Cancer Research Targets Metabolism
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cai, D.
Right arrow Articles by Shapiro, G. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cai, D.
Right arrow Articles by Shapiro, G. I.
[Cancer Research 66, 435-444, January 1, 2006]
© 2006 American Association for Cancer Research

Experimental Therapeutics, Molecular Targets, and Chemical Biology

AZ703, an Imidazo[1,2-a]Pyridine Inhibitor of Cyclin-Dependent Kinases 1 and 2, Induces E2F-1-Dependent Apoptosis Enhanced by Depletion of Cyclin-Dependent Kinase 9

Dongpo Cai1,2, Kate F. Byth3 and Geoffrey I. Shapiro1,2

1 Department of Medical Oncology, Dana-Farber Cancer Institute; 2 Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts; and 3 AstraZeneca, Alderley Park, Cheshire, United Kingdom

Requests for reprints: Geoffrey I. Shapiro, Dana-Farber Cancer Institute, Dana 810A, 44 Binney Street, Boston, MA 02115. Phone: 617-632-4942; Fax: 617-632-1977; E-mail: geoffrey_shapiro{at}dfci.harvard.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Preclinical studies were performed of a novel selective imidazopyridine cyclin-dependent kinase (cdk) inhibitor, AZ703. In vitro kinase assays showed that IC50 values for AZ703 against purified cyclin E/cdk2 and cyclin B/cdk1 were 34 and 29 nmol/L, respectively. In contrast, the IC50 against cdk4 was >10 µmol/L. AZ703 also inhibited cdk7 and cdk9 with IC50 values of 2.1 µmol/L and 521 nmol/L, respectively. Treatment of U2OS, NCI-H1299, and A549 cells for 24 hours resulted in growth arrest involving multiple cell cycle phases. At low drug concentrations (<2 µmol/L), G2 arrest predominated, whereas at higher concentrations (≥2 µmol/L), S-G2 arrest was observed. When cells were synchronized in G1 by starvation and released into AZ703, a block in G1 occurred that was not evident in exponentially growing cells. Cell cycle arrest was associated with reduced phosphorylation of the retinoblastoma protein and p27Kip1 at cdk2 phospho-sites. Following longer exposures, apoptosis was evident. Cells were further sensitized to AZ703 following recruitment to S phase by synchronization. Consistent with the inhibition of cdks during S and G2 that modulate the activity and stability of E2F-1, AZ703 treatment induced E2F-1 expression. In U2OS and NCI-H1299 cells engineered to inducibly express the dominant-negative mutant E2F-1 (1-374), expression of the mutant decreased AZ703-mediated apoptosis, indicating dependence on E2F-1 transcriptional targets. AZ703-induced apoptosis in NCI-H1299 cells was enhanced by small interfering RNA–mediated depletion of cdk9, which caused reduced levels of Mcl-1 and XIAP, suggesting that cdk2, cdk1, and cdk9 represent a rational subset of family members for drug targeting. (Cancer Res 2006; 66(1): 435-44)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cyclin-dependent kinases (cdk) are a family of enzymes divided into two groups based on their roles in cell cycle progression and transcriptional regulation (1). Members of the first group comprise core components of the cell cycle machinery and include cyclin D–dependent kinases 4 and 6, as well as cyclin E/cdk2 complexes, which sequentially phosphorylate the retinoblastoma protein (Rb) to facilitate the G1-S transition (2). Cyclin A–dependent kinases 2 and 1 and cyclin B/cdk1 complexes are required for orderly S-phase progression and the G2-M transition, respectively (3). Cdks are regulated by phosphorylation, as well as by their association with cyclins and endogenous Cip/Kip or INK4 inhibitors (4). In malignant cells, the overexpression of cyclins or diminished levels of cdk inhibitors provides a selective growth advantage (5). Ectopic expression of cdk inhibitors in tumor cell lines restores cell cycle control, usually causing G1 and G2 arrest, or in some cases, apoptosis (6). These observations have prompted the development of pharmacologic cdk inhibitors that could potentially produce similar antitumor effects (7).

One of the first generation cdk inhibitors, flavopiridol, causes arrest at both the G1 and G2 phases of the cell cycle of a large proportion of solid tumor cell lines, consistent with inhibition of cdk2, cdk4, cdk6, cdk1, and cdk7 (8). However, cells are sensitized to flavopiridol if they are first recruited to S phase, either by synchronization or by chemotherapy-imposed S-phase delay (9). Following cdk-mediated phosphorylation of Rb, E2F-1 activity is derepressed, and E2F-1 is released so that it can direct transcription of genes required for S phase. However, this transcription is activated only transiently. Orderly S-phase progression requires the down-regulation of E2F-1 activity, accomplished in part by cdk-mediated phosphorylation (10–12). Inhibition of cdk activity during S phase is expected to result in inappropriately persistent E2F-1, causing S-phase delay and apoptosis (13). Consistent with this model, we have recently shown that flavopiridol-induced apoptosis during S phase is E2F-1–dependent (14).

Several cdk holoenzymes have been reported to contribute to E2F-1 phosphorylation during S-G2 and participate in the appropriately timed neutralization of E2F-1 activity. Cyclin A/cdk2 stably interacts with the NH2 terminus of E2F-1 and directs the phosphorylation of both E2F-1 and DP-1, which inhibits the DNA-binding activity of the dimer (10–12, 15). E2F-1 is also phosphorylated by cyclin A/cdk1, which may promote the formation of Rb/E2F-1 complexes, contributing to the turn off of E2F-1 activity late in the cell cycle (16). Finally, cyclin H/cdk7/MAT-1 kinase activity associated with the general transcription factor TFIIH multisubunit protein complex phosphorylates E2F-1, so that it is targeted for degradation (17). Therefore, the inhibition of cdk2, cdk1, and cdk7 by flavopiridol could all contribute to the inappropriately persistent E2F-1 activity critical for the apoptotic response during S phase.

The involvement of multiple cdks in the phosphorylation of critical targets, such as Rb during the G1 phase and E2F-1 during S phase, may explain the recent result in which small interfering RNA (siRNA) used to down-regulate only cdk2 has not induced effects on cell cycle progression or apoptosis in many cell types (18), confirmed by analysis of unsynchronized cdk2–/– mouse embryonic fibroblasts (MEF; refs. 19, 20). During the G1-S transition, cdk4 may substitute for Rb phosphorylation normally mediated by cyclin E/cdk2; during S-G2 progression, cyclin A/cdk1 complexes may have overlapping function and provide compensation. Other experiments designed to down-regulate cdk2 activity have employed ectopic expression of p27Kip1 (21), inhibitory peptides (22), dominant-negative cdk2 mutants (23), or targeted degradation of cyclin A (24). These are all designed to inhibit cdk2 but may also inhibit the activity of other cdks, including cdk1, either directly or indirectly. These approaches have been associated with profound effects on S-G2 progression as well as with apoptosis, suggesting that combined pharmacologic inhibition of both cdk2 and cdk1 may affect the proliferation and viability of transformed cells.

Notably, inducible expression of dominant-negative cdk2 mutants or introduction of cdk2 inhibitory peptides has induced only weak arrest at the G1-S boundary, unmasked only by prior synchronization and release from a nocodazole-induced mitotic block or starvation-induced early G1 arrest (22, 23, 25), a similar phenomenon in cdk2–/– MEFs (19, 20). Therefore, a pharmacologic cdk2/cdk1 inhibitor may be less prone to induce cytostatic G1 arrest than an inhibitor also targeting cdk4/6 and more likely to cause S/G2 delay, modulation of E2F-1 activity, and cytotoxicity.

In addition to their role in cell cycle progression, cyclin/cdk complexes also participate in the regulation of RNA transcription. For example, cyclin H-cdk7/MAT1 and cyclin T/cdk9 (pTEFb) phosphorylate the heptapeptide repeat YSPTSPS found in the COOH-terminal domain (CTD) of RNA polymerase II and control transcriptional initiation and efficient transcriptional elongation, respectively (26–29). Inhibition of transcriptional cdks, also potently achieved by flavopiridol, preferentially depletes cells of mRNAs with short half-lives, including those encoding D cyclins, c-myc, and antiapoptosis family members, such as XIAP and Mcl-1 (30–33). These effects may contribute to flavopiridol-mediated apoptosis in certain cellular contexts.

