In vitro metabolism of 1,2,3,3,3-pentafluoropropene (PFP) was investigated in the present study. PFP was metabolized via cytochrome P450-catalyzed oxidative dehalogenation in liver microsomes and glutathione transferase (GST)-catalyzed conjugation in liver microsomes and cytosol. Two oxidation products, 2,3,3,3-tetrafluoropropionaldehyde (TPA) and 3,3,3-trifluoropyruvaldehyde (TFPA), and two GSH conjugates, S-(2,3,3,3-tetrafluoropropenyl)-GSH (TFPG) and S-(1,2,3,3,3-pentafluoropropyl)-GSH (PFPG) were identified. Enzyme kinetic parameters for the formation of TFPA, TFPG, and PFPG were obtained in male and female rat, mouse, dog, and human liver microsomes and cytosol and were confirmed using freshly isolated male rat hepatocytes. For the TFPA pathway, dog microsomes exhibited much larger Km values than rat, mouse, and human microsomes. Sex differences in the rates of metabolism within a given species were minor and generally were less than 2-fold. Across the species, liver microsomes were the primary subcellular fraction for GSH S-conjugation and the apparent reaction rates for the formation of TFPG were much greater than those for PFPG in liver microsomes. PFPG was unstable and had a half-life of approximately 3.9 h in a phosphate buffer (pH 7.4 and 37°C). The intrinsic clearance values for the formation of TFPA were much greater than those for the formation of GSH S-conjugates, suggesting that cytochrome P450-mediated oxidation is the primary pathway for the metabolism of PFP at relatively low PFP concentrations. Because saturation of the GST-mediated reactions was not reached at the highest possible PFP concentration, GSH S-conjugation may become a much more important pathway at higher PFP concentrations (relative to the Km for TFPA).
1,2,3,3,3-Pentafluoropropene (PFP) belongs to a group of fluoroalkenes having chemical and environmental properties that make them attractive alternatives to substances with high global warming potential. A number of fluoroalkenes are found to be nephrotoxic in rats, such as hexafluoropropene (Koob and Dekant, 1990), tetrafluoroethylene (Odum and Green, 1984), and 2-(fluoromethoxy)-1,1,3,3,3-pentafluoro-1-propene (compound A) (Iyer and Anders, 1997). The mechanism is considered to involve a multistep bioactivation pathway including hepatic GSH S-conjugation of the fluoroalkenes, followed by enzymatic hydrolysis of the GSH S-conjugates to cysteine S-conjugates, renal uptake of cysteine S-conjugates, and formation of reactive species via the renal cysteine S-conjugate β-lyase pathway (Anders and Dekant, 1998). Compared with rats, humans have much less risk for fluoroalkene-induced nephrotoxicity (Anders and Dekant, 1998; Altuntas and Kharasch, 2001, 2002; Altuntas et al., 2003), because the renal β-lyase activity in humans is much lower than that in rats (Anders and Dekant, 1998; Altuntas and Kharasch, 2001, 2002; Altuntas et al., 2003).
In addition to GSH S-conjugation, cytochrome P450 (P450)-mediated oxidation is also a possible metabolic pathway for fluoroalkenes. This oxidative pathway was not observed for some perfluorinated alkenes, such as hexafluoropropene (Koob and Dekant, 1990) and tetrafluoroethylene (Odum and Green, 1984), but appears to be sensitive to the degree of fluorination and is enhanced by chloro substitutions (Bolt et al., 1982; Baker et al., 1987).
To understand the relative importance of the two metabolic pathways for PFP in different species, we investigated the oxidation and GSH S-conjugation potentials of PFP in rat liver microsomes and cytosol and compared enzyme kinetics for the formation of one stable oxidation product and two GSH S-conjugates in male and female mouse, rat, dog, and human liver microsomes and cytosol. Freshly isolated male rat hepatocytes were used to confirm the metabolic pathways and enzyme kinetic parameters obtained in the liver subcellular fractions.
