See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/26289734 Metabolism of carbosulfan. I. Species differences in the in vitro biotransformation by mammalian hepatic microsomes including... Article in Chemico-biological interactions · July 2009 DOI: 10.1016/j.cbi.2009.06.001 · Source: PubMed CITATIONS READS 18 114 4 authors: Khaled M. Abass Petri Reponen 28 PUBLICATIONS 316 CITATIONS 16 PUBLICATIONS 241 CITATIONS University of Oulu University of Oulu SEE PROFILE SEE PROFILE Sampo Mattila Olavi Pelkonen 43 PUBLICATIONS 601 CITATIONS 514 PUBLICATIONS 15,895 CITATIONS University of Oulu SEE PROFILE University of Oulu SEE PROFILE All content following this page was uploaded by Khaled M. Abass on 23 December 2016. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Chemico-Biological Interactions 181 (2009) 210–219 Contents lists available at ScienceDirect Chemico-Biological Interactions journal homepage: www.elsevier.com/locate/chembioint Metabolism of carbosulfan. I. Species differences in the in vitro biotransformation by mammalian hepatic microsomes including human Khaled Abass a,∗ , Petri Reponen a,b , Sampo Mattila b , Olavi Pelkonen a a b Pharmacology and Toxicology Unit, Institute of Biomedicine, P.O. Box 5000, FIN-90014 University of Oulu, Oulu, Finland Department of Chemistry, P.O. Box 3000, FIN-90014 University of Oulu, Oulu, Finland a r t i c l e i n f o Article history: Received 1 April 2009 Received in revised form 29 May 2009 Accepted 3 June 2009 Available online 11 June 2009 Keywords: Pesticides Toxicokinetics P450 In vitro Metabolism Risk assessment a b s t r a c t The in vitro metabolism of carbosulfan, a widely used carbamate insecticide, by hepatic microsomes from human, rat, mouse, dog, rabbit, minipig, and monkey was studied. Altogether eight (8) phase I metabolites were detected by LC–MS; phase II metabolites were not found in human homogenates fortified with appropriate cofactors. The primary metabolic pathways were the initial oxidation of sulfur to carbosulfan sulfinamide (‘sulfur oxidation pathway’) and the cleavage of the nitrogen sulfur bond (N–S) to give carbofuran and dibutylamine (‘carbofuran pathway’). Carbofuran was further hydroxylated to 3-hydroxycarbofuran and/or 7-phenolcarbofuran, which were further oxidized to 3-ketocarbofuran or 3-hydroxy-7-phenolcarbofuran, respectively, and finally to 3-keto-7phenolcarbofuran. 3-Hydroxycarbofuran was the main metabolite in all species, but otherwise there were some qualitative interspecies differences in carbofuran pathway metabolites. Only rabbit liver microsomes were able to metabolize carbofuran via hydroxylation to 7-phenolcarbofuran. Carbofuran was not detected in dog liver microsomes due to rapid further metabolism. In general, liver microsomes from all seven species produced more toxic products (carbofuran, 3-hydroxy-carbofuran, 3-ketocarbofuran) more rapidly than a detoxification product (carbosulfan sulfinamide). Differences in intrinsic hepatic clearances (CLint ) between the lowest and highest species were moderate; 2-fold for the carbofuran pathway, 2.7-fold for carbosulfan sulfinamide and 6.2-fold for dibutylamine. Our studies, although restricted to in vitro metabolic data from human and animal hepatic preparations, provide valuable quantitative carbosulfan-specific data for risk assessment, which suggest that interspecies differences, for carbosulfan active chemical moiety, in toxicokinetics are within the standard applied factor for species extrapolation in toxicokinetics. These results will be valuable in further defining the risks associated with exposure to carbosulfan. © 2009 Elsevier Ireland Ltd. All rights reserved. 1. Introduction Carbosulfan belongs to the carbamates class of pesticides and to carbofuran sulfonylated derivatives. Carbosulfan is a systemic insecticide with contact and stomach actions, extensively used in the United States, Europe, and Asia for pest control for a wide range of crops. Carbosulfan toxicological evaluation was compiled by JMPR in 2003 [1]. Xenobiotic biotransformation is the process by which lipophilic foreign compounds are biotransformed through enzymatic processes to hydrophilic metabolites that are eliminated directly via renal elimination or after conjugation with endogenous cofactors. It is well recognized that theses enzymatic processes may lead to detoxification and/or activation reactions [2]. Studies on pesticide metabolism are an important issue, because even more ∗ Corresponding author. Tel.: +358 8 537 5231; fax: +358 8 537 5247. E-mail address: khaled.megahed@oulu.fi (K. Abass). 0009-2797/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.cbi.2009.06.001 toxic metabolites may arise than the parent compound such as in the case of organophosphorus insecticides, which are activated via cytochrome P450 enzymes-mediated desulfuration [3–9]. The cytochrome P450 (CYP) enzymes are important in the metabolism of various endogenous substrates as well as a wide range of xenobiotics. During the last few years, a growing number of papers on the role of CYPs in the metabolism of pesticides have appeared [10–14]. In the environment, the metabolism of carbosulfan involves hydroxylation and/or oxidation reactions, to be metabolized first to carbofuran [15] and then to 3-hydroxycarbofuran and 3-ketocarbofuran [16]. Other metabolites of carbosulfan, 3-hydroxy-7-phenolcarbofuran, 3-keto-7-phenolcarbofuran, 7-phenolcarbofuran, and dibutylamine, have also been detected [17]. In plants, carbofuran, 3-hydroxycarbofuran, 3-ketocarbofuran, and dibutylamine are the main degradation products detected in oranges by LC–MS [18]. In the rat, carbosulfan is metabolized by initial oxidation of the sulfur atom and by N–S bond cleavage. Metabolites were analyzed by TLC and the principal metabolites were the conjugated 3-keto-7- Author's personal copy K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 phenolcarbofuran, 3-ketocarbofuran, and dibutylamine [19]. In rat stomach, carbosulfan was slowly converted into a variety of products. Major alteration products were carbofuran and biscarbofuran