In the present study, we characterize a novel cdk inhibitor, AZ703, a member of the recently reported cdk inhibitor class of imidazo[1,2-a]pyridines (34, 35). This compound has been reported to selectively inhibit cdk1 and cdk2 over cdk4 as well as an extensive panel of cellular kinases, including epidermal growth factor receptor, insulin-like growth factor receptor, kinase insert domain protein receptor, c-Jun NH2-terminal kinase 1, and protein kinase A (36). Structure-activity relationship data showed that the para-N-alkylsulfonamyl aniline group imparts potency for cdk2 and selectivity over cdk4 (35). In addition, inhibition of cyclin A/cdk2, cyclin E/cdk2, and cyclin B/cdk1 was more potent than that achieved by either flavopiridol or roscovitine (35, 36). Here, we confirm that AZ703 selectively inhibits both cdk1 and cdk2 at low nanomolar concentration in vitro, whereas inhibition of cdk4 requires concentrations ~100-fold higher. Consistent with predictions for a cdk2/cdk1 inhibitor, AZ703 primarily affects S/G2 progression and induces apoptosis in an E2F-dependent manner. AZ703 has intermediate potency against cdk7 and cdk9, although effects on representative transcriptional cdk targets are not immediate and occur at relatively high concentration in cells. However, cells are further sensitized to AZ703 following siRNA-mediated cdk9 depletion, with associated reduction in antiapoptotic proteins. Our results indicate that combined inhibition of cdk2, cdk1, and cdk9 induces cell cycle arrest and apoptosis and suggest that these cdks represent a promising subset against which small molecule inhibitors should be targeted.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Tumor cell lines. U2OS osteosarcoma cells and NCI-H1299 and A549 NSCLC cell lines were obtained from the American Type Culture Collection (Rockville, MD) and maintained in the recommended medium.

Generation of U2OS cells and NCI-H1299 cells inducibly expressing E2F-1 (1-374). NCI-H1299 TetOn cells and the pTRE-E2F1 (1-374) plasmid were gifts from Dr. Andrew Phillips (Medical College of Georgia, Augusta, GA; ref. 37). NCI-H1299 and U2OS (Invitrogen, Carlsbad, CA) TetOn cells were maintained in DMEM supplemented with 10% tetracycline-free fetal bovine serum (FBS; BD Biosciences/Clontech, Palo Alto, CA) and 200 µg/mL G418 and transfected with both pTRE-E2F1 (1-374) and pBabe-Puro plasmids at a 10:1 ratio. Single-cell colonies were selected and amplified in DMEM containing 10% tetracycline-free FBS and 1 µg/mL puromycin. Cells were treated with or without 5 µg/mL doxycycline for 24 hours, and E2F-1 (1-374) protein expression was tested by Western blotting using an anti-E2F1 monoclonal mixture containing both KH20 and KH95 (Upstate Biotechnology, Lake Placid, NY), capable of detecting truncated E2F-1 (1-374).

Generation of NCI-H1299 cells inducibly expressing siRNA targeting cdk9. The NCI-H1299 TetR siRNA starter line was generated by transfecting parental cells with the pcDNA6/TR plasmid followed by selection in 10 µg/mL Blasticidin S HCl. Western blot analysis with an anti-Tet Repressor antibody (MoBiTec, Göttingen, Germany) was used to identify the clone with the highest expression of Tet Repressor. Oligonucleotides containing the siRNA sequence targeting cdk9, GAACCAAAGCTTCCCCCTA, were purchased from Sigma Genosys (The Woodlands, TX) annealed, and ligated into the pSuperior.puro vector (Oligoengine, Seattle, WA) precut with BglII and XhoI. Plasmid insert was sequenced for confirmation, and the pSuperior.puro/cdk9 siRNA vector was transfected into the NCI-H1299 TetR starter line and maintained in DMEM and 10% tetracycline-free FBS followed by selection in 1 µg/mL puromycin. Individual clones were tested for reduced cdk9 expression after addition of 5 µg/mL doxycycline for 72 hours.

Drug treatment. AZ703 was synthesized by AstraZeneca (Cheshire, United Kingdom; ref. 35). A stock solution was prepared in DMSO at a concentration of 10 mmol/L and maintained at –20°C. Drug was diluted in DMSO for working solutions and used at concentrations ranging from 2 nmol/L to 5 µmol/L in kinase assays and from 0.5 to 5 µmol/L for the treatment of cell lines. Cells were synchronized at the G1-S boundary by treatment with 1 mmol/L hydroxyurea (Sigma-Aldrich Co., St. Louis, MO) for 24 hours.

In vitro kinase assays. Cyclin B/cdk1, cyclin E/cdk2, cyclin A/cdk2, and cyclin H/cdk7/MAT1 recombinant kinases were obtained from Upstate Biotechnology. Histone H1 (Roche Applied Science, Indianapolis, IN), GST-Rb, and GST-cdk2 (Santa Cruz Biotechnology, Santa Cruz, CA) and GST-CTD were used as substrates. The plasmid encoding GST-CTD was the gift of Judith Garriga (Temple University, Philadelphia, PA; ref. 38). Antibodies used to recover cdks by immunoprecipitation included anti-cdk2 (M2), anti-cdk4 (C-22), and anti-cdk9 (D-7; Santa Cruz Biotechnology). Recombinant or immunoprecipitated kinases were used in assays performed according to Upstate Biotechnology protocols, or as previously described (39, 40). All kinase assays were done in the presence of 0.5 µmol/L cold ATP, 10 µCi of [{gamma}-33P]ATP, and 2.5 µg of the appropriate substrate. Samples were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes. Phosphorylated substrates were visualized by autoradiography and bands were quantitated using ImageJ Software (NIH). Membranes were subsequently used for Western blotting to demonstrate equal amounts of either recombinant or immunoprecipitated kinases in the assays.

Western blot analysis. Whole-cell and nuclear lysates were prepared as previously described (14, 37) and supplemented with protease and phosphatase inhibitor I and II cocktails (Calbiochem, San Diego, CA). Protein concentrations were determined using the Bradford assay (Bio-Rad, Richmond, CA), and equivalent amounts (10-50 µg) were subjected to SDS-PAGE. Western blotting was done as previously described (14, 40), with the following primary antibodies: anti-cdk1 (C-19), anti-cdk2 (M2), anti-cdk4 (C-22), anti-cdk9 (D-7), anti-E2F-1 (KH95), anti-cyclin D3 (C-16), anti-cyclin H (C-18), anti-p53 (DO-1), and anti-Mcl-1 (S-19), all from Santa Cruz Biotechnology; anti-phospho-specific Rb antibodies from Cell Signaling Technology (Beverly, MA) or Biosource International (Camarillo, CA); anti-Rb, clone G3-245 (PharMingen, San Diego, CA); anti-p27Kip1 (Signal Transduction Laboratories, Lexington, KY); anti-p27Kip1 [pT187] (Zymed Laboratories, South San Francisco, CA); anti-cleaved poly(ADP-ribose) polymerase (PARP) and anti-XIAP, both from Cell Signaling Technology; and anti-tubulin, clone DM 1A (Sigma-Aldrich Co.). Analysis of RNA polymerase II was performed after electrophoresis on 6% polyacrylamide gels with anti-RNA polymerase II [pSer2] (H5) and anti-RNA polymerase II [pSer5] (H14; ref. 41), both from Covance (Berkeley, CA) and designated RNA polymerase II Ser2 and Ser5, respectively. Results were confirmed with anti-phospho-CTD (clone CTD4H8) from Upstate Biotechnology, directed against a synthetic peptide containing 10 repeats of YSPT[pS]PS of the heptapeptide repeat of the CTD, recognizing predominantly phosphorylated RNA polymerase II (33), and designated phospho-CTD. Total RNA polymerase II was analyzed with either anti-RNA polymerase II (C-18) or anti-RNA polymerase II (A-10), both from Santa Cruz Biotechnology.

Fluorescence-activated cell sorting analysis, bromodeoxyuridine analysis, and detection of apoptosis by flow cytometry. These procedures have been previously described (9, 37). For cell cycle analysis, fixed cells were stained with propidium iodide and analyzed for DNA content by flow cytometry using the ModFit program (Verity Software House, Topsham, ME). In bromodeoxyuridine (BrdUrd) experiments, AZ703-treated cells were pulse labeled for 1 hour before fixation, denaturation, and incubation with FITC-conjugated anti-BrdUrd antibody (clone BMG 6H8, Roche Applied Bioscience). Cells were stained with propidium iodide and analyzed by two-color flow cytometry for BrdUrd incorporation and DNA content. For apoptosis assays, a fluorescein apoptosis detection kit was used (Promega, Madison, WI). Following formaldehyde and ethanol fixation of pooled adherent and nonadherent cells, cells were incubated with fluorescein-12 dUTP in the absence or presence of terminal deoxynucleotidyl transferase (TdT), stained with propidium iodide and analyzed for DNA content and apoptosis using two-color flow cytometry. Apoptosis was quantified as the percentage of cells shifting to fluorescein positivity in the presence of TdT.

