| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Laboratory of Metabolism [Y-S. S., S-Y. K., K. S. L.] and Laboratory of Cellular Carcinogenesis and Tumor Promotion [L. L., S. H. Y.], Center for Cancer Research, National Cancer Institute, NIH, Bethesda, Maryland 20892; LG Biomedical Institute, La Jolla, California 92037 [C. M., J. Y. Y., X. C., K. K., H-H. C.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Regulation of Cdks could be achieved either by directly inhibiting their catalytic activity or by indirectly modulating the activity of Cdk regulators or associated proteins. Among various approaches, the most effective way of inhibiting Cdk activity appears to be by small-molecule chemical compounds. For more than a decade, direct small-molecule Cdk inhibitors have been developed and characterized. Among these, two purine derivatives, olomucine and roscovitine, which exhibit potent inhibition against Cdks, have been relatively well characterized. Olomucine inhibits Cdk1 and Cdk2 with an IC50 of 7 µM (4
, 5)
, whereas roscovitine, which is derived from olomucine, exhibits even more potency, with an IC50 of 0.7 µM for Cdk1 and Cdk2 (4)
. Crystal structure analysis showed that roscovitine or olomucine binds to the ATP-binding site of Cdks (6
, 7)
. In addition, flavopiridol, a semisynthetic flavonoid derived from a plant alkaloid, rohitukine, exhibits nonspecific Cdk inhibitory activity and arrests cells in G1-S phase and at the G2-M boundary (8)
. Because of the conserved Cdks structures, flavopiridol exhibits an IC50 of 
100 nM against various Cdks (9
, 10)
. As with olomucine and roscovitine, flavopiridol binds to the ATP-binding sites of Cdks (11)
and competitively inhibits these enzyme activities (10)
. In addition, flavopiridol exhibits a potent in vitro antiproliferative activity when tested against 60 NCI human tumor cell lines and is currently under clinical trials (reviewed in Refs. 3
and 12
).
Although regulation of cellular proliferation through the modulation of Cdk activity is an attractive approach, questions still remain as to how effective their biological activities are or whether more than one inhibitor should be combined to accomplish an effective physiological outcome. In an effort to generate a new class of Cdk inhibitors, we synthesized aryl aminopyrimidines substituted with additional aromatic heterocycles. Here we report the initial characterization of one of these inhibitors, BMI-1026, that exhibited IC50 <10 nM against various Cdks in vitro. Our results suggest that BMI-1026 induces a potent mitotic arrest, which is accompanied by mitotic catastrophe and apoptotic cell death in cultured cells. In addition, proliferating mouse keratinocytes, but not differentiated mouse keratinocytes, exhibit a selective sensitivity to BMI-1026. These data suggest that BMI-1026 could be developed as a promising small-molecule inhibitor specific for mitotic Cdk activity.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
for structures). Both BMI-1026 and BMI-1042 were analyzed by nuclear magnetic resonance to confirm the structure (data not shown). Synthesis of these compounds will be published elsewhere. A stock solution (1 mg/ml) of either BMI-1026 or BMI-1042 was made in DMSO and used at the indicated concentration.
|
Primary mouse keratinocytes were freshly isolated from newborn BALB/c mice and cultured in Eagle’s MEM (Invitrogen) supplemented with 0.05 mM CaCl2 and 8% chelexed fetal bovine serum (low calcium medium) as described previously (13) . Terminal differentiation of mouse primary keratinocytes was induced with the same medium containing 1.4 mM CaCl2 (high calcium medium). SP1 (14) , a tumorigenic mouse keratinocyte cell line, was also maintained in the low-calcium medium.
Flow Cytometry Analysis.
Flow cytometry analyses were carried out with FACSCalibur (Becton Dickinson, San Jose, CA) as reported previously (15)
. Data were analyzed by CellQuest and Modfit software (Becton Dickinson).
Immunoblotting and in Vitro Kinase Assays.
