Mutations leading to inappropriate activation of Akt isoforms contribute to proliferation and survival of a significant proportion of human cancers. Akt is activated by phosphorylation of its T-loop residue (Thr308) by PDK1 (3-phosphoinositide-dependent kinase-1) and its C-terminal hydrophobic motif (Ser473) by mTORC2 [mTOR (mammalian target of rapamycin) complex 2]. Potent PDK1 inhibitors such as GSK2334470 have recently been elaborated as potential anti-cancer agents. However, these compounds were surprisingly ineffective at suppressing Akt activation. In the present study we demonstrate that resistance to PDK1 inhibitors results from Akt being efficiently recruited to PDK1 via two alternative mechanisms. The first involves ability of Akt and PDK1 to mutually interact with the PI3K (phosphoinositide 3-kinase) second messenger PtdIns(3,4,5)P3. The second entails recruitment of PDK1 to Akt after its phosphorylation at Ser473 by mTORC2, via a substrate-docking motif termed the PIF-pocket. We find that disruption of either the PtdIns(3,4,5)P3 or the Ser473 phosphorylation/PIF-pocket mechanism only moderately impacts on Akt activation, but induces marked sensitization to PDK1 inhibitors. These findings suggest that suppression of Ser473 phosphorylation by using mTOR inhibitors would disrupt the PIF-pocket mechanism and thereby sensitize Akt to PDK1 inhibitors. Consistent with this, we find combing PDK1 and mTOR inhibitors reduced Akt activation to below basal levels and markedly inhibited proliferation of all of the cell lines tested. Our results suggest further work is warranted to explore the utility of combining PDK1 and mTOR inhibitors as a therapeutic strategy for treatment of cancers that harbour mutations elevating Akt activity.

INTRODUCTION

PDK1 (3-phosphoinositide-dependent protein kinase-1) phosphorylates and activates a group of protein kinases belonging to the AGC kinase [cAMP-dependent, cGMP-dependent and PKC (protein kinase C) kinase] family, that include three isoforms of Akt, S6K (S6 kinase) and SGK (serum- and glucocorticoid-induced protein kinase) [1,2]. PDK1 stimulates these enzymes by phosphorylating the T-loop residue of their kinase domains (Thr308 in Akt1) [1]. Maximal activation of the kinases also requires phosphorylation of a residue lying C-terminal to the kinase domain, known as the hydrophobic motif (Ser473 in Akt1) [36]. mTORC [mTOR (mammalian target of rapamycin) complex] 1 phosphorylates the hydrophobic motif of S6K isoforms, whereas mTORC2 phosphorylates the hydrophobic motif of Akt [7] and SGK isoforms [5].

Activation of AGC kinases, such as S6K and SGK isoforms that do not possess a 3-phosphoinositide-binding PH (pleckstrin homology) domain, is triggered through phosphorylation of their hydrophobic motif by mTORCs [8]. This does not directly activate these enzymes, but instead promotes their interaction with a motif on the PDK1 catalytic domain termed the PIF-pocket [8]. PDK1 then phosphorylates the T-loop residue eliciting activation. Key evidence to support this model is provided by the finding that knockin mutations that disrupt the ability of the PIF-pocket to recognize the phosphorylated hydrophobic-motif residues prevents activation of SGK and S6K isoforms by all agonists tested [911]. Moreover, inhibition of hydrophobic-motif phosphorylation employing mTOR inhibitors [12], mutation of the hydrophobic motif [6,13], or knockdown/knockout of the mTORC1 (S6K) or mTORC2 (SGK) subunits [14,15] also ablates activation of the S6K and SGK isoforms.

Activation of the Akt isoforms that possess a PH domain is dependent on prior activation of PI3K (phosphoinositide 3-kinase) and generation of the second messenger PtdIns(3,4,5)P3 [1618]. Both Akt and PDK1 possess a PH domain that binds to PtdIns(3,4,5)P3 and this promotes interaction of these enzymes at the plasma membrane. Moreover, binding of Akt to PtdIns(3,4,5)P3 induces a conformational change that facilitates phosphorylation of Thr308 by PDK1 [1622].

Previous work demonstrated that knockin mutations that disrupt the PIF-pocket either in ES (embryonic stem) cells [9] or in mouse muscle [22], whilst abolishing activation of S6K and SGK isoforms, did not markedly hinder activation of Akt isoforms. This suggested that the PIF-pocket mechanism was not rate-limiting for PDK1-mediated activation of Akt. Consistent with this conclusion, in mouse embryonic fibroblasts knockout of the mTORC2 components Rictor (rapamycin-insensitive companion of mTOR) [23], Sin1 (stress-activated protein kinase-interacting 1) [24] or mLST8 (mTOR-associated protein, LST8 homologue) [25] ablated hydrophobic motif Ser473 phosphorylation and therefore the ability of Akt to bind the PDK1 PIF-pocket, but had little impact on PDK1-mediated Akt Thr308 phosphorylation. Finally, in vitro studies undertaken in the presence of lipid vesicles containing PtdIns(3,4,5)P3 indicated that mutation of Ser473 to alanine had no effect on phosphorylation of Thr308 and activation of Akt [6].

The majority of all human cancers harbour mutations that lead to inappropriate activation of the Akt, S6K and SGK isoforms [2628]. This has stimulated much research to devise therapeutic strategies to suppress the activation of these enzymes. The therapeutic effectiveness of numerous PI3K, Akt and mTOR inhibitors is actively being investigated in clinical trials [29]. One strategy to inhibit the PI3K pathway is to deploy PDK1 inhibitors, which would be expected to suppress activation of Akt, as well as other AGC kinases that are stimulated by PDK1, such as the S6K and SGK isoforms that have been implicated in cancer [27,28]. The finding that PDK1 hypomorphic mice expressing ~20% of the normal level of PDK1 in all tissues were protected from developing tumours when crossed to PTEN (phosphatase and tensin homologue deleted on chromosome 10)+/− heterozygous mice, supported the notion that PDK1 inhibitors might have utility for the prevention of PI3K pathway-driven tumours [30].

The development of PDK1 inhibitors has lagged behind other enzymes of the PI3K pathway, partly due to the challenges of elaborating potent and specific PDK1 inhibitors [31]. A highly specific ATP-competitive PDK1 inhibitor termed GSK2334470 (IC50 ~15 nM) was described recently [32,33]. GSK2334470 efficiently inhibited the T-loop phosphorylation and activation of S6K and SGK isoforms, but was ineffective at suppressing Thr308 phosphorylation and hence Akt activation, especially by stimuli that induced strong activation of the PI3K pathway, such as IGF-1 (insulin-like growth factor 1) [32]. Other PDK1 inhibitors such as Sunesis-33, a non-ATP competitive inhibitor, have also recently been elaborated and found to be relatively ineffective at suppressing Akt activation [34,35]. Moreover, a recent study has also found that reducing PDK1 levels with an inducible RNAi (RNA interference) approach failed to significantly suppress Akt activation of tumour formation driven by loss of the PTEN tumour suppressor [36].

In the present study, we provide evidence that, in contrast with the previous work discussed above, phosphorylation of Ser473 does indeed play a role in regulating Thr308 phosphorylation, and this accounts for Akt's resistance to PDK1 inhibitors. Our data suggest that in vivo both the PtdIns(3,4,5)P3- and PIF-pocket-dependent mechanisms operate as two alternative ways to make Akt activation highly efficient and difficult to suppress by PDK1 inhibitors. Our findings suggest employing mTOR inhibitors, to suppress Ser473 phosphorylation mediated by mTORC2, will inhibit the PDK1 PIF-pocket mechanism and therefore markedly sensitize to PDK1 inhibitors. Consistent with this, we report that combining PDK1 and mTOR inhibitors ablates Akt activation and cancer cell proliferation more effectively than deployment of either inhibitor alone. Our results indicate that further work is warranted to explore the utility of employing combinations of PDK1 and mTOR inhibitors as a strategy for treating cancers possessing elevated Akt activity.

MATERIALS AND METHODS

Materials

We synthesized GSK2334470 as described previously [33]. Glutathione–Sepharose was purchased from Amersham Biosciences. [γ-32P]ATP was from PerkinElmer. IGF-1 was from Cell Signaling Technology. DMSO and Tween 20 were from Sigma. AZD8055 was from Axon Medchem. CellTiter 96® Non-Radioactive Cell Proliferation assay {MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide]} was from Promega. Enhanced chemiluminescence reagent was from GE Healthcare Lifesciences. The wild-type control and SIN1-knockout MEFs (mouse embryonic fibroblasts) have been described previously [24] and were provided by Dr Bing Su (Yale University School of Medicine, New Haven, CT, U.S.A.). The wild-type control and Rictor-knockout MEFs have been described previously [23] and were provided by Dr Mark Magnuson (Vanderbilt University School of Medicine, Nashville, TN, U.S.A.).

