![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Drug Discovery Research, Johnson & Johnson Pharmaceutical Research & Development LLC, Spring House, Pennsylvania
(Received January 5, 2006; accepted January 27, 2006)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). UGT enzymes are normally divided into two gene families, UGT1A and UGT2B, based on their sequence homologies (Mackenzie et al., 1997). UGT1A enzymes have been shown to catalyze glucuronidation of a vast number of substrates at a diverse array of functional groups to form a variety of glucuronides. UGT1A-mediated N-glucuronidation represents an important metabolic pathway, as evidenced by the fact that many compounds, such as alkylamines, arylamines, hydroxylamines, and various heterocyclic amines, form many different N-glucuronides both in vitro and in vivo (Chiu and Husky, 1998; Green and Tephly, 1998). These types of conjugates can be broadly categorized as aliphatic and/or aromatic amine-linked glucuronides (N-glucuronide) and quaternary ammonium-linked glucuronides (N+-glucuronide). In contrast to the UGT1A family, the UGT2B enzymes are primarily responsible for the metabolism of steroids, except for UGT2B7, which has been shown to metabolize many different drugs (Radominska-Pandya et al., 2001).
Most UGT1A proteins are expressed predominantly in the liver, but they have been also found in several extrahepatic tissues. For example, UGT1A1, 1A3, 1A4, 1A6, and 1A9 are highly abundant in the liver, whereas UGT1A7, 1A8, and 1A10 are found exclusively in extrahepatic tissues such as gastric, colon, and biliary tissue, respectively (Strassburg et al., 1997). Identification of individual UGTs responsible for the metabolism of a particular drug is important for predicting the major phase II elimination route of the drug in humans.
A tricyclic indenopyrazole compound, 6,7-(dimethoxy-2,4-dihydroindeno[1,2-c]pyrazol-3-yl)-(3-fluoro-phenyl)-amine (JNJ-10198409; Fig. 1), was discovered as a potent inhibitor of platelet-derived growth factor receptor tyrosine kinase (Ho et al., 2005). Overexpression of platelet-derived growth factor and its receptors have been implicated as drivers of tumor cell proliferation in some cancers. Anticancer drugs such as paclitaxel normally act as antiproliferative agents that inhibit tumor cell growth directly (Inoue et al., 2000). During the past decade, the antiangiogenesis agents blocking the formation of neovasculature in a tumor in situ have shown great promise as cancer drugs (Bocci et al., 2004; Li, 2004). JNJ-10198409 represents a novel anticancer agent that, as a single chemical entity, has both antiangiogenic and tumor cell antiproliferative activities (Mei et al., 2004; Tuman et al., 2004; D'Andrea et al., 2005).
|
JNJ-10198409 contains a pyrazole ring and an aniline, which are the potential functional groups for N-glucuronidation. Although the pyrazole ring is a common structural moiety, N-glucuronidation at the nitrogens on the pyrazole has not been documented in the literature. Therefore, it was of interest to investigate the glucuronidation of JNJ-10198409. More importantly, information about glucuronidation would be valuable for both evaluating potential drug-drug interactions and predicting the drug metabolism fate in humans.
In the present study, in vitro N-glucuronidation was characterized as the major metabolic pathway of JNJ-10198409 by metabolic profiling, liquid chromatography-tandem mass spectrometry (LC-MS/MS) and nuclear magnetic resonance (NMR) analyses. Furthermore, UGT enzymes that are responsible for formation of N-glucuronides were characterized by using human cDNA-expressed UGTs and by enzyme kinetic analyses.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Other reagents were purchased from Sigma (St. Louis, MO), which included alamethicin UDPGA, Escherichia coli ß-glucuronidase, propofol, ß-nicotinamide adenine dinucleotide phosphate (NADP+), glucose 6-phosphate, and glucose-6-phosphate dehydrogenase. All HPLC and NMR solvents were obtained from standard vendors. Oxidative Metabolism of JNJ-10198409 in Liver Microsomes. All microsomal incubations were performed at 37°C in a water bath. JNJ-10198409 was separately mixed with human, monkey, or rat liver microsomal protein in 50 mM phosphate buffer (pH 7.4). After a 5-min preincubation at 37°C, reactions were initiated by the addition of a NADPH-generating system to give a final volume of 1.0 ml. The final reaction mixture contained 10 µM JNJ-10198409, 1 mg/ml microsomal protein, 1.3 mM NADP+, 3.3 mM glucose 6-phosphate, 0.4 U/ml glucose-6-phosphate dehydrogenase, and 3.3 mM magnesium chloride. After a 60-min incubation, reactions were terminated by the addition of 250 µl of acetonitrile. Samples were centrifuged at 10,000g for 10 min to pellet the precipitated protein, and supernatants were subjected to LC-MS/MS analysis of oxidative metabolites formed by P450s.