For immunoblotting analyses, anti-Cdc25C antibody (Santa Cruz Biotechnologies, Santa Cruz, CA), anti-Cdc27 antibody (Santa Cruz Biotechnologies), anti-cyclin B1 antibody (Santa Cruz Biotechnologies), anti-Cdk1 antibody (Upstate Biotechnology Inc., Lake Placid, NY), anti-Plk1 COOH-terminal antibody (Zymed, South San Francisco, CA), and anti-PARP antibody (Santa Cruz Biotechnologies) were used at 0.5 µg/ml. Immunoblotting was carried out as described previously (15)
.
In vitro kinase assays for immunoprecipitated Cdk1 or Plk1 were carried out as described previously (16) using histone H1 (Calbiochem, La Jolla, CA) and casein (Sigma) as substrates for Cdk1 or Plk1, respectively. For assays with recombinant Cdk1 or Plk1, recombinant Cdk1/GST-cyclin B1 (a gift of H. Piwnica-Worms, Washington University, St. Louis, MO) and GST-Plk1 (a gift of F. R. Yarm and R. L. Erikson, Harvard University, Cambridge, MA) were purified from Sf9 cells using GSH-agarose beads (Sigma).
To determine the IC50 values for the Cdks, GST-Cdk1/GST-cyclin B1 and GST-Cdk2/GST-cyclin A were purified from Sf9 cells, whereas GST-CDK5/GST-p25 was purified from Escherichia coli. These enzyme complexes were then reacted with a synthetic peptide derived from histone H1 (PKTPKKAKKLRRR). Both His6-Plk1 expressed in Sf9 cells and His6-Aurora A expressed in E. coli were purified by IMAC affinity chromatography (Clontech, Palo Alto, CA), and then reacted with casein or histone H3, respectively, as a substrate. PKA assays were carried out with the SignaTECT cAMP-dependent protein kinase assay system (Promega, Madison, WI) using PKA catalytic subunit and a biotinylated Kemptide (LRRASLG). PKC
assays were carried out with a PKC assay system (Panvera, Madison, WI) using 200 µg/ml phosphoserine, 20 µg/ml diacylglycerol, and a PKC substrate (RFARKGALRQKNV). GST-Erk1 was purified from Sf9 cells by use of GSH-agarose beads, and reactions were carried out with myelin basic protein as a substrate.
Immunofluorescence Microscopy.
For indirect immunofluorescence studies, U-2 OS cells were grown on poly-L-lysine-coated (Sigma) glass coverslips and then fixed with 4% paraformaldehyde for 10 min. These cells were then treated with PBS containing 0.5% Triton X-100 and 0.1 µg/ml DAPI (Sigma) to visualize the chromosomal DNA. Fluorescent images were collected with a Zeiss LSM510 confocal microscope.
In Vivo Time-Lapse Microscopy.
An U-2 OS cell line expressing GFP-histone H2B (15)
was cultured on a 35-mm dish on the stage of an Axiovert S-100 inverted microscope equipped with an environmental chamber (Zeiss, Thornwood, NY). Time-lapse images were captured by a SenSys digital camera (Photometrics, Tucson, AZ) and analyzed by Openlab software (Improvision, Coventry, United Kingdom).
APO-BrdU Assay.
To measure apoptosis, samples were prepared using the APO-BrdU assay kit (Biosource, Camarillo, CA). Briefly, U-2 OS cells were fixed in 1% paraformaldehyde in PBS buffer and washed; the cell number was then adjusted to 2 x 106/ml in 70% ice-cold ethanol. These cells were incubated with DNA-labeling solution for 1 h at 37°C and then further incubated with fluorescein-labeled PRB-1 antibody for 30 min. Samples were then incubated with propidium iodide/RNase solution before flow cytometry analysis. To determine the percentage of BrdU-positive cells, the flow cytometry data were analyzed by the CellQuest program (Becton Dickinson) according to the manufacturer’s protocol.
Cell Viability Assay.