N,N′ -disulfide. In addition, five polysulfide derivatives of carbosulfan or biscarbofuran N,N′ -disulfide were detected as minor products by TLC [20]. In male and female rats in vivo, 10 metabolites were identified by TLC and HPLC and major metabolites were confirmed by GC–MS. 3-Hydroxy-7-phenolcarbofuran, 3-hydroxycarbofuran, 3-keto-7-phenolcarbofuran, and 7-phenolcarbofuran were among the major metabolites, while carbofuran and 3-ketocarbofuran were also detected [1]. As it has been shown, carbosulfan is metabolized in the environment, plants, and mammals to carbofuran, 3-hydroxycarbofuran, and 3-ketocarbofuran. While all are actually toxic compounds, carbofuran is found to be more persistent and toxic than carbosulfan itself. This is a special case in which a less toxic pesticide (carbosulfan, LD50 250 and 185 mg kg−1 for male and female rats, respectively) is transformed into a more toxic one (carbofuran, LD50 8 mg kg−1 for male and female rats) [21–23]. Carbofuran metabolism was reviewed for economic animals by Akhtar [24] and its toxicity and metabolism was extensively reviewed by Gupta [25]. Risk assessment needs reliable scientific information and one source of information is the characterization of metabolic factors and toxicokinetics. Indeed, quantitative toxicokinetic data in humans are needed for human risk assessment to make reliable comparisons between individuals or between species [26,27]. No information is available on the human metabolism and mammalian hepatic enzyme kinetics of carbosulfan. In addition to toxicological evaluation, in vitro toxicokinetic data for the active chemical moiety will be valuable in further development of the proposed default subdivision of the usual uncertainty factor to quantitative toxicokinetic chemical-specific assessment factors (CSAFs) [28]. In the present study, the identification and quantification of carbosulfan metabolites by seven mammalian liver microsomes including human were investigated in vitro by LC/MS–MS to examine species differences in the phase I biotransformations of carbosulfan. 2. Materials and methods 2.1. Chemicals Carbosulfan (2,3-dihydro-2,2-dimethylbenzofuran-7yl(dibutylaminothio) methylcarbamate), carbofuran (2,3-dihydro2,2-dimethylbenzofurany-7-yl methylcarbamate), 3-hydroxycarbofuran (2,3-dihydro-3-hydroxy-2,2-dimethylbenzofuran-7-yl methylcarbamate), 3-ketocarbofuran (2,3-dihydro-3-oxy-2,2dimethylbenzofuran-7-yl methylcarbamate), 3-keto-7phenolcarbofuran (2,3-dihydro-2,2-dimethyl-3-oxobenzofuran-7-ol), 3-hydroxy-7-phenolcarbofuran (2,3-dihydro2,2-dimethylbenzofuran-3,7-diol), and 7-phenolcarbofuran (2,3-dihydro-2,2 dimethylbenzofuran-7-ol) were purchased from ChemService (West Chester, PA). Dibutylamine was purchased from Sigma–Aldrich (Germany) and carbaryl was a kind gift from Agrochem (Eg). HPLC-grade solvents were obtained from Rathburn (Walkerburn, UK) and Labscan (Dublin, Ireland). All other chemicals used were from the Sigma Chemical Company (St. Louis, MO) and were of the highest purity available. Water was freshly prepared in-house with the Simplicity 185 (Millipore S.A., Molsheim, France) water purification system and was UP-grade (ultra pure, 18.2 M). 2.2. Human liver homogenates and mammalian liver microsomes Human liver samples used in this study were obtained from the University Hospital of Oulu as surplus from organ donors. The 211 collection of surplus tissue was approved by the Ethics Committee of the Medical Faculty of the University of Oulu, Finland. All liver samples were of Caucasian race including 4 females and 6 males between the ages of 21 and 62. Intracerebral hemorrhage was the primary cause of death. Detailed characteristics of the liver samples are presented in our previous publication [12]. The livers were transferred to ice immediately after the surgical excision and cut into pieces, snap-frozen in liquid nitrogen, and stored at −80 ◦ C. Human liver homogenate was prepared from livers of 10 individuals by homogenizing liver tissue in four volumes of 0.1 M phosphate buffer (pH 7.4), i.e. the homogenate contained 200 mg of hepatic tissue per ml. Male DBA/2 mouse, Sprague–Dawley rat, Beagle dog, Cynomolgus monkey, Göttingen minipig, and New Zealand white rabbit liver samples were obtained after approval of the Ethics Committee of the University of Oulu, Finland. All microsomes were prepared by standard differential ultracentrifugation [29]. The final microsomal pellet was suspended in 100 mM phosphate buffer, pH 7.4. Protein content was determined by the Bradford method [30]. 2.3. In vitro assay of carbosulfan metabolites The standard incubation mixture contained 100 ␮M carbosulfan, 0.15 mg pooled liver microsomal protein (n = 10), and 1 mM NADPH in a final volume of 200 ␮l of 0.1 M phosphate buffer (pH 7.4). Carbosulfan was prepared once a week in dimethylsulfoxide (DMSO; final amount in the reaction medium 1.0%). After a 2-min incubation at +37 ◦ C in a shaking incubator block (Eppendorf Thermomixer 5436, Hamburg, Germany), the reaction was started by adding NADPH. The mixture was incubated at +37 ◦ C for 20 and 60 min and the reaction was stopped with 600 ␮l of ice cold acetonitrile containing an internal standard. All incubations were carried out in triplicate. After centrifugation at 10,000 × g for 15 min, the supernatant was collected and stored at −20 ◦ C until analyzed. To measure the production of potential metabolites, human liver homogenate incubations were prepared containing the same final carbosulfan concentration as the microsomal incubations. In addition to 40 ␮l of human liver homogenates (contains approximately 0.14 mg microsomal protein), the other components in homogenate incubates were 5 mM uridine 5′ -diphosphoglucuronic acid (UDPGA), 1 mM glutathione, 1.2 mM adenosine-3′ -phosphate5′ -phosphosulfate (PAPS), and 1 mM NADPH in a final volume of 200 ␮l of 0.1 M phosphate buffer (pH 7.4). The mixture was incubated at +37 ◦ C for 20 and 60 min and the reaction was stopped with 600 ␮l of ice cold acetonitrile containing an internal standard. The analytical method was similar to the microsomal preparations. Results are expressed as a mean ± standard deviation for three replicates. It has to be stressed here that while liver homogenates fortified by appropriate cofactors may be suitable for qualitative profiling of metabolites, it is not optimized for all of the enzymes such that kinetic parameters could be obtained. 