Abstract
Despite their substantially lower levels relative to hepatic tissue, pulmonary cytochrome P-450 (CYP) monooxygenases play an important role in the metabolic activation of substrates that cause lung injury. The target- and species-selective toxicity of a number of pulmonary toxicants has been attributed to the presence and distribution of activating enzymes with highkcat in target airways of susceptible species. However, experimental demonstration of these concepts and quantitative assessment of the contribution of individual CYP isoforms is lacking. This study was undertaken to characterize the catalytic activities of CYP2F2 with naphthalene, a murine Clara cell toxicant, as well as with other xenobiotics that either undergo metabolic activation to cytotoxic intermediates or that function as “isoform-selective” substrates. Recombinant CYP2F2 was produced using the baculovirus expression vector system in Spodoptera frugiperda andTrichoplusia ni cells, accounting up to ∼20% of the total cellular protein. Incubations containing naphthalene, recombinant CYP2F2, NADPH-cytochrome P-450 oxidoreductase, and NADPH-regenerating system metabolized naphthalene with a high degree of stereoselectivity to 1R,2S-naphthalene oxide (66:1 enantiomeric ratio). The Km andkcat values, along with the specificity constant, for naphthalene metabolism by recombinant CYP2F2 were 3 μM, 104 min−1, and 5.8 × 105M−1 s−1, respectively. Recombinant CYP2F2 also metabolized ethoxyresorufin, pentoxyresorufin,p-nitrophenol, and 1-nitronaphthalene at easily detectable levels. The results from this work suggest that CYP2F2 1) plays a key role in the species- and cell-selective toxicity of naphthalene and 2) efficiently metabolizes a number of other substrates, including the lung toxicant 1-nitronaphthalene.
Numerous studies with laboratory animals have demonstrated the importance of respiratory epithelial cells as targets for both inhaled and ingested chemicals. Specifically, nonciliated bronchiolar epithelial (Clara) cells are very sensitive to toxicants, and this appears to be due to the metabolic capabilities of the Clara cell (Plopper, 1993; Gram, 1997). With several lung toxicants such as trichloroethylene, nitronaphthalene, 3-methylindole, and naphthalene (see Gram, 1997,Yost, 1997; for review), metabolism by cytochrome P-450 (CYP) monooxygenases is an obligate step in the formation of toxicologically active derivatives. The eventual outcome from exposure of a particular tissue/cell appears to depend on a complex interplay among exposure levels, the rate of metabolism of both parent compound and reactive metabolite, and the sensitivity of the cell to the biologically active derivative. A total of seven CYP monooxygenases have been reported in the lung, including CYP1A1, CYP1B1, CYP2B, CYP2E1, CYP2F, CYP3A, and CYP4B1/2. The catalytic activities of a few of these isoforms have been reported using either purified proteins or recombinant enzymes (Buckpitt and Cruikshank, 1997, and Yost, 1997, for reviews; Willey et al., 1996). With the exception of 4-ipomeanol, where the quantitative contributions of CYP2B4 and CYP4B1 in the rabbit were assigned a number of years ago (Wolf et al., 1982; Domin et al., 1986), the roles of each of the pulmonary isoforms in substrate turnover have not been assessed fully.
Substantial species differences have been observed in sensitivity to a variety of lung toxicants. For example, parenteral administration of naphthalene, 2-methylnaphthalene, dichloroethylene, and trichloroethylene results in Clara cell necrosis in mice but not in rats (Plopper, 1993). One of the possible explanations for these species differences is that there is a pulmonary CYP isoform in the mouse that shows high catalytic activity with a number of these agents. The mouse is a particularly sensitive species for naphthalene toxicity, with the lung being the primary target (Plopper et al., 1992). Previous work has demonstrated a correlation between the presence of CYP2F2 in mouse Clara cells with the stereoselective epoxidation of naphthalene and the site-selective cytotoxicity of this chemical in murine Clara cells (Buckpitt et al., 1995). A cDNA copy of CYP2F2 mRNA was isolated from mouse liver and the presence of this mRNA was shown in mouse lung by Northern blot analysis (Ritter et al., 1991). Immunohistochemical localization with antibodies against purified CYP2F2 also confirmed its presence in mouse lung Clara cells (Buckpitt et al., 1995). Furthermore, an inhibitory anti-CYP2F2 antibody markedly alters the ratio of epoxide enantiomers from 10:1 to 1:1 in mouse lung microsomes (Nagata et al., 1990). CYP2F2 has been expressed at low levels in the yeast expression system using Saccharomyces cerevisiae and has been shown to metabolize naphthalene with a high degree of stereoselectivity (Ritter et al., 1991). This evidence localizes CYP2F2 to the mouse lung Clara cell and strongly suggests its involvement in the metabolism and, possibly, the toxicity of naphthalene in mouse lung. Although the yeast-expressed CYP2F2 has been partially characterized with naphthalene, the system has never been optimized and kinetic parameters have not been determined. Accordingly, the objective of this work was to express CYP2F2 at high levels in insect cells and, using optimized conditions, determine the catalytic activity of the recombinant protein with both naphthalene and other relevant substrates.