The effect of BMI-1026 on the viability of cultured mouse primary keratinocytes was determined using CellTiter 96 Non-Radioactive Cell Proliferation Assay (Promega). In brief, primary mouse keratinocytes maintained in the low-calcium medium (0.05 mM CaCl2) in 24-well plates were divided into three groups. To induce differentiation, cells were shifted to the high-calcium medium (1.4 mM CaCl2) for 18 h. These cells were then treated with various concentrations of BMI-1026 in the same medium. To examine the effect of BMI-1026 on induced differentiation, keratinocytes maintained in the low-calcium medium were directly transferred to the high-calcium medium supplemented with various concentrations of BMI-1026. To determine the effect of BMI-1026 on the proliferating cells, keratinocytes maintained in the low-calcium medium were transferred to the same low-calcium medium supplemented with BMI-1026. Twenty-two h after the addition of BMI-1026, samples were harvested and subjected to the cell viability assay. The rates of cell survival were determined by comparing each group of treated cells with the corresponding untreated cells.
Online Supplemental Material.
Videos of cells depicted in Figs. 3, A and B
, and Fig. 4B
are provided as online supplemental data. A supplementary video (S1) shows two mitotic U-2 OS cells exhibiting precocious mitotic exit in the presence of 200 nM BMI-1026. In this case, BMI-1026 was added directly to the normal U-2 OS culture medium (time 0), and then mitotically rounded cells were closely monitored. Both cells exhibited premature contraction in the absence of sister-chromatid separation, leading to the generation of a "cut" morphology. Trapped chromosomal DNA visualized by GFP-histone H2B (green) was evident at intracellular bridges between the two dividing cells.
|
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. As expected, Cdk1/cyclin B, Cdk2/cyclin A, and Cdk5/p25 were potently inhibited by <10 nM BMI-1026 or BMI-1042 (Table 1)
. In contrast, both mitotic polo-like kinase Plk1, which plays a critical role in various mitotic events (reviewed in Ref. 17
), and Aurora-A, whose overexpression leads to centrosome amplification and chromosomal instability (18
, 19)
, were largely insensitive to BMI-1026 and BMI-1042 (Table 1)
. In addition, BMI-1026 did not inhibit PKA, PKC, and Erk1 significantly (Table 1)
.
|
. At 80 nM, a small fraction of cells exhibited 8N DNA content, suggesting a failure in cytokinesis. In addition, accumulation of sub-G1 population was evident in the presence of 80 or 100 nM of BMI-1026 (Fig. 2A
, arrows). The same concentration of BMI-1042 induced a similar, but somewhat weaker, cell cycle alteration (data not shown). This observation suggests that although BMI-1042 is a more potent Cdk1 inhibitor than BMI-1026 in vitro (Table. 1
), it is less effective in cultured cells. Thus, BMI-1026 was chosen for further characterization.
|
, accumulation of G2-M cells was evident in all three concentrations of BMI-1026 examined (Fig. 2B)
. At higher concentrations (100 and 200 nM) of BMI-1026, however, significant delays in the G1 and S phases of the cell cycle were also manifest. A small but significant fraction of cells with sub-G1 DNA content was also reproducibly detectable under these conditions. These data suggest that BMI-1026 primarily interferes with proper G2-M progression and that a higher dosage of BMI-1026 can also interfere with early stages of the cell cycle.
Induction of Mitotic Catastrophe and Cell Death by BMI-1026.
