Drug Metabolism and Disposition Fast Forward
First published on November 29, 2007; DOI: 10.1124/dmd.107.019232
0090-9556/08/3603-561-570$20.00
DMD 36:561-570, 2008
Metabolism of the Trisubstituted Purine Cyclin-Dependent Kinase Inhibitor Seliciclib (R-Roscovitine) in Vitro and in Vivo
Steven J. McClue, and
Iain Stuart
Cyclacel Ltd., James Lindsay Place, Dundee, United Kingdom
(Received October 16, 2007;
accepted November 28, 2007)
 |
Abstract
|
|---|
Seliciclib (R-roscovitine, CYC202) is a small molecule inhibitor of cyclin-dependent kinases currently in phase II clinical trials as an anticancer agent. We examined the metabolism of seliciclib in vitro and in vivo. Using radiolabeled seliciclib we found that cytochrome P450 (P450)-mediated metabolism in liver microsomes from human, rat, mouse, rabbit, monkey, and dog was rapid to a number of metabolic species, one of the most prevalent being a carboxylate previously identified in urine from rats and mice dosed with seliciclib. Metabolism was fastest in mouse microsomes and slowest in microsomes from dog. Using characterized human microsomes, we identified the P450s responsible for this metabolism as CYP3A4 and CYP2B6. Glucuronidation of seliciclib and its metabolites was shown to be a major elimination process in bile duct-cannulated rats dosed with [14C]seliciclib at 10 mg/kg. Elimination by the fecal route accounted for up to 65% of the administered dose, whereas urinary excretion accounted for up to 43%. Almost half of the administered dose was found to be eliminated via the bile, and elimination was found to be rapid, with up to 88% of the dose being excreted within the first 24 h. Preliminary experiments indicated that UDP-glucuronosyltransferase (UGT) 1A3, 1A9, and 2B7 were involved in the conjugation of seliciclib. Seliciclib was further shown in vitro to inhibit the activity of some of the enzymes responsible for its metabolism. Cytochrome P450s CYP3A4 and CYP2C9 and UGT1A1 were all inhibited at concentrations achieved in human trials, which raises the possibility of drug-drug interactions in the clinic.
Purine derivatives have been under investigation for their antiproliferative properties since the finding that 6-dimethylaminopurine could inhibit cell division in sea urchin embryos and its subsequent identification as an inhibitor of the cyclin-dependent kinase cdc2 (CDK1) (Meijer and Pondaven, 1988
; Neant and Guerrier 1988). One of the first selective purine inhibitors of CDKs was olomoucine, a 2,6,9-trisubstituted purine (Vesel
et al., 1994). Further analogs, bohemine and roscovitine, were synthesized shortly thereafter and shown to be more potent inhibitors of CDK2, with the R-enantiomer of roscovitine having CDK2 inhibitory activity superior to that of the S-enantiomer (Havlícek et al., 1997; Wang et al., 2001). R-Roscovitine appears to act by a proapoptotic mechanism to induce cell death in human tumor cells and has shown efficacy in human tumor xenograft models (McClue et al., 2002a,b). R-Roscovitine is currently under development by Cyclacel Ltd under the name seliciclib as an anticancer agent and is in phase II clinical trials as a single agent and in combination with approved anticancer agents (Fischer and Gianella-Borradori, 2005). The use of seliciclib in combination with standard chemotherapy drugs requires an understanding of the routes of its metabolism and excretion to predict likely drug-drug interactions and possible resulting toxicities in the clinic. Seliciclib is from a different chemical class than other CDK inhibitors currently in clinical trials and has already shown significant differences in both mechanism of action and toxicity (Shapiro, 2004; Fischer and Gianella-Borradori, 2005; Benson et al., 2007).
The metabolism of the related trisubstituted purine, bohemine, has been studied in some detail in vitro, and routes of primary metabolism and glucuronidation have been determined (Chmela et al., 2001; Rypka et al., 2002; Cervenková et al., 2003). Metabolism of roscovitine has also been examined after i.v. administration to rats (Vita et al., 2005) and to mice (Nutley et al., 2005). We report here the results of our studies on metabolism of seliciclib both in vitro and in vivo. We demonstrate in vivo that glucuronidation accounts for a higher percentage of compound metabolism and excretion than does P450-mediated metabolism after oral dosing of seliciclib to rats. We show that P450-mediated metabolism of seliciclib occurs in vitro principally via cytochrome P450 CYP3A4 to a number of metabolic species, one of the most prevalent being a less active and more soluble carboxylate. Seliciclib is further shown in vitro to inhibit the activity of CYP3A4. We show comparative in vitro metabolic profiles from nonhuman species, illustrating differences that may inform the choice of species for long-term toxicity studies.