Materials and Methods
Reagents.
The CYP2F2 cDNA clone in pBluescript SK− was a gift from Dr. J. Ritter (Virginia Commonwealth University, Richmond, VA). Spodoptera frugiperda (Sf9) and Trichoplusia ni (Tn5) cells were obtained from American Type Culture Collection (Manassas, VA) and Invitrogen (Carlsbad, CA), respectively. Ex-Cell 405 medium was purchased from JRH Biosciences (Lenexa, KS). Grace’s medium was from Life Technologies (Rockville, MD). Mouse NADPH-cytochrome P-450 oxidoreductase (reductase) was purified from liver using standard procedures (Strobel and Dignam, 1978). The specific activity of the purified reductase preparation was 3.6 U/mg protein (1 U = 1 μmol cytochrome c reduced/min). Reductase was quantified by the standard cytochrome c reduction assay (Guengerich, 1994). Conversions from cytochrome c units to molar concentrations are based on the assumption that pure reductase reduces 55 μmol cytochrome c/min/mg of the protein under the above assay conditions. Glutathione-S-transferases (GST) from mouse liver cytosol were purified by standard procedures using affinity column chromatography (Simons and Vander Jagt, 1981), and activities were assessed with 1-chloro-2,4-dinitrobenzene as substrate. Recombinant human cytochrome b5 was a generous gift from Dr. M. Shet and Dr. R. Estabrook (University of Texas, Dallas, TX). Unless otherwise stated, all other reagents were purchased from commercial vendors and were of reagent/analytical grade.
Construction of Recombinant Baculovirus.
The cDNA encoding murine CYP2F2 was directionally cloned from pBluescript SK− containing the complete CYP2F2 cDNA (pBluescript SK−/2F2) into the multiple cloning site of pFastBac1 (Life Technologies, Rockville, MD) using two steps. 1) Polymerase chain reaction (PCR) was used to incorporate aBamHI site immediately before the translational start codon of CYP2F2 as well as for introducing an intentional silent mutation (T→C) at position 288 to incorporate a NotI site. The PCR product was digested with BamHI/NotI, gel purified, and ligated into the BamHI/NotI sites in the multiple cloning site of pFastBac1 using T4 DNA ligase. A recombinant construct (pFastBac1/2F2-5′) containing the 5′ 288 bp of the CYP2F2 open reading frame was isolated and sequenced to ensure that unwanted PCR errors had not occurred during amplification. 2) The construct (pFastBac1/2F2-5′) was linearized with XbaI, blunt-ended with T4 DNA polymerase, and digested with NotI, leaving an available NotI end at position 288 and a blunt end downstream from the multiple cloning site of pFastBac1. The remainder of the CYP2F2 open reading frame was prepared from pBluescript SK−/2F2 by digestion withEaeI (leaving a NotI-compatible end as with the PCR product) and with ScaI (leaving a blunt end two nucleotides after the stop codon). This fragment was gel purified and ligated into the linearized pFastBac1/2F2-5′.
A construct (pFastBac1/2F2) containing nucleotides 62→1539 of pBluescript SK−/2F2 was isolated and sequenced to ensure the integrity and orientation of the construct. This construct (with a 1478-bp insert) consists of the complete CYP2F2 ORF (1476 bp) plus two 3′ nucleotides of untranslated sequence. The sequence contains one intentional silent mutation, as mentioned.
Recombinant CYP2F2 Expression.
