Clinical Cancer Research Versailles No Abst 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 Raynaud, F. I.
Right arrow Articles by Workman, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Raynaud, F. I.
Right arrow Articles by Workman, P.
Clinical Cancer Research Vol. 11, 4875-4887, July 1, 2005
© 2005 American Association for Cancer Research

Cancer Therapy: Preclinical

In vitro and In vivo Pharmacokinetic-Pharmacodynamic Relationships for the Trisubstituted Aminopurine Cyclin-Dependent Kinase Inhibitors Olomoucine, Bohemine and CYC202

Florence I. Raynaud1, Steven R. Whittaker1, Peter M. Fischer2, Steven McClue2, Michael I. Walton1, S. Elaine Barrie1, Michelle D. Garrett1, Paul Rogers1, Simon J. Clarke1, Lloyd R. Kelland1, Melanie Valenti1, Lisa Brunton1, Suzanne Eccles1, David P. Lane2 and Paul Workman1

Authors' Affiliations: 1 Cancer Research UK Centre for Cancer Therapeutics at The Institute of Cancer Research, Haddow Laboratories, Belmont, Sutton, United Kingdom and 2 Cyclacel Ltd., James Lindsay Place, Dundee, United Kingdom

Requests for reprints: Paul Workman, Cancer Research UK Centre for Cancer Therapeutics, The Institute of Cancer Research, Haddow Laboratories, 15 Cotswold Road, Sutton, Surrey, SM2 5NG, United Kingdom. Phone: 44-208-722-4301; Fax: 44-208-722-4324; E-mail: paul.workman{at}icr.ac.uk.


   Abstract
Top
 Abstract
Materials and Methods
 Cell culture
 Results
 Discussion
 References
 
Purpose: To investigate pharmacokinetic-pharmacodynamic relationships for the trisubstituted aminopurine cyclin-dependent kinase inhibitors olomoucine, bohemine, and CYC202 (R-roscovitine; seliciclib) in the HCT116 human colon carcinoma model.

Experimental Design: The in vitro activity of the agents was determined in a human tumor panel using the sulforhodamine B assay. The concentration and time dependence was established in HCT116 cells. Molecular biomarkers, including RB phosphorylation and cyclin expression, were assessed by Western blotting. Pharmacokinetic properties were characterized in mice following analysis by liquid chromatography-tandem mass spectrometry. Based on these studies, a dosing regimen was developed for CYC202 that allowed therapeutic exposures in the HCT116 tumor xenograft.

Results: The antitumor potency of the agents in vitro was in the order olomoucine (IC50, 56 µmol/L) < bohemine (IC50, 27 µmol/L) < CYC202 (IC50, 15 µmol/L), corresponding to their activities as cyclin-dependent kinase inhibitors. Antitumor activity increased with exposure time up to 16 hours. The agents caused inhibition of RB and RNA polymerase II phosphorylation and depletion of cyclins. They exhibited relatively rapid clearance following administration to mice. CYC202 displayed the slowest clearance from plasma and the highest tumor uptake, with oral bioavailability of 86%. Oral dosing of CYC202 gave active concentrations in the tumor, modulation of pharmacodynamic markers, and inhibition of tumor growth.

Conclusions: CYC202 showed therapeutic activity on human cancer cell lines in vitro and on xenografts. Pharmacodynamic markers are altered in vitro and in vivo, consistent with the inhibition of cyclin-dependent kinases. Such markers may be potentially useful in the clinical development of CYC202 and other cyclin-dependent kinase inhibitors.

The cell cycle is coordinated through the activities of the cyclin-dependent kinases (CDKs; refs. 1, 2). The CDKs require their partner cyclins for activity and these are expressed in a cell cycle–dependent manner. In addition, regulatory phosphorylation events and binding of the cyclin-dependent kinase inhibitors (CDKI) ensure timely activation or inhibition of the CDK complexes (1, 2). CDKs are deregulated in numerous ways in cancer. Overexpression of cyclin E has been observed in human tumors and is known to result in a poor prognosis in breast cancer (3). Loss of the CDKI proteins such as p16INK4A is found in many malignancies and can predispose to melanoma (4). Therefore, small-molecule pharmacologic CDK inhibitors are being developed to block cell cycle progression and hence to inhibit tumor growth (5, 6). Although the validity of inhibiting CDK2 alone as a cancer drug target has been questioned by recent data (7), other studies showing selective killing of transformed cells by a peptide inhibitor of CDK2/cyclin A binding to E2F-1 have suggested the possibility that CDK2 inhibitors may not only block tumor cell growth but might also preferentially induce apoptosis in tumor cells (8, 9). Furthermore, a number of small-molecule CDK2 inhibitors that are in development also show inhibitory activity towards CDK1 as well as the transcriptional kinases CDK7 and CDK9 (6). Combinatorial inhibition of more than one CDK may provide greater antitumor activity and overcome resistance that may be associated with blockade of, for example, CDK2 alone.

Through an initial screen for CDK1/cyclin B inhibitors, olomoucine was discovered to be a relatively specific and potent inhibitor compared with the structurally similar but more general kinase inhibitor N6-dimethyl-aminopurine (10). Further exploration of the trisubstituted aminopurine structure led to the more potent and selective roscovitine (11). Potency and selectivity for CDK2 kinase inhibition was further improved by synthesis and purification of the R-enantiomer of roscovitine (CYC202; refs. 10, 12–15). Roscovitine and other closely related trisubstituted aminopurine analogues can induce cell cycle arrest and apoptosis in a range of human tumor cell lines (13–16).

The objective of the present work was to extend our previous mechanistic studies (15) and in particular to translate our investigations on the in vitro properties of CYC202 and its analogues into the in vivo animal model setting. We describe the comparative in vitro and in vivo properties of olomoucine, bohemine, and CYC202 (Fig. 1) with particular regard to pharmacokinetic-pharmacodynamic relationships and how this information may be used to support the development of CDK inhibitors. Initially, relative potencies of the analogues against a panel of human cancer cell lines in vitro were determined and the influence of length of compound exposure on in vitro cancer cell proliferation was defined. To show that the compounds were acting in a manner consistent with the proposed mechanism of CDK inhibition, RB phosphorylation was assessed in tumor cells in vitro and the cell cycle effects were determined. In view of the potential importance of the depletion of cyclin D1 and other cyclins (15), the effects of compound treatment on these regulatory proteins was also determined. In parallel, comparative effects on cell cycle and cell death were determined by flow cytometry. Following this in vitro characterization, the pharmacokinetic properties of the three compounds were characterized following administration by the i.v., i.p., and oral routes, and the data were reviewed in relation to the exposures required for in vitro antitumor activity. Comparing the in vitro results and the in vivo pharmacokinetic behavior, CYC202 emerged as the analogue exhibiting the best profile in both respects. Based on the aforementioned properties, a dosing regimen was developed for CYC202, which led to the achievement of antitumor activity in the HCT116 human colon cancer xenograft model. Inhibition of RB phosphorylation and depletion of cyclin D1 was shown in the tumor xenografts, revealing in vivo pharmacokinetic-pharmacodynamic relationships. These pharmacokinetic-pharmacodynamic relationships and other therapeutic studies (13) provide support for the development of CYC202, which is now undergoing phase II clinical trials in cancer patients.



View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Chemical structures of olomoucine, bohemine, and CYC202.

 

   Materials and Methods
Top
 Abstract
 Materials and Methods
Cell culture
 Results
 Discussion
 References
 
Materials. Unless otherwise stated, materials were from Sigma Chemical Ltd. (Poole, Dorset, United Kingdom).

Test compounds. 2-(6-Benzylamino-9-methyl-9H-purin-2-ylamino)-ethanol (olomoucine), (R)-2-(6-benzylamino-9-isopropyl-9H-purin-2-ylamino)butan-1-ol (R-roscovitine; seleciclib; CYC202), and 3-(6-benzylamino-9-isopropyl-9H-purin-2-ylamino)-propan-1-ol (bohemine) were obtained from Cyclacel Ltd. (Dundee, United Kingdom). For structures, see Fig. 1.