|
Materials and Methods
|
|---|
Seliciclib (R-roscovitine, CYC202) and [14C]seliciclib ([7-14C]roscovitine) were synthesized as published previously (Wang et al., 2001). [14C]Seliciclib was incubated with pooled or individual human liver microsomal preparations (0.5 mg/ml) at concentrations shown in the text for 0, 10, 20, 40, and 60 min. Incubation mixtures contained [14C]seliciclib (20 µl, 18 kBq/ml), microsomal preparation (50 µl), and phosphate buffer (1730 µl). Mixtures were incubated for 5 min at 37°C before reaction was started by addition of NADPH cofactor (200 µl). The final incubation volume was 2 ml. Control incubations contained NaHCO3 (2% w/v) in place of the cofactor or phosphate buffer in place of the microsomal preparation. The reaction was terminated by removing 250-µl aliquots into chilled methanol. After centrifugation at 1200 rpm to pellet protein, supernatants were removed for HPLC analysis. HPLC analysis was performed using a Prodigy ODS-3 HPLC column (250 x 4.6 mm; 5 µm; Phenomenex, Torrance, CA) and a mobile phase of 0.7% triethylamine (pH 2.8 with formic acid)-acetonitrile (70:30) with a flow rate of 1 ml/min. 14C-Labeled moieties were detected with a β-RAM3 radiochemical detector (500-µl liquid flow cell; Lab Logic, Sheffield, South Yorkshire, UK) using Monoflow 3 scintillant (National Diagnostic, East Riding of Yorkshire, UK) at a flow rate of 3 ml/min. Under these conditions seliciclib and its carboxylate metabolite PMF30-128 had retention times of 18 and 20 min, respectively (Fig. 1). For species comparison experiments, the mobile phase was 0.1 M ammonium formate-acetonitrile (70:30). Under these conditions seliciclib and PMF30-128 had retention times of 37 and 22 min, respectively (see Fig. 4).
View larger version (14K):
[in this window]
[in a new window]
|
FIG. 1. Human liver microsomal metabolism of seliciclib: HPLC analysis. A, no cofactor, 10 µM [14C]seliciclib, 60-min incubation. B, 1 µM [14C]seliciclib, NADPH cofactor, 20-min incubation. C, 10 µM [14C]seliciclib, NADPH cofactor, 20-min incubation. D, 100 µM [14C]seliciclib, NADPH cofactor, 20-min incubation. Seliciclib is indicated on these traces as CYC202.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
FIG. 4. Species comparison of microsomal metabolism. Microsomes from human, rat, mouse, rabbit, cynomolgus monkey, and dog livers were incubated with [14C]seliciclib and assayed by reversed-phase HPLC. Seliciclib is indicated on these traces as CYC202.
|
|
Cryopreserved hepatocytes and human, rat, mouse, and dog microsomes were obtained from In Vitro Technologies (Leipzig, Germany) or from Human Biologics International (Scottsdale, AZ). Microsomal stability experiments were performed as follows: 250 µg of microsomes were diluted in a final volume of 445 µl of Sorenson's PB (100 mM dipotassium hydrogen phosphate with pH adjusted to 7.4 with 100 mM potassium dihydrogen orthophosphate). To each tube was added 5 µl of 500 µM seliciclib (diluted in acetonitrile), and the reactions were preincubated at 37°C with gentle shaking. After 5 min all reactions were initiated, with the exception of T = 0, by the addition of cofactor solution consisting of 2% (w/v) NaHCO3, 1.7 mg/ml NADP, 7.8 mg/ml glucose 6-phosphate, and 6 units of glucose-6-phosphate dehydrogenase. For T = 0, 50 µl of ice-cold acetonitrile was added before the addition of cofactor solution. Samples were taken at T = 5, 15, 30, 60, and 120 min, and all reactions were stopped at the appropriate time points by the addition of 50 µl of ice-cold acetonitrile. Samples were snap-frozen immediately and submitted for analysis by LC-MS/MS. Control reactions contained seliciclib diluted in phosphate-buffered saline to final concentrations of 5, 1, 0.5, and 0 µM.
Analysis was performed on LC-MS/MS system comprising a PAL HTS autoinjector (CTC Analytics, Zurich, Switzerland), a series 600 binary HPLC pump and a Quattro Ultima tandem mass spectrometer (both Waters Limited, Elstree, Hertfordshire, UK). Chromatography was performed using a Jupiter Proteo HPLC column (250 x 4.6 mm, 4 µm; Phenomenex) at a flow rate of 1 ml/min and the following gradient program [90% 0.2% formic acid (aqueous)-10% acetonitrile (A); 10% 0.2% formic acid (aqueous)-90% acetonitrile (B)]: 0 to 3 min 100% A, 3 to 18 min linear gradient to 20% A/80% B and held to 25 min; and 25 to 30 min reequilibrate. Pseudomolecular and product ion scanning experiments were performed over the mass ranges 200 > 600 and 50 > 400 Da, respectively.