Materials and Methods
PFP (CAS 2252-83-7, purity >99.9%) was provided by DuPont Fluoroproducts (Newark, DE). Its identity and purity were confirmed by NMR, a GC-flame ionization detector, and GC-mass selective detector methods. 3,3,3-Trifluoropyruvaldehyde (TFPA) (CAS 91944-47-7, purity >99%) was purchased from Apollo Scientific (Bredbury, UK). 2,3,3,3-Tetrafluoropropanoic acid (TFPAA) (CAS 359-49-9, purity >97%) was purchased from Oakwood Products (West Columbia, SC). Pooled liver microsomes and cytosol from rat, mouse, dog, and human were purchased from either In Vitro Technologies (Baltimore, MD) or XenoTech, LLC (Lenexa, KS). Cell culture media and buffers were obtained from Invitrogen (Carlsbad, CA). 2,4-Dinitrophenylhydrazine (DNPH) was purchased from Sigma-Aldrich (St. Louis, MO). All other chemicals, if not specified in the article, were obtained from Sigma-Aldrich.
Male Crl:CD(SD)IGS BR rats were obtained from Charles River Laboratories (Wilmington, MA). Upon arrival, all animals were housed in quarantine for at least 4 days. Animals were provided tap water ad libitum and fed Certified Rodent LabDiet 5002 (PMI Nutrition International, St. Louis, MO) ad libitum. At the time of hepatocyte isolation, rats were 6 to 8 weeks of age. Animal rooms were maintained at a temperature of 18–26°C and a relative humidity of 30 to 70% and were artificially illuminated (fluorescent light) on a 12-h light/dark cycle.
Rat Hepatocyte Isolation.
The procedures for isolation of rat hepatocytes were described previously (Nabb et al., 2006; Mingoia et al., 2007). Cells were suspended in Leibovitz L-15 medium at pH 7.4. Cells were counted in the presence of 0.04% trypan blue. The viabilities of the cells were approximately 90%.
Phosphate Buffer-Air Partition of PFP.
PFP dose concentrations were prepared by diluting pure gaseous compound in Tedlar bags (SKC Inc., Eighty Four, PA) containing known volumes of room air. Approximately 35 ml of known concentration PFP gas was purged through a 10-ml gas-tight glass vial that contained 2 ml of 0.1 mM potassium phosphate buffer (pH 7.4). After incubation at 37°C for 2 h, 1 ml of the phosphate buffer was quickly transferred via an air-tight syringe to another PFP-free vial. The second vial was heated at 75°C for approximately 20 min. The PFP concentration in the headspace of the second vial, which corresponds to the concentration of dissolved PFP in the phosphate buffer, was determined by a GC-flame ionization detection method on an Agilent HP 6890 instrument (Agilent Technologies, Santa Clara, CA) equipped with a Agilent HP-5 column (30 m × 320 μm, 0.25-μm film thickness) and a Gerstel MP2 autosampler.
Biosynthesis of TFPG.
In a 40-ml air-tight glass flask, male rat microsomes (2 mg protein/ml) were incubated at 37°C with pure PFP gas and 10 mM GSH in 10 ml of 0.1 M potassium phosphate buffer, pH 7.4, containing 1% ethanol. After a 2-h incubation, the sample was concentrated using a MF C18 Isolute solid-phase extraction column (Biotage, Uppsala, Sweden) in acetonitrile-water (1:1). TFPG was purified by an LC-MS method. The sample was evaporated to dryness on a speed vacuum and redissolved in approximately 1 ml of CD3CN-D2O (2:8) for NMR analysis.
All enzymatic reactions were stopped by the addition of a 10% volume of 35% perchloric acid (PCA) and with brief centrifugation to remove the protein precipitants.