In vitro kinase assays. These were carried out by measurement of incorporation of radioactive phosphate from ATP into appropriate polypeptide substrates by purified recombinant human protein kinases and kinase complexes, as described (17). Assays were done using 96-well plates and appropriate assay buffers [typically 25 mmol/L ß-glycerophosphate, 20 mmol/L MOPS, 5 mmol/L EGTA, 1 mmol/L DTT, 1 mmol/L Na3VO3 (pH 7.4)], into which were added 2 to 4 µg of active enzyme with appropriate substrates. The reactions were initiated by addition of Mg/ATP mix (15 mmol/L MgCl2 and 100 µmol/L ATP with 30-50 kBq per well of [{gamma}-32P]-ATP) and mixtures were incubated as required at 30°C. Reactions were stopped on ice followed by filtration through p81 filterplates or GF/C filterplates (Whatman Polyfiltronics, Kent, United Kingdom). After washing thrice with 75 mmol/L aqueous orthophosphoric acid, plates were dried, scintillant added, and incorporated radioactivity measured in a scintillation counter (TopCount, Packard Instruments, Pangbourne, Berks, United Kingdom). Compounds for kinase assay were made up as 10 mmol/L stocks in DMSO and diluted into 10% DMSO in assay buffer. Data were analyzed using curve-fitting software (GraphPad Prism version 3.00 for Windows, GraphPad Software, Inc., San Diego, CA) to determine IC50 values (concentration of test compound which inhibits kinase activity by 50%).


   Cell culture
Top
 Abstract
 Materials and Methods
 Cell culture
Results
 Discussion
 References
 
Human tumor cell lines, including the HCT116 (National Cancer Institute, Bethesda, MD) human colon carcinoma, were grown in DMEM (Invitrogen, Paisley, United Kingdom) supplemented with 10% fetal bovine serum (Invitrogen, Paisley, United Kingdom) in an atmosphere of 5% CO2. A matched pair of HCT116 cells isogenic for TP53 was kindly provided by Professor Bert Vogelstein, The Johns Hopkins University School of Medicine, Baltimore MD (18). For drug treatment, 2 x 106 cells were seeded into a T75 flask (Corning, Acton, MA) and left to attach for 36 hours. Compounds were dissolved in DMSO and diluted directly into the culture media when required. The total concentration of DMSO in the medium did not exceed 0.4% (v/v) during treatments.

Sulforhodamine B assay. Drugs were made up at 20 to 50 mmol/L in DMSO. Growth inhibition assays were done in 96-well microtiter plates using the sulforhodamine B assay as described previously (19). Briefly, cells were seeded at 3 to 8 x 103 cells per well (dependent upon doubling time) in 160 µL growth medium and allowed to attach overnight. Agents were then added at 10 different concentrations (typically from 2.5 nmol/L to 100 µmol/L) in quadruplicate wells and incubated at 37°C for 96 hours. Where the effect of duration of exposure was investigated, compounds were added for the times indicated and washed off and the cells incubated in fresh growth medium for the remainder of the 96-hour period. Fifty microliters of sulforhodamine B (0.4%, w/v) dissolved in 1% acetic acid were then added and the absorbance at 492 nm was determined using a Multiscan MCC/340 MKII (Titertek, Huntsville, AL). Absorbance of treated wells was expressed as a percentage of control wells. IC50 levels were calculated graphically and mean IC50 levels derived. Results are the mean of at least three determinations.

Animals. Female BALB/c– mice were maintained on SDA Expanded Rodent diet (Harlan UK Ltd., Bicester, Oxon, United Kingdom) and water ad libitum. They were housed in an Individual Ventilated Caging System manufactured by Thoren. Female CD1 nude mice were supplied by Charles River UK Ltd. (Martgate, United Kingdom) and maintained on Harlan Teklad 9607 R&M diet and water ad libitum. All animal procedures complied with local and national animal welfare guidelines (20).

Maximum tolerated dose. BALB/c– mice were given increasing doses of 50, 100, and 200 mg/kg of the three compounds i.v. in 50 mmol/L HCl/saline. Animals were checked daily on several occasions and monitored for a period of 5 days. For oral administration BALB/c– mice were given a single dose of 500 or 2,000 mg/kg orally in 50 mmol/L HCl/saline.

Pharmacokinetic experiments: Pharmacokinetics of olomoucine, bohemine, and CYC202 iv. Compounds were given i.v. in a volume of 0.1 mL/10 g bodyweight in 50 mmol/L HCl/saline at 50 mg/kg to BALB/c– mice bearing the s.c. colon 26 murine tumor. A 1-mm3 brei was implanted under anesthetic using a Bashford syringe into the flank of 100 BALB/c– female mice. Fourteen days post-implantation, mice bearing comparably sized tumors (diameters of ~6 mm) were randomized into either a control group of six mice or treatment groups of three mice per time point. Plasma, liver, kidney, spleen, and tumor were collected at 0.25, 0.5, 1, 3, 6, and 24 hours after administration.

Pharmacokinetics of CYC202 i.p., i.v., and orally. CYC202 at a dose of 50 mg/kg (0.1 mL/10 g, 5 mg/mL) in 9% (w/v) Lutrol (polyethylene glycol 660-12 hydroxystearate), 3.5% (w/v) Solutol HS15 (BASF), and 87% water, was given i.v., i.p., and orally to BALB/c– mice. Solutol was weighed and warmed at 70°C until liquid. Lutrol was weighed and added to the Solutol. CYC202 was added and the solution kept at 70°C until dissolved. The appropriate volume of sterile water was then added to the mixture. Blood was collected after 5, 15, 30 minutes, 1, 2, 4, 6, and 24 hours after administration. The Lutrol/Solutol vehicle was used because HCl/saline was unsuitable for the i.p. route and because a drug solution was required for i.v. administration. CYC202 (50, 500, and 2,000 mg/kg) was given orally in 50 mmol/L HCl/saline in a volume of 0.1 mL/10 g body weight to female BALB/c– mice. Blood was collected at 5, 15, 30 minutes, 1, 2, 4, 6, and 24 hours after administration.

Pharmacokinetic calculations and simulation experiments. Pharmacokinetic variables were calculated using Winnonlin Professional software (21), version 3.2 (Pharsight, Mountain View, CA). Variables presented were derived from noncompartmental analysis. Model 200 (extravascular administration) was used for plasma following i.p. and oral administration and for tissues following i.v. administration. Model 201 (i.v. bolus model) was used for plasma following i.v. administration. Area under the concentration versus time curve to the last time point (AUC), Cmax (maximum concentration observed), Cl (clearance), t1/2 (half-life), and Vz (volume of distribution based on the terminal phase) were evaluated. For simulation studies, variables derived from one-compartment analysis following oral administration of 50 mg/kg were used.

Protein binding experiments. Fresh human or BALB/c– mouse plasma was incubated at 37°C for 5 and 30 minutes, 2 and 6 hours with 2 and 20 µmol/L CYC202. Plasma and standard curve samples in PBS were then ultrafiltered by centrifugation for 40 minutes at 4°C at 1,500 x g using 10,000 MW exclusion membranes (Amicon, Dorset, United Kingdom). Samples were subsequently frozen at –20°C until analysis. Control experiments were carried out to show that test compound did not bind to the filter membranes.

Analytic method. Drug measurements were done by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Initially, a 150 mm x 2.1 mm zwitterionic ABZ+ column (Supelco, Poole, Dorset, United Kingdom) was used and drugs eluted with 10 mmol/L ammonium acetate and 70% methanol at a flow rate of 0.2 mL/min; total run time was generally 3 minutes. Tissue homogenates and plasma were treated with 3 volumes of methanol to precipitate protein followed by dichloromethane extraction. Standard curves were made in the appropriate matrix and analyzed at the level of 1, 10, 100, 1,000, 10,000, and 100,000 ng/mL. Samples were reconstituted in 200 µL mobile phase and 10 µL injected onto the column. Although this analytic method was specific, sensitive, and reproducible, it was found that the column could not withstand a large number of extracts. Therefore, a second method was used for subsequent routine CYC202 analysis, as follows. In this method, analyses were done on a 50 mm x 4.6 mm zwitterionic ABZ+ column (Supelco) and the drugs eluted with a gradient of 80% to 0.1% formic acid, 20% to 100% methanol over 5 minutes. Standard curves were made in mouse plasma at the level of 10, 100, 1,000, 10,000, and 100,000 nmol/L. Quality controls were made in control mouse plasma. The assay was validated in mouse plasma by running three precision batches of five replicates at the levels of 25, 90, 500, and 50,000 nmol/L on three separate days. Plasma (100 µL aliquots) was added to 30 µl of internal standard (500 ng/mL olomoucine in methanol) incubated for 30 minutes and treated with 3 volumes of methanol. Following centrifugation, the supernatant was transferred to high-performance liquid chromatography vials and 10 µL injected onto the column. For ultrafiltrate analysis, standard curve and quality controls were made in plasma ultrafiltrates.

Detection was achieved by multiple reaction monitoring on a triple sector mass spectrometer (TSQ700, Thermoquest Ltd., Hemel Hempstead, Herts, United Kingdom). Multiple reaction monitoring of the sum of two daughter ions of the pseudomolecular ion [M+H]+ 299 (99 and 177 AMU) for olomoucine, 355 (99 and 233 AMU) for CYC202, and 341 (91 and 206) for bohemine. Peak areas were monitored and plotted against concentration (GraphPad Software). When internal standard was used, relative areas were monitored.

