Abstract
The Polycyclic Aromatic Hydrocarbon Naphthalene Is An Environmental Pollutant, A Component Of Jet Fuel, And, Since 2000, Has Been Reclassified As A Potential Human Carcinogen. Few Studies Of The In Vitro Human Metabolism Of Naphthalene Are Available, And These Focus Primarily On Lung Metabolism. The Current Studies Were Performed To Characterize Naphthalene Metabolism By Human Cytochromes P450. Naphthalene Metabolites From Pooled Human Liver Microsomes (Phlms) Were Trans-1,2-Dihydro-1,2-Naphthalenediol (Dihydrodiol), 1-Naphthol, And 2-Naphthol. Metabolite Production Generated KM Values Of 23, 40, And 116 μM And VMax Values Of 2860, 268, And 22 Pmol/Mg Protein/Min, Respectively. P450 Isoform Screening Of Naphthalene Metabolism Identified Cyp1A2 As The Most Efficient Isoform For Producing Dihydrodiol And 1-Naphthol, And Cyp3A4 As The Most Effective For 2-Naphthol Production. Metabolism Of The Primary Metabolites Of Naphthalene Was Also Studied To Identify Secondary Metabolites. Whereas 2-Naphthol Was Readily Metabolized By Phlms To Produce 2,6- And 1,7-Dihydroxynaphthalene, Dihydrodiol And 1-Naphthol Were Inefficient Substrates For Phlms. A Series Of Human P450 Isoforms Was Used To Further Explore The Metabolism Of Dihydrodiol And 1-Naphthol. 1,4-Naphthoquinone And Four Minor Unknown Metabolites From 1-Naphthol Were Observed, And Cyp1A2 And 2D6*1 Were Identified As The Most Active Isoforms For The Production Of 1,4-Naphthoquinone. Dihydrodiol Was Metabolized By P450 Isoforms To Three Minor Unidentified Metabolites With Cyp3A4 And Cyp2A6 Having The Greatest Activity Toward This Substrate. The Metabolism Of Dihydrodiol By P450 Isoforms Was Lower Than That Of 1-Naphthol. These Studies Identify Primary And Secondary Metabolites Of Naphthalene Produced By Phlms And P450 Isoforms.
The polycyclic aromatic hydrocarbon naphthalene is an environmental pollutant, a component of jet fuel, and, since 2000, has been reclassified as a potential human carcinogen (Riviere et al., 1999; White, 1999; McDougal et al., 2000; Preuss et al., 2003). Naphthalene also has been used in the production of phthalate plasticizers and resins, azo dyes, dispersants, and tanning agents in the rubber and leather industries (Preuss et al., 2003). Naphthalene is volatile and is discharged into the environment through incomplete burning of fossil fuels as well as domestic and industrial uses of products containing this chemical.
The toxicity of naphthalene has been studied in vitro and in vivo. In the presence of NADPH and human liver microsomes, 100 μM naphthalene produced significant cytotoxicity in human blood mononuclear leukocytes, but not genotoxicity (Tingle et al., 1993). However, naphthalene has been reclassified as a potential human carcinogen due to evidence of its carcinogenic activity in rats (Preuss et al., 2003). Naphthalene has also been reported to induce oxidative stress, resulting in lipid peroxidation and DNA damage in a cultured macrophage cell line, J774A.1 (Bagchi et al., 1998). Lipid peroxidation in mitochondria and glutathione decreases in hepatic and brain tissues are observed in naphthalene-dosed rats (Vuchetich et al., 1996). DNA single-strand breaks are caused by naphthalene in hepatic tissues in the same studies (Vuchetich et al., 1996). In addition, the p53 tumor suppressor gene may be related to the toxicity of naphthalene, including enhanced production of superoxide anion and DNA fragmentation (Bagchi et al., 2000). Early stage toxicological indicators of naphthalene exposure in the mouse include perturbation of nonciliated bronchiolar (Clara) epithelial cell membranes, changes of cell ultrastructure, including swollen smooth endoplasmic reticulum and cytoplasmic blebbing, and intracellular glutathione depletion (Van Winkle et al., 1999; Plopper et al., 2001).
