| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 36, 31924-31935, September 9, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶
From the
¶Department of Molecular Pharmacology, St. Jude Children's Research Hospital, Memphis, Tennessee 38105, the Department of Chemistry, Western Illinois University, Macomb, Illinois 61455, and the ||Department of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee Health Sciences Center, Memphis, Tennessee 38163
Received for publication, June 28, 2005 , and in revised form, July 11, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxyethyl)piperazino]butyl]-2-chlorophenoxazine (15B) (in vitro, IC50 
1-2 µM) were studied further. Inhibition of Akt phosphorylation correlated with inhibition of its kinase activity as determined in vitro after immunoprecipitation. Akt inhibitory phenoxazines did not inhibit the activity of recombinant phosphatidylinositol 3'-kinase, PDK1, or SGK1 but potently inhibited the kinase activity of recombinant Akt and Akt
PH, a mutant lacking the pleckstrin homology domain. Akt inhibitory phenoxazines blocked IGF-I-stimulated nuclear translocation of Akt in Rh1 cells and suppressed growth of Rh1, Rh18, and Rh30 cells (IC50 2-5 µM), whereas "inactive" derivatives were 
10-fold less potent inhibitors of cell growth. In contrast to rapamycin analogs, Akt inhibitory phenoxazines induced significant levels of apoptosis under serum-containing culture conditions at concentrations of agent consistent with Akt inhibition. Thus, the cellular responses to phenoxazine inhibitors of Akt appear qualitatively different from the rapamycin analogs. Modeling studies suggest inhibitory phenoxazines may bind in the ATP-binding site, although ATP competition studies were unable to distinguish between competitive and noncompetitive inhibition. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The design and development of small molecules that specifically inhibit the kinase activity of Akt and its signal transduction pathway in cancer cells are thus attractive approaches for the development of new cancer therapeutic agents. Currently, considerable effort is being placed into the development of novel specific inhibitors of Akt. The compounds include a number of phosphatidylinositol (PI) analogs (26-29), H-89 analogs (30), azapane derivatives (31), peptide inhibitors (32), Akt/protein kinase B signaling inhibitor-2 (33), and compounds containing planar aromatic heterocyclics (34). This latter group includes phenothiazine derivatives. Isozyme-selective inhibitors have been described and tested for their ability to inhibit both PI 3-kinase and Akt activities (35).
Previously, we have reported (36-40) the chemistry and biology of N10-substituted phenoxazines synthesized originally as modulators of P-glycoprotein-mediated multidrug resistance (MDR). Some of these N10-substituted phenoxazines demonstrated significant cytotoxicity per se (hence were not studied further), whereas several derivatives enhanced vincristine toxicity in cells with undetectable levels of P-glycoprotein. From these results, we concluded that at least part of the activity of some of these phenoxazine MDR modulators is mediated through an unknown, but P-glycoprotein-independent, mechanism. As it is now established that Akt signaling protects against cellular stress, including cytotoxic agents, we have investigated whether phenoxazine derivatives inhibit Akt and induce apoptosis. We screened a number of phenoxazine derivatives for their effects on Akt activation in cells derived from pediatric cancers. From among these compounds, we report the identification of a small group of novel lead compounds, which at low micromolar concentrations specifically block Akt activation and downstream signaling to substrates such as mTOR, p70 S6 kinase, and ribosomal protein S6 (S6). These agents do not affect activation of the upstream kinases, PDK-1, PI 3-kinase, or other kinases downstream of ras such as Erk-1/2. Furthermore, at low micromolar concentrations, under normal growth conditions, these small molecule inhibitors induce apoptosis in rhabdomyosarcoma cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Lines and Growth Conditions—The human cell lines Rh1, Rh18, and Rh30 have been described (41). Briefly, Rh1, Rh18, or Rh30 cells were grown in antibiotic-free RPMI 1640 medium (BioWhittaker, Walkersville, MD) supplemented with 10% fetal bovine serum (HyClone Laboratories, Logan, UT) and 2 mM L-glutamine (BioWhittaker) at 37 °C in an atmosphere of 5% CO2. For serum-free experiments, cells were cultured in modified N2E (MN2E) medium (DMEM/F-12; 1:1 mixture) supplemented with 1 µg/ml human holotransferrin, 30 nM sodium selenite, 20 nM progesterone, 100 µM putrescine, 30 nM vitamin E phosphate, and 50 µM ethanolamine. Cells in MN2E medium containing 5 µg/ml bovine fibronectin (Sigma) were plated and allowed to attach overnight at 37 °C in a humidified 5% CO2 atmosphere.
Cellular Screening for Inhibitors—Rh1, Rh18, or Rh30 cells were seeded at a density of 4 x 106/10-cm plate in serum-free medium for overnight attachment. Cells were exposed to 0.1% Me2SO or each of the phenoxazine derivatives for 1 h and then stimulated with IGF-I (10 ng/ml) for 10 min.