Pharmacokinetic-pharmacodynamic relationships in human tumor xenografts. Nude mice bearing established (~130 mm3) HCT116 human colon tumors as a s.c. xenograft in the flank were given 500 mg/kg CYC202 orally, twice daily for 5 days. This regimen was based on the pharmacokinetic simulation models. CYC202 was dissolved in 10% DMSO, 5% Tween 20, and 85% of 50 mmol/L HCl/saline. For pharmacodynamic profiling, three mice were sacrificed per time point. Time points comprised 3 days vehicle control, 1 day of treatment (harvested 4 hours after the second dose), 3 days treated (again 4 hours after the second dose), and 5 days treated (also 4 hours after the second dose). Plasma and tumor were recovered from each animal and immediately either homogenized in cold lysis buffer or snap-frozen in liquid nitrogen and stored at –80°C. For antitumor therapy studies, 10 mice were treated with vehicle control alone and eight mice were treated with CYC202 for 5 days, using the same protocol as described above. Tumor volume was determined from measurement of two orthodiagonal diameters and antitumor activity was determined by the comparison of volumes for treated and control groups. The difference in the means of paired samples was determined using a two-tailed Student's t test and Prism software (GraphPad Software) and Ps < 0.05 were considered statistically significant.

Western blotting. To harvest cells, the medium was removed and cells were incubated with 5 mL trypsin for 5 minutes at 37°C to detach them from the plastic. The cells were then pelleted, washed in ice-cold PBS, and resuspended in ice-cold lysis buffer containing 50 mmol/L HEPES (pH 7.4), 250 mmol/L NaCl, 0.1% NP40, 1 mmol/L DTT, 1 mmol/L EDTA, 1 mmol/L NaF, 10 mmol/L ß-glycerophosphate, 0.1 mmol/L sodium orthovanadate, and one complete protease inhibitor cocktail tablet (Roche, East Sussex, United Kingdom) per 10 mL of lysis buffer for 30 minutes on ice. Lysates were centrifuged at ~18,000 x g for 10 minutes at 4°C to remove cellular debris. The supernatant was stored at –80°C before use. The protein concentration of lysates was determined using the protein assay reagent; bicinchoninic acid protein assay (Pierce, Rockford, IL). Proteins were separated by SDS-PAGE using Novex precast tris-glycine gels (Invitrogen, Groningen, the Netherlands) and transferred to Immobilon-P membranes (Millipore, Bedford, MA). Membranes were blocked for 1 hour in TBSTM [50 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 0.1% Tween 20 (Sigma Chemical)] and 3% milk. Immunoblotting with primary antibodies diluted in TBSTM was done at 4°C overnight followed by a 1-hour incubation with horseradish peroxidase–conjugated secondary antibodies at room temperature. Membranes were washed with enhanced chemiluminescence reagents and exposed to Hyperfilm (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom). Antibodies used were phospho-RB Ser780 1:5,000, phospho-ERK1/2 1:1,000, phospho-CDK2 Thr160 1:1,000 (Cell Signaling Technologies, Beverly, MA); total RB SC-50 1:2,000 (Santa Cruz Biotechnology, Santa Cruz, CA); phospho-RB Ser608 (22) 1:2,000 (Dr. Sibylle Mittnacht, Institute of Cancer Research, London, United Kingdom); nonphosphorylated RB Ser608 (underphosphorylated RB) 1:1,000 (PharMingen, San Diego, CA); phospho-RB Ser807/811 1:5,000 (Sigma Chemical); phospho-RB Thr821 1:1,000 (Biosource, Nivelles, Belgium); cyclin B1 Ab-1 1:200, cyclin D1 Ab-1 1:200, cyclin A Ab-6 1:200, cyclin E Ab-1 1:200, CDK1 Ab-1 1:200, CDK2 Ab-4 1:200, CDK4 1:200 Ab-1 (Neomarkers, Fremont, CA); poly(ADP-ribose) polymerase C-2-10 1:1000 (BD Biosciences, Oxford, United Kingdom); total RNA polymerase II AB817 1:5000 (Abcam, Cambridge, United Kingdom); total glyceraldehyde-3-phosphate dehydrogenase 1:5000 (Chemicon, Temecula, CA); goat anti-rabbit and goat anti-mouse horseradish peroxidase–conjugated secondary antibodies 1:5,000 (Bio-Rad, Hercules, CA). Western blots were quantified by densitometry with Image Quant (Amersham Biosciences). Using a high-throughput 96 well plate-based "in-cell western" approach, RB phosphorylation was assessed in intact cells as described previously (23).

Flow cytometry. HCT116 cells (2.5 x 105) were seeded into a T25 flask and left for 36 hours to attach to the plastic. Drug treatments were done as above. Cells were harvested in 1 mL trypsin-versene by incubating at 37°C for 5 minutes to form a cell suspension, gently pelleted and resuspended in 100 µL ice-cold PBS, and fixed by slow addition of 1 mL ice-cold 70% ethanol while vortexing. Cells were resuspended in 200 µL of 0.02 mg/mL propidium iodide (Molecular Probes, Cambridge, United Kingdom)/0.25 mg/mL RNase A (Sigma Chemical) and incubated at 37°C for 30 minutes. Samples were analyzed on a Beckman Coulter Elite ESP (Beckman Coulter, High Wycombe, United Kingdom) and cell cycle analysis was done with WinMidi2.8 software (Scripps Research Institute, La Jolla, CA).


   Results
Top
 Abstract
 Materials and Methods
 Cell culture
 Results
Discussion
 References
 
Activity against recombinant kinases in vitro. The ability of the trisubstituted aminopurine CDK inhibitors olomoucine, bohemine, and CYC202 to inhibit recombinant kinases was assessed and the results are reported in Table 1. CYC202 was clearly the most potent inhibitor against all kinases tested. The kinase most potently inhibited by CYC202 was CDK2/cyclin E (IC50, 0.13 µmol/L), closely followed by CDK7/cyclin H (IC50, 0.46 µmol/L) and CDK9/cyclin T1 (IC50, 0.78 µmol/L). The inhibition of CDK1, CDK4, CDK6, and extracellular signal–related kinase 2 (ERK2) was much less.


View this table:
[in this window]
[in a new window]

  		Table 1. Inhibition of recombinant or purified kinases in vitro by olomoucine, bohemine, and CYC202

 
Activity against human tumor lines in cell culture. Table 2 shows the activities of the three analogues against a range of human tumor cell lines. Mean IC50 values across the panel were 56 µmol/L for olomoucine, 27 µmol/L for bohemine, and 15 µmol/L for CYC202. The mean value for CYC202 was identical to that of 15 µmol/L found in an independent panel (13). These relative potencies for the three aminopurines correlated with the increased potency of CYC202 over bohemine and olomoucine observed in CDK inhibition assays (Table 1). Bohemine and CYC202 showed a broad spectrum of in vitro anticancer activity, which was not directed towards any particular tumor cell type.


View this table:
[in this window]
[in a new window]

  		Table 2. In vitro growth inhibitory activity of olomoucine, bohemine, and CYC202 in a panel of cancer cell lines

 
Included within the cell line panel were sublines possessing acquired drug resistance to cisplatin (CisR lines; ref. 24) and doxorubicin (DoxR line, with resistance mediated by overexpression of P-glycoprotein; ref. 25). In these paired cell lines, no cross-resistance was observed to the CDK inhibitors. Also of interest is the possible influence of RB and TP53 status in determining cellular sensitivity to CDK inhibitors. A comparison of sensitivity of the widely used pair of human osteosarcoma lines (SAOS-2, RB negative and U2-OS, RB positive; ref. 26) suggests little or no effect of RB status on sensitivity. In contrast, when comparing two breast cancer cell lines which differ in TP53 status (MCF-7, wild-type TP53 and MDA-MB-231, mutant TP53; ref. 27) the mutant cell line was ~2-fold more resistant than the wild-type line. Previous studies in a human cancer cell line panel showed a similar modest difference (13). However, the sensitivity to CYC202 of an isogenic variant of the HCT116 human colon carcinoma cell line in which TP53 was knocked out by homologous recombination (HCT116 TP53–/–; ref. 18) was essentially no different from that of the vector control (HCT116 TP53+/+) or parental cell line (Fig. 2A).