The appearance of a sub-G1 population could be the consequence of mitotic interference imposed by BMI-1026. Alternatively, it could be induced by a mechanism independent of mitotic inhibition. To distinguish these possibilities, U-2 OS cells stably expressing GFP-histone H2B were treated with a low dose (80 nM) of BMI-1026 and then closely monitored by time-lapse microscopy. Within 48 h after the addition of BMI-1026, most of the cells proceeding through M phase (95.6%, n = 46) developed an apparent cut morphology with ingression between unseparated chromosomal DNA (Fig. 3
, arrows). Close examination of these cells revealed that 54% (25 of 46) of these cells prematurely proceeded through anaphase without an apparent chromosomal congregation at the metaphase plate, whereas 41% (19 of 46) exhibited the cut morphology after an apparent metaphase alignment (Table 2)
. These observations suggest that, under these conditions, sister chromatid separation and exit from mitosis are not coordinated, resulting in a precocious cytokinetic event. In both cases, these cells frequently exhibited multiply lobulated morphology (85%; 39 of 46) and died before the next round of mitosis (Fig. 3)
. These data suggest that treatment of cells with 80 nM BMI-1026 results in multiple mitotic failures and that this mitotic catastrophe may ultimately lead to cell death in the subsequent cell cycle.
|
. More than 60% of the cells exhibited this morphology 67 h after treatment. However, consistent with the slow cell cycle progression observed in Fig. 2B
, there appeared to be a significant delay in generating mitotically arrested cells under these conditions. Because Cdk1 activity is required for mitotic entry, the accumulation of the rounded morphology could be the result of a potent inhibition of Cdk1 by BMI-1026 at the time of mitotic onset.
We next asked whether BMI-1026 can inhibit Cdk1 activity in mitotically arrested cells, a stage when Cdk1 is maximally active. Cdk1 activity is required for maintaining the mitotic status and down-regulation of Cdk1 activity is a prerequisite for mitotic exit and the onset of cytokinesis. Thus, if the inhibition of Cdk1 is the primary effect induced by the treatment of BMI-1026, then a forced inactivation of mitotic Cdk1 by BMI-1026 may induce exit from mitosis. To test this possibility, U-2 OS cells arrested at prometaphase by treatment with 200 ng/ml nocodazole for 16 h were additionally treated with 200 nM BMI-1026 and then subjected to time-lapse microscopy. Approximately 40 min after the addition of BMI-1026, cells began to exhibit severely elongated morphologies (0:40; Fig. 4B
). After subsequent membrane blebbings (1:00), these cells exhibited a G1-like morphology with an apparent micronucleation (2:23; Fig. 4B
, arrows). Neither the elongated morphology nor the membrane blebbing were observed in control cells arrested with nocodazole alone for up to 24 h (data not shown), indicating that these phenotypic changes are BMI-1026 specific. These observations suggest that BMI-1026-dependent inhibition of Cdk1 is sufficient to induce a forced mitotic exit even in the presence of nocodazole. Interestingly, during the course of this experiment the morphologies of interphase cells did not appear to be significantly influenced (Fig. 4B)
, suggesting that mitotic cells are selectively sensitive to BMI-1026. In addition, the observed premature mitotic exit did not appear to be attributable to a nocodazole effect because mitotic cells in normal growing medium supplemented with BMI-1026 alone exhibited similar phenomena (Supplementary video S1).
BMI-1026 Inhibits Cdk1 but Not Plk1.
To confirm whether the premature mitotic exit correlates with the inhibition of Cdk1, U-2 OS cells arrested with nocodazole for 16 h were additionally treated with either 200 nM BMI-1026 or DMSO as a control. As a comparison, the same nocodazole-arrested cells were released into fresh medium. Samples were prepared at the indicated time points after the treatment, then subjected to immunoblotting and kinase assays to determine the mitotic status of these cells. When nocodazole-arrested cells were treated with DMSO, the phosphorylated forms of Cdc25 and Cdc27 were maintained for up to 7 h. Because Cdc25 and Cdc27 are in vivo substrates of both Cdk1 and Plk1 (20, 21, 22, 23, 24)
and are phosphorylated at an early stage of mitosis, these data indicate that DMSO alone did not interfere with nocodazole-induced mitotic arrest under these conditions. In contrast, addition of BMI-1026 into the nocodazole-treated U-2 OS cells induced dephosphorylation of both Cdc25 and Cdc27 as early as 30 min after treatment (Fig. 5A)
. Consistent with this observation, Cdk1 activity was also completely inhibited within 30 min even in the continuous presence of nocodazole (Fig. 5A)
. The level of mitotic cyclin B1 appeared to be maintained for up to 7 h (Fig. 5A)
, indicating that BMI-1026 inhibits Cdk1 without influencing the level of cyclin B1. When the cells were released from the nocodazole block into fresh medium, the phosphorylated forms of Cdc25 and Cdc27 disappeared 1 h after release. In addition, as with the diminishing levels of cyclin B1, the Cdk1/cyclin B1 activities decreased gradually (Fig. 5A)
. Taken together, these data indicate that BMI-1026 potently inhibits Cdk1 even in the nocodazole-arrested mitotic cells.
