LY294002 and LY303511 Sensitize Tumor Cells to Drug-Induced

Research Article

LY294002 and LY303511 Sensitize Tumor Cells to Drug-Induced Apoptosis via Intracellular Hydrogen Peroxide Production Independent of the Phosphoinositide 3-Kinase-Akt Pathway

Tze Wei Poh1 and Shazib Pervaiz1,2,3

1Department of Physiology, 2Oncology Research Institute, National University Medical Institute, Faculty of Medicine, and 3Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore, Singapore

Abstract then recruits Akt/PKB to the membrane where it becomes The phosphoinositide 3-kinase (PI3K)-Akt pathway is consti- phosphorylated and thus activated by the phosphatidyl-dependent tutively active in many tumors, and inhibitors of this kinase-1. Akt/PKB has been implicated in the regulation of a variety prosurvival network, such as LY294002, have been shown to of signal transduction pathways that mediate gene transcription sensitize tumor cells to death stimuli. Here, we report a novel, (5–10), cell cycle events (11–15), cell proliferation (16–19), glycogen PI3K-independent mechanism of LY-mediated sensitization of and protein metabolism (20–22), angiogenesis (23, 24), DNA repair LNCaP prostate carcinoma cells to drug-induced apoptosis. (25, 26), and cell survival (27). Recent evidence strongly suggests Preincubation of tumor cells to LY294002 or its inactive that activated Akt/PKB not only contributes to oncogenic analogue LY303511 resulted in a significant increase in proliferative ability but also through phosphorylation of downstream targets, such as Bad, confers resistance to drug-induced intracellular hydrogen peroxide (H2O2) production and enhanced sensitivity to nonapoptotic concentrations of the apoptosis (28, 29). chemotherapeutic agent vincristine. The critical role of Interestingly, aside from the conventional signals, such as activation of tyrosine receptor kinases, a causal relationship intracellular H2O2 in LY-induced death sensitization is corroborated by transient transfection of cells with a vector between intracellular reactive oxygen species production, in containing human catalase gene. Indeed, overexpression of particular hydrogen peroxide (H2O2), and activation of Akt/PKB catalase significantly blocked the amplifying effect of LY has been described recently; exogenous addition of H2O2 was pretreatment on caspase-2 and caspase-3 activation and cell shown to activate Akt/PKB phosphorylation (30–33). However, death triggered by vincristine. Furthermore, the inability of other data seem to suggest that activation of the PI3K-Akt pathway wortmannin, another inhibitor of PI3K, to induce an increase could generate other reactive species like nitric oxide and superoxide, which would further enhance the downstream effects in H2O2 production at doses that effectively blocked Akt phosphorylation provides strong evidence to unlink inhibition of this pathway (33–36). These are intriguing findings, given our previous reports highlighting the role of intracellular redox status of PI3K from intracellular H2O2 production. These data strongly support death-sensitizing effect of LY compounds in the sensitivity of cancer cells to death stimuli (37–44). We independent of the PI3K pathway and underscore the critical showed that a slight pro-oxidant intracellular milieu, invariably associated with the transformed phenotype, favors cell survival role of H2O2 in creating a permissive intracellular milieu for efficient drug-induced execution of tumor cells. (Cancer Res either by its positive effect on cell proliferation (45) or by impeding 2005; 65(14): 6264-74) death execution pathway(s) (41, 43, 44, 46). In contrast, a significant increase in intracellular H2O2 production and downstream Introduction acidification provides an environment conducive for apoptotic signaling (38, 39, 42). The process of oncogenesis is a function of the balance between In an effort to study downstream effectors of the Akt/PKB the rates of cellular proliferation and death. Consistent with this, signaling pathway, a variety of PI3K inhibitors are currently in use, deficient or defective apoptosis circuitry and/or constitutive but with varying effects. For example, wortmannin, a cell-permeant activation of prosurvival pathways are commonly associated with fungal metabolite, acts as a potent inhibitor of PI3K; however, it is most cancers. Examples of the former are overexpression of nonspecific as it also elicits strong inhibitory effect on mitogen- antiapoptotic members of the Bcl-2 family (Bcl-2 and Bcl-xL)or activated protein kinases (MAPK; ref. 47). LY294002 (LY29) is a silencing of the proapoptotic counterparts, such as Bax, and commonly used pharmacologic inhibitor of PI3K, where it acts on overexpression of the inhibitor of apoptosis proteins (IAP; cIAP1, the ATP-binding site of the PI3K enzyme, thus selectively inhibiting cIAP2, XIAP, and survivin; refs. 1–3). With respect to the prosurvival the PI3K-Akt nexus. Consistent with this observation, LY29 has signals, a common finding reported in many cancers is the been successfully used to enhance sensitivity of cancer cells to constitutive activation of Akt/protein kinase B (PKB), a down- drug-induced apoptosis, and this effect has generally been stream mediator of phosphoinositide 3-kinase (PI3K) activation attributed to its PI3K-Akt blocking activity (29, 48–53). Although signal. PI3K phosphorylates phosphatidylinositol (3,4)-diphosphate LY29 is comparatively more selective in inhibiting PI3K and would to form phosphatidylinositol (3,4,5)-triphosphate (PIP ; ref. 4). PIP 3 3 seem to be a more useful inhibitor of PI3K (54, 55), recent reports have suggested that this compound might have effects other than inhibiting PI3K (56–58). In this regard, LY29 and LY303511 (LY30), a Requests for reprints: Shazib Pervaiz, Department of Physiology, MD902-05, kinase inactive analogue of LY29 that contains a single atom Faculty of Medicine, National University of Singapore, Singapore, Singapore 117597. Fax: 65-67788161; E-mail: [email protected]. substitution in the morpholine ring, have each been shown to block I2005 American Association for Cancer Research. K(V) currents via a non-PI3K-dependent mechanism in rat

