| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Oral and Pharyngeal Cancer Branch, National Institute of Craniofacial and Dental Research [V. P., T. L., T. S., J. S. G., A. M. S.], and Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, National Cancer Institute [E. A. S.], NIH, Bethesda, Maryland 20892
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.6–8.9 µM. Cell cycle arrest at the G1-S and G2-M boundaries was observed in HN12, HN30, and HaCaT cells independent of p53 function, and this effect was preceded by loss in cdc2 and cyclin-dependent kinase (cdk) 2 activity. Analysis of cdk complexes in vitro demonstrated that perifosine, up to 20 µM, did not directly interfere with these enzymes. However, aphidicolin-synchronized HN12 cells released in the presence of perifosine (10 µM) demonstrated increased expression of total p21WAF1 and increased association of p21WAF1 with cyclin-cdk complexes resulting in reduced cdc2 activity. HCT116 isogenic cell lines were used to assess the role of p21WAF1 induction by perifosine. This compound (20 µM) induced both G1-S and G2-M cell cycle arrest, together with p21WAF1 expression in both p53 wild-type and p53-/- clones. By contrast, p21-/- variants demonstrated no p21WAF1 induction or cell cycle arrest. Similar results were obtained with other ALK congeners (miltefosine and edelfosine). These data, therefore, indicate that perifosine blocks cell cycle progression of head and neck squamous carcinoma cells at G1-S and G2-M by inducing p21WAF1, irrespective of p53 function, and may be exploited clinically because the majority of human malignancies harbor p53 mutations. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Refs. 7, 8, 9
).
|
Perifosine has been characterized for its antiproliferative effect in a number of tumor cell lines (8) . For instance, those derived from breast, colon, prostate, and larynx demonstrated reasonable sensitivity (8) . More remarkably, cell lines derived from certain HNSCCs, for instance KB and Hep-2 (larynx) and SAS (tongue), were found to be very sensitive to the effects of perifosine (8) . We have reported previously on a panel of HNSCC cell lines derived from primary and secondary cancer lesions of different clinical stages (T2–T4), which have been extensively characterized for alterations of the major tumor suppressor genes (p53 and p16INK4A) and components of the cell cycle (19, 20, 21) . Therefore, by using this model system of HNSCC, we evaluated in detail the antiproliferative action of perifosine.
In this study, we report that in vitro, perifosine is growth inhibitory in a representative panel of HNSCC cells, resulting in the blockade of cells in G1-S and G2-M. This effect is caused by modulation of cdk activity by the up-regulation and increased association of p21WAF1 with cdk/cyclin complexes, occurring in a p53-independent manner. We provide evidence that p21WAF1 up-regulation is required for perifosine-induced cell cycle arrest, because p21WAF1-/- cells were found to be insensitive to this effect of the drug.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Drugs.
ASTA Medica AG (Germany) and Aventis Pharmaceuticals (Bridgewater, NJ) provided perifosine and flavopiridol, respectively, to the Development Therapeutics Program, National Cancer Institute. For in vitro studies, perifosine was reconstituted in PBS at a stock concentration of 100 mM and further diluted in PBS to the working concentration (0.1–30 µM) for experimental procedures.
Assessment of Thymidine Incorporation in Perifosine-treated HNSCC Cells.
Cell proliferation studies by measuring the uptake of [3H]thymidine was performed as described (21)
. Briefly, HNSCC and HaCaT cells (1–2 x 104/well) were grown overnight in 24-well plates and exposed to either perifosine (0.1–30 µM) or PBS (control). After treatment (24–48 h), cells were pulsed with [3H]thymidine (1 µCi/well) for 4–6 h, fixed (5% trichloroacetic acid), and solubilized (0.5 M NaOH) before scintillation counting. Experiments were performed in triplicates.
Cell Cycle Analysis.
Analysis of cellular DNA content by flow cytometry was performed as described (21)
. Briefly, perifosine (0.1–30 µM) and control-treated cells (HNSCC, HaCaT, HCT116, and isogenic variants) were harvested after 24 h, washed briefly in ice-cold PBS, and fixed in 70% ethanol. DNA was stained by incubating the cells in PBS containing propidium iodide (50 µg/ml) and RNase A (1 mg/ml) for 30 min at 37°C. Fluorescence was measured and analyzed using FACSCaliber (Becton Dickinson Immunocytometry Systems, San Jose, CA) and ModFit (Verity Software, Topsham, ME), respectively. For time-dependent analysis, cells treated with perifosine (10 µM) were harvested at the indicated time (0–24 h) and processed as described.