|
. In contrast, consistent with the in vitro IC50 assays (Table 1)
, BMI-1026 did not significantly inhibit Plk1 activity from either anti-Plk1 immunoprecipitates or recombinant Plk1 (Fig. 5C)
. These observations suggest that BMI-1026 specifically inhibits the Cdk1 activity but not the Plk1 activity and that the various mitotic defects induced by the treatment of BMI-1026 are likely the result of specific inhibition of Cdk1.
To confirm the specific inhibition of Cdk1 by BMI-1026, we then examined whether expression of Cdk1/cyclin B1 could alleviate the mitotic arrest induced by BMI-1026. To this end, U-2 OS cells treated with 80 nM BMI-1026 for 24 h were infected with adenoviruses expressing Cdk1, cyclin B1, and the tTA tetracycline transactivator. Mild overexpression of Cdk1/cyclin B1 in the absence of doxycycline significantly alleviated the BMI-1026-induced mitotic arrest (Fig. 5, D and E)
. In contrast, cells treated with doxycycline, which represses the expression of Cdk1/cyclin B1, exhibited pronounced mitotic arrest with a small fraction of cells with 8N DNA content (Fig. 5, D and E)
. Under these conditions, the level of exogenously introduced Cdk1-HA was 2–3-fold greater than that of endogenous Cdk1 (Fig. 5E
, top panel). It should be noted that a low level of Cdk1-HA was detectable in the presence of doxycycline (Fig. 5E
, bottom panel), most likely due to leaky tTA activity under these conditions.
Induction of Apoptosis by BMI-1026.
In a dosage-dependent manner, treatment of U2-OS cells with BMI-1026 led to the generation of a sub-G1 population (Fig. 2)
suggestive of apoptosis. To quantitatively examine the ability of BMI-1026 to induce apoptosis, asynchronously growing U-2 OS cells were treated with 50, 100, or 200 nM BMI-1026, harvested, and then analyzed using an APO-BrdU assay system (see "Materials and Methods"). No significant BrdU-positive fraction was induced by the treatment of 50 nM BMI-1026. Treatment of cells with 100 nM of BMI-1026 for 96 h resulted in 23% of the BrdU-positive population as determined by the CellQuest program (see "Materials and Methods"). In the presence of 200 nM BMI-1026, 5% of the population became BrdU positive at the 48 h time point and 38% of the population became BrdU positive 96 h after treatment (Fig. 6A)
. These data suggest that BMI-1026 induces apoptosis in a dosage- and time-dependent manner.
|
for 48 h time point). Because Cdk1 inhibition occurs within 30 min after BMI-1026 treatment (Fig. 5A)
, these observations, together with the data shown in Fig. 6A
, suggest that BMI-1026 can induce apoptosis at a rate much slower than that of Cdk1 inhibition.
