Bioactivation of the Epidermal Growth Factor Receptor Inhibitor Gefitinib

Gefitinib is an inhibitor of the epidermal growth factor receptor (EGFR) tyrosine kinase and has been approved for the treatment of nonsmall cell lung...
0 downloads 0 Views 381KB Size
1736

Chem. Res. Toxicol. 2009, 22, 1736–1742

Bioactivation of the Epidermal Growth Factor Receptor Inhibitor Gefitinib: Implications for Pulmonary and Hepatic Toxicities Xiaohai Li,† Theodore M. Kamenecka,† and Michael D. Cameron*,†,‡ Translational Research Institute and Department of Molecular Therapeutics, Scripps Florida, The Scripps Research Institute, 130 Scripps Way, Jupiter, Florida 33458 ReceiVed July 27, 2009

Gefitinib is an inhibitor of the epidermal growth factor receptor (EGFR) tyrosine kinase and has been approved for the treatment of nonsmall cell lung cancer refractory to established cancer treatments. Several cases of adverse hepatic and pulmonary events have been reported, which led to discontinuation of therapy. While the mechanism of toxicity remains unknown, we present evidence that gefitinib accumulates in the liver and lung, and it can be bioactivated in hepatic, intestinal, and pulmonary microsomes to form a reactive metabolite. The reactive metabolite was trapped by the peptide mimetic GSH, indicating that the metabolite was sufficiently reactive to bind to the cysteine groups of proteins. Two cytochrome P450dependent gefitinib-GSH adducts were detected, and P450 1A1 and 3A4 were found to be the major enzymes responsible for adduct formation. The mechanism of bioactivation is proposed to involve oxidative defluorination of gefitinib to form a reactive quinone-imine. Clinical reports have noted an increase in adverse pulmonary events with patients who continued smoking. Consistent with the clinical toxicology data, a 12-fold increase in GSH adduct formation was detected in human pulmonary microsomes from smokers over nonsmokers, in agreement with P450 1A1 being induced by cigarette smoke. Introduction Gefitinib (GEF),1 a reversible inhibitor of the epidermal growth factor receptor tyrosine kinase (HER1/EGFR), is implicated in the proliferation and maintenance of certain cancers and has been approved for patients with nonsmall cell lung cancer who are refractory to established cancer treatments (1). Selective and multiple kinase inhibitors are being actively researched as chemotherapeutics because they target processes involved in tumor growth and maintenance and have demonstrated lower degrees of toxicity than traditional cytotoxic therapies (2, 3). While considered “well-tolerated”, GEF has been associated with common adverse effects such as skin rash (4, 5) and liver injury (6, 7). Additionally, GEF has been associated with pulmonary toxicity, such as interstitial lung disease (ILD), which occurs in low incidence but can be serious and sometimes fatal (8-15). The incidence of ILD in patients receiving GEF has been estimated at 2% in Japan and 0.3% in the United States (16). Takano et al. (15) reported an investigation from Japan where ILD developed in 5.4% of 112 patients receiving GEF therapy, leading to four deaths. Observations in patients that were removed from GEF due to ILD and later reintroduced to the drug revealed that ILD recurrence was more rapid and severe than the first episode (14, 17), indicating a possible immune system role rather than direct drug cytotoxicity. All reported deaths due to GEF-induced ILD occurred in current or former smokers (15). Here, we present evidence that P450 1A1, an enzyme that is upregulated in the lungs of smokers, catalyzes * To whom correspondence should be addressed. E-mail: cameron@ scripps.edu. † Translational Research Institute. ‡ Department of Molecular Therapeutics. 1 Abbreviations: ACN, acetonitrile; CE, collision energy; CID, collisioninduced dissociation; DP, declustering potential; EPI, enhanced product ion; GEF, gefitinib; HLM, human liver microsomes; IDA, information dependent acquisition; MRM, multiple reaction monitoring; PI, precursor ion scan.

the metabolism of GEF to a reactive quinone-imine that can bind covalently to protein. In humans, GEF was shown to be metabolized extensively by P450 3A4 and to a lesser degree by P450 3A5, 1A1, and 2D6. The major metabolic pathways included morpholine ring opening, stepwise removal of the morpholine ring, and oxidative defluorination to form a hydroxyaniline metabolite (18-21). The para-hydroxyaniline metabolite may undergo P450-mediated twoelectron oxidation, forming a toxic quinone-imine intermediate. To date, GEF has not been reported to be bioactivated to reactive intermediates. We have identified the human cytochrome P450 enzymes involved, demonstrated the formation of GSH adducts during the oxidative metabolism of GEF, and proposed that the mechanism of adduct formation involves the formation of a quinone-imine.