Microsomes genetically engineered to express a single human cytochrome P450 enzyme were purchased from Cypex Ltd (Dundee, UK). Recombinant human cytochrome P450s and UGT Supersomes were obtained from BD Biosciences (Erembodegem, Belgium). Recombinant human cytochrome P450 enzyme inhibition assays were performed using BD Gentest P450 inhibition kits. Recombinant P450s (0.25 µg/ml) were incubated with seliciclib over a range of 0.137 to 300 µM and assayed for activity at 37°C using a fluorescent method. The reaction phenotype was determined using incubations of [14C]seliciclib (10 µM) with liver microsomes from 14 individual donors (0.5 mg/ml) for 10 min at 37°C. The reaction was terminated by addition of chilled acetonitrile, and samples were then centrifuged at 12,000 rpm to pellet protein, after which aliquots of the supernatant were analyzed using a reversed-phase HPLC method to determine either the loss of seliciclib or the appearance of metabolite PMF30-128 or Met-4. Probe substrates and their final concentrations in reactions were as follows: CYP1A2, [14C]phenacetin (20 µM); CYP2C9, [14C]tolbutamide (100 µM); CYP2C19, [14C]S-mephenytoin (100 µM); CYP2D6, bufuralol (5 µM); and CYP3A4, [14C]testosterone (50 µM).
View larger version (9K):
[in this window]
[in a new window]
|
FIG. 2. Rate of metabolism of seliciclib by pooled human liver microsomes. Human liver microsomes were incubated with seliciclib for up to 120 min. Loss of seliciclib was determined by LC-MS/MS analysis.
|
|
UDP-glycosyltransferase activity was assayed in two ways. 1) If samples were required for mass spectrometric analysis, drug candidates at various concentrations were incubated in a buffer containing 50 mM Tris-HCl (pH 7.5), 8 mM MgCl2, 2 mM UDP-glucuronic acid, and 25 µg of alamethicin/mg of protein. The mixture was warmed to 37°C before the addition of a suitable aliquot of UGT Supersomes to start the reaction and incubated at 37°C for 60 min. The final reaction volume was 25 µl, and the reaction was terminated by the addition of 10 µl of 80% MeCN/20% 0.5 M Tris. 2) For the fluorimetric assay, drug candidates at various concentrations were mixed in a buffer containing 10 µM scopoletin, 50 mM Tris-HCl (pH 7.5), 8 mM MgCl2, 2 mM UDPGA, and 25 µg of alamethicin/mg. The mixture was warmed to 37°C before the addition of a suitable aliquot of UGT isoform. The final reaction volume was 25 µl. The relative loss of fluorescence was followed at 37°C in a BMG PolarStar plate reader fitted with 390- and 510-nm filters for excitation and emission, respectively. Changes in the rate of reaction from an uninhibited reaction were calculated for each drug concentration.
Male and female Sprague-Dawley rats (two per sex, weighing 188–453 g at the time of the experiment) were supplied by Charles River (UK) Ltd. (Margate, Kent, UK). Animals were implanted with indwelling bile duct and duodenal cannulas under Isoflurane anesthesia. Cannulas were passed through the abdominal wall, truncated s.c., and externalized through the ventral surface of the tail. Animals were dosed with [14C]seliciclib in 50 mM HCl by oral gavage at a target dose of 10 mg/kg in a dose volume of 10 ml/kg.
All in vitro experiments were performed with duplicate determinations. Unless otherwise stated, results shown are representative of at least two independent determinations.
|
Results
|
|---|
In Vitro Identification of Metabolites of Seliciclib. Human microsomes were incubated with radiolabeled [14C]seliciclib to identify in vitro metabolites. Incubations used 1 to 100 µM [14C]seliciclib with pooled human liver microsomes (0.5 mg/ml) for 20 min, after which reactions were stopped, and samples were analyzed by reversed-phase HPLC. These incubations showed that seliciclib metabolism resulted in five major peaks, one of which coeluted with a carboxylic acid (code PMF30-128, molecular mass 368.2 Da) previously shown to be a major urinary metabolite of seliciclib in rodents (Nutley et al., 2005) (Fig. 1). The major peak observed was coded Met-4. The concentration dependence and time course of seliciclib metabolism were assessed by incubating microsomes with 1, 10, or 100 µM [14C]seliciclib for periods up to 60 min. Metabolism of seliciclib had an average half-life of 10 min, reaching a plateau by 20 min, whereafter metabolism did not increase with time. Interestingly the extent of metabolism was markedly reduced at the top concentration, falling from 56.6% loss of parent after 60 min at 1 µM seliciclib to 49.4% at 10 µM and to 17.3% at 100 µM. This result suggested that metabolism in human microsomes was saturated or possibly inhibited at high concentrations of seliciclib.