Cancer Res 2005; 65: (14). July 15, 2005 6264 www.aacrjournals.org

6 Figure 1. LY29 triggers intracellular H2O2 production in tumor cells. Tumor cells (1 Â 10 ) from a variety of cell lines were treated with 25 Amol/L LY29 for 1 hour in the presence or absence of 1,000 units/mL catalase. Cells were loaded with the H2O2-sensitive probe DCFH-DA (5 Amol/L) for 15 minutes and intracellular H2O2 was determined by the shift in DCFH-DA fluorescence detected by flow cytometry as detected by flow cytometry. pancreatic h cells (57) and inhibit monocyte chemoattractant LY30 (25 Amol/L) and then treated with 0.02 Amol/L vincristine overnight protein-1 (MCP-1) expression in human umbilical vein endothelial for 18 hours. Cytotoxicity was determined by the crystal violet assay. After cells (56). Thus, the ability of LY30 to behave like LY29 in these drug exposure, cells were washed once with PBS and crystal violet solution studies indicates that LY29 has other effects that are independent (0.25 mL) was added to each well and incubated for 15 minutes. The excess crystal violet solution was washed away with distilled water and the of its ability to inhibit PI3K activity. Based on these recent reports, remaining crystals were dissolved in a 1% SDS in 1Â PBS solution. Viability the effects of LY29 cannot be exclusively attributed to inhibition of was determined by absorbance at 595 nm wavelength using an automated the PI3K-Akt pathway. ELISA reader. Cell viability experiments were thus similarly done with Here, we propose an alternative role for LY29 as an inducer of 50 Amol/L caspase-2 and caspase-3 inhibitors and the pan-caspase intracellular H2O2 generation independent of its PI3K inhibitory inhibitor, Z-VDVAD-fmk, Z-DEVD-fmk (R&D Systems, Minneapolis, MN), activity. We show that the intracellular H2O2 produced is inde- and ZVAD-fmk (Biomol, Plymouth Meeting, PA), respectively. Cells were pendent of mitochondria in a variety of cell lines. At concentrations preincubated for 1 hour with the caspase inhibitors before addition of LY29/ used to inhibit PI3K activity, LY29 exposure results in a significant LY30 and vincristine. Flow cytometric analysis of intracellular hydrogen peroxide intracellular increase in H2O2 production in tumor cell lines. Furthermore, in LNCaP prostate carcinoma cells, we also show that concentration. Intracellular concentration of H2O2 was determined by staining with the redox sensitive dye 5-(and-6)-chloromethyl-2V,7V-dichloro- both LY29 and its kinase inactive analogue LY30 sensitize cancer fluorescin diacetate (CM-H2DCFDA; Molecular Probes, Eugene, OR), which cells to drug-induced apoptosis through intracellular H2O2 becomes fluorescent when oxidized by H2O2 and its free radical products production independent of the PI3K-Akt pathway. These findings (40). Briefly, cells were washed once with PBS, loaded with 5 Amol/L CM- provide a hitherto undefined activity of LY29 and LY30 and H2DCFDA at 37jC for 15 minutes, and analyzed by flow cytometry (Coulter highlight their potential for use for favorably tailoring the intra- EPICS Elite ESP) using an excitation wavelength of 488 nm. At least 10,000 cellular milieu of cancer cells for drug-induced death execution. events were analyzed. Detection of total Akt/protein kinase B and Akt/protein kinase B phosphorylation levels. LNCaP cells (1.5 Â 106-2 Â 106) were plated in a Materials and Methods T25 flask overnight followed by exposure to the various triggers. Cells were Determination of cell viability. The human lymph node carcinoma of harvested from the flask, washed once with PBS, and then lysed with cell the prostate (LNCaP) cell line was purchased from American Type Culture lysis buffer [150 mmol/L NaCl, Tris-HCl (pH 7.4), 1% NP40]. Cell lysate Collection (Rockville, MD) and maintained in RPMI 1640 supplemented (200 Ag) was then electrophoresed on an 8% acrylamide gel. Antibodies 473 with 10% fetal bovine serum, 1% L-glutamine, 1% penicillin, and 100 Amol/L were used to probe for total Akt and phosphorylated Akt/PKB at the Ser sodium pyruvate. In a typical survival assay, LNCaP cells (2.5 Â 105/well) position (Cell Signaling, Beverly, MA). Protein blots were probed with anti- plated in 24-well plates were preincubated for 1 hour with either LY29 or h-actin (Sigma-Aldrich, St. Louis, MO) to check for equal protein loading. www.aacrjournals.org 6265 Cancer Res 2005; 65: (14). July 15, 2005

Nonradioactive Akt/protein kinase B immunoprecipitation kinase (human bladder carcinoma) cells (Fig. 1). This increase in H2O2 also activity assay. Determination of the Akt/PKB kinase activity was done via persisted even in cells stably overexpressing Bcl-2 (CEM-Bcl-2; 6 the Akt kinase assay kit (Cell Signaling). LNCaP cells (3 Â 10 ) were plated Fig. 1). Similar experiments with LY29 were also done on isolated in a 100 mm Petri dish and treated with increasing concentrations of LY29. mitochondria; however, the increase in H O was not observed in The culture medium was then removed and the cells were washed once 2 2 this particular system. with ice-cold PBS and cell lysis buffer (1 mL) was added to each plate and incubated for 10 minutes. Cells were then scraped off the plates and LY294002-induced hydrogen peroxide production is inde- immobilized Akt monoclonal antibody was used to immunoprecipitate Akt pendent of its phosphoinositide 3-kinase-Akt inhibitory from cell extracts. Then, an in vitro kinase assay was done using glycogen activity in LNCaP cells. LNCaP prostate cancer cells that express synthase kinase (GSK)-3 fusion protein as a substrate. Phosphorylation of constitutively active form of Akt/PKB were incubated with GSK-3 was detected by Western blot analysis using a phospho-GSK-3a/h increasing concentrations of LY29 (1-25 Amol/L) and the effect 21 9 (Ser /Ser ) antibody (Cell Signaling). on Akt/PKB activation and intracellular H2O2 production was Determination of the colony-forming ability of tumor cells. LNCaP assessed. As expected, concentrations of LY in excess of 5 Amol/L cells (4,000-8,000) were pretreated with LY29 or LY30 and then treated with effectively blocked Akt phosphorylation, assayed by Western blot A 0.02 mol/L vincristine for 18 hours. As overnight treatment with LY29 or analysis with the phosphospecific antibody and by the ability of LY30 had a slight effect on cell proliferation, cell numbers were adjusted active Akt/PKB to phosphorylate its substrate GSK-3a/h (Fig. 2A). accordingly to take this into account when cells were plated. After treatment, cells were then washed and replated into 100 mm culture dishes Intriguingly, LY29 was still able to trigger intracellular H2O2 and left to form colonies over a period of 7 to 10 days. Culture dishes were production at a concentration (1 Amol/L; Fig. 2B) that did not stained with crystal violet and the number of colonies in a 2 Â 2 cm grid (on the culture plates) was scored to determine the colony-forming ability of cells. Only colonies containing >50 cells were counted. Determination of caspase-2 and caspase-3 activities. Caspase-2 and caspase-3 activities were assayed by using 7-amino-4-trifluoromethyl coumarin (AFC)–conjugated and amino-4-methyl coumarin (AMC)– conjugated substrates, respectively (Biomol). LNCaP cells (2.5 Â 105/mL) were preincubated with LY29 or LY30 (25 Amol/L) for 1 hour and then incubated with 0.02 Amol/L vincristine for various time points. Cells were then harvested and washed with 1Â PBS, resuspended in chilled cell lysis buffer (BD PharMingen, San Diego, CA), and incubated on ice for 10 minutes. Reaction buffer (50 AL; 2Â; 10 mmol/L HEPES, 2 mmol/L