Because the toxicity of naphthalene in cell culture and animal models is closely related to the metabolism of the compound, cytochrome P450 (P450) monooxygenases may play an important role in its toxicological effects. High concentrations of naphthalene (>500 μM) cause decreased viability in isolated murine Clara cells, but a P450 inhibitor, piperonyl butoxide, blocks the loss of cell viability on preincubation with Clara cells (Chichester et al., 1994). Naphthalene metabolism to naphthalene 1R,2S-oxide stereoselectively mediated by CYP2F2 is suggested to be closely related to species-specific and tissue-selective cytotoxicity of this chemical (Buckpitt et al., 1995). Metabolic formation of 1,2- and/or 1,4-naphthoquinone from 1-naphthol may be a direct cause (Stohs et al., 2002) or an intermediate step in the production of naphthosemiquinone radicals for the toxicity of 1-naphthol (Doherty et al., 1984). Susceptibility to naphthalene-induced injury is gender-dependent in the mouse, with female mice producing more dihydrodiol in primary injury sites than male mice (Van Winkle et al., 2002).
The metabolism of naphthalene has been studied primarily in experimental animals (Buckpitt et al., 1984, 1987, 1995, 2002; Buckpitt and Bahnson, 1986; Chichester et al., 1994). Metabolic characterization in humans has been investigated in only a few studies (Buckpitt and Bahnson, 1986; Tingle et al., 1993). Dihydrodiol and three glutathione conjugates are generated by human lung microsomes in the presence of glutathione and glutathione transferases (Buckpitt and Bahnson, 1986). In naphthalene metabolism using human liver microsomes, trans-1,2-dihydrodiol and 1-naphthol are generated (Tingle et al., 1993). However, detailed biochemical characterization and identification of the P450 isoforms most responsible for human naphthalene metabolism have not been reported.
In the present studies, we provide the kinetics of naphthalene metabolism by both human liver microsomes and a wide spectrum of human P450 isoforms. The secondary metabolism of naphthalene primary metabolites was also studied to identify metabolic pathways of naphthalene. A new potential biomarker for exposure to naphthalene was suggested by these studies.
Materials and Methods
Chemicals. Naphthalene, 1-naphthol, 2-naphthol, 1,4-naphthoquinone, 2,6-dihydroxynaphthalene, and 1,7-dihydroxynaphthalene were purchased from Sigma-Aldrich (St. Louis, MO). trans-1,2-Dihydro-1,2-naphthalenediol was a generous gift from Dr. Alan R. Buckpitt (University of California, Davis, CA). Acetonitrile, tetrahydrofuran, and phosphoric acid were purchased from Fisher Scientific Co. (Pittsburgh, PA).
Human Liver Microsomes and Human Cytochrome P450 Isoforms. Pooled human liver microsomes (pHLMs) and human P450 isoforms expressed in baculovirus-infected insect (Autographa californica) cells (BTI-TN-5B1-4) [CYP1A1, 1A2, 1B1, 2A6, 2B6, 2C8, 2C9*1(Arg114), 2C18, 2C19, 2D6*1(Val374), 2E1, 3A4, 3A5, 3A7, 4A11], and microsomal epoxide hydrolase (mEH) were purchased from BD Gentest (Woburn, MA).
In Vitro Naphthalene Metabolism by Pooled Human Liver Microsomes. Naphthalene metabolism mediated by pHLMs was tested in vitro. These assays were performed with an NADPH-generating system (0.25 mM NADP, 2.5 mM glucose 6-phosphate, and 2 U/ml glucose-6-phosphate dehydrogenase) in 100 mM potassium phosphate buffer containing 3.3 mM MgCl2 (pH 7.4). After the substrate was preincubated at 37°C for 5 min, the enzymatic reactions were initiated by the addition of ice-cold pHLMs (0.48 mg/ml) and incubated at 37°C for 10 min. Incubated vials were capped to prevent loss of substrate due to volatility. Metabolism of metabolites of naphthalene, 1-naphthol, 2-naphthol, and trans-1,2-dihydro-1,2-naphthalenediol, in pHLMs was also studied in the same conditions stated above for investigating the metabolic pathways of naphthalene with a series of substrate concentrations. In addition, 1-naphthol (200 μM) was incubated with a 2-fold higher amount of pHLMs (0.96 mg/ml) and the NADPH-generating system at 37°C for a longer time (30 min) to further examine its metabolism in pHLMs (duplicate). For controls, each substrate was incubated in the same buffer system containing pHLMs without the NADPH-generating system.