Analysis of cdk Activity in Perifosine-treated HNSCC Cells.
Assessment of in vitro cdc2 and cdk2 activity were as reported (21
, 22)
. Briefly, exponentially growing HNSCC and HaCaT cells were exposed to perifosine (10 µM) or PBS for 24 h and subsequently lysed as described above, and 500 µg of total cellular protein were used to immunoprecipitate cdc2 and cdk2 for 1 h at 4°C with appropriate antibodies (sc-954 and sc-163, respectively; Santa Cruz Biotechnology, Inc., Santa Cruz Biotechnology, CA). Gammabind G Sepharose (Pharmacia, Piscataway, NJ) was used to capture the immune complexes, and after several washes, kinase reactions were carried out in kinase assay buffer [50 mM HEPES (pH 7.5), 10 mM MgCl2, 1 mM DTT, and 25 µM ATP] containing [
-32P]ATP (3000 Ci/mmol; NEN, Boston, MA) and 0.2 mg/ml histone H1 (Boehringer Mannheim Biochemicals, Indianapolis, IN). Reactions were incubated at 37°C for 30 min and terminated by the addition of SDS-gel loading buffer. The resolved and dried gels were subjected to autoradiography and phosphorimaging (Molecular Dynamics, Sunnyvale, CA). To assess cdc2 activity in detail, aphidicolin-synchronized (5 µg/ml for 12–14 h) HN12 cells were released in the presence of perifosine (10 µM) or PBS and harvested at the indicated times (0–12 h), and kinase reactions were carried out as described above. Additional assessment of the in vivo activity of G1 cdks was carried out by Western blot analysis on lysates from cells treated as described above, using phospho-specific polyclonal antisera to pRb recognizing threonine 356 and serine 780, sites that are phosphorylated by cdk2 and cdk4/6, respectively (22)
.
Assessment of Perifosine on cdk Activity in Vitro.
Exponentially growing cells (HN12, HN30, and HaCaT) were lysed as described above, and 500 µg of total cellular protein were used to immunoprecipitate active cdc2 and cdk2 complexes. After capturing with gammabind G Sepharose and subsequent washes, the active immune complexes were assessed for activity in the presence of increasing concentrations of perifosine (0.1–30 µM) or flavopiridol (300 nM) in the kinase assay buffer containing [-32P]ATP (3000 Ci/mmol) and 0.2 mg/ml histone H1, 25 µM ATP. Reactions were incubated at 37°C for 30 min and terminated by the addition of SDS-gel loading buffer, resolved in SDS-PAGE, and dried gels were subjected to autoradiography and phosphorimaging.
Immunoblot Analysis of Perifosine-treated HNSCC Cells.
Western blot analysis of lysates (HNSCC, HaCaT, HCT116, and isogenic variants) or cdc2, cdk2, and p21WAF1 immunoprecipitates (HNSCC and HaCaT), prepared from cells treated as described, were carried out using appropriate antibodies to the indicated proteins (cdc2 and cdk2: as above, cyclin B1, 1:1000, sc-752; Santa Cruz Biotechnology, Inc.; cyclin A, 1:500, Novacastra, Newcastle, United Kingdom; p21WAF1 and p27KIP1, 1:750, 6B6 and G173–534, respectively, BD Transduction Laboratories, San Diego, CA), and reactions were detected by horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence.
Analysis of Mitosis.
For assessment of the mitotic index, cells treated with perifosine (10 µM) or vehicle control were harvested, washed in ice-cold PBS, resuspended in 0.5x PBS (10 min), and fixed in 0.5 ml of ethanol:acetic acid (3:1) for 10 min. Cell suspension was dropped onto glass slides, air dried, and mounted in medium containing 4',6-diamidino-2-phenylindole before analysis under fluorescence light. For each spread, 500 cells were analyzed for mitosis, and the percent number of cells in mitosis in each treatment was compared with the percent observed in vehicle-treated cells. Similarly, cells treated for 12–14 h with nocodazole (0.5 µg/ml) were used as positive control.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, exposure to perifosine (0.1–30 µM) for 24 h resulted in a dose-dependent inhibition of [3H]thymidine uptake in all cell lines tested. The IC50s (50% inhibitory concentration) for growth were between 0.6 and 8.9 µM, reaching IC80s of 10 µM. On the basis of this observation, we chose to use two malignant HNSCC phenotypes, HN12 and HN30, and immortalized HaCaT cells for subsequent experiments. The distinct biological features reported previously, for instance the insensitivity of HN30 to -irradiation and to certain anticancer agents (21)
and the remarkable tumorigenicity in vivo of HN12 cells (21)
, suggest that they may represent useful models for HNSCC.