Western Blot Analysis—Cells were rapidly washed with ice-cold phosphate-buffered saline (PBS), placed on ice, and lysed in mammalian protein extraction reagent (Pierce) containing one CompleteTM mini protease inhibitor tablet (Roche Applied Science), 1 mM phenylmethylsulfonyl fluoride, 1 mM Na3VO4, and 1 mM NaF. Cellular debris was pelleted by centrifugation at 17,500 x g for 10 min at 4 °C. Protein concentration of the supernatants was measured by the bicinchoninic acid assay using bovine serum albumin as the standard (Pierce).
For the analysis of Akt, Erk-1/2, mTOR, p70 S6 kinase, ribosomal protein S6, and GSK-3, equivalent amounts of protein were separated on a 12% SDS-polyacrylamide gel (Bio-Rad) by electrophoresis and subsequently transferred to a nitrocellulose membrane (Bio-Rad). After a 1-h incubation in 1x TBS containing 0.05% Tween 20 and 5% blocking reagent (skim milk) (Upstate%20Biotechnology">Upstate Biotechnology Inc., Lake Placid, NY) at room temperature, the wet nitrocellulose membranes were then incubated with the appropriate antibodies from Cell Signaling Technologies (Waltham, MA) at the dilutions indicated: rabbit polyclonal antiserum specific for the phosphorylated Ser-473 or Thr-308 of Akt (dilution 1:1000); rabbit polyclonal antiserum specific for the phosphorylated Thr-202/Tyr-204 of Erk-1/2 (dilution 1:1000); rabbit polyclonal anti-serum specific for the phosphorylated Ser-2448 or Ser-2481 of mTOR (dilution 1:1000); rabbit polyclonal antiserum specific for the phosphorylated Thr-389 of p70 S6 kinase (dilution 1:4000); rabbit polyclonal antiserum specific for the phosphorylated Ser-235/236 of S6 (dilution 1:1000); or rabbit polyclonal antiserum specific for the phosphorylated Ser-21/9 of GSK-3
/ (dilution 1:1000). The secondary antibody was the horseradish peroxidase-conjugated goat anti-rabbit IgG antibody (dilution, 1:10,000). Immunoreactive protein was visualized by using Western Lightning chemiluminescence reagent (PerkinElmer Life Sciences).
To ensure that equivalent amounts of protein were loaded on each gel, all immunoblots were treated with stripping buffer (62.5 mM Tris-HCl, pH 6.7; 2% SDS; and 100 mM -mercaptoethanol) for 30 min at 50 °C and then incubated with one of the appropriate antibodies as follows: rabbit polyclonal antibody to Akt (dilution 1:1000; Cell Signaling Technology) or mouse monoclonal antibody to -tubulin (dilution 1:2000; Sigma). The secondary antibodies and detection of bound antibody were as described above.
Determination of Cellular Akt Kinase Activity—Rh1 cells were seeded in serum-free medium at a density of 4 x 106 per 10-cm plate. After 24 h, cells were exposed to Me2SO (0.1%) or phenoxazine at 5 µM for 1 h. Cells were then stimulated with ± IGF-I (10 ng/ml) for 10 min and washed once with ice-cold PBS. Cells were lysed in 200 µl of ice-cold 1x lysis buffer (20 mM Tris, pH 7.5; 150 mM NaCl; 1 mM EDTA; 1 mM EGTA; 1% Triton X-100; 2.5 mM sodium pyrophosphate; 1 mM -glycerol phosphate; 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride; and 1 mM leupeptin) and incubated for 10 min on ice. The cell lysates were centrifuged for 10 min at 17,500 x g at 4 °C. The volumes of the supernatants were adjusted so that each sample contained an equal amount of protein (150 µg); the supernatants were then incubated with immobilized (cross-linked) anti-Akt antibody for 3 h at 4 °C. The immunoprecipitates were pelleted and washed twice in ice-cold cell lysis buffer and twice in kinase buffer (25 mM Tris, pH 7.5; 5 mM -glycerol phosphate; 2 mM dithiothreitol; 0.1 mM Na3VO4; and 10 mM MgCl2). The pellets were suspended in 40 µl of kinase buffer containing 200 µM ATP and 1 µg of a GSK-3 fusion protein, which served as the substrate. After the suspensions were incubated at 30 °C for 30 min, the reaction was terminated by the addition of 3x SDS sample buffer (187.5 mM Tris-HCl, pH 6.8; 6% SDS; 30% glycerol; 150 mM dithiothreitol; and 0.03% bromphenol blue). The samples were boiled for 5 min, and the proteins were separated on a 12% SDS-polyacrylamide gel and subsequently transferred to a nitrocellulose membrane. Membranes were incubated with rabbit polyclonal anti-phospho-GSK-3/ (Ser-21/9) antibody.