Cell viability assay. Cells (5 x 103 per well) were seeded on 96-well plates and allowed to grow in the presence or absence of drug for 48 hours. Dehydrogenase activity was determined by using the CCK-8 colorimetric assay (Dojindo, Gaithersburg, MD) in which 10 µL WST-8 substrate was added for 4 hours before absorbance measurement at 450 nm with a microplate reader. Untreated cells were defined as having 100% viability.

Statistical analysis. Statistical analysis was performed with the two-tailed, unpaired Student's t test. P < 0.05 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
AZ703 selectively inhibits cdk2 and cdk1 over cdk4. To confirm the selective inhibition of cdk2 and cdk1 by AZ703, in vitro kinase assays were performed. As shown in Fig. 1A, AZ703 potently inhibits the kinase activity of recombinant cdk2 holoenzymes and cyclin B/cdk1, with IC50 values of 34 and 29 nmol/L, respectively, using histone H1 as substrate. Similar results were obtained in kinase assays using cdk2 recovered from NCI-H1299 NSCLC cells by immunoprecipitation, using Rb as substrate, in which the IC50 was 84 nmol/L. In contrast, there was no effect on the kinase activity of cdk4 immunoprecipitated from NCI-H1299 cells at concentrations of AZ703 as high as 10 µmol/L (Fig. 1B).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 1. AZ703 inhibits cdk2, cdk1, cdk7, and cdk9. A, recombinant cyclin E/cdk2, cyclin A/cdk2, and cyclin B/cdk1 were used to direct the phosphorylation of histone H1 in the absence (DMSO, first lane) or presence of AZ703 at various concentrations, including 2, 5, 10, 25, 50, 100, 250, 500 nmol/L, 1, 3, and 5 µmol/L. Blots were subsequently probed with the appropriate anti-cdk antibody to demonstrate equal amounts of each recombinant kinase across the assays. B, cdk2 and cdk4 were recovered by immunoprecipitation of NCI-H1299 lysates and used to direct the phosphorylation of Rb in the absence (DMSO, first lane) or presence of AZ703. For the cdk2 immunoprecipitation kinase assays, concentrations of AZ703 were 10, 100, 500 nmol/L, 1, 2, and 10 µmol/L. For the cdk4 immunoprecipitation kinase assays, concentrations were 10, 100 nmol/L, 1, and 10 µmol/L. Anti-cdk2 and anti-cdk4 Western blot analyses were performed to demonstrate equal recovery of the respective kinases in immunoprecipitations. C, recombinant cyclin H/cdk7 was used to direct the phosphorylation of GST-cdk2 in the absence (DMSO, first lane) or presence of 100, 250, 500 nmol/L, 1, 1.5, 2, 3, 5, 10, 50, and 100 µmol/L AZ703. Cdk9 was recovered by immunoprecipitation of NCI-H1299 lysates and used to direct phosphorylation of GST-CTD in the absence (DMSO, first lane) or presence of 100, 250, 500 nmol/L, 1, or 10 µmol/L AZ703. Western blot analyses were performed to demonstrate equal amounts of cyclin H and cdk9 in the respective kinase assays.

 
AZ703 has intermediate potency for cdk7 and cdk9. To fully characterize the kinase inhibitory profile of AZ703, its effects on cdk7 and cdk9 activity were also determined in vitro. The cdk-activating kinase activity associated with cyclin H/cdk7 phosphorylates the other cdks in the activation loop of the kinase (at Thr160 in cdk1 and at corresponding sites in cdk2, cdk4, and cdk6). As shown in Fig. 1C, AZ703 inhibits the ability of recombinant cyclin H/cdk7 to phosphorylate cdk2 with an IC50 of 2.1 µmol/L. Cyclin H/cdk7 has a non–cell cycle role as well, and like cyclin T1/cdk9, is a transcriptional cdk that phosphorylates the CTD of RNA polymerase II. AZ703 inhibits the ability of cdk9 recovered by immunoprecipitation from NCI-H1299 cells to phosphorylate a GST-CTD substrate with an IC50 of 521 nmol/L (Fig. 1C).

AZ703 affects cell cycle progression during multiple phases. We examined the effects of a 24-hour exposure of AZ703 on cell cycle progression in U2OS osteosarcoma cells and NCI-H1299 and A549 NSCLC cells (Fig. 2A; data not shown). Previous antiproliferative IC50s reported for AZ703 against multiple tumor cell lines are in the micromolar range (36). These data indicate that at low concentration (1 µmol/L), G2-M arrest predominates following drug exposure. At higher concentrations, there is an increase in S-phase DNA content, so that the arrest is best characterized as S-G2. To further assess the effects of AZ703 on S-phase progression, NCI-H1299 cells were synchronized at the G1-S boundary with hydroxyurea and released into S phase in the absence or presence of drug. Figure 2B shows that S-phase progression is delayed in a concentration-dependent manner over a 10-hour period.



View larger version (46K):
[in this window]
[in a new window]
  		Figure 2. AZ703 blocks cell cycle progression through multiple phases. A, U2OS or NCI-H1299 cells were harvested 24 hours after treatment with DMSO or the indicated concentrations of AZ703 and DNA content was analyzed by flow cytometry. Quantification of a representative experiment (bottom), showing predominantly G2-M arrest after exposure to 1 µmol/L drug and S-G2 arrest at higher concentrations. A concentration dependent increase in G1 DNA content is also observed. B, NCI-H1299 cells were treated with 1 mmol/L hydroxyurea (1H) for 24 hours, inducing a block at the G1-S boundary. Release into DMSO for the indicated times shows S-phase progression over a 10-hour period. Release into AZ703 shows a concentration-dependent slowing of S-phase progression. C, U2OS cells were treated with DMSO or with the indicated concentrations of AZ703 for 24 hours and pulsed with BrdUrd during the last hour of drug exposure. Fluorescein (y-axis) and propidium iodide (x-axis). The percentage of BrdUrd-positive cells declines after AZ703 exposure in a concentration-dependent manner. D, A549 cells were treated with DMSO or 2 µmol/L AZ703 for 24 hours and collected. DNA content was analyzed by flow cytometry, showing AZ703-induced S-G2 arrest, similar to results in other cell lines. Alternatively, following starvation in 0.1% BCS for 72 hours (S), cells were released into DMSO or 2 µmol/L AZ703 for 24 hours, showing a block in G1 progression unmasked by synchronization.

 
Although S-G2 effects predominated in exponentially growing cells, there was a concentration-dependent increase in G1 content (Fig. 2A). Consistent with a block in G1-S progression, BrdUrd synthesis was compromised in U2OS cells in a concentration-dependent manner (Fig. 2C). In addition, A549 cells, synchronized in early G1 by serum starvation, remained arrested in G1 after release in AZ703, while progressing into S phase following release into DMSO (Fig. 2D). These data suggest that the slowing of G1 progression by AZ703 is less pronounced and requires higher concentration than the effects on the S and G2 phases; however, effects on G1 progression are unmasked by prior cell synchronization.

