In Vitro Antitumor Properties of a Novel Cyclin-Dependent Kinase Inhibitor, P276-00

Home , CDK inhibitor, MAPK1, Protein kinase A

918

In vitro antitumor properties of a novel cyclin-dependent kinase inhibitor, P276-00

Kalpana S. Joshi,1 Maggie J. Rathos,1 Cdk4 and a decrease in Cdk4-specific pRb Ser780 Rajendra D. Joshi,1 Meenakshi Sivakumar,2 phosphorylation was observed. P276-00 produced potent Malcolm Mascarenhas,2 Shrikant Kamble,2 inhibition of Cdk4-D1 activity that was found to be Bansi Lal,1,2 and Somesh Sharma1,2 competitive with ATP and not with retinoblastoma protein. The compound also induced apoptosis in human 1Department of Pharmacology, 2Department of Medicinal promyelocytic leukemia (HL-60) cells, as evidenced by the Chemistry, Nicholas Piramal Research Centre, Nicholas Piramal induction of caspase-3 and DNA ladder studies. These India Limited, 1-Nirlon Complex, Goregaon, Mumbai, India data suggest that P276-00 has the potential to be developed as an anti-Cdk chemotherapeutic agent. [Mol Abstract Cancer Ther 2007;6(3):918–25] Cyclin-dependent kinases (Cdk) and their associated pathways represent some of the most attractive targets Introduction for the development of anticancer therapeutics. Based on antitumor activity in animal models, a variety of Cdk The ability of eukaryotic cells to proliferate in response to a inhibitors are undergoing clinical evaluation either as a growth signal is tightly controlled by a complex network of single agent or in combination with other approved drugs. ordered biochemical events collectively known as the cell In our anticancer drug discovery program, a novel series of cycle (1–3). The cell cycle can be divided into four phases: flavones have been synthesized for evaluation against the G1 (Gap1) during which cellular events are preparing the activity of Cdk4-D1. This enzyme catalyzes the phosphor- cell for DNA synthesis; S (Synthesis) phase in which ylation of retinoblastoma protein, thus inhibiting its DNA replication occurs; G2 (Gap2), preparing the cell for function. We have identified a series of potent Cdk4-D1 division; and M (Mitosis), which is the effective division of the mother cell into two daughter cells (3). Cyclin- inhibitors with IC50 below 250 nmol/L. In this report, we have described the properties of one of the best dependent kinases (Cdk) are critical regulators of cell cycle compound, P276-00 of the flavone’s series. P276-00 progression and are constitutively expressed throughout shows 40-fold selectivity toward Cdk4-D1, compared the cell cycle (3). Expression and interaction of Cdks, with Cdk2-E. The specificity toward 14 other related and cyclins, and their activators and inhibitors, account to a unrelated kinases was also determined. P276-00 was great extent for the timing of events in cell cycle. Specific Cdks operate in distinct phases of the cell cycle. The major found to be more selective with IC50s <100 nmol/L for Cdk4-D1, Cdk1-B, and Cdk9-T1, as compared with other regulatory checkpoint in this process controls the transition Cdks, and less selective for non-Cdk kinases. It showed from G1 to S phase and is characterized by inactivation potent antiproliferative effects against various human and phosphorylation of the retinoblastoma (Rb) gene product (4). Inactivation of Rb leads to the release of the cancer cell lines, with an IC50 ranging from 300 to 800 nmol/L and was further compared for its antiproliferative transcriptional factor E2F1 and the activation of E2F- activity against cancer and normal fibroblast cell lines. responsive genes necessary for progression to S phase P276-00 was found to be highly selective for cancer cells (5–8). Complexed with their respective D-type cyclins, as compared with normal fibroblast cells. To delineate its Cdk4 and Cdk6 are responsible for the progression of the mechanism of action, the effect of P276-00 on cell cycle cells through G1 (9–11). A complex of Cdk2 and cyclin E is proteins was studied in human breast cancer cell line required for the progression of the cells from G1 to S phase. (MCF-7) and human non–small cell lung carcinoma (H- A single, most defining feature of cancer is unchecked 460). A significant down-regulation of cyclin D1 and growth. In malignant cells, altered expression of Cdks and their modulators, including overexpression of cyclins and loss of expression of natural Cdk inhibitors, results in dereg- ulated Cdk activity, providing a selective growth advantage Received 10/4/06; revised 12/4/06; accepted 2/1/07. (12–14). Because of their critical role in cell cycle progres- The costs of publication of this article were defrayed in part by the sion, as well as the association of their activities with the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to processes of differentiation and apoptosis, the Cdks com- indicate this fact. prise an attractive set of targets for novel antineoplastics. Requests for reprints: Kalpana S. Joshi, Department of Pharmacology, Several drug discovery programs have produced potent Nicholas Piramal Research Center, Nicholas Piramal India Limited, small-molecule Cdk inhibitors (15–20). A variety of 1-Nirlon Complex, Goregaon (E), Mumbai 400 063, India. Phone: 91-22-3081-8421; Fax: 91-22-3081-8411; chemical classes, which include purine analogues (21–25), E-mail: [email protected] pyrimidine analogues (26–28), indenopyrazoles (29, 30), Copyright C 2007 American Association for Cancer Research. pyridopyrimidines (31–33), pyrazolopyridines (34, 35), indo- doi:10.1158/1535-7163.MCT-06-0613 locarbazoles (36), pyrrolocarbazoles (37, 38), oxindoles