2.4. Kinetic parameters To measure the enzyme kinetic parameters in microsomal samples, the standard incubation mixture contained carbosulfan (final concentrations 2.5–300 ␮M). Incubation mixtures and methods were the same as mentioned above, except the incubation times were 20 min for microsomal samples. Samples were analyzed by LC–MS–MS. The kinetic parameters Vmax and Km were calculated using Prism 5.0 (GraphPad Software, Inc., San Diego, CA) by nonlinear regression. These values were used to calculate the intrinsic clearance value (Vmax /Km ). All results are expressed as mean ± standard error for three replicates. In the standard experimental conditions used for carbosulfan metabolites, reaction rates were linear at least up to 0.15 mg of microsomal protein/ml and 60 min of incubation time. Author's personal copy 212 K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 2.5. Chromatography of the carbosulfan metabolites Samples were centrifuged before analysis for 10 min at 10,000 × g. Chromatographic separation was carried out with the Waters Alliance 2690 HPLC system (Waters Corp., Milford, MA). The column used was a Waters Atlantis T3 (2.1 mm × 100 mm, particle size of 3 ␮m) together with a Phenomenex C18 2.0 mm × 4.0 mm precolumn (Phenomenex, Torrance, CA). The temperature of the column oven was 45 ◦ C. The eluent flow rate was 0.4 ml/min. The eluents used were ultrapure-grade water containing 0.1% acetic acid (A) and methanol (B). A linear gradient elution from 5% B to 75% B in 8 min was applied. Solvent B was thus maintained at 98% for 3 min before re-equilibration (6 min). The total analysis time was 17 min. 2.6. Mass spectrometry The initial screening of the compounds’ present and accurate mass measurements were carried out using a Micromass LCT (Micromass, Altrincham, UK) time of flight (TOF) mass spectrometer equipped with a Z-Spray ionization source. A generic positive electrospray ionization method was used for all substrates and metabolites. The capillary voltage was 4000 V, cone voltage 23 V, and desolvation and source temperatures 300 and 150 ◦ C, respectively. Nitrogen was used as the desolvation and cone gas with flow rates of 780 and 300 L/h. The mass spectrometer and HPLC system were operated under Micromass MassLynx 3.4 software. For exact mass measurements the lock mass was N-1-naphthylphthalimide ([M+H]+ at m/z 274.0868) and it was delivered into the ionization source through a T-union using a syringe pump (Harvard Apparatus, Holliston, MA). The quantification (multiple reaction monitoring, MRM) and fragmentation measurements were performed with a Micromass Quattro II triple quadrupole instrument equipped with a Z-spray ionization source. The capillary voltage was 4000 V, and desolvation and source temperatures 280 and 150 ◦ C, respectively. The collision gas was argon with a CID gas cell pressure of 2.0 × 103 mbar. Nitrogen was used as the drying and nebulizing gas with flow rates of 450 and 15 L/h. The fragmentation reactions monitored (MRM), collision energies, and sample cone voltages for metabolites and the internal standard are presented in Table 1. External standards were measured in the beginning, middle, and end of the experiment to ensure the quality of the analysis. The lower limit of quantitation was 0.5 ␮M for all compounds. Intraday coefficients of variation were less than 20% throughout the quantitation range of 2.5–300 ␮M. 3. Results 3.1. Identification of carbosulfan metabolites produced in vitro by mammalian liver microsomes Incubations of pooled human (humanLM), rat (ratLM), mouse (mouseLM), dog (dogLM), rabbit (rabbitLM), minipig (minipigLM), and monkey (monkeyLM) liver microsomes with various con- Table 1 The names, molecular weights (MW), exact masses, fragmentations, sample cone voltages (SC), collision energies (CE), and retention times (RT) of analytes used in the measurements by LC–MS. Analytes MW M+H+ Exact mass Calculated mass Fragmentationsa SC (V) CE (eV) RT (min) Carbosulfan 380 381.1 381.2228 381.2212 222.0 (10) 164.9 (10) 160.0 (100) 118.0 (90) 25 15 12.9 Carbofuran 221 222.1 222.1132 222.1130 165.0 (100) 123.0 (40) 20 10 9.5 3-Hydroxycarbofuran 237 220.0b 238.1062 238.1079 163.0 (100) 135.0 (40) 25 10 7.5 3-Hydroxy-7-phenolcarbofuran 180 163.0b 163.0726c 163.0759 135.1 (100) 107.0 (75) 25 15 6.6 3-Ketocarbofuran 235 236.1 236.0956 236.0923 179.0 (100) 161.0 (60) 151.0 (50) 20 10 8.5 3-Keto-7-phenolcarbofuran 178 179.0 179.0723 179.0708 161.0 (100) 151.0 (15) 123.0 (20) 25 15 8.3 7-Phenolcarbofuran 164 165.0 165.0944 165.0916 123.0 (100) 105.1 (15) 95.0 (20) 77.0 (50) 20 15 9.7 Carbosulfan sulfinamide 396 397.2 397.2133 397.2161 163.1 (20) 160.1 (100) 128.1 (40) 118.0 (90) 20 20 12.4 Dibytylamine 129 130.1 130.1600 130.1596 74.1 (50) 69.0 (100) 57.1 (75) 41.2 (35) 25 15 4.4 Carbaryl (IS) 201 202.0 202.0824 202.0868 161.6 (20) 145.0 (90) 141.0 (100) 15 10 9.8 a b c Fragmentations monitored in the quantification are presented in bold. Metabolites were quantified as the protonated dehydrated molecule [M−H2 O+H]+ due to significant in-source fragmentation. Exact mass could be measured only from the protonated dehydrated molecule [M−H2 O+H]+ . Author's personal copy K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 Fig. 1. The overall in vitro scheme of carbosulfan metabolism in mammalian liver microsomes. 