Experimental Procedures Materials. Reagents and solvents used were of the highest grade commercially available. Analytical grade GEF (>99%) was purchased from LC Laboratories (Woburn, MA). The following chemicals were purchased from Sigma-Aldrich (St. Louis, MO): midazolam, carbamazepine, R-naphthoflavone, phenacetin, diclofenac, (S)-mephenytoin, and ketoconazole. Pooled human liver microsomes (HLMs), human intestinal microsomes, human pulmonary microsomes (NS, nonsmoker; S, smoker), and recombinant P450s (EasyCYP bactosomes, prepared from Escherichia coli coexpressing recombinant human NADPH-P450 reductase) were purchased from XenoTech LLC (Lenexa, KS). Formic acid, dimethyl sulfoxide (DMSO), and acetonitrile (ACN) were purchased from Fisher Scientific (Fair Lawn, NJ). Recombinant human microsomal epoxide hydrolase was purchased from BD Gentest (Woburn, MA). Instrumental Methods. LC-MS/MS analyses of GSH adducts was performed on an API 4000 Q-Trap mass spectrometer equipped with a Turboionspray source (Applied Biosystems, Foster City, CA) as previously described (22). Briefly, a negative precursor ion (PI)

10.1021/tx900256y CCC: $40.75  2009 American Chemical Society Published on Web 10/05/2009

P450-Mediated BioactiVation of Gefitinib

Chem. Res. Toxicol., Vol. 22, No. 10, 2009 1737

Figure 1. Generation of the GEF para-hydroxyaniline metabolite and GSH adducts with recombinant cytochrome P450. In incubations containing recombinant P450 bactosomes, NADPH, and GEF, the relative amount of para-hydroxygefitinib formed was determined (A). When GSH was added to the incubations, the formation of GEF-GSH adducts was measured (B).

scan of m/z 272, corresponding to GSH fragmenting at the thioether bond, was used to detect GSH adducts. This triggered positive ion enhanced resolution and enhanced product ion (EPI) scans. Chromatographic separation was achieved by using an Agilent Eclipse XDB C18 column (3.5 µm, 3.0 mm × 150 mm). The HPLC mobile phase A was water with 0.1% formic acid, and mobile phase B was ACN with 0.1% formic acid. An Agilent 1200 HPLC system was used with the following gradient elution profile: 5% solvent B for 3 min, and then, solvent B was rapidly ramped to 10% in 0.5 min, followed by 10-50% B in 19.5 min and 50-80% B in 5 min; at 28 min, the column was flushed with 80% B for 2 min before equilibrating to initial conditions. The HPLC flow rate was 0.4 mL/min. Metabolites, including GSH adducts, were initially characterized by comparing incubated samples to control samples without NADPH, GSH, or substrate. Structural information was generated from the collision-induced dissociation (CID) spectra of the corresponding protonated molecular ions in EPI scans. For characteristic LC/MS/MS analysis of metabolites, incubation samples were stopped by adding 500 µL of ACN and concentrated by a SpeedVac. The samples were reconstituted in 100 µL of 30% ACN, and 10 µL was used for metabolite profiling experiments by the EMS-information-dependent acquisition (IDA)-EPI method. Briefly, EMS spectra were obtained from m/z 350 to 760, and the most intense ion above 1000 counts per second triggered an EPI scan to acquire MS/MS data. To improve detection sensitivity and specificity, metabolites and GSH adducts were also characterized using the multiple reaction monitoring (MRM) triggered EPI (MRM-IDA-EPI) mode following preset MRM transitions. The MRM methods were set to the most intense ion pairs for each adduct, which were transitions from m/z 750.2f477.1 and m/z 752.2f479.1. For all potential GSH adducts, the same declustering potential (DP) (70 V), collision energy (CE) (40 eV), and CE spread ((20 eV) were applied. For relative comparison of GSH adduct levels and the correlated hydroxylated metabolites formed by individual recombinant P450 enzymes, the internal standard carbamazepine (m/z 237.3f194.2) was used. MRM transitions were simultaneously monitored for GSH adducts and the hydroxyaniline metabolites. The hydroxyaniline metabolite of GEF was followed using m/z 445.2f128.1. Data were analyzed using Analyst 1.4.2 version software (Applied Biosystems, Foster City, CA).

Figure 2. Detection of GEF-GSH adducts in human liver microsomal incubations. NADPH-supplemented human liver microsomal incubations of GEF in the presence of GSH (5 mM) were examined using MRM methods by LC-MS/MS (A). The peaks between 10 and 11 min are isomers with a molecular ion of m/z 750 AMU. The MS/MS spectrum for the larger peak is shown in B. The detected GEF-GSH adduct has the para-position fluorine replaced by a hydroxyl group. The MS/MS spectrum of the larger of the two isomers between 12 and 13 min is shown in C. Both peaks had molecular ions of m/z 752 AMU and are similar to the 10-11 min peaks but with an oxidized morpholine group.