In Vitro Inhibition of Recombinant Akt, AktPH, and SGK1—In vitro kinase assays were performed using active recombinant full-length Akt1/PKB (Upstate%20Biotechnology">Upstate Biotechnology, Inc.), active recombinant Akt lacking the pleckstrin homology domain, Akt1PH (Upstate%20Biotechnology">Upstate Biotechnology, Inc.), or active recombinant SGK1 (Upstate%20Biotechnology">Upstate Biotechnology, Inc.). Briefly, 10 ng of recombinant enzyme in 25 µl of 1x kinase buffer (25 mM Tris, pH 7.5; 5 mM -glycerol phosphate; 2 mM dithiothreitol; 0.1 mM Na3VO4; and 10 mM MgCl2), was mixed with 2.5 µl of Me2SO or phenoxazine derivative (5 µM). Samples were incubated on ice for 1 h at which time 1 µg of GSK-3 fusion protein, which served as the substrate, was added followed by ATP (200 µM) to each reaction mixture. After the suspensions were incubated at 30 °C for 30 min, the reaction was terminated by the addition of 3x SDS sample buffer (187.5 mM Tris-HCl, pH 6.8; 6% SDS; 30% glycerol; 150 mM dithiothreitol; and 0.03% bromphenol blue). The samples were boiled for 5 min, and the proteins were separated on a 12% SDS-polyacrylamide gel and subsequently transferred to a nitrocellulose membrane. Membranes were incubated with rabbit polyclonal anti-phospho-GSK-3/ (Ser-21/9) antibody.
Competition Experiments with ATP and Phenoxazines—Concentrations of compound 15B were prepared as 10x stocks in Me2SO ranging from 25 µM to 50 mM to give a final reaction concentration range of 2.5 µM to 5 mM. An ATP master mix was prepared containing 0.75 µl of [
-33P]ATP (PerkinElmer Life Sciences NEG302H), 0.5 µl of 10 mM ATP, and 1.25 µl of 1x kinase buffer for each sample. An enzyme/substrate master mix was prepared containing 10 µl of 1x kinase buffer, 5 µl of Akt peptide substrate (670 ng/µl) (Upstate%20Biotechnology">Upstate Biotechnology, Inc.), and 5 µl of active Akt (10 ng/µl) (Upstate%20Biotechnology">Upstate Biotechnology, Inc.) diluted from stock using 1x kinase buffer. The reactions were set up by adding 2.5 µl of the phenoxazine to the bottom of the tube followed by the addition of 2.5 µl of ATP mix near the bottom of the tube. The reaction was initiated by the addition of 20 µl of the enzyme/substrate master mix. After the master mix was added to all of the tubes, the samples were placed at 30 °C for 30 min. After incubation the samples were centrifuged briefly and spotted onto phosphocellulose in the same order as the addition of the master mix. After 2 min these samples were added to a beaker with 0.75% phosphoric acid in the same order as above. The samples were washed for 5 min three times in 0.75% phosphoric acid followed by 5 min in acetone. The squares were then placed on Whatman paper and allowed to dry. The radioactivity was quantitated by scintillation counting.
Molecular Modeling—Structures of inhibitors were built, and energy was minimized by using SYBYL 6.9.1 two-dimensional modeling software (version 6.9.1, Tripos Associates, St. Louis, MO). The GOLD program (version 2.1.2) (42) from Cambridge Crystallographic Data Center, UK, was used to dock the inhibitors into the ATP site of Akt, the three-dimensional coordinates of which were imported from the Protein Data Bank (code 1O6K
[PDB]
). GOLD uses a genetic algorithm to explore ligand conformational flexibility. The program also optimizes the torsion angles of serine and threonine hydroxyl groups as well as lysine NH+3 groups to achieve limited receptor flexibility. The ATP site was defined as residues lying within 15 Å of Thr-292. Up to 10 different docking solutions were obtained for each molecule, the docking being terminated when the top three solutions were within a root mean square deviation of 
1.5.
Photoaffinity Labeling of Recombinant Akt—The photoaffinity labeling experiments were performed on ice in MES buffer (20 mM MES, pH 6.1) using 200 ng of recombinant active Akt (Upstate%20Biotechnology">Upstate Biotechnology, Inc.) and 20 µM 8-azidoadenosine 5'-[-32P] triphosphate (8N3[-32P]ATP) (Altcorp) in a total reaction volume of 25 µl. All of the samples, including samples containing phenoxazine 15B (200 µM), were incubated on ice for 30 min. After 30 min of incubation, 8N3[-32P]ATP was added and allowed to bind for 1 min. In the sample containing ATP (200 µM or 1 mM), the ATP was premixed with 8N3[-32P]ATP prior to its addition to the reaction. Immediately after binding, the samples were photocross-linked using a handheld model UVLS-28 (UVP) UV lamp set at 254 nm at a distance of 4 cm from the sample surface for 1 min. After cross-linking, 15 µl of 3x SDS sample loading buffer was added to each sample. The entire sample was electrophoresed on a 4-20% Tris glycine gel followed by one rinse with H2O and overnight fixation (50% MeOH, 40% H2O, 10% acetic acid). After fixation the gel was washed two times for 30 min with H2O. The bands were detected using a STORM 860 PhosphorImager (Amersham Biosciences). To verify even loading of Akt, the gels were stained with Coomassie Brilliant Blue (Sigma).
