Novel Small Molecule Inhibitors of 3-Phosphoinositide-dependent Kinase-1
2005; Elsevier BV; Volume: 280; Issue: 20 Linguagem: Inglês
10.1074/jbc.m501367200
ISSN1083-351X
AutoresRichard I. Feldman, James M. Wu, Mark A. Polokoff, Monica J. Kochanny, Harald Dinter, Daguang Zhu, Sandra L. Biroc, Bruno Alicke, Judi Bryant, Shendong Yuan, Brad O. Buckman, Dao Lentz, Mike Ferrer, Marc Whitlow, Marc Adler, Silke Finster, Zheng Chang, Damian O. Arnaiz,
Tópico(s)Ubiquitin and proteasome pathways
ResumoThe phosphoinositide 3-kinase/3-phosphoinositide-dependent kinase 1 (PDK1)/Akt signaling pathway plays a key role in cancer cell growth, survival, and tumor angiogenesis and represents a promising target for anticancer drugs. Here, we describe three potent PDK1 inhibitors, BX-795, BX-912, and BX-320 (IC50 = 11–30 nm) and their initial biological characterization. The inhibitors blocked PDK1/Akt signaling in tumor cells and inhibited the anchorage-dependent growth of a variety of tumor cell lines in culture or induced apoptosis. A number of cancer cell lines with elevated Akt activity were >30-fold more sensitive to growth inhibition by PDK1 inhibitors in soft agar than on tissue culture plastic, consistent with the cell survival function of the PDK1/Akt signaling pathway, which is particularly important for unattached cells. BX-320 inhibited the growth of LOX melanoma tumors in the lungs of nude mice after injection of tumor cells into the tail vein. The effect of BX-320 on cancer cell growth in vitro and in vivo indicates that PDK1 inhibitors may have clinical utility as anticancer agents. The phosphoinositide 3-kinase/3-phosphoinositide-dependent kinase 1 (PDK1)/Akt signaling pathway plays a key role in cancer cell growth, survival, and tumor angiogenesis and represents a promising target for anticancer drugs. Here, we describe three potent PDK1 inhibitors, BX-795, BX-912, and BX-320 (IC50 = 11–30 nm) and their initial biological characterization. The inhibitors blocked PDK1/Akt signaling in tumor cells and inhibited the anchorage-dependent growth of a variety of tumor cell lines in culture or induced apoptosis. A number of cancer cell lines with elevated Akt activity were >30-fold more sensitive to growth inhibition by PDK1 inhibitors in soft agar than on tissue culture plastic, consistent with the cell survival function of the PDK1/Akt signaling pathway, which is particularly important for unattached cells. BX-320 inhibited the growth of LOX melanoma tumors in the lungs of nude mice after injection of tumor cells into the tail vein. The effect of BX-320 on cancer cell growth in vitro and in vivo indicates that PDK1 inhibitors may have clinical utility as anticancer agents. 3-Phosphoinositide-dependent kinase 1 (PDK1) 1The abbreviations used are: PDK1, 3-phosphoinositide-dependent kinase-1; GSK3β, glycogen synthase kinase 3β; HMEC, human mammary epithelial cell; IGF-I, insulin-like growth factor I; MOPS, 4-morpholinepropanesulfonic acid; PI 3-kinase, phosphoinositide 3-kinase; PKA, protein kinase A; PKC, protein kinase C; PrEC, prostate epithelial cell; PRK, PKC-related kinase; PtdIns, phosphatidylinositol; S6K, p70 ribosomal S6-kinase; SGK, serum and glucocorticoid-regulated kinase; PTEN, phosphatase and tensin homologue deleted on chromosome 10. is a Ser/Thr protein kinase that can phosphorylate and activate a number of kinases in the AGC kinase superfamily (named after family members, protein kinase A, protein kinase G, and protein kinase C), including Akt/protein kinase B, protein kinase C (PKC), PKC-related kinases (PRK1 and PRK2), p70 ribobsomal S6-kinase (S6K1), and serum and glucocorticoid-regulated kinase (SGK) (1Mora A. Komander D. Van Aalten D.M. Alessi D.R. Semin. Cell Dev. Biol. 2004; 15: 161-170Crossref PubMed Scopus (669) Google Scholar). The first identified and best characterized PDK1 substrate is the proto-oncogene Akt (2Mitsiades C.S. Mitsiades N. Koutsilieris M. Curr. Cancer Drug Targets. 