213 Author's personal copy 214 K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 centrations of carbosulfan were analyzed by LC–TOF-MS. Eight metabolites were detected from the extracted mass chromatograms and seven of them were identified with the help of reference standards. Biotransformations by cytochrome P450-mediated carbosulfan metabolism are presented in Fig. 1 and the exact masses of the analytes in Table 1. Fragmentations of the metabolites were determined by triple quadrupole MS. The fragmentations produced and the collision energy used in the method are presented in Table 1. The hydroxymetabolites, 3-hydroxycarbofuran and 3-hydroxy-7phenolcarbofuran, were quantified as the protonated dehydrated molecules [M−H2 O+H]+ due to significant in-source fragmentation. The unidentified metabolite could be either a sulfonamide derivative or a hydroxylated carbosulfan. Either of the metabolites or their derivatives have been detected by TLC in two different studies [1,19]. The carbosulfan sulfoxide metabolite has been characterized with an unknown method in human, mouse, and rat liver microsomes [31]. In the ion source of a mass spectrometer, hydroxy metabolites of both carbofuran and 7-phenolcarbofuran produced protonated dehydrated molecules. There is no reason why this should not also happen to the corresponding carbosulfan hydroxy metabolite. Based on previous reports and missing dehydration of the unidentified metabolite, we have assumed that it is carbosulfan sulfinamide even though fragmentation of the analytes also gives some indication of the hydroxylation of carbosulfan. For the exact identification of this unidentified metabolite, more experimental work should be performed. Good chromatographic separation of the detected metabolites by LC was produced, even if it is not necessarily needed because of MS–MS separation. Retention times of the analytes are presented in Table 1. In some measurements a broad dibutylamine peak was observed because dibutylamine is a polar molecule and thus elutes in the beginning of the chromatogram. Since carbofuran produces some in-source fragmentation of 7-phenolcarbofuran, part of it appears in the same chromatograph as 7-phenolcarbofuran (retention time differs by only 0.2 min). Occasionally, a very large carbofuran peak could make the detection of the small 7phenolcarbofuran peak difficult. 3.2. Metabolism of carbosulfan in vitro by mammalian liver microsomal samples Carbosulfan metabolite formation as a function of incubation time (20 and 60 min) using humanLM, ratLM, mouseLM, dogLM, rabbitLM, minipigLM, and monkeyLM at carbosulfan concentrations of 100 ␮M were analyzed by LC–MS–MS. The results are presented in Fig. 2. In 20-min incubations, mouseLM had the highest formation rate of carbofuran, while at 60 min humanLM were the highest (2.5 and 1.9 nmol/(mg protein min), respectively). 7-Phenolcarbofuran was detected only with rabbitLM at both 20 and 60 min (0.18 and 0.004 nmol/(mg protein min), respectively). The highest formation rate of 3-hydroxycarbofuran was detected with monkeyLM (15.2) and minipigLM (5.4 nmol/(mg protein min) at 20 and 60 min, respectively). Among the seven mammalian liver microsomes tested, minipigLM was the most active one in carbosulfan biotransformation to 3-ketocarbofuran (0.6 and 0.4 nmol/(mg protein min)) and 3-keto-7-phenolcarbofuran (0.6 and 0.3 nmol/(mg protein min)) for both incubation times. The corresponding concentrations of carbosulfan sulfinamide formed by dogLM were 10.2 and 2.0 nmol/(mg protein min) for both incubation times displaying the uppermost values. For dibutylamine, humanLM had the highest metabolite formation rate after 20 min, while dogLM had the highest after 60 min of incubation. 3-Hydroxy7-phenolcarbofuran was not detected in both ratLM and mouseLM, Fig. 2. The formations of carbosulfan metabolites in the presence of different mammalian liver microsomes (LM) as a function of incubation time (A) 20 min and (B) 60 min. Columns represent the means of three separate determinations and the error bars represent S.D. while 3-keto-7-phenolcarbofuran was not detected in ratLM. 7Phenolcarbofuran was detected only in rabbitLM. No phase II enzyme-associated metabolites were observed after incubation with human hepatic homogenates (data not shown). Alongside metabolite formation, reduction in the amount of the parent compound was observed with increasing incubation time (data not shown). It is worthy to notice that the vast majority of metabolite amounts, not the formation rates, were increased by time in all mammalian liver samples preparations in correspondence to the disappearance of the parent compound. 3.3. Carbosulfan depletion rate Disappearance or depletion of carbosulfan was quantified by triple quadrupole mass spectrometry from incubations of varying concentrations (from 2.5 to 150 ␮M) with seven mammalian microsomes including human. Fig. 3 summarizes the concentration-dependent depletion curves. The depletion curve for mouseLM was the best one for Michaelis–Menten analysis. The rate of NADPH-dependent elimination of carbosulfan was only 22% of the highest carbosulfan concentration. The depletion curve for dogLM increased practically linearly as a function of carbosulfan concentration and was not suitable for the calculation of Author's personal copy K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 215 est capacity (2.5, 2.5 and 2.0 nmol dibutylamine/(mg protein min), respectively). The CLint rate illustrated that ratLM was the most efficient of the mammalian liver microsomes for carbosulfan metabolism to dibutylamine, whereas rabbitLM was the least (114.5 and 18.3 ␮l/(mg protein min), respectively). 3.4.3. Species differences in the carbofuran pathway The results are presented in Table 4. Sums of all the metabolites of the carbofuran pathway were used for total intrinsic clearance calculations. HumanLM and monkeyLM displayed similar affinity (Km values 34.4 and 38.0 ␮M) and capacity (Vmax values 15.6 and 17.2 nmol/(mg protein min)). DogLM had the lowest affinity corresponding to the highest Km value (72.4 ␮M), while mouseLM and ratLM had the highest affinity (3.4 and 5.5 nmol/(mg protein min)). MonkeyLM, dogLM, and humanLM had the highest capacity corresponding to the highest Vmax , while the reverse was true for mouseLM, ratLM, and rabbitLM. The catalysis efficiency (CLint ) for minipigLM (471.4 ␮l/(mg protein min)) was higher than those for other mammalian liver microsomes, whereas dogLM had the lowest efficiency (223.9 ␮l/(mg protein min)). The rank order of the CLint values in different species for the carbofuran pathway was dogLM < mouseLM < ratLM < rabbitLM < monkeyLM < humanLM < minipigLM. Fig. 3. The concentration-dependent depletion rate of carbosulfan in mammalian liver microsomes. Michaelis–Menten kinetics. Carbosulfan was rapidly and efficiently metabolized at all concentrations; elimination was 85% of the highest carbosulfan concentration. The depletion curve for ratLM more closely resembled the one for mouseLM, whereas depletion curves for monkeyLM, minipigLM, rabbitLM, and humanLM were closer to the dogLM curve. 