Effect of Perifosine on Cell Cycle Progression at G1/S and G2/M in HNSCC Cells.
Although the ALKs are known to be antiproliferative, the exact mechanism by which they induce this effect is still unknown. We sought to investigate whether perifosine may be targeting the cell cycle regulatory mechanisms of HNSCC cells by initially determining the DNA content of HNSCC cells exposed to perifosine using FACS analysis. Dose-response analysis was performed where HNSCC cells were exposed to increasing concentrations of perifosine (0.1–30 µM) for 24 h and subsequently processed for cell cycle analysis. As illustrated in Fig. 2A
, in HN12 cells a minimal sensitivity to 0.1–1 µM perifosine was observed when compared with control. Surprisingly, a consistent increase in G2-M along with a concomitant decline in the S-phase fraction was notable in all cell types tested (HN12, HN30, and HaCaT) at concentrations 
3 µM perifosine, with maximal effects observed with 10 µM. On the basis of these data, we chose 10 µM, unless specified otherwise, as an effective concentration (IC80) for further characterization of this anticancer agent. For time-dependent studies, perifosine (10 µM)-treated HN12 cells were harvested at the indicated times (0–24 h) and processed for cell cycle analysis. As shown in Fig. 2B
, a clear effect on cell cycle progression was observed as early as 4 h after treatment, with an increased accumulation of cells at G2-M. With prolonged exposure (4–12 h), a progressive accumulation of cells in G2-M (69% compared with baseline of 39%) is observed with a decline in those cells in S-phase (10% compared with a baseline of 19%). This effect became even more apparent at longer time points (12–24 h). The data indicate that perifosine (10 µM) is able to cause an accumulation of cells at G1 and G2-M phases of the cell cycle. Similar observations were made for HaCaT and HN30 cells (data not shown).
|
(right panel), most (>50%) HaCaT and HN12 cells treated with nocodazole (0.5 µg/ml) demonstrated loss of nuclear membrane and chromosome condensation, features that are typical of cell cycle arrest in metaphase when caused by microtubule disruptors. In contrast, most (>90%) HNSCC cells exposed to either vehicle (Fig. 3
, left panel) or 10 µM perifosine (Fig. 3
, middle panel) showed intact nuclear membrane without an increase in mitotic figures. Similar observations were seen with HN30 cells (data not shown). Thus, increased G2-M DNA content provoked by perifosine in HNSCC cells results from arrest of cells in G2 (interphase) with loss of entry into mitosis.
|
, treatment of HN12 cells with perifosine resulted in a dose-dependent reduction in the activity of both cdc2 and cdk2. To ensure that these observations were not attributable to loading differences, immunoprecipitates (5 µl) were assessed by western blotting using appropriate antibodies, and as shown (Fig. 4A)
, drug treatment did not result in altered cdk protein levels. Quantitation of relative activities of all cdks by phosphorimaging indicated (Fig. 4A
, lower panel) a reduction of approximately 50 and 80–90% with 5 and 10 µM perifosine, respectively. Similar results were observed for HaCaT and HN30 cells (data not shown). These findings suggest that the loss in cdc2 and cdk2 activity may explain the effects of perifosine in G1-S and G2-M cell cycle progression.
|
, exposure to perifosine (12–24 h) results in a decrease in pRb phosphorylation on threonine 356 and serine 780, suggesting that perifosine diminishes effectively the in vivo activity of G1 cdks (cdk2 and cdk4/6).
To determine whether perifosine could modulate the activity of cdks by direct interaction with the catalytic subunit of these kinases, lysates obtained from exponentially growing HN30 cells were immunoprecipitated for in vitro kinase reactions, and properly activated cdc2 and cdk2 were incubated with increasing concentrations of perifosine in the kinase reaction. As a positive control, cdc2 and cdk2 complexes were incubated with flavopiridol (300 nM), a known direct inhibitor of cdks (22, 23, 24, 25)
. As shown in Fig. 4C
, no effect was observed on cdc2 and cdk2 kinase activity up to 20 µM perifosine, whereas 300 nM flavopiridol demonstrated a characteristic inhibition of cdk activity under similar experimental conditions (Fig. 4C
, last lane). In contrast, as described in Fig. 4A
, kinase complexes obtained from HN30 cells previously exposed to perifosine (10 µM) demonstrated reduced intrinsic activity (data not shown). Thus, the effect of perifosine on cdk activity in intact cells may occur by the modulation of upstream signals important for cdk activation.