In Vitro Screening and Enzyme Kinetics for the Metabolic Activity of Human Cytochrome P450 Isoforms. The metabolic activity of the human P450 isoforms (50 pmol/ml) listed above was determined at a substrate concentration of 300 μM naphthalene. The enzymatic assays were performed in the same manner as above, with a modified incubation time of 15 min. Generation of each metabolite mediated by individual P450 isoforms was compared. Sf9 insect cell microsomes from wild-type baculovirus-infected cells (BD Gentest) were used as a control for these assays.
Based on the screening for metabolic activity of human P450 isoforms, enzyme kinetics of the most efficient isoforms (CYP1A1, 1A2, 2B6, 2E1, and 3A4; 40 pmol/ml) for naphthalene metabolism were also studied using a series of substrate concentrations and an incubation time of 10 min.
The metabolic activity of human P450 isoforms (40 pmol/ml) for 80 μM 1-naphthol or trans-1,2-dihydro-1,2-naphthalenediol (dihydrodiol) was also screened. To identify metabolites, the retention time and the spectra of each metabolite were closely compared. A small number of minor metabolites could not be identified in these studies.
Because pooled human liver microsomes generated a higher ratio of trans-1,2-dihydro-1,2-naphthalenediol to 1-naphthol than was observed using P450 isoforms, the potential contribution of naturally occurring epoxide hydrolase to the generation of this product was explored using purified CYP1A2. The kinetics of the formation of 1-naphthol, 2-naphthol, and trans-1,2-dihydro-1,2-naphthalenediol were examined by incubating various concentrations of naphthalene with CYP1A2 (40 pmol/ml) in the absence or presence of human microsomal epoxide hydrolase (0.2 mg/ml).
To compare the metabolic efficiency of human P450 isoforms for 1-naphthol with that for trans-1,2-dihydro-1,2-naphthalenediol, the metabolic activities for two of the most efficient isoforms, based on the isoform screening (CYP1A2 and 2D6*1 for 1-naphthol, and CYP2A6 and 3A4 for trans-1,2-dihydro-1,2-naphthalenediol), were compared by measuring residual parent chemical after incubation at 37°C for 15 min. The controls did not include the NADPH-generating system.
All assay reactions were terminated by addition of an equal volume (250 μl) of acetonitrile and vortexing. After a 5-min centrifugation at 15,000 rpm (21,000g), the supernatant was collected for metabolite characterization using an HPLC system. No metabolites were detected in controls from which the NADPH-generating system was absent.
Analysis of Metabolites by HPLC. The generation of metabolites was analyzed using a Waters 2695 HPLC system equipped with a 2996 photodiode array (PDA) detector (Waters, Milford, MA). This HPLC system was equipped with a degasser and an autoinjector, and data were collected and analyzed using Waters Empower software version 5.00. The solution for pump A was composed of 3% tetrahydrofuran, 0.2% O-phosphorus acid (85%), and 96.8% water, and for pump B, 100% acetonitrile. The gradient in the mobile phase was designed as follows: 0 to 2 min, 20% B; 2 to 22 min, gradient to 80% B; 22 to 25 min, 80% B; and 25 to 30 min, gradient to 20% B. The flow rate was 1.0 ml/min. Metabolites were separated by a reversed phase C12 column (Synergi 4μ Max-RP, 250 × 4.6 mm; Phenomenex, Torrance, CA) and detected using a PDA detector operated from 190 to 350 nm. Optimal wavelengths for 1-naphthol, 2-naphthol, trans-1,2-dihydro-1,2-naphthalenediol, 1,4-naphthoquinone, 1,7-dihydroxynaphthalene, and 2,6-dihydroxynaphthalene were selected as 232.7, 225.6, 262.2, 251.6, 239.8, and 228 nm, respectively. Standards of metabolites were prepared in acetonitrile, and 50 μl of standard or sample was injected into the HPLC system.