For metabolite identification experiments, 1 mg/ml microsomes or cytosol from male rats was incubated for 1 h at 37°C with 120,000 ppm PFP in 0.1 M potassium phosphate buffer, pH 7.4, containing 10 mM GSH. The microsomal samples also contained a NADPH-regenerating system consisting of 0.1 mM EDTA, 10 mM glucose 6-phosphate, 3.6 U/ml glucose 6-phosphate dehydrogenase, 15 mM MgCl2, and 0.525 mM NADP. After PCA precipitation and centrifugation, the samples were analyzed directly by the LC-MS method. Aliquots of the microsomal samples were also mixed with an equal volume of DNPH derivatization solution (0.3% DNPH in water-concentrated HCl-acetonitrile, 51:29:20) before analysis for the identification of ketone or aldehyde metabolites (Olson and Swarin, 1985; van Leeuwen et al., 2004).
The enzyme kinetics for the formation of TFPA were determined in 0.1 M potassium phosphate buffer, pH 7.4, containing 0.5 mg/ml microsomal proteins, the NADPH-regenerating system, and varied PFP concentrations from 250 to 50,000 ppm. The reactions were equilibrated at 37°C for 10 min before being initiated with the addition of 0.525 mM (final concentration) of NADP and were stopped 20 min after the initiation. The samples were mixed with an equal volume of the DNPH solution for LC-MS quantification. TFPA standards were prepared in a heat-inactivated microsomal matrix that contained exact compositions of the microsomal reactions mentioned above except for the PFP dose and were also derivatized with an equal volume of the DNPH solution before quantification.
The enzyme kinetics for the formation of TFPG and PFPG were determined in 0.1 M potassium phosphate buffer, pH 7.4, containing 0.2 mg/ml microsomal proteins or 0.5 mg/ml cytosol proteins and PFP concentrations ranging from 100,000 to 1,000,000 ppm. For spontaneous reactions, heat-inactivated microsomes or cytosol at the same protein concentrations was used instead. The reactions were equilibrated at 37°C for 10 min before being initiated with the addition of 10 mM (final concentration) GSH and were stopped 60 min after the initiation. Immediately after PCA precipitation and brief centrifugation to remove the precipitants, the samples were quickly chilled to 4°C and analyzed by the LC-MS methods described under Sample Analysis. Biosynthesized TFPG was prepared in a heat-inactivated microsomal or cytosol matrix and was used as a quantification standard for both TFPG and PFPG samples.
For enzymatic reactions in isolated rat hepatocytes, PFP at various concentrations was incubated with 2 × 106 cells/ml hepatocytes in Leibovitz L-15 medium at pH 7.4 and 37°C for 20 (TFPA samples) or 60 min (TFPG and PFPG samples). The samples were analyzed in the same way as described above.
In a 10-ml gas-tight glass vial, pure PFP gas was allowed to react with 10 mM GSH in 3 ml of 0.1 mM potassium phosphate buffer at pH 7.4. After incubation at 37°C for 2 h, the phosphate buffer was transferred to another PFP-free vial that was preincubated at 37°C. The vial was loosely capped, and aliquots of the buffer were taken at different time points over the incubation period. The PFPG level at each time point was determined by the LC-MS method. Spontaneous degradation of PFPG in the buffer was assumed, following first-order kinetics: where t is incubation time in hours; C(0) and C(t) are PFPG concentrations at time 0 and t, respectively; and t1/2 is PFPG half-life in hours.
NMR data were acquired on a Bruker Avance 400 MHz spectrometer using a 19F, 1H, 13C triple resonance observe probe optimized for 19F detection. Fluorine TFPG spectra were acquired in CD3CN-D2O (2:8) with a spectra window of 100 KHz, an acquisition time of 1.3 s, and 9431 scans. Probe calibrations for pulse width and signal/noise ratio were performed with 0.05% α,α,α-trifluorotoluene in C6D6. The fluorine chemical shift was calibrated using CFCl3. Each sample was tuned and matched before data acquisition. Spectra were acquired at 27°C. The concentration of the biosynthesized TFPG, which was used as an analytical standard for the enzyme kinetic studies, was determined by comparing the intensity of the CF3 resonance of TFPG (−69.8 ppm) against the CF3 resonance of a set of TFPAA (−74.3 ppm) standards at known molar concentrations.