PI 3-Kinase Assay—20 ng of recombinant p110 enzyme (AG Scientific, San Diego, CA), Me2SO (5 µl), phenoxazine derivative (5 µM), or wortmannin (5 µM) were placed on ice for 1 h in 100 µl of 1x kinase buffer (10 mM Tris, pH 7.4; 100 mM NaCl; and 5 mM MgCl2). 10 µg of phosphatidylinositol (Sigma) was then added to each sample, and the incubation was continued on ice for an additional 15 min. ATP (final concentration of 25 µM containing 30 µCi of [-32P]ATP) was added to each sample, and the reaction mixtures were incubated at 37 °C for 10 min. Reactions were terminated by adding 20 µl of 6 N hydrochloric acid. The sample was vortexed, and the lipids were extracted into 300 µl of MeOH/CHCl3 (1:1) mixture. After mixing gently and spinning at 10,000 x g for 5 min, 50 µl of the organic phase was spotted onto a silica-coated TLC plate (EMD Biosciences Inc., La Jolla, CA) and developed using a solvent system containing CHCl3/MeOH/H2O/NH4OH (60:47:11.3:2). The TLC plate was allowed to dry, and the bands were analyzed using a Storm 860 PhosphorImager (Amersham Biosciences).
PDK1 Kinase Assays—In vitro PDK1 activity assays were performed using a PDK1 assay kit (Upstate%20Biotechnology">Upstate Biotechnology, Inc.) with a slight modification of the manufacturer's instructions. Briefly, 10 ng of recombinant PDK1 enzyme, Me2SO (5 µl), or phenoxazines (5 µM) were incubated in 80 µl of 1x PDK assay dilution buffer on ice. After 1 h, 100 ng of SGK1 was added to each sample and incubated on ice for an additional 10 min. The samples were transferred to a 30 °C water bath and incubated for an additional 15 min. Then SGK1 substrate peptide (245 µM) followed by ATP (40 µM containing 10 µCi of [-32P]ATP) were added, and the reaction mixture was gently vortexed. Samples were incubated at 30 °C for 15 min with a gentle vortexing every 2 min. Samples were centrifuged, and 40 µl of the reaction mixture was spotted onto the center of a PE 81 phosphocellulose paper square. After 30 s, the filter was washed four times with 0.75% phosphoric acid and twice with acetone. The filter was drained and transferred into a scintillation vial to which 5 ml of scintillation mixture was added. The amount of incorporated radioactivity into the substrate was determined by scintillation counting. The assay for SGK1 kinase activity was performed as described above for the PDK1 assay.
Translocation of Akt in Rh1 Cells—Rh1 cells (2 x 105 per chamber) were grown on 2-well glass chamber slides (Falcon, Franklin Lakes, NJ) in serum-free medium containing fibronectin (10 µg/ml). After 20 h, cells were exposed to Me2SO (0.1%, vehicle control) or phenoxazine derivative (5 µM) for 1 h and then stimulated with IGF-I (10 ng/ml) for 20 min. Cells were washed twice with PBS and fixed in 4% formaldehyde for 30 min at room temperature. The samples were then rinsed twice with PBS and permeabilized with 1% Triton X-100 for 5 min at room temperature. After rinsing twice with PBS, the cells were incubated with an anti-Akt antibody (Rockland, West Chester, PA; 1:50 dilution) for 45 min at 37 °C. After rinsing three times with PBS, the slides were then incubated with an anti-IgG rabbit secondary antibody coupled to Alexa 488 (Molecular Probes, Eugene, OR) at a dilution of 1:50. The slides were washed and incubated with RNase. After rinsing twice with PBS, the slides were mounted in media containing TOPRO-3 (Molecular Probes, Eugene, OR) and then analyzed on a Leica confocal microscope.
Cell Growth Inhibition—Rh1, Rh18, and Rh30 cells at a density of 6000, 50,000 and 10,000 cells, respectively, were plated per well in 6-well flat-bottom tissue culture plates (Falcon) in complete medium. After 24 h at 37 °C, medium was replaced with fresh medium containing Me2SO (0.1%) or phenoxazines at concentrations ranging from 100 nM to 25 µM. The cells were further incubated for 6 days. Growth was assessed after lysing cells and counting nuclei. All measurements were made in triplicate.