Baculovirus stocks were prepared using the Bac-to-Bac expression system (Life Technologies, Rockville, MD). The cDNA encoding CYP2F2 from pFastBac1/2F2 was site-specifically transposed into the baculovirus genome as described in the manufacturer’s protocol. The high-molecular-weight genomic DNA was purified from recombinant clones and used to transfect Sf9 cells for the production of baculovirus stocks from various recombinant plasmid constructs. Viral stocks were amplified and stored at 4°C. All baculovirus stocks generated overexpressed a protein product with the same mobility in SDS-polyacrylamide gel electrophoresis (PAGE), yielding similar CYP spectra, and all were capable of metabolizing naphthalene with similar metabolic profiles. Therefore, one stock was chosen for further characterization. Plaque assays showed the viral titer to be 1.2 × 108 plaque-forming units (pfu)/ml.
Sf9 cells grown in complete Grace’s insect cell culture medium and Tn5 cells grown in complete Ex-Cell 405 insect cell culture medium were used as hosts for infection. Infections were performed by incubating monolayers of host cells at a density of 2 × 105 cells/cm2 in 6-well plates or culture flasks with recombinant virus at a multiplicity of infection (m.o.i.) of 1 pfu/cell. Hemin chloride was added 24 h postinfection (p.i.) to a final concentration of 5 μg/ml. Alternatively, 5-aminolevulinic acid (ALA) and ferric citrate (FC) were both added at a concentration of 100 μM at the time of infection to avoid a substantial 422-nm absorption during spectral analysis due to contaminating hemin in the preparations. Recombinant CYP2F2 was harvested at 72 h p.i. Cell lysates were prepared by pelleting cells at 1000g for 10 min followed by washing twice with ice-cold PBS. Pellets were frozen immediately in liquid nitrogen for 5 to 10 min to lyse cells, thawed at 37°C in the appropriate buffer supplemented with 100 μg/ml phenylmethylsulfonyl fluoride, and homogenized in a glass-glass tissue grinder on ice. Buffers consisted of 0.1 M phosphate buffer, pH 7.4, when cell lysates were to be used immediately; 0.1 M phosphate buffer, pH 7.4, with 20% glycerol and 0.1 mM EDTA for storage at −80°C; or 0.1 M phosphate buffer, pH 7.4, with 0.25 M sucrose and 0.1 mM EDTA for microsomal preparations. Cell lysate preparations were adequate for conducting metabolism experiments. Microsomes were prepared by differential centrifugation of cell lysates at 10,000g for 15 min at 4°C followed by 100,000g for 90 min at 4°C. CYP levels were determined by obtaining difference spectra of sodium dithionite-reduced versus CO-bubbled samples at 500 to 400 nm according to Omura and Sato (1964). Noninfected Tn5 or Sf9 cells were used as negative controls and were supplemented and prepared exactly as infected cells.
SDS-PAGE and Western Blotting.
Proteins from either Tn5 cell lysates or microsomes were separated by SDS-PAGE using premade 1-mm 10% Tris-glycine gels (Novex, San Diego, CA). Bands were identified by Coomassie blue staining and by Western blotting. Western blot analysis was performed using the method of Laemmli (1970). Recombinant CYP2F2 was identified using rabbit anti-CYP2F2 (Nagata et al., 1990). Blots were blocked (30 min at room temperature) with 3% BSA/1% nonfat dried milk, hybridized (overnight at 4°C) with rabbit anti-CYP2F2 (1:20,000 dilution); hybridized (15 min at room temperature) with goat anti-rabbit IgG linker antibody (1:100 dilution), hybridized (15 min at room temperature) with a rabbit peroxidase anti-peroxidase complex (1:2000 dilution) (Cappel, Organon Teknika, Durham, NC), and stained (5 min at room temp) with 3,3′-diaminobenzidine.
Measurement of Naphthalene Metabolism.
The rates of naphthalene metabolism were assessed in incubations containing recombinant CYP2F2 (quantity determined spectrally), reductase (quantity determined by cytochrome c activity), NADPH-regenerating system (0.25 U of glucose-6-phosphate dehydrogenase, 14 mM glucose-6-phosphate, 2.18 mM NADP, and 1 mM MgCl2), 1 mM glutathione (GSH), and 2.5 U GST in a total volume of 250 μl (0.1 M Na2HPO4, pH 7.4). CYP2F2 and reductase concentrations are specified in the figure legends. GSH and GST were included for the trapping of reactive epoxides as stable GSH conjugates (Fig. 1). Incubations were performed in a shaking water bath at 37°C for the times specified in the figure legends. Reactions were quenched on ice by the addition of two volumes of methanol. Protein was removed by centrifugation, and the supernatants were evaporated under reduced pressure. GSH conjugates were separated by reversed phase HPLC and were quantified by peak areas using a HP1100 UV detector at 260 nm (Buckpitt et al., 1987). A variation to this method was a change to a new mobile phase system (0.06% triethylamine, pH 3.1/acetonitrile) with a gradient of acetonitrile ranging from 5% to 7% over the first 60 min of a 100-min run. GSH conjugate standards were prepared by synthesis from naphthalene oxide and GSH and were purified by preparative HPLC.