Samples for metabolite identification experiments were analyzed using two LC-MS/MS systems. The first LC-MS/MS system included an Applied Biosystems 4000 QTrap mass spectrometer (Applied Biosystems, Foster City, CA), an Agilent 1100 high-performance liquid chromatograph), and a CTC PAL autosampler (LEAP Technologies, Carrboro, NC). Samples were eluted on an Agilent Zorbax SB-C18 column (2.1 × 30 mm, 3.5-μm) by maintaining 98% eluent A (2 mM ammonium acetate in water) for 0.5 min and with a linear gradient, changing to 100% eluent B (methanol) within the next 13 min. The flow rate was kept at 0.3 ml/min. Turbo spray ionization in negative and positive ion modes was used for the QTrap spectrometer, and the probe temperature was set at 450°C. The second LC-MS/MS system included a Waters Acquity Ultra Performance liquid chromatograph and Q-Tof II mass spectrometer (Waters, Milford, MA). Samples were analyzed with the same LC conditions except that a 1.8-μm Zorbax column was used instead. The electrospray source conditions for the Q-Tof spectrometer were 2.5 kV for the capillary, 35 V for the cone, 120°C for the source temperature, and 350°C for the desolvation temperature.
Isolation of biosynthesized TFPG was conducted on an Agilent 1100 LC system and a Waters ZQ mass spectrometer. Samples were separated on a Zorbax SB-C18 column (2.1 × 150 mm, 5-μm) by maintaining 98% eluent A (0.1% formic acid in water) for 1 min and with a linear gradient, changing to 40% eluent B (0.1% formic acid in acetonitrile) within the next 4 min. The flow rate was kept at 0.5 ml/min.
Samples for enzyme kinetics measurements were analyzed on a Waters Quattro Micro mass spectrometer and a Waters 2795 high-performance liquid chromatograph. DNPH-derivatized TFPA samples were separated on a Waters XTerra column (C18, 2.5-μm, 2.1 × 30 mm) within 6 min by linear gradient from 80% eluent A (2 mM ammonium acetate in water) to 80% eluent B (acetonitrile) at a flow rate of 0.3 ml/min. Data were acquired in negative ion mode with a capillary voltage at 3 kV, a cone voltage at 25 V, source temperature at 120°C, and desolvation temperature at 350°C by single ion recording of the deprotonated molecular ion (M − H)− of 305 m/z.
The LC conditions for TFPG and PFPG samples were analogous to the method used for TFPG isolation. Data were acquired in positive ion mode with a capillary voltage at 3 kV, a cone voltage at 15 V, source temperature at 120°C, and desolvation temperature at 350°C by single ion recording of the protonated molecular ions (M + H)+ of 420 m/z (TFPG) and 440 m/z (PFPG).
The velocity versus substrate concentration, V/[S], plots for metabolite TFPA were fitted to the Michaelis-Menten equation in software package Origin (version 7.0; OriginLab Corp., Northampton, MA) to obtain Vmax and Km values. Intrinsic clearance (Vmax/Km) values for the GSH S-conjugates were the average of the V/[S] ratios at one to four different PFP concentrations that were low relative to the Km.
All the supporting figures for the metabolite identification of PFP are summarized in Supplemental Figs. 1 to 10. Figure 1 shows PFP metabolites identified in male rat liver microsomes in the presence of the NADPH-regenerating system and GSH. In negative ion mode, DNPH-derivatized TFPA has a molecular ion of 485 m/z with major fragments of 465 (loss of HF), 445 (loss of 2HF), 302, 279, and 182 m/z. Its product ion spectrum and postulated fragments were presented in S1 and S2, respectively. The retention time and the product ion spectrum of the TFPA metabolite matched those of the authentic standard (data not shown). DNPH-derivatized TPA has a molecular ion of 309 m/z with major fragments of 289 (loss of HF), 269 (loss of 2HF), 241, 231, 214, and 181 m/z. Its product ion spectrum and postulated fragments were presented in S3 and S4, respectively. No authentic standard was available for this metabolite, and there was a possibility that this metabolite was an isomeric ketone. However, the fragmentation pattern, especially multiple losses of HF to form ion m/z 269 or 241, suggests that the metabolite is an aldehyde. The rationale is mainly based on the findings that a similar type of fragmentation pattern was observed for 3,3,3-trifluoropropanal derivatized with DNPH, but not for DNPH-derivatized 1,1,1-trifluoroacetone (data not shown).
