Substrate Specificity of Rabbit Aldehyde Oxidase for Nitroguanidine

Unique and Common Metabolites of Thiamethoxam, Clothianidin, and Dinotefuran in Mice. Kevin A. Ford and John E. Casida. Chemical Research in Toxicolog...
0 downloads 0 Views 186KB Size
Download PDF
38

Chem. Res. Toxicol. 2006, 19, 38-43

Substrate Specificity of Rabbit Aldehyde Oxidase for Nitroguanidine and Nitromethylene Neonicotinoid Insecticides 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 August 18, 2005

The nitroguanidine or nitromethylene moiety of the newest major class of insecticides, the neonicotinoids, is important for potency at insect nicotinic receptors and selectivity relative to mammalian receptors. Aldehyde oxidase (AOX) was recently identified as the imidacloprid nitroreductase of mammalian liver, producing both nitrosoguanidine and aminoguanidine metabolites. The present study considers the ability of AOX, partially purified from rabbit liver, to reduce five commercial nitroguanidine (i.e., imidacloprid, thiamethoxam, clothianidin, and dinotefuran) and nitromethylene (i.e., nitenpyram) neonicotinoid insecticides and three derivatives thereof (i.e., the N-methyl and nitromethylene analogues of imidacloprid and desmethylthiamethoxam). LC/MS/MS was used to demonstrate that AOX reduces nitroguanidines to both nitroso- and aminoguanidines, while nitromethylenes are reduced only to the corresponding nitroso metabolites. Additionally, nitrosonitenpyram was found to spontaneously dehydrate to form a 2-cyanoamidine metabolite, mimicking a predominant photoreaction. The substrate specificity of AOX was characterized as follows: Neonicotinoids with a tertiary nitrogen (N-methylimidacloprid and thiamethoxam) are poor substrates; nitroguanidines are metabolized faster than nitromethylenes; and clothianidin is the most rapidly reduced. Kinetic constants were measured for reduction of three nitroguanidines at two concentrations of AOX. At 2 mg protein/mL, only nitroso metabolites were detected, with Km values of 1.03, 2.99, and 2.41 mM and Vmax values of 5.13, 2.54, and 0.98 nmol/min/mg protein measured for clothianidin, imidacloprid, and dinotefuran, respectively. At 5 mg protein/mL, both amino and nitroso metabolites were detected. However, with each nitroguanidine, the formation of nitroso metabolites did not saturate at substrate levels up to 4 mM, whereas amino metabolite formation exhibited Km values of 0.052, 0.16, and 0.084 mM with corresponding Vmax values of 0.80, 1.24, and 0.79 nmol/min/mg protein for clothianidin, imidacloprid, and dinotefuran, respectively. These in vitro observations show large structural differences in the rates of AOX-catalyzed reduction and help to interpret the extensive studies on in vivo metabolism of neonicotinoid insecticides. Introduction The control of insect pests on crops and pets is increasingly dependent on a new generation of insecticides, the neonicotinoids (1-3). These compounds contain an electronegative pharmacophore that imparts binding potency for insect nicotinic acetylcholine receptors (nAChR)1 and selectivity when compared to mammals. The commercially relevant nitroneonicotinoids include four nitroguanidines, i.e., imidacloprid (IMI), thiamethoxam (THI), clothianidin (CLO), and dinotefuran (DIN), and one nitromethylene, i.e., nitenpyram (NIT) (2-4) (Figure 1). Additional nitroneonicotinoids of interest include the high potency nitromethylene analogue of IMI (chIMI) (2, 3, 5), the proinsecticide N-methyl-IMI (nmIMI) (5), and the THI metabolite desmethylthiamethoxam (dmTHI) (6, 7) (Figure 1). * To whom correspondence should be addressed. Tel: 510-642-5424. Fax: 510-642-6497. E-mail: [email protected]. 1 Abbreviations: ACN, acetonitrile; AOX, aldehyde oxidase; chIMI, nitromethylene analogue of IMI; CLO, clothianidin; DAD, diode array detection; DIN, dinotefuran; dmTHI, desmethyl THI; EtOAc, ethyl acetate; IMI, imidacloprid; nAChR, nicotinic acetylcholine receptor; -NdC(CH3)2, acetone imine adduct of aminoguanidine metabolite; -NdCH2, formaldehyde imine adduct of aminoguanidine metabolite; -NH2, amino metabolite; NIT, nitenpyram; NIT-CN, 2-cyanoamidine metabolite of NIT; nmIMI, N-methyl IMI; NMN, N-methylnicotinamide; -NO, nitroso metabolite; ppAOX, partially purified AOX; THI, thiamethoxam; tR, retention time.