Mol Cancer Ther 2007;6(3). 2007

(39, 40), and aminothiozoles (41) have been developed as Cdk mmol/L MgCl2, 1 mmol/L EGTA], and then the solution inhibitors. Several compounds that inhibit Cdk activity are was removed by vacuum. With filter plate on vacuum currently in clinical trials, including flavopiridol, R-roscovi- manifold, 50 AL GST-Rb bound to GSH-Sepharose beads in tine (CYC-202), UCN-01 (7-hydroxystaurosporine), and BMS- kinase buffer (0.5 Ag GST-Rb/50 AL) was added to each well, 387032 either as single agents or in combination. Most of and vacuum was applied to the filter plate. About 25 ALofa these compounds, however, inhibit multiple Cdks, with Cdk2 reaction mix containing ATP (cold + hot) and 4Â phospha- being a particularly common target. The present report tase inhibitor mix (40 Amol/L unlabeled ATP, 10 ACi/mL describes a potent inhibitor of Cdk4-D1, Cdk1-B, and Cdk9- g32P-ATP, 40 mmol/L h-glycerophosphate, 4 mmol/L DTT, T1 with f40-fold selectivity over Cdk2-E in in vitro enzyme 0.4 mmol/L NaF, 0.4 mmol/L sodium orthovanadate) assays. Furthermore, we describe the antiproliferative activity diluted in kinase buffer was added to each well. The test of this compound in a broad spectrum of human cancer cell compound (4Â final concentration in kinase buffer) or kinase lines. We also provide evidence of down-regulation of cyclin buffer alone (control) was then added in an additional 25 AL D1 and Cdk4 protein levels with decreased phosphorylation volume. To each well, 50 AL (100 ng) of human Cdk4-D1/ of pRb at Ser780, an event that was preceded by the loss of Cdk2-E enzyme in kinase buffer was added to initiate the Cdk4-D1 activity. The ability of the compound to induce reaction, which was allowed to continue for 30 min at 30jC. apoptosis and the effect on Cdk4-D1 enzyme kinetics were When the reaction was completed, vacuum was applied also studied. again, and the plate was washed with the TNEN buffer [20 mmol/L Tris (pH, 8.0), 100 mmol/L NaCl, 1 mmol/L EDTA, 0.5% nonidet-P40] thrice; the filter plate was air-dried and Materials and Methods was placed in a Multiscreen adapter plate. Packard Micro- Compound Used for the Studies scint-O cocktail (30 AL) was added, and the plate was Compounds P276-00 and flavopiridol were synthesized covered with a Top-Seal A film. Quantitation of 32P-GST-Rb at Nicholas Piramal Research Centre, Nicholas Piramal in 96-well filter plates was carried out by Top Count India Limited, Mumbai, India. scintillation counter (Packard). All compounds were tested Cell Lines and Cell Culture Conditions initially at 1 Amol/L concentration. Compounds showing All human cancer cell lines were obtained either from the more than or equal to 50% inhibition were further profiled American Type Culture Collection (Manassas, VA) or from for IC50 determination. Flavopiridol, a pan-Cdk inhibitor, National Centre for Cell Sciences, Pune, India. Cells were was used as a standard in all experiments. maintained in the appropriate media such as RPMI 1640, Other In vitro Kinase Assays for SpecificityAnalysis DMEM, McCoy’s or F12 K Hams modified medium supple- In vitro kinase assays in the presence of increasing mented with fetal bovine serum, 2 mmol/L L-glutamine, concentrations of P276-00 and appropriate substrates were and penicillin (100 units/mL) and streptomycin (10 units/ carried out using a Kinase Profiler (Protocol Guide) at j mL) in a humidified 37 C incubator with 5% CO2. Upstate (Dundee, United Kingdom). Cdk4-Cyclin D1and Cdk2-Cyclin EEnzyme Assays In vitro Cell Proliferation Assay Human cyclin-dependent kinase-4/2 was coexpressed Exponentially growing cultures of human cancer cell with cyclin D1/E in a baculovirus expression system. To lines were used to assess the effect of compounds on accomplish this, 1 Â 10 7 Sf9 cells were coinfected with proliferation and survival. All the compounds showing V A baculoviruses containing human glutathione S-transferase IC50 1 mol/L in Cdk4-D1 enzyme assay were evaluated (GST)–tagged Cdk-4/2 and cyclin D1/E genes. The cells in the in vitro cell proliferation 3H-thymidine uptake assay. were lysed after 72 h in 500 AL of the lysis buffer [50 mmol/ The cells were seeded at a density of 3,000–5,000 cells per A L HEPES (pH, 7.5), 10 mmol/L MgCl2, 1 mmol/L DTT, well, depending on cell type in 180 L of culture medium 5 Ag/mL of aprotinin, 5 Ag/mL of leupeptin, 0.1 mmol/L in 96-well plate and incubated overnight to allow the NaF, 0.2 mmol/L phenylmethylsulfonyl fluoride, and cells to adhere. Varying concentrations of compounds sodium orthovanadate]. The cell lysate was centrifuged, were added to the wells and incubated for 48 h at 37jC. and the supernatant was purified on a GST-sepharose 3H-thymidine (0.25 ACi) was added to each well, and column. Purity of the proteins was checked by SDS-PAGE incorporation of the radiolabel was allowed to proceed for followed by Western blots using antibodies specific to Cdk4 5 to 7 h. Following this incubation, cells were harvested or Cdk2 (Santa Cruz Biotechnology, Santa Cruz, CA). onto GF/B unifilter plates (Packard) using a Packard GST-retinoblastoma (amino acids 776–928) fusion pro- Filtermate Universal harvester, and the plates were tein was expressed in Escherichia coli and purified by counted in a Packard Top Count 96-well liquid scintillation reduced glutathione (GSH)-Sepharose affinity chromatog- counter. raphy. GST-Rb bound to these beads served as the Immunoprecipitation substrate in the assay. H-460 and MCF-7 cells were treated with 1.5 Amol/L The Cdk4-D1/Cdk2-E enzyme assay was run in 96-well P276-00 for 3, 6, 9, 12, and 24 h. Following this incubation, format using Millipore Multiscreen filtration plates (Uni- the cells were lysed in chilled cell lysis buffer [50 mmol/L filter plates, Packard, Meriden, CT). All assay steps are done HEPES (pH, 7.5), 150 mmol/L NaCl, 1 mmol/L EDTA, in a single filter plate. The filtration wells were prewetted 2.5 mmol/L EGTA, 1 mmol/L DTT, 0.1% Tween with 100 AL of kinase buffer [50 mmol/L HEPES (pH, 7.5), 10 20, 10% glycerol, 0.1 mmol/L phenylmethylsulfonyl

