944
Chem. Res. Toxicol. 2006, 19, 944-951
Chloropyridinyl Neonicotinoid Insecticides: Diverse Molecular Substituents Contribute to Facile Metabolism in Mice Kevin A. Ford and John E. Casida* EnVironmental Chemistry and Toxicology Laboratory, Department of EnVironmental Science, Policy and Management, UniVersity of California, Berkeley, California 94720-3112 ReceiVed March 31, 2006
Chloropyridinyl neonicotinoid insecticides play a major role in crop protection and flea control on cats and dogs. Imidacloprid (IMI), nitenpyram (NIT), thiacloprid (THI), and acetamiprid (ACE) have in common the 6-chloro-3-pyridinylmethyl group but differ in the nitroguanidine, nitromethylene, or cyanoamidine substituent on an acyclic or cyclic moiety. Earlier metabolism studies were made with rats, goats, and hens but not with mice or under conditions suitable to compare metabolic pathways or pharmacokinetics. In this investigation, IMI, NIT, THI, and ACE were individually administered ip to mice at 10 or 20 mg/kg for analysis of brain, liver, and plasma at 15-240 min and 0-24 h urine by HPLC/DAD, LC/MSD, and LC/MS/MS. Maximum levels of the parent compounds in brain were 11-16 ppm (NIT and THI), 6 ppm (IMI), and 3 ppm (ACE). Persistence in the tissues was greater for ACE than the other neonicotinoids. Urinary excretion of the parent compound was greatest with NIT and IMI. Each of the compounds was cleaved to the same eight urinary metabolites derived from the chloropyridinylmethyl moiety (i.e., chloropyridinecarboxylic acid and its methylthio-, hydroxy-, and N-acetylcysteinyl derivatives and glycine, Oglucuronide, and sulfate conjugates thereof). Three nitro- or cyano-containing fragments were identified from the rest of the molecule for IMI, NIT, and ACE and one for THI. IMI gave nitrosoguanidine, aminoguanidine, guanidine (desnitro), olefin, methyltriazinone, and hydroxy- and dihydroxyimidazole derivatives. NIT metabolism involved N-demethylation, conversion to a cyano derivative via a nitrosomethylene intermediate, and oxidation at the nitromethylene carbon to the carboxylic acid. THI yielded olefin, imine (descyano), descyano olefin, amide, and hydroxythiazolidine derivatives and a ring-opened and methylated THI sulfoxide. ACE formed N-desmethyl, acetamide, amide, chloropyridinylmethylamine, and N-methylchloropyridinylmethylamine derivatives. Despite their common metabolites, these neonicotinoids differ greatly in their molecular sites and rates of metabolism in mice.
Introduction Neonicotinoids are the newest major class of insecticides (13). They are increasingly replacing the organophosphate and methylcarbamate acetylcholinesterase inhibitors which are losing their effectiveness because of selection for resistant pest populations. The neonicotinoids are systemic in plants to manage crop pests and in animals to control fleas on cats and dogs, currently accounting for 11-15% of the total insecticide market. They act as agonists at the nicotinic acetylcholine receptor (nAChR)1 and are selectively toxic to insects versus mammals in large part because of higher potency on insect than mammalian nAChRs (4, 5). * To whom correspondence should be addressed. Phone: (510) 6425424. Fax: (510) 642-6497. E-mail:
[email protected].
Imidacloprid (IMI) is the most important neonicotinoid (13). It has a 6-chloro-3-pyridinylmethyl substituent and is sometimes referred to as a chloronicotinyl compound (3). Related insecticides are nitenpyram (NIT), thiacloprid (THI), and acetamiprid (ACE) (Figure 1). The toxicology of each of these compounds has been extensively studied in mice, rats, and dogs to determine the no-observed-adverse-effect levels important for extrapolating safe dietary intakes and granting residue tolerances (6-9). The available knowledge on chloropyridinyl neonicotinoid insecticide metabolism in mammals comes mostly from studies of metabolites in tissues and excreta of rats, lactating goats, and laying hens required for registration and granting tolerance values. These investigations on individual compounds are reported more completely for IMI (10-12) and THI (13) than for NIT (14) and ACE (8), but in all cases, they are lacking in details on metabolite localization, separation, and identification. Species differences are an important aspect of understanding neonicotinoid toxicology and metabolism. Although extensively used in the toxicology investigations, no chloropyridinyl metabolism studies are reported for mice, and no pharmacokinetic data are available for any species that allow a direct comparison of one compound with another. This investigation helps fill this gap by defining the metabolic pathways and pharmacokinetics in mice of IMI, NIT, THI, and ACE administered ip at 10 or 20 mg/kg with analyses of brain, liver, and plasma at 15-240 min and 0-24 h urine. Most metabolites are characterized and analyzed by comparison with authentic standards from synthesis using liquid chromatography coupled with a mass selective detector (LC/MSD) or by HPLC with a diode array detector (HPLC/DAD). The goal of this research is to define common metabolites useful in recognizing chloropyridinyl neonicotinoid insecticide exposure and unique metabolites associated with specific compounds. Each neonicotinoid has a great variety of molecular substituents, and collectively, they provide an intriguing array of possible sites for metabolic attack.
Materials and Methods Chemicals. IMI and ACE were from Chem Service (West Chester, PA) and NIT and THI from Sigma-Aldrich (St. Louis, MO). Structures and designations for the observed and candidate metabolites are given in Figures 2-7. Asterisks designate com1 Abbreviations: ACE, acetamiprid; acet, acetamide derivative; DAD, diode array detector; dm, desmethyl derivative; ESI, electrospray ionization; gluc, glucuronide; GSH, glutathione; IC50, concentration for 50% inhibition; IMI, imidacloprid; IMI-tri, methyltriazinone derivative of IMI-NNH2; IS, internal standard; mAU, milliabsorbance unit; MSD, mass selective detector; nAChR, nicotinic acetylcholine receptor; NH, guanidine or imine derivative; NNH2, aminoguanidine derivative; NIT, nitenpyram; NNO, nitrosoguanidine derivative; ole, olefin derivative; ppm equiv, ppm equivalents based on the absorbances at 254 nm and recovery values of the parent compounds; SIM, selected ion monitoring; TFA, trifluoroacetic acid; THI, thiacloprid.
10.1021/tx0600696 CCC: $33.50 © 2006 American Chemical Society Published on Web 06/08/2006
Chloropyridinyl Neonicotinoid Metabolism
Chem. Res. Toxicol., Vol. 19, No. 7, 2006 945
Figure 1. Chloropyridinyl neonicotinoid insecticides with nitroguanidine, nitromethylene, cyanoguanidine, and cyanoamidine substituents. Table 1. Chloropyridinyl Neonicotinoid Insecticides and Metabolites in Mouse Tissues and Urine compounda
MW
tR (min) HPLC/DAD LC/MSD
identification criteriab
IMI IMI-NH IMI-NNO IMI-NNH2 IMI-tri IMI-ole IMI-5-OH IMI-diol IMI-urea*c IMI-de*
IMI and metabolites 255.7 18.6 210.7 11.6 239.7 13.8 225.7 12.5 278.7 13.1 253.7 15.2 271.7 16.1 287.7 15.6 211.7 15.4 229.6 14.9
18.6 3.0 14.1 4.1 13.2 15.3 16.1 15.6 15.4 -
1 1 1 1 1 1 1 1 5 5
NIT NIT-dm NIT-dm-COOH NIT-CN
NIT and metabolites 270.7 9.2 256.7 14.0 243.7 236.7 17.8
9.2 14.0 11.7 22.0
1 1 4 2
THI THI-NH THI-ole THI-ole-NHd THI-4-OH THI-NCONH2 THI-4-OH-NCONH2* THI-SO* THI-SO3H-NCONH2* THI-SOMe
THI and metabolites 252.7 23.0 227.7 9.5 250.7 22.1 225.7 10.0 268.7 22.1 270.7 16.5 286.7 14.6 268.7 16.1 336.8 24.1 300.8 18.9
22.9 3.0 22.1 3.2 20.3 16.2 16.2 18.9
1 2 1 2 1 2 5 5 5 3
ACE ACE-dm ACE-acet ACE-dm-acet ACE-dm-NCONH2 ACE-U
ACE and metabolites 222.7 19.8 208.7 17.8 198.7 16.2 184.6 226.7 268.3 14.3
19.8 19.2 16.2 15.8 15.5 14.3
1 2 2 4 4 6
a See Figures 3-5 and 7 for structures and detection in brain, liver, plasma, and urine. b See Results on Metabolite Identification. c Cyclic urea or imidazolidinone. d Detected in liver on administration of THI-ole but not THI.
pounds available as standards but not observed as metabolites. IMI metabolites in Table 1 were either synthesized in the Environmental Chemistry and Toxicology Laboratory at Berkeley (IMI-NH, IMINNO, IMI-NNH2, and IMI-tri) (15) or provided by Bayer Agrochemicals (Leverkusen, Germany and Stilwell, KA) (the other compounds). Candidate metabolites described earlier were NITdm (16, 17) and NIT-CN (18) from NIT; THI-NH and THI-oleNH from THI (19, 20). THI-ole, THI-4-OH, THI-NCONH2, THI4-OH-NCONH2, and THI-SO3H-NCONH2* were provided by Bayer. THI-SO* was obtained by treatment of THI (0.2 mmol) with equimolar m-chloroperoxybenzoic acid in chloroform (2 mL) for 5 h at room temperature and recrystallization from ethyl acetate-hexane (Lee, D. L., and Casida, J. E., unpublished result). ACE-acet was prepared by acetylation of r with acetyl chloride in anhydrous acetonitrile and ACE-dm was provided by Nippon Soda
Co. (Tokyo, Japan). Three cleavage product candidate metabolites (f, h, and i) shown in Figure 2 were synthesized in the Berkeley laboratory, while a*, b*, c, and g were from Sigma-Aldrich. Compound k was also from Sigma-Aldrich, and l, m, q, and u were from preparation at Berkeley. Cleavage products k, l, m, q, and w exist in tautomeric forms, only one of which is shown. Treatment of Mice and Collection of Tissue Samples and Excreta. Male albino Swiss-Webster mice (25-30 g) from Harlan Laboratories (Indianapolis, IN) were administered the test compounds ip at 10 mg/kg (IMI, THI, and ACE) or 20 mg/kg (NIT only) using Me2SO (1 µL/g mouse weight) as the carrier vehicle, or Me2SO alone was injected as the control. The ip route was used to emphasize differences in metabolism rather than absorption, and the dose was near the maximum for asymptomatic animals. Isoflurane was employed for anesthesia and did not produce interfering peaks in tissue and excreta extracts from control mice. One set of studies emphasized tissues and another the excreta. For tissues, mice were sacrificed at 15, 30, 60, 120, and 240 min after treatment, blood was drawn by cardiac puncture, and brain and liver were dissected out. The whole blood was centrifuged (5000g, 10 min) to recover the plasma. To collect urine and feces, the treated mice were placed in all-glass metabolism cages for 24 h with food and water ad libitum. Tissues were analyzed fresh, and excreta were held for up to 2 days at -80 °C before extraction and analysis. Extraction of Tissues and Excreta. Whole brain (350-375 mg), liver (750-800 mg), or plasma (100 µL) was transferred to an icecold 50 mL polypropylene conical tube. Acetonitrile (5 mL) and NaCl (250 mg) were added, and the internal standard (IS) was introduced (1 ng/mg tissue) in 100 µL of 75:25:0.1 acetonitrile/ water/trifluoroacetic acid (TFA). Urine (100-200 µL; one-tenth of the 0-24 h sample) and feces (100 mg) were processed in the same way as the tissues, except no IS was added. Homogenates were prepared using a Sonic Dismembrator (Fisher Scientific, Pittsburgh, PA) at maximum power output for 2-3 min, followed by vigorous vortex treatment for 2-3 min to ensure complete tissue disintegration. Centrifugation at 2000g for 15 min produced a precipitate and upper acetonitrile fraction which was collected in an 8-mL glass tube and evaporated to dryness (about 3 h) on a Savant SVC 200H Centrifugal Evaporator (Farmingdale, NY). The extract at this stage was stored overnight at 4 °C or preferably analyzed directly as below. HPLC/DAD Analysis. The evaporated extract as a yellow film with some precipitate was dissolved or suspended in 75:25:0.1 acetonitrile/water/TFA (reagent grade) (300 µL), sonicated, and filtered through a nylon membrane (0.45 µm) in an Acrodisc syringe filter (13 mm) (Pall Life Sciences, East Hills, NY). One-third of the sample (100 µL) was subjected to HPLC analysis on a Luna C-18 column (5 µm, 100 Å, 250 mm × 4.6 mm) with a precolumn filter (Phenomenex, Torrance, CA). The Hewlett-Packard model 1050 liquid chromatograph was fitted with a quaternary HPLC pump, a vacuum degasser, a DAD equipped with a deuterium lamp, and an auto sampler. Gradient development was with acetonitrile (HPLC grade)/pure deionized water (18.2 MΩ‚cm) containing 0.1% TFA, beginning with 5% acetonitrile and steadily increasing to 80% over a period of 30 min. This gradient generally resolved the neonicotinoids and metabolites, then an additional 10 min with 5% acetonitrile eluted interfering materials. The flow rate was 1 mL/ min, and absorbance measurements were at 254 nm. Tissue levels (ppm) were determined by peak area comparison for the neonicotinoid analyte with the IS. Each neonicotinoid gave a recovery of 80% or greater. The IS was THI for mice treated with IMI, NIT, and ACE (see Table 1 for tR values) and dinotefuran (a commercial neonicotinoid lacking a chloropyridinyl moiety) (tR 10.6 min) for THI-treated mice. The sensitivity of the method is indicated by the peak areas [milliabsorbance units (mAU‚s)] for 0.1 µg of neonicotinoid, that is, 346 for IMI, 59 for NIT, 99 for THI, and 137 for ACE. Metabolite levels are given as ppm equivalent (ppm equiv) based on the absorbances at 254 nm and recovery values of the parent compounds. Urine levels of parent neonicotinoids (percent of administered dose) were determined by comparing samples from treated mice with the same aliquot of the correspond-
946 Chem. Res. Toxicol., Vol. 19, No. 7, 2006
Ford and Casida
Table 2. Chloropyridinyl Neonicotinoid Cleavage Products as Metabolites in Mouse Brain, Liver, and Urine tR (min) compounda a* b* c d e f g h i j
MW
HPLC/DAD
LC/MSD
identification criteriab
Common to IMI, NIT, THI, and ACE (Figure 2) 141.6 11.9 5 143.6 16.2 5 157.6 18.0 3.0 1 333.7 20.5 4 284.3 22.4 3 169.2 20.2 20.2 2 139.1 1.7 2 214.6 12.5 13.4 1 226.3 16.4 16.4 2 219.2 20.2 4
source b, l, u u b, l, u u b, l, u l, u u u
104.1 130.1 128.1
Unique to IMI (Figure 3) 3.3 2.0 3.5 5.4 4.6 5.0
4 3 3
b, l, u u b, u
n o p
131.1 145.2 321.3
Unique to NIT (Figure 4) 3.9 3.5 25.8
4 4 4
u u u
q
125.2
Unique to THI (Figure 5) 6.6 6.6
3
b, l
156.6 142.6 170.6 97.1 273.3 83.1
Unique to ACE (Figure 7) 6.6 4.2 12.5 3.4 11.2 5.5 18.7 18.7 3.8
3 4 5 4 4 4
u u u b, l, u b, l, u
k l m
r sc t* u v w
a See Figure 2 for structures and detection in brain, liver, plasma, and urine. b See Results on Metabolite Identification. c Detected only with ACE, although also a possible metabolite of the other neonicotinoids.
ing control urine fortified with the administered neonicotinoid. Control urine gave four prominent HPLC peaks at tR 11.9, 13.5, 16.9, and 19.6 min, the latter two of which partially overlap i at 16.4 min and ACE at 19.8 min. Control feces gave three prominent peaks at 12.8, 14.8, and 16.3 min. The endogenous materials, partially interfering with UV analysis, did not affect the LC/MSD or LC/MS/MS characterization. LC/MSD Analysis. The quadrupole ion analyzer HewlettPackard 1100 LC/MSD system consisted of a binary pump, thermostated column compartment at 25 °C, vacuum degasser, a DAD with a deuterium lamp, and an autosampler. The acetonitrile/ water gradient and C-18 column were the same as those used for HPLC/DAD analysis except that TFA was replaced with 0.01% formic acid (HPLC grade). The ions used for selected ion monitoring (SIM) of each compound gave a strong signal with positive mode electrospray ionization (ESI). The conditions for LC/ MS/MS were as previously described (21). Bioactivity. Potency at the R4β2 nAChR primary target of agonist action was compared as concentration for 50% inhibition (IC50) for the parent compounds and metabolites using data from our earlier studies (5, 15, 22).
Results Chromatography. Chromatographic properties of the chloropyridinyl neonicotinoid insecticides, their cleavage products, and metabolites are given in Tables 1 and 2. The same column and elution gradient were used for HPLC/DAD and LC/MSD with the single exception that TFA was employed for the former and formic acid for the latter analyses. Accordingly, the tR values of most compounds are essentially the same in both systems with major differences only for the protonated compounds (IMINH, IMI-NNH2, THI-NH, THI-ole-NH, c, and s). Metabolite Identification. Metabolites were recognized by a new peak in the treated sample compared with the control.
Figure 2. Chloropyridinecarboxylic acid and its metabolites in mice treated ip with imidacloprid, nitenpyram, thiacloprid, or acetamiprid. The first intermediate from N-methylene hydroxylation (leading to aldehyde a*) is not shown. Metabolites c, e, and g were observed in brain, liver, and urine of mice treated with the four neonicotinoids. The glycine conjugate of g was not observed on monitoring urine by LC/MSD following treatment with g or its precursors. Partial pathways were confirmed by administering c, f, and g and observing their subsequent metabolites. Brackets designate hypothetical but not observed intermediates. Metabolic pathways of the GSH conjugate to e, f, and g indicated by single arrows are multistep reactions.
Only metabolites absorbing at 254 nm were detected in the HPLC/DAD studies, while additional ones were analyzed by LC/MSD. Six categories were used for metabolite identification based on the indicated criteria: (1) identical with synthetic standard by LC/MSD (tR, [M+], and 35Cl/37Cl ratio) and also by HPLC/DAD cochromatography; (2) identical with synthetic standard by LC/MSD (tR, [M+], and 35Cl/37Cl ratio); (3) previously reported metabolite (see below) not available as synthetic standard but observed here by LC/MSD (M+ and 35Cl/37Cl ratio when chlorine present); (4) new metabolite at indicated tR with tentative structure proposed based on LC/MSD (M+ and 35Cl/37Cl ratio when chlorine present); (5) available as synthetic standard but not detected in tissues or urine by LC/ MSD (tR and [M+]); and (6) identity unknown. Metabolites in category 3 (previously reported) were as follows: e, l, and m (10-12); q and THI-SOMe (13); a series of THI metabolites with modified cyano substituents (Figure 6) (13); r, ACE-dm, and ACE-acet (8, 23). Five IMI metabolites (IMI-NH, IMINNO, IMI-NNH2, IMI-tri, and IMI-ole in liver and IMI-NH and IMI-NNO in brain) were confirmed by LC/MS/MS for [M + H]+ and [M + H + CH3CN]+ plus fragmentation patterns (15, 18, 21). Chloropyridinecarboxylic Acid Metabolism (Figure 2). Chloropyridinecarboxylic acid (c) and its glycine conjugate (h) were prominent urinary metabolites of IMI, NIT, THI, and ACE (Table 3). There were also six other metabolites (d, e, f, g, i, and j) common to each of these insecticides. The sequence of metabolite formation (precursor-product relationship) was established by administering (ip) c, f, and g (20 mg/kg) to individual mice and analyzing the 0-24 h urine. Compound c gave d-j; f yielded g, i, and j; and g gave only j. Metabolites c, e, g, and h were also observed in brain or liver. Imidacloprid Metabolism (Figure 3, Tables 1-3). IMI gave 10 identified derivatives (including three cleavage products) in addition to chloropyridinecarboxylic acid and seven of its further metabolites. The brain contained not only IMI but also IMINH and IMI-NNO. Liver analysis revealed IMI, IMI-NH, IMINNO, IMI-NNH2, IMI-tri, IMI-ole, IMI-5-OH, and IMI-diol,
Chloropyridinyl Neonicotinoid Metabolism
Chem. Res. Toxicol., Vol. 19, No. 7, 2006 947
Table 3. Chloropyridinyl Neonicotinoid Urinary Excretion, Tissue Levels, and Persistence in Mice Treated ip at 10 or 20 mg/kg Neonicotinoid (dose, mg/kg) parameter parenta c h other
IMI (10)
NIT (20)
THI (10)
Urinary Products 0-24 h (% equivalent ( SD) 22 ( 6 46 ( 6 1.3 ( 0.9 3(2 4.8b 12 ( 6 2.0b 2.9b 5.0b IMI-5-OH NIT-dm THI-4-OH 22 ( 7 7.3 ( 0.9 3.5 ( 0.8
Parent Compound in Tissues maximum level (ppm) brain 6 16 11 liver 18 105 29 plasma 8 12 t1/2 relative to maximum level (min) brain 90 80 40 liver 30 45 50 plasma 80 50 b
ACE (10) 1.6 ( 1.1 2.4 ( 0.4 8.0b
3 12 6 >240 >240 >240
a Parent in feces