Figure 1. Nitroguanidine and nitromethylene neonicotinoids showing common names or designations and abbreviations.

10.1021/tx050230x CCC: $33.50 © 2006 American Chemical Society Published on Web 12/15/2005

Substrate Specificity of Rabbit Aldehyde Oxidase

The neonicotinoids as a whole are the only major new class of insecticides of the last three decades and have a generally favorable safety profile in mammals (7). Nitroreduction is a major metabolic pathway for IMI in mammals (8, 9). Using IMI as the prototype, the “neonicotinoid nitroreductase” of mammalian liver (10) was recently identified as the molybdoflavo-enzyme aldehyde oxidase (AOX) (EC 1.2.3.1) (11). This enzyme reduces IMI to nitroso (IMI-NO) and amino (IMI-NH2) metabolites when a specific electron donor [such as N-methylnicotinamide (NMN), benzaldehyde, or 2-hydroxypyrimidine] is present (11). Cytochrome P450s are also capable of reducing IMI in vitro (12). While the electronegative nitroneonicotinoids display excellent selectivity for insect vs mammalian nAChRs, their positively charged aminoguanidine and guanidine metabolites demonstrate the opposite selectivity (3). Thus, nitroreduction is thought to be an activation event in terms of mammalian neurotoxicity. The substrate specificity of AOX for neonicotinoids is not known but may play a crucial role in their efficacy, toxicology, and persistence. In this study, AOX partially purified from rabbit liver (ppAOX) is used to generate neonicotinoid metabolites by coincubation with NMN, which was used as an electron donor substrate. The parent compounds and metabolites are analyzed by HPLC with diode array detection (DAD) and identified by LC/MS/MS. Kinetic analysis of nitroreduction as loss of the parent compound is made for all eight neonicotinoids, while IMI, CLO, and DIN are used to measure MichaelisMenten constants for metabolite formation. These are the first steps in defining the neonicotinoid substrate specificity of AOX.

Materials and Methods AOX and Neonicotinoid Derivatives. ppAOX was prepared from rabbit liver cytosol by heat treatment (60 °C for 10 min), ammonium sulfate (50% saturation) precipitation, and dialysis (11). The bicinchoninic acid reagent (Pierce, Rockford, IL) was used according to the manufacturer’s directions to determine protein concentrations. Sources of the nitroguanidine and nitromethylene neonicotinoids and protocols for synthesis of relevant nitroso and amino derivatives thereof were described previously (10, 11, 13). N-(6-Chloropyridin-3-ylmethyl)-N-ethyl-N′-methyl-2-cyanoamidine (NIT-CN). As detailed later, NIT metabolism gave a major product with LC/MS/MS fragmentation consistent with the NITCN structure. This metabolic conversion would be analogous to the reported photochemical degradation of nithiazine (the original nitromethylene neonicotinoid) to 2-cyanothioimidate (14; Scheme 1). The conditions used for this nithiazine photochemical conversion were applied to NIT. More specifically, NIT (8.9 mg) in CH2Cl2 (2.1 mL) was irradiated at 300 nm for 50 min. The residue from solvent evaporation was chromatographed on silica gel developed with 5% MeOH in CH2Cl2 to give a mixture (2 mg) of NIT-CN and NIT as major components. NIT-CN: LC/MS/MS: m/z 237 [M + H]+, 210 [M - CN]+, 126, 85. HR-EI-MS: calcd for C11H13N4Cl [M+], 236.08285; found, 236.08416. IR (on a 3 M disposable IR card of thin microporous polyethylene): 2224 cm-1 weak (cyano). Metabolite Identification. Metabolites of nitroneonicotinoids were generated and identified using previously described methods (10, 11). Briefly, 1 mg of ppAOX was incubated in 100 mM Scheme 1. Photodecomposition of Nithiazine

Chem. Res. Toxicol., Vol. 19, No. 1, 2006 39 potassium phosphate buffer (pH 7.5) with 0.5 mM neonicotinoid and 10 mM NMN for 120 min at 37 °C. NMN-deficient reactions were used as negative controls. Protein was then precipitated with 2 volumes of cold acetonitrile (ACN), followed by centrifugation at 20000g at 4 °C for 10 min. Supernatants were evaporated and desalted using Oasis HLB columns (Waters, Milford, MA). Samples were then injected into a Finnigan TSQ 700 electrospray mass spectrometer (Thermo Finnigan, San Jose, CA), which was set to operate in full scan mode (200-500 amu) (11). By comparison of NMN-containing and deficient reactions, possible AOX metabolites were identified. LC/MS/MS was then used for confirmation of molecular structure (11); molecular masses of purported metabolites were used to scan for daughter ions. These data were also compared to those of authentic standards, when available. Kinetics of Neonicotinoid Reduction. Half-lives (t1/2) of the neonicotinoids incubated with ppAOX were determined. Neonicotinoids were added (50 µM final concentration) to 100 mM potassium phosphate buffer (pH 7.4) containing 1 mg of ppAOX. Samples were then prewarmed in a 37 °C water bath for 3 min. Reactions were started with the addition of NMN (final concentration 10 mM) to bring the volume to 200 µL. Reactions were stopped at predetermined time points by the addition of 400 µL of ice-cold ethyl acetate (EtOAc) containing 2 nmol of thiacloprid (a cyanoguanidine neonicotinoid used here as an internal standard). Samples were then vortexed, left on ice for 10 min, and centrifuged for 5 min at 20000g. The organic layer was removed, and the aqueous phase was extracted again with EtOAc (no internal standard). The organic layers were pooled, evaporated, resuspended in ACN:0.1% trifluoroacetic acid (25:75), and injected into a 1050 HPLC-DAD (Agilent, Palo Alto, CA) for analysis. A Luna C18 5 µm, 250 mm × 4.6 mm column (Phenomenex, Torrance, CA) was used with a flow rate of 1 mL/min and mobile phases A (0.1% trifluoroacetic acid) and B (ACN). The following isocratic conditions were used with the corresponding neonicotinoids: 27% B for DIN and NIT; 30% B for THI; 35% B for IMI, dmTHI, and CLO; and 37% B for nmIMI. Parental neonicotinoid peak areas were quantified at 254 nm and compared to those of NMN-deficient controls. The t1/2 value was calculated by fitting data to eq 1: y ) yo + ae-kt. Kinetic Constant Measurement. Two different protein concentrations (2 and 5 mg/mL) were used to calculate Vmax and Km values for IMI, CLO, and DIN metabolite formation. To 100 mM potassium phosphate buffer (pH 7.4), ppAOX and varied concentrations of neonicotinoid were added. Reactions were prewarmed for 3 min at 37 °C and then started with the addition of a saturating concentration (10 mM) of NMN to give a final volume of 200 µL. Incubations containing 2 and 5 mg protein/mL were stopped after 15 and 20 min, respectively, by the addition of 600 µL of ice-cold acetone containing 5 nmol of dmTHI (internal standard). The addition of acetone not only served to precipitate protein but also efficiently (>95%) derivatized aminoguanidine neonicotinoid metabolites (Figure 2). These derivatives displayed higher extinction coefficients and longer tR values and, thus, were more amenable for analysis by HPLC-DAD. Samples were then incubated on ice for 10 min and centrifuged at 20000g for 5 min, and their supernatants were evaporated. Residues were then resuspended in ACN:0.1% trifluoroacetic acid (25:75), filtered through a 0.22 µm filter, and analyzed using the same HPLC instrument and conditions as above but with the following changes. For IMI, CLO, and their metabolites, a 25% B isocratic gradient was used and gave the following tR values: IMI-NdC(CH3)2, 4.7 min; IMI-NO, 5.7 min; IMI, 10.8 min; CLO-NO, 4.6 min; CLO-NdC(CH3)2, 5.5 min; CLO, 9.7 min; and dmTHI, 13.7 min. A previously described gradient (11) was used for DIN, giving tR values of: DIN-NO, 7.3 min; DIN-NdC(CH3)2, 10.1 min; DIN, 10.4 min; and dmTHI, 19.8 min. Authentic standards of each of the metabolites or derivatives were used to generate standard curves at 254 nm. Vmax and Km constants were calculated using nonlinear regression analysis by Hyper.exe.

40 Chem. Res. Toxicol., Vol. 19, No. 1, 2006

Dick et al. Table 1. LC/MS/MS Characteristics of Neonicotinoid Insecticide AOX-Generated Metabolites and Derivatives Thereofa substrates, m/z metabolites, tR + and derivatives (min) [M + H] [M + H + ACN]+ daughter ionsb chloropyridinyls-nitroguanidines 24.0 256 297

IMI IMI-NOd

22.0

240

281

IMI-NH2

11.2

226

267

IMI-NdCH2d

17.5

238

279

chIMI

chloropyridinyls-nitromethylenes 22.2 255 296

chIMI-NO NIT

11.1 19.0

239 271

280 312

NIT-NO NIT-CNd

16.4 23.2

255 237

296 278

THI

chlorothiazolyls-nitroguanidines 22.3 292 333

THI-NdCH2 dmTHI

19.2 25.2

274 278

315 319

dmTHI-NO dmTHI-NH2

22.2 21.5

262 248

303 289

CLO

24.1

250

291

CLO-NOd CLO-NH2d

19.1 9.1

234 220

275 261

DIN

20.1

DIN-NOd DIN-NH2d

17.3 2.6

d

Figure 2. Reactions involved in AOX-catalyzed reduction of a representative nitroguanidine (CLO) to CLO-NO and CLO-NH2 metabolites and derivatization of CLO-NH2 with acetone for analysis. Generally, the nitroso metabolite may serve as the direct precursor for amino metabolites.

Results Metabolite Identification. The neonicotinoids were incubated with either ppAOX and NMN or ppAOX alone. No degradation products were detected by LC/MS from incubations lacking NMN. Peaks apparent in LC/MS chromatograms of incubations containing NMN, but not in NMN-deficient controls, were assumed to be generated through reduction by AOX. Both parent [M + H]+ and ACN adduct [M + H + ACN]+ ions were detected in duplicate or triplicate experiments. Fragmentation patterns obtained using LC/MS/MS were used to confirm the molecular structures of metabolites. Data were then compared to those of synthetic standards as specified. Because nmIMI was not a substrate for AOX under the conditions used, it was not examined extensively. The following fragmentation patterns were helpful in defining molecular structures: loss of Cl gave -35 m/z; loss of NO2 gave -46 m/z; loss of both gave -81 m/z; chlorothiazoles produced a daughter ion of 132 m/z; and chloropyridinyls gave 126 m/z. Further interpretations of fragmentation patterns are given in the Supporting Information. The nitroguanidine IMI was reduced to both IMI-NO and IMI-NH2 (Table 1) in accord with our earlier study (11). Formaldehyde, here a contaminant of unknown origin, was found to readily adduct with the hydrazine-like moiety of IMINH2 to form IMI-NdCH2 (13). Amino (-NH2) and nitroso (-NO) metabolites were observed for all other nitroguanidines (illustrated with CLO in Figure 2) except for THI, for which only trace amounts of THI-NdCH2 were detected. Only two-electron reductions of the nitromethylene neonicotinoids by AOX, to produce nitroso metabolites (chIMI-NO and NIT-NO), were observed (Table 1). With the more extensively metabolized NIT, an additional major product was detected with a mass [M + H]+ of 237 m/z, i.e., a loss of 18 amu from NIT-NO. A unique daughter ion of 210 m/z from the 237 m/z compound indicates the loss of 26 amu and corresponds to cleavage and loss of a cyanide ion. This metabolite is therefore proposed to be 2-cyanoamidine NIT-CN and rationalized to be formed via the loss of water from NIT-NO (Figure 3). No NIT-CN was detected in AOX incubations that lacked NMN. Nithiazine is known to form a 2-cyanothioimidate as a major photoproduct (14; Scheme 1). The 237 m/z metabolite of NIT was also found to be a major photoproduct of this compound (identical LC/MS/MS spectra for the metabolite and

tefuryls-nitroguanidines 203 244 187 173

228 214

209, 175, 128, 84c 209, 175, 128, 126, 84c 190, 168, 126, 100, 99c 209, 175, 126, 84 221, 208, 174, 126 126 225, 210, 196, 169, 126, 99 171, 126, 85 210, 126, 85 246, 211, 152, 132 161, 132 232, 197, 167, 132 197, 175, 132 207, 177, 132, 119 169, 142, 132, 113 169, 132, 113 132, 113 157, 129, 127, 114, 100 114, 100, 83 83

a See Materials and Methods for metabolite identification. b Base peak in bold type. See Supporting Information for proposed structures. c Data from ref 10. d Structure confirmed by LC/MS comparison of tR [M + H]+ and [M + H + ACN]+ with photoproduct (NIT-CN, see Methods and Materials) or standard from synthesis (13).

Figure 3. Proposed mechanism of AOX-catalyzed or photochemical conversion of NIT to NIT-CN.

photoproduct). Further analytical data on the photoproduct (see Materials and Methods) support the assignment of the NIT-CN structure for both the photoproduct and the metabolite. Kinetics of Neonicotinoid Reduction. Reduction of the nitroneonicotinoids by AOX was characterized kinetically by monitoring the loss of parent compound with HPLC-DAD (Figure 4). They were incubated at 50 µM with ppAOX (1 mg protein/mL) and a saturating amount of NMN (10 mM). The percentage of parent compound remaining was measured using

Substrate Specificity of Rabbit Aldehyde Oxidase

Chem. Res. Toxicol., Vol. 19, No. 1, 2006 41

Figure 4. Kinetics of AOX-catalyzed neonicotinoid reduction. Neonicotinoids (50 µM) were incubated with 1 mg ppAOX/mL and NMN (10 mM). The percent parent remaining was quantified using HPLC-DAD and NMN-deficient controls. Data from duplicates in the same experiment.

Figure 5. Kinetics of AOX-catalyzed nitroso- and aminoguanidine formation. Neonicotinoids incubated with 2 mg ppAOX/mL for 15 min or 5 mg ppAOX/mL for 20 min and a saturating concentration of NMN (10 mM). Kinetic constants are given in Table 2 based on the data in this figure with single determinations at 7-10 different concentrations.

NMN-deficient controls. The t1/2 values were calculated with eq 1. Of the eight compounds tested, nmIMI, THI, and chIMI did not reach their half-lives within the time of the assay. Comparison of the chloropyridinyl compounds shows that IMI (t1/2 ) 15 min) is reduced at more than eight times the rate of its nitromethylene (chIMI) and N-methyl (nmIMI) analogues (t1/2 > 120 min). Of the nitromethylenes, the acyclic NIT (t1/2 ) 31 min) is reduced considerably faster than its cyclic analogue, chIMI (t1/2 > 120 min). The retarding effect of the tertiary nitrogen, i.e., IMI vs nmIMI, is repeated in the chlorothiazolyl neonicotinioids, i.e., dmTHI (t1/2 ) 8.5 min) and THI (t1/2 > 120 min). The chlorothiazolyl substituent of CLO is preferred over the tefuryl moiety of DIN as evidenced by a more than 2-fold decrease in half-life. Of the neonicotinoids examined, CLO is reduced the fastest (t1/2 ) 4.7 min) followed by dmTHI, indicating a steric effect of the bulky oxadiazinane ring. Kinetics of Nitroso- and Aminoguanidine Metabolite Formation. Kinetic constants (Km and Vmax) were measured for reduction of IMI, CLO, and DIN by ppAOX to nitroso- and aminoguanidine metabolites. In each experiment, a saturating concentration of NMN was used. Reactions were stopped in the linear range of metabolite formation by the addition of

acetone. This solvent served not only to denature and precipitate protein but also to derivatize aminoguanidine metabolites for analysis (Figure 2). Nitrosoguanidine metabolites of IMI, CLO, and DIN were easily detected with ppAOX at 2 mg protein/mL, while derivatized aminoguanidines were not (Figure 5). Saturation of nitroso metabolite formation occurred at low millimolar concentrations, and nonlinear regression analysis gave the following values for IMI, CLO, and DIN, respectively: Km, 2.99, 1.03, and 2.41 mM; Vmax, 2.54, 5.13, and 0.98 nmol/min/mg protein (Table 2). CLO displays both the lowest Km and the highest Vmax under these conditions (11) (Figure 5). Aminoguanidine metabolites were detected at 5 mg ppAOX protein/mL (11) (Figure 5). Saturation of aminoguanidine formation was reached at much lower substrate concentrations, and similar analysis of data gave the following values for IMI, CLO, and DIN, respectively: Km, 0.16, 0.052, and 0.084 mM; Vmax, 1.24, 0.80, and 0.79 nmol/min/mg protein. Interestingly, the rate of aminoguanidine formation decreased significantly after reaching saturation, an indication of possible substrate inhibition. At this higher protein level, nitroso metabolite formation was linear with substrate concentrations up to 4 mM. Rates of nitroso formation at 5 mg protein/mL were also

42 Chem. Res. Toxicol., Vol. 19, No. 1, 2006

Dick et al.

Table 2. Kinetic Constants for Nitroso- and Aminoguanidine Formation from Three Neonicotinoids neonicotinoid IMI CLO DIN IMI CLO DIN

Km (mM)a

Vmax (nmol/min/ mg protein)a

nitrosoguanidine formationb 2.99 2.54 1.03 5.13 2.41 0.98 aminoguanidine formationc 0.16 1.24 0.052 0.80 0.084 0.79

Vmax/Km 0.85 4.98 0.41 7.8 15.4 9.4

a Calculated from data in Figure 5. b A 2 mg ppAOX/mL amount, 15 min of incubation, which yields nitrosoguanidines but not aminoguanidines. c A 5 mg ppAOX/mL amount, 20 min of incubation, which yields both nitroso- and aminoguanidines.

consistently lower (on the order of 2-fold) than in the 2 mg protein/mL experiments.

Discussion An enzyme that catalyzes the nitroreduction of IMI, first designated the “neonicotinoid nitroreductase” (10), was subsequently identified as AOX (11), again using only IMI as a prototypical nitroneonicotinoid. The present study investigates the metabolism of a set of eight nitroneonicotinoids, consisting of commercial compounds and derivatives, establishing that AOX is truly a neonicotinoid, and not merely IMI, nitroreductase. AOX is a homodimeric enzyme with a molecular mass of approximately 300 kDa (15). Each subunit contains one molybdopterin cofactor, two 2Fe/2S clusters, and one FAD molecule (15). Oxidations of aldehydes and charged and neutral azaheterocycles occur in the molybdenum cofactor active site (16). Electrons are then passed through the iron-sulfur clusters to the flavin, which catalyzes electron transfer to molecular oxygen, the terminal electron acceptor (17). However, a wide variety of xenobiotics, including sulfoxides (18), N-oxides (19), nitrosamines (20), azo dyes (21), isoxazoles (22), isothiazoles (22), and nitro compounds (23), are capable of serving as surrogate electron acceptors. This study shows that AOX reduces nitroguanidine neonicotinoids by two electrons to form nitroso metabolites and by six electrons to form aminoguanidines. The hydrazine-like functional group of the latter class of metabolites reacts readily with aldehydes, as demonstrated by formaldehyde and acetaldehyde adducts detected in this and previous studies (10, 11), and ketones. This propensity was exploited by using high concentrations of acetone to precipitate protein and derivatize analytes in kinetic investigations. In contrast with the nitroguanidines, the nitromethylenes were reduced by AOX only to nitroso metabolites. This might result from a difference in chemical characteristics between C- and N-nitro compounds or intermediates as they pertain to AOX binding or catalysis. It is also possible that, as demonstrated with NIT, the nitrosomethylene readily dehydrates via its tautomer to the 2-cyanoamidine before the reduction of nitroso can take place (see Supporting Information). An analogous reaction was not observed with chIMI possibly because the low level of reduction would yield an amount of chIMI-CN below the limit of detection in this study. The substrate specificity of ppAOX was investigated through the comparison of nitroneonicotinoid metabolic stability. It appears that AOX strongly prefers nitroguanidine substrates over nitromethylenes and, additionally, acyclic neonicotinoids over cyclics (NIT vs chIMI) and chlorothiazolyls over tefuryls (CLO