Mol Cancer Ther 2007;6(3). 2007

fluoride, 100 Ag/mL leupeptin, 20 Ag/mL aprotinin, kinase activity as part of the anticancer-screening program. 10 mmol/L h-glycerphosphate, 1 mmol/L NaF, and Several compounds exhibiting below 200 nmol/L in vitro 0.1 mmol/L sodium vanadate]. The cell lysate was cleared anti-Cdk4 activity were selected. Optimization of the leads by centrifugation at 10,000 Â g for 15 min at 4jC. Cdk4 was with structure-activity relationship studies yielded the F immunoprecipitated using specific Cdk4 antibody (Santa most potent compound P276-00 with an IC50 of 63 18.5 Cruz Biotechnology) with protein A-sepharose beads at nmol/L in Cdk4-D1 enzyme assay. This compound was 4jC from 400 Ag protein from cell lysates. The immune then screened for inhibition of Cdk2 activity. The results of complexes were washed thrice with NET-N [20 mmol/L these experiments confirmed that P276-00 exhibited the Tris-HCl (pH, 7.5), 100 mmol/L NaCl, 1 mmol/L EDTA, highest potency and selectivity for Cdk4-D1 in comparison and 0.5% NP40] and twice with kinase buffer. Kinase assay with Cdk2-E (Table 1). For Cdk4-D1 and Cdk2-E enzyme was then done as described in the method above using this assays, flavopiridol is used as standard and showed IC50 of immune complex and the Rb protein as a substrate. 40 F 15 and 150 F 50 nmol/L, respectively. Western Blot Analysis Selectivity of P276-00 against Different Kinases H-460 and MCF-7 cells were treated with P276-00 at To ascertain the selectivity of P276-00 for cyclin-depen- 1.5 Amol/L for 3, 6, 9, 12, and 24 h. HL-60 cells were dent and non-Cdks, it was tested against a panel of Cdks treated with 0.5, 1.0, 2.5, and 5.0 Amol/L concentrations of and kinases other than Cdks by performing in vitro kinase P276-00 for 6 and 24 h. Cells were lysed, and protein assays in the presence of increasing concentrations of P276- content of the lysate was estimated using the Bradford 00 and the appropriate substrates as per the Kinase Profiler reagent. An equal amount of protein was loaded on at Upstate (Table 1). P276-00 exhibited greater selectivity SDS-PAGE, and Western blot analysis was carried out as (IC50s below 100 nmol/L) to Cdk4-D1, Cdk1-B, and Cdk9- described previously (42) using specific antibody to cyclin T1, as compared with other Cdks tested. P276-00 showed 780 D1, Cdk4, pRb, and phospho-pRb Ser antibodies (Santa the highest potency for Cdk9-T1 with an IC50 of 20 nmol/L. Cruz Biotechnology) for H-460 and MCF-7 cell lysate. When compared with other Cdks that are directly involved Similarly, specific antibody to cleaved caspase-3 (Santa in cell cycle progression, it was found to be more selective Cruz Biotechnology) was used for HL-60 cell lysate. for Cdk4-D1 and Cdk1-B inhibition than Cdk2-E and Caspase-3 Cellular Activity Cdk7-H (f40 times specific for Cdk4-D1). In addition, it HL-60 (human promyelocytic leukemia) cells were trea- was not active against many non-Cdk enzymes such as ted with different concentrations of P276-00 for 24 h and mitogen-activated protein kinase 1 (MAPK1), MAPK2, lysed, and the lysate was assayed using the caspase-3 PKCa, PKA, PKC, LCK, and GSK3h and cSRC (Table 1). cellular activity kit (Calbiochem, La Jolla, CA). Potency against HumanTumor Lines DNA Ladder Formation Based on the potency in in vitro Cdk4-D1 enzyme assay, HL-60 cells were treated with different concentrations of P276-00 was selected as the best compound for further P276-00 for 24 and 48 h. Cells were lysed, and DNA was studies. Its antiproliferative effect was also determined in a isolated and subjected to 1.5% agarose gel electrophoresis panel of 12 tumor cell lines in comparison with two normal as described previously (43). fibroblast cell lines using the 3H-thymidine uptake assay. Quantitation of DNA Fragmentation The range of IC50 values for the cell line panel spanned from The amount of DNA fragmentation was quantified using 300 to 800 nmol/L in the 3H-thymidine uptake assay. To the method of Xia et al. (44). This method separates ascertain whether the compound had any selectivity for fragmented and genomic DNA and stains it with DNA normal versus cancer cells, human normal lung fibroblast intercalating fluorescent dye, 4¶-6-diamidino-2-phenylin- WI-38 and MRC-5 were treated with P276-00. The IC50 value dole. Excitation at 365 nm and emission at 454 nm were for normal fibroblast cell lines was significantly higher than read using a Flurometer (POLARstar Optima from BMG, those obtained for human tumor cell lines (Table 2). Labtach GmbH, Offenburg, Germany). Loss of Cdk4 Activity Induced by P276-00 Precedes Enzyme Kinetic Studies Depletion of Cyclin D1, Cdk4, and Rb Hypophosphory- Cdk4-D1 kinase activity was measured by varying the lation concentrations of Rb 200 Ag/mL (1, 2, 4, 6, 8, and 10 AL) or Cdks, their activators, and natural inhibitors are respon- ATP (10, 30, 50, 70, 100, and 150 Amol/L) as substrates in sible for the regulation of the cell cycle. Cell homeostasis is the presence of different concentrations of P276-00 (30, 60, and 90 nmol/L). The Cdk4-D1 assay was carried out as described above. The enzyme inhibitory activity of P276-00 was determined by measuring the phosphorylation of Rb in comparison with the absence of inhibitors.