Glucuronidation of JNJ-10198409 in Liver Microsomes. JNJ-10198409 was separately mixed with human, monkey, or rat liver microsomal protein in 50 mM Tris buffer (pH 7.4) supplemented with alamethicin in aqueous solution (stock solution 1 mg/ml). After a 5-min preincubation at 37°C, reactions were initiated by the addition of UDPGA to give a final volume of 0.5 ml. The final reaction mixture contained 2 mM UDPGA, 25 µg/ml alamethicin, JNJ-10198409, and microsomal protein at desired concentrations. After incubation for designated times, reactions were terminated by the addition of 250 µl of acetonitrile. Samples were centrifuged at 10,000g for 10 min to pellet the precipitated protein, and supernatants were subjected to LC-MS/MS for direct analysis of glucuronides.
Incubations of Glucuronides with ß-Glucuronidase. Glucuronides were first generated by the in vitro incubation with rat liver microsomes as described above. The total volume of incubation was 1 ml. After centrifugation, the supernatant was transferred to a 1.5-ml tube, and the sample was concentrated to approximately 100 µl in a spin-vacuum drier. A total volume of 900 µl of water was added to the concentrated sample, and the reconstituted sample was subsequently divided into three fresh tubes. One tube was stored at –80°C as a control. The second tube was incubated with ß-glucuronidase (500 U/ml) for 30 min at 37°C, and then was stored at –80°C. The third tube remained at room temperature for 6 h. Finally, all three samples were analyzed for glucuronides using LC-MS/MS as described below.
Isolation of Glucuronides. A pilot study was first conducted with ovine liver microsomes, and results confirmed that JNJ-10198409 formed three glucuronides identical to those formed in rat and human liver microsomes. To be cost-effective, a large-scale incubation was performed using ovine liver microsomal UDP-glucuronyltransferase and UDP-glucuronate using a literature procedure (Kren et al., 2000; Hubl and Stevenson, 2001). The reaction mixture contained ovine liver microsomal protein (2 mg/ml), UDPGA (50 mM), and dithiothreitol (0.16 mg/ml) in 200 mM Tris buffer (pH 8.0) containing 6 mM CaCl2. The reaction was started by gradually adding JNJ-10198409 in DMSO to the reaction mixture. Incubation was performed for 24 h in a water bath set at 35°C. After centrifugation, the supernatant was applied to a 3.5 x 20 cm column of Merck (Darmstadt, Germany) LiChroprep RP-18 silica that was preequilibrated with 5% methanol supplemented with 0.025% ammonium trifluoroacetate. After washing with 100 ml of 5% methanol, a gradient of 5 to 65% methanol was applied over 40 min. Fractions containing glucuronides were collected and pooled. Preparation was repeated many times to accumulate enough material. Solvents were removed by rotary evaporation. The resulting solid from the RP-18 column was dissolved in Milli-Q water, and purified by preparative HPLC using a 250 x 21.4 cm Jupiter C-18 column (15 µ, 100 Å) (Phenomenex, Torrance, CA) and running a gradient of 2.5 to 50% methanol over 45 min. Fractions containing individual glucuronides were collected and dried by rotary evaporation to obtain solid materials. Purity of isomeric glucuronides was checked by LC-MS/MS and NMR analysis.
Glucuronidation of JNJ-10198409 by cDNA-expressed UGTs. Glucuronide formation rates were determined using Supersomes containing cDNA-expressed individual UGTs (UGT1A1, 1A2, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 2B4, 2B7, 2B15, and 2B17). All incubations were performed in triplicate in a 96-well plate at 37°C in a water bath. JNJ-10198409 at desired concentrations was preincubated for 5 min with Supersomal protein (0.25 mg/ml) in 50 mM Tris buffer (pH 7.4, 3.5 mM MgCl2). The conjugation reaction was initiated by the addition of 2 mM UDPGA. At designated incubation times, the reaction was terminated by the addition of ice-cold acetonitrile. Samples were centrifuged at 10,000g for 20 min to pellet the precipitated protein, and the resulting supernatants were transferred to HPLC vials for LC-MS/MS analysis of formed glucuronides.