The product ion spectra and postulated structures of fragment ions for the two GSH S-conjugates, TFPG and PFPG, were presented in S5 to S8. The molecular ions for TFPG and PFPG in negative ion mode were 418 and 438 m/z, respectively. The majority of the fragment ions observed in the spectra (m/z 179, 210, 254, and 272) originated from the fragmentation of the GSH moiety (Naylor et al., 1988). Proton decoupled and coupled 19F NMR spectra of a biosynthesized TFPG sample were shown in S9 and S10. The CF3 and the single fluorine signals were at approximately −69.8 and −130.8 ppm, respectively. In both proton coupled and decoupled spectra, the CF3 group was a clean doublet. In addition, the single fluorine signal does not appear to be germinally coupled to a proton. These data were consistent with the single fluorine position we proposed for TFPG. We could not obtain an acceptable PFPG NMR spectrum due to the instability of PFPG (see PFPG Stability). The structure of PFPG was proposed on the basis of its molecular mass, the similarities between the mass spectra from TFPG and PFPG, and the addition mechanism for the formation of GSH S-conjugates of fluoroalkenes (Koob and Dekant, 1990; Anders and Dekant, 1998).
We also obtained the authentic standard for TFPAA and looked for possible existence of this metabolite in the hepatocyte samples that were used for the enzyme kinetic study. We did not find a corresponding peak in the hepatocyte samples at the retention time of the standard.
Phosphate Buffer-Air Partition of PFP.
Partitioning of the gaseous compound PFP to 0.1 M potassium phosphate buffer (pH 7.4) was determined at 37°C. Figure 2 shows that concentrations of PFP in the buffer were proportional to the compound concentrations in the gas phase up to 1,000,000 ppm (the pure gas). Therefore, the substrate concentrations of PFP in the enzyme kinetics studies were calculated using 1.153 nM (in the buffer)/ppm in the gas phase. The effect of protein on the solubility of PFP in the buffer was considered negligible because the metabolite formation was linear over a wide range of the protein concentrations from 0.1 up to 2 mg/ml (see Enzyme Kinetics for the Formation of TFPA in Liver Microsomes and Enzyme Kinetics for the Formation of GSH S-Conjugates in Liver Microsomes and Cytosol for details).
Enzyme Kinetics for the Formation of TFPA in Liver Microsomes.
The velocities of TFPA formation in male rat liver microsomes were found to be linear to the microsomal protein concentrations between 0.1 and 1 mg/ml and up to 30-min incubation time (data not shown). The representative V/[S] curves of TFPA obtained in male rat, mouse, dog, and human liver microsomes (0.5 mg/ml) and the 20-min incubation are shown in Fig. 3. All of the V/[S] curves were fitted to the Michaelis-Menten equation to obtain the enzyme kinetic parameters (Table 1). All species produced TFPA in their liver microsome samples. The order of the Vmax values was dog > human > mouse > rat. The sex differences were minor relative to the differences between the species. Km values were similar for rat, mouse, and human and ranged from 1.33 to 2 μM. Dog had much larger Km values: ∼61 μM for male and ∼88 μM for female. As a result, intrinsic clearance rates (Vmax/Km) for the dog were the lowest compared with those for other species. Humans in general had the highest rate of intrinsic clearance via the oxidative pathway to TFPA, with the exception of the minor difference between female mouse and female human.
Enzyme Kinetics for the Formation of GSH S-Conjugates in Liver Microsomes and Cytosol.