3.4. Kinetic parameters of the carbosulfan metabolism in mammalian liver microsomal samples Metabolites of carbosulfan were quantified by triple quadrupole mass spectrometry from incubations with seven mammalian microsomes including human. Carbosulfan biotransformation by mammalian liver microsomes followed Michaelis–Menten kinetics as demonstrated by Eadie-Hofstee plots (V versus V/S). The kinetic parameters for the two primary pathways of carbosulfan metabolism, i.e. sulfur oxidation and N–S bond cleavage (‘carbofuran’), in liver microsomes were determined using a wide concentration range (2.5–300 ␮M) of carbosulfan. 3.4.1. Species comparison of carbosulfan sulfinamide formation Kinetic parameters of seven mammalian liver microsomes with respect to the production of the carbosulfan sulfinamide are shown in Table 2. MouseLM exhibited the highest affinity, corresponding to the lowest Km (89.2 ␮M), and the lowest capacity, corresponding to the lowest Vmax (2.9 nmol carbosulfan sulfinamide/(mg protein min)). MonkeyLM displayed the lowest affinity (316.3 ␮M), whereas dogLM exhibited the highest capacity (13.5 nmol carbosulfan sulfinamide/(mg protein min)). DogLM shows the highest CLint rate, whereas humanLM and ratLM displayed the lowest rate (51.2, 18.7 and 11.9 ␮l/(mg protein min), respectively). 3.4.2. Species differences in dibutylamine formation The cleavage of the N–S bond of carbosulfan yields dibutylamine and carbofuran and its more distal metabolites. In different mammalian liver microsomes, dibutylamine was produced with different kinetics (Table 3). RatLM, mouseLM, and humanLM showed the highest affinity (22.0, 31.8, and 32.1 ␮M) and low- 3.4.3.1. Cross-species comparison of the formation rates of distal carbofuran metabolites. In view of the fact that reliable kinetic parameter estimates could not be obtained for more distal metabolites when using carbosulfan as a substrate, metabolite formation rates were used for species comparisons in these cases (Fig. 4). In some of the seven species, the amount of metabolites formed was below the limit of quantification at low carbosulfan concentrations. Among mammalian liver microsomes examined for carbosulfan metabolism, mouseLM and ratLM had the lowest 3-hydroxycarbofuran formation rates, while dogLM, monkeyLM, humanLM and minipigLM exhibited highest formation rates. The 3-hydroxycarbofuran formation rate displayed a sevenfold difference from 2.0 to 14.2 nmol/mg protein/min at 300 ␮M carbosulfan. It is worth noting that the formation rate of dogLM increased in an almost linear manner with increasing carbosulfan concentration. 3-Ketocarbofuran formation rates displayed about a 14fold variation between species (0.04–0.57 nmol/mg protein/min). RatLM and mouseLM displayed the lowest rate similarly to 3hydroxycarbofuran. DogLM and minipig exhibited the highest formation rates, while human rabbitLM and monkeyLM showed a roughly similar trend. Since ratLM and mouseLM had the lowest 3-hydroxy- and 3ketocarbofuran formation rates, their distal metabolites, 3-hydroxy and 3-keto-7-phenolcarbofuran, were not detected at any carbosulfan concentration. The highest 3-hydroxy-7-phenolcarbofuran formation rate was observed in dogLM (0.13 nmol/mg protein/min), while humanLM and rabbitLM displayed the lowest activity. RabbitLM had the lowest activity toward 3-keto-7phenolcarbofuran formation, while monkeyLM and minipigLM had the highest (0.19 nmol/mg protein/min). Generally, in all the different species, the amounts of 3-keto-7-phenolcarbofuran decreased with increasing carbosulfan concentration. Only rabbitLM produced 7-phenolcarbofuran. The maximal formation rate was 0.19 nmol/(mg protein/min). 4. Discussion This is the first study in which hepatic cytochrome P450dependent carbosulfan metabolism has been extensively compared in microsomes representing seven mammalian species including human, and qualitative and quantitative interspecies differences Author's personal copy 216 K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 Table 2 Kinetic parameters of carbosulfan sulfinamide formation obtained with different mammalian liver microsomesa . Liver microsomes Vmax (nmol/(mg protein min)) Km (␮M) HumanLM RatLM MouseLM DogLM RabbitLM MinipigLM MonkeyLM 5.0 3.2 2.9 13.5 8.8 4.5 8.6 268.6 264.8 89.2 262.6 229.4 177.4 316.3 a b ± ± ± ± ± ± ± 0.4 0.7 0.1 4.7 2.5 0.4 1.3 ± ± ± ± ± ± ± 42.8 103.5 8.2 50.7 23.3 34.9 82.0 CLint (␮l/(mg protein min)) Interspecies differencesb 18.7 11.9 33.0 51.2 38.2 25.2 27.3 0.64 1.76 2.74 2.04 1.35 1.46 CLint (␮l/(mg protein min)) Interspecies differencesb 63.2 114.5 80.0 86.7 18.3 57.4 47.4 1.81 1.27 1.37 0.29 0.91 0.75 Each value represents the mean ± std. error of three determinations. Interspecies differences represents animal to human fold differences in toxicokinetics. Table 3 Kinetic parameters of dibutylamine formation obtained with different mammalian liver microsomesa . Liver microsomes Vmax (nmol/(mg protein min)) Km (␮M) HumanLM RatLM MouseLM DogLM RabbitLM MinipigLM MonkeyLM 2.0 2.5 2.5 8.7 2.8 7.9 8.1 32.1 22.0 31.8 100.3 150.5 137.3 169.9 a b ± ± ± ± ± ± ± 1.2 0.3 0.3 0.6 0.8 0.5 2.3 ± ± ± ± ± ± ± 7.3 10.0 11.8 17.5 53.6 20.5 70.0 Each value represents the mean ± std. error of three determinations. Interspecies differences represents animal to human fold differences in toxicokinetics. Table 4 Kinetic parameters of the carbofuran–metabolic pathway obtained with different mammalian liver microsomesa . Liver