EDTA, 10 mmol/L KCl, 1.5 mmol/L MgCl2, 10 mmol/L DTT) and fluorogenic caspase-specific substrate (1 AL; DEVD-AFC for caspase-2 and VDVAD-AMC for caspase-2) were added to each sample and incubated at 37jC for 1 hour. Detection of caspase-3 activity was determined by the relative fluorescence intensity at 505 nm following excitation at 400 nm using a spectrofluo- rometer, whereas detection of caspase-2 activity was determined by the relative fluorescence intensity at 460 nm following excitation at 380 nm. Propidium iodide staining for DNA fragmentation. Cells were fixed then stained with propidium iodide for DNA content analysis as described elsewhere (40). A total of 10,000 events were analyzed by flow cytometry using an excitation wavelength set at 488 nm and emission at 610 nm. Transient transfection of LNCaP cells with human catalase. LNCaP cells were transiently transfected with the pzeoSV plasmid vector with the 1.6-kb human catalase cDNA cloned into the HindIII sites. Briefly, the pzeoSV-h-catalase (generous gift from Dr. J.D. Lambeth, Emory University School of Medicine, Atlanta, GA) was transfected into LNCaP cells using the SuperFect Transfection reagent from Qiagen GmbH (Germany) as per the manufacturer’s instructions. Overexpression of human catalase occurred at 24 hours after transfection and decreased to basal levels subsequently. Cells transfected with the empty vector pzeoSV were used as controls. Expression of human catalase protein was detected using human anti-catalase antibody (Calbiochem, Darmstadt, Germany), which recognizes the 60- to 65-kDa catalase subunit.

Figure 2. LY29-induced H2O2 production is independent of its PI3K-Akt inhibitory activity in LNCaP cells. A, ability of the varying doses of LY29 to Results dephosphorylate and deactivate Akt was shown by Western blot. LNCaP cells (1.5 Â 106-2 Â 106) were incubated with varying doses of LY29 (1-25 Amol/L) LY294002 triggers intracellular hydrogen peroxide produc- for 1 hour and cell lysates were obtained for Western blot analysis of the tion in tumor cells. Incubation of human prostate carcinoma Akt phosphorylation status in these cell extracts. In a similar experiment, LNCaP cells (3 Â 106) were incubated with varying doses of LY29 (1-25 Amol/L) (LNCaP) cells with LY29 (25 Amol/L) for 1 hour (concentration for 1 hour, after which the cell lysates were obtained and the Akt commonly used to show PI3K inhibitory activity) resulted in a immunoprecipitation kinase activity assay was done on these cell extracts as described in Materials and Methods. The total Akt and total GSK bands indicated significant increase in DCF fluorescence by flow cytometry (Fig. 1), equal loading of protein for analysis of the Akt phosphorylation status and which could be reversed by 1,000 units/mL catalase, thus indicating activity, respectively. B, LNCaP cells (0.8 Â 106) were treated with 1 Amol/L LY29 for 1 hour in the presence or absence of 1,000 units/mL catalase. Cells intracellular H2O2 production. More importantly, this effect was not were then loaded with DCFH-DA (5 Amol/L) for 15 minutes and the amount of exclusive to only LNCaP cells as evidenced by the similar effect of intracellular H2O2 generated was indicated by the shift in fluorescence as LY29 on HL60 (human leukemia), CEM (human leukemia), and T24 detected by flow cytometry.

Cancer Res 2005; 65: (14). July 15, 2005 6266 www.aacrjournals.org

Figure 3. Wortmannin inhibits PI3K but has no effect on intracellular H2O2 production in LNCaP cells. A, LNCaP cells (1.5 Â 106-2 Â 106) were incubated with varying concentrations of wortmannin (10-200 nmol/L) for 1 hour and cell lysates were obtained for Western blot analysis of the Akt phosphorylation status. B, LNCaP cells (0.8 Â 106) were treated with varying doses of wortmannin (10-200 nmol/L) for 1 hour. Cells were then loaded with DCFH-DA (5 Amol/L) for 15 minutes and intracellular H2O2 was assayed by flow cytometry.

affect Akt/PKB phosphorylation status. Furthermore, the fact (Fig. 5C). The involvement of caspases in this system was further that LY-induced H2O2 production was independent of PI3K activity evidenced by the use of the tetrapeptide inhibitors of caspase-2 was corroborated by the inability of wortmannin, another (VDVAD-fmk), caspase-3 (DEVD-fmk), and the pan-caspase commonly used PI3K inhibitor, to trigger increase in intracellular inhibitor (ZVAD-fmk). Interestingly, addition of ZVAD-fmk was H2O2 (Fig. 3B) at doses that effectively inhibited Akt phosphory- able to block LY-induced sensitization of LNCaP cells to drug- lation (Fig. 3A). induced apoptosis more effectively than the addition of caspase-2 LY303511, a LY294002 analogue, also produces hydrogen or caspase-3 inhibitor alone, as seen most significantly in an peroxide, inhibits proliferation, and sensitizes LNCaP cells to increase in cell viability over a longer period of study at 48 hours drug-induced apoptosis. To provide further evidence that the (Fig. 5D). The activation of caspase-8 or caspase-9 was minimal ability of LY29 to produce H2O2 was not linked to its PI3K and a similar sensitizing effect of LY30 on LNCaP cells was inhibitory effect, we investigated an analogue of LY29 (i.e., LY30), observed with other commonly used chemotherapeutic com- which has a high degree of structural similarity with LY29 (Fig. 4A) pounds, etoposide and daunorubicin (data not shown). As an but completely lacks the ability to inhibit the PI3K enzyme (refs. 56, additional marker of apoptosis, analysis of fragmented DNA was 57, 59, 60; Fig. 4C). Indeed, similar to the results obtained with also carried out. LNCaP cells incubated with 0.02 Amol/L LY29, incubation of LNCaP cells for 1 hour with varying vincristine were again preincubated in the presence or absence concentrations of LY30 (1-25 Amol/L) resulted in a significant of LY29 or LY30. Incubation with 0.02 Amol/L vincristine alone did increase in intracellular H2O2 (Fig. 4B). not alter the cell cycle profile; however, addition of LY29 or LY30 LNCaP cells were treated for 18 hours with a low dose to vincristine-treated cells increased the sub-G1 and G2-M (0.02 Amol/L) of vincristine that does not, by itself, significantly fractions of these cells, albeit to different extents at 18 and affect tumor cell viability (Fig. 5A). However, preincubation of 48 hours. Extent of DNA fragmentation is shown as a histogram cells with LY29 and LY30 significantly increased cell sensitivity where LY30 increased the amount of DNA fragmentation observed to 0.02 Amol/L vincristine as indicated by the reduction in the in vincristine-treated cells to levels similar to that observed with viable cell fraction (Fig. 5B). At the same time, incubation of cells LY29 at 48-hour incubation (Fig. 5E). Addition of caspase-2 with LY29 or LY30 overnight for 18 hours alone reduced cell inhibitor (VDVAD-fmk), caspase-3 inhibitor (DEVD-fmk), and the viability by f40% and 15%, respectively, suggesting perhaps an pan-caspase inhibitor (ZVAD-fmk) reduced the amount of DNA antiproliferative effect of these compounds on the doubling rate fragmentation as shown in Fig. 5E and cell cycle profiles (Fig. 5F of the cells as well (Fig. 5B). In addition, whereas treatment with and G). The pan-caspase inhibitor ZVAD-fmk reduced the extent 0.02 Amol/L vincristine alone did not induce intracellular caspase of DNA fragmentation effectively as depicted in the cell cycle activation, pretreatment of cells with LY29 or LY30 triggered profiles (Fig. 5F and G). Additionally, whereas a low nonlethal robust increases in caspase-2 and caspase-3 activities on dose of vincristine (0.02 Amol/L) did not produce any detectable subsequent exposure to the same concentration of vincristine H2O2, preincubation of tumor cells with 25 Amol/L LY30 produced www.aacrjournals.org 6267 Cancer Res 2005; 65: (14). July 15, 2005