View larger version (11K):
[in this window]
[in a new window]
|
FIG. 3. Major human liver microsomal metabolites of seliciclib. Metabolites were determined by LC-MS/MS after incubation of seliciclib with human liver microsomes for up to 120 min.
|
|
A second set of experiments used nonradiolabeled seliciclib at 1 µM incubated with pooled human liver microsomes for up to 120 min (0.1–0.5 mg of protein/incubation). Samples were analyzed by LC-MS/MS to identify the major metabolic species. Under these conditions the rate of parent disappearance was comparable with that observed with radiolabeled seliciclib, with an average half-life of 16 min. Figure 2 shows results from a typical experiment. Major metabolic species determined by LC-MS/MS analysis in addition to PMF30-128 (mass 368.2 Da) included a peak at mass 352.2 (an oxidized form of seliciclib) and multiple peaks at mass 370.2, which might represent either 8-keto oxidation or ring-hydroxylated forms of seliciclib (Fig. 3).
Species Comparison of Seliciclib Microsomal Metabolism. We compared the metabolic profiles of [14C]seliciclib after incubations with microsomes prepared from rat, rabbit, mouse, dog, cynomolgus monkey, and man. [14C]Seliciclib was incubated at a concentration of 5 µM at 37°C, and samples were taken at time points up to 90 min. Samples were analyzed by reversed-phase HPLC, and examples of chromatograms are shown in Fig. 4. Results suggested that seliciclib was extensively metabolized in all species examined and that PMF30-128 was one of the major metabolites detected in all species, except in dog. Relative rates of metabolism of seliciclib (peak area ratios based on loss of parent over 30 min) are shown in Table 1. These data suggest that microsomal metabolism of seliciclib is rapid in most species, being highest in mouse and lowest in dog. Indeed, by 30 min all of the added seliciclib had been metabolized by mouse microsomes. Table 2 illustrates results from further incubations with seliciclib at 5 µM in dog, human, rabbit, cynomolgus monkey, rat, and mouse microsomes, showing the relative abundance of the nine most prevalent peaks determined by HPLC over the first 15 min.
View this table:
[in this window]
[in a new window]
|
TABLE 1 Rate and extent of metabolism of seliciclib in liver microsomes from six species Relative rates of metabolism are shown assayed as peak area ratios except for seliciclib and PMF30-128.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2 Relative abundance of metabolites from dog, rat, mouse, and human liver microsomes Data are percent of total eluted material except for seliciclib, which is percent metabolized after 15 min.
|
|
Seliciclib Is Metabolized by Cyp3a4 in Human Microsomes. [14C]Seliciclib was incubated (10 µM for 10 and 20 min) with 14 individual donor human microsome preparations, and the extent of microsomal metabolism (loss of parent seliciclib and formation of PMF30-128 and Met-4) and the (predetermined) activities of the major P450 enzymes therein were correlated. A significant correlation was obtained for loss of parent and with dextromethorphan N-demethylation (r = 0.89) and testosterone 6β-hydroxylation (r = 0.98). These reactions are selective for CYP3A4 and CYP3A4/5, respectively. The rate of seliciclib metabolism also showed a significant correlation with CYP2B6 activity (r = 0.87) (Fig. 5). Seliciclib metabolism did not correlate with CYP1A2, 2D6, 2A6, 2C9, 2E1, or 4A11 activity; however, one donor microsomal preparation was an extensive metabolizer of seliciclib, and this preparation also showed a correlation with CYP2C19 (r = 0.66), a result not obtained with any other microsomal fraction.
View larger version (9K):
[in this window]
[in a new window]
|
FIG. 5. Correlation analysis of metabolism of seliciclib with cytochrome P450 CYP3A4 and CYP2B6 levels. Different human liver microsome preparations characterized for cytochrome P450 activity were incubated with [14C]seliciclib and assayed for loss of seliciclib and appearance of PMF30-128 (data not shown) by LC-MS/MS.
|
|
The role of CYP3A4 in metabolism of seliciclib was confirmed by incubating pooled human liver microsomes with the CYP3A4 inhibitor ketoconazole. Metabolism of seliciclib was reduced from 615.1 to 114.9 pmol/min/mg in the presence of 2 µM ketoconazole, a reduction of 81.4% (Fig. 6). Furthermore, incubations of seliciclib with recombinant human CYP3A4 resulted in metabolic profiles comparable with those observed from incubations with pooled human microsomes, generating the five major metabolites, including Met-4 and PMF30-128 (data not shown). These results confirm the predominant role of cytochrome CYP3A4 in P450-mediated metabolism of seliciclib.