Antibodies

The following antibodies were raised in sheep and affinity purified against the indicated antigens: anti-Akt1 (S695B, third bleed; raised against residues 466–480 of human Akt1 RPHFPQFSYSASGTA, used for immunoblotting and immuno-precipitation), anti-S6K (S417B, second bleed; raised against residues 25–44 of human S6K AGVFDIDLDQPEDAGSEDEL, used for immunoblotting and immunoprecipitation), anti-PRAS40 (proline-rich Akt substrate of 40 kDa; S115B, first bleed; raised against residues 238–256 of human PRAS40 DLPRPRLNTSDFQKLKRKY, used for immunoblotting), anti-(phospho-PRAS40 Thr246) (S114B, second bleed, raised against residues 240–251 of human PRAS40 CRPRLNTpSDFQK, used for immunoblotting), anti-NDRG1 (N-myc downstream regulated 1; S276B third bleed; raised against full-length human NDRG1, used for immunoblotting). The following commercially available antibodies were used in the present study: anti-(phospho-Akt Ser473) (catalogue number 9271), anti-(phospho-Akt Thr308) (catalogue number 4056), anti-(phospho-S6K Thr389) (catalogue number 9234), anti-(phospho-S6 ribosomal protein Ser235/Ser236) (catalogue number 4856), anti-(phospho-S6 ribosomal protein Ser240/Ser244) (catalogue number 4838), anti-(total S6 ribosomal protein) (catalogue number 2217) and anti-(phospho-NDRG1 Thr346) (catalogue number 5482) were purchased from Cell Signaling Technology. For immunoblotting of the phosphorylated T-loop of S6K1 we employed the pan-PDK1 site antibody from Cell Signaling Technology (catalogue number 9379) as described previously [10]. Anti-GST (glutathione transferase)–HRP (horseradish peroxidase) conjugate was purchased from Abcam (catalogue number ab58626). The secondary antibodies coupled to HRP used for immunoblotting were obtained from Thermo Scientific.

General methods

Tissue culture, immunoblotting, restriction enzyme digests, DNA ligations and other recombinant DNA procedures were performed using standard protocols. DNA constructs used for transfection were purified from Escherichia coli DH5α cells using a Qiagen plasmid Maxi prep kit according to the manufacturer's protocol. For transient transfections, ten 10-cm-diameter dishes of HEK (human embryonic kidney)-293 cells were cultured and each dish was transfected with 5–10 μg of the indicated plasmids using the polyethylenimine method [37].

Buffers

The following buffers were used: lysis buffer [40 mM Tris/HCl (pH 7.5), 1% Triton X-100, 120 mM NaCl, 0.27 mM sucrose, 1 mM EDTA, 50 mM NaF, 10 mM 2-glycerophosphate, 5 mM sodium pyrophosphate, 1 mM sodium orthovanadate (added prior to lysis), 1 mM benzamidine (added prior to lysis), 1 mM PMSF (added prior to lysis), 0.1% 2-mercaptoethanol (added prior to lysis) and 1 μM microcystin-LR (added prior to lysis)], TBST buffer [50 mM Tris/HCl (pH 7.5), 0.15 M NaCl and 0.1% Tween 20] and sample buffer [50 mM Tris/HCl (pH 6.8), 6.5% (v/v) glycerol, 1% (w/v) SDS and 1% (v/v) 2-mercaptoethanol].

Cell treatments and lysis

HEK-293 cells were grown on 10-cm-diameter dishes. Cells were cultured in 10% (v/v) FBS (fetal bovine serum) in DMEM (Dulbecco's modified Eagle's medium; high glucose) and treated with or without different inhibitors as described in the Figure legends. Following treatment, cells were rinsed with 5 ml of ice-cold PBS and lysed using 0.5 ml of the freshly prepared lysis buffer and plastic scrapers. Lysates were clarified by centrifugation (18000 g at 4°C for 10 min) and supernatants were snap-frozen in liquid nitrogen and stored at −80°C until required. Protein concentration was determined using Coomassie Protein Assay reagent (catalogue number 1856209, Thermo Scientific).

Immunoblotting

Total cell lysate (20 μg) or pull-down samples were heated at 95°C for 5 min in sample buffer, subjected to SDS/PAGE (10% gel) and then electrotransferred on to nitrocellulose membranes. Membranes were blocked for 1 h in TBST buffer containing 5% (w/v) skimmed milk. The membranes were probed with the indicated antibodies in TBST containing 5% (w/v) skimmed milk or 5% (w/v) BSA for 16 h at 4°C. Detection was performed using HRP-conjugated secondary antibodies and the enhanced chemiluminescence reagent.

Cell growth assays

Cells were seeded (4000 per well) into 96-well plates and allowed to attach for 24 h. Cells were then treated with various concentrations of AZD8055 and/or GSK2334470. After 72 h, cell viability was determined using CellTiter 96® Non-Radioactive Cell Proliferation assay (MTT), according to the manufacturer's instructions.

RESULTS

Mutation of hydrophobic motif Ser473 renders Akt more susceptible to inhibition by PDK1 inhibitor

Consistent with our previous work [32], treatment of IGF-1-stimulated HEK-293 cells with high concentrations of the PDK1 inhibitor GSK2334470 (1–10 μM) did not significantly inhibit Thr308 phosphorylation of either endogenous Akt (Figure 1A) or overexpressed GST–Akt (Figure 1B), despite inhibiting the activation of S6K1. As the hydrophobic motif phosphorylation of AGC kinases plays such a vital role in regulating their activity we decided to test whether mutation of the hydrophobic-motif residue of Akt (Ser473) might influence sensitivity of Akt towards GSK2334470. We therefore mutated Ser473 to alanine and found that this markedly sensitized Thr308 phosphorylation to GSK2334470, causing near-complete inhibition of Thr308 phosphorylation at 0.3–1 μM doses of PDK1 inhibitor (Figure 1B). A mutant of Akt, in which Ser473 was changed to aspartic acid to mimic phosphorylation and induce constitutive binding to the PDK1 PIF-pocket, behaved more like the wild-type Akt displaying resistance to GSK2334470 (Figure 1B). We also stimulated HEK-293 cells with serum, conditions that induce ~6-fold lower activation of Akt than IGF-1 [12]. Under these conditions of lower PI3K pathway activation, mutation of Ser473 to alanine also increased the sensitivity of Thr308 phosphorylation to the PDK1 inhibitor GSK2334470 10-fold (Figure 1C).

Mutation of hydrophobic motif phosphorylation site of Akt (S473A) results in enhanced inhibition Akt Thr308 phosphorylation upon treatment with GSK2334470

Figure 1
Mutation of hydrophobic motif phosphorylation site of Akt (S473A) results in enhanced inhibition Akt Thr308 phosphorylation upon treatment with GSK2334470

(A) HEK-293 cells were serum-starved for 16 h and, after a 30 min pre-treatment of the cells with the indicated concentrations of the GSK2334470, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. Cell lysates were immunoblotted with the indicated antibodies. (B) As (A), except the cells were transfected with indicated GST-tagged Akt constructs. At 24 h post-transfection, GST–Akt was affinity purified on glutathione–Sepharose and subjected to immunoblot analysis using the indicated antibodies. (C) As (B), only the cells were maintained in serum and not serum starved or stimulated with IGF-1. Similar results were obtained from two independent experiments. Quantification of the Akt phospho-Thr308 (p-T308) and S6K phospho-Thr229 (p-T229) immunoblots shown in this Figure is presented in Supplementary Figure S1 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Figure 1
Mutation of hydrophobic motif phosphorylation site of Akt (S473A) results in enhanced inhibition Akt Thr308 phosphorylation upon treatment with GSK2334470

(A) HEK-293 cells were serum-starved for 16 h and, after a 30 min pre-treatment of the cells with the indicated concentrations of the GSK2334470, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. Cell lysates were immunoblotted with the indicated antibodies. (B) As (A), except the cells were transfected with indicated GST-tagged Akt constructs. At 24 h post-transfection, GST–Akt was affinity purified on glutathione–Sepharose and subjected to immunoblot analysis using the indicated antibodies. (C) As (B), only the cells were maintained in serum and not serum starved or stimulated with IGF-1. Similar results were obtained from two independent experiments. Quantification of the Akt phospho-Thr308 (p-T308) and S6K phospho-Thr229 (p-T229) immunoblots shown in this Figure is presented in Supplementary Figure S1 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Inhibition of mTOR leaves Akt more susceptible to inhibition by the PDK1 inhibitor GSK2334470