a significant burst of intracellular H2O2 as seen with a rightward transiently transfected with a pzeoSV plasmid vector containing shift in DCF fluorescence (Fig. 5H). a 1.6-kb human catalase cDNA. Overexpression of catalase was LY303511 inhibits colony-forming ability of LNCaP cells confirmed by Western blot analysis and by the decrease in basal treated with vincristine. Not only could both LY29 and LY30 intracellular level of H2O2 by DCF fluorescence analysis (Fig. 7A). sensitize cells to vincristine-induced cell death but they were also Indeed, overexpression of catalase was able to significantly revert effective in reducing the colony-forming ability of LNCaP cells LY-induced increases in the sensitivity (Fig. 7B) and caspase when treated with vincristine (Fig. 6A). LNCaP cells were activity (Fig. 7C) of tumor cells on exposure to 0.02 Amol/L pretreated with LY29 or LY30 before treatment with 0.02 Amol/L vincristine. vincristine for 18 hours. As both LY29 and LY30 had an effect on cell proliferation overnight, the cell numbers were adjusted accordingly to take that into consideration. The cells were then Discussion replated onto 100 mm culture plates and incubated for 8 to 14 days LY294002 and LY303511 sensitize LNCaP cells to drug- to allow colonies to form. Cells pretreated with LY29 or LY30 induced apoptosis independently of the phosphoinositide (before vincristine exposure) showed a significant reduction in the 3-kinase pathway. Defects or deficiencies in apoptotic signaling number of colonies formed compared with cells treated with pathway(s) are a common feature of many tumors; therefore, ways vincristine alone (Fig. 6B). It should be pointed out that treatment to combat and overwhelm these blocks constitute a major effort of of tumor cells with LY30 alone for 18 hours did not seem to biologists and clinicians alike. One of the signaling pathways shown significantly reduce the colony-forming ability of cells; however, to be constitutively active in many tumor cells, thereby providing incubation for the same period with LY29 alone affected the them with a survival advantage over their normal counterparts, is colony-forming ability of LNCaP cells, suggesting an even stronger the PI3K-Akt survival circuitry. An essential component of this sensitizing effect of LY30, which unlike LY29 has no effect on the pathway is the active kinase Akt/PKB, which once phosphorylated PI3K-Akt pathway. can signal recruitment of many downstream targets involved in Transfection with human catalase inhibits caspase activity either cell survival signaling or inhibition of apoptosis. Therefore, and increase in apoptosis sensitivity induced by LY294002 and identification and development of compounds with the ability LY303511. Having shown that the LY compounds not only to inhibit upstream PI3K activation and subsequent Akt phosphor- increased sensitivity to drug-induced apoptosis but also stimulated ylation are an attractive strategy for enhancing tumor cell sensitivity a significant increase in intracellular H2O2 production, we next to apoptosis. To that end, LY29 has long been used as a somewhat asked if LY-induced increases in caspase-2 and caspase-3 activities selective inhibitor of PI3K-mediated phosphorylation of Akt/PKB and cell sensitivity to drug-induced apoptosis were a result of (55), and most of its biological activity relevant to apoptosis intracellular H2O2 production. To do so, LNCaP cells were sensitization has been attributed to its effect on the PI3K-Akt

6 Figure 4. LY30, a LY29 analogue, also produces H2O2 in LNCaP cells. A, structures of both LY29 and LY30. B, LNCaP cells (0.8 Â 10 ) were treated with varying doses (1-25 Amol/L) of LY30 for 1 hour in the presence or absence of 1,000 units/mL catalase. Cells were then loaded with DCFH-DA (5 Amol/L) for 15 minutes 6 6 and intracellular H2O2 was detected by flow cytometry. C, LNCaP cells (1.5 Â 10 -2 Â 10 ) were incubated with 25 Amol/L LY29 or LY30 for 1 hour and cell lysates were obtained for Western blot analysis of the Akt phosphorylation status.

Cancer Res 2005; 65: (14). July 15, 2005 6268 www.aacrjournals.org

Figure 5. LY30 can reduce cell proliferation and sensitize cells treated with low doses of vincristine to apoptosis via an increase in caspase activity. A, LNCaP cells (2.5 Â 105) were exposed to doses of 0.02 to 1 Amol/L vincristine for 18 hours, after which the crystal violet assay was done to establish a cell viability dose-response curve. B, LNCaP cells (2.5 Â 105) were exposed to 0.02 Amol/L vincristine (vin) for 18 hours in the presence or absence of a 1-hour pretreatment either with 25 Amol/L LY29 or LY30. The crystal violet assay was then done on these cells as described in Materials and Methods to determine cell viability after 18 hours. C, LNCaP cells (2.5 Â 105) were treated similarly for a time period of 6 to 18 hours and lysed to obtain whole-cell lysates. Caspase-2 and caspase-3 activities were assayed by using AFC- and AMC-conjugated substrates, respectively, on the whole-cell lysates. D, for experiments with the caspase inhibitors, cells were preincubated for 1 hour with 50 Amol/L of the respective caspase inhibitors before treatment with LY29/LY30 and vincristine for 18 and 48 hours, after which cell viability was assessed using the crystal violet assay. Columns, mean of at least three independent experiments; bars, SD. Viability of cells treated with vincristine and LY29/LY30 was calculated as a percentage against cells treated only with vincristine. E, LNCaP cells were incubated with 0.02 Amol/L vincristine in the presence or absence of 25 Amol/L LY30 together with various caspase inhibitors as described previously and stained with propidium iodide for cell cycle analysis by flow cytometry as described in Materials and Methods. The percentage of cells in the sub-G1 population was shown as an indication of the extent of DNA fragmentation occurring in the cells. Columns, mean of at least three independent experiments; bars, SD. survival network. Indeed, the realization that the PI3K-Akt axis is constitutively active Akt/PKB to drug-induced apoptosis. Exposure abnormally active in a variety of tumor cells has been the impetus of tumor cells to LY29 and more importantly to its inactive analogue behind preclinical trials involving inhibitory agents, such as LY29. LY30 (no effect on PI3K activation and Akt phosphorylation) Here, we show a novel mechanism by which LY compounds (LY29 resulted in a significant increase in intracellular H2O2 production and LY30) sensitize LNCaP prostate carcinoma cells expressing and sensitization of prostate carcinoma cells to nonlethal www.aacrjournals.org 6269 Cancer Res 2005; 65: (14). July 15, 2005