Perifosine Arrests Cells at the G2-M Transition by Up-Regulation of p21WAF1 and Loss of cdc2 Activity in HNSCC Cells.
Because perifosine treatment results in arrest and accumulation of cells in G2, we sought to assess the kinetics of cdc2 kinase activity after drug exposure in more detail. As an approach, G1-S-synchronized HN12 cells were released in the presence or absence of 10 µM perifosine and harvested (0–12 h) for the assessment of cdc2 activity as described above. As shown in Fig. 5A
, HN12 cells demonstrated minimal cdc2 activity upon release from the aphidicolin block (0–3 h), which is indicative of cells in S-phase. Further cell cycle progression (6–9 h) results in significant elevation of this activity, which peaked at 12 h after release. This increased cdc2 activity represents the required activation of the kinase necessary for G2-M progression (26)
. However, cells released in the presence of perifosine initially showed a minimal increase in cdc2 activity (3–6 h, see below) but further elevations were prevented (9–12 h). Thus, the data indicate that the accumulation of cells with a G2 DNA content by perifosine treatment may result from a loss in cdc2 activity. As a positive control, G1-S-synchronized HN12 cells released in the presence of nocodazole, a known mitotic blocker, resulted in a characteristic activation of cdc2 by 12 h (27)
. Parallel lysates were analyzed by Western blot analysis and, as indicated in Fig. 5B
, showed no differences in cdc2 expression that may explain the loss of activity of this kinase by perifosine. Similarly, cyclin A levels were mostly unaltered and, as expected, those of cyclin B1 were initially absent at the G1-S transition (0 h) but showed a gradual increase in its expression, reaching a maximal level at 12 h that is crucial for cdc2 activation (26)
. Of note, cyclin B1 expression appears to be induced earlier after perifosine treatment (Fig. 5B
, third lane). This elevation may explain the apparent increase in cdc2 activity observed in Fig. 5A
. At 9 h after release, perifosine-treated cells demonstrated unaltered cdc2, cyclin A, or cyclin B1 expression, despite lack of cdc2 activity. However, by 12 h, a slight decline in cyclin B1 was observed, which is unlikely to explain fully the total loss in cdc2 activity observed 9–12 h after treatment.
|
Increased p21WAF1 Expression and Its Association with Cdk Complexes in Perifosine-treated HNSCC Cells Results in Loss of Kinase Activity.
To further characterize the effects of perifosine on p21WAF1 induction and to determine whether increased levels of this protein associates with the cdk complexes, we initially compared the effects of perifosine to -irradiation, an established DNA-damaging agent that provokes accumulation of cell in G2 phase in HNSCC cells (21)
. As observed elsewhere, treatment of HaCaT cells with -irradiation provokes accumulation of cells in G2 phase because of loss in cdc2 activity (21)
. As shown in Fig. 6A
, -irradiation was unable to induce p21WAF1. In contrast, treatment of HaCaT cells with perifosine provokes a rapid (3-h) induction of p21WAF1, followed by a decline in cyclin B1 by 24 h. As stated elsewhere (19
, 20)
, these cells lack normal p53 function; thus, the p21WAF1 induction observed occurs, again, through a p53-independent pathway. Furthermore, steady-state expression levels of cdc2, cdk2, and ß-actin were unaltered by either treatment. Similar effects were observed in the other three remaining HNSCC cells (data not shown).
|
shows that perifosine treatment (12–24 h) resulted in increased association of p21WAF1 with the cdc2 complex. However, no differences in cyclin A association with cdc2 was observed with perifosine treatment, whereas levels of immunoprecipitated cdc2 proteins were unaltered. Similarly, cdk2 immunoprecipitates demonstrated a clear increase in p21WAF1 association. A limited increase in p27KIP1 association was also observed, but no differences were noted in the relative levels of immunoprecipitated cdk2 protein upon treatment. Finally, p21WAF1 immunoprecipitates demonstrated that the total amount of protein immunoprecipitated increased (12–24 h), reflecting the induction of p21WAF1 provoked by perifosine. Interestingly, the association of p21WAF1 with either cdc2 or cyclin A was dramatically elevated. The data indicate that p21WAF1 associates with cdc2 and cyclin A upon treatment of cells with perifosine. The observed association could lead to a loss in cdc2 activation at an earlier point in the G2 phase, thus preventing the later activation of cdc2/cyclin B1. Collectively, the data suggest that the loss in cdk activity provoked by perifosine at either G1-S or G2-M may result from the induction and increased association of p21WAF1 to cyclin/cdk partners, which would explain a delay in cell cycle progression from G2 to M and potentially in G1 phase as well.