2004; 4: 235-256Crossref PubMed Scopus (236) Google Scholar). PDK1 phosphorylates the activation loop of Akt (also called the T-loop) on residue Thr308, which promotes the formation of the enzyme conformation with catalytic activity (3Alessi D.R. James S.R. Downes C.P. Holmes A.B. Gaffney P.R. Reese C.B. Cohen P. Curr. Biol. 1997; 7: 261-269Abstract Full Text Full Text PDF PubMed Google Scholar). Numerous studies have found a high level of activated Akt in a large percentage (30–60%) of common tumor types, including melanoma and breast, lung, gastric, prostate, hematological, and ovarian cancers. When activated in tumor cells, Akt has multiple effects that promote disease progression, including suppression of apoptosis and stimulation of tumor cell proliferation, metabolism, and angiogenesis (4Hanada M. Feng J. Hemmings B.A. Biochim. Biophys. Acta. 2004; 1697: 3-16Crossref PubMed Scopus (623) Google Scholar, 5Shiojima I. Walsh K. Circ. Res. 2002; 90: 1243-1250Crossref PubMed Scopus (850) Google Scholar). The PDK1/Akt signaling pathway thus represents an attractive target for the development of small molecule inhibitors that may be useful in the treatment of cancer. Akt comprises a family of Ser/Thr protein kinases containing three highly homologous members (AKT1, AKT2, and AKT3) (2Mitsiades C.S. Mitsiades N. Koutsilieris M. Curr. Cancer Drug Targets. 2004; 4: 235-256Crossref PubMed Scopus (236) Google Scholar, 6Downward J. Semin. Cell Dev. Biol. 2004; 15: 177-182Crossref PubMed Scopus (678) Google Scholar). Activation of Akt in cells by PDK1 requires stimulation of phosphoinositide 3-kinase (PI 3-kinase) whose activity becomes highly elevated in many tumors through the up-regulation or mutation of upstream signaling molecules such as epidermal growth factor receptors, Ras, Src, and c-ABL or by overexpression of PI 3-kinase itself (2Mitsiades C.S. Mitsiades N. Koutsilieris M. Curr. Cancer Drug Targets. 2004; 4: 235-256Crossref PubMed Scopus (236) Google Scholar, 4Hanada M. Feng J. Hemmings B.A. Biochim. Biophys. Acta. 2004; 1697: 3-16Crossref PubMed Scopus (623) Google Scholar). Loss of the tumor suppressor PTEN/MMAC1 is also a common mechanism of Akt activation in tumor cells (7Sansal I. Sellers W.R. J. Clin. Oncol. 2004; 22: 2954-2963Crossref PubMed Scopus (819) Google Scholar). PTEN has D-3 phosphoinositide phosphatase activity and acts as a negative regulator of PI 3-kinase function by removing its products, phosphatidylinositol (PtdIns)-3,4-P2 or PtdIns-3,4,5-P3 (8Simpson L. Parsons R. Exp. Cell Res. 2001; 264: 29-41Crossref PubMed Scopus (569) Google Scholar). Phosphoinositides produced by PI 3-kinase bind directly to the regulatory pleckstrin homology (PH) domain of Akt, driving a conformational change in the molecule which enables the activation loop of AKT1 to be phosphorylated by PDK1 at Thr308 (Thr309 for AKT2 and Thr305 for AKT3) (9Milburn C.C. Deak M. Kelly S.M. Price N.C. Alessi D.R. Van Aalten D.M. Biochem. J. 2003; 375: 531-538Crossref PubMed Scopus (222) Google Scholar, 10Scheid M.P. Woodgett J.R. FEBS Lett. 2003; 546: 108-112Crossref PubMed Scopus (346) Google Scholar). Activation of AKT1 is also associated with phosphorylation of Ser473 (Ser474 for AKT2 and Ser472 for AKT3) within a C-terminal hydrophobic motif characteristic of kinases in the AGC kinase family. This modification serves to further enhance the kinase activity of Akt (11Alessi D.R. Andjelkovic M. Caudwell B. Cron P. Morrice N. Cohen P. Hemmings B.A. EMBO J. 1996; 15: 6541-6551Crossref PubMed Scopus (2517) Google Scholar). Although the role of