microsomes Vmax (nmol/(mg protein min)) Km (␮M) CLint (␮l/(mg protein min)) Interspecies differencesb HumanLM RatLM MouseLM DogLM RabbitLM MinipigLM MonkeyLM 15.6 5.5 3.4 16.2 8.5 13.7 17.2 34.4 16.9 13.2 72.4 25.2 29.0 38.0 454.9 326.6 253.2 223.9 335.7 471.4 450.8 0.72 0.56 0.49 0.74 1.03 0.99 ± ± ± ± ± ± ± 1.1 0.4 0.3 0.8 0.6 0.9 1.2 ± ± ± ± ± ± ± 7.9 4.4 4.6 9.5 6.6 7.0 8.8 a Each value represents the mean ± std. error of three determinations. Sums of all the metabolites of the carbofuran pathway were used for the calculation of kinetic parameters. b Interspecies differences represents animal to human fold differences in toxicokinetics. were observed. The metabolites were quantified with the advantage of LC/MS/MS for high sensitivity and selectivity that enabled the analysis of the metabolites at very low concentrations. From eight detected metabolites, seven were identified. These metabolites were carbofuran, 3-hydroxycarbofuran, 3-hydroxy-7phenolcarbofuran, 3-ketocarbofuran, 3-keto-7-phenolcarbofuran, 7-phenolcarbofuran, carbosulfan sulfinamide and dibutylamine. These metabolites have been detected before by TLC in rat in vivo with the exception of carbosulfan sulfinamide. Instead of carbosulfan sulfinamide its further oxidized metabolite, carbosulfane sulfone, was detected [19]. Moreover, carbofuran and polysulfide derivatives of carbosulfan were detected in rat stomach by TLC [20]. In another study of carbosulfan metabolism in male and female rats, 10 metabolites were identified by TLC and HPLC and the major metabolites were confirmed by GC–MS [1]. Seven metabolites of the 10 rats in vivo metabolites were detected in our study. 7Phenolcarbofuran was detected only as a rabbit metabolite, while seven metabolites were described as human metabolites for the first time in this study. Our analyses allow the construction of the metabolic route map for carbosulfan in animal and human hepatic preparations. Carbosulfan was metabolized via two main metabolic pathways as shown in Fig. 1. The first metabolic pathway was the initial oxidation of sulfur to carbosulfan sulfinamide. The second metabolic pathway was via the cleavage of the nitrogen–sulfur bond (N–S) to give carbofuran and/or dibutylamine. Carbofuran was subsequently further metabolized. No phase II metabolites were detected in human liver homogenates and tested liver microsomes although screening of the potential metabolites had been carried out by LC–MS. Most metabolites that had been identified in vivo in rat were also presented in our in vitro study with hepatic microsomes. However, discrepancies were observed with some metabolites. The sulfate/glucuronide conjugation was not detected in vitro. However it should be kept in mind that the metabolism in rats in vivo was determined 15 days after administration of carbosulfan, whereas the maximum incubation period in vitro was only 60 min. The metabolic fate of carbofuran has been earlier investigated in in vitro experiments with human, rat, and mouse liver microsomes and metabolites were detected by LC–UV. Carbofuran was metabolized by cytochrome P450, leading to the production of a major metabolite, 3-hydroxycarbofuran, and two minor metabolites, 3-ketocarbofuran and an unknown metabolite, and finally by a non-enzymatic pathway to 3-keto-7-phenolcarbofuran [32]. The reported metabolites were detected in our studies in addition to 3-hydroxy-7-phenolcarbofuran as a further hydroxylation from 3-hydroxycarbofuran in all mammalian microsomes except rat and mouse. Only rabbitLM might able to produce 3-hydroxy-7-phenolcarbofuran via 3-hydroxycarbofuran and/or 7phenolcarbofuran hydroxylation. However, we did not observe any non-enzymatic hydrolysis of either 3-ketocarbofuran or the reference standards. A high stability of reference standard solutions has also been reported by Soler et al. [33]. Based on these findings, further metabolism of 3-ketocarbofuran may occur by enzymatic hydroxylation, which is probably catalyzed by a cytochrome P450- Author's personal copy K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 217 Fig. 4. The formation of carbosulfan metabolites in the presence of different mammalian liver microsomes (LM) as a function of carbosulfan concentration. Columns represent the means of three separate determinations and the error bars represent S.D. dependent reaction, although the involved isoforms were not studied. Moreover, cross-species differences in the formation of 3-keto-7-phenolcarbofuran support this finding. Some qualitative interspecies differences were observed in metabolite profiles. 3-Keto-7-phenolcarbofuran was not detected in ratLM, while 3-hydroxy-7-phenolcarbofuran was not detected in either ratLM or mouseLM. 7-Phenolcarbofuran was a rabbitLMspecific metabolite, not detected in any other tested species. However, 7-phenolcarbofuran has been detected as a carbosulfan metabolite in vivo in rat urine and feces and in goat’s milk, liver, and kidney [1]. The probable reason may be the short duration of in vitro incubations, not allowing enough time for the formation of minor metabolites. Carbofuran was not detected in dogLM. On the other hand, carbofuran formation is an initial step of the metabolic pathway leading finally to other more distal metabolites. In view of the fact that the further metabolites were detectable with dogLM (and at relatively high concentrations compared with other species), the most probable explanation of this is that carbofuran-metabolizing enzyme(s) in dogLM are so efficient that the carbofuran formed is immediately metabolized further into more distal metabolites. In general, the intrinsic clearance rates indicated that liver microsomes of all seven species metabolize carbosulfan via the carbofuran–metabolic pathway more rapidly than via the detoxification pathway, carbosulfan sulfinamide. It is worth noting that the carbofuran–metabolic pathway contains products (carbofuran, 3hydroxy-carbofuran, 3-ketocarbofuran), which are more toxic than the parent carbosulfan [19,22,34]. Carbofuran acts as a substrate for acetylcholinesterase, forming