Figure 5 Continued. F, actual cell cycle profiles as analyzed by flow cytometry of LNCaP cells treated for 18 hours to show the effect of ZVAD on cells treated with 0.02 Amol/L vincristine and LY29/LY30. Representative of at least three independent experiments. G, actual cell cycle profiles as analyzed by flow cytometry of LNCaP cells treated for 48 hours to show the effect of ZVAD on cells treated with 0.02 Amol/L vincristine and LY29/LY30. Representative of at least three independent experiments. H, LNCaP cells (0.8 Â 106) were treated with 0.02 Amol/L vincristine for 1 hour in the presence or absence of a 1-hour preincubation with 25 Amol/L LY30. Cells were then loaded with DCFH-DA (5 Amol/L) for 15 minutes and intracellular H2O2 was detected by flow cytometry.

Cancer Res 2005; 65: (14). July 15, 2005 6270 www.aacrjournals.org

augmentation of drug-induced apoptosis sensitivity is corroborated by transient transfection with a vector containing human catalase, scavenger of intracellular H2O2; transient transfection with catalase virtually completely reverted the death-promoting effect of LY29 and LY30. It should be pointed out that apoptosis-sensitizing effect of LY compounds was not exclusive to vincristine but applied similarly to daunorubicin and etoposide (data not shown). Considering the strong inhibitory effect of LY29 on PI3K activation, one could argue that it is not surprising that preincubation of tumor cells resulted in an increase in sensitivity to apoptosis. However, the data reported here strongly support a novel, PI3K-independent mechanism for sensitizing tumor cells to drug-induced apoptosis through intracellular production of H2O2. Corroborating this is the ability of low concentrations of LY29 (1-5 Amol/L) that do not affect PI3K activity but still induce intracellular H2O2 production and sensitize tumor cells to drug-induced apoptosis (data not shown), thereby excluding the involvement of the PI3K pathway in this system. This is further supported by a strong sensitizing effect of LY30, which lacks PI3K inhibitory activity but, similar to LY29, triggers a significant increase in intracellular H2O2 production. Finally, the fact that wortmannin, another commonly used inhibitor of PI3K, had no effect on intracellular H2O2 production provides evidence that LY-induced H2O2 production is not a function of PI3K inhibition. In fact, incubation of LNCaP cells with wortmannin actually reduced the levels of H2O2 in the cell. Although the data at this point in time are not sufficient to explain this observation, it could be attributed to the ability of wortmannin to nonspecifically inhibit MAPK (47) and phospholipase D. Figure 6. LY30 inhibits the colony-forming ability of cells treated with vincristine. Regarding the relationship between inhibition of PI3K and A, LNCaP cells (4,000-8,000) were exposed to 0.02 Amol/L vincristine for intracellular H2O2 production, a plausible scenario, considering 18 hours in the presence or absence of a 1-hour pretreatment with either the previously shown ability of LY29 to bind to the ATP-binding site 25 Amol/L LY29 or LY30. The cells were then seeded in 100 mm Petri dishes and allowed to form colonies over 8 to 14 days, after which they were stained with of other protein kinases like the DNA-dependent protein kinase in crystal violet as described in Materials and Methods to show colony formation. the nucleus (61), could be that LY compounds affect transcription of Representative of nine independent experiments, each done in triplicate. B, extent of colony formation for three independent experiments, each done intracellular redox control genes, like catalase and superoxide in triplicate. dismutase. However, the burst of intracellular H2O2 generated by the LY compounds in LNCaP cells was detected as early as 15 minutes concentration of the chemotherapeutic agent vincristine via after incubation (data not shown), thereby rendering this possibility caspase-2 and caspase-3 activation, increased DNA fragmentation, highly unlikely. and a strong inhibitory effect on the colony-forming ability of tumor Intracellular generation of hydrogen peroxide: a novel cells. The reduced viability of cells treated overnight for 18 hours mechanism of action of LY compounds. H2O2 has long been with only LY29 or LY30 (60% and 85% viability for LY29 and LY30, shown to play vastly different roles in the cell, where subtle changes respectively) could be an indication of the inhibitory effect of the LY in the delicate intracellular redox status could greatly affect cell compounds on cell proliferation as the doubling time of LNCaP physiology. For example, depending on its concentration, H2O2 can prostate carcinoma cells is f24 hours. This could be extrapolated induce cell proliferation, mediate drug-induced apoptosis in many to mean that proliferation of all of the LY29-treated cells was models (particularly in tumor cells), or at overwhelming levels affected and proliferation of f15% of LY30 treated cells was bring about necrotic cell death. Indeed, there is a fair amount of affected. LY30 also significantly enhanced the extent of DNA controversy with respect to the relationship between PI3K fragmentation in vincristine-treated cells, especially over a longer signaling and H2O2 production, with data supporting both an period of 48 hours. The ability of the tetrapeptide caspase inhibitors upstream and a downstream role of H2O2 in PI3K signaling (62). to block the observed DNA fragmentation showed again that this However, in tumor cells, intracellular generation of H2O2, was a caspase-dependent pathway independent of the PI3K-Akt specifically in response to drug exposure, is independent of PI3K. axis. The pan-caspase inhibitor ZVAD-fmk almost completely Data presented here not only corroborate those findings but also attenuated the process of DNA fragmentation observed in these underscore the inherent ability of the LY compounds to trigger a cells, much more effectively than the presence of either caspase-2 or significant increase in intracellular H2O2 and provide a novel caspase-3 inhibitor alone, suggesting that it is probably the mechanism for their sensitizing effect on tumor cells. More synergistic inhibition of caspase-2 and caspase-3 by ZVAD-fmk significantly, although previous studies have shown that H2O2 is a that protected the cells from DNA fragmentation. Interestingly, mediator of vincristine-induced apoptotic cell death, we show here there was no significant increase in either caspase-8 or caspase-9 that incubation of cells with a nonlethal dose of vincristine did not activation in this system, suggesting a direct effect of H2O2 on generate H2O2 in the cell, an observation commensurate with the caspase-2 and/or caspase-3 activation resulting in death execution. fact that there was no cell death with such low doses of vincristine. Furthermore, the critical involvement of H2O2 in LY-induced It was the preincubation with LY compounds that significantly www.aacrjournals.org 6271 Cancer Res 2005; 65: (14). July 15, 2005