Determination of Apoptosis—We used the ApoAlertTM annexin V-FITC apoptosis kit (Clontech) to evaluate the extent of apoptosis within cell populations. Cells (Rh1, 350,000 per 75-cm2 flask; Rh18, 800,000 per 75-cm2 flask; or Rh30, 500,000 per 75-cm2 flask) were grown overnight in complete medium. On day 1, cells were treated with Me2SO (0.1%; vehicle control) or phenoxazine derivative. After 4 days, the cells were trypsinized, washed with PBS, and resuspended in 200 µl of binding buffer. Cells were incubated with 10 µl of annexin V-FITC (final concentration, 1 µg/ml) and 500 ng of propidium iodide in a final volume of 410 µl. Cells were incubated at room temperature in the dark for 10 min before flow cytometric analysis (FACSCalibur, BD Biosciences) was performed as described (41).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
To determine whether these effects were general, the studies were extended to other cell lines. Rh18 or Rh30 rhabdomyosarcoma cells grown in serum-free medium were exposed to compounds 10B, 14B, 15B, or 20B at 5 µM for 1 h and then stimulated with IGF-I for 10 min. Phospho-Akt (Ser-473), phospho-mTOR (Ser-2448 or Ser-2481), or phospho-S6 (Ser-235/236) was checked using the respective phospho-specific antibody as shown in Fig. 2C for Rh18 or Fig. 2D for Rh30 cells. In both cell lines, the IGF-I-induced phosphorylation of Akt, mTOR, and S6 was effectively blocked by all of the compounds except 20B. After the membrane was stripped of bound antibody, it was incubated with the anti-Akt antibody to determine the total amount of Akt. The data confirm that equal amounts of protein were loaded. Taken together, these results further confirm that Akt-mediated activation of mTOR/p70 S6 kinase/S6 pathway in three cancer cell lines is blocked by phenoxazine derivatives.
Phenoxazine Derivatives Inhibit Akt Kinase Activity in Cells—These studies strongly suggest that phenoxazines inhibit Akt activation. To determine this directly, we studied the effects of compounds 10B and 15B on the Akt kinase activity in cells. Activation of Akt by IGF-I was evaluated by assessing the phosphorylation of Akt (Ser-473) or the in vitro kinase activity of protein immunoprecipitated by the anti-Akt antibody. To determine whether changes in Akt phosphorylation were correlated with alterations in kinase activity, we examined the phosphorylation status of a target downstream from Akt, e.g. GSK-3. Rh1 cells grown in serum-free medium were exposed to 0.1% Me2SO or 5 µM of 10B or 15B for 1 h and then stimulated with IGF-I for 10 min. Cell lysates were then immunoprecipitated with immobilized Akt antibody. Immunoprecipitates were used in vitro to phosphorylate a GSK-3 fusion protein. Phosphorylation was completely inhibited in cells treated with 10B or 15B (Fig. 3). These results indicate that phenoxazines effectively block the activity of endogenous Akt in Rh1 cells.
Phenoxazines Do Not Inhibit PI 3-Kinase Activity in Vitro—As discussed above, the finding that the phenoxazine derivatives do not inhibit IGF-I-induced phosphorylation of Erk1/2 implies that they do not inhibit PI 3-kinase. However, cells treated with phenoxazines 10B or 15B exhibit many of the effects observed in cells treated with PI 3-kinase inhibitors such as wortmannin (data not shown). This is not surprising because PI 3-kinase is required for both the association of Akt with the cell membrane via the PH domain of Akt and activation of the Akt kinase function through the phosphorylation of Ser-308 by the 3-phosphoinositide-dependent protein kinase PDK1. To verify that the phenoxazines were not targeting PI 3-kinase, we performed in vitro kinase assays using recombinant enzyme. By using phosphatidylinositol (PI) as the substrate and [-32P]ATP as the phosphate donor, the kinase activity of samples treated with wortmannin, 10B, or 15B at 5 µM was compared with an untreated sample. The lipids were resolved by thin layer chromatography and quantitated using a PhosphorImager. As expected, the untreated sample showed robust phosphorylation of PI as can be seen by the levels of phosphatidylinositol 3-phosphate detected (Fig. 4, 1st lane). The PI 3-kinase activity in samples incubated with 10B or 15B was comparable with the untreated sample (Fig. 4, 2nd and 3rd lanes), whereas the wortmannin-treated sample had barely detectable levels of activity as shown by the lack of visible phosphatidylinositol phosphate (Fig. 4, 4th lane) and the low level of detectable counts. These results indicate that the phenoxazines do not inhibit the activity of PI 3-kinase.