PDK1 in Thr308 phosphorylation is well established, the mechanism of Ser473 phosphorylation is controversial. A number of candidate enzymes responsible for this modification have been put forward, including integrin-linked kinase (12Persad S. Attwell S. Gray V. Mawji N. Deng J.T. Leung D. Yan J. Sanghera J. Walsh M.P. Dedhar S. J. Biol. Chem. 2001; 276: 27462-27469Abstract Full Text Full Text PDF PubMed Scopus (416) Google Scholar), PDK1 when in a complex with the kinase PRK2 (13Balendran A. Casamayor A. Deak M. Paterson A. Gaffney P. Currie R. Downes C.P. Alessi D.R. Curr. Biol. 1999; 9: 393-404Abstract Full Text Full Text PDF PubMed Scopus (384) Google Scholar), Akt itself, through autophosphorylation (14Toker A. Newton A.C. J. Biol. Chem. 2000; 275: 8271-8274Abstract Full Text Full Text PDF PubMed Scopus (422) Google Scholar), PKCα (15Partovian C. Simons M. Cell. Signal. 2004; 16: 951-957Crossref PubMed Scopus (102) Google Scholar), PKCβII (16Kawakami Y. Nishimoto H. Kitaura J. Maeda-Yamamoto M. Kato R.M. Littman D.R. Rawlings D.J. Kawakami T. J. Biol. Chem. 2004; 279: 47720-47725Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar), DNA-dependent kinase (17Feng J. Park J. Cron P. Hess D. Hemmings B.A. J. Biol. Chem. 2004; 279: 41189-41196Abstract Full Text Full Text PDF PubMed Scopus (414) Google Scholar), and the rictor-mTOR complex (18Sarbassov D.D. Guertin D.A. Ali S.M. Sabatini D.M. Science. 2005; 307: 1098-1101Crossref PubMed Scopus (5257) Google Scholar). Besides Akt, many other kinases in the AGC kinase superfamily, including PKC, PKA, S6K1, RSK1, SGK, and PRK1, have a homologous region of their activation loops containing a consensus substrate recognition site for PDK1 (19Storz P. Toker A. Front. Biosci. 2002; 7: 886-902Crossref PubMed Scopus (105) Google Scholar). S6K1 is an important regulator of cell growth which controls the synthesis of ribosomal and other proteins in response to growth factors and nutrients (20Dufner A. Thomas G. Exp. Cell Res. 1999; 253: 100-109Crossref PubMed Scopus (607) Google Scholar). Activation of S6K1 requires two PDK1-dependent modifications: phosphorylation of Thr389 (21Li Y. Corradetti M.N. Inoki K. Guan K.L. Trends Biochem. Sci. 2004; 29: 32-38Abstract Full Text Full Text PDF PubMed Scopus (335) Google Scholar), which can be mediated by PDK1 through activation of Akt, and the direct phosphorylation of Thr229 within the activation loop of S6K1 by PDK1 (22Pullen N. Dennis P.B. Andjelkovic M. Dufner A. Kozma S.C. Hemmings B.A. Thomas G. Science. 1998; 279: 707-710Crossref PubMed Scopus (727) Google Scholar, 23Alessi D.R. Kozlowski M.T. Weng Q.P. Morrice N. Avruch J. Curr. Biol. 1998; 8: 69-81Abstract Full Text Full Text PDF PubMed Scopus (516) Google Scholar). Promising preclinical and clinical studies with rapamycin analogs that block S6K1 through inhibition of mTOR (e.g. CCI-779) also implicate S6K1 as a cancer target (24Atkins M.B. Hidalgo M. Stadler W.M. Logan T.F. Dutcher J.P. Hudes G.R. Park Y. Liou S.H. Marshall B. Boni J.P. Dukart G. Sherman M.L. J. Clin. Oncol. 2004; 22: 909-918Crossref PubMed Scopus (891) Google Scholar). PDK1 can phosphorylate several isoforms of PKC (e.g. PKCδ, PKCζ, and PKCβII) in vitro or in cells cotransfected with PKC isoforms together with PDK1 (25Le Good J.A. Ziegler W.H. Parekh D.B. Alessi D.R. Cohen P. Parker P.J. Science. 1998; 281: 2042-2045Crossref PubMed Scopus (973) Google Scholar, 26Chou M.M. Hou W. Johnson J. Graham L.K. Lee M.H. Chen C.S. Newton A.C. Schaffhausen B.S. Toker A. Curr. Biol. 1998; 8: 1069-1077Abstract Full Text Full Text PDF PubMed Google Scholar, 27Toker A. Methods Mol. Biol. 2003; 233: 171-189PubMed Google Scholar). Consistent with these studies, deletion of both PDK1 alleles in embryonic mouse fibroblast cells results in impaired phosphorylation and activation of Akt, S6K1, and PKCζ (28Williams M.R. Arthur J.S. Balendran A. van der Kaay J. Poli V. Cohen P. Alessi D.R. Curr. Biol. 2000; 10: 439-448Abstract Full Text Full Text PDF PubMed Scopus (396) Google Scholar). Other isoforms of PKC, including PKCα, PKCβII, and PKCδ, showed reduced protein levels (29Balendran A. Hare G.R. Kieloch A. Williams M.R. Alessi D.R. FEBS Lett. 2000; 484: 217-223Crossref PubMed Scopus (181) Google Scholar). The established role of PDK1 in the regulation of Akt, S6K1, SGK, and PKC points to its importance as a critical regulator of cell signaling in cancer cells and tumor progression. In light of the compelling evidence for the role of PI 3-kinase/PDK1/Akt signaling in cancer progression, we sought to discover novel small molecule inhibitors that block this pathway for evaluation as anticancer drugs. Our strategy was to screen chemical libraries for lead compounds using a coupled assay measuring PtdIns-3,4-P2- and PDK1-mediated activation of AKT2, with the final assay readout being phosphorylation of a peptide substrate by activated AKT2. From our screen, we identified different mechanistic classes of compounds that blocked PDK1 or AKT2 or that interfered with phosphoinositide-dependent activation of AKT2, possibly through inhibition of the PH domain function. In this paper, we report on the biological characterization of optimized compounds that directly inhibit the activity of PDK1 in vitro and in cells while showing selectivity against a panel of other kinases. These compounds block the growth of a variety of tumor cell lines in culture, in soft agar, and in an animal model, supporting their further evaluation as anticancer agents. Materials—Polyclonal antibodies against phospho-Thr308-Akt, phospho-Ser473-Akt, Akt, phospho-Thr389-S6K1, S6K1, phospho-Ser241-PDK1, PDK1, phospho-Thr505-PKCδ, PKCδ, and phospho-Ser21/9-GSK3β, were obtained from Cell Signaling Technologies (Beverly, MA). Recombinant AKT2 and PDK1, containing N-terminal His tags, were expressed in insect cells and purified by Ni2+ affinity chromatography. λ-Phosphatase was obtained from Upstate Biotechnology (Lake Placid, NY). All other chemicals were of reagent grade quality. BX-320, BX-795, and BX-912 were synthesized as disclosed previously (42.Bryant, J., Kochanny, M., Yuan, S., Khim, S.-K., Buckman, B., Arnaiz, D., Bömer, U., Briem, H., Esperling, P., Huwe, P., Kuhnke, J., Schäfer, M., Wortmann, L., Kosemund, D., Eckle, E., Feldman, R., and Phillips, G. (June 10, 2004) International Patent WO2004048343Google Scholar). Cell Culture Conditions—The cell lines MDA-468, MDA-453, HCT-116, U87-MG, U2OS, PC-3, B16F10, and MiaPaCa were obtained from the American Type Culture Collection (ATCC). LOX amelanotic human melanoma cells were obtained as a gift from the Norwegian Cancer Institute (Oslo, Norway), and HeLa cells were obtained from the Institute of Molecular Biology (Zurich, Switzerland). Growth media and supplements were obtained from Invitrogen and Irvine Scientific (Santa Ana, CA) and were combined following ATCC guidelines. Primary human mammary epithelial cells (HMECs) and prostate epithelial cells (PrECs) and their respective growth media were purchased from Clonetics/Cambrex (Baltimore, MD). Kinase Assays—PDK1 was assayed in a direct kinase assay and a coupled assay format measuring PDK1- and PtdIns-3,4-P2-mediated activation of AKT2. For the coupled assay, the final assay mixture (60 μl) contained: 15 mm MOPS, pH 7.2, 1 mg/ml bovine serum albumin, 18 mm β-glycerol phosphate, 0.7 mm dithiothreitol, 3 mm EGTA, 10 mm MgOAc, 7.5 μm ATP, 0.2 μCi of [γ-33P]ATP, 7.5 μm biotinylated peptide substrate (biotin-ARRRDGGGAQPFRPRAATF), 0.5 μl of PtdIns-3,4-P2-containing phospholipid vesicles (3Alessi D.R. James S.R. Downes C.P. Holmes A.B. Gaffney P.R. Reese C.B. Cohen P. Curr. Biol. 1997; 7: 261-269Abstract Full Text Full Text PDF PubMed Google Scholar), 60 pg of purified recombinant human PDK1, and 172 ng of purified recombinant human AKT2. After incubation for 2 h at room temperature, the biotin-labeled peptide was captured from 10 μl of the assay mixture on streptavidin-coated SPA beads, and product formation was measured by scintillation proximity in a Wallac MicroBeta counter. The product formed was proportional to the time of incubation and to the amount of PDK1 and inactive AKT2 added. PDK1 was added at suboptimal levels so that the assay could sensitively detect inhibitors of AKT2 activation as well as direct inhibitors of PDK1 or AKT2. To measure PDK1 activity directly, the final assay mixture (60 μl) contained 50 mm Tris-HCl, pH 7.5, 0.1 mm EGTA, 0.1 mm EDTA, 0.1% β-mercaptoethanol, 1 mg/ml bovine serum albumin, 10 mm MgOAc, 10 μm ATP, 0.2 μCi of [γ-33P]ATP, 7.5 μm substrate peptide (H2N-ARRRGVTTKTFCGT), and 60 ng of purified recombinant human PDK1. After 4 h at room temperature, we added 25 mm EDTA and spotted a portion of the reaction mixture on Whatman P81 phosphocellulose paper. The filter paper was washed three times with 0.75% phosphoric acid and once with acetone. After drying, the filter-bound labeled peptide was quantified using a Fuji phosphorimager. Immunoblotting—After treating cells (roughly 80% confluent) as described in the figures, we washed them once with phosphate-buffered saline and then lysed them with MPER reagent (Pierce), supplemented with protease and phosphatase inhibitors (Halt protease inhibitor mixture, Pierce; 1 mm sodium fluoride, 1 mm sodium orthovanadate). Lysates were analyzed by Western blotting using the antibodies indicated in the figures. Growth Assays—Cells seeded at a low density (1,500–3,000 cells/well, 0.1 ml/well, 96-well plates) were incubated overnight. Compound treatments were made by adding 10 μl/well of the compound in 1% dimethyl sulfoxide and growth medium (final concentration of dimethyl sulfoxide, 0.1%), followed by brief shaking. Treated cells were incubated for 72 h, and viability was measured by the addition of 10 μl of the metabolic dye WST-1 (Roche Applied Science). The WST-1 signal was read in a plate reader at 450 nm, and a no cell, or zero time cell, background was subtracted to calculate the net signal. Results are reported as the average ± S.E. of two or more replicates. Apoptosis Assays—Cells were seeded and treated identically to cells in the proliferation assay. After a 48-h treatment, apoptosis was measured by the addition of 0.1 ml/well Apo-One reagent (Promega, Madison, WI). Plates were read in a fluorescence plate reader (Wallac Victor2 1420 Multilabel Counter) after 1 h. The fluorescent readout measured the cleavage of a specific caspase-3/caspase-7 fluorescent substrate (excitation 485 nm, emission 535 nm), indicating cells that were undergoing apoptosis. Results are reported as the average ± S.E. of two or more replicates. Soft Agar Assays—Sterile 35-mm plates were prepared with bottom layers of 0.5% agar in growth medium (1.5 ml/plate, agar temperature 48 °C). Trypsinized cells, dispersed into single cells using a 21-gauge needle, were seeded in a 1.5-ml volume of 0.3% agar in growth medium (42 °C temperature) added over the bottom agar layer and allowed to incubate overnight. Compound treatments were applied by adding 1 ml of 0.3% agar in growth medium (containing the compound) on top of the previously established agar layers. Cells incubated for 1 week were treated a second time (1 ml/well). After the 2nd week of treatment, colony growth was measured by adding 0.3 ml nitro blue tetrazolium (0.5 mg/ml stock solution in phosphate-buffered saline) (Sigma). Viable colonies stained dark blue after 24–48 h of incubation. Digitally scanned plates were assessed for total stained area using ImagePro 4.0 software (MediaCybernetics, Silver Spring, MD). Plotted results are the averages of duplicate plates. Cell Cycle Analyses—Subconfluent cells (0.5–2 million cells/100-mm plate, 5 ml/plate) were allowed to incubate overnight. Compound treatments were applied by adding 50 μl of compound in dimethyl sulfoxide (final vehicle concentration, 1%) per plate. After 18 h, cells were fixed, stained for DNA with propidium iodide, and analyzed on a BD Biosciences FACSCalibur cell sorter. The data were quantitated using CellQuest software (BD Biosciences). Blood-borne Metastasis Animal Model—Athymic (nu/nu) female mice (Simonsen, Gilroy, CA), 6–8 weeks old, were inoculated intravenously (lateral tail vein) on day 0 with 1 × 106 LOX cells, suspended at 1 × 107 cells/ml in 10 mm glucose in phosphate-buffered saline (calcium and magnesium-free). Groups (n = 26) were dosed by oral gavage twice daily (12 h apart) with vehicle (20% (w/v) hydroxypropyl-β-cyclodextrin, Cavitron 82005, Cargill, Cedar Rapids, IA) or BX-320 dissolved in vehicle (20 mg/ml) adjusted to pH 4. The dose volume was 10 ml/kg (200 mg/kg), and dosing began 2–4 h before cell inoculation and continued until the end of the study. A small, but equal, number of animals died in the vehicle- and BX-320-treated groups during the study, which was related to the dosing regimen and not to compound toxicity. Animals showed no observable side effects of compound treatment. On day 21 (22 days of dosing), mice were euthanized, and the lungs were removed and frozen for future DNA extraction. All experiments were conducted in accordance with the principles and procedures approved by the Berlex Animal Care and Use Committee. Statistical analyses were performed with Statview 5.0 (SAS Institute, Inc, Cary, NC). Significance was determined by the nonparametric Mann-Whitney U test. p values < 0.05 were considered significant. Quantification of Tumor Burden by Quantitative Real Time PCR—We determined the tumor burden in the mouse lungs by extracting total DNA, then measuring the amount of the human CCR5 gene present by quantitative real time PCR. To extract and purify total DNA, each lung was homogenized in 1 ml of Qiagen ATL buffer (Valencia, CA) with a Mini-Bead Beater-8 (BioSpec) using zirconia silicate beads (2.5 mm; BioSpec, Bartlesville, OK). We performed DNA extractions using the Qiagen QiaAmp DNA Mini Kit according to the manufacturer's instructions. For quantitative PCR, we used the following human CCR5-specific primer set and probes: forward primer, 5′-CCTCTTAGTCTTCTTCCAAAAGCAC-3′; reverse primer, 5′-TCGGGAGCCTCTTGCTG-3′; and probe, 5′-FAM-CCAAACGCTTCTGCAAATGCTGTTCTATT-TAMRA-3′. Thermocycling was performed for 40 cycles at 95 °C for 15 s and 63 °C for 1 min after initial UNG treatment at 50 °C for 2 min with subsequent UNG inactivation and TaqGold activation for 10 min at 95 °C. A standard curve was generated using human genomic DNA (Promega), which was used to calculate the number of human cells/mouse lung, assuming two copies of the gene/diploid cell. Data Curve Fitting—Data were curve fit to a four-parameter logistic equation using Kaleidagraph software, version 3.51 (Synergy Software, Reading, PA). Identification of Novel Aminopyrimidine Compounds That Block PDK1 Activity in Vitro and in Cells—From screening of compound libraries and lead optimization, we identified three potent and selective inhibitors of PDK1 termed BX-795, BX-912, and BX-320 which contain a common aminopyrimidine backbone (Fig. 1). The compounds were identified in a coupled assay measuring PDK1- and PtdIns-3,4-P2-mediated Akt activation, which could detect inhibitors of PDK1, AKT2, or other steps critical for activation of AKT2 (Fig. 2). We also found that the compounds potently inhibited PDK1 enzyme activity in a direct kinase assay format (IC50 values for BX-795, BX-912, and BX-320 were 11, 26, and 30 nm, respectively), although they failed to block preactivated AKT2 activity (IC50 > 10 μm) (not shown). These data indicate that BX-795, BX-912, and BX-320 are direct inhibitors of PDK1.Fig. 2Potencies of compounds for inhibiting PDK1 kinase activity in vitro. BX-320 (○), BX-795 (□), and BX-912 (▵) were assayed for inhibition of PDK1-mediated AKT2 activation and the direct inhibition of PDK1 as described under “Experimental Procedures.” Representative assay data are shown in the figure. The IC50 values for inhibition of the coupled PDK1/AKT2 assay, averaged from n independent experiments, are BX-320, 39 nm (n = 12); BX-795, 6 nm (n = 2); BX-912, 12 nm (n = 14). The average IC50 values for inhibition of the direct PDK1 assay are BX-320, 30 nm (n = 5); BX-795, 11 nm (n = 2); BX-912, 26 nm (n = 6).View Large Image Figure ViewerDownload Hi-res image Download (PPT) We also performed kinetic studies indicating that BX-912, BX-320, and BX-795 are competitive inhibitors of PDK1 activity with respect to its substrate, ATP, suggesting that they bind to the ATP binding pocket of PDK1 (data not shown). To investigate further how the inhibitors bind to PDK1, we prepared crystals of the PDK1 kinase domain (residues 51–359), as described by Biondi et al. (30Biondi R.M. Komander D. Thomas C.C. Lizcano J.M. Deak M. Alessi D.R. Van Aalten D.M. EMBO J. 2002; 21: 4219-4228Crossref PubMed Scopus (166) Google Scholar). We then introduced compounds into PDK1 crystals and determined the structures of enzyme-inhibitor complexes by x-ray diffraction methods. These studies confirmed that BX-912 and BX-320 bind to the ATP binding site of PDK1. The structure of BX-320 in the active site of PDK1, shown in Fig. 3, reveals two hydrogen bonds between two aminopyrimidine nitrogens and the amino acid backbone of Ala162, which lies in the hinge region of the molecule. The aminopyrimidine backbone of BX-912 adopts a similar orientation in the active site of PDK1 (data not shown). Several published structures of aminopyrimidines bound to the active site of cyclin-dependent kinase 2 show an analogous mode of binding (31Beattie J.F. Breault G.A. Ellston R.P. Green S. Jewsbury P.J. Midgley C.J. Naven R.T. Minshull C.A. Pauptit R.A. Tucker J.A. Pease J.E. Bioorg. Med. Chem. Lett. 2003; 13: 2955-2960Crossref PubMed Scopus (91) Google Scholar, 32Sayle K.L. Bentley J. Boyle F.T. Calvert A.H. Cheng Y. Curtin N.J. Endicott J.A. Golding B.T. Hardcastle I.R. Jewsbury P. Mesguiche V. Newell D.R. Noble M.E. Parsons R.J. Pratt D.J. Wang L.Z. Griffin R.J. Bioorg. Med. Chem. Lett. 2003; 13: 3079-3082Crossref PubMed Scopus (75) Google Scholar). To examine the kinase selectivity of our PDK1 inhibitors, we determined their effects on the in vitro activity of 10 different Ser/Thr and tyrosine kinases, including the related AGC kinases PKA and PKCα. As shown in Table I, BX-912, BX-320, and BX-795 displayed greater than 20-fold selectivity for PDK1 relative to the kinases in this panel, with one exception; BX-912 was 9-fold selective for PDK1 relative to PKA. BX-795 and BX-320 displayed significantly higher selectivity against PKA (140- and 35-fold, respectively).Table IPotencies of compounds for inhibition of PDK1 and other kinases in vitroKinaseBX-912BX-795BX-320IC50-Fold selectiveIC50-Fold selectiveIC50-Fold selectiveμmμmμmPDK1aCoupled AKT2 activation assays were performed as described under “Experimental Procedures.” Other kinase assays with purified recombinant enzymes used an ATP concentration at or below its Km value0.01210.00610.0391Chck10.83690.51900.8221PKA0.1190.841401.435PKC1.251059.316005.7146GSK3β7.46000.621004.0102CDK2bCDK2, cyclin-dependent kinase 2/cyclin E0.651050.43701.538EGFRcEGFR, epidermal growth factor receptor>10>850>10>1,700>10>256Insulin receptor6.1500>10>1,700>10>256c-Kit0.85700.32500.8923T-Fyn2.11756.41,100>10>256KDRdKDR, kinase insert domain-containing receptor0.41351.11901.421a Coupled AKT2 activation assays were performed as described under “Experimental Procedures.” Other kinase assays with purified recombinant enzymes used an ATP concentration at or below its Km valueb CDK2, cyclin-dependent kinase 2c EGFR, epidermal growth factor receptord KDR, kinase insert domain-containing receptor Open table in a new tab BX-320, BX-795, and BX-912 Block PDK1 Activity in Cells— We next characterized the effect of our compounds on the PI 3-kinase/PDK1/Akt/S6K1 pathway in PC-3 prostate cancer cells. PTEN-negative PC-3 cells display constitutive activation of Akt which is reflected in high levels of the PDK1 product, phospho-Thr308-Akt. PC-3 cells also express high levels of phospho-Thr389-S6K1, whose formation is dependent on PI 3-kinase and is thought to result from Akt signaling, although there are conflicting results in some situations (20Dufner A. Thomas G. Exp. Cell Res. 1999; 253: 100-109Crossref PubMed Scopus (607) Google Scholar, 33Balendran A. Currie R. Armstrong C.G. Avruch J. Alessi D.R. J. Biol. Chem. 1999; 274: 37400-37406Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar, 34Dufner A. Andjelkovic M. Burgering B.M. Hemmings B.A. Thomas G. Mol. Cell. Biol. 1999; 19: 4525-4534Crossref PubMed Scopus (147) Google Scholar). We treated PC-3 prostate cancer cells with and without compounds for 18 h and then determined the level of phospho-Thr308-Akt and phospho-Thr389-S6K1 in cell extracts by Western blotting. As shown in Fig. 4A, BX-795 and BX-320 greatly reduced the amount of both phospho-Thr308-Akt and phospho-Thr386-S6K1. BX compounds were as effective in blocking Akt and S6K1 phosphorylation as the PI 3-kinase inhibitor LY-294002 (100 μm) (Fig. 4A). We also saw similar effects after shorter times of compound treatment (2 h or 15 min, not shown). Quantification of Western blot signals revealed that BX-795 inhibited both Thr308-Akt and Thr389-S6K1 phosphorylation with an IC50 value of 300 nm, whereas BX-320 showed IC50 values of 1–3 μm (Fig. 4B). The more potent cell-based activity of BX-795 is consistent with its lower IC50 for inhibition of PDK1 activity in vitro. BX-912 displayed intermediate potencies (not shown). As a control, we stained Western blots for total Akt and S6K1 proteins. Compound treatment had no significant effect on Akt levels, whereas we observed a slight decrease in the level of total S6K1 protein which was too small to account for the dramatic loss of phospho-Thr389-S6K1 which we observed. In addition to the phosphorylation of Thr308 within the activation loop by PDK1, full activation of Akt requires the phosphorylation of Ser473, which is located in the C-terminal hydrophobic motif. The kinase responsible for this modification has not been clearly identified. As shown in Fig. 4A, we analyzed the effect of BX-795 and BX-320 on phospho-Ser473 levels in extracts prepared from compound-treated PC-3 cells. Both compounds reduced the levels of phospho-Ser473, although the potency of this effect appeared to be lower than that for inhibition of phospho-Thr308-Akt levels (Fig. 4A). These data indicate that PDK1 activity is critical for phosphorylation of both Ser473 and Thr308 of Akt, at least in PC-3 cells. We cannot exclude the possibility, however, that the less potent inhibition of Ser473 phosphorylation displayed by BX compou
Referência(s)