Results Inhibition of Cdk4-D1and Cdk2-EEnzyme Activity In an effort to generate novel anti-Cdk4 inhibitors, several flavone derivatives (Fig. 1) were evaluated for their in vitro Figure 1. General structure of P276-00 series of flavones.

Mol Cancer Ther 2007;6(3). 2007

Table 1. Activity against a panel of cyclin-dependent and non- Cdks in the presence of compound P276-00

Serial no. Kinase P276-00 IC50 (Amol/L)

1 Cdk4-D1 0.063 F 0.018 2 Cdk2-E 2.540 F 0.409 3 Cdk1-B 0.079 Figure 2. Effect of P276-00 on the Cdk4 enzyme activity in human 4 Cdk2-A 0.224 breast carcinoma (MCF-7) and human non – small cell lungcarcinoma 5 Cdk6-D3 0.396 (H-460) cells by immunoprecipitation. Lane 1, control; lane 2, 1.5 Amol/L 6 Cdk7-H 2.870 P276-00. 7 Cdk9-T1 0.020 8 c-Src tyrosine kinase <10 9 Glycogen synthase kinase-3h 2.771 reduction in cyclin D1, Cdk4, and Rb levels as seen by 10 Lymphocyte-specific protein tyrosine kinase 40 Western blot analysis (Fig. 3A). However, decrease in Rb 780 11 Mitogen-activated protein kinase 1 22 phosphorylation at Ser was seen at 3 h itself (Fig. 3A). 12 Mitogen-activated protein kinase 2 >100 Similar treatment of MCF-7 cells shows a loss in Cdk4-D1 13 Protein kinase Ca 10 enzyme activity at 3 h onward (Fig. 2), whereas protein 13 Protein kinase A 575 levels of cyclin D1 and Cdk4 levels starts decreasing at 6 and 14 Protein kinase C (total) 25 9 h, respectively (Fig. 3B). This was also accompanied by a decrease in phosphorylation of Rb at Ser780 from 6 h onward, followed by reduced Rb levels at 24 h (Fig. 3B). Therefore, maintained as long as these factors are in the right from these results, it seems that the loss of Cdk4-D1 activity proportions. Deregulation of these proteins may lead to at the early time point is not due to the specific decline in cancer. P276-00 was studied for its effect on the regulation of cyclin D1 and Cdk4 levels. Instead, loss of Cdk4 activity levels of cell cycle proteins in human non–small cell lung precedes cyclin D1 and Cdk4 depletion. cancer (H-460) and human breast cancer (MCF-7) cell lines. To address whether P276-00 decreases the abundance of cyclin D1 indirectly because of effects on G1 Cdk function, catalytic Cdk4 enzyme activity was measured from intact H-460 and MCF-7 cell lines exposed to P276-00. Cdk4 kinase activity immunoprecipitated from growing MCF-7 and H-460 cells following exposure to 1.5 Amol/L P276-00 for various times is shown in Fig. 2. Shortly after addition of P276-00 to H-460 cells (3, 6, 9, 12, and 24 h), significant reduction in enzyme activity was seen at 6 h onward (Fig. 2). This decline in Cdk4-D1 activity in H-460 was followed by a

Table 2. Inhibition of proliferation of human cancer cells by P276-00 as determined by thymidine uptake assay

Cell line Cell type 3H-thymidine (IC50) (nmol/L) P276-00

HCT-116 Colon carcinoma 310 U2OS Osteosarcoma 400 H-460 Lung carcinoma 800 HL-60 Promyelocytic 750 HT-29 Colon carcinoma 600 SiHa Cervical carcinoma 420 MCF-7 Breast carcinoma 520 Colo-205 Colon carcinoma 650 SW-480 Colon carcinoma 760 PC-3 Prostate carcinoma 560 Caco2 Colon carcinoma 650 Figure 3. Effect of P276-00 on cell cycle protein levels. A, cyclin D1, Cdk4, pRb, and the phosphorylation status of pRb at Ser780 in human T-24 Bladder carcinoma 390 non – small cell lungcarcinoma cell line (H-460). Lane 1, control; lane 2, WI-38 Normal lung fibroblast 16,500 1.5 Amol/L P276-00. B, cyclin D1, Cdk4, pRb, and the phosphorylation MRC-5 Normal lung fibroblast 11,500 status of pRb at Ser780 in human breast carcinoma cell line (MCF-7). Lane 1, control; lane 2, 1.5 Amol/L P276-00.

