotting for AMPKa1 expression in total cell lysates from U937 cells transfected with control siRNA or AMPKa1 siRNA is shown. B, (Left) Kasumi-1 cells were transfected with AMPKα1 siRNA and then plated in methylcellulose in the presence or absence of As₂O₃ (1μM) and leukemic progenitor colony formation (CFU-L) was assessed in clonogenic assays in methylcellulose. Data are expressed as a percentage of the untreated control siRNA transfected cells. Shown are the means+ SE of six independent experiments. ** p < 0.01 using a paired t-test. (Right) Immunblotting for AMPKa1 expression in total cell lysates from Kasumi-1 cells transfected with control siRNA or AMPKa1 siRNA is shown. C, Mononuclear cells isolated from AML patients were plated in clonogenic assays in methylcellulose to assess the effects of siRNA-mediated AMPKα1 knockdown in the generation of the suppressive effects of As₂O₃ (0.5μM) on primary leukemic progenitor colony formation (CFU-L). Data are expressed as a percentage of the untreated control siRNA transfected cells. Shown are the means+ SE of nine experiments performed with samples from different AML patients. ** p < 0.01 using a paired t-test.

Figure 5

Co-treatment with the direct AMPK activator A769662 antagonizes the cytotoxic effects of As₂O₃ in vitro. A, U937 cells were treated with varying concentrations of As₂O₃, A769662 or As₂O₃ + A769662 for 4 days and viability was assessed by a WST-1 assay. Data are expressed as a percentage of vehicle control treated cells. Shown are means + SE of three independent experiments. B, U937 cells were plated in clonogenic assays in methylcellulose in the presence of As₂O₃ (1μM), A769662 (200 μM) or As₂O₃ +A769662, as indicated and CFU-L leukemic colony formation was assessed. Shown are means + SE of six independent experiments. * p < 0.05 using a paired t-test. C, Mononuclear cells isolated from AML patients were plated in clonogenic assays in methylcellulose in the presence of As₂O₃ (1μM), A769662 (200 μM) or As₂O₃ +A769662, as indicated and primary leukemic progenitor colony formation (CFU-L) was assessed. Shown are the means+ SE of eight experiments performed with samples from different AML patients. * p < 0.05 using a paired t-test. D, U937 cells were treated with As₂O₃ (2μM), A769662 (300 μM), or As₂O₃ +A769662 for 2 hours. Total cell lysates were resolved by SDS-PAGE and immunoblotted with the indicated antibodies. The immunoblots with antibodies against the phosphorylated forms of the proteins or against the total proteins were from lysates from the same experiments analyzed in parallel by SDS-PAGE. E, U937 cells were treated with As₂O₃ (2μM), A769662 (300 μM) or As₂O₃ +A769662 for 2 hours. Total cell lysates were resolved by SDS-PAGE and immunoblotted with the indicated antibodies. The immunoblots with antibodies against the phosphorylated forms of the proteins or against the total proteins were from lysates from the same experiments analyzed in parallel by SDS-PAGE.

Figure 6

A769662 antagonizes the effects of As₂O₃ in vivo. A, U937 cells were injected into the left flank of athymic nude mice. The indicated numbers of animals were treated with control, A769662, As₂O₃ and the combination of As₂O₃+A769662. Survival was determined as the time to euthanasia. ** p < 0.01, using a Log-rank (Mantel-Cox) test. B, Tumors from the mice in A were analyzed by H&E and IHC for TUNEL as a marker of apoptosis. Boxed regions show viable cells presented at higher magnification in the images below. Top panel images are at ×25 and the lower panel images are at ×200 magnification. C, The percentage area of necrosis was measured from 5 different fields covering a majority of the tumor sample at x25 magnification. Area was measured using Image J. *p< 0.05 using a one-way ANOVA analysis with a Bonferroni correction.