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. A, effect of TP53 status on sensitivity to CYC202 in HCT116 human colon carcinoma cells. HCT116 parental cells (PAR, WT TP53), vector control (WT, WT TP53), or TP53 knock out (NULL, TP53 null) were exposed to increasing doses of CYC202 for 96 hours. Cell growth was assayed using the sulforhodamine B assay and IC50 values were calculated graphically. Columns, mean (n = 3); bars, ±SD. Statistical analysis of the mean IC50 values by a one-way ANOVA test showed no significant differences (P > 0.05). B, effect of time of exposure on response to CDKIs. HCT116 cells in logarithmic growth were exposed to increasing concentrations of either bohemine or CYC202 for 4, 8, 16, 24, and 96 hours, the compounds were washed off and cell growth assayed by sulforhodamine B after a total of 96 hours. Points, mean IC50 values (n = 3); bars, ±SD. The IC50 following a 4-hour exposure to bohemine was >200 µmol/L, as indicated by the vertical arrow.

 
Time of exposure required for in vitro activity. The effect of varying times of exposure to CYC202 and bohemine was investigated using the HCT116 parental human colon cancer cell line in monolayer culture. Following the times of exposure shown in Fig. 2B, the compounds were washed off and cell growth inhibition determined at 96 hours using the sulforhodamine B assay. Between 8 and 16 hours of drug exposure was required to achieve the maximum growth inhibitory effect. Similar data were obtained for the human ovarian carcinoma cell line A2780 when exposed to olomoucine, bohemine, and CYC202 (data not shown).

Molecular pharmacodynamic markers in vitro. We investigated potential molecular pharmacodynamic markers of response to the three aminopurines in the HCT116 cell line (Fig. 3B). This line was selected because it was among the most sensitive of those tested (Table 2) and was suitable for use as a xenograft in subsequent in vivo studies. Figure 3B shows the effect of CYC202, bohemine, and olomoucine on the phosphorylation status of RB following treatment of cells for 24 hours with 3 x IC50 of these agents as measured by the sulforhodamine B assay at 96 hours (Table 2). Cell counts were done at this time in the same samples from which molecular analysis was carried out. Cell number was reduced by 60% to 70%, confirming that similarly effective concentrations of the three agents were being used (Fig. 3A). It is clear that RB phosphorylation was reduced by pharmacologically active concentrations of the three aminopurines. This was shown both by a gel mobility shift to the hypophosphorylated form and also by a major reduction (≥70%) in phosphorylation at all sites tested, including the proposed CDK2-preferred site at Thr821 (Fig. 3B). The decrease in RB phosphorylation was shown not to be due to a loss of total RB protein as demonstrated by the use of the antibody that recognized the Ser608 site in the nonphosphorylated state (Fig. 3B), the signal for which increased as phosphorylation was lost. Concentrations of the three inhibitors that elicited a similar inhibition of cell growth exhibited comparable inhibition of RB phosphorylation (Fig. 3C). Inhibition of RB phosphorylation in HCT116 cells by CYC202 was confirmed and quantified using a DELFIA-based assay measuring RB phosphorylation at Ser608 following a 24-hour exposure (Fig. 3D). The IC50 for inhibition of RB phosphorylation by CYC202 was 18 µmol/L. This correlates well with the 50% to 60% inhibition of proliferation seen at this concentration, as determined by cell counts.



View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. Cellular and molecular responses to CDK inhibitors. HCT116 cells were exposed to equiactive doses of CYC202, bohemine or olomoucine (3 x 96 hours IC50) for 24 hours. A, attached cell number following exposure to the CDKIs (n = 6 from two independent experiments). B, Western blots for the indicated proteins following analysis of cell lysates (50 µg protein). GAPDH was used as loading control. C, densitometric analysis of the Western blots in (B) as carried out in Image Quant. Columns, mean (n = 3); bars, ±SE. D, inhibition of RB phosphorylation at Ser608 in HCT116 cells using a 96-well plate-based DELFIA assay. Points, mean from one of two independent experiments; bars, ±SD. E, cell cycle distribution of fixed cells using propidium iodide and flow cytometry. Columns, mean from 2 independent experiments (n = 6); bars, ± SD.

 
In addition to decreases in RB phosphorylation, all three aminopurine compounds reduced the expression of cyclin D1 (77-100% decrease; Fig. 3B-C). Cyclins A and B1 also showed decreased expression but to a lesser extent (8-75% reduction). Cyclin E was least sensitive to inhibition by the three compounds but was also reduced (4-52% loss). The expression of CDK1 and CDK2 showed a modest decrease in response to the compounds (16-40% reduction). However, the expression of CDK4 was greatly reduced following exposure to all three inhibitors (~80% loss). The expression of phosphorylated RNA polymerase II was also reduced by all three compounds, as shown by a switch to the lower, hypophosphorylated form. This is consistent with inhibition of CDKs 7 and 9. In support of inhibition of CDK7 (28) Thr160 phosphorylation on CDK2 was reduced following 8 to 24 hours exposure to 3x IC50 CYC202 (data not shown). In addition, a partial cleavage of poly(ADP-ribose) polymerase was observed following a 24-hour exposure to these compounds, indicative of apoptosis (Fig. 3B-C).

In contrast to previous results in HT29 and KM12 human colon cancer cells (15), no increase in phosphorylated ERK1/2 was observed in response to the three compounds (Fig. 3B-C), possibly because the HCT116 cell line has a Kirsten-RAS mutation (29).

Treatment of HCT116 cells with CYC202, bohemine and olomoucine resulted in a 11% to 15% loss of cells from the S phase of the cell cycle with a 6% to 17% increase in the proportion of cells in the G2-M phase after 24 hours of treatment, as determined by propidium iodide staining and analysis by flow cytometry (Fig. 3E). Cells in the S phase were confirmed to have ceased DNA replication as determined by bromodeoxyuridine incorporation (data not shown). A sub-G1 fraction was also seen with all three agents, indicative of apoptosis. Taken together, the above results are consistent with the agents acting as CDK inhibitors although other possibilities cannot be ruled out.

Maximum tolerated dose in BALB/c– mice. Given i.v., the single dose maximum tolerated dose for both bohemine and CYC202 was 100 mg/kg. For olomoucine, the maximum tolerated dose was higher at 200 mg/kg. Subsequent i.v. studies were done with 50 mg/kg of the agents. At 150 mg/kg, CYC202 was well tolerated when given i.p. Following a single oral administration, CYC202 was well tolerated up to 2,000 mg/kg.

Analytic results and pharmacokinetic variables. LC-MS/MS proved to be specific, sensitive, and reproducible in both analytic systems described in Materials and Methods. Whereas only 100 ng/mL of each trisubstituted aminopurine could be detected with UV, the limit of quantification with LC-MS/MS was 3.5 ng/mL or 10 nmol/L. The interbatch precision and accuracy at 25, 90, 5,000, and 50,000 nmol/L were below 15%. These values are in accordance with the established guidelines for assay validation.

Tables 3 and 4 show the pharmacokinetic variables for CYC202, olomoucine, and bohemine in plasma, liver, kidney, and colon 26 tumor in BALB/c– mice following a dose of 50 mg/kg i.v. Variables were obtained using noncompartmental analysis. Plasma levels decayed in a biexponential fashion in all three cases (Fig. 4). All compounds cleared rapidly from plasma. CYC202 had the highest plasma concentrations and exhibited the longest half-life (1.19 hours) and the slowest clearance of 43 mL/h. Drug uptake from the general circulation was very rapid for all three compounds with Cmax observed in tissues at the sampling time of 0.25 minutes (Figs. 4 and 5). The tumor to plasma AUC ratios were 0.71, 0.45, and 0.18 for CYC202, olomoucine, and bohemine, respectively. These ratios were higher than those observed for kidney and spleen but lower than the corresponding liver to plasma ratios. The terminal half-life for the elimination of CYC202 from tissues was greater than that observed for the other two analogues (Table 4). Overall, CYC202 exhibited the highest AUC, the longest plasma t1/2, and the highest tissue to plasma ratio of the three analogues.


View this table:
[in this window]
[in a new window]

  		Table 3. Plasma pharmacokinetic variables derived from Winnonlin noncompartmental analysis following administration of 50 mg/kg i.v. olomoucine, bohemine, and CYC202 to BALB/c– mice

 

View this table:
[in this window]
[in a new window]

  		Table 4. Tissue pharmacokinetic variables for kidney, liver, and colon 26 tumor derived from Winnonlin noncompartmental analysis following administration of 50 mg/kg i.v. olomoucine, bohemine, and CYC202 to BALB/c– mice

 


View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 4. Plasma (A), colon 26 tumor (B), liver (C), and kidney (D) concentration versus time curves following administration of 50 mg/kg olomoucine ({square}), CYC202 ({lozenge}), and bohemine ({circ}), i.v. in 50 mmol/L HCl saline to BALB/c mice.