Mol Cancer Ther 2007;6(3). 2007

Figure 4. Induction of caspase-3 in human promyelocytic leukemia (HL-60) cells by treatment with 0.5, 0.75, and 1.0 Amol/L of P276-00 for 24 h. Actinomycin served as a positive control. Figure 6. DNA fragmentation as seen by ladder formation in human A Caspase-3 CellularActivity andWestern Blot Analysis promyelocytic leukemia (HL-60) cells after treatment with 1 and 2.5 mol/L P276-00. Lane 1, untreated control; lane 2, 1 Amol/L P276-00, 24 h; lane Caspase-3, an executioner caspase, is activated during 3, 2.5 Amol/L P276-00, 24 h; lane 4, 2.5 Amol/L P276-00, 48 h. apoptotic signaling events by upstream proteases, including caspase-6, caspase-8, and cytotoxic T cell–derived granzyme B. Caspase-3 is one of the principal caspases biochemical hallmark of apoptosis. HL-60 cells treated found in apoptotic cells. Hence, the ability of P276-00 to with 1 Amol/L of P276-00 for 24 h and 2.5 Amol/L for 24 induce caspase-3 activity in HL-60 cells was examined. and 48 h induced apoptosis as evident by the DNA ladder A HL-60 cells were treated with 0.5, 0.75, and 1 mol/L of formation (Fig. 6). P276-00 for 24 h. Actinomycin, a known inducer of Quantitation of DNA Fragmentation apoptosis, was used as a positive control. As compared HL-60 (promyelocytic leukemia) cells treated with 0.25, with untreated control, a dose-dependent increase was seen 0.5, and 0.75 Amol/L of P276-00 for 48 h showed increase in in cellular caspase-3 activity after 24 h of treatment with apoptosis from 30% to 80% at 48 h with increasing P276-00 (Fig. 4). The specific activity of caspase-3 at 0.5, concentration of P276-00 (Fig. 7). The induction of DNA 0.75, and 1 Amol/L of P276-00 were 17.2, 116.5, and 138.2 A fragmentation was also compared with the normal lung pmol/min/ g, respectively. The caspase activity increases fibroblast cell line WI-38 versus the cancerous non–small from 3.6- to 29-fold over the control activity of 4.763 pmol/ A cell lung carcinoma cell line H-460. H-460 cells showed 85% min/ g. Western blot analysis of caspase-3 protein using and 73% apoptosis at 2.5 and 5.0 Amol/L, respectively, as cleaved caspase-3–specific antibody indicates an increase in the levels of cleaved caspase-3 after 24 h of treatment with P276-00 (Fig. 5). DNA Ladder Formation During apoptosis, cells undergo widespread morpho- logic changes (chromatin condensation and cytoplasmic blebbing), which is associated with the incidence of nucleosome excision from chromatin. This endonuclease- mediated nucleosome excision is observed as a DNA ladder (multimers of f180-200 bp) in agarose gels. This characteristic form of DNA degradation is a major

Figure 5. Western blot analysis of caspase-3 following6 and 24 h treatment of human promyelocytic leukemia (HL-60) cells with 0.5 Figure 7. Quantitation of DNA fragmentation in HL-60 following Amol/L (lane 1), 1.0 Amol/L (lane 2), 2.5 Amol/L (lane 3), 5 Amol/L (lane 4) treatment with 0.25, 0.5, and 0.75 Amol/L of P276-00. Actinomycin of P276-00, and C-control (lane 5). was used as a positive control.