initially a Michaelis-like reversible complex [35], and it has been reported to be teratogenic and embry- otoxic [25]. In addition, metabolites of the dibutylamine moiety may enter the carbon pool and be incorporated into natural constituents of the body [1]. The interspecies differences in carbosulfan metabolite formation as a function of time correlated with the variation in kinetic parameters. Based on metabolic efficiency CLint , it is clear that the carbofuran–metabolic pathway is the main pathway in all the different species, and monkeyLM, dogLM, humanLM and minipigLM displayed the highest metabolic capacity (Vmax ). There is a relatively small variability, based on animal to human efficiency comparisons, in the level of differences in CLint values for all of the tested species regarding carbosulfan metabolite formations. MinipigLM, monkeyLM, and humanLM metabolized carbosulfan to carbofuran almost equally, while dogLM and mouseLM had the smallest values (0.49- and 0.56-fold). On the other hand, dogLM was the most efficient in carbosulfan sulfinamide and dibutylamine formation (2.7- and 1.4-fold to human, respectively). It is, however, necessary to emphasize that maximum differences between the lowest and highest rates were not conspicuously large; about 2-fold for the carbofuran pathway, 2.7-fold for carbosulfan sulfinamide and 6.2-fold for dibutylamine. These findings illustrate that in vitro screening of metabolite profiles and metabolic activities for the proper interpretation (and perhaps prediction) of animal model studies is desirable, because species differences in carbamate metabolism and disposition could affect the accuracy of the extrapolation approach. Noticeably, the ratio between intrinsic clearance values for carbofuran (activation) and carbosulfan sulfinamide (detoxification) clearly indicate that the most active microsomes in bioactivation compared to detoxification were ratLM (the ratio, 27.4) and Author's personal copy 218 K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 humanLM (24.4) followed by minipigLM (18.7) and monkeyLM (16.5), while rabbitLM (8.8), mouseLM (7.7) and dogLM (4.4) were proportionally the least active in carbosulfan activation. Interspecies differences in the carbosulfan metabolic pathway might be related to the variability in the constitutive expressions of CYP enzymes in the liver of investigated animal species. Moreover, a comprehensive evaluation of cytochrome P450 activities of the same seven mammalian species illustrated a wide variability between species [36]. Differences in efficiencies between rat, minipig, and human liver microsomes have been reported by Lang et al. [37] in the metabolism of triazine herbicides such as atrazine and ametryn. Also with other pesticides such as chlorpyrifos (organophosphorus insecticide) and carbofuran (carbamate insecticide), human liver microsomes showed lower rates of metabolism than mouse and rat liver microsomes [4,32]. In the case of profenofos, an organophosphorothiolate insecticide, humanLM were more efficient than ratLM and less efficient than mouseLM in profenofos bioactivation via desthiopropylation [12]. The 100-fold uncertainty factor (UF) is used to convert a noobserved-adverse-effect level (NOEAL) from an animal toxicity study to a safe value for human intake. Furthermore, this UF has to allow for a 10-fold interspecies difference (which is subdivided into a factor of 100.6 (4.0) for toxicokinetics and 100.4 (2.5) for toxicodynamics) and a 10-fold interindividual variation (which is divided equally into two subfactors each of 100.5 (3.16)) [38]. The International Program on Chemical Safety (IPCS) proposes the use of chemical-specific toxicological data instead of default assessment factors, whenever possible. In our results the highest animal to human differences in toxicokinetics for the carbofuran–metabolic pathway, carbosulfan sulfinamide and dibutylamine was 1.03-, 2.74-, and 1.81-fold, respectively. Toxicokinetics data for the active chemical moiety (carbofuran–metabolic pathway) will be valuable in further development of the proposed default subdivision of the usual uncertainty factor to quantitative toxicokinetic chemicalspecific assessment factors (CSAFs) [28]. It must be stressed here that we measured only the hepatic metabolism of carbosulfan; however, metabolism is often the most important factor contributing to interspecies differences in toxicokinetics. In order to obtain quantitative toxicokinetic data for comparisons between individuals or between animals and humans, human data are needed [27,39]. Moreover, risk assessment has to be carried out using quantitative chemical-specific data which will influence the toxicokinetics and toxicodynamics [40]. Our studies, although restricted to metabolic data from human and animal liver preparations, provide valuable quantitative carbosulfan-specific data for risk assessment, which suggest that interspecies differences, for carbosulfan active chemical moiety, in toxicokinetics are within the standard applied factor for species extrapolation in toxicokinetics. These results will be valuable in further defining the risks associated with exposure to carbosulfan. In conclusion, we have studied carbosulfan metabolism by seven mammalian liver microsomes including human and from the eight metabolites quantified, seven were identified with the advantage of using LC/MS/MS with comparison to reference standards. Carbosulfan was metabolized by initial oxidation of sulfur to yield carbosulfan sulfonamide, by the cleavage of the nitrogen–sulfur bond (N–S) to give carbofuran and dibutylamine, and by the oxidation of carbofuran to 3-hydroxycarbofuran, 3-hydroxy-7-phenolcarbofuran, 3-ketocarbofuran, and 3-keto-7phenolcarbofuran. These studies demonstrated some species differences in the in vitro pathways profile of carbosulfan biotransformation in rabbitLM, ratLM, and mouseLM, while dogLM illustrated excessive activity. Studies conducted in experimental animals for human risk assessment may be misleading due to species differences. Our quantitative data on