Microsomal Incubations. Stock solutions of GEF were prepared in DMSO. The final concentration of organic solvent in the incubations was 0.2% (v/v). All incubations were performed at 37 °C in a shaking incubator. Pooled microsomes and recombinant P450 (EasyCYP Bactosomes) were thawed on ice prior to the experiment. The total incubation volume was 0.5 mL. GEF (40 µM) was mixed with microsomes (2 mg/mL protein) in 100 mM potassium phosphate buffer (pH 7.4) fortified with 5 mM GSH. After a 4 min preincubation at 37 °C, the reactions were initiated by the addition of 1 mM NADPH. Reactions were stopped by the addition of 500 µL of ACN (with or without IS. added depending on the purpose of the analysis) after 60 min of incubation. Incubations with recombinant P450 were performed similarly except that liver microsomes were replaced by EasyCYP bactosomes (100 pmol/mL). Control samples containing no NADPH or substrate were included. Samples were centrifuged at 10000g for 10 min at 4 °C to pellet proteins, supernatants were dried down using a SpeedVac, and the dried samples were reconstituted in 100 µL of 30% ACN. In some incubations, the selective P450 3A4/5 inhibitor ketoconazole was added at a final concentration of 1 µM, R-naphthoflavone at 20 µM, or microsomal epoxide hydrolase at 1 mg/mL. All other conditions were the same. Purification of para-Hydroxygefitinib. para-Hydroxygefitinib was purified from human liver microsomal incubations. Incubations

1738

Chem. Res. Toxicol., Vol. 22, No. 10, 2009

Li et al.

Scheme 1. Proposed Pathways of GEF-GSH Adduct Formationa

a

The nomenclature for metabolites H and N is from McKillop et al. (19).

contained 40 µM GEF, 1 mg/mL HLMs, and 1 mM NADPH in a final volume of 1 mL and were incubated at 37 °C for 1 h prior to the addition of 150 µL of 10% trichloroacetic acid. Metabolites were isolated by solid-phase extraction (Strata C18-E, 300 mg/3 mL cartridges, Phenomenex). Multiple reactions were pooled and further purified by manual collection of HPLC peaks. The incorporation of a hydroxyl group and loss of GEF fluorine were verified by high-resolution mass spectrometry. Time- and Concentration-Dependent Inactivation of P450s. Time- and concentration-dependent loss of P450 activity was assayed as previously described (22). The P450 3A4 activity was evaluated in both HLMs and recombinantly expressed enzymes. Other cytochrome P450s, 1A1, 1A2, 2D6, 2C9, and 2C19, were evaluated with recombinantly expressed enzyme according to validated assays (23) with slight modification. Plasma and Tissue Concentration after Oral Dosing of GEF. The tissue distribution of GEF was evaluated in C57Bl6 mice (n ) 3). Compounds were dosed at 10 mg/kg orally, and after 2 h, blood, liver, lung, and brain were collected. Tissues were not perfused to reduce the risk that GEF would be eluted from the tissue during the perfusion process. Plasma was generated using standard centrifugation techniques, and the plasma and tissues were frozen at -80 °C. Plasma and tissues were mixed with ACN (1:5 v:v or 1:5 w:v, respectively). Tissues were sonicated with a probe tip sonicator. Samples were analyzed for drug levels by LC-MS/MS. Plasma drug levels were determined against standards made in plasma and tissue levels against standards made in the appropriate blank tissue matrix. All procedures were approved by the Scripps Florida IACUC and Scripps vivarium is fully AAALAC accredited.

Results GEF Tissue Distribution. GEF concentrations used for this study were intended to approximate the maximum GEF concentration likely to be encountered by a subset of patients under clinical conditions. A study by Nakagawa et al. (24) reported that GEF levels increase upon repeat dosing and that the steady state Cmax of GEF after 14 days of dosing with 225 or 400 mg of GEF was 0.76 and 1.75 µM, respectively. This study had only 4-6 patients per dose group, so interindividual variability could not be evaluated. A single dose study by Swaisland et al. (25) using 24 subjects reported similar median Cmax but noted a 15-fold variability between subjects. While plasma levels are routinely monitored, concentrations in the liver and the lung are not; however, liver and lung concentrations are more important to GEF metabolism and reactive intermediate formation evaluated in this manuscript. Published mouse positron emission tomography studies utilizing [18F]GEF showed significant accumulation in the liver and lung, with levels 9- and 3-fold higher than plasma 2 h after intravenous injection of GEF (26). Lung levels were highest immediately after injection and decreased as GEF became concentrated in the liver and then the gallbladder. Because the positron emission tomography study measured the isotopic fluorine and not GEF, it cannot differentiate between GEF and its metabolites. We evaluated tissue levels by LC-MS/MS using oral dosing to more closely simulate clinical conditions. Three mice were

P450-Mediated BioactiVation of Gefitinib

Figure 3. GEF metabolites in GSH-supplemented incubations of HLMs, P450 1A1, and 3A4. LC-UV chromatograms were detected at 254 nm for incubations with (A) HLMs, (B) P450 1A1, and (C) P450 3A4. Peaks labeled with an M are believed to be metabolites of GEF as they had MS/MS fragmentation patterns similar to GEF. Peaks labeled with GEF-G1 are GSH adducts, and the peak labeled GEF is remaining GEF.