View larger version (9K):
[in this window]
[in a new window]
|
FIG. 6. Effect of ketoconazole incubation on seliciclib metabolism by human liver microsomes. A, 10 µM [14C]seliciclib, NADPH cofactor, 10-min incubation. B, 10 µM [14C]seliciclib, 2 µM ketoconazole, NADPH cofactor, 10-min incubation.
|
|
Seliciclib Routes of Elimination in Vivo. To determine the rate and routes of excretion of seliciclib we examined bile, feces, and urine from Sprague-Dawley rats given a single oral dose of 10 mg/kg [14C]seliciclib and sampled at 10 time points for up to 48 h. The primary route of total radioactivity excretion was found to be via the feces, with 65.1 and 53.2% of the administered dose, whereas urinary excretion accounted for 32.4 and 43.0% of the administered dose (percentages for males and females, respectively). Retention of radioactivity in the body was minimal after 48 h, as was elimination via respired CO2.
In bile duct-cannulated rats dosed with 10 mg/kg [14C]seliciclib, the major route of excretion was found to be through the bile (46.0 and 37.5% of administered dose, respectively), with urinary and fecal excretion now accounting for 24.1 and 20.7% of total radioactivity, respectively, in male rats and 30.3 and 19.4% in female rats (Table 3). Elimination appeared to be rapid, with almost 88% of the dose being eliminated in the first 24 h. Some small sex differences in elimination were noted: males excreted more radioactivity through the bile than females (46.0 versus 37.5%, respectively).
View this table:
[in this window]
[in a new window]
|
TABLE 3 Excretion of [14C]seliciclib by bile duct-cannulated rats Two male and female rats were bile duct-cannulated and dosed once with [14C]seliciclib at 10 mg/kg. At the time points indicated samples of urine, feces, or bile were taken and assayed for total radioactivity. Figures represent percent administered dose.
|
|
Appearance of Secondary Metabolites of Seliciclib in Rat Bile. We then examined the bile and urine samples obtained in the radiochemical study by LC-MS/MS to determine the identity and extent of seliciclib metabolites therein. Figure 7 shows the relative abundance of the six major species identified in bile and urine. With the exception of seliciclib and PMF30-128, quantities were assessed on the basis of peak area ratios. The species detected almost exclusively in rat urine was the carboxylate metabolite of seliciclib, PMF30-128, accounting for 99.5 and 97.4% of material detected in male and female rats, respectively. This finding is in agreement with the previous determination by Nutley et al. (2005) who showed that PMF30-128 is the principal component of mouse urine after seliciclib treatment. A small amount of seliciclib appeared to be excreted unchanged in urine (2.6% and 0.5% in female and male rats, respectively).
View larger version (14K):
[in this window]
[in a new window]
|
FIG. 7. Determination of seliciclib metabolites in rat bile and urine. The relative abundance of seliciclib-derived metabolites in rat urine (A) and bile (B) was determined by LC-MS/MS (peak area ratios). Figures show percent relative abundance. Gluc, glucuronide.
|
|
A broader spectrum of metabolites was found in rat bile (Fig. 7B). Seliciclib and PMF30-128 were detected, as well as a glucuronidated form of seliciclib and three other species, one with a mass of 529 (subsequently identified as the glucuronide of a reduced form of seliciclib) and two with a mass of 547 (identified as glucuronidated forms of a phenol ring-hydroxylated metabolite of seliciclib). To date we have failed to detect a direct glucuronide of PMF30-128, and neither have we detected evidence of acyl glucuronides.
We attempted to make a preliminary identification of the likely UGT isoforms involved in glucuronidation of seliciclib by incubating recombinant human UGTs with seliciclib and assaying for presence of the O- and N-glucuronides of seliciclib. Only with UGT1A9 was N-glucuronidation observed, and only with UGT2B7 and 1A3 were O-glucuronides detected (data not shown).
Inhibition of Primary and Secondary Metabolism by Seliciclib. Using recombinant human cytochrome P450s, we determined the sensitivity of these enzymes to inhibition by seliciclib (Table 4). Activity was confirmed by the use of appropriate substrates for each enzyme, whereas positive control inhibitor compounds confirmed the potential of the enzymes to be inhibited. No inhibition due to seliciclib was observed with CYP2A6 or CYP2E1; however, the remaining seven enzymes were inhibited with IC50 values ranging from 3.2 to 124 µM. Seliciclib was most potent as an inhibitor of CYP3A4 and CYP2C9/1 with IC50 values of 3.2 and 3.8 µM, respectively, and was a moderate inhibitor of CYP2D6 and CYP2C19 with IC50 values of 17.0 and 19.2 µM, respectively.