enhanced the levels of H2O2 in the cell, which was then responsible Although the current data seem to exclude the mitochondria as for tipping the cell toward eventual apoptosis. a possible source of H2O2 accumulation by these LY compounds, it The ability of the LY compounds to generate intracellular H2O2 does not exclude the possibility of the activation of other oxidase- could perhaps be further understood by a closer look at their like complexes in the cell, such as those found in the endoplasmic structures. The structures of the LY compounds were based on reticulum or activation of the copper/zinc superoxide dismutases quercetin, a naturally occurring bioflavonoid that was shown in the cytosol itself. As these compounds possess structural previously to inhibit PI3K. However, quercetin was also shown to be similarity to quercetin, a bioflavonoid, it is perhaps not surprising able to inhibit other kinases like PI4K and several tyrosine and that they generate H2O2 in the cell, given the controversial nature serine/threonine kinases. LY29 was synthesized using quercetin as a of flavonoids to act as both an antioxidant with free radical lead compound where the dihydroxyphenyl group at the 2-position scavenging properties and a pro-oxidant. of the chromone ring in quercetin was replaced with a morpholine Interestingly, recent reports have shown biological effects of LY moiety (55) as seen in LY29, thus conferring on LY29, its specificity compounds that are independent of its PI3K-inhibiting activity in a in the inhibition of PI3K as it was then subsequently unable to variety of biological systems (56, 57). For instance, both LY29 and inhibit PI4K, epidermal growth factor receptor tyrosine kinase, or LY30 have been shown to block K(V) currents via a direct non- other ATP-requiring kinases and ATPases. A simple replacement of PI3K-dependent mechanism in rat pancreatic h cells (57) and to oxygen with nitrogen in the 2-position of the morpholine ring gave inhibit MCP-1 expression in human umbilical vein endothelial cells the compound LY30, which displayed a dramatic reduction in the (56). Considering our findings, one is tempted to speculate that ability of this compound to inhibit PI3K. LY29 inhibits PI3K to <1% these PI3K-independent activities of LY compounds might be of its original activity at a dose of 100 Amol/L; however, LY30 at the linked to the intracellular generation of H2O2. same dose allows PI3K to retain almost all of its original activity Intracellular hydrogen peroxide production: a permissive (55). Although the H2O2 generated by both LY compounds is environment for sensitization of cells to drug-induced functionally relevant, it is probably the synergistic effect of the apoptosis. We have highlighted previously the critical role of ability of LY29 to inhibit PI3K effectively that results in its greater intracellular H2O2 in sensitization of tumor cells to apoptotic potency compared with LY30 when it is used. stimuli by creating an intracellular milieu permissive for caspase

Figure 7. Transfection of human catalase into cells reduces the increase in caspase-2 and caspase-3 activity in cells treated with vincristine and LY29 or LY30. A, LNCaP cells were transiently transfected with human catalase and the empty zeocin vector. Overexpression of the catalase was detected by Western blot analysis as described in Materials and Methods. Catalase overexpression resulted in decreased basal levels of H2O2 in LNCaP cells assayed by DCFH- DA loading and flow cytometry. B and C, at 24 hours post-transfection, LNCaP cells (0.8 Â 106) were treated with LY30/29 and 0.02 Amol/L vincristine as described previously. Activation of caspases and proliferation rate of these cells were then assayed 18 hours later as described previously to compare the effects of catalase- transfected and empty vector–transfected cells.

Cancer Res 2005; 65: (14). July 15, 2005 6272 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2005 American Association for Cancer Research. Apoptosis-Sensitizing Activity of LY Compounds activation and death execution. In addition, we showed that findings, the structures of the LY compounds could therefore be priming of tumor cells with subapoptotic concentrations of H2O2 used as a model to develop chemotherapeutic compounds that significantly enhanced the sensitivity of tumor cells to drug- could trigger intracellular H2O2, which in turn would either induced apoptosis (63), and alternatively, blocking intracellular sensitize tumor cells to drug-induced apoptosis or directly induce H2O2 production inhibited death execution (38, 44, 46). Along the apoptosis itself. same lines, intracellular production of H2O2 seems to be a common denominator in drug-induced apoptosis of tumor cells. Judging from these in vitro data, it is logical that chemothera- Acknowledgments peutic efficacy could be significantly improved in an environment Received 1/17/2005; revised 4/13/2005; accepted 5/2/2005. Grant support: National Medical Research Council Singapore grants NMRC/0539/ made conducive for death execution by the presence of H2O2. 2001 and NMRC/0738/2003. Therefore, identification of compounds that trigger a significant The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance increase in intracellular H2O2 and their use in conjunction with with 18 U.S.C. Section 1734 solely to indicate this fact. chemotherapy agents could be an attractive strategy to enhance We thank Drs. Marie-Veronique Clement, Andrea Holme, and Alan Kumar for the sensitivity of tumor cells to drug therapy. In the light of these useful discussions.

