The Akt activation inhibitor TCN-P inhibits Akt phosphorylation by binding to the PH domain of Akt and blocking its recruitment to the plasma membrane

Abstract

Persistently hyperphosphorylated Akt contributes to human oncogenesis and resistance to therapy. Triciribine (TCN) phosphate (TCN-P), the active metabolite of the Akt phosphorylation inhibitor TCN, is in clinical trials, but the mechanism by which TCN-P inhibits Akt phosphorylation is unknown. Here we show that in vitro, TCN-P inhibits neither Akt activity nor the phosphorylation of Akt S473 and T308 by mammalian target of rapamycin or phosphoinositide-dependent kinase 1. However, in intact cells, TCN inhibits EGF-stimulated Akt recruitment to the plasma membrane and phosphorylation of Akt. Surface plasmon resonance shows that TCN, but not TCN, binds Akt-derived pleckstrin homology (PH) domain (K(D): 690 nM). Furthermore, nuclear magnetic resonance spectroscopy shows that TCN-P, but not TCN, binds to the PH domain in the vicinity of the PIP3-binding pocket. Finally, constitutively active Akt mutants, Akt1-T308D/S473D and myr-Akt1, but not the transforming mutant Akt1-E17K, are resistant to TCN and rescue from its inhibition of proliferation and induction of apoptosis. Thus, the results of our studies indicate that TCN-P binds to the PH domain of Akt and blocks its recruitment to the membrane, and that the subsequent inhibition of Akt phosphorylation contributes to TCN-P antiproliferative and proapoptotic activities, suggesting that this drug may be beneficial to patients whose tumors express persistently phosphorylated Akt.

Figure 1

TCN-P does not inhibit the activity and phosphorylation of purified Akt in vitro. Kinase assays were performed with purified or immunoprecipitated proteins in the presence of 200 μM ATP in the absence or presence of increasing TCN-P concentrations, followed by Western blotting with the indicated antibodies as described in Materials and Methods. All experiments were performed independently at least three times. (a) Up to 30 μM TCN-P does not inhibit in vitro the kinase activity (as measured by phosphorylation of the substrate GSK3-β of Akt1 immunoprecipitated from HEK293T cells transfected with Myc-tagged Akt. Exposure to 10 μM Akt inhibitor X (lane 6) demonstrates that Akt activity could be inhibited efficiently. (b) Up to 30 μM TCN-P does not interfere with the phosphorylation of Akt1 at T308 by PDK-1. (c) Up to 30 μM TCN-P does not interfere with the phosphorylation of Akt1 at S473 by immunoprecipitated mTOR.

Figure 2

TCN inhibits EGF-induced recruitment of AKT1 to the plasma membrane. (a-c) Serum-starved MDA-MB-468 cells exogenously expressing Myc-tagged wild-type Akt1 were treated with vehicle, EGF, TCN or TCN + EGF, and further analyzed by immunofluorescence (a-b) and subcellular fractionation (c) as described in Materials and Methods. (a) Fluorescent images representative of three independent experiments show that in control cells, Akt (red) was mainly found in the cytosol, while in EGF-stimulated cells, Akt clearly relocated to the membrane. Cells treated with TCN showed cytosolic staining with or without stimulation by EGF. (b) Quantification of membrane localization of Akt1 was carried out as follows. Approximately 100 cells were scored in each of three randomly selected fields. Results are presented as percentage of Myc-Akt1-transfected cells that are positive for plasma membrane-located Akt1. Each column represents the mean of three independent experiments ± standard deviation. (c) Membrane and cytosolic fractions were prepared as described in Materials and Methods and analyzed by Western blotting with the indicated antibodies (representative of two independent

Figure 3

TCN-P, but not TCN, binds to the PH domain of Akt. The interaction between TCN and TCN-P, respectively, and Akt PH domain was investigated with the Biacore T100 system. The graphs shown are representative of at least two independent experiments. (a-b) Akt1-derived PH domain was immobilized to the surface via ligand thiol coupling as described in Materials and Methods. (a) Representative example of sensorgrams for the interaction between TCN-P and Akt1-PH domain, showing an increased response with increasing concentrations. (b) Representative example of sensorgrams for the interaction between TCN and Akt1-PH domain, indicating that there is no significant interaction. (c) Dose-dependent interaction at 300 sec with NTA-captured His-tagged Akt1 PH domain was observed between TCN-P (blue), but not TCN (red).