Measurement of Ethoxyresorufin, Pentoxyresorufin, andp-Nitrophenol Metabolism.
Incubations withp-nitrophenol, ethoxyresorufin, and pentoxyresorufin were carried out in 250-μl volumes (0.1 M Na2HPO4, pH 7.4) containing 5 pmol of recombinant CYP2F2, 8.5 pmol of mouse liver reductase, and NADPH-regenerating system. Incubations were performed in a shaking water bath at 37°C for 20 min. Ethoxyresorufin and pentoxyresorufin reactions were quenched on ice with two volumes of methanol. Insoluble material was removed by centrifugation, and the supernatants were diluted with an equal volume of water for HPLC analysis. Resorufin was analyzed by reversed phase HPLC with fluorescence detection as described previously (λex = 535 nm; λem = 585 nm) (Plopper et al., 1993).
p-Nitrophenol incubations were quenched with the addition of 5 μl of trifluoroacetic acid. Samples were centrifuged at 4°C for 30 min to pellet insoluble material. The supernatant was used directly for analysis. The metabolite, nitrocatechol, was separated by reversed phase HPLC and detected electrochemically on an ESA Coulochem II electrochemical detector equipped with a 5010 analytical cell. This procedure has been described in detail by Watt et al. (1997).
Measurement of 1-Nitronaphthalene Metabolism.
Incubations with 1-nitronaphthalene were carried out in 250-μl volumes containing recombinant CYP2F2, reductase, and NADPH-regeneration system as described above. Incubations were performed in the presence of3H-GSH (0.25 mM, 6500 dpm/nmol) and GST (2.5 U) for the trapping of any reactive epoxides as GSH conjugates. Incubations were performed in a shaking water bath at 37°C for 30 min. Reactions were stopped, and samples prepared for analysis as with naphthalene (see above). GST conjugates were separated by reversed phase HPLC using a variation of a method by Watt et al. (1998). The change in the method was to a 0.06% triethylamine, pH 3.1/acetonitrile mobile phase system, with an acetonitrile gradient ranging from 5% to 16% over the first 60 min of a 110-min run. The mobile phase flow rate was 1 ml/min (K. C. Watt, personal communication). Column effluent was monitored at 256 nm using a HP1100 UV detector. Fractions were collected at 0.5-min intervals and counted using a Beckman LS5000TD scintillation counter.
Results
Baculovirus Expression of Recombinant CYP2F2.
The expression of recombinant CYP2F2 was assessed by measuring CO-difference spectra, by SDS-PAGE with Coomassie blue staining, by Western blotting, and, in some instances, by catalytic activity with naphthalene as substrate. Experimental conditions, including the host cell line (Sf9 and Tn5 cells), m.o.i., harvest time after infection, and the use of various supplements, were varied to optimize the generation of functional recombinant protein. Tn5 cells infected with recombinant baculovirus and incubated for 72 h in medium supplemented with ALA/FC produced easily detectable quantities of CYP with a Soret maximum at 451 nm (Fig. 2). The levels of CYP measured by difference spectra were higher in Tn5 cells than in Sf9 cells despite the finding that similar quantities of the 50-kDa protein were present (determined by Coomassie staining; data not shown). Spectra could be obtained by supplementing Tn5 cells with a heme source (hemin chloride or ALA/FC), although the addition of hemin chloride yielded a substantial 422-nm absorption that was also present in negative controls. Supplementation with ALA/FC gave very little or no 422-nm absorption but resulted in lower overall levels of CYP.