AZ703 affects phosphorylation of cdk targets. Consistent with the inhibition of cdk2 by AZ703, the phosphorylation of both Rb and p27Kip1 was compromised in treated NCI-H1299 cells. Figure 3A shows that the phosphorylation of Rb at Thr821, Thr356, and Ser249/Thr252 was decreased in a concentration-dependent manner, consistent with cdk2 inhibition at low concentration (1 µmol/L) by 24 hours in these cells. In contrast, Rb sites known to be phosphorylated by both cdk2 and cdk4 or cdk4 alone were not affected after exposure to AZ703, suggesting that Rb dephosphorylation is incomplete and correlating with the less pronounced effects of this drug on G1 progression. In addition, cyclin E/cdk2 phosphorylates p27Kip1 at Thr187, targeting the protein for degradation; inhibition of cdk2 is expected to stabilize p27Kip1. As predicted for the effects of a cdk2 inhibitor, there was decreased abundance of phospho-p27Kip1 and relatively stable levels of total p27Kip1 after AZ703 treatment (Fig. 3B).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 3. AZ703 inhibits phosphorylation of cdk2 targets and affects expression of proteins frequently modulated after inhibition of cdk9. A, NCI-H1299 cells were plated; 24 hours after plating (time 0), cells were harvested or treated for an additional 12 or 24 hours with DMSO (D) or AZ703 at the indicated micromolar concentration. Nuclear lysates were subjected to SDS-PAGE and Western blotting with the indicated antibodies, showing reduced phosphorylation of Rb at the Ser249/Thr252, Thr356, and Thr821 Rb phospho-sites. B, NCI-H1299 cells were treated as in (A). Whole-cell lysates were subjected to SDS-PAGE and Western blotting with the indicated antibodies, showing reduced phosphorylation of p27Kip1 at the Thr187 phospho-site, as well as concentration-dependent accumulation of E2F-1 after AZ703 treatment. C, left, NCI-H1299 cells were untreated or treated with the indicated micromolar concentration of AZ703 for 24 hours. Whole-cell lysates were subjected to SDS-PAGE and Western blotting, showing reduced phosphorylation of the RNA polymerase II COOH-terminal domain at the highest concentrations, as well as slightly reduced expression of cyclin D3 and XIAP. Right, the analysis was confirmed in nuclear extracts from NCI-H1299 cells treated with DMSO or the indicated concentrations for 24 hours, again showing that phospho-Ser2 and phospho-Ser5 RNA polymerase II forms are reduced. In both whole-cell and nuclear extracts, total RNA polymerase II levels are largely unchanged; a small amount of phosphorylated RNA polymerase II is identified with the A-10 antibody in nuclear extracts (arrow). Bottom, A549 cells were treated similarly, and whole-cell lysates were subjected to SDS-PAGE and Western blotting, showing concentration-dependent accumulation of p53 in response to AZ703.

 
We also examined the effects of AZ703 on the phosphorylation of RNA polymerase II. Figure 3C shows a decrease in phospho-Ser2 and phospho-Ser5 forms in the absence of changes in total RNA polymerase II levels after 24 hours of drug treatment in NCI-H1299 cells, most evident at the highest concentrations. Cyclin D3 and XIAP expression are similarly decreased, likely targets of cdk9 inhibition. In addition, in A549 cells that express wild-type p53, induction of p53 is evident. This is a common signature of inhibitors of transcription, including inhibitors of cdk9, attributable to a decrease in mdm2 mRNA synthesis and p53 stabilization (42). Therefore, AZ703 affects the phosphorylation of canonical cdk targets consistent with its in vitro cdk inhibitory profile.

Induction of E2F-1 by AZ703. Because AZ703 modulates several cdks that contribute to E2F-1 phosphorylation events that affect its activity and stability, we examined whether E2F-1 is induced following drug treatment. As shown in Fig. 3B, E2F-1 levels are increased in AZ703-treated cells. E2F-3 is abundantly expressed in the cell lines examined, and levels also increased slightly in response to AZ703 (data not shown).

AZ703 induces apoptosis in tumor cell lines. To determine whether S-G2 arrest and E2F-1 induction were events followed by apoptosis, cells were exposed to AZ703 for longer time periods. Treatment of U2OS and NCI-H1299 cells for 48 or 72 hours induced apoptosis, as quantified by TdT-mediated nick-end labeling (TUNEL) assay (Fig. 4A, B, and D), the appearance of sub-G1 DNA content as assessed by flow cytometry (Fig. 4C) and evidence of PARP cleavage (Fig. 4E). In addition, by 72 hours, there was a more pronounced concentration-dependent decrease in levels of phospho-RNA polymerase II, especially phospho-Ser2 RNA polymerase II, accompanied by substantial depletion of the antiapoptotic proteins XIAP and Mcl-1 (Fig. 4E). Induction of E2F-1 was also more marked at the later time point. Interestingly, the degree of apoptosis was greatest after treatment with 3 µmol/L AZ703 and declined slightly after exposure to 5 µmol/L drug, perhaps related to the increased G1 DNA content more apparent in exponentially growing cells at the high concentration (Fig. 2A).



View larger version (43K):
[in this window]
[in a new window]
  		Figure 4. AZ703 induces apoptosis. A, U2OS cells were treated with the indicated concentrations of AZ703 for 24, 48, and 72 hours. % Apoptosis was quantified by TUNEL assay. Compilation of a minimum of three experiments at each condition. Bars, SD. B, NCI-H1299 cells treated with the indicated concentrations of AZ703 for 24 and 48 hours, and % apoptosis was quantified by TUNEL assay. Compilation of a minimum of three experiments at each condition. Bars, SD. At the 48-hour time point in both cell lines, treatment with DMSO resulted in <5% apoptosis. Cell death is maximal after exposure to 3 µmol/l AZ703 and is less substantial in response to 5 µmol/L AZ703; the latter concentration causes more G1 arrest (as shown in Fig. 2). C, U2OS cells treated with either DMSO or the indicated concentration of AZ703 for 72 hours were collected, and DNA content was analyzed by flow cytometry, showing a substantial sub-G1 DNA content. Similar data were obtained for NCI-H1299 cells (data not shown). D, example of TUNEL assay data. NCI-H1299 cells were treated with 2.5 µmol/L AZ703 for the indicated times; cells were collected, fixed, and subjected to TUNEL assay. Fluorescein (x-axis) and propidium iodide (y-axis); top right, G1, S, and G2-M fractions are demarcated. At 24 hours, a small percentage of the population (range, 7-18% over three experiments) is fluorescein positive, with the majority of fluorescein-positive cells (70-86%) derived from the late S-G2 population. Similar early fluorescein shift of the late S-G2 population was also seen in U2OS cells (data not shown). By 72 hours, there is substantial fluorescein shift, with a large percentage of the apoptotic cells with late S and G2-M DNA content; additional fragmentation probably contributes to cell populations with lower DNA content, including sub-G1. E, NCI-H1299 cells were treated with DMSO or 1 to 5 µmol/L AZ703 for 72 hours. Nuclear lysates were subjected to Western blotting with the RNA polymerase II Ser2 and Ser5 phospho-specific antibodies; whole-cell lysates were subjected to Western blotting with the remainder of the indicated antibodies. At this time point, concentration-dependent reduction in phosphorylated RNA polymerase II, most notably the phospho-Ser2 form, is observed with a concomitant decline in levels of Mcl-1 and XIAP. No change in level of total RNA polymerase II was detected in nuclear (data not shown) or whole-cell extracts. E2F-1 continues to accumulate compared with the 24-hour time point (Fig. 3); PARP cleavage is also evident, with the maximal amount of the cleaved product evident after exposure to 3 µmol/L AZ703. F, cells are sensitized to AZ703-induced apoptosis following recruitment to S phase. Left, U2OS cells were treated with PBS or 1 mmol/L hydroxyurea for 24 hours; after a PBS rinse, cells were treated with DMSO or 3 µmol/L AZ703 for an additional 24 hours; after hydroxyurea-mediated recruitment of cells to the G1-S boundary, release into AZ703 induces cell death at an early time point. Right, apoptosis was quantified for the indicated conditions by TUNEL assay. Compilation of a minimum of three experiments for each condition. The difference between apoptosis induced by 3 µmol/L AZ703 following hydroxyurea compared with PBS was statistically significant (P = 0.00058). Bars, SD.

 
In our previous work with the pan-cdk inhibitor flavopiridol, circumvention of G1 arrest by recruitment to S phase sensitized cells and significantly increased apoptotic cytotoxicity (9). Although G1 arrest induced by AZ703 is less pronounced, the data in Fig. 2D indicate that slowing of the G1-S transition does occur. In addition, as shown in Fig. 4D, TUNEL assays in NCI-H1299 cells indicate that cells with S-G2 DNA content begin to undergo apoptosis at an early time point with continued DNA fragmentation over time, suggesting that S-G2 cells are the most sensitive to AZ703. To further assess this issue, apoptosis induced by AZ703 was examined following recruitment of cells to S phase (Fig. 4F). After release from a hydroxyurea-induced block at the G1-S boundary, cells underwent apoptosis over the next 24 hours, so that cell death was readily detected significantly earlier than when exponentially growing cells were treated at the same concentration. Therefore, cells can be sensitized to AZ703-induced apoptosis by prior recruitment to S phase.