Assessment of Perifosine on the Cell Cycle Progression in Isogenic HCT116 Carcinoma Cells.
To assess the contribution of p21WAF1 to the cell cycle effects of perifosine, we took advantage of the availability of isogenic HCT116 colon carcinoma variant cells (p53+/+, p53-/-, and p21WAF1-/-; Ref. 30
). As demonstrated in Fig. 7, A and B
, perifosine is able to block cell cycle progression at G1-S and G2-M in the p53+/+ variant HCT116 cells. Furthermore, perifosine caused the fraction of cells in G2-M phase to increase dramatically from 17 to 67%, whereas those in S-phase declined from 29 to 10%. This effect of perifosine in p53 +/+ cells was associated with an induction in expression of p21WAF1 (Fig. 7C)
. Furthermore, p53 protein levels were also induced. To evaluate whether the cell cycle arrest and the induction of p21WAF1 was dependent on p53 function, we tested the effect of perifosine in the p53 -/- HCT116 variant cells. As demonstrated (Fig. 7)
, G2-M cell fraction increased from 19 to 72% upon perifosine treatment, with a loss in S-phase (from 41 to 11%), and induction of p21WAF1, despite lack of p53. Thus, p53 function is dispensable for the cell cycle effects of perifosine and for the induction of p21WAF1, which is in line with our previous observations in "p53-defective" HaCaT and HN12 cells (19
, 20
, 31)
. To establish conclusively the role of p21WAF1 in the cell cycle effects of perifosine, we also tested the HCT116 p21WAF1-/- variants. As observed in Fig. 7
, perifosine failed to induce cell cycle arrest in these cells, which was associated with the lack of p21WAF1 induction despite normal expression of p53 in these cells. To determine whether the effects on cell cycle progression are shared by two other known ALKs, edelfosine and miltefosine, we exposed HCT116 cell lines to these two compounds. Prominent G1-S and G2-M arrest was observed along with p21WAF1 increase in the p53+/+ and p53-/- variants (data not shown). Furthermore, when HCT116 p21-/- cells were exposed to these compounds, similar to perifosine, no cell cycle effects were observed (data not shown). Collectively, we can conclude that ALKs, including perifosine, may block cell cycle progression by the p53-independent induction of p21WAF1.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-irradiation (16
, 18
, 34)
and can induce differentiation in a number of leukemia models (35)
. Although the exact mechanism of action of ALK is unknown, it is thought that these molecules can modulate signal transduction pathways at the membrane level by modulating protein kinase C and phospholipase ß, among others (15
, 17
, 36)
. Although members of the ALK family are able to induce apoptosis in combination with -irradiation (16)
, the role of ALKs on cell cycle progression is still unknown, which prompted us to investigate the cell cycle effects of perifosine, the only p.o.-acting ALK in clinical trials. Initially, we determined the antiproliferative properties of perifosine in HNSCC. Potent antiproliferative effects were observed in this panel of cell lines with IC50s in the 1–10 µM range, irrespective of the presence of a functional p53. Similar activity of perifosine was observed in other keratinocyte models (9) . To examine the antiproliferative effect of perifosine in more detail, cell cycle progression of HNSCC lines exposed to perifosine was determined. Surprisingly, accumulation of cells with G1 and G2-M DNA content was observed in all cell types tested. Time-dependent studies showed that 4 h of exposure to perifosine is sufficient to promote the arrest of cells with G1 and G2-M DNA content and relative loss in S-phase population. To determine more precisely where in G2-M perifosine may be arresting cells, mitotic index analyses were conducted and showed that perifosine arrests cells at G2 with interphase nuclei, in contrast to nocodazole, a known microtubule inhibitor, which arrested cells at the M phase.