There were no significant differences in expression levels determined spectrally between m.o.i. values of 1, 5, 10, and 20 pfu/cell (data not shown), and accordingly, an m.o.i. of 1 pfu/cell was used for infections in all subsequent experiments. Tn5 cell monolayers were infected in 6-well culture plates supplemented with hemin chloride, and cell lysates were prepared at various times after infection for CYP determinations, SDS-PAGE, Western blotting experiments, and measurement of catalytic activity with naphthalene. Optimal harvest times were 72 to 84 h p.i., and there was relatively good correspondence between CYP levels and catalytic activity of CYP2F2 with naphthalene (Fig.3). A yield of 2.7 nmol CYP2F2/mg protein was observed at 72 h. At various times p.i., expression yielded a 50-kDa protein easily detectable by Coomassie staining (Fig.4A). This band approximated 3%, 11%, 20%, 22%, and 29% of the total protein per lane at the 24, 48, 72, 96, and 120 h time points, respectively, as determined by densitometry. This 50-kDa protein was not detected in lysates from either wild-type Autographa californica nuclear polyhedrosis virus (AcNPV)-infected or noninfected cultures. Yield of recombinant CYP2F2 (∼11% at 48 h) was similar to that of the native polyhedrin protein (37.5 kDa) of wild-type AcNPV-infected cells (∼15% at 48 h). The 50-kDa band corresponded with a band identified by Western blotting probed with anti-CYP2F2 antibodies (Fig.4B). Blots containing lysates of noninfected cells did not stain with the same antibody (data not shown).
Metabolism of naphthalene by recombinant CYP2F2 under the conditions described in the methods proceeded with a high degree of stereoselectivity (Fig. 5) as has been shown previously with S. cerevisiae-expressed CYP2F2 (Ritter et al., 1991). The 1R,2S-naphthalene oxide (conjugate 2) was produced at a 66:1 enantiomeric ratio over the 1S,2R-naphthalene oxide (conjugates 1 and 3). No naphthalene metabolites were obtained from incubations containing noninfected cell lysates under similar conditions. Reactions containing recombinant CYP2F2 were inhibited by either bubbling carbon monoxide before incubations or deleting the NADPH-regenerating system (data not shown).
Optimization of Naphthalene Metabolism.
Before conducting kinetic studies, optimal incubation conditions for naphthalene metabolism by recombinant CYP2F2 were established. Naphthalene metabolism was not detected in incubations containing noninfected Tn5 cell lysates, CYP reductase, NADPH-regenerating system, and naphthalene. Similar incubations containing CYP2F2 but without supplemental reductase revealed very low rates of naphthalene metabolism (corresponding to ∼3% of total naphthalene metabolism in complete incubations), indicating a small amount of endogenous electron donor for CYP2F2. Increasing quantities of CYP reductase (assayed by cytochrome c reduction) were added to incubations containing 2.5 pmol of CYP2F2 and other components necessary to trap naphthalene epoxides as GSH conjugates. As the ratio (nmol/nmol) of CYP reductase to CYP was increased from 0.1:1 to 1.4:1, the rates of naphthalene metabolism increased from less than 2 nmol/min/nmol CYP2F2 to more than 90 nmol/min/nmol CYP2F2 (Fig. 6). No further increases were noted with ratios above 1.4:1, and, accordingly, a ratio of 1.7:1 was used for all further metabolism experiments. No alterations in the ratio of 1R,2S- to 1S,2R-epoxide were noted at the varying quantities of reductase in the incubation. Preincubation of the reductase with CYP2F2 was not necessary to obtain optimal catalytic activities (data not shown).
Experiments were conducted to determine whether the addition of cytochrome b5 altered the rates of naphthalene metabolism under optimal incubation conditions. Incubations containing CYP2F2 and optimal amounts of reductase were supplemented with recombinant human cytochrome b5at 1:1 and 3:1 (nmol cytochromeb5/nmol CYP). The addition of cytochrome b5 did have a slight effect on the metabolism of naphthalene by CYP2F2 (Table1).
Kinetics of Naphthalene Metabolism by CYP2F2.
Before kinetic studies, linearity was established with protein and with time, and all kinetic studies were conducted in the linear portion of the curves. The kinetic parameters (Km andkcat) for the metabolism of naphthalene by recombinant CYP2F2 were established with this optimized system. A preliminary experiment indicated a very low apparentKm; accordingly, incubation times were shortened to 4 min to decrease the percentage of naphthalene consumed in the reaction. Substrate concentrations varied from 0.0025 to 1.0 mM naphthalene. Data were analyzed by nonlinear regression (Fig.7) for the estimation of kinetic parameters. The Km of CYP2F2 with naphthalene was 3 μM with a kcat of 104 min−1. The specificity constant of CYP2F2 for naphthalene was calculated to be 5.8 × 105M−1s−1. No alterations in the ratio of 1R,2S- to 1S,2R-epoxide were noted at the varying concentrations of naphthalene in the incubation.