View this table:
[in this window]
[in a new window]
|
TABLE 4 Inhibition of recombinant human cytochrome P450 enzymes by seliciclib Recombinant human cytochrome P450 enzymes were incubated with seliciclib (0.137–300 µM) and assayed for activity.
|
|
In a similar fashion, a number of isoforms of recombinant human glucuronidating enzymes (UGTs) were tested for sensitivity to seliciclib (Table 5). Seliciclib was shown to inhibit UGT1A1 most potently, with an IC50 of 7.1 µM. Several isoforms were not inhibited by seliciclib at concentrations in excess of 100 µM, including UGTs 1A7, 1A8, and 1A10.
View this table:
[in this window]
[in a new window]
|
TABLE 5 In vitro inhibition of human recombinant UGT isoforms by seliciclib Human recombinant UGT enzymes were incubated with seliciclib at concentrations up to 200 µM for 60 min.
|
|
|
Discussion
|
|---|
Our results show that seliciclib (R-roscovitine, CYC202) is metabolized extensively in microsomes from several species, including man. One of the major metabolites in most species tested here was a carboxylate (PMF30-128), which was previously shown to have lower potency against the target CDK2 and was the principal species observed in urine from rats and mice dosed with seliciclib (Nutley et al., 2005) and also was detected in urine and tissues from rats dosed i.v. roscovitine (Vita et al., 2005).
We further show here that during oral dosing of seliciclib in rats the urinary route of excretion is efficient, accounting for 23 to 29% of the total radioactivity dosed to rats in a mass balance study. In a study in which roscovitine was dosed to rats i.v. at 25 mg/kg, the proportion of roscovitine in urine was found to be <1% of the administered dose, with the carboxylate metabolite accounting for 10-fold more than the parent (Vita et al., 2005). In our study we measured only total excreted radioactivity and did not break this down into its constituent components (seliciclib plus metabolites), but it may be that in rats a higher proportion of elimination occurs by the urinary route when roscovitine is dosed orally than when it is dosed i.v. Our findings are supported by those of Nutley et al. (2005) who showed that in mice elimination of seliciclib in urine accounted for 0.02% of a 50 mg/kg dose, but that PMF30-128 could account for 65 to 68% of the dose. However, this high proportion of PMF30-128 elimination in urine did not appear to depend on the route of administration.
The other major metabolites of seliciclib determined in vivo are conjugates, with glucuronidation being the principal route of metabolism and excretion of seliciclib in rats, particularly through the action of UGTs 1A1, 1A3, and 2B7. Up to 46% of dosed radioactivity in rats appears in the bile, whereas a further 20% of dosed radioactivity appears to pass, presumably unchanged, through the digestive tract. After a single dose of [14C]seliciclib, almost 90% of the total radioactivity has been excreted by urinary or faecal routes within 24 h of dosing. Preliminary investigations have shown that glucuronides of seliciclib are also observed in plasma from dogs dosed with the drug and in vitro in human hepatocytes incubated with seliciclib (data not shown). Conjugation therefore appears to be a potential route of metabolism after oral dosing of seliciclib in species other than rat.
Patterns of microsomal metabolism in different species were observed to be largely similar. The carboxylate metabolite of seliciclib (PMF30-128) was found in microsomes from all species except dogs, being particularly prominent in mouse, monkey, and human. However, PMF30-128 has subsequently been assayed in plasma samples from dogs dosed with seliciclib (data not shown). The lack of detection of PMF30-128 in microsomal fractions may reflect the slower rate of metabolism of seliciclib in dog microsomes. In subsequent toxicology studies, dog and rat were chosen as test species: rats having a rate of metabolism higher than that of humans, and dogs having a rate slower than that of humans. Vita et al. (2005) showed production of three major metabolites after i.v. injection of roscovitine to rats. These three metabolites correspond to the carboxylate PMF30-128, the 8-keto metabolite of mass 370, and the dealkylated species of mass 312, which we too observed in incubations with human microsomes. We also detected the carboxylate metabolite PMF30-128 in urine and bile from rats. Interestingly, Vita et al. (2005) did not mention the detection of glucuronides after i.v. administration.
Oxidative phase I metabolism of seliciclib appears to be a common and efficacious route of compound metabolism, being observed in microsomes from several species. The formation of the seliciclib carboxylate is shown here to occur via the cytochromes P450 CYP3A4 and CYP2B6. Our results suggested that CYP2B6-mediated metabolism can occur at low levels of the enzyme, whereas the rate of CYP3A4-mediated metabolism is low until higher levels of the enzyme are present. CYP3A4 is responsible for metabolism of many xenobiotics (Guengerich, 1999) and has been previously shown to be at least partly responsible for metabolism of bohemine, a trisubstituted purine derivative with a close structural relationship to seliciclib (Rypka et al., 2002). We also made the interesting observation that seliciclib inhibits CYP3A4 in vitro at micromolar concentrations that can be achieved in the clinic (de la Motte and Gianella-Borradori, 2004). Because many anticancer drugs in clinical use are metabolized by CYP3A4, as are concurrent medications such as antibiotics and pain-relieving drugs, careful clinical monitoring of potential drug-drug interactions when seliciclib is dosed in combinations is important. To date, however, no evidence for drug-drug interactions has been noted in the clinic (Benson et al., 2007). Inhibition of its own phase I metabolism might also result in seliciclib accumulation in the blood plasma. No evidence of such accumulation has been found in human trials (Benson et al., 2007.)