The progression of the cell cycle is governed by the cyclical activation of serine-threonine kinases, cdks, that are formed by the cdk catalytic subunit (cdk), positive cofactors (cyclins), and negative cofactors, endogenous cdk inhibitors (26 , 37 , 38) . Modulation of cdks by direct interaction with the catalytic subunit, or indirectly by modulating the upstream cofactors necessary for cdk activation or by the up-regulation of endogenous cdk inhibitor, can lead to loss in cdk activity with arrest in cell cycle progression (24 , 25 , 37 , 39) . To determine whether perifosine has the capacity to modulate cdk activity, HNSCCs were exposed to perifosine, and cdk2 and cdc2 kinase activity was assessed by immunocomplex kinase reactions. Loss in cdk activity was demonstrated with perifosine. However, differences observed in the rate of loss of activity between the two kinases may reflect, in part, increased sensitivity of cdk2 to the inhibitory effects of perifosine, resulting from raised intrinsic kinase activity attributable to a high number of asynchronous cells traversing the G1 and S phases (>80%) at any given time. Additionally, this difference could result from a complex interplay of rate of induction of p21WAF1 (at least 6 h), the molar concentration of active cdk2 or cdc2 kinases, and the time required for completion of S-phase. Furthermore, to confirm the loss of cdk activity in this cellular system, we measured the phosphorylation status of the protein product of the Rb tumor suppressor gene (pRB), a known endogenous substrate for cdks, by site specific phospho-specific antisera. Indeed, loss in phosphorylation at specific cdk2 and cdk4/cdk6 sites were observed with perifosine, suggesting that the drug has the capacity to inhibit both G1 and G2 cdks, when assessed by either in vitro kinase assays or loss in pRb phosphorylation status. To determine whether this novel ALK targets the catalytic subunit of cdks, active cdc2 and cdk2 obtained from exponentially growing cells were immunoprecipitated, and perifosine was added to the kinase reaction. Under these conditions, perifosine did not inhibit cdks, although the cdk activity from intact cells treated with perifosine was significantly diminished. Thus, the loss in cdk activity may reflect an indirect effect of perifosine on the cdk complex.
To further examine the effects of perifosine on cdks, aphidicolin-synchronized cells were released in the presence of perifosine. We found that perifosine prevented the activation of cdc2 necessary for the G2-M transition. Of note, examination of the molecular complexes containing cdc2 revealed that the loss of cdc2 activity was clearly preceded by up-regulation of p21WAF1. p21WAF1 belongs to the cip/kip family of endogenous cdk inhibitors (p21, p27, and p57), which regulates directly the activity of cdks, and was initially identified as a mediator of p53-induced growth arrest (40, 41, 42, 43) . This family of endogenous inhibitors can prevent the activation of cdks. Moreover, even activated cyclin/cdk complexes are readily inhibited by cip family members (44) . Although initially thought to play a unique role in G1-S transition, it has become apparent recently that at least p21WAF1 and p27KIP1 also have a clear role in the G2-M transition (28 , 45) . Furthermore, biochemical and genetic studies demonstrated that p21WAF1 family members at low concentrations may promote the assembly and activation of D-type cyclin kinase, whereas at higher concentrations, they are able to suppress cdk activity (45 , 46) . Besides cell cycle control, p21WAF1 has reported roles in transcription, DNA repair, differentiation, and apoptosis (47, 48, 49, 50) . A major transcriptional regulator of p21WAF1 is the tumor suppressor gene p53, which binds to specific p53-binding sites within the p21WAF1 promoter (41) , although additional p53-independent mechanisms have also been described (51, 52, 53) .
To assess the biological importance for the up-regulation of p21WAF1 induced by perifosine, we compared the effects of a known DNA-damaging agent, -irradiation, in HaCaT cells, an immortalized keratinocyte cell line with nonfunctional p53 (31)
. Although both treatments provoked the arrest of cells in G2-M, only perifosine up-regulated p21WAF1. Thus, the induction observed is not dependent on a functional p53. To determine whether the increased p21WAF1 is indeed associated with cdk complexes, HNSCC cells exposed to perifosine demonstrated a clear increased association between p21WAF1 and both G1 and G2 cyclin/cdk complexes. p21WAF1 was also associated with cyclin A/cdk complexes. This association may explain the lack of cdc2 activation observed in aphidicolin-synchronized cells upon perifosine treatment, because cyclin A, the predominant cyclin in S-phase, initially binds and activates cdk2 in this phase, and subsequently, cyclin A associates with cdc2 in late S and early G2 phases (26
, 38)
. Thus, induced p21WAF1 may prevent the proper activation of cdc2/cyclin A, leading to a block in early G2 phase, and may represent a good candidate to explain the observed loss of cdk activity elicited by the treatment with perifosine.
To determine the exact role of p21WAF1 in the cell cycle effects of perifosine, we used the HCT116 cell lines isogenic for wild-type, p53-/-, and p21WAF1-/- genotypes (30) . As expected, wild-type cells exposed to perifosine demonstrated clear evidence of G1-S and G2-M arrest. Again, this effect was accompanied by induction of p21WAF1. When the p53-/- cell lines were exposed to perifosine, nearly identical effects to those displayed in wild-type cells were observed, reinforcing the notion that the p21WAF1 up-regulation is not dependent on p53 function, and that the cell cycle effects of perifosine are not attributable to p53 modulation. Finally, cells devoid of p21WAF1 but with intact p53 failed to arrest upon perifosine treatment. Together, these data provide the first evidence that perifosine can arrest cells G2-M transition and delay in the progression of G1 by up-regulating p21WAF1 in a p53-independent fashion, thus leading to a loss in cdk activity. The actual mechanism whereby perifosine may modulate the expression of p21WAF1 is still unclear and under current investigation. Furthermore, the detailed mechanism of effects in G1 will require analysis of cdk2 activation in background controlled for Rb expression.