Figure 4

TCN-P, but not TCN, causes chemical shift perturbation of amino acids in the Akt PH domain. (a) Superposition of the HSQC spectra of the Akt PH domain in presence of TCN (blue) and of TCN-P (green) onto the spectrum of apo-protein (red). (b) Summary of the chemical shift perturbation values in the presence of TCN-P (gray) and TCN (green). Three dashed lines indicate Δppm positions of 0.015, 0.03, and 0.06 respectively. (c) Structure of TCN-P (top) and model of Akt PH domain (ribbons) in complex with TCN-P (bottom). The pose of TCN-P was obtained from the docking software GOLD. The overall ribbon structures of Akt PH domain in complex with inositol (1,3,4,5)-tetrakisphosphate (IP4) (PDB code: 1H10) were colored in gray except for residues showing significant chemical shift perturbation. These were colored based on Δppm values: red, Δppm > 0.06; magenta, 0.06 > Δppm > 0.03; yellow, 0.03 > Δppm >0.015. (d) Ribbon representation of the structure of Akt PH domain in complex with IP4 (PDB code: 1H10) . The overall orientation is the same as that in (c). (e) Surface representation of the docking complex Akt/TCN-P and (f) of the complex Akt/IP4. In both panels, the overall surface of Akt was colored according to lipophilicity (blue, less lipophilic; brown, more lipophilic). The hydrogen bonds predicted to be formed between Akt and the compounds are displayed as yellow dashed lines. The position of residues involved in the formation of hydrogen bonds is shown in white and yellow. In (e), the position of residues showing chemical shifting perturbation in (c) is also shown using the color code as in (c).

Figure 5

Constitutively active forms of Akt1 rescue cells from TCN-induced loss of viability and apoptosis. COS-7 cells expressing vector, Myc-Akt1, Myc-Akt1-DD, Myc-Akt1-E17K or myr-Akt1 were exposed to TCN. (a) The level of expression of Akt1 constructs is shown in the left panel. In the right panel, exogenously expressed Akt1 was immunoprecipitated from cells with Myc antibody, and in vitro Akt1 kinase activity was assayed by its ability to phosphorylate S9 in GSK3-β. (b) Serum-starved cells were treated with different doses of TCN for 1 h, with or without subsequent stimulation by 30 ng/ml EGF for 20 min. Levels of endogenous phospho-Akt (S473) and phospho-GSK3-β(S9) were examined by Western blotting, with vinculin serving as loading control. The Myc blots demonstrate that the recombinant proteins were being expressed at comparable levels. (c) COS-7 cells expressing Akt1 constructs were exposed to TCN for 24 h and then subjected to MTT assay to determine their proliferation/viability. The graph is representative of three independent experiments. (d) COS-7 cells have very low basal levels of phospho-Akt. To assess basal levels of S473 phosphorylation in Akt of COS-7 cells, we compared the phospho-Akt levels in COS-7 cells stably transfected with vector or wild-type Akt1, following exposure to vehicle (0.1 % DMSO, lanes 1 and 6) or increasing concentrations of TCN for 24 h. Cells were lysed as described in Materials and Methods, and analyzed by Western blotting with the indicated antibodies. (e) COS-7 cells expressing vector, wt-Akt1, Akt1-DD, Akt1-E17K or myr-Akt1 were exposed to TCN for 24 h and then lysed. Triplicate aliquots of these lysates were analyzed by caspase-3 assay as described in Materials and Methods. This was calculated, separately for each of the three cell types, by dividing the average fluorescence intensity at a given TCN concentration by the average fluorescence intensity in the absence of TCN. The graph is representative of two independent experiments.