Sample Preparation for GC/MS Analysis. Naphthalene (300 μM) was incubated in a total volume of 500 μl of 100 mM potassium phosphate buffer containing 3.3 mM MgCl2 (pH 7.4) with pHLMs (0.96 mg/ml) and the NADPH-generating system mentioned above for 10 min at 37°C after 5-min preincubation. Immediately after incubation, sample tubes were centrifuged at 15,000 rpm (21,000g) for 5 min, and 470 μl of supernatant from each tube was transferred into a fresh tube. Dichloromethane (DCM) (100 μl) was added to the fresh tube containing supernatant, and each tube was vigorously shaken for 1 min. The lower (DCM) layer was then collected for analysis after the tubes were centrifuged at 5000 rpm for 3.5 min. This extraction process with DCM was performed three more times, and the supernatants were combined for the GC/MS analysis.
Analysis of Metabolites by GC/MS. The generation of naphthalene metabolites by pHLMs was confirmed by analysis with an Agilent GC/MS system consisting of a 6890 GC and a 5973 Mass Selective Detector (Agilent Technologies, Palo Alto, CA). A 30-m capillary column with a 0.25-mm nominal diameter (Restek Rtx-5MS; Restek, Bellefonte, PA) was used for the analyses with an injection volume of 2 μl and a constant flow of helium gas (1 ml/min carrier gas). The oven temperature was programmed as follows: initially 40°C with a 1-min hold, increased to 100°C at a rate of 25°C/min, followed by an increase to 300°C at a rate of 10°C/min, followed by a 10-min hold. The total running time was 33.4 min and electron impact was used for the ionization of metabolites.
These analyses were performed as a confirmatory process for the HPLC analysis for the production of primary metabolites of naphthalene metabolism by pHLMs. Throughout the GC/MS analyses, naphthalene, 1-naphthol, and trans-1,2-dihydro-1,2-naphthalenediol were detected at retention times of 7.7, 11.6, and 12.0, respectively, and their fragmentation patterns were compared with those of standards. Detection of 2-naphthol was not successful in these analyses, probably due to the combination of its low level of production and potential loss during the extraction process.
Data Analysis and Statistics. The apparent Vmax and Km parameters were calculated using a nonlinear regression curve fitted to the Michaelis-Menten equation. The coefficient of determination (R2), a measure of how well a regression model describes the data, is shown in the tables. Data means were obtained by at least three determinations. The percentages of total normalized rate (%TNR) were determined as described previously (Rodrigues, 1999). The nominal specific contents of individual P450 proteins in native human livers (10 donors) for calculating the %TNR were obtained from BD Gentest (2003 product catalog) except for the contents of CYP2C8 and CYP2C18, which were from Rodrigues (1999). Statistical significance of the data was determined with one-way analysis of variance followed by Tukey's multiple comparisons.
Results
Three metabolites were detected in metabolism studies of naphthalene by pooled human liver microsomes, 1-naphthol, 2-naphthol, and trans-1,2-dihydro-1,2-naphthalenediol (dihydrodiol). As presented in Table 1, dihydrodiol was the most abundant metabolite, followed in order by 1-naphthol and 2-naphthol. The Km value for dihydrodiol was in the same range as that for 1-naphthol, but the Km for 2-naphthol was significantly higher than those for the other two metabolites. The intrinsic clearance (CLint) of dihydrodiol was significantly higher than those for 1-naphthol and 2-naphthol (Table 1). The Michaelis-Menten curves and metabolic rates for production of the three naphthalene metabolites are shown in Fig. 1.