orally dosed with 10 mg/kg GEF, and the drug levels were determined in the plasma, liver, lung, and brain at 2 h. Significant tissue accumulation was observed in the lung and in the liver, each having GEF concentrations over 10 times the level in plasma. Brain levels were similar to previous reports and were included to act as a control (26). Relevant GEF concentrations in human liver and lung are inferred based on the assumption of similar tissue distribution and of high interindividual variability as noted by Swaisland et al. (25). Hepatic GEF levels of 10-20 µM would be likely with most patients with some having hepatic Cmax of 40 µM or higher. On the basis of this, subsequent incubations were done using 40 µM GEF. para-Hydroxyaniline Formation. Previously published work using radiolabeled GEF and mass spectrometry identified numerous GEF metabolites including the replacement of fluorine at the para-position of the aniline with a hydroxy group (18-20, 27). We observed similar metabolite profiles. The resulting para-hydroxyaniline was formed primarily by P450 1A1 and 3A4. The relative amount of the para-hydroxyaniline metabolite is shown in panel A of Figure 1. Relative concentrations were determined via comparison to an internal standard because an authentic standard for the para-hydroxyaniline metabolite was not available. When GSH was added to the

Chem. Res. Toxicol., Vol. 22, No. 10, 2009 1739

Figure 4. Comparisons of the relative amounts of GEF-GSH adducts formed in different human microsomal incubations with or without chemical inhibitor added. (A) The formation of the major GSH adduct, GEF-G1, in microsomal incubations was normalized to HLMs without ketoconazole added. (B) The formation of the major adduct in lung microsomal incubations was normalized to that formed in human lung microsomes from smokers without inhibitor added. The values were an average of duplicate incubations.

incubations, P450s that generated the para-hydroxyaniline metabolite also catalyzed the formation of GSH adducts, panel B of Figure 1. In contrast to these experiments, which use equivalent amounts of recombinant P450 enzyme, P450 3A4 is likely to be the major enzyme responsible for the formation of the para-hydroxyaniline metabolite and GSH adduct formation in hepatic and intestinal microsomes, due to its high level of expression. GSH Adducts of GEF. At least four GSH adducts of GEF were observed in incubations with HLMs (Figure 2). All adducts required NADPH for their formation, and GSH was covalently bound to the aniline ring. Two adducts appear to be secondary metabolites where the GEF morpholine ring is oxidized (Scheme 1). Similar oxidation of the morpholine ring had been previously reported (19) in human liver microsomal incubations. The MS/ MS fragmentation pattern of the most abundant GEF-GSH adduct with m/z of 750.3 was consistent with replacement of the GEF fluorine with a hydroxy group at the para-position, as depicted in Figure 2. An additional adduct was detected at m/z of 752.3. This is likely a secondary metabolite formed by oxidation of the morpholine to its ring-opened form (19). Quantification of GEF-GSH Adducts. Nominal quantification of GEF metabolites and GSH adducts was based on LCUV chromatograms. The extinction coefficients at 254 nm were assumed to be the same for all metabolites. The identities of individual peaks were determined based on MS/MS data. The fractions of the total metabolites that were GEF-GSH for HLMs, P450 1A1, and P450 3A4 were estimated to be 11.1,

1740

Chem. Res. Toxicol., Vol. 22, No. 10, 2009

Li et al.

Scheme 2. Mechanism of GEF Bioactivation

44.2, and 4.7%, respectively (Figure 3). Data for other enzymes, such as P450 2D6, which extensively dealkylates GEF (18-20, 27), are not shown since they generate only small amounts of GSH adducts. Identification of GEF-GSH Adducts in Human Hepatic, Intestinal, and Pulmonary Microsomes. On the basis of the data presented in Figure 1, P450 1A1 has the highest specific activity with respect to GSH adduct formation. The specific activity of P450 3A4 was only 10.6% of that seen for P450 1A1; however, there is a significantly higher content of P450 3A4 than 1A1 in intestinal and hepatic microsomes (28). Only low levels of adduct formation were detected in incubations with other enzymes, including P450 1A2, 2C9, 2C19, 2D6, and 3A5 (less than 1.5% of that formed by P450 1A1). Comparisons of bioactivation of GEF in human liver and intestinal microsomes, plus lung microsomes from smokers versus nonsmokers, were performed. The effect of smoking is not commonly evaluated for drug metabolism studies; however, these were included due to the observed smoking effect on GEF pulmonary toxicity and the known increase in P450 1A1 expression in smokers. When normalized for the amount of microsomal protein, hepatic microsomes generated almost twice the amount of GSH adduct as intestinal microsomes and many fold more than pulmonary microsomes (Figure 4). The addition of the selective P450 3A4/5 inhibitor ketoconazole (1 µM) significantly inhibited GSH adduct formation in hepatic and intestinal microsomes by 91 and 97%, respectively. The addition of R-naphthoflavone, which inhibits P450 1A1/2, did not alter the formation rate of GSH adducts in either hepatic or intestinal microsomes. Results