Mol Cancer Ther 2007;6(3). 2007

compared with 5% apoptosis in the normal cell line at potent inhibitor of cell proliferation. It was further tested in 5.0 Amol/L of P276-00 (Table 3). These results further cell proliferation assay, i.e., 3H-thymidine uptake assay in a confirm our previous observations that P276-00 is not panel of 12 cancerous cell lines and two normal fibroblast active in the normal fibroblast cell line. cell lines. We were able to show a selective antitumor effect, Kinetics of Inhibition of Cdk4-D1 by P276-00 with an average potency of 568 nmol/L against a panel of In an attempt to study the role of P276-00 with respect to human tumor cell lines, that compared with potency of Cdk4 inhibition, enzyme kinetics was studied. The enzyme almost 5-fold or higher against nonproliferating cells, kinetics (competitive, uncompetitive, or mixed type) was suggesting that a therapeutic window could be achieved determined from inhibition measured against varying because they might have a selective cytotoxic effect on concentrations of one substrate (Rb or ATP) and fixed rapidly proliferating cells over quiescent or slowly prolif- concentration of the other substrate. As seen from the erating normal cells, a property that would make them Lineweaver-Burk plot (45), Vmax remains constant with suitable for use in anticancer therapies. increasing concentration of ATP, indicating that P276-00 With P276-00 being a potent Cdk4-D1 inhibitor, the is a competitive inhibitor with respect to ATP (Fig. 8A). cellular levels of cyclin D1, Cdk4, Rb, and phospho-Rb In the case of Rb at varying concentrations of the inhibitor, Ser780 were investigated. The phosphorylation mediated by all have the same slope, indicating a proportional decrease Cdk4 on Rb protein is required for the cells to progress in the value of Km and Vmax. This indicates that it is an from G1 to S phase in those cells possessing a functional uncompetitive inhibition relative to Rb (Fig. 8B). phospho-Rb. As an indication of P276-00 inhibiting the Cdk4 within the cells, we have determined the phosphor- Discussion ylation status of Rb gene product following exposure to P276-00 in exponentially growing H-460 and MCF-7 cells. It is well established that eukaryotic cell proliferation is a Over a 24-h exposure to P276-00 in both the cell lines, we tightly regulated system controlled by a network of cyclin- found that Cdk4 enzyme activity is inhibited as seen by Cdk complexes (46–48). In epithelial cells, the major immunoprecipitation (Fig. 2). This is also confirmed by regulatory checkpoint in this process is the transition from Western blot, which shows that the Rb protein changed G to the S phase. This transition is characterized by the 1 from mostly hyperphosphorylated pRb at 3 h in H-460 and phosphorylation of pRb, and the Cdk4-D1 enzyme complex at 6 h in MCF-7 cells to a completely dephosphorylated catalyzes the reaction. Because most carcinomas are of form at 24 h (Fig. 3). In this study, we show that P276-00, a epithelial origin, they exhibit a general dependence on potent Cdk inhibitor, induces a decline in cyclin D1 and Cdk4-D1 activity for their proliferation. Inhibition of Cdk4- Cdk4 levels. The depletion of cyclin D1 and Cdk4 is D1 results in the inhibition of cell proliferation as evidenced preceded by the loss of Cdk4-D1 activity and Rb by the experiments using antisense or antibodies to cyclin phosphorylation, thus indicating that the P276-00 may D1 (49–53). These observations suggest that an agent that exert its antiproliferative effect at an early stage through can block the activity of this enzyme complex may be of inhibition of Cdk4-D1 in a competitive manner with ATP. therapeutic value against most carcinomas. This evidence is also provided by enzyme kinetic studies, In this study, we report the results of our efforts to find which show that P276-00 competes with ATP for binding at inhibitors of Cdks and their possible therapeutic utility. We the catalytic site of the Cdk4 enzyme and, thus, inhibiting identified potent compounds from a series of synthetic its kinase activity. Alternatively, inhibition of Cdk9-T1 flavones screened in an in vitro Cdk4-D1 enzyme assay. globally affects cellular transcription with the most The best compound P276-00 was profiled. To establish the profound effects on synthesis of mRNA with short lives, selectivity of P276-00, it was tested against a representative including those encoding growth and apoptosis regulators. panel of serine, threonine, and tyrosine kinases. It potently Effect of Cdk9-T1 may underline the transcriptional inhibited Cdk9-T1 enzyme activity, which is a known repression of cyclin D1 by P276-00 as seen in flavopiridol transcription activator. Other than potent activity IC 50 (54, 55), which needs to be proven. To define the below 100 nmol/L in Cdk4-D1, Cdk1-B, and Cdk9-T1, the mechanism by which P276-00 induces cell death, caspase- compound had little or no activity against other Cdks and 3 cellular activity, DNA ladder formation and quantitation non-Cdks. These in vitro enzyme assay results also of DNA fragmentation were studied. The results showed suggested that the compound P276-00 would be the most that caspase-3 activity increased with increasing concentrations of P276-00. This was further confirmed by demonstrating DNA ladder formation, a biochemical Table 3. Percentage DNA fragmentation in H-460 versus WI-38 hallmark of apoptosis. Western blot analysis also indicated after 48 h of treatment with P276-00 the increase in caspase-3 protein levels as an executioner protein for apoptosis. Apoptotic DNA was quantitated Samples H-460 (%) WI-38 (%) using the fluorescent dye 4¶-6-diamidino-2-phenylindole. Control 9 0 An increase in apoptotic DNA was seen on treatment of the P276-00 (2.5 Amol/L) 85 0 cells with P276-00. These results showed that the anti- P276-00 (5.0 Amol/L) 73 5 proliferative effect of P276-00 is through the induction of apoptosis in tumor cell lines. The enzyme kinetic studies

Mol Cancer Ther 2007;6(3). 2007

Figure 8. Enzyme kinetic studies of Cdk4-D1 inhibition. A, P276-00 inhibition of recombinant Cdk4- cyclin D1: Lineweaver-Burk plot showinginhibition with a competitive characteristic when ATP was used as a substrate. B, P276-00 inhibition of recombinant Cdk4- cyclin D1: Lineweaver-Burk plot showinginhibition with an uncompetitive characteristic when Rb was used as a substrate.

indicate that P276-00 is a competitive inhibitor of Cdk4-D1 illustrated that it is necessary in this process to incorporate enzyme and competes with ATP for the ATP binding site both in vitro enzyme and cell-based proliferation assays to on the enzyme. evaluate compounds for their therapeutic potential. Pre- P276-00, like flavopiridol, belongs to the flavone class of dictive evaluations of potency and efficacy in cell systems compounds. However, unlike flavopiridol, it shows better and/or in animal models are essential adjuncts to primary selectivity toward Cdk1-B, Cdk4-D1, and Cdk9-T1 as screening in the drug discovery process. Compound P276- compared with Cdk7-H and Cdk2-E. It inhibits Cdk4-D1, 00, a novel Cdk inhibitor, is currently in phase I/II clinical Cdk9-T1, and Cdk1-B with almost the same potency as trials. Outcome of clinical trial results will decide the flavopiridol. However, the inhibition of tumor cell growth beneficial effects of Cdk inhibitors. in culture is twice to thrice higher than flavopiridol, which may be attributed to its not being a potent pan-Cdk References inhibitor and, hence, is also less toxic than flavopiridol (56), 1. Morgan DO. Cell cycle control in normal and neoplastic cells. Curr Opin although in comparison to other Cdk inhibitors like Genet Dev 1992;2:33 – 7. kenpaullone, olomoucin, and roscovitine, P276-00 is sub- 2. Pardee AB. G1 events and regulation of cell proliferation. Science stantially more potent in inhibiting tumor cell growth in 1989;246:6038. culture. 3. Pines J. Cyclins and their associated cyclin-dependent kinases in the The results presented in this paper show that Cdks are a human cell cycle. Biochem Soc Trans 1993;21:921 – 5. viable target for anticancer therapy. By focusing on specific 4. Zhou P, JiangW, ZhangYJ, et al. Antisense to cyclin D1 inhibits growth and reverses the transformed phenotype of human esophageal mechanisms of tumorigenesis, we have developed a cancer cells. Oncogene 1995;11:571 – 80. primary screening assay for discovering compounds that 5. Trimarchi JM, Lees JA. Siblingrivalry in the E2F family. Nat Rev Mol may be developed for cancer chemotherapy. We have also Cell Biol 2002;3:11 – 20.