Metabolic activities of 15 human P450 isoforms for naphthalene were evaluated (Fig. 2). Among those tested, CYP1A2 was found to be the most efficient for the production of 1-naphthol and dihydrodiol, whereas CYP3A4 was the most efficient for the production of 2-naphthol. The individual isoforms showed varying degrees of efficiency for the production of each metabolite. P450 isoforms such as 2C8, 2C9, 2C18, 3A5, 3A7, and 4A11 showed minimal or no activity for naphthalene metabolism (Fig. 2). CYP1A2 was the only isoform to generate 1,4-naphthoquinone from naphthalene in a detectable amount (data not shown). CYP1A2, 3A4, and 2E1 showed the highest total normalized rates (%TNR) for 1-naphthol and 2-naphthol generated in naphthalene metabolism, and CYP1A2, 2A6, and 3A4 showed the highest %TNR for dihydrodiol (Table 2).
The five most efficient human P450 isoforms for naphthalene metabolism as shown in Fig. 2 were selected to further characterize their metabolic activity for naphthalene. As expected, CYP1A2 was identified as the most efficient isoform for generating 1-naphthol and dihydrodiol, showing the highest Vmax values for these metabolites (Table 3). The Vmax and Km for the production of 1,4-naphthoquinone from naphthalene by this CYP1A2 isoform were 2.3 pmol/pmol/min and 29 μM, respectively. In general, more 1-naphthol than dihydrodiol was produced from naphthalene by these isoforms, which is in contrast to naphthalene metabolism by pHLMs. CYP2E1 has higher affinity (i.e., lower Km values) for naphthalene in the production of 1- or 2-naphthol compared with other isoforms. For CLint of 1-naphthol, CYP2E1, 1A2, and 2B6 in that order showed higher values than 3A4 and 1A1. CYP2E1 and 3A4 were higher for the intrinsic clearance of 2-naphthol than other isoforms. CYP3A4 had the highest Vmax for the production of 2-naphthol, and CYP2E1 had the lowest Km, accounting for their greater CLint values observed relative to the other isoforms. The Vmax and CLint values of 1A2 and 1A1 were higher for dihydrodiol production than those of the other isoforms (Table 3). Naphthalene metabolism by CYP1A2 produced one unknown minor metabolite (retention time = 14.3 min), for which the area under the curve (AUC) was less than 1% of the total metabolite AUC.
To investigate apparent discrepancies between amounts of dihydrodiol and 1-naphthol as observed in pHLM-compared with P450 isoform-mediated naphthalene metabolism, naphthalene metabolism by CYP1A2 in the presence of human mEH was studied. These results were compared with the naphthalene metabolism mediated only by CYP1A2 (Table 4; Fig. 3). The production of 1-naphthol and 2-naphthol was significantly reduced in the presence of mEH, whereas the production of dihydrodiol was increased based on Vmax and CLint values. Km values for 1- and 2-naphthol production were significantly increased in the presence of mEH, although the Km value for dihydrodiol production did not change (Table 4). The significant changes in the catalytic velocities by the addition of mEH are also shown in the fitted curves in Fig. 3.
The secondary metabolism of naphthalene was tested by incubating 1-naphthol, 2-naphthol, or dihydrodiol with either pHLMs or CYP1A2. 1-Naphthol was poorly metabolized by pHLMs with about 11% reduction of parent chemical after metabolism, and four unknown minor metabolites were produced. 2-Naphthol was metabolized to produce 2,6- and 1,7-dihydroxynaphthalene, and two unknown minor metabolites (about 3% based on the AUC). In contrast with pHLMs, however, 1-naphthol was metabolized by CYP1A2 to generate 1,4-naphthoquinone and four unknown metabolites (about 56% based on the AUC). 2-Naphthol metabolism by CYP1A2 also produced the same metabolites as those by pHLMs and three additional unknown metabolites (about 6% based on AUC). More 2,6-dihydroxynaphthalene than 1,7-dihydroxynaphthalene was generated by both pHLMs and CYP1A2. Neither pHLMs nor CYP1A2 metabolized dihydrodiol. Metabolism of 1,4-naphthoquinone by pHLMs resulted in one unknown metabolite without the NADPH-generating system and two unknown metabolites additionally with the system. The incubation of 1,4-naphthoquinone with pHLMs (0.48 mg/ml) for 10 min caused the disappearance of this substrate up to about 38% and 51% in the absence and in the presence of the NADPH-generating system, respectively. The kinetic parameters for this secondary metabolism are shown in Table 5.