Metabolism of Other Substrates by CYP2F2.
Ethoxyresorufin, pentoxyresorufin, and p-nitrophenol are thought to be isoform-selective substrates for other CYPs in the lung (CYP1A1, CYP2B, and CYP2E, respectively). Incubations containing recombinant CYP2F2, reductase, and the necessary cofactors metabolize all three substrates at easily detectable rates under similar conditions as with naphthalene. Comparing the rates of metabolism for these substrates with published activities for their respective isoforms indicated that the turnover of both ethoxyresorufin and pentoxyresorufin by recombinant CYP2F2 is dramatically slower than that for CYP1A1/1B1 or CYP2B, respectively (Table 2). In contrast, metabolism of p-nitrophenol by CYP2F2 and CYP2E1 occurs at very similar rates.
Incubations containing recombinant CYP2F2, reductase, NADPH-regeneration system, GSH, and GST metabolized the lung toxicant 1-nitronaphthalene to metabolites capable of reacting with GSH. 1-Nitronaphthalene was metabolized to three major metabolites at a combined rate of 1.03 ± 0.27 nmol/min/nmol CYP2F2. Incubations containing mouse liver microsomes metabolized 1-nitronaphthalene (0.45 nmol/min/nmol total CYP) under the same conditions and yielded a similar metabolite profile as with recombinant CYP2F2. Incubations containing noninfected insect cell lysates produced no detectable metabolites under the same conditions as described above.
Discussion
A number of pulmonary toxicants, including naphthalene, show highly species-selective toxicity. Naphthalene shows differential toxicity to the mouse and the rat. The i.p. administration of naphthalene at doses as low as 50 mg/kg produces detectable Clara cell necrosis in mice; toxicity is not discernible in rat lungs even at doses of 1600 mg/kg (Plopper et al., 1992). These species differences in susceptibility correlate well with the rates of naphthalene metabolism to the epoxides (see Fig. 1) in both lung microsomes (Buckpitt et al., 1992) and dissected airway explants (Buckpitt et al., 1995). Furthermore, there are no striking differences in either epoxide hydrolase- or GST-mediated detoxification pathways for naphthalene oxide, which would account for the unusual susceptibility of the mouse compared with the rat (Lorenz et al., 1984). These factors suggest that the initial metabolic activation of naphthalene is a key event in determining the species susceptibility to this compound. The data presented here, demonstrating very high turnover numbers with naphthalene as substrate (kcat ≈ 104 min−1) and a biologically relevantKm (3 μM), support the importance of CYP2F2 in determining species-selective toxicity of naphthalene.
A number of CYP isoforms have been expressed at high levels using the baculovirus-insect cell system, including CYP2A6 (Chen et al., 1997), CYP2D6 (Paine et al., 1996), CYP2E1 (Patten and Koch, 1995), and CYP3A4 (Buters et al., 1994). The advantages of the system, namely that of high levels of expression and targeting to the membrane fraction in comparison with other heterologous expression systems, have been described in detail by others (Lee et al., 1996). In this work, the levels of CYP2F2 in insect cell lysates (2.7 nmol/mg protein) are higher than those reported for CYP2D6 (0.1–0.2 nmol/mg cell lysate), CYP2E1 (0.5–0.8 nmol/mg cell lysate), and CYP3A4 (0.4 nmol/mg cell lysate). The production of recombinant CYP2F2 was similar in yield to that of the polyhedrin protein of wild-type AcNPV-infected cells. The time of optimal expression was similar to that of studies reported previously with other CYP monooxygenases at ∼72 h p.i. These data confirm the efficient expression of recombinant CYP2F2 in the baculovirus expression system. The data showing cross-reactivity with anti-CYP2F2 antibody confirm that recombinant CYP2F2 possesses the antigenic properties of the native pulmonary enzyme. Furthermore, the metabolism data confirm that recombinant CYP2F2 possesses the functional characteristics of the native enzyme-metabolizing naphthalene to the 1R,2S-oxide over the 1S,2R-oxide at a ratio of 66:1, which is similar to the ratio of 30:1 in mouse lung microsomes (Buckpitt et al., 1992).