AZ703-induced apoptosis is E2F-1 dependent. To determine whether the cell death induced was dependent on the activity of E2F-1, a dominant-negative E2F-1 mutant, E2F-1 (1-374), was inducibly expressed in U2OS and NCI-H1299 cells (Fig. 5A). The E2F-1 (1-374) mutant lacks transactivation activity but is capable of binding to DNA and inhibits wild-type E2F-1 activity by competing for DNA-binding sites (43, 44). In a reporter assay, induction of expression of the E2F-1 (1-374) mutant in these cell lines partially inhibited transcription directed by a promoter containing the E2F-1 binding site linked to luciferase, associated with a slight slowing of G1-S progression (data not shown). Treatment of U2OS cells in the absence of doxycycline with AZ703 for 48 hours induced cell death, as assessed by TUNEL assay; both overall degree of flourescein shift as well as sub-G1 DNA content were significantly reduced in the presence of doxycycline (Fig. 5A). Similar data were obtained with NCI-H1299 cells (data not shown). To insure that subtle effects on G1-S progression upon induction of expression of the E2F-1 (1-374) mutant were not contributing to the decrease in cell death in response to AZ703, the experiments were also done following release from a hydroxyurea-induced block at the G1-S boundary. Release into AZ703 in the absence of doxycycline induced apoptosis, which was sharply reduced in the presence of doxycycline (Fig. 5B).



View larger version (34K):
[in this window]
[in a new window]
  		Figure 5. AZ703-induced apoptosis is E2F-1-dependent. A, NCI-H1299 and U2OS were engineered to inducibly express E2F-1 (1-374), a dominant-negative mutant, in the presence of doxycycline. Cells were untreated or treated with doxycycline for 24 hours before collection and preparation of lysates, which were subjected to Western blotting with an E2F-1 antibody detecting wild-type and truncated forms. B, U2OS cells were grown in the absence or presence of doxycycline for 24 hours followed by treatment with DMSO, 3 or 5 µmol/L AZ703 for 48 hours in the absence or presence or doxycycline. Cells were collected, fixed, and subjected to TUNEL assay. A minimum of three experiments for each condition. For both 3 and 5 µmol/L AZ703, the degree of apoptosis was significantly greater in the absence of doxycycline (P = 0.011 and 0.00038, respectively). Bars, SD. C, TUNEL assays data for a representative experiment in U2OS cells collected after exposure to DMSO or 5 µmol/L AZ703 in the absence or presence of doxycycline. D, NCI-H1299 cells were treated with 1 mmol/L hydroxyurea for 24 hours. At the 16-hour time point, cells continued in hydroxyurea in the absence or presence of doxycycline for an additional 8 hours. After a PBS rinse, cells were then exposed to DMSO, 1 or 3 µmol/L AZ703 for an additional 24 hours in the absence or presence of doxycycline, collected, fixed, and subjected to TUNEL assay. Compilation of two and three experiments, respectively, for treatment with DMSO and 1 µmol/L. A single experiment was done using 3 µmol/L drug. At 1 µmol/L AZ703, the amount of apoptosis induced was significantly greater in the absence of doxycycline (P = 0.0039). Bars, SD.

 
Cells are sensitized to AZ703 following induction of siRNA targeting cdk9. The in vitro kinase assays (Fig. 1) and effects on cdk9 targets (Fig. 4) indicate that AZ703 has moderate potency against the transcriptional cdks. Antiapoptotic proteins, including Mcl-1 and XIAP, are depleted to a small extent after 24 hours (Fig. 3C) or 48 hours (data not shown) at high concentration and more significantly after 72 hours of drug exposure (Fig. 4E). To determine whether apoptosis induced by AZ703 could be accentuated by further depletion of cdk9 activity, we inducibly expressed siRNA targeting cdk9 in NCI-H1299 cells (Fig. 6A). Induction of this siRNA resulted in ~80% reduction in the level of cdk9, but alone had no appreciable effect on cell cycle distribution (Fig. 6B). CTD phosphorylation was slightly decreased, although there was no effect on the expression of proteins encoded by cdk9 mRNA targets, including Mcl-1 and XIAP (Fig. 6D). In the absence of doxycycline, exposure to 3 µmol/L AZ703 for 48 hours results in reduced levels of phospho-CTD, with slight diminution in XIAP and no change in Mcl-1 (Fig. 6D), similar to the data in parental cells. In the presence of doxycycline, depletion of these antiapoptotic proteins is more marked and evident after exposure to a lower concentration (1 µmol/L) of AZ703. Expression of cyclin D3, another likely transcriptional target of cdk9, is depleted in a manner similar to XIAP in the absence and presence of doxycycline. Figure 6B shows that cell death is increased after exposure to 1 µmol/L AZ703 in the presence of doxycycline compared with the absence of doxycycline, as indicated by the increased sub-G1 DNA content, confirmed by quantitation of apoptosis by TUNEL assay. The effect of siRNA-mediated cdk9 depletion was less apparent following exposure to 3 µmol/L AZ703, a concentration of drug that alone reduces levels of cdk9 targets and antiapoptotic proteins. These results were confirmed by assessment of viable cell count after a 48-hour exposure to varying concentrations of drug in the absence or presence of doxycycline, showing that depletion of cdk9-sensitized NCI-H1299 cells to AZ703 (Fig. 6C).



View larger version (29K):
[in this window]
[in a new window]
  		Figure 6. AZ703-induced apoptosis is enhanced by depletion of cdk9. A, NCI-H1299 cells were engineered to inducibly express an siRNA targeting cdk9. Cells were grown for 72 hours in the absence or presence of doxycycline before collection and preparation of lysates, which were subjected to Western blotting, showing ~80% reduction in the level of cdk9 after expression of the siRNA. B, left, cells were grown in the absence or presence of doxycycline for 72 hours, replated, and treated with either DMSO or 1 µmol/L AZ703 for an additional 48 hours in the absence or presence of doxycycline. In the absence of doxycycline, S-G2 arrest occurs; in the presence of doxycycline, cell death occurs as indicated by the sub-G1 peak. A small G1 population is present as well after cdk9 depletion. Right, cells were grown in the absence or presence or doxycycline for 72 hours, replated, and treated with DMSO, 1 or 3 µmol/L AZ703 for an additional 48 hours, with or without doxycycline, and subsequently collected, fixed, and subjected to TUNEL assay. Compilation of three experiments for each condition. Differences in the amount of apoptosis in the absence or presence of doxycycline were not statistically different after treatment with DMSO (P = 0.46) but were statistically significant after treatment with 1 µmol/L (P = 0.00010) or 3 µmol/L AZ703 (P = 0.032). Bars, SD. C, cells were grown as in (B) and treated with the indicated concentrations of AZ703 for 48 hours in the absence or presence of doxycycline followed by CCK-8 assay for assessment of viability. Compilation of six experiments for each condition. Differences in viability in the absence or presence of doxycycline reached statistical significance at 0.5 µmol/L (P = 0.0022), 1 µmol/L (P = 0.0017), 2 µmol/L (P = 0.0027), and 4 µmol/L (P = 0.005) AZ703. Bars, SE. D, lysates were prepared after growth for 72 hours in the absence or presence of doxycycline followed by treatment with 0, 1, or 3 µmol/L AZ703 for 48 hours and subjected to Western blotting with the indicated antibodies. In the absence of doxycycline, there is decreased phospho-CTD and slightly decreased XIAP and cyclin D3 after treatment with 3 µmol/L AZ703; in the presence of doxycycline, more substantial depletion of these proteins is observed even after exposure to 1 µmol/L drug, along with depletion of Mcl-1 as well.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have characterized AZ703, a novel imidazo[1,2-a]pyridine cdk inhibitor that is most selective for cdk2 and cdk1 and far less active against cdk4. Cdk7 and cdk9 are inhibited with intermediate potency. AZ703 has a lower IC50 in cdk2 and cdk1 kinase assays than both flavopiridol and roscovitine; the IC50 against cdk9 lies between that of the other drugs (8, 31, 45, 46). The cellular IC50 of AZ703 against a large panel of tumor cell lines ranges from 0.2 to 1.1 µmol/L (36). Pharmacokinetic analyses after 50 mg/kg oral dosing in the AP rat and the AP mouse indicate Cmax values of 4.4 and 10.1 µmol/L, respectively. However, a high degree of protein binding (>99%) may ultimately limit its clinical use. Nonetheless, AZ703 is representative of dual specificity cdk inhibitors targeting cdk2 and cdk1, and its in vitro characterization has allowed us to probe mechanisms by which such compounds induce apoptosis, while similar imidazo[1,2-a]pyridine cdk inhibitors with superior pharmacokinetic properties are developed (35).

During the first 24 hours of drug treatment, 1 µmol/L AZ703 induces G2-M arrest in both U2OS and NCI-H1299 cells. At higher concentrations (2-3 µmol/L), S-G2 arrest is evident and following a hydroxyurea-induced block at the G1-S boundary, release of cells into AZ703 slows S-phase progression over the first 10 hours of drug exposure. Although a concentration-dependent reduction in BrdUrd incorporation is observed, effects in G1 are best shown following synchronization, in the case of A549 cells after release from a starvation-induced block.