 


View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 5. CYC202 plasma concentration versus time curves following 50 mg/kg i.v. ({square}), i.p. ({circ}), oral administration ({lozenge}) in Solutrol Lutrol vehicle to female BALB/c– mice.

 
Figure 5 shows the plasma concentration versus time curves following i.p., i.v., and oral administration of 50 mg/kg CYC202. The pharmacokinetic data in Table 5 show that rapid absorption occurred from the peritoneal cavity, with maximal plasma concentrations observed 5 minutes after administration. The concentrations obtained by this route of administration were very similar to those observed by the i.v. route with an overall i.p. bioavailability of 67% (Table 5). There was no increase in terminal half-life by this route of administration suggesting that delayed absorption did not occur.


View this table:
[in this window]
[in a new window]

  		Table 5. Plasma pharmacokinetic variables derived from Winnonlin noncompartmental analysis following administration of 50 mg/kg CYC202, i.p., i.v., and orally to BALB/c– mice

 
CYC202 concentrations peaked 15 minutes after oral administration, and the terminal half-life increased compared with that measured after i.v. or i.p. administration (2.77 hours for oral dosing versus 1.38 hours by the i.v. route). The oral bioavailability was 86%. Following increasing oral doses of CYC202, the half-life was extended from 2.88 to 6.67 hours and 11.9 hours. Plasma concentrations were maintained above 15 µmol/L (the mean IC50 level across the human tumor cell line panel) for 4, 12, and 24 hours following 50, 500, and 2,000 mg/kg, respectively (Table 6; Fig. 6). Following increased oral doses, the AUC increased 10-fold from 50 to 500 mg/kg but only a further 2-fold from 500 to 2,000 mg/kg. This difference was also reflected in the Cmax which did not increase linearly with dose. It should be noted that the plasma AUC measured following 50 mg/kg CYC202 in the Solutol/Lutrol vehicle was 35% lower than that observed using the HCl saline vehicle at the same dose level. This could be explained by differences in Cmax as the t1/2 values were identical. The results suggest the possibility that acid conditions may favor oral absorption of CYC202.


View this table:
[in this window]
[in a new window]

  		Table 6. Plasma pharmacokinetic variables derived from Winnonlin noncompartmental analysis following administration of 50, 500, and 2,000 mg/kg CYC202 orally to BALB/c– mice

 


View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 6. CYC202 plasma concentrations versus time curves following administration of 50 ({circ}), 500 ({square}), and 2,000 mg/kg ({lozenge}) CYC202 orally to BALB/c– mice.

 
Computer simulation experiments derived from variables in Table 5 showed that to maintain a concentration of 15 µmol/L over 24 hours (mean IC50 level), CYC202 would need to be given orally thrice a day at 200 mg/kg or twice a day at 500 mg/kg (Fig. 7). Protein binding was similar at 2 and 20 µmol/L. In mouse plasma 97.2% of CYC202 was bound to proteins as early as 5 minutes after incubation and the binding was the same 6 hours after incubation. In human plasma, the protein binding increased slightly but gradually from 92% to 96% over a period of 5 minutes to 6 hours.



View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 7. Pharmacokinetic simulations for CYC202 following administration of (A) 500 mg/kg orally twice daily and (B) 200 mg/kg orally thrice daily.

 
Pharmacokinetic-pharmacodynamic analysis. To explore the pharmacokinetic-pharmacodynamic relationships in vivo, the HCT116 human colon cancer xenograft model in nude mice was used to determine the antitumor effect, tumor drug concentrations, and alterations in molecular biomarkers. Given the high oral bioavailability for CYC202 reported here and the attractiveness of oral dosing for clinical application, this route was selected for the pharmacokinetic-pharmacodynamic studies. Previous studies showed that CYC202 was tolerated in tumor-bearing nude mice at a dose of 500 mg/kg thrice a day (13). Based on the pharmacokinetic simulation (Fig. 7), mice were treated with 500 mg/kg orally CYC202 twice per day for 5 days. In the experiment shown in Fig. 8, the mean body weights (expressed as percentage of the starting weight on day 0) were 102%, 101%, 101%, 101%, and 95% on days 1 to 5 compared with 99%, 98%, 98%, 98%, and 98% for the vehicle controls. One of the eight animals died in the treated group during this period (day 2). CYC202 caused a reduction in the growth of established tumors (~130 mm3) compared with controls for the duration of treatment (Fig. 8A). %T/C values were 79%, 75%, 72%, and 65% at days 2, 3, 4, and 5, respectively (P < 0.05). This is comparable with the T/C value of 47% that was reported in the MESSA-DX5 human uterine carcinoma xenograft using a comparable schedule of 500 mg/kg orally thrice a day for 4 days (13). To determine if the inhibition of tumor growth was related to and consistent with the proposed mechanism of action of CYC202, tumor lysates were prepared and subjected to analysis by Western blotting for RB phosphorylation (Fig. 8B). Vehicle-treated tumor lysates exhibited readily detectable levels of total and phosphorylated RB. Mice treated with CYC202 on the above schedule for 1 day showed a decrease in total RB expression when measured at 4 hours after the second of the two daily doses (30% reduction as determined by densitometry). After 5 days of treatment, RB protein expression decreased further (55% reduction). Inhibition of RB phosphorylation at specific phosphorylation sites was determined using phospho-specific antibodies. Whereas the signal for all sites was markedly decreased after 3 to 5 days of treatment, the proposed CDK2-preferred Thr821 site was the most sensitive (95% inhibition on day 5). In addition, cyclin D1 levels were decreased by 40% to 77% after 3 to 5 days of treatment. Both the RB and cyclin D1 biomarker changes were similar to those described earlier for HCT116 cells treated in vitro. Also in agreement with the in vitro data, no increase in ERK phosphorylation was observed in vivo in response to CYC202.



View larger version (42K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 8. Pharmacokinetic-pharmacodynamic relationships for CYC202 in the HCT116 human colon cancer carcinoma xenograft model. A, effect of 500 mg/kg orally CYC202 twice daily on the growth of established HCT116 human colon carcinoma tumor xenografts (treated at ~130mm3) growing s.c. in nude mice. Points, mean relative tumor volume; bars, ±SE. %T/C values are indicated. P < 0.05 for treated versus control tumors. B, changes in various molecular biomarkers in the HCT116 tumor xenograft in response to CYC202 treatment as assessed by Western blotting. GAPDH was used as the loading control. C, tumor and plasma concentrations of CYC202 corresponding to samples used in (A) and (B) as determined by LC-MS/MS. The in vitro IC50 of 6.9 µmol/L for HCT116 cell growth inhibition is marked by a horizontal line. D, densitometric analysis of the Western blots in (C), as carried out in Image Quant.

 
To relate the molecular marker changes to drug exposure, CYC202 concentrations in the HCT116 tumor and plasma were determined by LC-MS/MS using samples from the same group of mice used for the biomarker measurements (Fig. 8C). Plasma concentrations were in accordance with predicted levels from the computer simulation described earlier. Tumor concentrations measured at 4 hours after the second of the twice daily doses were greater than those required to inhibit the growth of HCT116 cells in cell culture (IC50 of 6.9 µmol/L; Table 2) on days 1 and 3. However, the concentration of CYC202 was markedly reduced on day 5 (4 hours after the last dose) in both plasma and tumor. In a single oral dose experiment at 500 mg/kg, tumor concentrations were above cell proliferation IC50 levels for >12 hours and the final half-life was 3 hours (data not shown).


   Discussion
Top
 Abstract
 Materials and Methods
 Cell culture
 Results
 Discussion
References
 
The purpose of this study was to investigate the properties of three trisubstituted aminopurine CDK inhibitors, olomoucine, bohemine, and CYC202, with particular emphasis on pharmacokinetic-pharmacodynamic relationships for CYC202. A detailed understanding of the pharmacokinetic-pharmacodynamic properties should facilitate rational selection of a dosing schedule that would support therapeutic activity in tumor xenograft models (30). Identification of appropriate pharmacokinetic-pharmacodynamic variables allows construction of a "pharmacologic audit trail" (31, 32) to aid interpretation of preclinical data and for use in subsequent clinical trials.

CYC202 was the most potent of the three aminopurine analogues in terms of in vitro CDK2 inhibition and growth inhibition in a human cancer cell line panel. Interestingly, it was active in cell lines resistant to doxorubicin and cisplatin. Also of note is the lack of dependence on RB status as exemplified by similar IC50 values in the SAOS-2 versus U2-OS osteosarcoma cell lines. Similarly, TP53 status had little effect, if any, on the growth inhibition. Comparison of the HCT116 isogenic cell pair, differing only in TP53 status, showed near-identical IC50 values for CYC202. Together with previous data (13), these results suggest that the aminopurine analogues may exhibit broad-spectrum antitumor activity. The differences in the CDK inhibitory potency among the three purine analogues were reflected in their relative cancer cell growth inhibitory activities, CYC202 being more potent than bohemine which was in turn more active than olomoucine. This is consistent with CDK inhibition being a major contributor to the antiproliferative effects but does not exclude other possible cellular effects.