To further investigate the unknown metabolites from 1-naphthol or dihydrodiol, and to determine which human P450 isoforms are efficient in secondary metabolism, a series of human P450 isoforms was used for 1-naphthol or dihydrodiol metabolism. 1-Naphthol was metabolized to 1,4-naphthoquinone and four unknown metabolites by most P450 isoforms (Fig. 4). Dihydrodiol metabolism generated three unknown metabolites primarily due to activity of CYP2A6 and 3A4 (data not shown). Based on the total AUC of metabolites for 1-naphthol or dihydrodiol, the percentage of total normalized rate (%TNR) of their metabolites for each P450 isoform was calculated (Table 6). CYP3A4, 1A2, and 2C19 showed the highest %TNR for 1-naphthol metabolite, and CYP3A4, 2A6, and 2C8 had the highest %TNR for dihydrodiol metabolite (Table 6). Identification of unknown metabolites was not successful because of the lack of potential standards. To investigate which substrate, 1-naphthol or dihydrodiol, is more effectively metabolized by individual P450 isoforms, two isoforms among the most efficient for each substrate, CYP1A2 and 2D6 for 1-naphthol, and 2A6 and 3A4 for dihydrodiol, were selected. Because there were unknown metabolites from this metabolism, substrate disappearance after metabolism by each P450 enzyme was compared for evaluating the metabolic efficiency of each isoform. For 1-naphthol, 16.3 ± 0.5 and 19.6 ± 1.1% (mean ± S.E.M.) of the parent chemical were metabolized by 1A2 and 2D6, whereas for dihydrodiol, 0.4 ± 0.1 and 2.7 ± 0.2% were metabolized by 2A6 and 3A4, respectively.
Discussion
In naphthalene metabolism by pooled human liver microsomes, about 10 times more trans-1,2-dihydro-1,2-naphthalenediol (dihydrodiol) was generated than 1-naphthol, and generation of the latter was about 10 times higher than that of 2-naphthol. The observation of the predominant production of the dihydrodiol metabolite in these studies agrees with a previous report, in which about 8.6 times more dihydrodiol was generated than 1-naphthol in human liver microsomes (Tingle et al., 1993). Whereas the previous studies of human naphthalene metabolism were performed with microsomes obtained from a limited number of organ donors [1–6; individual human lung microsomes, or pHLMs or pooled human lung microsomes)] (Buckpitt and Bahnson, 1986; Tingle et al., 1993; Wilson et al., 1996), the human liver microsomes used in these studies were commercially prepared from organs donated from as many as 46 people. Therefore, the potential bias caused by individual variation was significantly reduced in the current studies. Although the generation of 2-naphthol in naphthalene metabolism has been known, the observation of 2-naphthol generation by human liver microsomes has not been previously reported. The predominant generation of the trans-form of dihydrodiol is probably related to the catalytic mechanism of human epoxide hydrolase (Morisseau and Hammock, 2005).
The P450 isoform screen in the current studies revealed the most efficient isoforms for producing naphthalene metabolites. Although there have been a few metabolic studies of naphthalene using human microsomes (Buckpitt and Bahnson, 1986; Tingle et al., 1993; Wilson et al., 1996), naphthalene metabolism using a series of individual human P450 isoforms has not been previously studied. CPY1A2 was identified as the most effective isoform for naphthalene metabolism. Total P450 protein content of the 1A2 isoform in human liver ranges from approximately 8 to 13% (Shimada et al., 1994; Rodrigues, 1999). Using the mean specific protein contents of P450 isoforms obtained from BD Gentest (2003 product catalog) and Rodrigues (1999), the calculated %TNR of CYP1A2 demonstrates its important role in naphthalene metabolism in human liver along with CYP3A4, 2E1, and 2A6. Although CYP3A4 showed generally lower metabolic activity toward naphthalene than CYP1A2 (Table 3; Fig. 2), the %TNR of CYP3A4 was approximately 50 and 25% of those for CYP1A2 for 1-naphthol and dihydrodiol generation, respectively, because it had the highest abundance in the human liver microsomes (Table 2). Furthermore, CYP3A4 was the dominant isoform for 2-naphthol formation, not only in the absolute generation of this metabolite but also in the %TNR value, showing that about three fourths of 2-naphthol formed from naphthalene in human liver is associated with CYP3A4. It is known that formation of 1-naphthol and 2-naphthol can be achieved by spontaneous, nonenzymatic rearrangement from the chemically unstable intermediate, naphthalene-1,2-epoxide (Van Bladeren et al., 1984; Buckpitt et al., 2002; Preuss et al., 2003). In the current studies, however, the P450 isoforms tested showed various metabolite ratios produced from naphthalene. These results lead to the conclusion that the production of 1-naphthol and 2-naphthol may be, at least in part, either enzymatic or influenced by the enzyme environment.