The initial cell cycle effects after exposure to AZ703 are similar to those described after inducible expression of a dominant-negative cdk2 mutant (dn-cdk2) in U2OS cells (23). In exponentially growing cells, low-level expression of dn-cdk2 resulted in G2 arrest; induction of higher levels caused arrest during S-G2. Effects on G1 progression were only observed when cells were synchronized and released from a nocodazole-induced mitotic block. However, these results must be reconciled with the recent demonstration that siRNA-mediated depletion of cdk2 from most malignant cell types has no effect on cell cycle distribution or cellular proliferation (18). One possible explanation is that in cells expressing dn-cdk2, cyclin B/cdk1 activity is also diminished (23). As altered progression through multiple cell cycle phases is probably not caused by inhibition of cdk1 alone,4 it is likely that the S-G2 effects induced by dn-cdk2 or AZ703 are mediated by combined cdk2 and cdk1 inhibition. We are currently confirming this hypothesis with cell lines engineered to inducibly express siRNAs targeting both cdk2 and cdk1 together.

The inhibition of cdk2 by AZ703 results in reduced phosphorylation of Rb, most prominently at the Thr356 and Ser249/Thr252 phospho-sites. Although these sites were initially described as cyclin D/cdk4 sites, the phosphorylation of Rb by baculovirus-produced cyclin E/cdk2 did generate a phosphopeptide spot that could have arisen from the phosphorylation of Thr5, Thr252, or Thr356 (47). Our data suggest that Thr356 and Thr252 are likely phosphorylated by cdk2, and that their phosphorylation is particularly sensitive to AZ703. Similarly, SU9516, a 3-substituted indolinone cdk2 inhibitor, inhibits Thr356 phosphorylation (48). Reduced phosphorylation at another cdk2 phospho-site, Thr821 (47), was detected early (i.e., 12 hours) but did not persist, suggesting that this phosphorylation event may be compensated later by other cdks. In contrast, other Rb phospho-sites, including Ser780 and Ser807/Ser811 have been reported to be solely phosphorylated by cyclin D-/cdk4 (47, 49), and as expected, AZ703 had no effect at these sites. Ser795 is phosphorylated by cdk4, although there is disagreement about whether this site is also phosphorylated by cdk2 (47, 50). In NCI-H1299 cells, AZ703 did not affect Ser795 phosphorylation. In any case, the partial dephosphorylation of Rb observed after 24 hours of AZ703 exposure is consistent with the relatively weak arrest observed at the G1-S boundary compared with effects of the compound during other cell cycle phases.

As predicted for an inhibitor of cyclin E/cdk2, AZ703 also inhibits phosphorylation of p27Kip1 at Thr187. Decreased phosphorylation of p27Kip1 at this site should result in stabilization of the protein (51, 52). Although levels of total p27Kip1 did not decline to the same degree as phospho-p27Kip1, we did not observe accumulation of total p27Kip1 in the cell types examined. Effects of AZ703 on cdk9/pTEFb at high concentration could have affected p27Kip1 transcription (53). Alternatively, cleavage of p27Kip1 by caspases early after the induction of apoptosis may result in the production of an unstable NH2-terminal product, not detected by COOH-terminal anti-p27Kip1 antibodies used in our analysis (54).

Although cell death was not observed in U2OS cells inducibly expressing dn-cdk2 (23), experiments only characterized the effects for 24 hours and may not have been long enough to detect apoptosis. In our experiments, a small amount of apoptosis was detected after 24 hours, with a progressive increase in the percentage of apoptotic cells at 48 and 72 hours after drug exposure. The induction of apoptosis is consistent with results described following the introduction of cdk2 inhibitory peptides into cells (22) or after proteasomal degradation of cyclin A/cdk2 (23). SU9516 has similarly been reported to induce G2 arrest followed by apoptosis (48). Again, antisense or siRNA-mediated depletion of cdk2 alone does not result in apoptosis (18), suggesting that the combined inhibition of cdk2 and cdk1 may be mediating these effects. The cdk2 inhibitory peptides were capable of cdk1 inhibition at high concentration (22), and the proteasomal degradation of cyclin A could indeed affect cyclin A/cdk1 activity as well. SU9516 was reported to have only 1.8-fold selectivity for cdk2 over cdk1 (48). Therefore, in all of these scenarios, inhibition of cdk1 could have played a role in the induction of cell death, as is likely with AZ703.

The data in Fig. 4D indicate that apoptosis is initially detected in cells with S-G2 DNA content, suggesting that S-G2 cells may be the most sensitive to AZ703. As shown in Fig. 4F, cells are sensitized to AZ703 if they are treated following release from a hydroxyurea-induced block at the G1-S boundary. Under these conditions, apoptosis occurred to a significantly greater degree at an earlier time point (i.e., 24 hours) after drug exposure. These results suggest that other conditions that achieve S-phase recruitment, including chemotherapy-imposed S-phase delay, should also sensitize cells to AZ703 treatment. Furthermore, G1 arrest likely protects cells from AZ703-induced apoptosis. As BrdUrd incorporation and slowing of G1 progression are most marked at the highest concentration studied (i.e., 5 µmol/L), this may explain why apoptosis was induced to a greater degree by 3 µmol/L AZ703. Ultimately, at any particular drug concentration, the amount of apoptosis observed may represent a balance between the induction of cell death pathways and competing effects at the G1-S boundary.

The inhibition of cyclin A/cdk1 (16), cyclin A/cdk2 (10–12, 15), and cyclin H/cdk7 (17) are all expected to contribute to inappropriately persistent activity and stability of E2F-1 during the S and G2 phases. As shown in Fig. 3B and Fig. 4E, E2F-1 accumulates in AZ703-treated cells in a dose- and time-dependent manner. Induction of E2F-1 has also been reported with other cdk inhibitors as well, including roscovitine and BMS-387032 (55), and may be a common property among compounds of this class. Apoptosis induced by these agents is likely at least in part E2F-1 dependent and selective for transformed cells (9, 22).

For AZ703, we have shown that cell death is E2F-1 dependent using cells engineered to inducibly express the E2F-1 dominant-negative mutant, E2F-1 (1-374). This mutant lacks the transactivation domain but retains the ability to bind DNA and likely inhibits wild-type E2F-1 activity by competitively binding to E2F sites (43, 44). Induction of this mutant can result in apoptosis due to its ability to inhibit receptor-mediated survival signals (56). However, in U2OS and NCI-H1299 cells, this process occurs at most to a small degree and with significantly slower kinetics than AZ703-mediated apoptosis, so that we were able to examine the effect of E2F-1 (1-374) expression on the response to AZ703 after 24 to 48 hours of drug exposure and to show its ability to abrogate AZ703-induced apoptosis. To insure that effects of the mutant on G1 progression were not accounting for the reduced cell death observed, we also performed experiments after recruitment of cells to S phase. Induction of E2F-1 (1-374) after release from a hydroxyurea-induced block at the G1-S boundary also abrogated AZ703-induced apoptosis. Therefore, it is likely that the E2F-1 protein is playing a primary role in the apoptotic response to AZ703, and that the effects seen are not secondary to altered cell cycle distribution in engineered cells.

As the effects with E2F-1 (1-374) occur in p53-deleted NCI-H1299 cells, the E2F-mediated apoptosis is likely p53 independent, requiring transcriptional transactivation activity. Recently, the p53 homologue p73 has been implicated in p53-independent E2F-1-induced apoptosis (57, 58), and dn-p73 reduced cell death induced by proteasomal degradation of cyclin A/cdk2 (24). It will be of interest to determine whether p73 is also essential for AZ703-mediated cytotoxicity.