In summary, our data indicate that the ALK perifosine promotes cell cycle arrest at either G1-S or G2-M because of a p53-independent up-regulation of p21WAF1. On the basis of these findings, we present a working model (Fig. 8)
by which ALKs may modulate cdks and block cell cycle progression. In this model, ALKs induce p21WAF1 expression by still unknown p53-independent mechanisms, and accumulated p21WAF1 protein leads to a loss in cdk activity by associating with cdk/cyclin complexes, resulting in the accumulation of cells in G1 and G2-M, even in tumor types with abnormal p53 gene status. The predicted capacity to inhibit cdk2/cyclin E will require experimental confirmation in appropriate models. This novel effect preventing cell cycle progression, along with the known apoptotic properties of ALKs, suggests that these anticancer agents may represent good candidates for further evaluation in the treatment and/or prevention of a variety of human neoplasms.
|
|
FOOTNOTES |
|---|
1 To whom requests for reprints should be addressed, at Oral and Pharyngeal Cancer Branch, National Institute of Craniofacial and Dental Research, NIH, 30 Convent Drive, Building 30, Room 212, Bethesda, MD 20892-4330. Phone: (301) 496-6259; Fax: (301) 402-0823; E-mail: sendero{at}helix.nih.gov. 
2 The abbreviations used are: HNSCC, head and neck squamous cell carcinoma; ALK, alkylphospholipid; cdk, cyclin-dependent kinase.
3 T. Lahusen, unpublished results.
Received 8/15/01. Accepted 1/ 3/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B by cyclin-dependent kinases associated with the p300 coactivator. Science (Wash. DC), 275: 523-527, 1997.This article has been cited by other articles:
|
|
 |
|
  H. A. Elrod, P. Yue, F. R. Khuri, and S.-Y. Sun Celecoxib antagonizes perifosine's anticancer activity involving a cyclooxygenase-2-dependent mechanism Mol. Cancer Ther., September 1, 2009; 8(9): 2575 - 2585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Leleu, J. Eeckhoute, X. Jia, A. M. Roccaro, A.-S. Moreau, M. Farag, A. Sacco, H. T. Ngo, J. Runnels, M. R. Melhem, et al. Targeting NF-{kappa}B in Waldenstrom macroglobulinemia Blood, May 15, 2008; 111(10): 5068 - 5077. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Vinall, K. Hwa, P. Ghosh, C.-X. Pan, P. N. Lara Jr., and R. W. de Vere White Combination Treatment of Prostate Cancer Cell Lines with Bioactive Soy Isoflavones and Perifosine Causes Increased Growth Arrest and/or Apoptosis Clin. Cancer Res., October 15, 2007; 13(20): 6204 - 6216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Elrod, Y.-D. Lin, P. Yue, X. Wang, S. Lonial, F. R. Khuri, and S.-Y. Sun The alkylphospholipid perifosine induces apoptosis of human lung cancer cells requiring inhibition of Akt and activation of the extrinsic apoptotic pathway Mol. Cancer Ther., July 1, 2007; 6(7): 2029 - 2038. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hideshima, L. Catley, H. Yasui, K. Ishitsuka, N. Raje, C. Mitsiades, K. Podar, N. C. Munshi, D. Chauhan, P. G. Richardson, et al. Perifosine, an oral bioactive novel alkylphospholipid, inhibits Akt and induces in vitro and in vivo cytotoxicity in human multiple myeloma cells Blood, May 15, 2006; 107(10): 4053 - 4062. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Vink, S. Lagerwerf, E. Mesman, J. H.M. Schellens, A. C. Begg, W. J. van Blitterswijk, and M. Verheij Radiosensitization of Squamous Cell Carcinoma by the Alkylphospholipid Perifosine in Cell Culture and Xenografts Clin. Cancer Res., March 1, 2006; 12(5): 1615 - 1622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Momota, E. Nerio, and E. C. Holland Perifosine Inhibits Multiple Signaling Pathways in Glial Progenitors and Cooperates With Temozolomide to Arrest Cell Proliferation in Gliomas In vivo Cancer Res., August 15, 2005; 65(16): 7429 - 7435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. F. Aguero, M. M. Facchinetti, Z. Sheleg, and A. M. Senderowicz Phenoxodiol, a Novel Isoflavone, Induces G1 Arrest by Specific Loss in Cyclin-Dependent Kinase 2 Activity by p53-Independent Induction of p21WAF1/CIP1 Cancer Res., April 15, 2005; 65(8): 3364 - 3373. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rahmani, E. Reese, Y. Dai, C. Bauer, S. G. Payne, P. Dent, S. Spiegel, and S. Grant Coadministration of Histone Deacetylase Inhibitors and Perifosine Synergistically Induces Apoptosis in Human Leukemia Cells through Akt and ERK1/2 Inactivation and the Generation of Ceramide and Reactive Oxygen Species Cancer Res., March 15, 2005; 65(6): 2422 - 2432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Van Ummersen, K. Binger, J. Volkman, R. Marnocha, K. Tutsch, J. Kolesar, R. Arzoomanian, D. Alberti, and G. Wilding A Phase I Trial of Perifosine (NSC 639966) on a Loading Dose/Maintenance Dose Schedule in Patients with Advanced Cancer Clin. Cancer Res., November 15, 2004; 10(22): 7450 - 7456. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. P. Dasmahapatra, P. Didolkar, M. C. Alley, S. Ghosh, E. A. Sausville, and K. K. Roy In vitro Combination Treatment with Perifosine and UCN-01 Demonstrates Synergism against Prostate (PC-3) and Lung (A549) Epithelial Adenocarcinoma Cell Lines Clin. Cancer Res., August 1, 2004; 10(15): 5242 - 5252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Facchinetti, A. De Siervi, D. Toskos, and A. M. Senderowicz UCN-01-Induced Cell Cycle Arrest Requires the Transcriptional Induction of p21waf1/cip1 by Activation of Mitogen-Activated Protein/Extracellular Signal-Regulated Kinase Kinase/Extracellular Signal-Regulated Kinase Pathway Cancer Res., May 15, 2004; 64(10): 3629 - 3637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. De Siervi, M. Marinissen, J. Diggs, X.-F. Wang, G. Pages, and A. Senderowicz Transcriptional Activation of p21waf1/cip1 by Alkylphospholipids: Role of the Mitogen-Activated Protein Kinase Pathway in the Transactivation of the Human p21waf1/cip1 Promoter by Sp1 Cancer Res., January 15, 2004; 64(2): 743 - 750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Fukuda, C. R. Mantel, and L. M. Pelus Survivin regulates hematopoietic progenitor cell proliferation through p21WAF1/Cip1-dependent and -independent pathways Blood, January 1, 2004; 103(1): 120 - 127. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. B. Kondapaka, S. S. Singh, G. P. Dasmahapatra, E. A. Sausville, and K. K. Roy Perifosine, a novel alkylphospholipid, inhibits protein kinase B activation Mol. Cancer Ther., November 1, 2003; 2(11): 1093 - 1103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Lapillonne, M. Konopleva, T. Tsao, D. Gold, T. McQueen, R. L. Sutherland, T. Madden, and M. Andreeff Activation of Peroxisome Proliferator-activated Receptor {gamma} by a Novel Synthetic Triterpenoid 2-Cyano-3,12-dioxooleana-1,9-dien-28-oic Acid Induces Growth Arrest and Apoptosis in Breast Cancer Cells Cancer Res., September 15, 2003; 63(18): 5926 - 5939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Frederich, M. Bentires-Alj, M. Tits, L. Angenot, R. Greimers, J. Gielen, V. Bours, and M.-P. Merville Isostrychnopentamine, an Indolomonoterpenic Alkaloid from Strychnos usambarensis, Induces Cell Cycle Arrest and Apoptosis in Human Colon Cancer Cells J. Pharmacol. Exp. Ther., March 1, 2003; 304(3): 1103 - 1110. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, HaCaT; 
, HN4; 
, HN8; 
, HN12; 
, HN19; and 
, HN30) were grown overnight in 24-well plates and exposed to perifosine (0.1–30 µM) or PBS as control for an additional 24 h. Cells were then pulsed with [3H]thymidine for 4–6 h, fixed, and solubilized, and samples were measured in a scintillation counter. Data represent the percentage of inhibition of [3H]thymidine incorporation into cellular DNA relative to control. The results are the means of three independent experiments and are used to calculate the IC50s of each cell line tested. SD between experiments was <10%.