Phenoxazines Do Not Inhibit the Kinase Activity of PDK-1—Although direct inhibition of Akt by the phenoxazines would be consistent with the observed effects on the Akt signaling pathway, inhibition of PDK1 and the consequent failure of PDK1 to phosphorylate and activate Akt would also be consistent with these effects. In addition, both Akt and PDK1 are members of the AGC family of serine-threonine kinases. To address the question of whether or not PDK1 was inhibited by 10B or 15B, an in vitro coupled-kinase assay was performed. Recombinant PDK1 was preincubated for 1 h with phenoxazines 10B, 15B, or 0.1% Me2SO (control). After this preincubation, the PDK1 substrate SGK1 was added along with ATP and incubated for 15 min followed by the addition of an SGK1 substrate peptide and [-32P]ATP. PDK1 activity was measured by the extent to which SGK1 transferred phosphate from [-32P]ATP to the substrate peptide. The amount of incorporation of phosphate into the substrate was determined by binding the substrate to phosphocellulose filters and quantitating with scintillation counting. The results indicate that PDK1 is not inhibited by the addition of either 10B or 15B (Fig. 5).
Phenoxazines Inhibit Recombinant Akt Kinase Activity in Vitro—To study the Akt inhibitory activity of phenoxazines, we have used the phosphorylation status of GSK-3 protein, a substrate of Akt, as a read out. Akt1/PKB active enzyme (recombinant protein expressed in Sf21 cells, 10 ng/reaction) was preincubated with compound 10B or 15B at 5 µM for 2 h on ice prior to initiating the kinase assay. The results show that compound 15B completely blocked the phosphorylation of GSK-3 (Fig. 6, 2nd and 3rd lanes), whereas inhibition by 10B was nearly complete. These results indicate that phenoxazine derivatives are targeting Akt directly and inhibit its kinase function.
Phenoxazines Do Not Block Activation of Akt via Its PH Domain—All three Akt isoforms consist of a conserved domain structure as follows: an amino-terminal pleckstrin homology (PH) domain, a central kinase domain, and a carboxyl-terminal regulatory domain that contains the hydrophobic motif, a characteristic of AGC kinases. The PH domain is a phosphoinositide-binding motif found in a number of signal-transducing proteins, including Akt that gives membrane-binding properties to the host proteins. The PH domain interacts with membrane lipid products such as PtdIns(3,4,5)P3 produced by PI 3-kinase (45). Biochemical analysis revealed that the PH domain of Akt binds preferentially PtdIns(3,4,5)P3 (46, 47), resulting in the recruitment of Akt to the plasma membrane (22, 47, 48). Binding induces a conformational change in Akt, exposing the critical Thr-308 residue in the activation loop to phosphorylation by PDK-1. For full activation, Akt is subsequently phosphorylated at Ser-473 by an as yet unidentified kinase termed PDK-2 (3, 22, 47). The observation that the phenoxazines do not inhibit PDK1 activity suggests that interaction with the PH domain of Akt may not be a requirement for the inhibitory effect observed with these compounds. To determine whether the absence of the PH domain of Akt affects the ability of 10B or 15B to inhibit Akt kinase activity, in vitro kinase assays were performed using a recombinant Akt lacking the PH domain. Because GSK-3 is one of the major downstream phosphorylation targets of Akt, we used GSK-3 phosphorylation as an indication of Akt activity. In agreement with previous studies, the deletion of the PH domain results in a higher level of kinase activity than the full-length protein (Fig. 6, compare 1st and 4th lanes). Despite the higher kinase activity in the control sample, the ability of 10B and 15B to inhibit Akt kinase activity was unaffected by the deletion of the PH domain (Fig. 6, 5th and 6th lanes), indicating that these compounds are unlikely to mediate their effects by blocking association of Akt with the membrane.
Phenoxazines Do Not Inhibit Recombinant SGK1 Kinase Activity in Vitro—Among the AGC family of kinases to which Akt belongs, SGK1 has the highest degree of homology to Akt. In addition to conservation of homologous sites at threonine 308 and serine 473, SGK1 also has a high degree of homology in the catalytic kinase domain and the carboxyl-terminal Ser/Thr kinase extension. To determine whether SGK1 was also a target of phenoxazines 10B or 15B, a kinase assay using 10 ng/reaction of recombinant active SGK1 was performed under the identical conditions described for the Akt kinase experiments. The results indicate that neither 10B nor 15B interfere with SGK1 kinase activity (Fig. 6B). This result is consistent with the results for the coupled enzyme assay measuring PDK1 activity where no difference was observed in the phosphorylation of the substrate peptide by SGK1 in the phenoxazine-treated versus -untreated samples.
|
|
|
|
|
The presence of electronegative groups in the distal amine group at the end of the hydrocarbon chain bridge as in compounds 2B, 5B, 5C, 9B, 12B, and 18B renders the molecules inactive due possibly to repulsive interactions. The loss of activity in analogs where the flexible hydrocarbon bridge is replaced by a shorter and rigid amidomethyl moiety as in compounds 16B, 17B, 18B, 19B, 20B and 21B may also stem from unfavorable steric and electrostatic interactions in the narrow groove below the glycine-rich loop.