obtained with pulmonary microsomes showed a 12-fold increase in GSH adduct formation in smokers relative to nonsmokers. This increase is consistent with an up regulation of P450 1A1 within pulmonary microsomes from smokers as the addition of R-naphthoflavone caused a large decrease in adduct formation where ketoconazole had minimal effect. Addition of Epoxide Hydroxylase. GSH adduct formation was positively correlated to formation of the para-hydroxyaniline metabolite, and the formation of this metabolite may require the formation of an intermediate epoxide. To test this hypothesis and to evaluate the possibility that GSH directly reacts with the epoxide, incubations with recombinant P450 1A1 and 3A4 were augmented with recombinantly expressed microsomal epoxide hydrolase (1 mg/mL protein). In duplicate reactions, epoxide hydrolase did not alter the rate of formation of GEF-GSH adducts (m/z ) 750) with P450 3A4, but the rate of formation was decreased by 26% with P450 1A1. Incubation of the para-Hydroxyaniline Metabolite with Recombinant P450. The para-hydroxyaniline metabolite of GEF was purified from human liver microsomal incubations. The role of epoxide formation and the potential of P450 to oxidize the hydroxyaniline back to a reactive quinone imine, Scheme 2, were tested using recombinant enzymes. Recombinant P450 1A1 and 3A4 were incubated with the parahydroxyaniline metabolite of GEF in the presence of GSH. Both enzymes catalyzed the formation of monohydroxylated GEFGSH adducts. No dihydroxylated adducts were detected, indicating that both enzymes could catalyze the formation of the quinone-imine from the para-hydroxyaniline.

P450-Mediated BioactiVation of Gefitinib

Evaluation of Time-Dependent Inactivation of CYP3A4 and CYP1A1 by GEF. Numerous compounds that generate reactive intermediates have been demonstrated to inactivate the cytochrome P450s (22, 29-32). To test the potential timedependent inhibitory effect of GEF, recombinantly expressed P450 1A1 and 3A4 and HLMs were evaluated. None of the enzymes were inactivated under the conditions tested (up to 200 µM GEF).

Discussion To our knowledge, the present investigation is the first report on P450-catalyzed bioactivation of GEF. GEF is bioactivated through oxidative metabolism likely through a reactive quinoneimine. Further bioactivation catalyzed by P450 1A1, an inducible cytochrome P450, may occur via quinone-imine or epoxide formation. The presence of fluorine at the position of metabolism in GEF did not prevent the formation of the reactive metabolite. Oxidative dehalogenation of aromatics by P450 is not uncommon and has been demonstrated with 4-halogenated anilines (33), fluoroanilines (34), and dasatinib and its structural analogues (22). Besides the research reports on the involvement of P450 3A4, 3A5, and 2D6 and to a lesser degree of P450 1A1 and 1A2 in the overall metabolism of GEF (19-21, 35), our study demonstrated major roles for these enzymes in the metabolic activation of GEF in incubations with human liver, intestinal, and pulmonary microsomes and with recombinantly expressed P450s. We believe that the major reactive intermediate is a quinone-imine. The quantitative correlation between the formation of the para-hydroxyaniline metabolite and GSH adduct formation supports the idea, as does the observation that all of the adducts were defluorinated. If GSH reacted directly with an epoxide that was generated at C4/C5 of the aniline, only a single defluorinated adduct would be expected. The observed 26% decrease in adduct formation upon the addition of epoxide hydrolase to incubations with recombinant P450 1A1 may be due to a decrease in the formation rate of the para-hydroxyaniline metabolite. However, it is quite possible that a reactive epoxide is the source of some of the GSH adduct with P450 1A1. A proposed mechanism showing both bioactivation routes is depicted in Scheme 2. The quinone-imine is a reactive electrophile and can react with nucleophiles such as GSH and protein. The quinone-imine may also react with NADPH or be reduced by microsomal reductases to form the para-hydroxyaniline (33, 36, 37). Conversely, the two-electron oxidation of the para-hydroxyaniline can be catalyzed by P450 to reform the quinone-imine [as seen with acetaminophen (38-40)] and demonstrated with recombinant P450 1A1 and 3A4 in this investigation. P450 1A1 has been extensively studied for its role in the metabolism of polyaromatic hydrocarbons (PAH). Significant levels of PAHs are released in cigarette smoke, in addition to natural environmental exposure. It is well-established that the levels of P450 1A1 mRNA and protein in human lung are greatly induced by exposure to tobacco smoke. P450 1A1 levels are often near or below the level of detection in noninduced samples (41, 42). In this study, P450 1A1 was demonstrated to bioactivate GEF in lung microsomes of smokers, and it is possible that this may influence risks in developing ILD in patients. If P450 1A1 is a factor in GEF-induced ILD, a P450 1A1 inhibitor may decrease ILD incidence as the rate of GEF bioactivation in pulmonary microsomes was reduced by over 80% when the P450 1A1/2 inhibitor R-naphthoflavone was added.