vs DIN). Interestingly, compounds with a tertiary (methylated) nitrogen ceased to be suitable substrates (nmIMI vs IMI and THI vs dmTHI or CLO). Perhaps this moiety requires an enzymatic hydrogen bond donor. Interestingly, this characteristic also translates to binding potency for nAChRs and, as such, dealkylation at this position is considered an activation event (3); thus, THI is considered a proinsecticide. The bulkiness of the oxadiazinane ring of dmTHI seemed to be tolerated, but not as well as the acyclic configuration of CLO. Of the neonicotinoids investigated, CLO was reduced the fastest, presumably because it contains the optimal mixture of beneficial characteristics (i.e., nitroguanidine, acyclic, chlorothiazolyl, and secondary methylated nitrogen). These structure-activity relationships for neonicotinoid metabolism are generally supported by findings on their activation and detoxification in a coupled nicotinic receptor-AOX system (24). Because synthetic standards were available (13), rates of IMI, CLO, and DIN metabolite formation were investigated. With 2 mg ppAOX protein/mL, only nitroso metabolites were detected for IMI, CLO, and DIN, and each gave Km values in excess of 1 mM. In accord with the metabolic stability assay, CLO again proved to be the best AOX substrate in terms of Vmax and Km, while IMI and DIN were roughly equivalent. With guidance from a previous study involving rabbit liver cytosol (11), it was assumed that an increase in ppAOX concentration would allow for the detection of aminoguanidine metabolites. Indeed, with 5 mg ppAOX/mL, aminoguanidines were found to be the major metabolites at low substrate concentrations. Km values for aminoguanidine formation were on the order of 20-fold lower than those measured for nitroso metabolites in the 2 mg ppAOX/ mL experiments. Vmax values were severalfold lower; however, Vmax/Km indicate that aminoguanidines could be the predominant metabolites at low substrate concentrations, given that AOX concentrations are suitable. Nitroso formation did not saturate at substrate concentrations up to 4 mM in the 5 mg ppAOX/ mL experiments. In our earlier report with rabbit liver cytosol, ratios of aminoguanidine to nitrosoguanidine metabolite formation shifted over 60-fold with a 4-fold increase in protein concentration, while total metabolite formation increased proportionately (11). It is not known why metabolic profiles differ so greatly with protein concentration and which conditions best represent hepatic physiology. The disparity in Km values of nitroso- and aminoguanidine formation is also puzzling. Coupled with the finding that reduction of IMI to IMI-NH2 is oxygen sensitive while reduction to IMI-NO is not, one is left to question whether the two activities take place in different active sites, perhaps on different enzymes. AOX and two homologues (AOH1 and AOH2) have been identified in mice (17); however, the rabbit has not been subjected to as extensive genomic investigation. Nevertheless, production of the IMI metabolites requires AOX-specific electron donor substrates and is sensitive to AOX-specific inhibitors (11). Substrate inhibition for aminoguanidine formation is apparent in the 5 mg ppAOX/mL experiments. However, the immediate substrate for aminoguanidine formation is not known. It is possible that AOX could reduce a nitroguanidine directly to an aminoguanidine or that free nitrosoguanidine metabolite could serve as its direct precursor. The latter possibility was examined using IMI-NO. This compound was found to be a mechanismbased inactivator of AOX with a very low partition ratio (∼3) and a KI of 1.3 mM. In summary, this investigation offers insight into the substrate specificity and catalytic mechanism of the reductive activity of

Substrate Specificity of Rabbit Aldehyde Oxidase

AOX and helps to interpret the extensive studies on in vivo metabolism of neonicotinoid insecticides (7). Acknowledgment. We thank Richard Staub, Daniel SchulzJander, Jennie Chin, and Gary Quistad for advice and assistance. The project described was supported by Grant R01 ES08424 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.

Chem. Res. Toxicol., Vol. 19, No. 1, 2006 43

(11) (12)

(13)

(14)

Supporting Information Available: LC/MS/MS fragmentation patterns for neonicotinoids and their metabolites. Proposed scheme for formation and reactions of nitrosomethylene derivatives. This material is available free of charge via the Internet at http:// pubs.acs.org.

(15) (16)

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, 300 pp, Springer-Verlag, Tokyo, Japan. (3) Tomizawa, M., and Casida, J. E. (2003) Selective toxicity of neonicotinoids attributable to specificity of insect and mammalian nicotinic receptors. Annu. ReV. Entomol. 48, 339-364. (4) Tomlin, C. D. S., Ed. (2003) The Pesticide Manual, 13th ed., 1344 pp, British Crop Protection Council, Alton, Hampshire, United Kingdom. (5) Yamamoto, I., Tomizawa, M., Saito, T., Miyamoto, T., Walcott, E. C., and Sumikawa, K. (1998) Structural factors contributing to insecticidal and selective actions of neonicotinoids. Arch. Insect Biochem. Physiol. 37, 24-32. (6) Nauen, R., Ebbinghaus-Kintscher, U., Salgado, V. L., and Kaussmann, M. (2003) Thiamethoxam is a neonicotinoid precursor converted to clothianidin in insects and plants. Pestic. Biochem. Physiol. 76, 5569. (7) Tomizawa, M., and Casida, J. E. (2005) Neonicotinoid insecticide toxicology: Mechanisms of selective action. Annu. ReV. Pharmacol. Toxicol. 45, 247-268. (8) Klein, O. (1994) The metabolism of imidacloprid in animals. Eighth IUPAC International Congress of Pesticide Chemistry: Abstract 367, Washington DC. (9) Solecki, R. (2001) Toxicological evaluations: Imidacloprid. www.inchem.org/documents/jmpr/jmpmono/2001pr06.htm. (10) Schulz-Jander, D. A., Leimkuehler, W. M., and Casida, J. E. (2002) Neonicotinoid insecticides: Reduction and cleavage of imidacloprid

(17)

(18) (19)

(20) (21) (22)

(23) (24)

nitroimine substituent by liver microsomal and cytosolic enzymes. Chem. Res. Toxicol. 15, 1158-1165. 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. Schulz-Jander, D. A., and Casida, J. E. (2002) Imidacloprid insecticide metabolism: Human cytochrome P450 isozymes differ in selectivity for imidazolidine oxidation versus nitroimine reduction. Toxicol. Lett. 132, 65-70. 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. Kleier, D., Holden, I., Casida, J. E., and Ruzo, L. O. (1985) Novel photoreactions of an insecticidal nitromethylene heterocycle. J. Agric. Food Chem. 33, 998-1000. 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. Ghafourian, T., and Rashdi, M. R. (2001) Quantitative study of the structural requirements of phthalazine/quinazoline derivatives for interaction with human liver aldehyde oxidase. Chem. Pharm. Bull. 49, 1066-1071. Garattini, E., Mendel, R., Roma˜o, M. J., Wright, R., and Terao, M. (2003) Mammalian molybdo-flavoenzymes, an expanding family of proteins: Structure, genetics, regulation, function and pathophysiology. Biochem. J. 372, 15-32. Tatsumi, K., Kitamura, S., and Yamada, H. (1983) Sulfoxide reductase activity of liver aldehyde oxidase. Biochim. Biophys. Acta 747, 8692. Kitamura, S., and Tatsumi, K. (1984) Reduction of tertiary amine N-oxides by liver preparations: Function of aldehyde oxidase as a major N-oxide reductase. Biochem. Biophys. Res. Commun. 121, 749754. Tatsumi, K., Yamada, H., and Kitamura, S. (1983) Reductive metabolism of N-nitrosodiphenylamine to the corresponding hydrazine derivative. Arch. Biochem. Biophys. 226, 174-181. Stoddart, A. M., and Levine, W. G. (1992) Azoreductase activity by purified rabbit liver aldehyde oxidase. Biochem. Pharmacol. 43, 22272235. Dalvie, D. K., Kalgutkar, A. S., Khojasteh-Bakht, S. C., Obach, R. S., and O’Donnell, J. P. (2002) Biotransformation reactions of fivemembered aromatic heterocyclic rings. Chem. Res. Toxicol. 15, 269299. Wolpert, M. K., Althaus, J. R., and Johns, D. G. (1973) Nitroreductase activity of mammalian liver aldehyde oxidase. J. Pharmacol. Exp. Ther. 185, 202-213. Honda, H., Tomizawa, M., and Casida, J. E. (2006) Neonicotinoid metabolic activation and detoxification established with coupled nicotinic receptorsCYP3A4 andsaldehyde oxidase systems. Toxicol. Lett. In press.

TX050230X