Spontaneous, nonenzymatic GSH S-conjugations of PFP to form TFPG and PFPG were observed. Example of TFPG spontaneous formation is shown in Fig. 4. The rates obtained in heat-inactivated microsomes and cytosol were lower than the rates from live microsomes and cytosol and roughly ranged from one-tenth to one-half the enzymatic reactions (data not shown). The velocities of TFPG and PFPG formation were linear to the protein concentrations between 0.1 and 1 mg/ml for male rat microsomes and between 0.25 and 2 mg/ml for male rat cytosol within 60 min of incubation (data not shown). In experiments to determine the Km, there was a linear increase in the velocities of TFPG and PFPG formation up to the highest possible substrate concentration in the gas phase (pure PFP gas at 1,000,000 ppm) (TFPG formation is shown in Fig. 4). The intrinsic clearance rates for PFP via the GSH S-conjugation pathway were determined by linear regression, and the contribution from the enzymatic reactions is summarized in Table 2. For all species, TFPG was formed predominantly in the microsomal samples, whereas PFPG was formed at a relatively higher rate in the cytosol. TFPG, if judged by the combined rates of formation in microsomes and cytosol, was the major product for all species. The order of PFP clearance rates via TFPG formation in the microsomes was female rat > male rat ≈ human > dog ≈ mouse.
Enzyme Kinetics for the Formation of TFPA, TFPG, and PFPG in Freshly Isolated Hepatocytes.
The parameters for the formation of TFPA, TFPG, and PFPG in male rat hepatocytes are summarized in Table 3. The intrinsic clearance value for TFPA pathway was more than 500 times larger than the combined values for the GSH S-conjugates. For the two GSH S-conjugates, intrinsic clearance was approximately 5 times faster for the TFPG pathway than that for the PFPG pathway. The contributions of spontaneous reaction to the clearance rates of TFPG and PFPG were not characterized because of the lack of appropriate controls for the hepatocyte samples.
GSH S-conjugate PFPG was unstable and underwent spontaneous degradation in both acetonitrile and aqueous solutions. Figure 5 shows the disappearance of PFPG as a function of time in 0.1 M potassium phosphate buffer at pH 7.4 and 37°C. The half-life of PFPG under such conditions was approximately 3.9 h. No effort was made to identify the degradation product of PFPG.
Comparative in vitro metabolism of PFP was investigated in the present study. PFP underwent GST and P450-mediated metabolism to form the GSH S-conjugates TFPG and PFPG and oxidation products TFPA and TPA (Fig. 1). Human and rat intrinsic clearance values for the formation of GSH S-conjugates were quite comparable and greater than those for mouse and dog (Table 2). For the TFPA pathway, dog exhibited somewhat different enzyme kinetics relative to that of the other species tested, and human had the highest intrinsic clearance values (Table 1). The differences between the genders were minor (Tables 1 and 2).
We confirmed the results obtained from liver subcellular fractions with freshly isolated male rat hepatocytes. Both oxidation and conjugation pathways were observed in rat hepatocytes. Quantitatively, the intrinsic clearance value for the TFPA pathway was approximately 3 times higher in the hepatocytes than that in the microsomes, mainly because of the 3 times smaller Km value observed in the hepatocytes (Table 1 versus Table 3). For the two GSH S-conjugates, the clearance values obtained from the hepatocytes were quite comparable to the values from liver subcellular fractions (Table 2 versus Table 3). Overall, we have demonstrated that our combined microsomal and cytosol data were in good agreement with the results obtained from intact liver cells.
The mechanism of P450-mediated oxidation of halogenated alkenes was described previously (Testa, 1995) as the formation of a carbocation intermediate, which collapsed to an epoxide and an aldehyde or ketone. The oxidation of PFP in liver microsomes is consistent with this mechanism. For PFP, the formation of the epoxide could be transient and readily hydrolyzed to the stable ketoaldehyde product TFPA via CYP2E1 (Schuster et al., 2008). The aldehyde product TPA, as opposed to a ketone, was also identified in our study (Fig. 1), which, however, could be a product of direct P450 oxidation without bypassing an epoxide intermediate (Schuster et al., 2008). We did not find the downstream acid product TFPAA in our in vitro samples but cannot rule out the possibility of this acid being formed in vivo.