Chem. Res. Toxicol., Vol. 22, No. 10, 2009 1741 Table 1. GEF Tissue Distribution in Mousea tissue GEF concentration (µM) GEF

plasma

brain

liver

lung

2.2 ( 0.31

0.33 ( 0.03

22.0 ( 4.5

26.0 ( 12.5

a Two hours after a single 10 mg/kg oral dose in C57Bl/6 mice. Tissues were not perfused to assure that GEF was not washed out during perfusion.

The clinical evidence to date indicates that GEF is associated with rare idiosyncratic hepatotoxicity. The mechanism of hepatotoxicity has not been established, but reactive metabolites have been implicated in numerous examples of idiosyncratic hepatotoxicity, and covalent modification of cellular proteins are implicated in a number of drug-related adverse effects (43, 44). In addition, redox cycling of quinone imines has been shown to produce oxidative stress (45). Both of these metabolismbased events may potentially contribute to the clinically observed idiosyncratic hepatotoxicity. The high tissue distribution of GEF to the liver and lung is likely to increase the incidence of toxicity. The current study implicates that P450-mediated bioactivation of GEF may result in covalent binding to proteins, oxidative stress, or eventually lead to hepatotoxicity or drug-induced ILD. We believe that the mechanism of bioactivation is through the formation or a reactive quinone-imine, which is primarily catalyzed by P450 3A4/5 in the intestine and liver. The inducible enzyme, P450 1A1, is the primary enzyme leading to GEF bioactivation in the lung, with 12-fold higher levels in smokers than in nonsmokers. This work opens new avenues for understanding the hepatotoxicity and pulmonary toxicities induced by GEF.

References (1) Siegel-Lakhai, W. S., Beijnen, J. H., and Schellens, J. H. (2005) Current knowledge and future directions of the selective epidermal growth factor receptor inhibitors erlotinib (Tarceva) and gefitinib (Iressa). Oncologist 10, 579–589. (2) Grunwald, V., Soltau, J., Ivanyi, P., Rentschler, J., Reuter, C., and Drevs, J. (2009) Molecular targeted therapies for solid tumors: management of side effects. Onkologie 32, 129–138. (3) Muller, B. A. (2009) Imatinib and its successorssHow modern chemistry has changed drug development. Curr. Pharm. Des. 15, 120– 133. (4) Dudek, A. Z., Kmak, K. L., Koopmeiners, J., and Keshtgarpour, M. (2006) Skin rash and bronchoalveolar histology correlates with clinical benefit in patients treated with gefitinib as a therapy for previously treated advanced or metastatic non-small cell lung cancer. Lung Cancer 51, 89–96. (5) Pascual, J. C., Belinchon, I., Sivera, F., and Yuste, A. (2005) Severe cutaneous toxicity following treatment with gefitinib (ZD1839). Br. J. Dermatol. 153, 1222–1223. (6) Carlini, P., Papaldo, P., Fabi, A., Felici, A., Ruggeri, E. M., Milella, M., Ciccarese, M., Nuzzo, C., Cognetti, F., and Ferretti, G. (2006) Liver toxicity after treatment with gefitinib and anastrozole: Drugdrug interactions through cytochrome p450? J. Clin. Oncol. 24, e60– 61. (7) Yoshimoto, A., Kasahara, K., Kimura, H., Kita, T., Fujimura, M., and Nakao, S. (2004) Transient liver injury caused by gefitinib. Nihon Kokyuki Gakkai Zasshi 42, 56–61. (8) Cohen, M. H., Williams, G. A., Sridhara, R., Chen, G., and Pazdur, R. (2003) FDA drug approval summary: Gefitinib (ZD1839) (Iressa) tablets. Oncologist 8, 303–306. (9) Ieki, R., Saitoh, E., and Shibuya, M. (2003) Acute lung injury as a possible adverse drug reaction related to gefitinib. Eur. Respir. J. 22, 179–181. (10) Inomata, S., Takahashi, H., Nagata, M., Yamada, G., Shiratori, M., Tanaka, H., Satoh, M., Saitoh, T., Sato, T., and Abe, S. (2004) Acute lung injury as an adverse event of gefitinib. Anticancer Drugs 15, 461–467. (11) Kitajima, H., Takahashi, H., Harada, K., Kanai, A., Inomata, S., Taniguchi, H., Saikai, T., and Abe, S. (2006) Gefitinib-induced

1742

(12) (13) (14) (15)

(16) (17) (18)

(19) (20) (21) (22)

(23)

(24)

(25)

(26)

(27)

Chem. Res. Toxicol., Vol. 22, No. 10, 2009

interstitial lung disease showing improvement after cessation: Disassociation of serum markers. Respirology 11, 217–220. Ohyanagi, F., Ando, Y., Nagashima, F., Narabayashi, M., and Sasaki, Y. (2004) Acute gefitinib-induced pneumonitis. Int. J. Clin. Oncol. 9, 406–409. Shih, Y. N., Chiu, C. H., Tsai, C. M., and Perng, R. P. (2005) Interstitial pneumonia during gefitinib treatment of non-small-cell lung cancer. J. Chin. Med. Assoc. 68, 183–186. Suzuki, M., Asahina, H., Konishi, J., Yamazaki, K., and Nishimura, M. (2008) Recurrent gefitinib-induced interstitial lung disease. Intern. Med. 47, 533–536. Takano, T., Ohe, Y., Kusumoto, M., Tateishi, U., Yamamoto, S., Nokihara, H., Yamamoto, N., Sekine, I., Kunitoh, H., Tamura, T., Kodama, T., and Saijo, N. (2004) Risk factors for interstitial lung disease and predictive factors for tumor response in patients with advanced non-small cell lung cancer treated with gefitinib. Lung Cancer 45, 93–104. Liu, V., White, D. A., Zakowski, M. F., Travis, W., Kris, M. G., Ginsberg, M. S., Miller, V. A., and Azzoli, C. G. (2007) Pulmonary toxicity associated with erlotinib. Chest 132, 1042–1044. Inoue, A., Saijo, Y., Maemondo, M., Gomi, K., Tokue, Y., Kimura, Y., Ebina, M., Kikuchi, T., Moriya, T., and Nukiwa, T. (2003) Severe acute interstitial pneumonia and gefitinib. Lancet 361, 137–139. McKillop, D., Hutchison, M., Partridge, E. A., Bushby, N., Cooper, C. M., Clarkson-Jones, J. A., Herron, W., and Swaisland, H. C. (2004) Metabolic disposition of gefitinib, an epidermal growth factor receptor tyrosine kinase inhibitor, in rat, dog and man. Xenobiotica 34, 917– 934. McKillop, D., McCormick, A. D., Miles, G. S., Phillips, P. J., Pickup, K. J., Bushby, N., and Hutchison, M. (2004) In vitro metabolism of gefitinib in human liver microsomes. Xenobiotica 34, 983–1000. McKillop, D., McCormick, A. D., Millar, A., Miles, G. S., Phillips, P. J., and Hutchison, M. (2005) Cytochrome P450-dependent metabolism of gefitinib. Xenobiotica 35, 39–50. Li, J., Zhao, M., He, P., Hidalgo, M., and Baker, S. D. (2007) Differential metabolism of gefitinib and erlotinib by human cytochrome P450 enzymes. Clin. Cancer Res. 13, 3731–3737. Li, X., He, Y., Ruiz, C. H., Koenig, M., and Cameron, M. D. (2009) Characterization of dasatinib and its structural analogs as CYP3A4 mechanism-based inactivators and the proposed bioactivation pathways. Drug Metab. Dispos. 37, 1242–1250. Perloff, E. S., Mason, A. K., Dehal, S. S., Blanchard, A. P., Morgan, L., Ho, T., Dandeneau, A., Crocker, R. M., Chandler, C. M., Boily, N., Crespi, C. L., and Stresser, D. M. (2009) Validation of cytochrome P450 time-dependent inhibition assays: a two-time point IC50 shift approach facilitates kinact assay design. Xenobiotica 39, 99–112. Nakagawa, K., Tamura, T., Negoro, S., Kudoh, S., Yamamoto, N., Takeda, K., Swaisland, H., Nakatani, I., Hirose, M., Dong, R. P., and Fukuoka, M. (2003) Phase I pharmacokinetic trial of the selective oral epidermal growth factor receptor tyrosine kinase inhibitor gefitinib (‘Iressa’, ZD1839) in Japanese patients with solid malignant tumors. Ann. Oncol. 14, 922–930. Swaisland, H. C., Smith, R. P., Laight, A., Kerr, D. J., Ranson, M., Wilder-Smith, C. H., and Duvauchelle, T. (2005) Single-dose clinical pharmacokinetic studies of gefitinib. Clin. Pharmacokinet. 44, 1165– 1177. Su, H., Seimbille, Y., Ferl, G. Z., Bodenstein, C., Fueger, B., Kim, K. J., Hsu, Y. T., Dubinett, S. M., Phelps, M. E., Czernin, J., and Weber, W. A. (2008) Evaluation of [(18)F]gefitinib as a molecular imaging probe for the assessment of the epidermal growth factor receptor status in malignant tumors. Eur. J. Nucl. Med. Mol. Imaging 35, 1089–1099. McKillop, D., Partridge, E. A., Hutchison, M., Rhead, S. A., Parry, A. C., Bardsley, J., Woodman, H. M., and Swaisland, H. C. (2004) Pharmacokinetics of gefitinib, an epidermal growth factor receptor tyrosine kinase inhibitor, in rat and dog. Xenobiotica 34, 901–915.

Li et al. (28) Ding, X., and Kaminsky, L. S. (2003) Human extrahepatic cytochromes P450: function in xenobiotic metabolism and tissue-selective chemical toxicity in the respiratory and gastrointestinal tracts. Annu. ReV. Pharmacol. Toxicol. 43, 149–173. (29) Hargus, S. J., Amouzedeh, H. R., Pumford, N. R., Myers, T. G., McCoy, S. C., and Pohl, L. R. (1994) Metabolic activation and immunochemical localization of liver protein adducts of the nonsteroidal anti-inflammatory drug diclofenac. Chem. Res. Toxicol. 7, 575– 582. (30) Decker, C. J., and Doerge, D. R. (1992) Covalent binding of 14Cand 35S-labeled thiocarbamides in rat hepatic microsomes. Biochem. Pharmacol. 43, 881–888. (31) Jushchyshyn, M. I., Kent, U. M., and Hollenberg, P. F. (2003) The mechanism-based inactivation of human cytochrome P450 2B6 by phencyclidine. Drug. Metab. Dispos. 31, 46–52. (32) Kang, P., Liao, M., Wester, M. R., Leeder, J. S., Pearce, R. E., and Correia, M. A. (2008) CYP3A4-Mediated carbamazepine (CBZ) metabolism: formation of a covalent CBZ-CYP3A4 adduct and alteration of the enzyme kinetic profile. Drug Metab. Dispos. 36, 490– 499. (33) Cnubben, N. H., Vervoort, J., Boersma, M. G., and Rietjens, I. M. (1995) The effect of varying halogen substituent patterns on the cytochrome P450 catalysed dehalogenation of 4-halogenated anilines to 4-aminophenol metabolites. Biochem. Pharmacol. 49, 1235–1248. (34) Rietjens, I. M., and Vervoort, J. (1989) Microsomal metabolism of fluoroanilines. Xenobiotica 19, 1297–1305. (35) Ling, J., Johnson, K. A., Miao, Z., Rakhit, A., Pantze, M. P., Hamilton, M., Lum, B. L., and Prakash, C. (2006) Metabolism and excretion of erlotinib, a small molecule inhibitor of epidermal growth factor receptor tyrosine kinase, in healthy male volunteers. Drug Metab. Dispos. 34, 420–426. (36) Rietjens, I. M., Tyrakowska, B., Veeger, C., and Vervoort, J. (1990) Reaction pathways for biodehalogenation of fluorinated anilines. Eur. J. Biochem. 194, 945–954. (37) Rietjens, I. M., and Vervoort, J. (1991) Bioactivation of 4-fluorinated anilines to benzoquinoneimines as primary reaction products. Chem.Biol. Interact. 77, 263–281. (38) Dahlin, D. C., Miwa, G. T., Lu, A. Y., and Nelson, S. D. (1984) N-acetyl-p-benzoquinone imine: A cytochrome P-450-mediated oxidation product of acetaminophen. Proc. Natl. Acad. Sci. U.S.A. 81, 1327– 1331. (39) Harvison, P. J., Guengerich, F. P., Rashed, M. S., and Nelson, S. D. (1988) Cytochrome P-450 isozyme selectivity in the oxidation of acetaminophen. Chem. Res. Toxicol. 1, 47–52. (40) Thummel, K. E., Lee, C. A., Kunze, K. L., Nelson, S. D., and Slattery, J. T. (1993) Oxidation of acetaminophen to N-acetyl-p-aminobenzoquinone imine by human CYP3A4. Biochem. Pharmacol. 45, 1563– 1569. (41) Hukkanen, J., Pelkonen, O., Hakkola, J., and Raunio, H. (2002) Expression and regulation of xenobiotic-metabolizing cytochrome P450 (CYP) enzymes in human lung. Crit. ReV. Toxicol. 32, 391–411. (42) Thum, T., Erpenbeck, V. J., Moeller, J., Hohlfeld, J. M., Krug, N., and Borlak, J. (2006) Expression of xenobiotic metabolizing enzymes in different lung compartments of smokers and nonsmokers. EnViron. Health Perspect. 114, 1655–1661. (43) Tang, W. (2007) Drug metabolite profiling and elucidation of druginduced hepatotoxicity. Expert Opin. Drug Metab. Toxicol. 3, 407– 420. (44) Guengerich, F. P., and MacDonald, J. S. (2007) Applying mechanisms of chemical toxicity to predict drug safety. Chem. Res. Toxicol. 20, 344–369. (45) Hinson, J. A., Reid, A. B., McCullough, S. S., and James, L. P. (2004) Acetaminophen-induced hepatotoxicity: Role of metabolic activation, reactive oxygen/nitrogen species, and mitochondrial permeability transition. Drug. Metab. ReV. 36, 805–822.

TX900256Y