To further examine whether BMI-1026 can induce a similar biological effect in epithelial cells, a tumorigenic mouse keratinocyte cell line, SP1 (14)
, was treated with various concentrations of BMI-1026 for 48 h. Treatment of SP1 cells with 50 nM BMI-1026 induced a rounded morphology in 30% of the population, whereas treatment with 200 nM BMI-1026 induced this morphology in nearly all of the population (Fig. 6C)
. To quantitatively assess the degree of apoptosis in these cells, cells treated with 200 nM BMI-1026 for the indicated length of time were analyzed to determine the fragmentation of PARP, a known apoptosis marker (26
, 27)
. An 85-kDa PARP cleavage product appeared as early as 12 h after treatment, and levels increased as proportional to the incubation time (Fig. 6D)
. These data indicate that BMI-1026 can also induce apoptosis in SP1 cells.
Proliferation-Specific Cell Death by BMI-1026.
Because BMI-1026 potently imposes a mitotic block and induces apoptotic cell death, we then examined whether the cell death is specific to cellular proliferation. To examine this possibility, we used primary mouse keratinocytes that can be induced to growth arrest and differentiate in the presence of a high concentration of calcium (13)
. While culturing under proliferation, differentiating, or already differentiated conditions, keratinocytes were treated with various concentrations of BMI-1026, and the cell survival rate was determined and compared with that of the untreated keratinocytes maintained under the same conditions. We observed that differentiated keratinocytes in a high-calcium medium exhibited a significantly better cell survival rate compared with proliferating keratinocytes in a low-calcium medium (Fig. 7)
. Cells induced to differentiate simultaneously with BMI-1026 exposure exhibited a somewhat lower sensitivity to BMI-1026 than the proliferating cells but were more sensitive than growth-arrested cells in a differentiated state (Fig. 7)
. These data suggest that BMI-1026 induces selective cell death on proliferating cells.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
200 nM), BMI-1026 also imposed G1-S- and S-phase blocks in U-2 OS cells (Fig. 2
, and data not shown). Because BMI-1026 can potently inhibit various Cdk complexes in vitro (Table 1)
, the observed G1-S- and S-phase arrest could be the result of inhibition of Cdk2, an enzyme whose activities are critical for G1- and S-phase progression.
Unlike the more potent Cdk2 inhibition observed in vitro (Table 1)
, treatment of U-2 OS cells with BMI-1026 resulted in the accumulation of rounded cellular morphologies with a potent arrest at G2-M (Figs. 2
and 4A
), a stage that requires the activity of Cdk1. In addition, treatment of nocodazole-arrested U-2 OS cells with BMI-1026 resulted in rapid inactivation of Cdk1/cyclin B1 and thereby precocious mitotic exit (Figs. 4
and 5A
), suggesting that potent inhibition of Cdk1 by BMI-1026 led to an exit from mitosis even in the presence of nocodazole. These observations suggest that the inhibitory effect of BMI-1026 is primarily on Cdk1/cyclin B1 activity in cultured cells. In support of this argument, expression of Cdk1/cyclin B1 alleviated the mitotic block induced by BMI-1026 in cultured U-2 OS cells (Fig. 5D)
. Because the effect on the cell cycle may depend on various factors, such as cell type, expression level, or the critical activity of an enzyme required for fulfilling a particular biochemical step, the apparent discrepancy between in vitro kinase assays and the effect of BMI-1026 in cultured cells could be attributable to the differences between these two assay systems. In addition, it is interesting to note that BMI-1026 induces potent G2-M arrest, whereas other Cdk inhibitors, such as flavopiridol and roscovitine, induce G1 arrest. Although we cannot eliminate the possibility that BMI-1026 interferes with other uncharacterized mitotic events, the data provided here suggest that BMI-1026 may have a better in vivo selectivity for Cdk1/cyclin B1 than other Cdk inhibitors.
Consistent with the potent G2-M phase arrest, treatment of U-2 OS cells with BMI-1026 resulted in various mitotic failures. In vivo time-lapse studies revealed that U-2 OS cells exposed to a low concentration (80 nM) of BMI-1026 exhibited mitotic catastrophe in a large fraction (96%) of mitotic cells. As a result, these cells frequently exhibited a cut morphology with ingression between unseparated sister chromatids (Fig. 3)
. In addition to this defect, a small but significant fraction of cells exhibited 8N DNA content under these conditions (Figs. 2
and 6A
), suggesting that BMI-1026 may also interfere with as yet uncharacterized cellular processes important for normal cytokinesis. At 200 nM BMI-1026, however, cells with 8N DNA content were not detectable, likely because of predominant mitotic defects and cell death under these conditions.