Determination of the Kinetic Constants of Glucuronides. The kinetic constants of Km and Vmax were determined for two glucuronides (designated Glu-A and -B) by incubating JNJ-10198409 at various concentrations (0, 0.13, 0.4, 1.2, 3.7, 11.1, 33.3, and 100 µM) with either pooled human liver microsomes (0.5 mg/ml) or Supersomes (0.25 mg/ml) containing individual UGTs under the same conditions described above. Three replicates of incubation were performed at each drug concentration. The data were analyzed by iterative nonlinear least-squares regression analysis using Prism 3.5 (GraphPad Software Inc., San Diego, CA), fitting the data to the Michaelis-Menten equation.
Propofol Inhibition of JNJ-10198409 Glucuronidation. The effect of propofol, a specific inhibitor of UGT1A9, on JNJ-10198409 glucuronidation was investigated using pooled human liver microsomes. Propofol (100 µM) was added to incubation mixtures containing JNJ-10198409 at various concentrations to investigate the involvement of UGT1A9. The microsomal protein concentration was 0.25 mg/ml, and the incubation time was 10 min. Formed glucuronides were analyzed by LC-MS/MS as described below. Glucuronidation rates were expressed relative to the control (no inhibitor).
LC-MS/MS Analysis of Glucuronides. LC-MS/MS analyses were performed on a Micromass (Manchester, UK) Quattro Micro triple quadrupole mass spectrometer interfaced to an Agilent 1100 (Agilent Technologies, Palo Alto, CA) HPLC system. LC-MS/MS analyses were conducted using electrospray ionization (ESI) under either the positive or negative ion mode. The capillary voltage was 3.1 kV, and the cone voltage was set at 20 V to minimize in-source dissociation of glucuronides (Yan et al., 2003). The source temperature was set at 120°C, and the desolvation temperature was 300°C. The collision gas was argon.
An Agilent Zorbax SB C18 column (2.1 x 100 mm) was used for the chromatographic separations. The starting mobile phase consisted of 95% water (0.5% acetic acid), and glucuronides were eluted using a linear gradient of 95% water to 95% acetonitrile over 7 min at a flow rate of 0.3 ml/min. At 7 min, the column was flushed with 95% acetonitrile for 2 min before reequilibration at initial conditions for 2 min. During the run, the divert valve was activated to divert the HPLC eluant to waste for the first minute of elution, and then switched to the mass spectrometer for analysis. LC-MS/MS analysis was carried out on 10-µl aliquots from incubations. To obtain a tandem spectrum, the mass spectrometer was operated in the product ion scan mode. For quantification of glucuronides, the multiple reaction monitoring mode was used to detect transitions at m/z 502
326.
For detecting glucuronides by ESI in the negative mode, 2 mM ammonium acetate was added to aqueous mobile phase. The same LC gradient profile was utilized for chromatographic separation. The MS transition m/z 500324 was used. Data were processed using the Masslynx v3.5 software from Micromass.