The results of the current study, demonstrating the requirement for the addition of saturable levels of CYP reductase to achieve optimal catalytic activity, was similar to other work with baculovirus-expressed CYPs such as CYP3A4 (Buters et al., 1994), CYP2D6 (Paine et al., 1996), and CYP2E1 (Patten and Koch, 1995). We are confident, having reached saturation with reductase, that maximum turnover rates of naphthalene were achieved with recombinant CYP2F2. Similar experiments in which mouse lung microsomes were supplemented with additional purified reductase yielded no enhancement of naphthalene turnover (data not shown), implying that CYP2F2 operates under saturating reductase conditions in microsomal preparations.
In subsequent metabolism experiments using CYP2F2 with different reductase preparations, we have found that the addition of 3-[(3-cholamidopropyl)dimethylammonio]propanesulfonate is necessary to reach the optimum turnover rates with naphthalene as reported here. The addition of 3-[(3-cholamidopropyl)dimethylammonio]propanesulfonate, however, does not stimulate metabolism beyond the catalytic activities presented here.
Earlier work has demonstrated that the addition of cytochromeb5 can markedly increase the catalytic activities of some CYP isoforms, including CYP2E1 (Patten and Koch, 1995) and CYP2F3 (Wang et al., 1998). There was a slight increase in the metabolism of naphthalene by CYP2F2 when supplemented with cytochrome b5. Although the differences were statistically significant (p < .05), metabolism of naphthalene increased only marginally by 18% to 20% at a 1:1 or 3:1 ratio (nmol cytochromeb5/nmol CYP). In comparison, the addition of cytochrome b5 at similar ratios with CYP2E1 incubations increased the metabolism ofN-nitrosodimethylamine, ethoxycoumarin, andp-nitrophenol by 3-, 6-, and more than 10-fold, respectively. The addition of cytochromeb5 at similar ratios with CYP2F3 incubations increased the metabolism of 3-methylindole by 6-fold. In relation to these examples, supplementation of CYP2F2 with cytochromeb5 did not increase the metabolism of naphthalene substantially.
Comparative data with CYPs isolated from different species must be interpreted cautiously; slight sequence variations between species may markedly influence the catalytic activities of the proteins. Ethoxyresorufin, pentoxyresorufin, and p-nitrophenol are used as isoform-selective substrates for CYP1A1, CYP2B, and CYP2E1, respectively. The current work shows that recombinant CYP2F2 metabolizes both ethoxyresorufin and pentoxyresorufin at rates that are considerably slower than the turnover observed with ethoxyresorufin-O-dealkylation by CYP1A1 (37 nmol/min/nmol CYP; Buters et al., 1995) and with pentoxyresorufin-O-dealkylation by CYP2B (1.18 nmol/min/nmol CYP2B4 or 3.15 nmol/min/nmol CYP2B5; Szklarz et al., 1996). These data indicate that the presence of CYP2F2 would likely contribute very little when metabolism assays are performed with these diagnostic substrates. In contrast, even under unoptimized incubation conditions, recombinant CYP2F2 metabolizes p-nitrophenol at a rate (2.5 nmol/min/nmol CYP) that is close to the rate reported with recombinant CYP2E1 under similar conditions (2.01 and 4.00 nmol/min/nmol for human and rat CYP2E1, respectively; Chen et al., 1996). These data indicate the need to evaluate results using “isoform-specific” substrates, at least in the case of CYP2E1, with more caution.
Recombinant CYP2F2 also metabolizes the lung toxicant 1-nitronaphthalene to metabolites capable of reacting with GSH (as with naphthalene metabolites). The rate of formation of these reactive species totaled 1.03 ± 0.27 nmol/min/nmol CYP, twice the turnover obtained from similar incubations containing mouse liver microsomes and similar to the turnover by mouse lung microsomes (1 nmol/min/nmol CYP; K. C. Watt, personal communication). These data provide preliminary evidence that CYP2F2 catalyzes the conversion of 1-nitronaphthalene to reactive metabolites at a rate similar to, if not higher than, that observed in mouse lung or liver tissues.
