Multidrug resistance protein 4 (MRP4/ABCC4) regulates cAMP cellular levels and controls human leukemia cell proliferation and differentiation

Abstract

Increased intracellular cAMP concentration plays a well established role in leukemic cell maturation. We previously reported that U937 cells stimulated by H2 receptor agonists, despite a robust increase in cAMP, fail to mature because of rapid H2 receptor desensitization and phosphodiesterase (PDE) activation. Here we show that intracellular cAMP levels not only in U937 cells but also in other acute myeloid leukemia cell lines are also regulated by multidrug resistance-associated proteins (MRPs), particularly MRP4. U937, HL-60, and KG-1a cells, exposed to amthamine (H2-receptor agonist), augmented intracellular cAMP concentration with a concomitant increase in the efflux. Extrusion of cAMP was ATP-dependent and probenecid-sensitive, supporting that the transport was MRP-mediated. Cells exposed to amthamine and the PDE4 inhibitor showed enhanced cAMP extrusion, but this response was inhibited by MRP blockade. Amthamine stimulation, combined with PDE4 and MRP inhibition, induced maximal cell arrest proliferation. Knockdown strategy by shRNA revealed that this process was mediated by MRP4. Furthermore, blockade by probenecid or MRP4 knockdown showed that increased intracellular cAMP levels induce maturation in U937 cells. These findings confirm the key role of intracellular cAMP levels in leukemic cell maturation and provide the first evidence that MRP4 may represent a new potential target for leukemia differentiation therapy.

FIGURE 1.

Efflux of cAMP in U937 cells. A–C, time course of icAMP (□) and ecAMP (○) levels in U937 cells in the presence of 10 μm amthamine (A), 10 μm amthamine + 10 μm rolipram (B), or 100 μm forskolin (C). The cyclic nucleotide was excluded in the presence or absence of PDE inhibitors in a time-dependent fashion. D and E, icAMP (D) and ecAMP (E) concentration-response curves to amthamine in U937 cells at different time points: 5 (□), 10 (○), 30 (●), 60 (■), and 120 (▴) min. A decrease in icAMP levels (D) was observed in parallel with its extracellular accumulation (E). F, ATP-dependent cAMP uptake in U937 cell vesicles is MRP-mediated. Membrane vesicles (15 μg) were incubated with 83 μm [³H]cAMP in the absence or presence of 4 mm ATP. The rate of net ATP-dependent transport was calculated by subtracting the uptake with 100 mm 5′-AMP from that in the presence of 4 mm ATP. The transport was analyzed in control (■) and 50 μm MK571 (MRPs selective inhibitor)-treated vesicles (□) at different time points. ***, p < 0.001 respect to control. Data represent mean ± S.E. (n = 3).

FIGURE 2.

MRPs are involved in cAMP transport in intact U937 cells. A and C, time course study of icAMP (A) and ecAMP (C) levels in U937 cells exposed to non-cytotoxic concentrations of different agents: 10 μm amthamine (A) alone or in combination with 500 μm probenecid (+P) and/or 10 μm rolipram (+R, +P+R). Data shown are representative of three independent experiments. B and D, percentage of icAMP (B) and ecAMP (D) levels at 90 min in U937 cells exposed to the same agents. Combined PDE4 and MRPs inhibition results in significant enhancement of icAMP in amthamine-stimulated U937 cells. **, p < 0.01, ***, p < 0.001. Data represent mean ± S.E. (n = 3).

FIGURE 3.

Blockade of MRPs enhances the inhibitory effect of icAMP on U937 cell proliferation. Cell proliferation was assessed in U937 cells exposed to 0.4 mm Bt₂cAMP, 10 μm amthamine (A), 500 μm probenecid (P), and 10 μm rolipram (R). The maximal inhibition in cell proliferation was observed with the combined blockade of MRPs and PDE4. **, p < 0.01, ***, p < 0.001 versus control. Data represent mean ± S.E. (n = 3).