The other P450 capable of metabolizing seliciclib is CYP2B6, which accounts for only 2 to 4% of overall drug metabolism (Evans and Relling, 1999); however, it is responsible for metabolism of anticancer drugs such as cyclophosphamide and ifosfamide (Code et al., 1997). CYP2B6 was found to be much less sensitive to inhibition by seliciclib and might therefore take over the functions of CYP3A4 if this were inhibited after repeated dosing of seliciclib.
Secondary metabolism of seliciclib was found to be a significant route of compound clearance, with UGTs 1A1, 1A3 and 2B7 shown to be involved in this process. UGT1A1 is involved in conjugation of bilirubin, and it is possible that seliciclib might compete with bilirubin at UGT1A1, which in extreme cases could lead to liver damage. The UGTs, like the P450s, also appear sensitive to inhibition by seliciclib. Several UGTs were inhibited at micromolar plasma levels likely to be encountered in the clinic, viz. UGT1A1, 1A6, and 1A9. Because UGT1A9 appears to be responsible for glucuronidation of seliciclib (at least in vitro), it seems possible that seliciclib could inhibit its own secondary metabolism after repeated dosing. However, we have determined in preliminary experiments not reported here that seliciclib can also induce expression of UGTs and of P450s; hence, the mixed inhibitory and inductive profile of seliciclib may ultimately combine to produce no major effects in man, and, indeed, in a clinical trial using seliciclib dosed twice daily for 7 days no evidence has been found for drug accumulation or for drug-drug interactions arising from inhibition of P450s or UGTs (Benson et al., 2007.)
|
Acknowledgments
|
|---|
We thank our colleagues Andy Plater and Lisa Logie for assistance with microsomal stability and UGT inhibition experiments. We thank Richard Cole, Guy Webber, and their colleagues at BioDynamics Research Ltd. (Rushden, Northants, UK) for assistance with the radiolabeled metabolism studies, David Mann and colleagues at Charles River Laboratories (Tranent, UK) for assistance with the radiolabeled distribution experiments, and Robert Powrie and colleagues at CXR Biosciences (Dundee, UK) for assistance with the studies on inhibition of cytochromes P450 and reaction phenotyping.
|
Footnotes
|
|---|
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.107.019232.
ABBREVIATIONS: CDK, cyclin dependent kinase; P450, cytochrome P450; HPLC, high-performance liquid chromatography; LC, liquid chromatography; MS/MS, tandem mass spectrometry; UGT, UDP glucuronosyltransferase; Met, metabolite.
Address correspondence to: Dr. Steven McClue, Cyclacel Ltd, James Lindsay Place, Dundee DD1 5JJ, UK. E-mail: smcclue{at}cyclacel.com
|
References
|
|---|
Benson C, White J, De Bono J, O'Donnell A, Raynaud F, Cruickshank C, McGrath H, Walton M, Workman P, Kaye S, Cassidy J, Gianella-Borradori A, Judson I, and Twelves C (2007) A phase I trial of the selective oral cyclin-dependent kinase inhibitor seliciclib (CYC202; R-roscovitine), administered twice daily for 7 days every 21 days. Br J Cancer 96: 29-37.[CrossRef][Medline]
Cervenková K, Belejova M, Chmela Z, Rypka M, Riegrova D, Michnova K, Michalikova K, Surova I, Brejcha A, Hanus J, et al. (2003) In vitro glycosidation potential toward olomoucine-type cyclin-dependent kinase inhibitors in rodent and primate microsomes. Physiol Res 52: 467-474.[Medline]
Chmela Z, Vesel J, Lemr K, Rypka M, Hanus J, Havlícek L, Krystof V, Michnova L, Fuksova K, and Lukes J (2001) In vivo metabolism of 2,6,9-trisubstituted purine-derived cyclin-dependent kinase inhibitor bohemine in mice: glucosidation as the principal metabolic route. Drug Metab Dispos 29: 326-334.[Abstract/Free Full Text]
Code EL, Crespi CL, Penman BW, Gonzalez FJ, Chang TK, and Waxman DJ (1997) Human cytochrome P4502B6: interindividual hepatic expression, substrate specificity, and role in procarcinogen activation. Drug Metab Dispos 25: 985-993.[Abstract/Free Full Text]