A total of four isoforms from the CYP2F family have been cloned, and limited studies of catalytic activity have been reported. CYP2F1, from human lung, has been expressed using vaccinia virus-infected HepG2 cells (Nhamburo et al., 1990). Recombinant CYP2F1 catalyzed the dealkylation of ethoxycoumarin, propoxycoumarin, pentoxyresorufin, and benzyloxyresorufin (Nhamburo et al., 1990) and the metabolic activation of 3-methylindole (Thornton-Manning et al., 1996). CYP2F3, cloned from goat lung and expressed in Escherichia coli (Wang et al., 1998), metabolizes 3-methylindole to reactive metabolites (trapped as mercapturic acid conjugates) with an apparentKm value of 0.34 mM and aVmax value of 0.55 nmol/min/nmol CYP. These kinetic data along with antibody inhibition studies that indicate ∼85% of the microsomal metabolism of 3-methylindole is catalyzed by CYP4B2 suggest that CYP4B2 is probably responsible for the metabolic activation of the substrate, at least in the goat. The high turnover rates observed in the current work (104 min−1) along with the low Km value (3 μM) supports the view that CYP2F2 is important in the metabolic activation of naphthalene in vivo.
A fourth member of the CYP2F family has been cloned from rat, sequenced, and expressed in Tn5 cells (CYP2F4; Baldwin et al., 1998). The deduced amino acid sequence is 93% and 80% identical with the mouse and human, respectively. Current studies that focus on defining the kinetics and stereochemistry of naphthalene metabolism by recombinant CYP2F4 will help determine whether the lack of sensitivity of rat lung to naphthalene is due to low CYP2F levels or to low catalytic activities of this protein.
Whether a particular isoform of CYP is responsible for the metabolic activation of a lung toxicant is dependent not only on the kinetics of metabolism of that toxicant but also on the abundance of the protein within the target cells. Using the techniques developed by Plopper et al. (1991) for preparing subcompartments of the lung, we mapped out the total CYP levels in airways of the mouse. Using the recombinant protein as a standard, we are currently determining the levels of CYP2F2 in these individual lung compartments of the mouse by Western blot analysis. Preliminary data indicate that CYP2F2 is a very abundant isoform, if not the majority of CYP, in mouse airways.
This study demonstrates high catalytic activity of CYP2F2 toward naphthalene at biologically relevant levels (Km = 3 μM) and provides limited data on the metabolism of the lung toxicant 1-nitronaphthalene. The data generated in this work, together with the relative abundance of CYP2F2 in mouse airways, and the sensitivity of mouse lung to toxicity raise an interesting question in regard to the possible involvement of CYP2F2 in rendering the mouse lung inherently sensitive to a number of metabolically activated compounds. Current work to determine kinetic parameters for CYP2F2 with a variety of metabolically activated lung toxicants (including those mentioned above) should address the possible in vivo importance of CYP2F2 in contributing to what appears to be the unique sensitivity of the mouse to metabolically activated lung toxicants.
Acknowledgments
We thank Dr. Joe Ritter (Virginia Commonwealth University) for providing the CYP2F2 cDNA clone; Dr. John Chandler for his advice in the subcloning of CYP2F2; Dr. Byung-Rae Jin for his assistance and advice in the production of recombinant CYP2F2; Dr. Manju Shet, Dr. Ron Estabrook, and Dr. Alan Conley for providing the cytochromeb5 necessary to conduct our experiments; and Dr. Katherine Watt for providing her method for the analysis of 1-nitronaphthalene metabolism.
Footnotes
-
Send reprint requests to: Dr. Michael Shultz, Department of Molecular Biosciences, School of Veterinary Medicine, University of California, Davis, CA 95616. E-mail: mashultz{at}ucdavis.edu
-
↵1 This work was supported by National Institute of Environmental Health Sciences Grants ES08408, ES04699, and ES04311. University of California, Davis is a Center for Environmental Health Sciences (Grant ES05711), and support for core facilities used in this work is gratefully acknowledged. M.A.S. is supported by NIEHS Predoctoral Fellowship ES05707.
- Abbreviations:
- Sf9
- Spodoptera frugiperda
- Tn5
- Trichoplusia ni
- pfu
- plaque-forming units
- GST
- glutathione-S-transferase
- AcNPV
- Autographa californica nuclear polyhedrosis virus
- PAGE
- polyacrylamide gel electrophoresis
- p.i.
- postinfection
- m.o.i.
- multiplicity of infection
- GSH
- glutathione
- ALA
- 5-aminolevulinic acid
- FC
- ferric citrate
- PCR
- polymerase chain reaction
- Received November 30, 1998.
- Accepted March 16, 1999.
- The American Society for Pharmacology and Experimental Therapeutics