FIGURE 4.

cAMP extrusion, cell proliferation, and MRP4 expression in myeloid leukemia cell lines. A, KG-1a (left) and HL-60 (right) cells were exposed for 60 min to 10 μm amthamine and 10 μm rolipram (A + R) or 10 μm amthamine, 10 μm rolipram, and 500 μm probenecid (A +R + P), and icAMP and ecAMP were evaluated. B, cell proliferation was assessed in KG-1a (left) and HL-60 (right) cells exposed to 10 μm amthamine (A), 500 μm probenecid (P), 10 μm amthamine and 10 μm rolipram (A + R), or 10 μm amthamine, 10 μm rolipram, and 500 μm probenecid (A + R + P); 0.4 mm Bt₂cAMP was used as a positive control. The maximal inhibition in cell proliferation was observed with combined blockade of MRPs and PDE4. **, p < 0.01, ***, p < 0.001 versus control. Data represent mean ± S.E. (n = 3). C, representative RT-PCR showing detection of MRP4 and RNP II mRNA in U937, HL-60, and KG-1a cells. D, Western blot analysis of MRP4 in U937, HL-60, and KG-1a cells. PP2A was used as a loading control.

FIGURE 5.

Effect of MRP4 shRNA on cAMP cellular levels and proliferation in U937 cells. A and B, quantitative real-time PCR (A) and Western blot (B) of U937 cells transfected with MRP4 shRNA or scramble shRNA. PP2A was used as a loading control. C, U937 cells transfected with MRP4 shRNA or scramble shRNA were exposed to 10 μm amthamine and 10 μm rolipram (A+R) or 10 μm amthamine, 10 μm rolipram, and 500 μm probenecid (A+R+P) for 60 min and icAMP (left) and ecAMP (right) were assessed. D, cell proliferation was evaluated after 48 h in scramble or MRP4 shRNA untreated cells (control) or exposed to 10 μm amthamine and 10 μm rolipram (A+R) or 10 μm amthamine, 10 μm rolipram, and 500 μm probenecid (A+R+P). E, cell proliferation was evaluated in scramble or MRP4 shRNA cells pretreated with 50 μm Rp-cAMPS and exposed 24 h to 10 μm amthamine, 10 μm rolipram, and 500 μm probenecid (A+R+P) or 10 μm amthamine and 10 μm rolipram (A+R). Control corresponds to scramble or MRP4 shRNA untreated cells. Data represent mean ± S.E. (n = 3). **, p < 0.01.

FIGURE 6.

Elevated icAMP levels resulting from MRP and PDE4 inhibition induce the expression of CD11b and CD88 in U937 cells, supporting their differentiation to monocyte-like cells. A, Western blot analysis of CD88 expression (lane C) in U937 cells exposed to 10 μm amthamine (A), 500 μm probenecid (P), and 10 μm rolipram (R); 0.4 mm Bt₂cAMP was used as a positive control. B–G, CD88 functionality assays. B and C, mobilization of intracellular calcium (B) and cell migration mediated by C5a (C) in U937 cells exposed to 10 μm amthamine (A), 500 μm probenecid (P), and 10 μm rolipram (R); 0.4 mm Bt₂cAMP was used as a positive control. D and E, CD11b (D) and CD14 (E) antigen expression was evaluated by flow cytometry at day 3 following cell exposure to 10 μm amthamine (A), 500 μm probenecid (P), and 10 μm rolipram (R); 0.4 mm Bt₂cAMP was used as a positive control. F and G, cell migration (F) and CD11b antigen expression (G) were evaluated in scramble or MRP4 shRNA cells exposed to 10 μm amthamine and 10 μm rolipram (A+R) or 10 μm amthamine, 10 μm rolipram, and 500 μm probenecid (A+R+P). ***, p < 0.001, **, p < 0.01 versus control. Data represent mean ± S.E. (n = 3). The maximal expression of the differentiation markers was observed with combined blockade of MRPs and PDE4. H, cell morphology was evaluated after May Grünwald-Giemsa staining in untreated cells (left panels) and treated cells with 10 μm amthamine, 10 μm rolipram, and/or 500 μm probenecid (right panels). Arrows indicate differentiated cells. Each picture is representative of three independent experiments. Bar, 10 μm.