NMR Analysis of Glucuronides. One-dimensional (1D) and two-dimensional (2D) NMR spectra were obtained on a Bruker-Biospin, Inc. (Billerica, MA) DRX-500 MHz or a DMX-600 MHz spectrometer using XWINNMR acquisition and processing software. The spectra were collected using a z-gradient inverse broadband probe (500 MHz) or a triple tuned three-axis gradient probe (600 MHz) at various temperatures. Typical 1D proton (1H) NMR spectra were obtained using standard pulse and acquisition sequences. Typical 1D carbon (13C) NMR (150.917 MHz) spectra were obtained using standard sequences with composite pulse decoupling techniques. The 2D homonuclear correlated spectroscopy gradient (g-COSY) experiment was used to obtain 1H-1H coupling correlations. A mixing pulse of 90 degrees was used and one scan was measured for 128 values of t1 to give a total matrix of 128 x 2048 complex points with a spectral width of 2670.940 Hz. The free induction decay was multiplied with a sine-bell function in both dimensions and made symmetrical about the diagonal. The 2D nuclear Overhauser enhancement spectroscopy gradient (g-NOESY) experiment was used to obtain 1H-1H spatial correlations. A mixing time of 500 µs was used and 64 scans were measured for 256 values of t1 to give a total matrix of 256 x 2048 complex points with a spectral width of 4370.629 Hz. The free induction decay was multiplied with a squared sine-bell in both dimensions. The 2D heteronuclear multiple bond correlated spectroscopy gradient (g-HMBC) experiment was used to obtain long-range 1H-13C couplings. The delay was optimized for 8-Hz couplings. A total of 2048 scans were measured for 128 values of t1 to give a total data matrix of 128 x 2048 complex points. The carbon F1 spectral width was 21,250 Hz and the proton F2 spectral width was 4370.629 Hz. Data were processed using squared sine-bell function in both dimensions. The compounds were dissolved in either dimethyl sulfoxide-d6 (DMSO-d6) or methanol-d4 (CD3OD) in 5-mm tubes. The chemical shifts (
) measured in parts per million (ppm) were referenced internally to DMSO (H 2.50 ppm and C 39.5 ppm) or CH3OH (H 3.30 ppm and C 49.15 ppm) (Tables 1 and 2).
|
|
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Yan and Caldwell, 2003). In rat liver microsomes, LC-MS/MS with ESI analyses detected three isomeric metabolites (MH+ at m/z 502), designated as Glu-A (RT 5.3 min), Glu-B (RT 5.6 min), and Glu-C (RT 4.8 min), respectively (Fig. 2A). In incubations with liver microsomes derived from monkey, Glu-B was more abundant than Glu-A (Fig. 2B); Glu-A and Glu-B were at a comparable level in human liver microsomes (Fig. 2C). The level of Glu-C was not detectable in either human or monkey liver microsomes.
|
Tandem MS spectra of all three components are shown in Fig. 3. The predominant product ion appeared at m/z 326 for all three metabolites. The cations at m/z 326 (M – 176)+ likely resulted from the cleavage of the glycosidic bond with transfer of a proton from the glucuronic acid moiety to the aglycone. The Glu-A MS spectrum (3A) also contained prominent product ions at m/z at 350 and 368. All three glucuronides were detected by ESI-MS at m/z 500 in the negative mode, although signal-to-noise ratios were significantly lower compared with that observed in the positive mode (data not shown).
|
). The NMR study also demonstrated that there was significant hindered rotation about the pyrazole-N(H)-phenyl ring systems. From a structural standpoint, if JNJ-10198409 is primarily in form II, it possesses two secondary amines that could result in the formation of N-glucuronides. Glucuronidation of the tertiary amine in form II would result in the formation of an N+-glucuronide. Since the equilibrium ratio of the tautomeric forms of JNJ-10198409 are dependent on environmental factors such as pH, temperature, and median ionic strength, it is likely that form I will also be present under our in vitro experimental conditions. In this case, instead of N+-glucuronidation at the tertiary amine N(2) in form II, glucuronidation of the secondary amine in form I would result in the formation of an N-glucuronide.
In the positive ESI mode, the two types of N-glucuronide metabolites cannot be distinguished from each other. For example, glucuronidation of JNJ-10148409 at a secondary amine would result in the formation of metabolites with molecular mass of 501 Da. Under positive ESI conditions, molecular ions at m/z 502 would be detected. Glucuronidation of JNJ-10148409 at a tertiary amine would result in the formation of positively charged metabolites with a molecular mass of 502 Da. Under positive ESI conditions, cations at m/z 502 would be detected. Thus, under positive ESI conditions, both types of N-glucuronides would appear at m/z 502. The two types of N-glucuronide metabolites can be distinguished by LC-ESI-MS in the negative mode. Glucuronidation of JNJ-10148409 at a secondary amine would result in the formation of metabolites with a molecular mass of 501 Da, which appear as molecular ions at m/z 500 (M – 1) in the negative ESI mode. Since the N+-glucuronide contains a positive charge, it would not be detected in the negative mode (Vashishtha et al., 2002). Our results indicated that all three glucuronides were formed at secondary amines. Definitive identification of individual glucuronides could not be made solely by MS/MS analysis.