We next studied whether inhibition of Ser473 phosphorylation induced by mTOR inhibitors would sensitize Akt Thr308 phosphorylation to the PDK1 inhibitor GSK2334470. We stimulated HCT116 (Figure 2A), MCF7 (Figure 2B) and HEK-293 (Figure 2C) cells with IGF-1 in the absence or presence of 0.03–0.1 μM mTOR inhibitor AZD8055 [38] and increasing doses of GSK2334470. Under these conditions, 0.03–0.1 μM AZD8055 or 3 μM GSK2334470 when deployed alone had little impact on Thr308 phosphorylation (Figures 2A–2C). However, combining 0.03–0.1 μM AZD8055 with increasing concentrations of GSK2334470 induced a marked dose-dependent inhibition of Thr308 phosphorylation. For example, in all of the cell lines tested, combination of 1–3 μM GSK2334470 with 0.03–0.1 μM AZD8055 inhibited Thr308 phosphorylation to a much greater extent than either inhibitor deployed alone (Figures 2A–2C). Similar inhibition trends were observed with the phosphorylation of PRAS40 (Thr246), an Akt substrate, whose phosphorylation was also effectively suppressed by combining AZD8055 and GSK2334470 (Figures 2A–2C). Comparable results were observed when cells were treated with increasing doses of AZD8055 in the presence of a fixed dose of 3 μM GSK2334470 (Figures 2D–2F).

Combination of mTOR and PDK1 inhibitors results in maximal inhibition of Akt Thr308 phosphorylation

Figure 2
Combination of mTOR and PDK1 inhibitors results in maximal inhibition of Akt Thr308 phosphorylation

(A) HCT116 cells were serum-starved for 16 h and, after a 30 min pre-treatment of the cells with the indicated concentrations of GSK2334470 and AZD8055, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. Cell lysates were immunoblotted using the indicated antibodies. (B) As (A), except MCF7 cells were used. (C) As (A), except HEK-293 cells were used. (DF) as (AC), except increasing concentrations of AZD8055 were used in combination with 3 μM GSK2334470 as indicated. Similar results were obtained from two independent experiments. Quantification of the Akt phospo-Thr308 (p-T308) immunoblots shown in this Figure is presented in Supplementary Figure S2 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Figure 2
Combination of mTOR and PDK1 inhibitors results in maximal inhibition of Akt Thr308 phosphorylation

(A) HCT116 cells were serum-starved for 16 h and, after a 30 min pre-treatment of the cells with the indicated concentrations of GSK2334470 and AZD8055, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. Cell lysates were immunoblotted using the indicated antibodies. (B) As (A), except MCF7 cells were used. (C) As (A), except HEK-293 cells were used. (DF) as (AC), except increasing concentrations of AZD8055 were used in combination with 3 μM GSK2334470 as indicated. Similar results were obtained from two independent experiments. Quantification of the Akt phospo-Thr308 (p-T308) immunoblots shown in this Figure is presented in Supplementary Figure S2 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Ablation of mTORC2 renders Akt more susceptible to inhibition by the PDK1 inhibitor GSK2334470

We next explored how genetic ablation of the mTORC2 complex, by knockout of its Rictor or SIN1 subunits, affected sensitivity to GSK2334470. In both Rictor- [23] (Figure 3A) and SIN1- [24] (Figure 3B) knockout MEFs that lack Ser473 phosphorylation, GSK23334470 more effectively suppressed Thr308 phosphorylation compared with the wild-type cells. For example, 1–3 μM GSK2334470 had little effect on phosphorylation of Thr308 in the wild-type MEFs, but markedly inhibited Thr308 phosphorylation in the mTORC2-deficient MEFs (Figure 3). Consistent with the loss of Akt activity in mTORC2-deficient cells treated with 3 μM GSK2334470, phosphorylation of PRAS40 at Thr246 was inhibited to a greater extent than observed in the wild-type cells (Figure 3). In Rictor-knockout cells, phosphorylation of the Akt sites of the transcription factor FoxO (forkhead box O) 1/O3 (Thr24/Thr32) was markedly inhibited in the presence of 1 or 3 μM GSK2334470, under conditions which resulted in insignificant inhibition in the wild-type cells (Figure 3A).

Lack of mTORC2 activity results in enhanced inhibition of Akt Thr308 phosphorylation upon inhibition of PDK1

Figure 3
Lack of mTORC2 activity results in enhanced inhibition of Akt Thr308 phosphorylation upon inhibition of PDK1

(A) Rictor wild-type (WT) and knockout (KO) MEFs [24] were serum-starved for 16 h and, after a 30 min pre-treatment of the cells with the indicated concentrations of GSK2334470, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. Cell lysates were immunoblotted using the indicated antibodies. (B) As (A), except SIN1 wild-type and knockout MEFs were used [23]. Quantification of the Akt phospho-Thr308 (p-T308), PRAS40 phospho-Thr246 (p-T246) and FoxO1/O3 phospho-Thr24/Thr32 (p-T24/32) immunoblots shown in this Figure is presented in Supplementary Figure S3 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Figure 3
Lack of mTORC2 activity results in enhanced inhibition of Akt Thr308 phosphorylation upon inhibition of PDK1

(A) Rictor wild-type (WT) and knockout (KO) MEFs [24] were serum-starved for 16 h and, after a 30 min pre-treatment of the cells with the indicated concentrations of GSK2334470, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. Cell lysates were immunoblotted using the indicated antibodies. (B) As (A), except SIN1 wild-type and knockout MEFs were used [23]. Quantification of the Akt phospho-Thr308 (p-T308), PRAS40 phospho-Thr246 (p-T246) and FoxO1/O3 phospho-Thr24/Thr32 (p-T24/32) immunoblots shown in this Figure is presented in Supplementary Figure S3 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Ablation of the PDK1 PIF-pocket in the PDK1L155E/L155E-knockin ES cell line makes Akt more susceptible to inhibition by the PDK1 inhibitor GSK2334470

We previously described the generation of a PIF-pocket in PDK1L155E/L155E-knockin ES cells that possess a homozygous mutation in the PDK1 gene (L155E) that disrupts the PIF-pocket and therefore prevents PDK1 from interacting with hydrophobic motifs of its AGC kinase substrates after they have become phosphorylated by mTOR [9]. To obtain further evidence that the PDK1 PIF-Ser473 phosphorylation mechanism is contributing to resistance to PDK1 inhibitors, we studied whether Akt activation was more sensitive to the PDK1 inhibitor GSK2334470 in PDK1L155E/L155E-knockin ES cells. We found that phosphorylation of Thr308 was indeed inhibited more robustly in the PIF-pocket mutant compared with the wild-type control ES cells (Figure 4). For example, at a dose of 0.3 μM PDK1 inhibitor GSK2334470, Thr308 phosphorylation was reduced to a much greater extent in the PDK1L155E/L155E cells compared with their littermate wild-type control cells (Figure 4). Similar inhibition trends were observed in the phosphorylation of the Akt substrate PRAS40. This is despite PDK1L155E/L155E ES cells possessing higher levels of phosphorylated Ser473, perhaps resulting from the loss of the negative feedback loop of S6K to IRS-1 in the PIF-pocket-deficient cells thereby enhancing activity of PI3K and mTORC2 [39,40].

Inhibition of Akt Thr308 phosphorylation upon treatment with GSK2334470 is enhanced by a PDK1 mutation (L155E) that disrupts PIF-pocket–hydrophobic motif interaction

Figure 4
Inhibition of Akt Thr308 phosphorylation upon treatment with GSK2334470 is enhanced by a PDK1 mutation (L155E) that disrupts PIF-pocket–hydrophobic motif interaction

PDK1 wild-type (WT) or PDK1L155E/L155E-knockin ES cells [9] were serum starved for 4 h, pre-incubated with the indicated concentrations of GSK2334470 for 30 min and stimulated with 50 ng/ml IGF-1 for 30 min. Cell lysates were immunoblotted with the indicated antibodies. Similar results were obtained from two independent experiments. Quantification of the Akt phospho-Thr308 (p-T308), PRAS40 phospho-Thr246 (p-T246) and FoxO1/O3 phospho-Thr24/Thr32 (p-T24/32) immunoblots shown in this Figure is presented in Supplementary Figure S4 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Figure 4
Inhibition of Akt Thr308 phosphorylation upon treatment with GSK2334470 is enhanced by a PDK1 mutation (L155E) that disrupts PIF-pocket–hydrophobic motif interaction

PDK1 wild-type (WT) or PDK1L155E/L155E-knockin ES cells [9] were serum starved for 4 h, pre-incubated with the indicated concentrations of GSK2334470 for 30 min and stimulated with 50 ng/ml IGF-1 for 30 min. Cell lysates were immunoblotted with the indicated antibodies. Similar results were obtained from two independent experiments. Quantification of the Akt phospho-Thr308 (p-T308), PRAS40 phospho-Thr246 (p-T246) and FoxO1/O3 phospho-Thr24/Thr32 (p-T24/32) immunoblots shown in this Figure is presented in Supplementary Figure S4 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Inhibiting ability of Akt or PDK1 to bind PtdIns(3,4,5)P3 makes the PIF-pocket-Ser473 phosphorylation pathway rate-limiting in activating Akt

The data suggest that Akt Thr308 phosphorylation in vivo is promoted through two alternative mechanisms involving PDK1 and Akt binding to PtdIns(3,4,5)P3 or the PDK1 PIF-pocket interacting with Ser473 after it is phosphorylated by mTORC2. If this is the case, then inhibiting the ability of either Akt or PDK1 to bind to PtdIns(3,4,5)P3 should result in the PIF-pocket-binding pathway becoming the rate-limiting mechanism controlling activation of Akt. Consistent with this, treatment of HEK-293 cells with the mTOR inhibitor AZD8055, to prevent Ser473 phosphorylation, potently inhibited Thr308 phosphorylation of an Akt mutant lacking the PH domain, under conditions in which high concentrations of 1 μM AZD8055 had no effect on Thr308 phosphorylation of the wild-type Akt (Figure 5A).

Disruption of PtdIns(3,4,5)P3-binding of Akt and PDK1 reveals importance of hydrophobic-motif (Ser473) phosphorylation of Akt in promoting Thr308 phosphorylation

Figure 5
Disruption of PtdIns(3,4,5)P3-binding of Akt and PDK1 reveals importance of hydrophobic-motif (Ser473) phosphorylation of Akt in promoting Thr308 phosphorylation

(A) HEK-293 cells were transfected with GST-tagged Akt or an Akt mutant lacking the PH domain (Akt-delPH) and allowed to express for 24 h. The cells were then serum-starved for 16 h and, after a 30 min pre-treatment of the cells with indicated concentrations of AZD8055, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. GST–Akt and GST–Akt-delPH were pulled down using glutathione–Sepharose beads and immunoblotted using the indicated antibodies. (B) PDK1 wild-type (WT) or PDK1K465E/K465E-knockin ES cells [22] were treated as in Figure 3, except that the cells were serum-starved for 4 h. Cell lysates were immunoblotted with the indicated antibodies. (C) PDK1 wild-type or PDK1K465E/K465E-knockin ES cells were transfected with the indicated constructs and treated with 50 ng/ml of IGF-1 for 30 min in duplicates. Cell lysates were immunoblotted with the indicated antibodies. Similar results were obtained from two independent experiments. Quantification of the Akt phospho-Thr308 (p-T308) and PRAS40 phospho-Thr246 (p-T246) immunoblots shown in this Figure is presented in Supplementary Figure S5 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

Figure 5
Disruption of PtdIns(3,4,5)P3-binding of Akt and PDK1 reveals importance of hydrophobic-motif (Ser473) phosphorylation of Akt in promoting Thr308 phosphorylation

(A) HEK-293 cells were transfected with GST-tagged Akt or an Akt mutant lacking the PH domain (Akt-delPH) and allowed to express for 24 h. The cells were then serum-starved for 16 h and, after a 30 min pre-treatment of the cells with indicated concentrations of AZD8055, the cells were stimulated with 50 ng/ml of IGF-1 for 30 min. GST–Akt and GST–Akt-delPH were pulled down using glutathione–Sepharose beads and immunoblotted using the indicated antibodies. (B) PDK1 wild-type (WT) or PDK1K465E/K465E-knockin ES cells [22] were treated as in Figure 3, except that the cells were serum-starved for 4 h. Cell lysates were immunoblotted with the indicated antibodies. (C) PDK1 wild-type or PDK1K465E/K465E-knockin ES cells were transfected with the indicated constructs and treated with 50 ng/ml of IGF-1 for 30 min in duplicates. Cell lysates were immunoblotted with the indicated antibodies. Similar results were obtained from two independent experiments. Quantification of the Akt phospho-Thr308 (p-T308) and PRAS40 phospho-Thr246 (p-T246) immunoblots shown in this Figure is presented in Supplementary Figure S5 at http://www.BiochemJ.org/bj/448/bj4480285add.htm. p-, phospho.

To test the effect of inhibiting the ability of PDK1 to interact with PtdIns(3,4,5)P3, we used a PH domain PDK1K465E/K465Eknockin ES cell line expressing a homozygous mutation that abolishes its interaction with PtdIns(3,4,5)P3 [21]. We previously observed in a PDK1K465E/K465E-knockin ES cell line as well as in mice [22] that Akt was still activated by IGF-1 and insulin, albeit at a 2–3-fold lower level than observed in the wild-type mice. The new data shown in the present study would suggest that activation of Akt observed in PDK1K465E/K465E-knockin cells is driven via mTORC2-mediated phosphorylation of Ser473. Consistent with this, we found that treatment of PDK1K465E/K465E cells with the AZD8055 mTOR inhibitor to suppress Ser473 phosphorylation inhibited Thr308 phosphorylation more potently compared with the wild-type control ES cells (Figure 5B). For example, 0.03 μM AZD8055 inhibited Thr308 phosphorylation of Akt in PDK1K465E/K465E cells more significantly than in the wild-type control cells. Comparable inhibition trends were observed in the phosphorylation of the Akt substrate PRAS40.

Lastly, we overexpressed the GST–Akt[S473A] mutant in wild-type and PDK1K465E/K465E-knockin ES cells and monitored Thr308 phosphorylation. Consistent with the model that phosphorylation of Ser473 is critical for Thr308 phosphorylation in PDK1K465E/K465E-knockin ES cells, no detectable phosphorylation of Thr308 was observed in the GST–Akt[S473A] mutant expressed in the PDK1 PH domain-deficient cells. In parallel experiments, the GST–Akt[S473A] mutant was phosphorylated at Thr308 when expressed in the control wild-type ES cells (Figure 5C). In contrast, mutation of Ser473 to an acidic aspartic acid residue, to boost interaction with the PDK1 PIF-pocket, resulted in GST–Akt[S473D] being significantly phosphorylated at Thr308 when expressed in PDK1K465E/K465E-knockin ES cells further emphasizing that the Thr308 phosphorylation observed in these cells is dependent upon a PIF-pocket-dependent mechanism (Figure 5C).

Combination of mTOR and PDK1 inhibitors results in enhanced cell proliferation suppression

Finally, we investigated the effect that dual inhibition of PDK1 and mTOR had on proliferation of nine cell lines that possess mutations that activate the PI3K pathway (Figure 6, the known mutations in each cell line shown are listed in the Figure legend). The impact of the inhibitor treatments on cell viability was assessed after 3 days using an MTT assay. We observed that combination of AZD8055 and GSK2334470 had a markedly greater impact on proliferation of all of the cell lines tested than individual deployment of inhibitors (Figure 6). For example, in the majority of cell lines combing low doses of 0.03–0.1 μM AZD8055 and 1 μM GSK2334470 induced near-maximal suppression of cell proliferation. The EC50 values and the percentage of growth inhibition values obtained from such combination are shown in Table 1 and Table 2 respectively.

Combination of PDK1 and mTOR inhibitors leads to enhanced suppression of cancer cell line proliferation

Figure 6
Combination of PDK1 and mTOR inhibitors leads to enhanced suppression of cancer cell line proliferation

Cells that possess mutations abnormally activating the PI3K pathway were used: MCF7 [PIK3CA (phosphatidyl-4, 5-bisphosphate 3-kinase)], A172 (PTEN), U87 (PTEN), HCT15 [PIK3CA and KRAS (v-Ki-ras2 Kirsten rat sarcoma viral oncogene homologue)], HCT116 (PIK3CA and KRAS), HEK-293T (overactive PI3K pathway, Hoxhaj, G., unpublished work, mutation unknown), ZR-75-1 (PTEN-deficient, [45]), HT29 [PIK3CA and BRAF (v-raf murine sarcoma viral oncogene homologue B1)] and NCI-H727 (KRAS). Cells were treated with increasing concentrations of GSK2334470 or AZD8055, or a combination of 3 μM GSK2334470 and increasing concentrations of AZD8055. Cell viability was assessed after 72 h using an MTT assay. Similar results were obtained from two independent experiments. Mutation information was obtained from the COSMIC database [46]. The error bars represent S.D.