We have shown that AZ703 also inhibits PTEFb/cdk9, and to a lesser extent cdk7, both in in vitro kinase assays and in intact cells. Cdk7 and cdk9 regulate transcriptional initiation and elongation, respectively, by phosphorylation of the heptapeptide repeat found in the CTD of RNA polymerase II. Cdk7 phosphorylates the Ser5 position, and cdk9 preferentially phosphorylates the Ser2 position, although an evolving literature suggests that cdk9 also phosphorylates the Ser5 position (26, 41). Inhibition of transcriptional cdks causes depletion of mRNAs of short half-life, often along with their encoded proteins. After 24 or 48 hours (Fig. 3C and Fig. 6D), these effects are evident only at the highest concentrations studied and result in reduced phosphorylation at Ser2 and Ser5 of the RNA polymerase II CTD, with slight reductions in cyclin D3 and XIAP in NCI-H1299 cells and induction of p53 (related to reduced mdm2) in A549 cells. By 72 hours, reduced CTD phosphorylation is seen at lower drug concentrations, with substantial depletion of Mcl-1 and XIAP. Although the effects on cdk9 targets are limited at the time when apoptosis is first detected, the depletion of antiapoptotic proteins might enhance cell death as time elapses. Consistent with this hypothesis, in NCI-H1299 cells engineered to inducibly express an siRNA-targeting cdk9, AZ703 causes depletion of antiapoptotic proteins earlier and at lower concentration, enhancing both the degree of apoptosis observed in TUNEL assay and significantly potentiating loss of viability. Further experiments will be required to confirm that the effects of cdk9 depletion are mediated via Mcl-1 and XIAP depletion, but these are reasonable candidates. Interestingly, in these cells, the synergistic effect of AZ703 and cdk9 depletion also reduces cyclin D3, the primary D cyclin in these cells, resulting in an increased G1 DNA content (Fig. 6B). Despite the increased G1 DNA content, the effect on antiapoptosis proteins seems to predominate, so that overall cell death is enhanced. Our results therefore suggest that combined inhibition of cdk2, cdk1, and cdk9 can produce substantial apoptosis in NCI-H1299 cells. This group comprises a promising subset of the cdk family for anticancer drug targeting.


   Acknowledgments
 
Grant support: AstraZeneca-sponsored Research Fellowship (D. Cai) and NIH grants P20 CA90578 (Dana-Farber/Harvard Cancer Center Specialized Program of Research Excellence in Lung Cancer) and R01 CA90687 (G.I. Shapiro).

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 with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Kathryn Folz-Donahue and Laura Prickett of the Dana-Farber Flow Cytometry Core for technical assistance.