NMR Identification of Glucuronides. To determine the structures of individual glucuronides by NMR, Glu-A, Glu-B, and Glu-C were purified from large-scale ovine liver microsomal incubations by semipreparative HPLC. The purified conjugates had HPLC retention times and MS/MS characteristic cations identical to those generated in rat, monkey, and human microsomal incubations.
The structures of the three glucuronides isolated from ovine liver microsomal incubations are shown in Fig. 4. The proton (1H) NMR and carbon (13C) chemical shift assignments of JNJ-10148409 and its three corresponding N-glucuronides are shown in Tables 1 and 2, respectively. A representative proton NMR spectrum (i.e., Glu-B) is shown in Fig. 5. The proton resonance assignments were based on expected proton coupling relationships, chemical shifts, g-COSY, and g-NOESY experiments. For the sugar moiety, the coupling constant of the H1' proton (i.e., J 1', 2' = 7.8 Hz) was indicative of ß-glucuronide stereochemistry (Smith and Benet, 1986). The carbon resonance assignments were based on 1H-13C g-HMBC experiments.
|
The structural assignments of the three N-glucuronides are based on a series of 2D NMR experiments involving g-NOESY (i.e., 1H-1H spatial correlation) and g-HMBC (i.e., 1H-13C long range coupling) experiments. The structural assignment of Glu-A was based primarily on the observation of a close-proximity 1H-1H spatial relationship between the H1' proton on the glucuronic acid moiety and the protons H15/H19 on the phenyl ring. In addition, the HMBC experiment indicated that there was a three-bond coupling between H1' and the quaternary carbon C14 on the phenyl ring. (i.e., H1'-N13-C14). These results are unique to the Glu-A structure. The structural assignment of Glu-B was based on the observation of a 1H-1H spatial relationship between the H1' proton on the glucuronic acid moiety and the H10 aromatic proton on the indenopyrazole ring. The HMBC experiment indicated that there was a three-bond coupling between H1' and the quaternary carbon C12 on the indenopyrazole ring. (i.e., H1'-N1-C12). These results are distinctive to the Glu-B structure. The structural assignment of Glu-C was based on the HMBC experiment, which indicated that there was a three-bond coupling between H1' and the quaternary carbon C3 on the pyrazole ring. (i.e., H1'-N2-C3).
Degradation of N-Glucuronides of JNJ-10198409 by ß-Glucuronidase. Three glucuronides were generated in microsomal incubations with rat liver microsomes. Glucuronides formed in incubation mixtures were treated with ß-glucuronidase. It was found that both Glu-A and Glu-C completely disappeared after treatment with ß-glucuronidase. Under the same conditions, Glu-B remained intact after a 30-min incubation at 37°C (data not shown). In the absence of ß-glucuronidase, both Glu-A and Glu-C underwent slow degradation; after sitting for 6 h at room temperature, 67% and 72% of Glu-A and Glu-C remained, respectively, but no significant degradation was observed for Glu-B under the same conditions. It was also found that purified Glu-A and Glu-C underwent degradation, and degradation was not inhibited by saccharolactone.
Kinetics of Formation of Glu-A and Glu-B in HLM. Due to clinical relevance in humans, kinetics studies were undertaken focused on the formation of Glu-A and Glu-B in human liver microsomes. Initial velocity conditions were established, which were linear with respect to both protein concentration and incubation time. Primary data also indicated that the formation rates of Glu-A and Glu-B were linear with incubation time up to 25 min and with microsomal protein concentration up to 0.75 mg/ml. Therefore, kinetic studies of glucuronidation were carried out at a protein concentration of 0.5 mg/ml over 10 min. Incubations of JNJ-10198409 over the concentration range of 1 to 100 µM with pooled human liver microsomes resulted in the concentration-dependent formation of Glu-A and Glu-B. Kinetic analyses showed that formation of Glu-A and Glu-B exhibited atypical kinetics. As shown in the Eadie-Hofstee transformation plot (Fig. 6), a nonlinear plot was observed for both GluA and Glu-B. A biphasic Michaelis-Menten model was applied to estimate the apparent kinetic parameters (Km1 and Vmax1) of the high affinity binding, with the assumption that two enzymes catalyzed the glucuronidation (see Table 3). The calculated intrinsic clearances (Vmax1/Km1) for the formation of Glu-A and Glu-B were 2002 nmol min–1 mg–1/µM and 2403 nmol min–1 mg–1/µM, respectively. Because of low solubility, data points at high substrate concentrations were too limited to estimate Michaelis-Menten kinetic parameters (Km2 and Vmax2) of the low affinity binding.
|
|
Glucuronide Formation in Recombinant UGT Enzymes. The formation of Glu-A and Glu-B was investigated in microsomes derived from baculovirus-infected insect cells (Supersomes) that expressed single recombinant human UGT enzymes. As shown in Table 4, at 100 µM JNJ-10198409, UGT1A9, 1A8, and 1A7 were the most active isozymes catalyzing formation of Glu-A with glucuronidation rates of 1869 nmol min–1 mg–1, 1361 nmol min–1 mg–1, and 1344 nmol min–1 mg–1, respectively, which is followed by UGT1A4 with a glucuronidation rate of 576 nmol min–1 mg–1. UGT1A1, 1A10, and 1A3 demonstrated lower glucuronidation with reaction rates of 146 nmol min–1 mg–1, 98 nmol min–1 mg–1, and 41 nmol min–1 mg–1, respectively. For Glu-B formation, UGT1A9 was the most active enzyme with a glucuronidation rate of 2384 nmol min–1 mg–1, which is followed by UGT1A3 (1003 nmol min–1 mg–1), UGT1A7 (834 nmol min–1 mg–1), and UGT1A4 (816 nmol min–1 mg–1). Although UGT1A1 and 1A8 also catalyzed formation of Glu-B, the rate was very low (UGT1A1, 155 nmol min–1 mg–1; UGT1A8, 144 nmol min–1 mg–1) compared with the other isoforms. In contrast to UGT1A enzymes, UGT2B members including UGT2B4, 2B7, 2B15, and 2B17 were all inactive in the glucuronidation of JNJ-10198409.
|
Kinetics of Glu-A and Glu-B by Individual UGTs. Kinetic studies were then focused on the formation of Glu-A and Glu-B by the most active UGT enzymes including UGT1A3, 1A4, 1A7, 1A8, and 1A9. Incubation conditions were initially optimized using various protein concentrations and for various times. It was found that formation of Glu-A and Glu-B was linear at 0.5 mg/ml protein within 30 min for UGT1A4, 1A7, 1A8, and 1A9. Formation of Glu-A by UGT1A3 was not significant, and kinetic parameters were not determined. For other tested UGT enzymes, incubations of JNJ-10198409 over the concentration range of 1 to 100 µM with individual UGTs resulted in a concentration-dependent formation of Glu-A and Glu-B. Kinetic analyses indicated that formation of Glu-A and Glu-B by all those UGT enzymes followed classic Michaelis-Menten kinetics with a correlation coefficient (r2) in the range of 0.95 to 0.99, suggesting that all enzymes have a single binding site. All kinetic constants are given in Table 5.
|
Propofol Inhibition of JNJ-10198409 Glucuronidation. UGT1A9 was shown to be the most active enzyme catalyzing the glucuronidation of JNJ-10198409. To determine the relative contribution of UGT1A9 to both Glu-A and Glu-B formation in human liver microsomes, the effect of propofol, a selective UGT1A9 competitive inhibitor (Girard et al., 2004), on the effect of JNJ-10198409 glucuronidation was determined (Table 6). At a lower drug concentration (1 µM), propofol effectively inhibited formation of both Glu-A and Glu-B. At higher drug concentrations (5 and 25 µM), propofol showed little inhibition of Glu-A and -B formation in human liver microsomes. Other UGT inhibitors have not been proven to be highly specific; therefore, inhibition activity toward formation of Glu-A and -B was not investigated in this study.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Initial results from our in vitro studies suggested that glucuronidation was the major elimination pathway for JNJ-10198409. In rat liver microsomal incubations, three N-glucuronides, Glu-A, -B, and -C, were detected by LC-MS/MS. Only Glu-A and -B were found in monkey and human liver microsomes. All three glucuronides were detected in rat plasma, whereas Glu-A and Glu-B were predominant glucuronides in plasma samples collected from monkeys (unpublished data). Thus, the in vitro and in vivo results were consistent. None of those oxidative metabolites identified in microsomal incubations were detected in either rat or monkey samples.
JNJ-10198409 has three nitrogen atoms. NMR analysis of isolated glucuronides indicated that all three conjugates are secondary N-glucuronides, which appeared to be consistent with the tautomerization of JNJ-10198409 (Fig. 1). Many aromatic heterocyclic amines have been reported to form N-glucuronides (Chiu and Husky, 1998), which include imidazole (Pahernik et al., 1995), pyridine (Sakamoto and Nakamura, 1993), tetrazole (Stearns et al., 1992; Huskey et al., 1994) and triazoles (Huskey et al., 1994). Glu-B appears to be the first documented example showing that the glucuronic acid moiety is linked to the pyrazole ring. Also, the results indicated that both aromatic nitrogen atoms of the pyrazole ring form N-glucuronides.
Species differences in glucuronidation of JNJ-10198409 were observed in the present study. In rat, three glucuronides were detected, and Glu-A is the most abundant conjugate, followed by Glu-C and Glu-B. In contrast, only two glucuronides, Glu-B and Glu-A, were formed in incubations with monkey and human liver microsomes. Similar species differences in N-glucuronidation were observed in rat and monkey dosed with JNJ-10198409. Previous studies have shown that N-glucuronidation exhibited species differences in vitro and in vivo, but species differences are generally less striking for compounds forming nonquaternary N-conjugates (Chiu and Husky, 1998; Kaji and Kume, 2005).
Stability differences were observed for the three glucuronides. Both Glu-A and Glu-C are hydrolyzed by ß-glucuronidase, and underwent slow degradation at neutral pH. In contrast, Glu-B was resistant to hydrolysis by bacterial ß-glucuronidases and did not decompose under mild conditions. Previous studies have shown that N-glucuronide metabolites of the primary/secondary amines and N-glucuronides of N-hydroxylated amines were hydrolyzed to the parent compounds and glucuronic acid under mild or acidic conditions (Kadlubar et al., 1977; Mohri et al., 2001), whereas the quaternary-ammonium glucuronides were hydrolyzed under basic conditions (Dulik and Fenselau, 1987; Kowalczyk et al., 2000). Among the several quaternary-ammonium glucuronides tested for stability in the presence of ß-glucuronidase from bovine liver and E. coli, only bacterial enzyme hydrolyzed all N-glucuronides (Kowalczyk et al., 2000). It is interesting to note that Glu-B showed distinct stability compared with Glu-C and Glu-A, two isomeric glucuronides.
Kinetic studies revealed that N-glucuronidation of JNJ-10198409 in human liver microsomes did not follow classic Michaelis-Menten kinetics. A likely explanation for the nonclassic Michaelis-Menten kinetics is the involvement of multiple UGT enzymes in glucuronidation of JNJ-10198409 in human liver microsomes. This argument is largely supported by the observation that several recombinant UGT1A enzymes (e.g., UGT1A9, 1A8, and 1A7) catalyzed glucuronidation of JNJ-10198409. In addition, kinetic analyses indicated that all glucuronidation reactions catalyzed by individual recombinant UGT enzymes followed classic Michaelis-Menten kinetics. At present, it is not clear whether incomplete removal of glucuronidation latency in human liver microsomes may also lead to atypical Michaelis-Menten kinetics of glucuronidation of JNJ-10198409. Atypical Michaelis-Menten kinetics has been reported previously in both glucuronidation (Kirkwood et al., 1998; Watanabe et al., 2002; Stone et al., 2003) and biotransformation mediated by P450s (Ekins et al., 1998; Hutzler et al., 2002).
Identification of the specific UGTs for N-glucuronidation of JNJ-10198409 would allow future investigation of potential drug-drug interaction as well as possible polymorphism in N-glucuronidation of JNJ-10198409. Although definitive identification of UGT isozymes is difficult at present, largely due to lack of enzyme-specific substrates and inhibitors for correlation and selective inhibition studies, characterization of UGT isoforms responsible for a glucuronidation reaction has become possible because of the availability of cDNA-expressed UGTs. Several UGT1A enzymes such as UGT1A4, UGT1A7, UGT1A8, and UGT1A9 were all active in catalyzing the formation of both Glu-A and Glu-B, which is probably due to efficient binding of JNJ-10198409 to those UGT1A enzymes. One should note that observations in individual recombinant UGTs may not simply be relevant to the liver tissue since it has been known that UGT activity and substrate specificity can change via heterodimerization with other isozymes.
It is interesting to note that, even though the pyrazole ring could undergo tautomerization (Fig. 1), glucuronidation of the pyrazole moiety by human UGT1A enzymes exclusively occurred at the N-1 position, which could be explained by the difference in the orientation of individual nitrogen atoms within the active sites of UGT1A. An alternative explanation is that the steric effect of the fluoro-phenylamine substitutent on the N-2 position of the pyrazole ring. However, this hypothesis could not explain the glucuronidation of the aliphatic amine that is also sterically hindered by the fluoro-phenyl moiety. Because of its unique difference in N-glucuronidation, it is our speculation that JNJ-10198409 could potentially serve as a model substrate for probing the structure-activity relationship of UGT1A enzymes such as UGT1A9.
Additionally, the present results seem to suggest that UGT1A9 may play a more important role than other isoforms in hepatic glucuronidation of JNJ-10198409 at low drug concentrations (<1 µM). This conclusion is supported by the following two observations. First, UGT1A9 is the most active enzyme catalyzing glucuronidation of JNJ-10198409 in human liver microsomes; second, at low drug concentrations, formation of Glu-A and Glu-B was effectively inhibited by the UGT1A9-selective inhibitor propofol. The inability to inhibit glucuronidation by propofol at high drug concentrations (>1 µM) suggested that multiple enzymes were likely involved in the glucuronidation of JNJ-10198409. Since UGT1A7 and 1A8, two enzymes highly expressed in extrahepatic tissues, were also active in glucuronidation of JNJ-10198409, it is likely that the first-pass phase II metabolism may also play a significant role in the clearance of the drug. Therefore, one should not simply interpret the in vivo relevance of the in vitro results. Because the drug is likely to be metabolized by multiple UGT1A enzymes expressed in different tissues, it is reasonable to speculate that neither UGT-mediated drug-drug interactions nor polymorphism of UGT genes is an obvious concern.
In conclusion, N-glucuronidation of JNJ-10198409, a novel potential anticancer agent, was elucidated using rat, monkey, and human liver microsomes. Species differences were observed; three glucuronides formed in rat, and two conjugates were detected in monkey and human liver microsomes. The Glu-B conjugate is the first documented example showing that the glucuronic acid moiety is linked to an aromatic nitrogen of the pyrazole ring. It was also found that multiple UGT enzymes are responsible for the N-glucuronidation of this compound in human liver microsomes.
|
Footnotes |
|---|
ABBREVIATIONS: UGT, uridine diphosphate glucuronosyltranferase; UDPGA, uridine 5'-diphosphoglucuronic acid; HLM, human liver microsomes; LC-MS/MS, liquid chromatography-tandem mass spectrometry; JNJ-10198409, 6,7-(dimethoxy-2,4-dihydroindeno[1,2-c]pyrazol-3-yl)-(3-fluoro-phenyl)-amine; P450, cytochrome P450; HPLC, high-performance liquid chromatography; DMSO, dimethyl sulfoxide; ESI, electrospray ionization; 1D, one-dimensional; 2D, two-dimensional; g-COSY, correlated spectroscopy gradient; g-NOESY, nuclear Overhauser enhancement spectroscopy gradient; g-HMBC, heteronuclear multiple bond correlated spectroscopy gradient; RT, retention time.
Address correspondence to: Zhengyin Yan, Drug Discovery, R2013, Johnson & Johnson Pharmaceutical Research & Development, LLC, Welsh & McKean Roads, Spring House, PA 19477-0776. E-mail: zyan{at}prdus.jnj.com
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Omura, T. Nakazawa, T. Sato, T. Iwanaga, and O. Nagata Characterization of N-Glucuronidation of 4-(5-Pyridin-4-yl-1H-[1,2,4]triazol-3-yl) pyridine-2-carbonitrile (FYX-051): A New Xanthine Oxidoreductase Inhibitor Drug Metab. Dispos., December 1, 2007; 35(12): 2143 - 2148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nakazawa, K. Miyata, K. Omura, T. Iwanaga, and O. Nagata Metabolic Profile of FYX-051 (4-(5-Pyridin-4-yl-1H-[1,2,4]triazol-3-yl)pyridine-2-carbonitrile) in the Rat, Dog, Monkey, and Human: Identification of N-Glucuronides and N-Glucosides Drug Metab. Dispos., November 1, 2006; 34(11): 1880 - 1886. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