Figure 6
Combination of PDK1 and mTOR inhibitors leads to enhanced suppression of cancer cell line proliferation

Cells that possess mutations abnormally activating the PI3K pathway were used: MCF7 [PIK3CA (phosphatidyl-4, 5-bisphosphate 3-kinase)], A172 (PTEN), U87 (PTEN), HCT15 [PIK3CA and KRAS (v-Ki-ras2 Kirsten rat sarcoma viral oncogene homologue)], HCT116 (PIK3CA and KRAS), HEK-293T (overactive PI3K pathway, Hoxhaj, G., unpublished work, mutation unknown), ZR-75-1 (PTEN-deficient, [45]), HT29 [PIK3CA and BRAF (v-raf murine sarcoma viral oncogene homologue B1)] and NCI-H727 (KRAS). Cells were treated with increasing concentrations of GSK2334470 or AZD8055, or a combination of 3 μM GSK2334470 and increasing concentrations of AZD8055. Cell viability was assessed after 72 h using an MTT assay. Similar results were obtained from two independent experiments. Mutation information was obtained from the COSMIC database [46]. The error bars represent S.D.

Table 1
EC50 values for the cell proliferation assays shown in Figure 6 

A, GSK2334470 dose response; B, AZD8055 dose response; C, AZD8055 dose response+1 μM GSK2334470.

 EC50 value (μM) 
Cell line 
MCF7 2.24±0.31 0.04±0.04 0.03±0.02 
A172 3.02±0.23 0.03±0.03 0.02±0.04 
U87 0.5±0.02 0.03±0.03 0.03±0.07 
HCT15 0.2±0.04 0.04±0.13 0.02±0.18 
HCT116 1.13±0.03 0.22±0.17 0.03±0.15 
HEK-293T 2.82±0.22 0.03±0.06 0.01±0.16 
ZR-75-1 0.44±0.04 0.02±0.08 0.01±0.16 
HT29 3.87±0.25 0.54±0.26 0.04±0.03 
NCI-H727 2.08±0.21 0.14±0.23 0.02±0.15 
 EC50 value (μM) 
Cell line 
MCF7 2.24±0.31 0.04±0.04 0.03±0.02 
A172 3.02±0.23 0.03±0.03 0.02±0.04 
U87 0.5±0.02 0.03±0.03 0.03±0.07 
HCT15 0.2±0.04 0.04±0.13 0.02±0.18 
HCT116 1.13±0.03 0.22±0.17 0.03±0.15 
HEK-293T 2.82±0.22 0.03±0.06 0.01±0.16 
ZR-75-1 0.44±0.04 0.02±0.08 0.01±0.16 
HT29 3.87±0.25 0.54±0.26 0.04±0.03 
NCI-H727 2.08±0.21 0.14±0.23 0.02±0.15 
Table 2
Percentage of cell growth inhibition at either 0.1 μM GSK2334470, 0.1 μM AZD8055 or combination of 1 μM GSK2334470 and 0.1 μM AZD8055 for the cell proliferation assays shown in Figure 6 

A, 0.1 μM GSK2334470; B, 0.1 μM AZD8055; C: 0.1 μM AZD8055+1 μM GSK2334470.

 Percentage inhibition 
Cell lines 
MCF7 27.2±1.4 61.5±1.1 73.7±0.6 
A172 23.5±5.3 47.5±0.6 56.8±1 
U87 50.9±4.7 60.1±2.8 72±0 
HCT15 66.1±2.9 58.8±12.4 67.4±0.2 
HCT116 56.3±11.4 17.2±0.5 64.6±8.2 
HEK-293T 19.2±22 78±10.7 87.8±2.3 
ZR-75-1 57.9±1.6 12.4±5.6 60.4±1.2 
HT29 34.4±8.6 3.8±0 39.7±5.4 
NCI-H727 52.2±9 14.6±19.3 64.3±0.5 
 Percentage inhibition 
Cell lines 
MCF7 27.2±1.4 61.5±1.1 73.7±0.6 
A172 23.5±5.3 47.5±0.6 56.8±1 
U87 50.9±4.7 60.1±2.8 72±0 
HCT15 66.1±2.9 58.8±12.4 67.4±0.2 
HCT116 56.3±11.4 17.2±0.5 64.6±8.2 
HEK-293T 19.2±22 78±10.7 87.8±2.3 
ZR-75-1 57.9±1.6 12.4±5.6 60.4±1.2 
HT29 34.4±8.6 3.8±0 39.7±5.4 
NCI-H727 52.2±9 14.6±19.3 64.3±0.5 

DISCUSSION

The results of the present study provide evidence that phosphorylation of Ser473 does play a significant role in regulating the phosphorylation of Thr308. Previous work had indicated that the ability of both Akt and PDK1 to bind PtdIns(3,4,5)P3 was the crucial rate-limiting mechanism regulating Thr308 phosphorylation. Evidence supporting this was that ablation of PtdIns(3,4,5)P3 binding by either mutation of the Akt or PDK1 PH domains had a dramatic effect on Akt activation both in vitro and in cells [17,22,41]. In contrast, mutation of PDK1 PIF-pocket (L155E) had little effect on PDK1-mediated phosphorylation of Akt in vitro [6]. Moreover, two different PDK1-knockin mutations that disrupted the PDK1 PIF-pocket, namely L155E [9] and R131M [10], had no significant effect on phosphorylation of Thr308 by PDK1. Moreover, Akt activation induced by insulin was also not influenced in the tissue-specific PDK1-knockin mice that expressed only the PDK1[L155E] mutation [22].

On the basis of the results of the present study, we suggest that Akt can be activated in vivo by either the PtdIns(3,4,5)P3 binding- or PIF-pocket-dependent pathways and inhibition of whichever pathway has only a moderate bearing on Akt activity. This dual mechanism presumably enables Akt to be efficiently and robustly activated by PDK1, in response to agonists that stimulate the PI3K and mTORC2 pathways and is likely to account for why Akt activation is resistant to PDK1 inhibitors (Figure 7). Most probably, in the presence of a PDK1 inhibitor, even a residual fraction of non-inhibited PDK1 will be capable of significantly activating Akt either exploiting the PtdIns(3,4,5)P3-binding or PIF-pocket mechanism. However, suppressing either the PIF-pocket or PtdIns(3,4,5)P3-binding mechanism reduces the efficiency at which Akt is activated and makes it much more sensitive to PDK1 inhibitors. We also consistently observe that GSK2334470 inhibits Thr308 phosphorylation of endogenous Akt more effectively than overexpressed Akt (compare Figures 1A and 1B). Our interpretation of this observation is that endogenous Akt is likely to be more efficiently activated by the dual PtdIns(3,4,5)P3 and the PIF-pocket mechanism than the overexpressed enzyme. These results indicate that in general it may be harder to inhibit the phosphorylation of substrates of protein kinases that have evolved multiple alternative mechanisms to interact with their downstream targets.

Two alternative PtdIns(3,4,5)P3-binding and PIF-pocket-dependent mechanisms enable Akt to be efficiently activated and account for resistance to PDK1 inhibitors

Figure 7
Two alternative PtdIns(3,4,5)P3-binding and PIF-pocket-dependent mechanisms enable Akt to be efficiently activated and account for resistance to PDK1 inhibitors

Our data indicate that the PI3K pathway promotes interaction of PDK1 and Akt activation via two alternative mechanisms. First, both Akt and PDK1 possess PH domains that interact with PtdIns(3,4,5)P3, facilitating the co-localization of these enzymes on two-dimensional membrane surfaces, thereby greatly increasing the rate at which PDK1 can activate Akt. Secondly, phosphorylation of Akt at Ser473 by mTORC2 enables PDK1 to interact with phosphorylated Ser473 through its PIF-pocket. Note that all available data indicate that Akt must interact with PtdIns(3,4,5)P3 in order for Thr308 to be phosphorylated by PDK1. We therefore hypothesise that in order for Akt phosphorylated at Ser473 to be activated by PDK1 through the PIF-pocket pathway it still needs to interact with PtdIns(3,4,5)P3. However, for this mechanism, PDK1 does not need to bind PtdIns(3,4,5)P3 as its interaction with Akt is mediated via the PIF-pocket. This explains why Akt is still partially activated in knockin mice that express a mutant form of PDK1 that is incapable of binding to PtdIns(3,4,5)P3 [22]. Our results show that interfering with either the PtdIns(3,4,5)P3-binding or PIF-pocket-dependent mechanism has no, or at best only moderate, impact on Akt activation, but makes Akt much more sensitive to PDK1 inhibitors such as GSK2334470. We provide evidence that inhibiting Ser473 phosphorylation using mTOR inhibitors greatly sensitizes Akt activation to PDK1 inhibitors. In future research it would be worthwhile to explore whether a combination of PDK1 and mTOR inhibitors could be used as an effective approach to inhibit Akt activity in cancer cells that have mutations that stimulate the PI3K pathway. HM, hydrophobic motif.

Figure 7
Two alternative PtdIns(3,4,5)P3-binding and PIF-pocket-dependent mechanisms enable Akt to be efficiently activated and account for resistance to PDK1 inhibitors

Our data indicate that the PI3K pathway promotes interaction of PDK1 and Akt activation via two alternative mechanisms. First, both Akt and PDK1 possess PH domains that interact with PtdIns(3,4,5)P3, facilitating the co-localization of these enzymes on two-dimensional membrane surfaces, thereby greatly increasing the rate at which PDK1 can activate Akt. Secondly, phosphorylation of Akt at Ser473 by mTORC2 enables PDK1 to interact with phosphorylated Ser473 through its PIF-pocket. Note that all available data indicate that Akt must interact with PtdIns(3,4,5)P3 in order for Thr308 to be phosphorylated by PDK1. We therefore hypothesise that in order for Akt phosphorylated at Ser473 to be activated by PDK1 through the PIF-pocket pathway it still needs to interact with PtdIns(3,4,5)P3. However, for this mechanism, PDK1 does not need to bind PtdIns(3,4,5)P3 as its interaction with Akt is mediated via the PIF-pocket. This explains why Akt is still partially activated in knockin mice that express a mutant form of PDK1 that is incapable of binding to PtdIns(3,4,5)P3 [22]. Our results show that interfering with either the PtdIns(3,4,5)P3-binding or PIF-pocket-dependent mechanism has no, or at best only moderate, impact on Akt activation, but makes Akt much more sensitive to PDK1 inhibitors such as GSK2334470. We provide evidence that inhibiting Ser473 phosphorylation using mTOR inhibitors greatly sensitizes Akt activation to PDK1 inhibitors. In future research it would be worthwhile to explore whether a combination of PDK1 and mTOR inhibitors could be used as an effective approach to inhibit Akt activity in cancer cells that have mutations that stimulate the PI3K pathway. HM, hydrophobic motif.

Although we have focused on the PIF-pocket- and PtdIns(3,4,5)P3-dependent devices that operate to bring Akt and PDK1 together, it is possible that other mechanisms that we have not analysed also play an important role. For example, using a Förster resonance energy transfer system, overexpressed Akt was found to interact with PDK1 in a PtdIns(3,4,5)P3-independent manner [19]. In future work it would be interesting to explore whether the PtdIns(3,4,5)P3-independent interaction observed in the present study was indeed due to PDK1 binding to Akt phosphorylated at Ser473 via PIF-pocket-dependent mechanism, or another type of interaction. More recently, evidence has also emerged that PDK1 is capable of homodimerzation in vivo [42] and it would thus be interesting to explore how disrupting the homodimerization interface would impact on sensitivity to PDK1 inhibitors. In vitro studies have shown that in the absence of PtdIns(3,4,5)P3, Akt is trapped in a closed conformation in which Thr308 is not accessible to be phosphorylated by PDK1 [1619,43]. Furthermore, the loss of binding of PtdIns(3,4,5)P3 to full-length Akt by mutating conserved Arg25 in the PH domain to cysteine, prevents Akt activation in response to the protein phosphatase inhibitor okadaic acid that potently stimulates Ser473 phosphorylation [19]. These results indicate that even if Akt is phosphorylated at Ser473, and presumably able to bind to the PIF-pocket of PDK1, Thr308 cannot be phosphorylated unless Akt can interact with PtdIns(3,4,5)P3. These results indicate that for the PIF-pocket-mediated activation pathway that Ser473-phosphorylated Akt still needs to interact with PtdIns(3,4,5)P3 before it can be phosphorylated at Thr308 by PDK1. In this circumstance our data indicate that PDK1 does not need to bind PtdIns(3,4,5)P3 as its interaction with Akt is mediated via the PIF-pocket. This would also explain how Akt can be at least partially activated in knockin mice that express a mutant form of PDK1 that is incapable of binding to PtdIns(3,4,5)P3 [22]. These results might also explain why PI3K inhibitors potently suppress Akt activation by virtually all stimuli that have been investigated.

A question that arises from our work is why is Akt activation not significantly reduced in knockin cells that express homozygous mutation that disrupts the PIF-pocket. It is possible that the lack of S6K activity, and hence the loss of inhibitory feedback loops controlled by S6K in the PIF-pocket-knockin cells, results in compensatory increases in PI3K activity and hence PtdIns(3,4,5)P3 production thereby promoting Thr308 phosphorylation. Although Akt is activated normally in PDK1L155E/L155E-knockin cells, we believe that its phosphorylation by PDK1 is less efficient as Akt Thr308 phosphorylation is more sensitive to GSK2334470 in the PDK1L155E/L155E-knockin compared with the wild-type cells (Figure 4). These findings stress that caution is required when interpreting data from genetic knockin models, especially if multiple alternate mechanisms operate to activate a signalling pathway. Ablating one of these mechanisms via a genetic mutation may have a limited effect, as the cell will adapt by making use of alternative pathway(s). Our findings further emphasize that thorough analysis of signal transduction pathways is best achieved using a combination of genetic and pharmacologic approaches.

In hindsight, evidence that Akt activation did not solely rely on the mutual binding of PDK1 and Akt to PtdIns(3,4,5)P3 was suggested by the unexpected finding that Akt was still phosphorylated at Thr308 in PDK1K465E/K465E PH domain-knockin mice or ES cells, despite PDK1 being unable to interact with PtdIns(3,4,5)P3 [22]. Our current data suggest that phosphorylation of Akt at Thr308 seen in the PDK1K465E/K465E PH domain-knockin cells is dependent upon the PIF-pocket mechanism. This is supported by the observation that treatment of PDK1K465E/K465E PH domain-knockin cells with low (0.03 μM) concentrations of the mTOR inhibitor AZD8055 to ablate Ser473 phosphorylation inhibits Akt Thr308 phosphorylation to a much greater extent than observed in the wild-type ES cells (Figure 5B). Consistent with the PIF-pocket pathway playing a significant role in controlling Akt activation, treatment of several, but not all, cell lines with mTOR inhibitors results in substantial inhibition of Thr308 phosphorylation [12,38,44]. It would be interesting to explore whether treatment of these cell lines with compounds that suppress mTOR over a prolonged time period results in Thr308 phosphorylation becoming insensitive to the mTOR inhibitor. This could result from a loss of S6K inhibitory feedback loops and/or other compensatory mechanisms that might kick in to restore activation of Akt in the presence of mTOR inhibitors. Understanding the pathways that operate to counteract inhibition of Ser473 phosphorylation induced by mTOR inhibitors and restore Akt Thr308 phosphorylation are likely to play critical roles in mediating innate resistance of cells to such compounds.

The implications of our findings is that it will be difficult to ablate Thr308 phosphorylation and hence activation of Akt, employing PDK1 or mTOR inhibitors as single agents. A solution is to combine PDK1 and mTOR inhibitors. Indeed in all of the cell lines we have tested, combinations of mTOR and PDK1 inhibitors potently suppresses Thr308 and Ser473 phosphorylation to below-basal levels. Furthermore, combination of PDK1 and mTOR inhibitors had much greater effects on inhibition of proliferation of all evaluated cancer cell lines, than treatments with either of the inhibitors alone. Our findings indicate that further work is warranted to explore the potential utility of combining of PDK1 and mTOR inhibitors as a therapeutic strategy to treat cancers harbouring mutations that induce inappropriate activation of Akt isoforms.

Abbreviations

     
  • AGC kinase

    cAMP-dependent, cGMP-dependent and PKC (protein kinase C) kinase

  •  
  • ES

    embryonic stem

  •  
  • FoxO

    forkhead box O

  •  
  • GST

    glutathione transferase

  •  
  • HEK

    human embryonic kidney

  •  
  • HRP

    horseradish peroxidase

  •  
  • IGF-1

    insulin-like growth factor 1

  •  
  • MEF

    mouse embryonic fibroblast

  •  
  • mTOR

    mammalian target of rapamycin

  •  
  • mTORC

    mTOR complex

  •  
  • MTT

    3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide

  •  
  • NDRG1

    N-myc downstream regulated 1

  •  
  • PDK1

    3-phosphoinositide-dependent kinase-1

  •  
  • PH

    pleckstrin homology

  •  
  • PI3K

    phosphoinositide 3-kinase

  •  
  • PRAS40

    proline-rich Akt substrate of 40 kDa

  •  
  • PTEN

    phosphatase and tensin homologue deleted on chromosome 10

  •  
  • Rictor

    rapamycin-insensitive companion of mTOR

  •  
  • SGK

    serum- and glucocorticoid-induced protein kinase

  •  
  • Sin1

    stress-activated protein kinase-interacting 1

  •  
  • S6K

    S6 kinase

AUTHOR CONTRIBUTION

Ayaz Najafov performed all of the experiments. Natalia Shpiro synthesized GSK2334470. Ayaz Najafov and Dario Alessi planned the experiments, analysed the experimental data and wrote the manuscript.