In our study, both addition and addition-elimination mechanisms (Anders and Dekant, 1998) were observed in the GST-catalyzed reaction of PFP to form PFPG and TFPG, respectively. Across species, the reaction rates for the formation of TFPG were much greater than that for PFPG in liver microsomes. In contrast, PFPG was generally the preferred product in cytosol. This pattern agreed with the previous observation for hexafluoropropene (Koob and Dekant, 1990). Overall, TFPG was the major GSH S-conjugation product as confirmed in our hepatocyte experiments (Table 3). PFPG was quantified using TFPG as the analytical standard, primarily because of the issue of stability of PFPG in the solvent (Fig. 5). We do not anticipate a huge difference in ionization efficiency between TFPG and PFPG, but it is important to note that absolute quantification of PFPG was not achieved in this study. We have also observed that, at physiological conditions (pH 7.4 and 37°C), PFPG had a half-life of approximately 3.9 h (Fig. 5). In contrast, TFPG was considerably more stable (data not shown). This observation, combined with our enzyme kinetic data for TFPG and PFPG, leads us to conclude that TFPG could be more physiologically and toxicologically relevant than PFPG.
Our results showed that intrinsic clearance of PFP via P450-catalyzed oxidation to form TFPA was much faster than the clearance from liver GST-catalyzed conjugation (Table 3), suggesting that the P450 mechanism is the primary pathway for the metabolism of PFP. It is important to note that the Km value for the formation of TFPA was quite small and usually less than 2 μM. This finding indicates that the GST pathway could become relatively more important at a higher substrate concentration. It can be estimated that in male rat hepatocytes, 260 μM PFP would allow the formation of TFPG at a rate of 65 pmol · min−1 · 106 cells−1, the Vmax value for TFPA formation (Table 3). In addition, we observed a spontaneous, nonenzymatic reaction between GSH and PFP (Fig. 4), which could contribute significantly to the overall formation of GSH S-conjugates of PFP in the body.
It has been proposed that the P450-mediated oxidation reactions of haloalkenes could form reactive intermediates that are hepatotoxic (Costa and Ivanetich, 1980; Bolt et al., 1982; Ortiz de Montellano et al., 1982; Baker et al., 1987; Yoshioka et al., 2002), and the GST-mediated GSH S-conjugation of haloalkenes could initiate a bioactivation pathway for nephrotoxicity (Odum and Green, 1984; Koob and Dekant, 1990; Lock and Ishmael, 1998). It was also suggested that the preference of P450 versus GST-mediated pathways for the haloalkenes to some extent correlated with their specific toxicities (hepatotoxicity versus nephrotoxicity) observed in vivo (Commandeur et al., 1987; Dekant et al., 1987). In this study, we confirmed both oxidation and GSH S-conjugation metabolism of PFP in vitro and obtained enzyme kinetic parameters for the formation of one oxidation product and two GSH S-conjugates in multiple animal species. The findings presented here provide a foundation for further study of the metabolism and toxicity of PFP in vivo.
Participated in research design: Han, Szostek, Himmelstein, and Jepson.
Conducted experiments: Han, Szostek, Yang, Cheatham, Mingoia, Nabb, and Gannon.
Contributed new reagents or analytic tools: Szostek and Cheatham.
Performed data analysis: Han, Yang, and Cheatham.
Wrote or contributed to the writing of the manuscript: Han, Szostek, and Himmelstein.
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
- cytochrome P450
- gas chromatography
- 2,3,3,3-tetrafluoropropanoic acid
- liquid chromatography
- mass spectrometry
- perchloric acid
- tandem mass spectrometry
- glutathione transferase.
- Received January 19, 2011.
- Accepted April 14, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics