Nitroso-Imidacloprid Irreversibly Inhibits Rabbit Aldehyde Oxidase

Nov 15, 2007 - The major neonicotinoid insecticide imidacloprid (IMI) is used worldwide for crop protection and pest control on pets. IMI is extensive...
4 downloads 0 Views 95KB Size
Download PDF
1942

Chem. Res. Toxicol. 2007, 20, 1942–1946

Nitroso-Imidacloprid Irreversibly Inhibits Rabbit Aldehyde Oxidase Ryan A. Dick,† David B. Kanne,‡ and John E. Casida* EnVironmental Chemistry and Toxicology Laboratory, Department of EnVironmental Science, Policy and Management, UniVersity of California, Berkeley, California 94720-3112 ReceiVed July 25, 2007

The major neonicotinoid insecticide imidacloprid (IMI) is used worldwide for crop protection and pest control on pets. IMI is extensively metabolized, oxidatively by cytochromes P450 and via aerobic nitroreduction by the molybdo-flavoenzyme aldehyde oxidase (AOX). Rabbit liver AOX is capable of reducing IMI to both its nitrosoguanidine (IMI-NO) and aminoguanidine (IMI-NH2) derivatives; however, when IMI-NO is used as a substrate, less than stoichiometric amounts of IMI-NH2 are detected while IMI-NO is completely consumed. The disappearance of IMI-NO requires both a source of AOX and an AOX-specific electron donor substrate and is not inhibited by the addition of catalase and superoxide dismutase. Experiments to evaluate IMI-NO as a possible time-dependent inactivator of AOX reveal the following four characteristics: First, partially purified AOX (ppAOX) is inactivated at a moderate rate by the electron donor substrate N-methylnicotinamide (NMN); second, AOX is inactivated by IMI-NO in an NMN-dependent manner at a 10-fold greater rate; third, IMI does not inactivate AOX; and finally, GSH protects AOX from inactivation but not to a degree greater than IMI-NO-deficient incubations. Values for the kinetic constants of KI and kinact are measured to be 1.3 mM and 0.35 min-1, respectively. Ultrafiltration is used to establish that IMI-NO inactivation is not reversible and to determine a partition ratio of 1.6. [3H]IMI-NO labeling shows that significant amounts (19%) of this molecule covalently bind to protein following reduction by ppAOX. The addition of 10 mM GSH attenuates this binding almost completely. These findings demonstrate that IMI-NO is metabolically activated by rabbit AOX to form both an irreversible inhibitor and a reactive intermediate that is capable of covalently binding to protein. Introduction 1

Imidacloprid (IMI) started a new era in pest insect control as a high potency nicotinic agonist selective for insect vs mammalian receptors (1, 2). It combines high insecticidal activity with low toxicity to mammals and is currently used worldwide for crop protection and flea and tick control on pets (3). Nitroso-imidacloprid (IMI-NO) (Figure 1) is the first product of the nitroreduction of IMI (4). It is a metabolite in soil (5) and plants (6), including agricultural crops, and is a proposed photodecomposition product (7). Additionally, it is detected in vivo in mice (8) and in vitro with liver preparations from various mammals and poultry (9). IMI-NO as a metabolic intermediate may be a contributing factor in the overall toxicology of IMI (6). Both IMI-NO and the aminoguanidine derivative of imidacloprid (IMI-NH2) are produced via enzymatic reduction of IMI by aldehyde oxidase (AOX), which serves as the neonicotinoid nitroreductase for most nitroguanidine insecticides (4, 9, 10). In a previous study, we sought to measure the kinetic constants for the conversion of IMI to IMI-NO and IMI-NH2 (10). At a concentration of 2 mg of partially purified AOX (ppAOX) per mL, Km and Vmax values of 3 mM and 2.5 nmol/min/mg protein, respectively, were measured for the formation of IMI-NO, while * To whom correspondence should be addressed. Tel: 510-642-5424. Fax: 510-642-6497. E-mail: [email protected]. † Current address: Ardea Biosciences, San Diego, CA. ‡ Current address: Oncologic, Inc., Berkeley, CA 94710-2224. 1 Abbreviations: AOX, aldehyde oxidase; DMAC, p-dimethylaminocinnamaldehyde; IMI, imidacloprid; IMI-NO, nitrosoguanidine metabolite of imidacloprid; IMI-NH2, aminoguanidine metabolite of imidacloprid; KPi, potassium phosphate buffer; NMN, N-methylnicotinamide; ppAOX, partially purified aldehyde oxidase; r, partition ratio; ROS, reactive oxygen species.

IMI-NH2 was not detected. At a 2.5-fold higher ppAOX concentration, both IMI-NO and IMI-NH2 were detected; however, the formation of IMI-NO did not saturate, and a 19fold lower Km was measured for the formation of IMI-NH2. To address this disparity between protein concentration and both metabolic profile and substrate Km values, IMI-NO was subjected to metabolic analysis. It was consumed by ppAOX when N-methylnicotinamide (NMN) was used as an electron donor; however, little IMI-NH2 was produced, and no novel metabolites were detected. The ultimate fate of IMI-NO, including the possibility of metabolic activation by AOX, is the subject of this investigation.

Materials and Methods Chemicals. IMI-NO and IMI-NH2 were obtained by the reduction of IMI as previously described (11). [3H]IMI-NO was prepared from [3H]IMI (12) by reduction with Raney Ni in ethanol (13) and was purified via preparative TLC on silica gel. More specifically, Raney Ni (80 mg) as a 50% slurry in water was washed twice with anhydrous ethanol (300 µL) and then transferred with anhydrous ethanol (2.6 mL) to a vial containing [3H]IMI (20.8 mg, 0.08 mmol; 20 µCi/mmol). A hydrogen balloon was attached with purging, and after 1.5 h, the reaction was stopped (at ∼50% completion) to minimize side products that chromatograph close to the nitroso compound. The mixture was filtered through celite, and the residue following solvent evaporation was chromatographed on a silica preparative TLC plate (2 mm) with 7.5% methanol in dichloromethane, giving good separation of IMI (Rf 0.45) and IMI-NO (Rf 0.34) in about equal amounts. The appropriate region of silica was recovered while still wet and placed in 5% methanol in dichloromethane for 2.5 h. Filtration and evaporation of solvent under N2 gave IMI-NO (20 µCi/mmol) of high (99%) radiopurity based on HPLC.

10.1021/tx700265r CCC: $37.00  2007 American Chemical Society Published on Web 11/15/2007

Nitroso-Imidacloprid Inhibits Aldehyde Oxidase

Chem. Res. Toxicol., Vol. 20, No. 12, 2007 1943 70 °C for 1 h, decolorized with hydrogen peroxide, added to Hionic-Fluor scintillation cocktail (Packard, Meriden, CT), and counted. Background counts were calculated from [3H]IMI-NOdeficient controls.

Figure 1. Structure of IMI-NO. The asterisk denotes the position of the tritium label.

Enzyme Preparation. AOX was partially purified from rabbit liver as previously described (10). Protein concentrations were measured using the bicinchoninic acid reagent (Pierce, Rockford, IL). Active AOX concentrations were measured by the procedures of Rajagopalan (14) and Branzoli and Massey (15). Briefly, the oxidation of 2.5 mM NMN was monitored spectrophotometrically at 300 nm and 25 °C. A turnover number of 755 mol substrate oxidized per min per mol of active site was used as a reference (14). HPLC conditions to monitor IMI-NO and IMI-NH2 were described previously (10). Time-Dependent Inactivation. For AOX inactivation studies, ppAOX (5 mg of protein) was incubated with 4 mM NMN in 1 mL of 100 mM potassium phosphate buffer, pH 7.4 (KPi). Unless otherwise noted, IMI and IMI-NO were dissolved in Me2SO and added to a final concentration of 500 and 400 µM, respectively. At predetermined time points, 100 µL was removed from the assay mix and added to a cuvette containing 1 mL of 50 µM pdimethylaminocinnamaldehyde (DMAC) in KPi (16). Cuvettes were immediately vortexed and placed in a Hewlett-Packard diode array spectrophotometer set in the kinetic mode to monitor the absorbance loss at 400 nm for 3 min. The AOX activity was linear under these conditions. GSH (10 mM) was used as an antioxidant and potential trapping agent and found to not inhibit DMAC oxidation or spontaneously react with IMI-NO under relevant conditions. The percent activity remaining was calculated by comparison of the slope to that obtained with 0 min preincubation samples. Data were fitted to eq 1:

y ) yo + ae-kt

(1)

Half-life values were calculated and used in a Kitz and Wilson plot (17) to determine KI and kinact values (18). Partition Ratio. The partition ratio (r) is the amount of released product (i.e., IMI-NH2) relative to the inhibited enzyme (18). To measure the r of IMI-NO, ppAOX in KPi was centrifuged for 60 min at 100000g and filtered with a 0.22 µm syringe filter to remove particulates that might interfere with the ultrafiltration. The AOX activity levels and protein concentration were reassessed. Varied concentrations (0–600 µM) of IMI-NO were added to a 200 µL KPi solution containing 1 mg of filtered ppAOX and 4 mM NMN. Samples were incubated at 37 °C for 25 min. Ice cold KPi buffer (300 µL) was added, and the samples were concentrated at 4 °C using Microcon YM-30 ultrafiltration devices as per the manufacturer’s instructions (Millipore, Billerica, MA). They were then resuspended in 500 µL of ice cold KPi and subjected to another round of ultrafiltration. This step was repeated once more. Concentrated samples were then resuspended in ice cold KPi and assayed for AOX activity using DMAC and protein concentration using bicinchoninic acid reagent. Activity measurements were normalized using protein concentration and compared to IMI-NOdeficient controls. Evaluation of Covalent Protein Binding. ppAOX (1 mg) was incubated with 160 µM [3H]IMI-NO and 4 mM NMN in 100 mM KPi buffer, pH 7.4, in a final volume of 200 µL at 37 °C, in triplicate. To certain samples, GSH was added to a final concentration of 10 mM. NMN and [3H]IMI-NO-deficient controls were also used. At predetermined time points, protein was precipitated with the addition of 800 µL of ice-cold acetone with centrifugation at 1000g and 4 °C for 10 min. After removal of the supernatant, the protein pellets were washed with 80% methanol. This step involved extensive vortexing and centrifugation with the above conditions and was repeated three times or until the supernatant counts were at background levels. Pellets were then dissolved in 1 N NaOH at

Results Characteristics of IMI-NO Disappearance. The stability of IMI-NO under various conditions was investigated using HPLC. No loss of IMI-NO was detected when incubated in 100 mM KPi, pH 7.4, for 1 h at 37 °C with and without 20 mg/mL bovine serum albumin. Likewise, little or no loss was detected in incubations of IMI-NO and ppAOX without NMN, rabbit liver cytosol without NMN, or NMN alone. However, the addition of NMN to incubations containing either ppAOX or rabbit liver cytosol caused a rapid loss of IMI-NO: A concentration of 43 µM was completely metabolized by ppAOX within 10 min. The addition of catalase and superoxide dismutase did not prevent the loss of IMI-NO. Using an acetone-based extraction procedure, nonstoichiometric amounts of acetone adduct of IMI-NH2 (10) were detected in ppAOX/NMN incubations. Time-Dependent Inhibition of AOX. To assess the timedependent inactivation, ppAOX was incubated with NMN and/ or IMI or IMI-NO at 37 °C. At predetermined time points, aliquots were removed and added to a cuvette containing a DMAC solution. Oxidation of this substrate was monitored spectrophotometrically and compared to controls. The AOX activity was stable throughout the assay when incubated with Me2SO (not shown), 400 µM IMI-NO (Figure 2A), or 500 µM IMI (Figure 2B) alone. However, the addition of 4 mM NMN alone (Figure 2A,B) lead to significant inactivation with a measured t1/2 of 77 min. This is thought to be the result of superoxide production as AOX utilizes molecular oxygen as a terminal electron acceptor. The addition of both IMI-NO and NMN to incubations lead to a drastic decrease in AOX t1/2 to 7 min. On the contrary, incubation with IMI and NMN did not produce significantly different results from NMNonly controls. This result is surprising since IMI-NO is an intermediate in the reduction of IMI to IMI-NH2 and has been detected at high levels in similar conditions (9, 11). Further experiments used GSH as an antioxidant and nucleophilic trapping agent (Figure 2C). The addition of 10 mM GSH to NMN control reactions afforded AOX a 1.8-fold increase in t1/2. This is likely the result of the superoxide quenching capacity of GSH. GSH also protected AOX in incubations of IMI-NO and NMN, however, not to a greater degree than NMN controls. These results suggest that GSH did not protect AOX from IMINO-mediated inactivation. AOX was incubated with NMN and concentrations of IMINO ranging from 0 to 625 µM (Figure 3A). Half-lives were calculated using eq 1 and were used to construct a Kitz and Wilson plot (Figure 3B), from which the following were calculated, KI ) 1.3 mM and Kinact ) 0.35 min-1. These data show that AOX inactivation by IMI-NO was both concentrationdependent and saturable. Partition Ratio. To measure the r and turnover number of AOX inactivation by IMI-NO, ppAOX (5.87 nmol) was preincubated with 0–600 µM IMI-NO for 25 min. Samples were then diluted with cold buffer and subjected to extensive ultrafiltration. They were assayed for activity and protein concentration; the latter was used for normalization. The activity was plotted against the molar ratio of IMI-NO to AOX (Figure 4). From this graph, it is clear that inactivation of AOX is not complete; extrapolation of data fit to eq 1 (solid line) predicts

1944 Chem. Res. Toxicol., Vol. 20, No. 12, 2007

Dick et al.

Figure 3. Saturation of IMI-NO inactivation and measurement of KI and kinact. ppAOX was incubated with 4 mM NMN and various concentrations of IMI-NO. (A) Half-lives were calculated using the data fitted to eq 1. (B) Kitz-Wilson plot of data from panel A to determine KI and kinact. Data points are averages of duplicates.

Figure 2. Time-dependent inhibition by IMI-NO. ppAOX was preincubated with the following substrates for predetermined intervals, and then, an aliquot was transferred to a DMAC solution to monitor the remaining enzyme activity. (A) A 400 µM concentration of IMI-NO alone (1), 4 mM NMN alone (b), or IMI-NO and NMN together (O). (B) A 500 µM concentration of IMI alone (1), NMN alone (b), or IMI and NMN together (O). (C) NMN alone (3), NMN with 10 mM GSH (1), IMI-NO and NMN together (O), or IMI-NO with both NMN and GSH (b). Data are fitted to eq 1, and each point is an average of duplicates.

maximal inactivation of 88%. This phenomenon may be due to product inhibition. Extrapolation of the lower-ratio, linear portion of the graph (dashed line) gives a turnover number of 2.6 and an r of 1.6. This is in accord with the yield of IMINH2 from incubations with IMI-NO used as a substrate (data not shown). IMI-NO Covalent Protein Binding. ppAOX was incubated with [3H]IMI-NO (159 µM) and either NMN (4 mM) or buffer alone. At predetermined time points, ice-cold acetone was added to denature and precipitate the protein. Pellets were obtained by centrifugation and washed extensively with 80% methanol until supernatant counts were at background levels. They were then dissolved, decolorized, and counted. Without the addition of NMN, very little [3H]IMI-NO (0.04 nmol) was bound to protein following a 10 min incubation (Figure 5). However, when NMN was included, 19% of labeled compound (6.2 nmol) was detected in the protein pellet. Levels were slightly lower in 30 min incubations but stable at this level for up to 2 h (data not shown); this decrease may be due to partial adduct instability. The addition of 10 mM GSH to 10 min incubations with NMN drastically decreased the amount of [3H]IMI-NO bound to 0.14 nmol, suggesting that an electrophilic intermediate is involved.

Figure 4. Partition ratio (r) for IMI-NO. ppAOX was incubated with varied molar ratios of IMI-NO (with 4 mM NMN) and then subjected to extensive ultrafiltration. The percent activity remaining was determined using DMAC, and data were normalized by protein concentration. Data are fitted to eq 1 (solid line), and extrapolation of the linear portion of the graph (dashed line) gives the r value. Data points are averages of duplicates.

Discussion The molybdo-flavoenzyme AOX (19) was recently identified as the neonicotinoid nitroreductase capable of reducing several neonicotinoids to nitroso- and aminoguanidine metabolites (9, 10). The reduction of IMI by rabbit liver AOX has been extensively characterized (4, 9, 10). When incubated with low concentrations of ppAOX (2 mg/mL), IMI-NO was the only metabolite detected giving Km and Vmax values of 3 mM and 2.5 nmol/min/mg protein, respectively. However, at 2.5-fold higher ppAOX concentrations, IMI-NH2 was the major metabolite formed at substrate concentrations less than 2 mM, with Km and Vmax values of 0.16 mM and 1.2 nmol/min/mg protein measured for its formation, respectively. This dependence of metabolic profile on AOX concentration was also observed when rabbit liver cytosol was used as the enzyme source. Such a large disparity in Km values for the formation of each metabolite is puzzling and could be the result of different mechanisms of formation, possibly with free IMI-NO serving as a direct

Nitroso-Imidacloprid Inhibits Aldehyde Oxidase

Figure 5. Covalent binding of [3H]IMI-NO. Minor amounts of radioactivity were detected in the protein pellet of incubations with ppAOX and [3H]IMI-NO alone following precipitation and exhaustive washing. However, when NMN was included in 10 min incubations with 32 nmol of [3H]IMI-NO, 6.2 nmol (19% of total) of label became covalently bound. The addition of GSH to complete 10 min reactions almost completely blocked binding. Incubations of complete reactions for 30 min suggest that covalent binding proceeded to completion by 10 min and may be slightly unstable.

precursor for IMI-NH2 formation. To investigate this scenario, IMI-NO was subjected to metabolic analysis. When IMI was incubated with ppAOX and electron donor (NMN), IMI-NH2 was detected; however, concentrations were far less than expected for the complete consumption of IMI-NO observed. Additionally, no other IMI metabolites were found by LC/MS/ MS or HPLC with diode array detection. This IMI-NO consumption appeared to be dependent on the presence of both an AOX source and an electron donor and not mediated by the reactive oxygen species (ROS) generated through the reduction of molecular oxygen by AOX (19, 20). The possibility that IMI-NO was metabolically activated by AOX was investigated next. In this scenario, IMI-NO disappearance could be accounted for by irreversible, possibly covalent, binding to AOX and/or other proteins of the ppAOX mixture. Using a dilution assay and DMAC, a colorimetric oxidative substrate, AOX was found to be inactivated by IMINO in a time-dependent and saturable manner. It was also found that AOX was unstable in incubations with NMN alone, however, to a much lesser degree than with IMI-NO and NMN. The NMN-based instability is likely to be mediated by the ROS produced. Interestingly, IMI with NMN did not inactivate AOX at a rate greater than NMN alone. Because IMI-NO is known to be produced under these conditions, one could envision IMINO exiting and reentering the active site to inactivate the enzyme. Perhaps IMI has a much greater affinity for AOX, and this is a matter of competition. Coincubation of IMI-NO and ppAOX with GSH afforded some protection from inactivation; however, it was to the same degree as observed with NMN alone. GSH is known to have antioxidative chemical properties (21). It was previously determined that relevant concentrations of GSH did not inhibit AOX activity nor affect IMI-NO chemical stability. By varying IMI-NO preincubation conditions, a KI of 1.3 mM and kinact of 0.35 min-1 were measured. While this KI is likely to be higher than normal exposures, the kinact is on par with such well-known cytochromes P450 inhibitors as ritonavir and mibefradil (22, 23). As such, it is unknown if IMINO could affect the in vivo metabolism of xenobiotics and endogenous compounds by AOX. Using exhaustive ultrafiltration and varying molar ratios of IMI-NO to AOX, a partition ratio of 1.6 was measured and AOX inactivation was confirmed

Chem. Res. Toxicol., Vol. 20, No. 12, 2007 1945

to be irreversible. This r is consistent with the amount of IMINH2 detected in incubations with ppAOX and IMI-NO. These experiments demonstrate that IMI-NO is an inactivator of rabbit liver AOX and that this process is time-dependent, saturable, and irreversible (18). Additionally, it requires an electron donor cofactor, and thus a catalytic step, and demonstrates a very low r. The nature of this reaction was examined using [3H]IMINO. This compound was incubated with ppAOX, both with and without NMN. Protein was subsequently precipitated with organic solvent, washed exhaustively, and assayed for tritium incorporation. About 19% of the 32 nmol of [3H]IMI-NO was found to be irreversibly bound to protein when NMN was included. However, when 10 mM GSH was added, irreversible binding dropped to nearly background levels. This result was perplexing because previous experiments showed that GSH did not augment AOX protection in AOX time-dependent stability assays. Perhaps AOX inactivation and covalent binding are distinct chemical events. This would be the case if, for instance, through its metabolic activation, IMI-NO is split with the unlabeled portion of the molecule serving to inactivate AOX and the labeled portion binding to proteins in the ppAOX mixture. The complexity of the structure and function of this molybdo-flavoenzyme and its poorly-defined physiological role (19) currently limit an understanding of the mechanism and toxicological relevance of IMI-NO inactivation of AOX. Acknowledgment. We thank Richard Staub, Thomas Kensler, Jennie Chin, and Gary Quistad for advice and assistance. The project was supported by Grant RO1 ES008424 from the National Institute of Environmental Health Sciences (NIEHS), NIH. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIEHS, NIH.

References (1) Casida, J. E., and Quistad, G. B. (1998) Golden age of insecticide research: Past, present, or future? Annu. ReV. Entomol. 43, 1–16. (2) Yamamoto, I., and Casida, J. E., Eds. (1999) Nicotinoid Insecticides and the Nicotinic Acetylcholine Receptor, p 300, Springer-Verlag, Tokyo. (3) Tomizawa, M., and Casida, J. E. (2005) Neonicotinoid insecticide toxicology: mechanisms of selective action. Annu. ReV. Pharmacol. Toxicol. 45, 247–268. (4) Schulz-Jander, D. A., Leimkuehler, W. M., and Casida, J. E. (2002) Neonicotinoid insecticides: Reduction and cleavage of imidacloprid nitroimine substituent by liver microsomal and cytosolic enzymes. Chem Res. Toxicol. 15, 1158–1165. (5) Krohn, J., and Hellpointer, E. (2002) Environmental fate of imidacloprid. Pflanzenschutz-Nachrichten Bayer 55, 3–26. (6) Solecki, R. (2001) Toxicological evaluations: imidacloprid. www. inchem.org/documents/jmpr/jmpmono/2001pr07.htm. (7) Scholz, K., and Reinhard, F. (1999) Photolysis of imidacloprid (NTN 33893) on the leaf surface of tomato plants. Pestic. Sci. 55, 633–654. (8) Ford, K. A., and Casida, J. E. (2006) Chloropyridinyl neonicotinoid insecticides: Diverse molecular substituents contribute to facile metabolism in mice. Chem. Res. Toxicol. 19, 944–951. (9) Dick, R. A., Kanne, D. B., and Casida, J. E. (2005) Identification of aldehyde oxidase as the neonicotinoid nitroreductase. Chem. Res. Toxicol. 18, 317–323. (10) Dick, R. A., Kanne, D. B., and Casida, J. E. (2006) Substrate specificity of rabbit aldehyde oxidase for nitroguanidine and nitromethylene neonicotinoid insecticides. Chem. Res. Toxicol. 19, 38–43. (11) Kanne, D. B., Dick, R. A., Tomizawa, M., and Casida, J. E. (2005) Neonicotinoid nitroguanidine insecticide metabolites: Synthesis and nicotinic receptor potency of guanidines, aminoguanidines, and their derivatives. Chem. Res. Toxicol. 18, 1479–1484. (12) Latli, B., Than, C., Morimoto, H., Williams, P. G., and Casida, J. E. (1996) [6-chloro-3-pyridylmethyl-3H]Neonicotinoids as high-affinity radioligands for the nicotinic acetylcholine receptor: Preparation using NaB3H4 and LiB3H4. J. Labelled Compd. Radiopharm. 38, 971–978.

1946 Chem. Res. Toxicol., Vol. 20, No. 12, 2007 (13) Novák, L., Hornyánszky, G., Király, I., Rohály, J., Kolonits, P., and Szántay, C. (2001) Preparation of new imidacloprid analogues. Heterocycles 55, 45–58. (14) Rajagopalan, K. V. (1980) Xanthine oxidase and aldehyde oxidase. In Enzymatic Basis of Detoxication (Jakoby, W. B., Ed.) Vol. 1, pp 295–309, Academic Press, Orlando, FL. (15) Branzoli, U., and Massey, V. (1974) Evidence for an active site persulfide residue in rabbit liver aldehyde oxidase. J. Biol. Chem. 249, 4346–4349. (16) Maia, L., and Mira, L. (2002) Xanthine oxidase and aldehyde oxidase: A simple procedure for the simultaneous purification from rat liver. Arch. Biochem. Biophys. 400, 48–53. (17) Kitz, R., and Wilson, I. B. (1962) Esters of methanesulfonic acid as irreversible inhibitors of acetylcholinesterase. J. Biol. Chem. 237, 3245–3249. (18) Silverman, R. B. (1988) Mechanism-based enzyme inactivation. Chemistry and Enzymology, p 288, CRC Press, Inc., Boca Raton, FL. (19) Garattini, E., Mendel, R., Romão, M. J., Wright, R., and Terao, M. (2003) Mammalian molybdo-flavoenzymes, an expanding family of

Dick et al.

(20)

(21)

(22)

(23)

proteins: Structure, genetics, regulation, function and pathophysiology. Biochem. J. 372, 15–32. Badwey, J. A., Robinson, J. M., Karnovsky, M. J., and Karnovsky, M. L. (1981) Superoxide production by an unusual aldehyde oxidase in guinea pig granulocytes. J. Biol. Chem. 256, 3479–3486. Wefers, H., and Seis, H. (1983) Oxidation of glutathione by the superoxide radical to the disulfide and the sulfonate yielding singlet oxygen. Eur. J. Biochem. 137, 29–36. Ernest, C. S., II, Hall, S. D., and Jones, D. R. (2005) Mechanismbased inactivation of CYP3A4 by HIV protease inhibitors. J. Pharmacol. Exp. Ther. 312, 583–591. Prueksaritanont, T., Ma, B., Tang, C., Meng, Y., Assang, C., Lu, P., Reider, P. J., Lin, J. H., and Baillie, T. A. (1999) Metabolic interactions between mibefradil and HMG-CoA reductase inhibitors: An in vitro investigation with human liver preparations. Br. J. Clin. Pharmacol. 47, 291–298.

TX700265R