We showed that olomoucine, bohemine, and CYC202 all reduced RB phosphorylation at pharmacologically relevant concentrations that inhibit cell growth in cell culture. Although functional RB was not required for an antiproliferative effect, RB phosphorylation was nevertheless a valuable indicator of CDK activity and could be useful in those tumors that are positive for RB. Loss of RB phosphorylation in HCT116 cells was concomitant with cell growth inhibition, as seen with CYC202 in HT29 and KM12 human colon cancer cells (15). Assessing all the combined data, we concluded that inhibition of RB phosphorylation is related to in vitro potency in RB-positive cancer cell lines, whether comparing between the different analogues in a given cell line or comparing the effects of CYC202 across different lines.

Inhibition of RB phosphorylation in HCT116 cells was shown by the gel mobility shift seen with the hypophosphorylated form of RB and also by antibodies to specific phosphorylation sites. Consistent with inhibition of CDK2, phosphorylation at Thr821 (a CDK2-preferred RB phosphorylation site) was most affected by CYC202. There was some loss of total RB protein, but this was not responsible for reduced phosphorylation, as shown by the increase in nonphosphorylated Ser608. Over the same concentration range and time course as the effects on RB phosphorylation were shown, the analogues increased cell number in G2-M phase and decreased the S-phase fraction. Use of a quantitative DELFIA assay showed that RB phosphorylation at Ser608 was inhibited by 50% following 24 hours exposure to 18 µmol/L CYC202, correlating well with the 50% to 60% decrease in cell number, compared with controls, observed with this treatment.

Inhibition of RB phosphorylation and the above cell cycle changes could be occurring through inhibition of CDK1/cyclin B or of CDK2. Introduction of dominant-negative CDK2 into cancer cell lines has been shown to cause an arrest in G2-M (33). We also showed that CYC202, bohemine, and olomoucine caused a marked decrease in cyclin D1 expression and a smaller loss of cyclins A and B1, with cyclin E affected much less. Decreased expression of cyclin D1 and other cyclins in HCT116 cells also occurred in HT29 and KM12 cells (15). In addition, CYC202 decreased CDK4 expression in HCT116 cells, an effect not seen in HT29 or KM12 cells previously (15). The data are consistent with the analogues acting directly or indirectly (or both) as inhibitors of CDKs, although other cellular mechanisms are not excluded. The compounds can inhibit activity of the transcription-related CDKs, CDK7 and CDK9 (Table 1), which are involved in phosphorylation of RNA polymerase II. Reduced phosphorylation of RNA polymerase II was observed in HCT116 cells in vitro. Inhibition of T160 phosphorylation on CDK2, mediated by CDK7, was also shown. Therefore, the reduction in cyclin expression may occur through transcriptional inhibition due to reduced CDK7 and CDK9 activity (15). Gene expression profiling can be used to evaluate such effects (34). The relative decreases in the individual cyclin proteins may reflect differential protein half-lives following an inhibition of mRNA expression downstream of CDK7 and CDK9 inhibition. Interestingly, the most marked effect occurred with the G1 phase-active cyclin D1, whereas the cells predominantly arrest in the G2-M phase of the cell cycle. Thus, CDK inhibition (see above) and/or loss of other proteins is more likely to contribute to the G2-M phase arrest. Loss of several cyclins may inhibit multiple CDK activities and may therefore block cell proliferation independent of CDK2 activity, which has been reported as dispensable for colon cancer cell proliferation (7). It should be emphasized that CDKs phosphorylate important cellular substrates other than RB (35–48). As noted previously (15), this may explain the activity of the aminopurine CDK inhibitors such as CYC202 on cells lacking RB.

A sub-G1 peak was seen in HCT116 cells after treatment with CYC202 and its analogues. An increase in poly(ADP-ribose) polymerase cleavage provided biochemical evidence of apoptosis, consistent with positivity in the terminal deoxynucleotidyl transferase–mediated nick-end labeling assay reported previously for CYC202 (13). Thus, over the concentration range in which olomoucine, bohemine, and CYC202 reduced RB phosphorylation and cyclin expression, these compounds caused not only a G2-M arrest and a decrease in the S-phase fraction but also the induction of apoptosis. Transformed and normal cells may differ in their requirements for CDK2 (8). Phosphorylation of E2F by CDK2/cyclin A is a major transcriptional control point, and modulation by CDK2 inhibitors such as CYC202, might cause deregulation of E2F and induction of apoptosis (8, 9).

Our previous studies have shown that ERK1/2 phosphorylation is induced by CYC202 in HT29, KM12, and NIH3T3 cells in a mitogen-activated protein kinase kinase 1/2–dependent manner (15). Interestingly, the effect was not seen here in HCT116 cells. This may be because the Kirsten-RAS mutation in HCT116 cells (29) causes constitutive activation of ERK1/2, reducing the opportunity for further activation by CYC202.

Pharmacokinetic properties are frequently critical in the translation of in vitro anticancer properties into therapeutic activity in an animal model, in the selection of a clinical development candidate and in the subsequent clinical evaluation. The growth inhibitory IC50 for bohemine and CYC202 decreased markedly with increased exposure time. Maximum growth inhibitory effect was seen after 8 to 16 hours. These results indicate that pharmacokinetic behavior is likely to be critical for in vivo activity. Here LC-MS/MS provided rapid, specific, and sensitive analysis of olomoucine, CYC202, and bohemine in biological tissues and fluids. As mentioned, the time dependence of the effects of CYC202 and bohemine in vitro emphasized the likely importance of pharmacokinetic behavior in vivo. All three compounds distributed quickly into mouse liver, kidney, and tumor tissue but also cleared quite rapidly from plasma and tissues. Bohemine cleared the fastest and CYC202 the slowest with olomoucine intermediate. Although CYC202 plasma protein binding was rapid and extensive (97%), it is important to note that this did not limit drug distribution, which was equally fast. Uptake into tumor tissue, as exemplified by the colon 26 transplantable syngeneic mouse tumor, was quite different for the three agents, with AUC values ranging from 18% of plasma for olomoucine to 71% for CYC202. Tissue clearance rates were comparable to those of plasma. CYC202 clearance is via oxidative metabolism of the side chain hydroxyl group to form the carboxylic acid, subsequently excreted in urine (49). The pharmacokinetics behaviour described here was predictive of that in healthy male subjects receiving single doses up to 800 mg CYC202 (50). Of the three analogues, CYC202 showed the best pharmacokinetic profile with the slowest clearance and highest tissue distribution. The nanomolar potency towards CDK2/cyclin E taken together with the high cellular potency and the better tissue distribution and clearance data, indicated that CYC202 had the most suitable overall properties for further therapeutic evaluation. It was therefore selected for pharmacokinetic-pharmacodynamic studies in the same HCT116 human colon cancer model as used in vitro.

Despite rapid distribution and high micromolar peak levels, tumor concentrations at the maximum tolerated single dose of CYC202 were not sustained above the in vitro IC50. This indicated the need to lengthen drug exposure by continuous infusion or repeated administration. The high bioavailability, both i.p. (67%) and orally (86%), suggested both routes would be suitable for multiple dosing in tumor xenografts. There seemed to be saturation in oral absorption from 500 to 2,000 mg/kg, but the AUC increased 10-fold with a 10-fold increase in dose from 50 to 500 mg/kg. Based on pharmacokinetic variables obtained with single doses, our multiple dose pharmacokinetic simulation study indicated that CYC202 should be given thrice a day at 200 mg/kg or twice a day at 500 mg/kg to sustain therapeutic exposures. Concentrations measured after 500 mg/kg orally twice daily showed that the simulation was quite predictive for days 1 to 4 of dosing. However, the decrease in plasma and tumor levels after 5 days suggests metabolic induction or decreased absorption. Importantly, tumor concentrations were well above the growth inhibitory concentrations before day 5.