Kinetic parameters were obtained for the five P450 isoforms showing the most efficient metabolism of naphthalene. In contrast to naphthalene metabolism in pHLMs, more 1-naphthol was produced by several P450 isoforms than was 2-naphthol or dihydrodiol. In pHLMs, dihydrodiol formation was higher than that of either 1- or 2-naphthol. Epoxide hydrolase is generally known to be involved in the production of dihydrodiol from naphthalene epoxide. Naphthalene assays with a mixture of CYP1A2 and human microsomal epoxide hydrolase showed that microsomal epoxide hydrolase in pHLMs contributes to the higher production of dihydrodiol. Epoxide hydrolase may not be the sole contributor to conversion of naphthalene-1,2-epoxide into the dihydrodiol, because individually expressed P450 isoforms also produced the dihydrodiol metabolite from naphthalene, possibly by nonenzymatic hydrolysis. However, because individual isoforms vary in the production of the dihydrodiol, the possibility exists for P450 involvement directly or indirectly in this transformation into dihydrodiol.
In the secondary metabolism of naphthalene, primary metabolites from naphthalene were used as substrates, and their metabolic reactions in pHLMs and P450 isoforms were investigated. Although 1-naphthol is readily metabolized by most P450 isoforms, this substance is a less favorable substrate for pHLMs than its parent chemical, naphthalene; 1-naphthol metabolism in pHLMs did not produce 1,4-naphthoquinone. In contrast to our results with pooled HLMs, individual P450 isoforms, including 1A2 and 2D6, were effective in metabolizing 1-naphthol to produce 1,4-naphthoquinone and unknown metabolites. 1,4-Naphthoquinone appears more effective as a substrate for pHLMs than its parent chemical, 1-naphthol. The significant reduction of 1,4-naphthoquinone in pHLMs may also be associated with either its metabolism by enzymes other than P450 isoforms, or covalent binding of metabolites derived from metabolism of naphthols to microsomal protein fraction (Hesse and Mezger, 1979), or both.
In a different way from 1-naphthol, 2-naphthol was readily metabolized by pHLMs and CYP1A2. More abundant production of 2,6-rather than 1,7-dihydroxynaphthalene from 2-naphthol indicates that hydroxylation at the carbon 6 position is kinetically more favorable than at the carbon 8 position after carbon 2 is hydroxylated. The product hydroxylated at carbons 2 and 8 is also named 1,7-dihydroxynaphthalene. Metabolism of 2-naphthol by pHLMs, which was more active than that of 1-naphthol and dihydrodiol, may be a factor influencing the higher apparent Km value for 2-naphthol production from naphthalene than those for 1-naphthol or dihydrodiol production in the metabolic system mediated by pHLMs.
trans-1,2-Dihydro-1,2-naphthalenediol (dihydrodiol) was not readily metabolized by either pHLMs or CYP1A2. Dihydrodiol is known to be converted into 1,2-dihydroxynaphthalene by dihydrodiol dehydrogenase and 1,2-naphthoquinone by further oxidation or 1,2-dihydroxy-3,4-epoxy-1,2,3,4-tetrahydronaphthalene by P450 (Penning et al., 1999; Buckpitt et al., 2002). However, neither 1,2-naphthoquinone nor 1,2-dihydroxy-3,4-epoxy-1,2,3,4-tetrahydronaphthalene was identified in this study. Because dihydrodiol dehydrogenase is a cytosolic enzyme, the absence of this enzyme in liver microsomes may explain why 1,2-naphthoquinone was not detected in the current study. Naphthalene metabolites, such as naphthols and dihydrodiol, can be further transformed into conjugation products with glucuronide/sulfate (Preuss et al., 2003), and epoxide can conjugate with glutathione (Smart and Buckpitt, 1983; Buckpitt et al.,1987; Preuss et al., 2003), and this is further transformed into mercapturic acid conjugate (Pakenham et al., 2002).