Mol Cancer Ther 2007;6(3). 2007

6. Mundle SD, Saberwal G. Evolvingintricacies and implications of E2F1 effects of PD 0183812. A potent inhibitor of the cyclin D-dependent regulation. FASEB J 2003;17:569 – 74. kinases CDK4 and CDK6. J Biol Chem 2001;276:16617 – 23. 7. Harbour JW, Dean DC. The Rb/E2F pathway: expandingroles and 33. Barvian M, Boschelli DH, Cossrow J, et al. Pyrido-[2,3-d]pyrimidin- emerging paradigms. Genes Dev 2000;14:2393 – 409. 7-one inhibitors of cyclin-dependent kinases. J Med Chem 2000;43: 8. Nevins JR. The Rb/E2F pathway and cancer. Hum Mol Genet 2001;10: 4606 – 16. 699 – 703. 34. Misra RN, Xiao H, Rawlins DB, et al. 1H-pyrazolol[3,4-b] pyridine 9. Chen Q, Lin J, Jinno S, Okayama H. Overexpression of Cdk6-cyclin D3 inhibitors of cyclin-dependent kinases: highly potent 2,6-difluorophenacyl highly sensitizes cells to physical and chemical transformation. Oncogene analogues. Bioorg Med Chem Lett 2003;13:2405 – 8. 2003;22:992 – 1001. 35. Misra RN, Rawlins DB, Xiao HY, et al. 1H-Pyrazolol[3,4-b] pyridine