In the pharmacokinetic-pharmacodynamic studies, RB phosphorylation was measured to determine whether CDKs were likely to have been inhibited by CYC202 and to assess whether this was a valid PD biomarker in the xenograft tumor model. As predicted from the in vitro sensitivity of HCT116 cells and the simulated and measured tumor pharmacokinetics, CYC202 showed antitumor activity against HCT116 tumor xenograft for the duration of the 5-day treatment with T/C values in the range of 79% to 65%. A comparable T/C value of 47% was reported for the MESS-DX5 human uterine carcinoma xenograft using a similar regimen of 500 mg/kg orally thrice a day for 4 days (13) and the same schedule has been used in a Phase I clinical trial of CYC202 (51, 52). More prolonged drug administration at the dose level used was not well tolerated. Progressive reduction of tumor RB phosphorylation and cyclin D1 expression occurred over the 5 days of treatment. As mentioned, a concentration of 18 µmol/L CYC202 is required to inhibit RB phosphorylation by 50% in cell culture after a 24-hour exposure. Tumor concentrations measured 4 hours after the second dose exceeded 18 µmol/L on days 1 and 3. Given that the exposure to CYC202 is not continuous in the mouse but rather is intermittent due to repeated dosing and subsequent clearance of the compound, it is probably not surprising that >24 hours is needed to achieve a measurable effect on RB phosphorylation in the HCT116 xenograft. The sustained PD effects at 5 days, which occur despite the lower CYC202 tumor concentrations at this time, remain unexplained. Differences in tumor microenvironment for in vivo tumor xenografts compared with in vitro cell culture may play a role.

It is important to stress that although the inhibition of RB phosphorylation in the human tumor xenograft is consistent with CYC202 acting as a CDK inhibitor in vivo, whether this is direct via inhibition of CDK catalytic activity or indirect via depletion of cyclins and CDK4 is unclear. Regardless of the precise mechanism, our results indicate that RB phosphorylation may be a useful pharmacodynamic marker that generally correlates with the antitumor activity of CYC202, and suggest that cyclin D1 may be an additional biomarker. Cyclin D1 may be especially useful in situations where RB phosphorylation is not detectable, as in RB-negative tumors. With respect to clinical application, these measurements require suitable tumor or surrogate tissue to be available before and after treatment.

We have shown that a detailed understanding of the in vitro CDK inhibitory potency and cellular activity of this group of trisubstituted aminopurines, coupled to a similarly detailed analysis of pharmacokinetic and pharmacodynamic properties in vivo, led to the rational selection of CYC202 for tumor xenograft studies. Based on the knowledge of the drug exposures required for antiproliferative activity and of the PK properties of CYC202, computer simulation of drug concentrations was helpful in developing an oral dosing schedule that allowed therapeutic activity to be shown at exposures causing inhibition of RB phosphorylation and cyclin D1 depletion in tumor tissue. The data presented constitute a "pharmacologic audit trail," particularly in relation to pharmacokinetic-pharmacodynamic relationships (31, 32). The construction of such an audit trail can improve the quality of decision-making in preclinical drug development. Together with the antitumor activity in other xenograft models reported elsewhere using similar schedules (13), these results support the ongoing development of CYC202, which is now undergoing Phase II clinical trial (51).


   Acknowledgments
 
We thank the members of the Signal Transduction and Molecular Pharmacology Team and the Cell Cycle Control Team in the Cancer Research UK Centre for Cancer Therapeutics, together with colleagues at Cyclacel Ltd., for valuable discussions; Jenny Titley of the Cell Cycle Control Team for flow cytometry; and the protein biochemistry and the assay development/screening teams at Cyclacel for kinase assays.


   Footnotes
 
Grant support: Cancer Research UK grant C309/A2187 (P. Workman), Cyclacel Ltd. (P. Workman and S.R. Whittaker), the Sir Samuel Scott of Yews Trust studentship (S.R. Whittaker), Cancer Research UK Gibb Fellowship (D.P. Lane), and Cancer Research UK Life Fellowship (P. Workman).

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.

Note: F.I. Raynaud and S.R. Whittaker, contributed equally to this work.

Received 11/ 9/04; revised 3/21/05; accepted 4/13/05.


   References
Top
 Abstract
 Materials and Methods
 Cell culture
 Results
 Discussion
 References
 