de la Motte S and Gianella-Borradori A (2004) Pharmacokinetic model of R-roscovitine and its metabolite in healthy male subjects. Int J Clin Pharmacol Ther 42: 232-239.[Medline]
Evans WE and Relling MV (1999) Pharmacogenomics: translating functional genomics into rational therapeutics. Science 286: 487-491.[Abstract/Free Full Text]
Fischer PM and Gianella-Borradori A (2005) Recent progress in the discovery and development of cyclin-dependent kinase inhibitors. Expert Opin Investig Drugs 14: 457-477.[CrossRef][Medline]
Guengerich FP (1999) Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu Rev Pharmacol Toxicol 39: 1-17.[CrossRef][Medline]
Havlícek L, Hanus J, Vesel J, Leclerc S, Meijer L, Shaw G, and Strnad M (1997) Cytokinin-derived cyclin-dependent kinase inhibitors: synthesis and cdc2 inhibitory activity of olomoucine and related compounds. J Med Chem 40: 408-412.[CrossRef][Medline]
McClue SJ, Blake D, Clarke R, Cowan A, Cummings L, Fischer PM, MacKenzie M, Melville J, Stewart K, Wang S, et al. (2002a) In vitro and in vivo antitumor properties of the cyclin dependent kinase inhibitor CYC202 (R-roscovitine). Int J Cancer 102: 463-468.[CrossRef][Medline]
McClue S, Fischer PM, Blake D, Clarke R, Duff S, Krauss E, Cummings L, Stewart K, Melville J, Wang S, et al. (2002b) Studies on the mechanism of action of CYC202 (R-roscovitine). Proc Am Assoc Cancer Res 43: 3303.
Meijer L and Pondaven P (1988) Cyclic activation of histone H1 kinase during sea urchin egg mitotic divisions. Exp Cell Res 174: 116-129.[CrossRef][Medline]
Neant I and Guerrier P (1988) 6-Dimethylaminopurine blocks starfish oocyte maturation by inhibiting a relevant protein kinase activity. Exp Cell Res 176: 68-79.[CrossRef][Medline]
Nutley BP, Raynaud FI, Wilson SC, Fischer PM, Hayes A, Goddard PM, McClue SJ, Jarman M, Lane DP, and Workman P (2005) Metabolism and pharmacokinetics of the cyclin-dependent kinase inhibitor R-roscovitine in the mouse. Mol Cancer Ther 4: 125-139.[Abstract/Free Full Text]
Rypka M, Vesel J, Chmela Z, Riegrova D, Cervenková K, Havlícek L, Lemr K, Hanus J, Cerny B, Lukes J, et al. (2002) In vitro biotransformation of 2,6,9-trisubstituted purine-derived cyclin-dependent kinase inhibitor bohemine by mouse liver microsomes. Xenobiotica 32: 1017-1031.[CrossRef][Medline]
Shapiro G. (2004) Preclinical and clinical development of the cyclin-dependent kinase inhibitor flavopiridol. Clin Cancer Res 10: 4270S-4275S.[Abstract/Free Full Text]
Vesel J, Havlícek L, Strnad M, Blow JJ, Donella-Deana A, Pinna L, Letham DS, Kato J, Detivaud L, and Leclerc S (1994) Inhibition of cyclin-dependent kinases by purine analogues. Eur J Biochem 224: 771-786.[Medline]
Vita M, Abdel-Rehim M, Olofsson S, Hassan Z, Meurling L, Siden A, Siden M, Pettersson T, and Hassan M (2005) Tissue distribution, pharmacokinetics and identification of roscovitine metabolites in rat. Eur J Pharm Sci 25: 91-103.[CrossRef][Medline]
Wang S, McClue SJ, Ferguson JR, Hull JD, Stokes S, Parsons S, Westwood R, and Fischer PM (2001) Synthesis and configuration of the cyclin-dependent kinase inhibitor roscovitine and its enantiomer. Tetrahedron Asymmetry 12: 2891-2894.[CrossRef]
This article has been cited by other articles:
|
|
|
|
 
M. Siller, P. Anzenbacher, E. Anzenbacherova, K. Dolezal, I. Popa, and M. Strnad
Interactions of Olomoucine II with Human Liver Microsomal Cytochromes P450
Drug Metab. Dispos.,
June 1, 2009;
37(6):
1198 - 1202.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
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]
|
|
|
![[Feedback]](/icons/banner/feedbackACT.gif)
![[For Subscribers]](/icons/banner/subscriberACT.gif)
![[Archive]](/icons/banner/archiveACT.gif)
![[Search]](/icons/banner/searchACT.gif)
![[Contents]](/icons/banner/tocACT.gif)
Top
Abstract
Materials and Methods