We thank Dr Mark Magnuson (Vanderbilt University School of Medicine, Nashville, TN, U.S.A.) and Dr Bing Su (Yale University School of Medicine, New Haven, CT, U.S.A.) for the provision of Rictor- and Sin1-deficient MEFs. We thank excellent technical support of the MRC-Protein Phosphorylation Unit (PPU) DNA Sequencing Service (co-ordinated by Nicholas Helps), the MRC-PPU cloning team (co-ordinated by Mark Peggie and Rachel Toth), the MRC-PPU tissue culture team (co-ordinated by Kirsten McLeod), the Division of Signal Transduction Therapy (DSTT) protein and antibody purification teams (co-ordinated by Hilary McLauchlan and James Hastie).

FUNDING

This work was supported by the Medical Research Council and the pharmaceutical companies supporting the Division of Signal Transduction Therapy Unit (AstraZeneca, Boehringer-Ingelheim, GlaxoSmithKline, Merck KgaA, Janssen Pharmaceutica and Pfizer).

References

References
1
Mora
A.
Komander
D.
Van Aalten
D. M.
Alessi
D. R.
PDK1, the master regulator of AGC kinase signal transduction
Semin. Cell. Dev. Biol.
2004
, vol. 
15
 (pg. 
161
-
170
)
2
Pearce
L. R.
Alton
G. R.
Richter
D. T.
Kath
J. C.
Lingardo
L.
Chapman
J.
Hwang
C.
Alessi
D. R.
Characterization of PF-4708671, a novel and highly specific inhibitor of p70 ribosomal S6 kinase (S6K1)
Biochem. J.
2010
, vol. 
431
 (pg. 
245
-
255
)
3
Alessi
D. R.
Andjelkovic
M.
Caudwell
B.
Cron
P.
Morrice
N.
Cohen
P.
Hemmings
B. A.
Mechanism of activation of protein kinase B by insulin and IGF-1
EMBO J.
1996
, vol. 
15
 (pg. 
6541
-
6551
)
4
Burnett
P. E.
Barrow
R. K.
Cohen
N. A.
Snyder
S. H.
Sabatini
D. M.
RAFT1 phosphorylation of the translational regulators p70 S6 kinase and 4E-BP1
Proc. Natl. Acad. Sci. U.S.A.
1998
, vol. 
95
 (pg. 
1432
-
1437
)
5
Garcia-Martinez
J. M.
Alessi
D. R.
mTOR complex 2 (mTORC2) controls hydrophobic motif phosphorylation and activation of serum- and glucocorticoid-induced protein kinase 1 (SGK1)
Biochem. J.
2008
, vol. 
416
 (pg. 
375
-
385
)
6
Biondi
R. M.
Kieloch
A.
Currie
R. A.
Deak
M.
Alessi
D. R.
The PIF-binding pocket in PDK1 is essential for activation of S6K and SGK, but not PKB
EMBO J.
2001
, vol. 
20
 (pg. 
4380
-
4390
)
7
Sarbassov
D. D.
Guertin
D. A.
Ali
S. M.
Sabatini
D. M.
Phosphorylation and regulation of Akt/PKB by the rictor–mTOR complex
Science
2005
, vol. 
307
 (pg. 
1098
-
1101
)
8
Pearce
L. R.
Komander
D.
Alessi
D. R.
The nuts and bolts of AGC protein kinases
Nat. Rev. Mol. Cell Biol.
2010
, vol. 
11
 (pg. 
9
-
22
)
9
Collins
B. J.
Deak
M.
Arthur
J. S.
Armit
L. J.
Alessi
D. R.
In vivo role of the PIF-binding docking site of PDK1 defined by knock-in mutation
EMBO J.
2003
, vol. 
22
 (pg. 
4202
-
4211
)
10
Collins
B. J.
Deak
M.
Murray-Tait
V.
Storey
K. G.
Alessi
D. R.
In vivo role of the phosphate groove of PDK1 defined by knockin mutation
J. Cell. Sci.
2005
, vol. 
118
 (pg. 
5023
-
5034
)
11
Bayascas
J. R.
Sakamoto
K.
Armit
L.
Arthur
J. S.
Alessi
D. R.
Evaluation of approaches to generate PDK1 tissue specific knock-in mice
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
28772
-
28781
)
12
Garcia-Martinez
J. M.
Moran
J.
Clarke
R. G.
Gray
A.
Cosulich
S. C.
Chresta
C. M.
Alessi
D. R.
Ku-0063794 is a specific inhibitor of the mammalian target of rapamycin (mTOR)
Biochem. J.
2009
, vol. 
421
 (pg. 
29
-
42
)
13
Kobayashi
T.
Cohen
P.
Activation of serum- and glucocorticoid-regulated protein kinase by agonists that activate phosphatidylinositide 3-kinase is mediated by 3- phosphoinositide-dependent protein kinase-1 (PDK1) and PDK2
Biochem. J.
1999
, vol. 
339
 (pg. 
319
-
328
)
14
Kim
D. H.
Sarbassov
D. D.
Ali
S. M.
King
J. E.
Latek
R. R.
Erdjument-Bromage
H.
Tempst
P.
Sabatini
D. M.
mTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machinery
Cell
2002
, vol. 
110
 (pg. 
163
-
175
)
15
Sarbassov
D. D.
Ali
S. M.
Kim
D. H.
Guertin
D. A.
Latek
R. R.
Erdjument-Bromage
H.
Tempst
P.
Sabatini
D. M.
Rictor, a novel binding partner of mTOR, defines a rapamycin-insensitive and raptor-independent pathway that regulates the cytoskeleton
Curr. Biol.
2004
, vol. 
14
 (pg. 
1296
-
1302
)
16
Alessi
D. R.
Deak
M.
Casamayor
A.
Caudwell
F. B.
Morrice
N.
Norman
D. G.
Gaffney
P.
Reese
C. B.
MacDougall
C. N.
Harbison
D.
, et al. 
3-Phosphoinositide-dependent protein kinase-1 (PDK1): structural and functional homology with the Drosophila DSTPK61 kinase
Curr. Biol.
1997
, vol. 
7
 (pg. 
776
-
789
)
17
Stokoe
D.
Stephens
L. R.
Copeland
T.
Gaffney
P. R.
Reese
C. B.
Painter
G. F.
Holmes
A. B.
McCormick
F.
Hawkins
P. T.
Dual role of phosphatidylinositol-3,4,5-trisphosphate in the activation of protein kinase B
Science
1997
, vol. 
277
 (pg. 
567
-
570
)
18
Milburn
C. C.
Deak
M.
Kelly
S. M.
Price
N. C.
Alessi
D. R.
Van Aalten
D. M.
Binding of phosphatidylinositol 3,4,5-trisphosphate to the pleckstrin homology domain of protein kinase B induces a conformational change
Biochem. J.
2003
, vol. 
375
 (pg. 
531
-
538
)
19
Calleja
V.
Alcor
D.
Laguerre
M.
Park
J.
Vojnovic
B.
Hemmings
B. A.
Downward
J.
Parker
P. J.
Larijani
B.
Intramolecular and intermolecular interactions of protein kinase B define its activation in vivo
PLoS Biol.
2007
, vol. 
5
 pg. 
e95
 
20
Currie
R. A.
Walker
K. S.
Gray
A.
Deak
M.
Casamayor
A.
Downes
C. P.
Cohen
P.
Alessi
D. R.
Lucocq
J.
Role of phosphatidylinositol 3,4,5-trisphosphate in regulating the activity and localization of 3-phosphoinositide-dependent protein kinase-1
Biochem. J.
1999
, vol. 
337
 (pg. 
575
-
583
)
21
Komander
D.
Fairservice
A.
Deak
M.
Kular
G. S.
Prescott
A. R.
Downes
C. P.
Safrany
S. T.
Alessi
D. R.
van Aalten
D. M.
Structural insights into the regulation of PDK1 by phosphoinositides and inositol phosphates
EMBO J.
2004
, vol. 
23
 (pg. 
3918
-
3928
)
22
Bayascas
J. R.
Wullschleger
S.
Sakamoto
K.
Garcia-Martinez
J. M.
Clacher
C.
Komander
D.
van Aalten
D. M.
Boini
K. M.
Lang
F.
Lipina
C.
, et al. 
Mutation of the PDK1 PH domain inhibits protein kinase B/Akt, leading to small size and insulin resistance
Mol. Cell. Biol.
2008
, vol. 
28
 (pg. 
3258
-
3272
)
23
Shiota
C.
Woo
J. T.
Lindner
J.
Shelton
K. D.
Magnuson
M. A.
Multiallelic disruption of the rictor gene in mice reveals that mTOR complex 2 is essential for fetal growth and viability
Dev. Cell
2006
, vol. 
11
 (pg. 
583
-
589
)
24
Jacinto
E.
Facchinetti
V.
Liu
D.
Soto
N.
Wei
S.
Jung
S. Y.
Huang
Q.
Qin
J.
Su
B.
SIN1/MIP1 maintains rictor–mTOR complex integrity and regulates Akt phosphorylation and substrate specificity
Cell
2006
, vol. 
127
 (pg. 
125
-
137
)
25
Guertin
D. A.
Stevens
D. M.
Thoreen
C. C.
Burds
A. A.
Kalaany
N. Y.
Moffat
J.
Brown
M.
Fitzgerald
K. J.
Sabatini
D. M.
Ablation in mice of the mTORC components raptor, rictor, or mLST8 reveals that mTORC2 is required for signaling to Akt-FOXO and PKCα, but not S6K1
Dev. Cell
2006
, vol. 
11
 (pg. 
859
-
871
)
26
Altomare
D. A.
Testa
J. R.
Perturbations of the AKT signaling pathway in human cancer
Oncogene
2005
, vol. 
24
 (pg. 
7455
-
7464
)
27
Fenton
T. R.
Gout
I. T.
Functions and regulation of the 70kDa ribosomal S6 kinases
Int. J. Biochem. Cell Biol.
2010
, vol. 
43
 (pg. 
47
-
59
)
28
Vasudevan
K. M.
Barbie
D. A.
Davies
M. A.
Rabinovsky
R.
McNear
C. J.
Kim
J. J.
Hennessy
B. T.
Tseng
H.
Pochanard
P.
Kim
S. Y.
, et al. 
AKT-independent signaling downstream of oncogenic PIK3CA mutations in human cancer
Cancer Cell
2009
, vol. 
16
 (pg. 
21
-
32
)
29
Emerling
B. M.
Akcakanat
A.
Targeting PI3K/mTOR signaling in cancer
Cancer Res.
2011
, vol. 
71
 (pg. 
7351
-
7359
)
30
Bayascas
J. R.
Leslie
N. R.
Parsons
R.
Fleming
S.
Alessi
D. R.
Hypomorphic mutation of PDK1 suppresses tumorigenesis in PTEN+/− mice
Curr. Biol.
2005
, vol. 
15
 (pg. 
1839
-
1846
)
31
Peifer
C.
Alessi
D. R.
Small-molecule inhibitors of PDK1
ChemMedChem
2008
, vol. 
3
 (pg. 
1810
-
1838
)
32
Najafov
A.
Sommer
E. M.
Axten
J. M.
Deyoung
M. P.
Alessi
D. R.
Characterization of GSK2334470, a novel and highly specific inhibitor of PDK1
Biochem. J.
2011
, vol. 
433
 (pg. 
357
-
369
)
33
Medina
J. R.
Becker
C. J.
Blackledge
C. W.
Duquenne
C.
Feng
Y.
Grant
S. W.
Heerding
D.
Li
W. H.
Miller
W. H.
Romeril
S. P.
, et al. 
Structure-based design of potent and selective 3-phosphoinositide-dependent kinase-1 (PDK1) inhibitors
J. Med. Chem.
2011
, vol. 
54
 (pg. 
1871
-
1895
)
34
Erlanson
D. A.
Arndt
J. W.
Cancilla
M. T.
Cao
K.
Elling
R. A.
English
N.
Friedman
J.
Hansen
S. K.
Hession
C.
Joseph
I.
, et al. 
Discovery of a potent and highly selective PDK1 inhibitor via fragment-based drug discovery
Bioorg. Med. Chem. Lett.
2011
, vol. 
21
 (pg. 
3078
-
3083
)
35
Bobkova
E. V.
Weber
M. J.
Xu
Z.
Zhang
Y. L.
Jung
J.
Blume-Jensen
P.
Northrup
A.
Kunapuli
P.
Andersen
J. N.
Kariv
I.
Discovery of PDK1 kinase inhibitors with a novel mechanism of action by ultrahigh throughput screening
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
18838
-
18846
)
36
Ellwood-Yen
K.
Keilhack
H.
Kunii
K.
Dolinski
B.
Connor
Y.
Hu
K.
Nagashima
K.
O’Hare
E.
Erkul
Y.
Di Bacco
A.
, et al. 
PDK1 attenuation fails to prevent tumor formation in PTEN-deficient transgenic mouse models
Cancer Res.
2011
, vol. 
71
 (pg. 
3052
-
3065
)
37
Durocher
Y.
Perret
S.
Kamen
A.
High-level and high-throughput recombinant protein production by transient transfection of suspension-growing human 293-EBNA1 cells
Nucleic Acids Res.
2002
, vol. 
30
 pg. 
E9
 
38
Chresta
C. M.
Davies
B. R.
Hickson
I.
Harding
T.
Cosulich
S.
Critchlow
S. E.
Vincent
J. P.
Ellston
R.
Jones
D.
Sini
P.
, et al. 
AZD8055 is a potent, selective, and orally bioavailable ATP-competitive mammalian target of rapamycin kinase inhibitor with in vitro and in vivo antitumor activity
Cancer Res.
2010
, vol. 
70
 (pg. 
288
-
298
)
39
Harrington
L. S.
Findlay
G. M.
Gray
A.
Tolkacheva
T.
Wigfield
S.
Rebholz
H.
Barnett
J.
Leslie
N. R.
Cheng
S.
Shepherd
P. R.
, et al. 
The TSC1–2 tumor suppressor controls insulin-PI3K signaling via regulation of IRS proteins
J. Cell Biol.
2004
, vol. 
166
 (pg. 
213
-
223
)
40
Shah
O. J.
Wang
Z.
Hunter
T.
Inappropriate activation of the TSC/Rheb/mTOR/S6K cassette induces IRS1/2 depletion, insulin resistance, and cell survival deficiencies
Curr. Biol.
2004
, vol. 
14
 (pg. 
1650
-
1656
)
41
Alessi
D. R.
James
S. R.
Downes
C. P.
Holmes
A. B.
Gaffney
P. R.
Reese
C. B.
Cohen
P.
Characterization of a 3-phosphoinositide-dependent protein kinase which phosphorylates and activates protein kinase Bα
Curr. Biol.
1997
, vol. 
7
 (pg. 
261
-
269
)
42
Masters
T. A.
Calleja
V.
Armoogum
D. A.
Marsh
R. J.
Applebee
C. J.
Laguerre
M.
Bain
A. J.
Larijani
B.
Regulation of 3-phosphoinositide-dependent protein kinase 1 activity by homodimerization in live cells
Sci. Signaling
2010
, vol. 
3
 pg. 
ra78
 
43
Thomas
C. C.
Deak
M.
Alessi
D. R.
van Aalten
D. M.
High-resolution structure of the pleckstrin homology domain of protein kinase b/akt bound to phosphatidylinositol (3,4,5)-trisphosphate
Curr. Biol.
2002
, vol. 
12
 (pg. 
1256
-
1262
)
44
Liu
Q.
Kirubakaran
S.
Hur
W.
Niepel
M.
Westover
K.
Thoreen
C. C.
Wang
J.
Ni
J.
Patricelli
M. P.
Vogel
K.
, et al. 
Kinome-wide selectivity profiling of ATP-competitive mammalian target of rapamycin (mTOR) inhibitors and characterization of their binding kinetics
J. Biol. Chem.
2012
, vol. 
287
 (pg. 
9742
-
9752
)
45
Weng
L. P.
Smith
W. M.
Dahia
P. L.
Ziebold
U.
Gil
E.
Lees
J. A.
Eng
C.
PTEN suppresses breast cancer cell growth by phosphatase activity-dependent G1 arrest followed by cell death
Cancer Res.
1999
, vol. 
59
 (pg. 
5808
-
5814
)
46
Forbes
S. A.
Tang
G.
Bindal
N.
Bamford
S.
Dawson
E.
Cole
C.
Kok
C. Y.
Jia
M.
Ewing
R.
Menzies
A.
, et al. 
COSMIC (the Catalogue of Somatic Mutations in Cancer): a resource to investigate acquired mutations in human cancer
Nucleic Acids Res.
2010
, vol. 
38
 (pg. 
D652
-
D657
)

Supplementary data