  1. Sherr CJ. Cancer cell cycles. Science 1996;274:1672–7.[Abstract/Free Full Text]
  2. Sherr CJ. The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 2000;60:3689–95.[Abstract/Free Full Text]
  3. Porter PL, Malone KE, Heagerty PJ, et al. Expression of cell-cycle regulators p27Kip1 and cyclin E, alone and in combination, correlate with survival in young breast cancer patients. Nat Med 1997;3:222–5.[CrossRef][Medline]
  4. Tsihlias J, Kapusta L, Slingerland J. The prognostic significance of altered cyclin-dependent kinase inhibitors in human cancer. Annu Rev Med 1999;50:401–23.[CrossRef][Medline]
  5. Fry DW, Garrett MD. Inhibitors of cyclin-dependent kinases as therapeutic agents for the treatment of cancer. Current Opinion in Oncologic. Endocrine and Metabolic Investigational Drugs 2000;2:40–59.
  6. Sausville EA. Complexities in the development of cyclin-dependent kinase inhibitor drugs. Trends Mol Med 2002;8:S32–7.[CrossRef][Medline]
  7. Tetsu O, McCormick F. Proliferation of cancer cells despite CDK2 inhibition. Cancer Cell 2003;3:233–45.[CrossRef][Medline]
  8. Chen YN, 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]
  9. Mendoza N, Fong S, Marsters J, Koeppen H, Schwall R, Wickramasinghe D. Selective cyclin-dependent kinase 2/cyclin A antagonists that differ from ATP site inhibitors block tumor growth. Cancer Res 2003;63:1020–4.[Abstract/Free Full Text]
  10. Vesely J, Havlicek L, Strnad M, et al. Inhibition of cyclin-dependent kinases by purine analogues. Eur J Biochem 1994;224:771–86.[Medline]
  11. De Azevedo WF, Leclerc S, Meijer L, Havlicek L, Strnad M, Kim SH. Inhibition of cyclin-dependent kinases by purine analogues: crystal structure of human cdk2 complexed with roscovitine. Eur J Biochem 1997;243:518–26.[Medline]
  12. Wang S, McClue SJ, Ferguson JR, et al. Synthesis and configuration of the cyclin-dependent kinase inhibitor roscovitine and its enantiomer. Tetrahedron: Asymmetry 2001;12:2891–4.[CrossRef]
  13. McClue SJ, Blake D, Clarke R, et al. In vitro and in vivo antitumor properties of the cyclin dependent kinase inhibitor CYC202 (R-roscovitine). Int J Cancer 2002;102:463–8.[CrossRef][Medline]
  14. 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]
  15. Whittaker SR, Walton MI, Garrett MD, Workman P. The cyclin-dependent kinase inhibitor CYC202 (R-roscovitine) inhibits retinoblastoma protein phosphorylation, causes loss of Cyclin D1, and activates the mitogen-activated protein kinase pathway. Cancer Res 2004;64:262–72.[Abstract/Free Full Text]
  16. Schutte B, Nieland L, van Engeland M, Henfling ME, Meijer L, Ramaekers FC. The effect of the cyclin-dependent kinase inhibitor olomoucine on cell cycle kinetics. Exp Cell Res 1997;236:4–15.[CrossRef][Medline]
  17. Wang S, Meades C, Wood G, et al. 2-Anilino-4-(thiazol-5-yl)pyrimidine CDK inhibitors: synthesis, SAR analysis, X-ray crystallography, and biological activity. J Med Chem 2004;47:1662–75.[CrossRef][Medline]
  18. Bunz F, Dutriaux A, Lengauer C, et al. Requirement for p53 and p21 to sustain G2 arrest after DNA damage. Science 1998;282:1497–501.[Abstract/Free Full Text]
  19. Kelland LR, Murrer BA, Abel G, Giandomenico CM, Mistry P, Harrap KR. Ammine/amine platinum(IV) dicarboxylates: a novel class of platinum complex exhibiting selective cytotoxicity to intrinsically cisplatin-resistant human ovarian carcinoma cell lines. Cancer Res 1992;52:822–8.[Abstract/Free Full Text]
  20. Workman P, Twentyman P, Balkwill F, et al. United Kingdom Co-ordinating Committee on Cancer Research (UKCCCR) guidelines for the welfare of animals in experimental neoplasia (second edition). Br J Cancer 1998;77:1–10.
  21. Gabrielsson J, Weiner D. Pharmacokinetic and pharmacodynamic data analysis, concepts and applications. 2nd ed. Swedish Pharmaceutical Press; 1997. Stockholm, Sweden.
  22. Tamrakar S, Mittnacht S, Ludlow JW. Binding of select forms of pRB to protein phosphatase type 1 independent of catalytic activity. Oncogene 1999;18:7803–9.[CrossRef][Medline]
  23. Barrie SE, Eno-Amooquaye E, Hardcastle A, et al. High-throughput screening for the identification of small-molecule inhibitors of retinoblastoma protein phosphorylation in cells. Anal Biochem 2003;320:66–74.[CrossRef][Medline]
  24. Mellish KJ, Kelland LR. Mechanisms of acquired resistance to the orally active platinum-based anticancer drug bis-acetato-ammine-dichloro-cyclohexylamine platinum (i v.) (JM216) in two human ovarian carcinoma cell lines. Cancer Res 1994;54:6194–200.[Abstract/Free Full Text]
  25. Sharp SY, Rowlands MG, Jarman M, Kelland LR. Effects of a new antioestrogen, idoxifene, on cisplatin- and doxorubicin-sensitive and -resistant human ovarian carcinoma cell lines. Br J Cancer 1994;70:409–14.[Medline]
  26. Ashizawa S, Nishizawa H, Yamada M, et al. Collective inhibition of pRB family proteins by phosphorylation in cells with p16INK4a loss or cyclin E overexpression. J Biol Chem 2001;276:11362–70.[Abstract/Free Full Text]
  27. Thompson AM, Steel CM, Chetty U, et al. p53 gene mRNA expression and chromosome 17p allele loss in breast cancer. Br J Cancer 1990;61:74–8.[Medline]
  28. Fisher RP, Morgan DO. A novel cyclin associates with MO15/CDK7 to form the CDK-activating kinase. Cell 1994;78:713–24.[CrossRef][Medline]
  29. Gayet J, Zhou XP, Duval A, et al. Extensive characterization of genetic alterations in a series of human colorectal cancer cell lines. Oncogene 2001;20:5025–32.[CrossRef][Medline]
  30. Garrett MD, Walton MI, McDonald E, Judson I, Workman P. The contemporary drug development process: advances and challenges in preclinical and clinical development. Prog Cell Cycle Res 2003;5:145–58.[Medline]
  31. Workman P. Challenges of PK/PD measurements in modern drug development. Eur J Cancer 2002;38:2189–93.
  32. Workman P. How much gets there and what does it do?: the need for better pharmacokinetic and pharmacodynamic endpoints in contemporary drug discovery and development. Curr Pharm Des 2003;9:891–902.[CrossRef][Medline]
  33. Hu B, Mitra J, van den Heuvel S, Enders GH. S and G2 phase 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]
  34. 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:RESEARCH0041.[Medline]
  35. Bueno A, Russell P. Dual functions of CDC6: a yeast protein required for DNA replication also inhibits nuclear division. EMBO J 1992;11:2167–76.[Medline]
  36. Byrd PJ, Cooper PR, Stankovic T, et al. A gene transcribed from the bidirectional ATM promoter coding for a serine rich protein: amino acid sequence, structure and expression studies. Hum Mol Genet 1996;5:1785–91.[Abstract/Free Full Text]
  37. Chen PL, Scully P, Shew JY, Wang JY, Lee WH. Phosphorylation of the retinoblastoma gene product is modulated during the cell cycle and cellular differentiation. Cell 1989;58:1193–8.[CrossRef][Medline]
  38. Gao G, Bracken AP, Burkard K, et al. NPAT expression is regulated by E2F and is essential for cell cycle progression. Mol Cell Biol 2003;23:2821–33.[Abstract/Free Full Text]
  39. Hua XH, Newport J. Identification of a preinitiation step in DNA replication that is independent of origin recognition complex and cdc6, but dependent on cdk2. J Cell Biol 1998;140:271–81.[Abstract/Free Full Text]
  40. Imai T, Yamauchi M, Seki N, et al. Identification and characterization of a new gene physically linked to the ATM gene. Genome Res 1996;6:439–47.[Abstract/Free Full Text]
  41. Imai T, Sugawara T, Nishiyama A, et al. The structure and organization of the human NPAT gene. Genomics 1997;42:388–92.[CrossRef][Medline]
  42. Jiang W, Wells NJ, Hunter T. Multistep regulation of DNA replication by Cdk phosphorylation of HsCdc6. Proc Natl Acad Sci U S A 1999;96:6193–8.[Abstract/Free Full Text]
  43. 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]
  44. Ma T, Van Tine BA, Wei Y, et al. Cell cycle-regulated phosphorylation of p220(NPAT) by cyclin E/Cdk2 in Cajal bodies promotes histone gene transcription. Genes Dev 2000;14:2298–313.[Abstract/Free Full Text]
  45. Petersen BO, Lukas J, Sorensen CS, Bartek J, Helin K. Phosphorylation of mammalian CDC6 by cyclin A/CDK2 regulates its subcellular localization. EMBO J 1999;18:396–410.[CrossRef][Medline]
  46. Shan B, Farmer AA, Lee WH. The molecular basis of E2F-1/DP-1-induced S-phase entry and apoptosis. Cell Growth Differ 1996;7:689–97.[Abstract]
  47. Zhao H, Piwnica-Worms H. ATR-mediated checkpoint pathways regulate phosphorylation and activation of human Chk1. Mol Cell Biol 2001;21:4129–39.[Abstract/Free Full Text]
  48. Zhao J, Dynlacht B, Imai T, Hori T, Harlow E. Expression of NPAT, a novel substrate of cyclin E-CDK2, promotes S-phase entry. Genes Dev 1998;12:456–61.[Abstract/Free Full Text]
  49. Nutley BP, Raynaud F, Wilson SC, et al. Metabolism and pharmacokinetics of the cyclin dependent kinase inhibitor CYC202 in the mouse. Mol Cancer Ther 2004;4:125–39.
  50. de la MS, Gianella-Borradori A. Pharmacokinetic model of R-roscovitine and its metabolite in healthy male subjects. Int J Clin Pharmacol Ther 2004;42:232–9.[Medline]
  51. Raymond E, Laurence V, Faivre S, et al. Proc AARC-NCI-EORTC Mol Targets Cancer Ther 2002;13 Abs 150.
  52. Guzi T. CYC-202. Curr Opin Investig Drugs 2004;5:1311–8.[Medline]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
A. B.Y. Hui, S. Yue, W. Shi, N. M. Alajez, E. Ito, S. R. Green, S. Frame, B. O'Sullivan, and F.-F. Liu
Therapeutic Efficacy of Seliciclib in Combination with Ionizing Radiation for Human Nasopharyngeal Carcinoma
Clin. Cancer Res., June 1, 2009; 15(11): 3716 - 3724.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
K. Bettayeb, H. Sallam, Y. Ferandin, F. Popowycz, G. Fournet, M. Hassan, A. Echalier, P. Bernard, J. Endicott, B. Joseph, et al.
N-&-N, a new class of cell death-inducing kinase inhibitors derived from the purine roscovitine
Mol. Cancer Ther., September 1, 2008; 7(9): 2713 - 2724.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
C. Zhang, K. Lundgren, Z. Yan, M. E. Arango, S. Price, A. Huber, J. Higgins, G. Troche, J. Skaptason, T. Koudriakova, et al.
Pharmacologic properties of AG-012986, a pan-cyclin-dependent kinase inhibitor with antitumor efficacy
Mol. Cancer Ther., April 1, 2008; 7(4): 818 - 828.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
S. Y. Sharp, C. Prodromou, K. Boxall, M. V. Powers, J. L. Holmes, G. Box, T. P. Matthews, K.-M. J. Cheung, A. Kalusa, K. James, et al.
Inhibition of the heat shock protein 90 molecular chaperone in vitro and in vivo by novel, synthetic, potent resorcinylic pyrazole/isoxazole amide analogues
Mol. Cancer Ther., April 1, 2007; 6(4): 1198 - 1211.
[Abstract] [Full Text] [PDF]

Home page
 Mol Cancer ResHome page
S. Mohapatra, D. Coppola, A. I. Riker, and W. J. Pledger
Roscovitine Inhibits Differentiation and Invasion in a Three-Dimensional Skin Reconstruction Model of Metastatic Melanoma
Mol. Cancer Res., February 1, 2007; 5(2): 145 - 151.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
N. F. Smith, F. I. Raynaud, and P. Workman
The application of cassette dosing for pharmacokinetic screening in small-molecule cancer drug discovery
Mol. Cancer Ther., February 1, 2007; 6(2): 428 - 440.
[Abstract] [Full Text] [PDF]

Home page
 Cold Spring Harb Symp Quant BiolHome page
P. WORKMAN
Drugging the Cancer Kinome: Progress and Challenges in Developing Personalized Molecular Cancer Therapeutics
Cold Spring Harb Symp Quant Biol, January 1, 2005; 70(0): 499 - 515.
[Abstract] [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 Raynaud, F. I.
Right arrow Articles by Workman, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Raynaud, F. I.
Right arrow Articles by Workman, P.

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