Phenoxazines Enhance the Binding of 8-Azidoadenosine 5'-[-32P]Triphosphate to Akt—To examine further the relationship between the inhibitory phenoxazines, and the ATP-binding site of Akt, photoaffinity labeling experiments were performed using 254 nm wavelength UV light to cross-link 8N3[-32P]ATP to Akt. In this experiment, 200 ng of recombinant Akt was incubated on ice for 30 min either with or without phenoxazine 15B prior to the addition of 20 µM 8N3[-32P]ATP. Because this assay measures binding as opposed to activity, the samples were incubated only briefly (1 min) on ice after the addition of 8N3[-32P]ATP and then UV cross-linked for 1 min. Photoactivation of 8N3[-32P]ATP resulted in only very low levels of cross-linking with Akt (Fig. 8A, 1st lane). Changes in the duration of photoactivation, pH of the buffer, incubation time, and temperature or the inclusion of divalent cations did not increase the extent of cross-linking (data not shown). The low level of cross-linking observed in these reactions may be because of the 8N3[-32P]ATP modification of ATP, which has been shown previously to adopt a conformation that fits poorly into the ATP binding domain of some proteins (50). The addition of a 10-fold excess of ATP (200 µM) relative to 8N3[-32P]ATP does not further reduce the extent of cross-linking (Fig. 8A, 2nd lane). Most surprisingly, the addition of a 10-fold excess of 15B (200 µM) dramatically increases the extent of covalent 8N3[-32P]ATP-Akt complexes formed by photoactivation (Fig. 8, A, 3rd lane, and B, 2nd lane). Furthermore, if a 5-fold excess of ATP (1 mM) relative to 15B is included in the reaction, the extent of cross-linking is reduced to almost the same level as is observed in the sample without 15B (Fig. 8B, 3rd and 1st lanes, respectively). These results clearly indicate that 15B facilitates the interaction of 8N3[-32P]ATP with Akt and that ATP is able to compete with 8N3[-32P]ATP or possibly 15B to reduce the interaction of 8N3[-32P]ATP with Akt.
|
|
52 and 75% apoptosis in Rh1, 34 and 33% apoptosis in Rh18, and 43 and 89% apoptosis in Rh30 cells. Cells exposed to phenoxazines appeared to lose membrane integrity late in apoptosis yet remained attached to the culture dish. Furthermore, a significant increase in the proportion of apoptotic cells was evident after treatment with 11B, 13B, and 14B (data not shown). To assess whether the apoptosis of these cells in response to 10B or 15B was because of a general toxic effect of the compound rather than because of inhibition of Akt, we compared 10B or 15B with other phenoxazines that poorly inhibit Akt in vitro (12B or 20B). Because of their overall chemical similarity with the Akt inhibitory phenoxazines, 12B and 20B might also be expected to exhibit cellular toxicity if the mechanism was general and independent of Akt. But in contrast to the effect observed with 10B and 15B, neither 12B nor 20B induced apoptosis (data not shown). Thus, there is a correlation between the ability of a compound to inhibit Akt in cells and its ability to induce apoptosis upon treatment of intact cells.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
50% but did not inhibit IGF-I stimulation of Erk1/2 phosphorylation. The two compounds 10-[4'-(N-diethylamino)butyl]-2-chlorophenoxazine (10B) and 10-[4'-[(-hydroxyethyl)piperazino]butyl]-2-chlorophenoxazine (15B) showed the greatest potency. In contrast, compounds such as 10-[4'-(N-morpholinobutyl]-2-chlorophenoxazine (12B) and 10-(N-pyrrolidinoacetyl)-2-chlorophenoxazine (20B) showed no inhibitory activity in this assay. Consistent with inhibition of Akt, the most potent inhibitors blocked IGF-I-stimulated phosphorylation of mTOR, p70 S6 kinase, and S6, downstream of Akt in the signaling cascade. More importantly, neither 12B nor 20B exerted any detectable effect on phosphorylation of Akt or downstream effectors. In serum-starved cells, IGF-I rapidly activated Akt, as demonstrated by in vitro kinase assay of immunoprecipitated enzyme using a GSK-3 peptide as substrate. In contrast, Akt immunoprecipitated from cells exposed to 10B or 15B the enzyme was inactive. Although we have not undertaken an exhaustive study of other kinases, neither 10B nor 15B was found to inhibit recombinant PI 3-kinase, PDK1, or the closest member of the AGC kinase family SGK1. However, these phenoxazines potently inhibited the activity of recombinant Akt and the mutant enzyme lacking the PH domain (AktPH). Thus, we conclude that the phenoxazines are not like the PH domain-dependent isozyme-specific Akt inhibitors reported recently (35). The precise mechanism by which phenoxazines inhibit Akt activity is unclear. Initial studies do not rule out the possibility that phenoxazines may be competing with ATP for binding to the enzyme. However, phenoxazines may also be locking the enzyme in an inactive conformation or acting in an allosteric manner. Modeling studies suggest that the inhibitory phenoxazines may bind in the ATP site, whereas those with shorter, more rigid amidomethyl side chains (i.e. 16B-20B) may be sterically hindered from binding. For compounds with electro-negative moieties (i.e. 2B, 5B, etc.), there may be repulsion from the site.