References 16. Holbro T, Beerli RR, Maurer F, Koziczak M, Barbas CF hydrogen peroxide in UV-induced signal transduction. III, Hynes NE. The ErbB2/ErbB3 heterodimer functions Mol Cell Biochem 2002;234–235:81–90. 1. Green DR. Apoptotic pathways: paper wraps stone as an oncogenic unit: ErbB2 requires ErbB3 to drive 31. Dong-Yun S, Yu-Ru D, Shan-Lin L, Ya-Dong Z, Lian W. blunts scissors. Cell 2000;102:1–4. breast tumor cell proliferation. Proc Natl Acad Sci U S A Redox stress regulates cell proliferation and apoptosis of 2. Altieri DC. Survivin, versatile modulation of cell 2003;100:8933–8. human hepatoma through Akt protein phosphorylation. division and apoptosis in cancer. Oncogene 2003;22: 17. Shelton JG, Steelman LS, Lee JT, et al. Effects of the FEBS Lett 2003;542:60–4. 8581–9. RAF/MEK/ERK and PI3K/AKT signal transduction 32. Esposito F, Chirico G, Montesano Gesualdi N, et al. 3. Korsmeyer SJ. BCL-2 gene family and the regulation of pathways on the abrogation of cytokine-dependence Protein kinase B activation by reactive oxygen species is programmed cell death. Cancer Res 1999;59:1693–700s. and prevention of apoptosis in hematopoietic cells. independent of tyrosine kinase receptor phosphoryla- 4. Brazil DP, Park J, Hemmings BA. PKB binding proteins. Oncogene 2003;22:2478–92. tion and requires SRC activity. J Biol Chem 2003;278: Getting in on the Akt. Cell 2002;111:293–303. 18. Chang F, Lee JT, Navolanic PM, et al. Involvement of 20828–34. 5. Stahl M, Dijkers PF, Kops GJ, et al. The forkhead PI3K/Akt pathway in cell cycle progression, apoptosis, 33. Thomas SR, Chen K, Keaney JF Jr. Hydrogen peroxide transcription factor FoxO regulates transcription of and neoplastic transformation: a target for cancer activates endothelial nitric-oxide synthase through Kip1 p27 and Bim in response to IL-2. J Immunol 2002; chemotherapy. Leukemia 2003;17:590–603. coordinated phosphorylation and dephosphorylation 168:5024–31. 19. Santos SC, Lacronique V, Bouchaert I, et al. Consti- via a phosphoinositide 3-kinase-dependent signaling 6. Wanzel M, Kleine-Kohlbrecher D, Herold S, et al. Akt tutively active STAT5 variants induce growth and pathway. J Biol Chem 2002;277:6017–24. and 14-3-3eta regulate Miz1 to control cell-cycle arrest survival of hematopoietic cells through a PI 3-kinase/ 34. Bae YS, Sung JY, Kim OS, et al. Platelet-derived after DNA damage. Nat Cell Biol 2005;7:30–41. Akt dependent pathway. Oncogene 2001;20:2080–90. growth factor-induced H(2)O(2) production requires the 7. Asano T, Yao Y, Zhu J, Li D, Abbruzzese JL, Reddy SA. 20. Sakoda H, Gotoh Y, Katagiri H, et al. Differing roles of activation of phosphatidylinositol 3-kinase. J Biol Chem The PI 3-kinase/Akt signaling pathway is activated due Akt and serum- and glucocorticoid-regulated kinase in 2000;275:10527–31. to aberrant Pten expression and targets transcription glucose metabolism, DNA synthesis, and oncogenic 35. Chen Q, Powell DW, Rane MJ, et al. Akt phosphor- factors NF-nB and c-Myc in pancreatic cancer cells. activity. J Biol Chem 2003;278:25802–7. ylates p47phox and mediates respiratory burst activity Oncogene 2004;23:8571–80. 21. Dufresne SD, Bjorbaek C, El-Haschimi K, et al. in human neutrophils. J Immunol 2003;170:5302–8. 8. Guo S, Sonenshein GE. Forkhead box transcription Altered extracellular signal-regulated kinase signaling 36. Goldshmit Y, Erlich S, Pinkas-Kramarski R. Neuregu- factor FOXO3a regulates estrogen receptor a expression and glycogen metabolism in skeletal muscle from p90 lin rescues PC12-ErbB4 cells from cell death induced by and is repressed by the Her-2/neu/phosphatidylinositol ribosomal S6 kinase 2 knockout mice. Mol Cell Biol H(2)O(2). Regulation of reactive oxygen species levels by 3-kinase/Akt signaling pathway. Mol Cell Biol 2004;24: 2001;21:81–7. phosphatidylinositol 3-kinase. J Biol Chem 2001;276: 8681–90. 22. Armstrong JL, Bonavaud SM, Toole BJ, Yeaman SJ. 46379–85. 9. Galetic I, Maira SM, Andjelkovic M, Hemmings BA. Regulation of glycogen synthesis by amino acids in 37. Pervaiz S, Clement MV. A permissive apoptotic Negative regulation of ERK and Elk by protein kinase B cultured human muscle cells. J Biol Chem 2001;276:952–6. environment: function of a decrease in intracellular modulates c-Fos transcription. J Biol Chem 2003;278: 23. Agarwal A, Das K, Lerner N, et al. The AKT/InB superoxide anion and cytosolic acidification. Biochem 4416–23. kinase pathway promotes angiogenic/metastatic gene Biophys Res Commun 2002;290:1145–50. 10. Pekarsky Y, Hallas C, Palamarchuk A, et al. Akt expression in colorectal cancer by activating nuclear 38. Ahmad KA, Iskandar KB, Hirpara JL, Clement MV, phosphorylates and regulates the orphan nuclear factor-nB and h-catenin. Oncogene 2005;24:1021–31. Pervaiz S. Hydrogen peroxide-mediated cytosolic acid- receptor Nur77. Proc Natl Acad Sci U S A 2001;98: 24. Kim HS, Skurk C, Thomas SR, et al. Regulation of ification is a signal for mitochondrial translocation of 3690–4. angiogenesis by glycogen synthase kinase-3h. J Biol Bax during drug-induced apoptosis of tumor cells. 11. Gao N, Flynn DC, Zhang Z, et al. G1 cell cycle Chem 2002;277:41888–96. Cancer Res 2004;64:7867–78. progression and the expression of G1 cyclins are 25. Kandel ES, Skeen J, Majewski N, et al. Activation of 39. Clement MV, Ponton A, Pervaiz S. Apoptosis induced regulated by PI3K/AKT/mTOR/p70S6K1 signaling in Akt/protein kinase B overcomes a G(2)/M cell cycle by hydrogen peroxide is mediated by decreased human ovarian cancer cells. Am J Physiol Cell Physiol checkpoint induced by DNA damage. Mol Cell Biol 2002; superoxide anion concentration and reduction of 2004;287:C281–91. 22:7831–41. intracellular milieu. FEBS Lett 1998;440:13–8. 12. DangiS,ChaH,ShapiroP.Requirementfor 26. Tran H, Brunet A, Grenier JM, et al. DNA repair 40. Hirpara JL, Clement MV, Pervaiz S. Intracellular phosphatidylinositol-3 kinase activity during progres- pathway stimulated by the forkhead transcription factor acidification triggered by mitochondrial-derived hy- sion through S-phase and entry into mitosis. Cell Signal FOXO3a through the Gadd45 protein. Science 2002;296: drogen peroxide is an effector mechanism for drug- 2003;15:667–75. 530–4. induced apoptosis in tumor cells. J Biol Chem 2001; 13. Alvarez B, Garrido E, Garcia-Sanz JA, Carrera AC. 27. Yu C, Rahmani M, Dai Y, et al. The lethal effects of 276:514–21. Phosphoinositide 3-kinase activation regulates cell pharmacological cyclin-dependent kinase inhibitors in 41. Pervaiz S, Cao J, Chao OS, Chin YY, Clement MV. division time by coordinated control of cell mass human leukemia cells proceed through a phosphatidy- Activation of the RacGTPase inhibits apoptosis in and cell cycle progression rate. J Biol Chem 2003;278: linositol 3-kinase/Akt-dependent process. Cancer Res human tumor cells. Oncogene 2001;20:6263–8. 26466–73. 2003;63:1822–33. 42. Clement MV, Hirpara JL, Pervaiz S. Decrease in 14. Shin I, Yakes FM, Rojo F, et al. PKB/Akt mediates cell- 28. Nam SY, Jung GA, Hur GC, et al. Upregulation of intracellular superoxide sensitizes Bcl-2-overexpressing cycle progression by phosphorylation of p27(Kip1) at FLIP(S) by Akt, a possible inhibition mechanism of tumor cells to receptor and drug-induced apoptosis threonine 157 and modulation of its cellular localiza- TRAIL-induced apoptosis in human gastric cancers. independent of the mitochondria. Cell Death Differ tion. Nat Med 2002;8:1145–52. Cancer Sci 2003;94:1066–73. 2003;10:1273–85. 15. Choudhury GG. Akt serine threonine kinase regu- 29. Mabuchi S, Ohmichi M, Kimura A, et al. Inhibition of 43. Pervaiz S, Ramalingam JK, Hirpara JL, Clement MV. lates platelet-derived growth factor-induced DNA phosphorylation of BAD and Raf-1 by Akt sensitizes Superoxide anion inhibits drug-induced tumor cell synthesis in glomerular mesangial cells: regulation of human ovarian cancer cells to paclitaxel. J Biol Chem death. FEBS Lett 1999;459:343–8. c-fos AND p27(kip1) gene expression. J Biol Chem 2001; 2002;277:33490–500. 44. Ahmad KA, Clement MV, Hanif IM, Pervaiz S. 276:35636–43. 30. Ding M, Li J, Leonard SS, et al. Differential role of Resveratrol inhibits drug-induced apoptosis in human