Treatment of either U-2 OS cells or SP1 cells with BMI-1026 led to apoptosis (Fig. 6, A and D)
. Because BMI-1026 inhibits Cdk1 and induces mitotic failure, we further examined whether BMI-1026 exhibits a differential effect on survival between proliferating cells and already differentiated cells. Our data showed that BMI-1026 significantly decreased the survival of the proliferating mouse keratinocytes compared with that of differentiated cells (Fig. 7)
. Because many human cancers are sensitive to mitotic stress and BMI-1026 can induce mitotic catastrophe and apoptotic cell death, BMI-1026 has the potential to be developed as a novel anti-Cdk inhibitor. In addition, the selective killing effect of BMI-1026 against cells that are actively proliferating or in mitosis could be exploitable in developing BMI-1026 as a potential antitumorigenic therapeutic agent, although additional studies will be required to test this generally. It will be interesting to investigate whether BMI-1026 induces a synergistic effect with previously characterized Cdk inhibitors or with other chemotherapeutic agents. Although we have shown that BMI-1026 is effective in inducing mitotic aberrations and apoptotic cell death in both human and murine tumor cells by itself, further studies in other tumor models in vitro and in vivo seem warranted to enhance our understanding and accelerate the application of this promising chemotherapeutic agent.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by National Cancer Institute Material Cooperative Research and Development Agreement (M-CRADA) No. 01552 with LG Biomedical Institute (San Diego, CA). 
2 Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org).
3 Present address: Department of Biochemistry, College of Medicine, Dankook University San 29, Anseodong, Chunan, Choongchungnamdo, South Korea.
4 To whom requests for reprints should be addressed, at Laboratory of Metabolism, Center for Cancer Research, National Cancer Institute, NIH, 9000 Rockville Pike, Building 37, Room 2D11, Bethesda, MD 20892. Phone: (301) 496-9635; Fax: (301) 496-8419; E-mail: kyunglee{at}pop.nci.nih.gov
5 The abbreviations used are: Cdk, cyclin-dependent kinase; PARP, poly(ADP-ribose) polymerase; GST, glutathione-S-transferase; GSH, reduced glutathione; PKA, cAMP-dependent protein kinase; PKC, protein kinase C; DAPI, 4',6-diamidino-2-phenylindole; BrdU, bromodeoxyuridine; GFP, green fluorescent protein; Erk, extracellular signal-regulated kinase.
Received 6/ 5/03. Revised 8/11/03. Accepted 8/20/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Niiya, X. Xie, K. S. Lee, H. Inoue, and T. Miki Inhibition of Cyclin-dependent Kinase 1 Induces Cytokinesis without Chromosome Segregation in an ECT2 and MgcRacGAP-dependent Manner J. Biol. Chem., October 28, 2005; 280(43): 36502 - 36509. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Scaife Selective and irreversible cell cycle inhibition by diphenyleneiodonium Mol. Cancer Ther., June 1, 2005; 4(6): 876 - 884. [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 |
) treated with BMI-1026 were prepared by continuously culturing the cells in the low-calcium medium supplemented with BMI-1026. Keratinocytes just entering the differentiation process (
) were prepared by transferring the culture from the low-calcium medium to the high-calcium medium supplemented with BMI-1026. Differentiated and growth-arrested keratinocytes (
) were prepared by transferring the cells from the low-calcium medium to the high-calcium medium 18 h before the treatment with BMI-1026 in the high-calcium medium. Twenty-two h after the BMI-1026 treatment, cell viability assays were carried out as described in "Materials and Methods" to determine the percentage of viable cells compared with the control in the same culture conditions. Error Bars, SEM.