The results from the photoaffinity cross-linking experiments, however, complicate the interpretation of the modeling data. The binding of 15B greatly facilitates the interaction of 8N3[-32P]ATP with Akt. The simplest interpretation of the data is that 15B binding to Akt results in the enzyme assuming a conformation that is more favorable for the binding of 8N3[-32P]ATP. The observation that ATP competes effectively with the 8N3[-32P]ATP would argue against the phenoxazines acting as competitive inhibitors for ATP as the modeling data suggest but rather as noncompetitive inhibitors acting allosterically.
Clearly, further studies will be required to determine the specific interaction of the phenoxazines with Akt. Additional studies involving co-crystallization of Akt along with 10B or 15B are planned to resolve the mode of phenoxazine binding to Akt.
Consistent with inhibition of Akt activation, 10B and 15B suppressed IGF-I-stimulated nuclear translocation of Akt. At low micromolar concentrations, these compounds were effective inhibitors of cell growth in three lines examined (Rh1, Rh18, and Rh30). In contrast, compounds 12B and 20B were at least 10-fold less potent. Of interest is that both Akt inhibitors induced very significant levels of apoptosis, even under serum-containing conditions of cell growth. This is important, as inhibition of mTOR signaling by rapamycin induces significant apoptosis only under serum-free conditions in Rh1 and Rh30 cells. Thus, these results demonstrate qualitative differences in cellular responses when different kinases are targeted within the same pathway. mTOR in the rictor-Lst8 (rapamycin-insensitive) complex has been proposed as the putative PDK2 (49). Our data do not formally address whether Akt-inhibitory phenoxazines inhibit this complex. However, preliminary studies show that in vitro neither 10B nor 15B inhibits mTOR kinase activity. Also of interest is that although treatment of cells with rapamycin increases the proportion of cells in the G1 phase of the cell cycle, no such accumulation was observed in cells exposed to either 10B or 15B (data not shown), further demonstrating the differences between Akt and mTOR inhibitors. Very preliminary in vivo toxicity testing with one compound (11B) suggests that it is well tolerated, causing sedation similar to that induced by phenoxazine, the parent compound (data not shown). Studies to determine the in vivo pharmacokinetics as well as initial studies to determine whether these compounds have anti-tumor properties are currently being undertaken.
In summary, we have identified a series of N10-substituted phenoxazines that are potent inhibitors of Akt activity in vitro and inhibit Akt signaling in cells. These compounds are growth inhibitory and induce significant apoptosis in cell lines under serum-containing conditions of culture and hence have differential cellular activities from mTOR inhibitors that induce only low level apoptosis under similar culture conditions.
|
FOOTNOTES |
|---|
Both authors contributed equally to this work.
** To whom correspondence should be addressed: Dept. of Molecular Pharmacology, St. Jude Children's Research Hospital, 332 North Lauderdale St., Memphis, TN 38105. Tel.: 901-495-3440; Fax: 901-495-4290; E-mail: peter.houghton{at}stjude.org.
1 The abbreviations used are: Akt, oncogene from AKR mouse thymoma; IGF-1, insulin-like growth factor 1; Erk-1/2, extracellular signal-regulated kinase 1/2; mTOR, mammalian target of rapamycin; Rh, rhabdomyosarcoma; PI, phosphoinositide; PDK1/2, phosphoinositide 3-phosphate-dependent kinase 1/2; SGK1, serum glucocorticoid-regulated kinase 1; PH, pleckstrin homology domain; PTEN, phosphatase and tensin homolog; MDR, multidrug resistance; GSK-3, glycogen synthase kinase 3; Me2SO, dimethyl sulfoxide; p70 S6K, p70 ribosomal protein S6 kinase; AGC, protein kinase A, protein kinase G, protein kinase C; S6, ribosomal protein S6; PBS, phosphate-buffered saline; FITC, fluorescein isothiocyanate; MES, 4-morpholineethanesulfonic acid; Ptd, phosphatidylinositol.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), 15B (
), 12B (
), or 20B (
). Growth was determined by lysing the cells and counting the nuclei. Data represent the mean ± S.D. values of three separate experiments. b, phenoxazines induce apoptosis. Rh1 cells (panel A), Rh18 cells (panel B), and Rh30 cells (panel C) were grown in complete medium and treated with Me2SO (0.1%; vehicle control) or 6.5 µM 10B or 15B for Rh1 or 7.5 µM 10B or 15B for Rh18 and Rh30 cell lines for 4 days. Apoptosis was evaluated by using the ApoAlert assay. Similar results were obtained in three independent experiments. Ordinate, uptake of propidium iodide (PI); abscissa, annexin V-FITC fluorescence (top panels). Viable cells are represented in the bottom left quadrant. The bottom panels show the corresponding distribution of annexin V-FITC staining of cell population.