www.aacrjournals.org 6273 Cancer Res 2005; 65: (14). July 15, 2005

leukemia cells by creating an intracellular milieu 51. O’Gorman DM, McKenna SL, McGahon AJ, Knox potently blocks K(V) currents via a direct mechanism. nonpermissive for death execution. Cancer Res 2004; KA, Cotter TG. Sensitisation of HL60 human leukae- FASEB J 2003;17:720–2. 64:1452–9. mic cells to cytotoxic drug-induced apoptosis by 58. Pasapera Limon AM, Herrera-Munoz J, Gutierrez- 45. Clement MV, Stamenkovic I. Superoxide anion is a inhibition of PI3-kinase survival signals. Leukemia 2000; Sagal R, Ulloa-Aguirre A. The phosphatidylinositol 3- natural inhibitor of FAS-mediated cell death. EMBO J 14:602–11. kinase inhibitor LY294002 binds the estrogen receptor 1996;15:216–25. 52. Ohigashi T, Mizuno R, Nakashima J, Marumo K, and inhibits 17h-estradiol-induced transcriptional 46. Ahmad KA, Clement MV, Pervaiz S. Pro-oxidant Murai M. Inhibition of Wnt signaling downregu- activity of an estrogen sensitive reporter gene. Mol Cell activity of low doses of resveratrol inhibits hydrogen lates Akt activity and induces chemosensitivity in Endocrinol 2003;200:199–202. peroxide-induced apoptosis. Ann N Y Acad Sci 2003; PTEN-mutated prostate cancer cells. Prostate 2005;62: 59. Ding J, Vlahos CJ, Liu R, Brown RF, Badwey JA. 1010:365–73. 61–8. Antagonists of phosphatidylinositol 3-kinase block 47. Ferby IM, Waga I, Hoshino M, Kume K, Shimizu 53. Thakkar H, Chen X, Tyan F, et al. Pro-survival activation of several novel protein kinases in neutro- T. Wortmannin inhibits mitogen-activated protein function of Akt/protein kinase B in prostate cancer cells. phils. J Biol Chem 1995;270:11684–91. kinase activation by platelet-activating factor Relationship with TRAIL resistance. J Biol Chem 2001; 60. Reddy MA, Prasadarao NV, Wass CA, Kim KS. through a mechanism independent of p85/p110-type 276:38361–9. Phosphatidylinositol 3-kinase activation and interaction phosphatidylinositol 3-kinase. J Biol Chem 1996;271: 54. Pene F, Claessens YE, Muller O, et al. Role of with focal adhesion kinase in Escherichia coli K1 11684–8. the phosphatidylinositol 3-kinase/Akt and mTOR/ invasion of human brain microvascular endothelial 48. Bortul R, Tazzari PL, Cappellini A, et al. Constitu- P70S6-kinase pathways in the proliferation and cells. J Biol Chem 2000;275:36769–74. tively active Akt1 protects HL60 leukemia cells from apoptosis in multiple myeloma. Oncogene 2002;21: 61. Izzard RA, Jackson SP, Smith GC. Competitive and TRAIL-induced apoptosis through a mechanism involv- 6587–97. noncompetitive inhibition of the DNA-dependent pro- ing NF-nB activation and cFLIP(L) up-regulation. 55. Vlahos CJ, Matter WF, Hui KY, Brown RF. A specific tein kinase. Cancer Res 1999;59:2581–6. Leukemia 2003;17:379–89. inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpho- 62. Jung DK, Bae GU, Kim YK, et al. Hydrogen peroxide 49. Clark AS, West K, Streicher S, Dennis PA. Constitutive linyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002). J Biol mediates arsenite activation of p70(s6k) and extracel- and inducible Akt activity promotes resistance to Chem 1994;269:5241–8. lular signal-regulated kinase. Exp Cell Res 2003;290: chemotherapy, trastuzumab, or tamoxifen in breast 56. Choi EK, Park HJ, Ma JS, et al. LY294002 inhibits 144–54. cancer cells. Mol Cancer Ther 2002;1:707–17. monocyte chemoattractant protein-1 expression 63. Pervaiz S, Seyed MA, Hirpara JL, Clement MV, 50. Nicholson KM, Quinn DM, Kellett GL, Warr JR. through a phosphatidylinositol 3-kinase-independent Loh KW. Purified photoproducts of merocyanine 540 LY294002, an inhibitor of phosphatidylinositol-3-kinase, mechanism. FEBS Lett 2004;559:141–4. trigger cytochrome c release and caspase 8-dependent causes preferential induction of apoptosis in human 57. El-Kholy W, Macdonald PE, Lin JH, et al. The apoptosis in human leukemia and melanoma cells. multidrug resistant cells. Cancer Lett 2003;190:31–6. phosphatidylinositol 3-kinase inhibitor LY294002 Blood 1999;93:4096–108.

Cancer Res 2005; 65: (14). July 15, 2005 6274 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2005 American Association for Cancer Research. LY294002 and LY303511 Sensitize Tumor Cells to Drug-Induced Apoptosis via Intracellular Hydrogen Peroxide Production Independent of the Phosphoinositide 3-Kinase-Akt Pathway

Tze Wei Poh and Shazib Pervaiz

Cancer Res 2005;65:6264-6274.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/65/14/6264

Cited articles This article cites 62 articles, 30 of which you can access for free at: http://cancerres.aacrjournals.org/content/65/14/6264.full#ref-list-1

Citing articles This article has been cited by 10 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/65/14/6264.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/65/14/6264. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.