As the major metabolites, naphthols can be used as biomarkers for exposure to naphthalene. 1-Naphthol and 2-naphthol are detected in urine of Wistar rats administered naphthalene intraperitoneally (Elovaara et al., 2003). However, these naphthols have also been detected in cases of exposure to environmental polycyclic aromatic hydrocarbons in humans and animals. A study of the urinary naphthol content in Japanese male workers, for instance, suggests that 1- and 2-naphthol can be used as biomarkers for exposure to airborne polycyclic aromatic hydrocarbons (Yang et al., 1999). In addition, personal preferences in lifestyle, including smoking, can provide significant variation in urinary naphthol content (Lee et al., 2001), and 1-naphthol is also generated as a metabolite when humans are exposed to the insecticide, carbaryl (Shealy et al., 1997). Furthermore, in the present studies, these naphthols were more readily metabolized by P450 isoforms than was dihydrodiol. Therefore, dihydrodiol can be a potential biomarker for exposure to naphthalene in humans due to the abundant generation and less effective conversion in human liver metabolism. The amount of dihydrodiol formed from naphthalene in mouse lung or liver microsomes in the presence of cytosolic proteins is not changed much over the range of 0 to 2 mg of cytosolic protein concentration (Buckpitt et al., 1984). This observation indicates that cytosolic enzymes may have minimal effects in the formation of dihydrodiol from naphthalene and in the conversion into downstream metabolites.
In summary, human naphthalene metabolism was extensively studied. The metabolic pathway of naphthalene by human liver microsomes and P450 isoforms is shown in Fig. 5. Naphthalene metabolism in pooled human liver microsomes produced trans-1,2-dihydro-1,2-naphthalenediol, 1-naphthol, and 2-naphthol in order of production. The most efficient and important isoforms in human naphthalene metabolism were identified through human P450 isoform screening. Based on the total normalized rates (%TNR), CYP1A2, 3A4, 2E1, and 2A6 are considered to be the most important isoforms in human liver naphthalene metabolism. In these studies, the secondary metabolism of naphthalene was investigated using the primary metabolites as substrates for pHLM and P450 isozymes. CYP1A2 and 2D6, and CYP2A6 and 3A4 were identified as the most efficient isoforms for metabolizing 1-naphthol and dihydrodiol, respectively. Based on the protein contents in human liver, CYP3A4, 1A2, and 2C19 are considered the important isoforms for 1-naphthol metabolism, and CYP3A4, 2A6, and 2C8 were important for dihydrodiol. Because dihydrodiol was less favorable for metabolism by P450 isoforms than naphthalene or other primary metabolites, this metabolite has potential as a biomarker for human exposure to naphthalene.
Acknowledgments
We thank Dr. Alan R. Buckpitt (University of California, Davis, CA) for the generous gift of trans-1,2-dihydro-1,2-naphthalenediol. We also thank Peter Lazaro for technical assistance in the GC/MS analysis.
Footnotes
-
This research was supported by a grant from the U.S. Army (DAMD 17-00-2-008). Part of this study was presented at the 44th Annual Meeting of the Society of Toxicology in New Orleans, LA, March 6–10, 2005.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.105.005785.
-
ABBREVIATIONS: P450, cytochrome P450; dihydrodiol, trans-1,2-dihydro-1,2-naphthalenediol; pHLM, pooled human liver microsome; mEH, microsomal epoxide hydrolase; HPLC, high performance liquid chromatography; GC/MS, gas chromatography-mass spectrometry; DCM, dichloromethane; %TNR, percentage of total normalized rate; CLint, intrinsic clearance; AUC, area under the curve.
- Received May 31, 2005.
- Accepted October 19, 2005.
- The American Society for Pharmacology and Experimental Therapeutics