10. Meterson M, Harlow E. Identification of G1 kinase activity for cdk6, inhibitors of cyclin-dependent kinases. BioorgMed Chem Lett 2003;13: a novel cyclin D partner. Mol Cell Biol 1994;14:2077 – 86. 1133 – 6. 11. Sherr CJ. D-type cyclins. Trends Biochem Sci 1995;20:187 – 90. 36. Zhu G, Conner SE, Zhou X, et al. Synthesis, structure-activity relationship, and biological studies of indolcarbazoles as potent cyclin 12. Hall M, Peters G. Genetic alterations of cyclins, cyclin-dependent D1-CDK4 inhibitors. J Med Chem 2003;46:2027 – 30. kinases, and Cdk inhibitors in human cancer. Adv Cancer Res 1996;68: 67 – 108. 37. Sanchez-Martinez C, Shih C, Faul MM, et al. Aryl [a]pyrrolo[3,4- c]carbazoles as selective cyclin-D1-CDK4 inhibitors. BioorgMed Chem 13. Kaelin WG, Jr. Alterations in G1-S cell-cycle control contributingto carcinogenesis. Ann N Y Acad Sci 1997;833:29 – 33. Lett 2003;13:3835 – 9. Sanchez-Martinez C, Shih C, Zhu G, et al. Studies on cyclin- 14. Bartek J, Lukas J, Batkova J. Perspective: defects in cell cycle control 38. dependent kinase inhibitors: indole-[2,3-a]pyrrolo[3,4-c]carbazoles versus and cancer. J Pathol 1999;187:95 – 9. bis-indolylmaleimides. BioorgMed Chem Lett 2003;13:3841 – 6. 15. Fry DW, Garrett MD. Inhibitors of cyclin-dependent kinases as Dermatakis A, Luk KC, DePinto W. Synthesis of potent oxindole therapeutic agents for the treatment of cancer. Curr Opin Oncol Endocr 39. Meatabol Invest Drugs 2000;2:40 – 59. CDK2 inhibitors. BioorgMed Chem 2003;11:1873 – 81. 40. Bramson HN, Corona J, Davis ST, et al. Oxindole-based inhibitors 16. Shapiro GI. Cyclin-dependent pathways as targets for cancer treatment. J Clin Oncol 2006;24:1770 – 83. of cyclin-dependent kinase 2 (CDK2): design, synthesis, enzymatic activities, and X-ray crystallographic analysis. J Med Chem 2001;44: 17. Toogood PL. Progress toward the development of agents to modulate 4339 – 58. the cell cycle. Curr Opin Chem Biol 2002;6:472 – 8. 41. Kim KS, Kimball SD, Misra RN, et al. Discovery of aminothiazole 18. Knockaert M, Greengard P, Meijer L. Pharmacological inhibitors of inhibitors of cyclin-dependent kinase 2: synthesis, X-ray crystallographic cyclin-dependent kinases. Trends Pharmacol Sci 2002;23:417 – 25. analysis, and biological activities. J Med Chem 2002;45:3905 – 27. 19. Senderowicz AM. Small-molecule cyclin-dependent kinase modula- 42. Lorraine LK, Nicolina E, Hull-Campbell, et al. Characterization of novel tors. Oncogene 2003;22:6609 – 20. inhibitors of cyclin-dependent kinases. Biochem Biophys Res Commun 20. Fischer PM, Endicott J, Meijer L. Cyclin-dependent kinase inhibitors. 1999;260:768 – 74. ProgCell Cycle Res 2003;5:235 – 48. 43. Herrmann M, Lorenz HM, Voll R, Grunke M, Woith W, Kaiden JR. A 21. McClue SJ, Blake D, Clarke R, et al. In vitro and in vivo antitumor rapid and simple method for the isolation of apoptotic DNA fragments. properties of the cyclin dependent kinase inhibitor CYC202 (R-roscovi- Nucleic Acids Res 1994;22:5506 – 7. tine). Int J Cancer 2002;102:463 – 8. 44. Xia Z, Karlsson H, DePierre JW, Lennart. Tricyclic antidepressants 22. Haesslein JL, Jullian N. Recent advances in cyclin-dependent kinase induce apoptosis in human T lymphocytes. Int J Immunopharmacol 1997; inhibition. Purine-based derivatives as anti-cancer agents. Roles and 11:645 – 54. perspectives for the future. Curr Top Med Chem 2002;2:1037 – 50. 45. Nelson D, Cox M. Lehninger principles of bichemistry. 3rd ed. W. H. 23. Davies TG, Bentley J, Arris CE, et al. Structure-based design of a Freeman and Co.; 2000. potent purine-based cyclin-dependent kinase inhibitor. Nat Struct Biol 46. Malumbres M, Barbacid M. To cycle or not to cycle: a decision in 2002;9:745 – 9. cancer. Nat Rev Cancer 2001;1:222 – 31. 24. ChangY-T, Gray NS, Rosania GR, et al. Synthesis and application of 47. Obaya AJ, Sedivy JM. Regulation of cyclin-Cdk activity in mammalian functionally diverse 2,6,9-trisubstituted purine libraries as CDK inhibitors. cells. Cell Mol Life Sci 2002;59:126 – 42. Chem Biol 1999;6:361 – 75. 48. Sherr CJ. Cancer cell cycles. Science 1996;275:1672 – 7. 25. Villerbu N, Gaben AM, Redeuilh G, Mester J. Cellular effects of purvalanol A: aspecific inhibitor of cyclin-dependent kinase activities. Int J 49. Morgan DO, Fisher RP, Espinoza FH, et.al. Control of eukaryotic cell Cancer 2002;97:761 – 9. cycle progression by phosphorylation of cyclin-dependent kinases. Cancer J Sci Am 1998;4:77 – 83. 26. Mesguiche V, Parsons RJ, Arris CE, et al. 4-Alkoxy-2,6-diaminopyr- imidine derivatives: inhibitors of cyclin dependent kinases 1 and 2. Bioorg 50. Morgan DO. Cyclin-dependent kinases: engines, clocks, and micro- Med Chem Lett 2003;13:217 – 22. processors. Annu Rev Cell Dev Biol 1997;13:261 – 91. 27. Breault GA, Ellston RP, Green S, et al. Cyclin-dependent kinase 4 51. Arber N, Doki Y, Han EK, et al. Antisense to cyclin D1 inhibits the inhibitors as a treatment for cancer. Part 2: identification and optimization growth and tumorigenicity of human colon cancer cells. Cancer Res 1997; of substituted 2,4-bis-anilino pyrimidines. BioorgMed Chem Lett 2003; 57:1569 – 74. 13:2961 – 6. 52. Driscoll B, Wu L, Buckley S, et al. Cyclin D1 antisense RNA 28. Beattie JF, Breault GA, Ellston, et al. Cyclin-dependent kinase 4 destabilizes pRb and retards lungcancer cell growth. Am J Physiol inhibitors as a treatment for cancer. Part 1: identification and optimization 1997;273:941 – 9. of substituted 4,6-bis-anilino pyrimidines. BioorgMed Chem Lett 2003; 53. Schrump DS, Chen A, Consoli U. Inhibition of lungcancer proliferation 13:2955 – 60. by antisense cyclin D. Cancer Gene Ther 1996;3:131 – 5. 29. Nugiel DA, Vidwans A, Etzkorn AM, et al. Synthesis and evaluation of 54. Brad C, Tyler L, Sheo S, et al. Down-regulation of cyclin D1 by indenopyrazoles as cyclin-dependent kinase inhibitors: 2. Probingthe transcriptional repression in MCF-7 human breast carcinoma cells induced indeno ringsubstituent pattern. J Med Chem 2002;45:5224 – 32. by Flavopiridol. Cancer Res 1999;59:4634 – 41. 30. Yue EW, Higley CA, DiMeo SV, et al. Synthesis and evaluation of 55. Carlson B, Lahusen T, Singh S, et al. Down-regulation of cyclin D1 by indenopyrazoles as cyclin-dependent kinase inhibitors: 3. Structure transcriptional repression of MCF-7 human breast carcinoma cells induced activity relationships at C3. J Med Chem 2002;45:5233 – 48. by flavopoirdol. Cancer Res 1999;59:4634 – 41. 31. Toogood PL. Cyclin-dependent kinase inhibitors for treating cancer. 56. Kaur G, Stevenson MS, Sebers S, et al. Growth inhibition with Med Res Rev 2001;2:487 – 98. reversible cell cycle arrest of carcinoma cells by flavone L86-8275. J Natl 32. Fry DW, Bedford DC, Harvey PH, et al. Cell cycle and biochemical Cancer Inst 1992;84:1736 – 40.

Mol Cancer Ther 2007;6(3). 2007

Kalpana S. Joshi, Maggie J. Rathos, Rajendra D. Joshi, et al.

Mol Cancer Ther 2007;6:918-925.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/6/3/918

Cited articles This article cites 55 articles, 8 of which you can access for free at: http://mct.aacrjournals.org/content/6/3/918.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/6/3/918.full#related-urls

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

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

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/6/3/918. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.