   Footnotes
 
4 D. Cai, V. Latham, and G. Shapiro, unpublished data. Back

Received 5/26/05. Revised 9/25/05. Accepted 10/26/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Sausville EA. Complexities in the development of cyclin-dependent kinase inhibitor drugs. Trends Mol Med 2002;8:S32–7.[CrossRef][Medline]
  2. Sherr CJ. G1 phase progression: cycling on cue. Cell 1994;79:551–5.[CrossRef][Medline]
  3. Pines J. Cyclins: wheels within wheels. Cell Growth Differ 1991;2:305–10.[Medline]
  4. Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 1995;9:1149–63.[Free Full Text]
  5. Hall M, Peters G. Genetic alterations of cyclins, cyclin-dependent kinases, and cdk inhibitors in human cancer. Adv Cancer Res 1996;68:67–108.[Medline]
  6. Shapiro GI, Harper JW. Anticancer drug targets: cell cycle and checkpoint control. J Clin Invest 1999;104:1645–53.[Medline]
  7. Senderowicz AM, Sausville EA. Preclinical and clinical development of cyclin-dependent kinase modulators. J Natl Cancer Inst 2000;92:376–87.[Abstract/Free Full Text]
  8. Sedlacek HH, Czech J, Naik R, et al. Flavopiridol (L86 8275; NSC 649890), a new kinase inhibitor for tumor therapy. Int J Oncol 1996;9:1143–68.
  9. Matranga CB, Shapiro GI. Selective sensitization of transformed cells to flavopiridol-induced apoptosis following recruitment to S-phase. Cancer Res 2002;62:1707–17.[Abstract/Free Full Text]
  10. Krek W, Ewen ME, Shirodkar S, et al. Negative regulation of the growth-promoting transcription factor E2F-1 by a stably bound cyclin A-dependent protein kinase. Cell 1994;78:161–72.[CrossRef][Medline]
  11. Dynlacht BD, Flores O, Lees JA, Harlow E. Differential regulation of E2F transactivation by cyclin-cdk2 complexes. Genes Dev 1994;8:1772–86.[Abstract/Free Full Text]
  12. Xu M, Sheppard KA, Peng CY, Yee AS, Piwinica-Worms H. Cyclin A/Cdk2 binds directly to E2F-1 and inhibits the DNA-binding activity of E2F-1/DP-1 by phosphorylation. Mol Cell Biol 1994;14:8420–31.[Abstract/Free Full Text]
  13. Krek W, Xu G, Livingston DM. Cyclin A-kinase regulation of E2F-1 DNA binding function underlies suppression of an S phase checkpoint. Cell 1995;83:1149–58.[CrossRef][Medline]
  14. Jiang J, Matranga CB, Cai D, et al. Flavopiridol-induced apoptosis during S phase requires E2F-1 and inhibition of cyclin A-dependent kinase activity. Cancer Res 2003;63:7410–22.[Abstract/Free Full Text]
  15. Kitagawa M, Higashi H, Suzuki-Takahashi I, et al. Phosphorylation of E2F-1 by cyclin A-cdk2. Oncogene 1995;10:229–36.[Medline]
  16. Peeper DS, Keblusek P, Helin K, toebes M, van der Eb AJ, Zantema A. Phosphorylation of a specific cdk site in E2F-1 affects its electrophoretic mobility and promotes pRB-binding in vitro. Oncogene 1995;10:39–48.[Medline]
  17. Vandel L, Kouzarides T. Residues phosphorylated by TFIIH are required for E2F-1 degradation during S-phase. EMBO J 1999;18:4280–91.[CrossRef][Medline]
  18. Tetsu O, McCormick F. Proliferation of cancer cells despite cdk2 inhibition. Cancer Cell 2003;3:233–45.[CrossRef][Medline]
  19. Ortega S, Prieto I, Odajima J, et al. Cyclin-dependent kinase 2 is essential for meiosis but not for mitotic cell division in mice. Nat Genet 2003;35:25–31.[CrossRef][Medline]
  20. Berthet C, Aleem E, Coppola V, Tessarollo L, Kaldis P. Cdk2 knockout mice are viable. Curr Biol 2003;13:1775–85.[CrossRef][Medline]
  21. Wang X, Gorospe M, Huang Y, Holbrook NJ. p27Kip1 overexpression causes apoptotic death of mammalian cells. Oncogene 1997;15:2991–7.[CrossRef][Medline]
  22. Chen YNP, Sharma SK, Ramsey TM, et al. Selective killing of transformed cells by cyclin/cyclin-dependent kinase 2 antagonists. Proc Natl Acad Sci U S A 1999;96:4325–9.[Abstract/Free Full Text]
  23. Hu B, Mitra J, van den Heuvel S, Enders GH. S and G2 roles for cdk2 revealed by inducible expression of a dominant-negative mutant in human cells. Mol Cell Biol 2001;21:2755–66.[Abstract/Free Full Text]
  24. Chen W, Lee J, Cho SY, Fine HA. Proteasome-mediated destruction of the cyclin A/cyclin-dependent kinase 2 complex suppresses tumor cell growth in vitro and in vivo. Cancer Res 2004;64:3949–57.[Abstract/Free Full Text]
  25. van den Heuvel S, Harlow E. Distinct roles for cyclin-dependent kinases in cell cycle control. Science 1993;262:2050–4.[Abstract/Free Full Text]
  26. Meinhart A, Kamenski T, Hoeppner S, Baumli S, Cramer P. A structural perspective of CTD function. Genes Dev 2005;19:1401–15.[Abstract/Free Full Text]
  27. Oelgeschlager T. Regulation of RNA polymerase II activity by CTD phosphorylation and cell cycle control. J Cell Physiol 2002;190:160–9.[CrossRef][Medline]
  28. Marshall NF, Peng J, Xie Z, Price DH. Control of RNA polymerase II elongation potential by a novel carboxyl-terminal domain kinase. J Biol Chem 1996;271:27176–83.[Abstract/Free Full Text]
  29. Garriga J, Grana X. Cellular control of gene expression by T-type cyclin/CDK9 complexes. Gene 2004;337:15–23.[CrossRef][Medline]
  30. Kitada S, Zapata JM, Andreeff M, Reed JC. Protein kinase inhibitors flavopiridol and 7-hydroxy-staurosporine down-regulate antiapoptosis proteins in B-cell chronic lymphocytic leukemia. Blood 2000;96:393–7.
  31. Chao SH, Price DH. Flavopiridol inactivates p-TEFb and blocks most RNA polymerase II transcription in vivo. J Biol Chem 2001;276:31793–9.[Abstract/Free Full Text]
  32. Lam LT, Pickeral OK, Peng AC, et al. Genomic-scale measurement of mRNA turnover and the mechanisms of action of the anti-cancer drug flavopiridol. Genome Biol 2001;2:research 0041.1–41.11.
  33. Gojo I, Zhang B, Fenton RG. The cyclin-dependent kinase inhibitor flavopiridol induces apoptosis in multiple myeloma cells through transcriptional repression and down-regulation of Mcl-1. Clin Cancer Res 2002;8:3527–38.[Abstract/Free Full Text]
  34. Byth KF, Cooper N, Culshaw JD, et al. Imidazo[1,2-b]pyridazines: a potent and selective class of cyclin-dependent kinase inhibitors. Bioorg Med Chem Lett 2004;14:2249–52.[CrossRef][Medline]
  35. Byth KF, Culshaw JD, Green S, Oakes SE, Thomas AP. Imidazo[1,2-a]pyridines. Part 2: SAR and optimisation of a potent and selective class of cyclin-dependent kinase inhibitors. Bioorg Med Chem Lett 2004;14:2245–8.[CrossRef][Medline]
  36. Byth K, Green S, Geh C, et al. Novel cell cycle inhibitors: characterization in tumor cell lines and normal cycling cells. Clin Cancer Res 2003;9:C58.
  37. Nakano K, Balint E, Ashcroft M, Vousden KH. A ribonucleotide reductase gene is a transcriptional target of p53 and p73. Oncogene 2000;19:4283–9.[CrossRef][Medline]
  38. Garriga J, Bhattacharya S, Calbo J, et al. CDK9 is constitutively expressed throughout the cell cycle, and its steady-state expression is independent of SKP2. Mol Cell Biol 2003;23:5165–73.[Abstract/Free Full Text]
  39. Shapiro GI, Edwards CD, Ewen ME, Rollins BJ. p16INK4A participates in a G1 arrest checkpoint in response to DNA damage. Mol Cell Biol 1998;18:378–87.[Abstract/Free Full Text]
  40. Shapiro GI, Koestner DA, Matranga CB, Rollins BJ. Flavopiridol induces cell cycle arrest and p53-independent apoptosis in non-small cell lung cancer cell lines. Clin Cancer Res 1999;5:2925–38.[Abstract/Free Full Text]
  41. Palancade B, Bensaude O. Investigating RNA polymerase II carboxyl-terminal domain (CTD) phosphorylation. Eur J Biochem 2003;270:3859–70.[Medline]
  42. Demidenko ZN, Blagosklonny MV. Flavopiridol induces p53 via initial inhibition of Mdm2 and p21 and, independently of p53, sensitizes apoptosis-reluctant cells to tumor necrosis factor. Cancer Res 2004;64:3653–60.[Abstract/Free Full Text]
  43. Hseih JK, Frdersdorf S, Kouzarides T, Martin K, Lu X. E2F-1-induced apoptosis requires DNA binding but not transactivation and is inhibited by the retinoblastoma protein through direct interaction. Genes Dev 1997;11:1840–52.[Abstract/Free Full Text]
  44. Phillips AC, Bates S, Ryan KM, Helin K, Vousden KH. Induction of DNA synthesis and apoptosis are separable functions of E2F-1. Genes Dev 1997;11:1853–63.[Abstract/Free Full Text]
  45. Meijer L, Borgne A, Mulner O, et al. Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5. Eur J Biochem 1997;243:527–36.[Medline]
  46. Wang D, de la Fuente C, Deng L, et al. Inhibition of human immunodeficiency virus type 1 transcription by chemical cyclin-dependent kinase inhibitors. J Virol 2001;75:7266–79.[Abstract/Free Full Text]
  47. Zarkowska T, Mittnacht S. Differential phosphorylation of the retinoblastoma protein by G1/S cyclin-dependent kinases. J Biol Chem 1997;272:12738–46.[Abstract/Free Full Text]
  48. Lane ME, Yu B, Rice A, et al. A novel cdk2-selective inhibitor, SU9516, induces apoptosis in colon carcinoma cells. Cancer Res 2001;61:6170–7.[Abstract/Free Full Text]
  49. Kitagawa M, Higashi H, Jung HK, et al. The consensus motif for phosphorylation by cyclin D1-Cdk4 is different from that for phosphorylation by cyclin A/E-Cdk2. EMBO J 1996;15:7060–9.[Medline]
  50. Connell-Crowley L, Harper JW, Goodrich DW. Cyclin D1/Cdk4 regulates retinoblastoma protein-mediated cell cycle arrest by site-specific phosphorylation. Mol Biol Cell 1997;8:287–301.[Abstract]
  51. Vlach J, Hennecke S, Amati B. Phosphorylation-dependent degradation of the cyclin-dependent kinase inhibitor p27. EMBO J 1997;16:5334–44.[CrossRef][Medline]
  52. Sheaff RJ, Groudine M, Gordon M, Roberts JM, Clurman BE. Cyclin E-cdk2 is a regulator of p27kipl. Genes Dev 1997;11:1464–78.[Abstract/Free Full Text]
  53. Medema RH, Kops GJPL, Bos JL, Burgering BMT. AFX-like forkhead transcription factors mediate cell-cycle regulation by Ras and PKB through p27kip1. Nature 2000;404:782–7.[CrossRef][Medline]
  54. Levkau B, Koyama H, Raines EW, et al. Cleavage of p21Cip1/Waf1 and p27Kip1 mediates apoptosis in endothelial cells through activation of cdk2: role of a caspase cascade. Mol Cell 1998;1:553–63.[CrossRef][Medline]
  55. Ma Y, Freeman SN, Cress WD. E2F4 deficiency promotes drug-induced apoptosis. Cancer Biol Ther 2004;3:1262–9.[Medline]
  56. Phillips A, Ernst M, Bates S, Rice N, Vousden K. E2F-1 potentiates cell death by blocking antiapoptotic signaling pathways. Mol Cell 1999;4:771–81.[CrossRef][Medline]
  57. Lissy NA, Davis PK, Irwin M, Kaelin WG, Dowdy SF. A common E2F-1 and p73 pathway mediates cell death induced by TCR activation. Nature 2000;407:642–5.[CrossRef][Medline]
  58. Irwin M, Marin MC, Phillips AC, et al. Role for the p53 homologue p73 in E2F-1 induced apoptosis. Nature 2000;407:645–8.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Molecular Cancer TherapeuticsHome page
T. Eguchi, H. Itadani, T. Shimomura, N. Kawanishi, H. Hirai, and H. Kotani
Expression levels of p18INK4C modify the cellular efficacy of cyclin-dependent kinase inhibitors via regulation of Mcl-1 expression in tumor cell lines
Mol. Cancer Ther., June 1, 2009; 8(6): 1460 - 1472.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
A Scholz, K Wagner, M Welzel, F Remlinger, B Wiedenmann, G Siemeister, S Rosewicz, and K M Detjen
The oral multitarget tumour growth inhibitor, ZK 304709, inhibits growth of pancreatic neuroendocrine tumours in an orthotopic mouse model
Gut, February 1, 2009; 58(2): 261 - 270.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
K. Bettayeb, O. M. Tirado, S. Marionneau-Lambot, Y. Ferandin, O. Lozach, J. C. Morris, S. Mateo-Lozano, P. Drueckes, C. Schachtele, M. H.G. Kubbutat, et al.
Meriolins, a New Class of Cell Death Inducing Kinase Inhibitors with Enhanced Selectivity for Cyclin-Dependent Kinases
Cancer Res., September 1, 2007; 67(17): 8325 - 8334.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
W. DePinto, X.-J. Chu, X. Yin, M. Smith, K. Packman, P. Goelzer, A. Lovey, Y. Chen, H. Qian, R. Hamid, et al.
In vitro and in vivo activity of R547: a potent and selective cyclin-dependent kinase inhibitor currently in phase I clinical trials.
Mol. Cancer Ther., November 1, 2006; 5(11): 2644 - 2658.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
D. Cai, V. M. Latham Jr., X. Zhang, and G. I. Shapiro
Combined depletion of cell cycle and transcriptional cyclin-dependent kinase activities induces apoptosis in cancer cells.
Cancer Res., September 15, 2006; 66(18): 9270 - 9280.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
G. I. Shapiro
Cyclin-Dependent Kinase Pathways As Targets for Cancer Treatment
J. Clin. Oncol., April 10, 2006; 24(11): 1770 - 1783.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
K. F. Byth, C. Geh, C. L. Forder, S. E. Oakes, and A. P. Thomas
The cellular phenotype of AZ703, a novel selective imidazo[1,2-a]pyridine cyclin-dependent kinase inhibitor.
Mol. Cancer Ther., March 1, 2006; 5(3): 655 - 664.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cai, D.
Right arrow Articles by Shapiro, G. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cai, D.
Right arrow Articles by Shapiro, G. I.

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