interspecies differences make an important contribution to carbosulfan risk assessment by developing a chemical-specific adjustment factor for interspecies differences in toxicokinetics restricted to metabolic data by human and animal liver preparations. Further studies with a number of human individual and cytochrome P450 isoforms are on-going to accomplish these findings. Conflicts of interest None of the authors has a conflict of interest related to this study. Acknowledgements This work was funded by the Ministry of Education-supported position from Finnish Graduate School in Toxicology (ToxGS) and was supported by grants from The Academy of Finland and The Finnish Granting Agency for Technological Research and Innovation (TEKES). References [1] FAO/WHO Pesticide residues in food – 2003, Toxicological evaluations, World Health Organization, Geneva, 2003. [2] F. Oesch, M.E. Herrero, J.G. Hengstler, M. Lohmann, M. Arand, Metabolic detoxification: implications for thresholds, Toxicol. Pathol. 28 (2000) 382–387. [3] C. Sams, H.J. Mason, R. Rawbone, Evidence for the activation of organophosphate pesticides by cytochromes P450 3A4 and 2D6 in human liver microsomes, Toxicol. Lett. 116 (2000) 217–221. [4] J. Tang, Y. Cao, R.L. Rose, A.A. Brimfield, D. Dai, J.A. Goldstein, E. Hodgson, Metabolism of chlorpyrifos by human cytochrome P450 isoforms and human, mouse, and rat liver microsomes, Drug Metab. Dispos. 29 (2001) 1201–1204. [5] T.S. Poet, H. Wu, A.A. Kousba, C. Timchalk, In vitro rat hepatic and intestinal metabolism of the organophosphate pesticides chlorpyrifos and diazinon, Toxicol. Sci. 72 (2003) 193–200. [6] W.A. Kappers, R.J. Edwards, S. Murray, A.R. Boobis, Diazinon is activated by CYP2C19 in human liver, Toxicol. Appl. Pharmacol. 177 (2001) 68–76. [7] F.M. Buratti, M.T. Volpe, A. Meneguz, L. Vittozzi, E. Testai, CYP-specific bioactivation of four organophosphorothioate pesticides by human liver microsomes, Toxicol. Appl. Pharmacol. 186 (2003) 143–154. [8] F.M. Buratti, A. D’Aniello, M.T. Volpe, A. Meneguz, E. Testai, Malathion bioactivation in the human liver: the contribution of different cytochrome P450 isoform, Drug Metab. Dispos. 33 (2005) 295–302. [9] F.M. Buratti, E. Testai, Evidences for CYP3A4 autoactivation in the desulfuration of dimethoate by the human liver, Toxicology 241 (2007) 33–46. [10] K.A. Usmani, E.D. Karoly, E. Hodgson, R.L. Rose, In vitro sulfoxidation of thioether compounds by human cytochrome P450 and flavin-containing monooxygenase isoforms with particular reference to the CYP2C subfamily, Drug Metab. Dispos. 32 (2004) 333–339. [11] E. Mutch, F.M. Williams, Diazinon, chlorpyrifos and parathion are metabolised by multiple cytochromes P450 in human liver, Toxicology 224 (2006) 22–32. [12] K. Abass, P. Reponen, J. Jalonen, O. Pelkonen, In vitro metabolism and interaction of profenofos by human, mouse and rat liver preparations, Pestic. Biochem. Physiol. 87 (2007) 238–247. [13] K. Abass, P. Reponen, M. Turpeinen, J. Jalonen, O. Pelkonen, Characterization of diuron N-demethylation by mammalian hepatic microsomes and cDNAexpressed human cytochrome P450 enzymes, Drug Metab. Dispos. 35 (2007) 1634–1641. [14] C. Leoni, F.M. Buratti, E. Testai, The participation of human hepatic P450 isoforms, flavin-containing monooxygenases and aldehyde oxidase in the biotransformation of the insecticide fenthion, Toxicol. Appl. Pharmacol. 233 (2008) 343–352. [15] L.P. de Melo Plese, L.C. Paraiba, L.L. Foloni, L.R. Pimentel Trevizan, Kinetics of carbosulfan hydrolysis to carbofuran and the subsequent degradation of this last compound in irrigated rice fields, Chemosphere 60 (2005) 149–156. [16] A. Detomaso, G. Mascolo, A. Lopez, Characterization of carbofuran photodegradation by-products by liquid chromatography/hybrid quadrupole time-of-flight mass spectrometry, Rapid Commun. Mass Spectrom. 19 (2005) 2193–2202. [17] G.R. Chaudhry, A. Mateen, B. Kaskar, M. Sardessai, M. Bloda, A.R. Bhatti, S.K. Walia, Induction of carbofuran oxidation to 4-hydroxycarbofuran by Pseudomonas sp. 50432, FEMS Microbiol. Lett. 214 (2002) 171–176. [18] C. Soler, J. Mañes, Y. Picó, Determination of carbosulfan and its metabolites in oranges by liquid chromatography ion-trap triple-stage mass spectrometry, J. Chromatogr. A 1109 (2006) 228–241. [19] P.J. Marsden, E. Kuwano, T.R. Fukuto, Metabolism of carbosulfan [2,3-dihydro2,2-dimethylbenzofuran-7-yl (di-n-butylaminothio)methylcarbamate] in the rat and house fly, Pestic. Biochem. Physiol. 18 (1982) 38–48. [20] N. Umetsu, T.R. Fukuto, Alteration of carbosulfan [2,3-dihydro-2,2-dimethyl-7benzofuranyl(di-n-butylaminosulfenyl)methylcarbamate] in the rat stomach, J. Agric. Food Chem. 30 (1982) 555–557. Author's personal copy K. Abass et al. / Chemico-Biological Interactions 181 (2009) 210–219 [21] C. Tomlin, The Pesticide Manual, 12th ed., BCPC Publications, Berkshire, UK, 2000. [22] M.J. Trevisan, L.R.P. Trevizan, G.C. de Baptista, A.A. Franco, Metabolism of carbosulfan to carbofuran and to 3-hydroxy-carbofuran in orange leaves, Rev. Bras. Toxicol. 17 (2004) 17–22. [23] M.R. Iesce, M. della Greca, F. Cermola, M. Rubino, M. Isidori, L. Pascarella, Transformation and ecotoxicity in carbamic pesticides in water, Environ. Sci. Pollut. Res. Int. 13 (2006) 105–109. [24] M.H. Akhtar, The fate of insecticides in economic animals, in: D.H. Hutson, T.R. Roberts (Eds.), Progress in Pesticide Biochemistry and Toxicology, vol. 4, Wiley, Chichester, UK, 1985. [25] R.C. Gupta, Carbofuran toxicity, J. Toxicol. Environ. Health 43 (1994) 383. [26] A.J. Paine, A.G. Renwick, Validity and reliability of in vitro systems in safety evaluation, Environ. Toxicol. Pharmacol. 2 (1996) 207–212. [27] A. Falk-Filipsson, A. Hanberg, K. Victorin, M. Warholm, M. Wallén, Assessment factors—applications in health risk assessment of chemicals, Environ. Res. 104 (2007) 108–127. [28] WHO/IPCS, Chemical-specific adjustment factors for interspecies differences and human variability: guidance document for use of data in dose/concentration–response assessment. World Health Organization, Geneva, 2005. [29] O. Pelkonen, E.H. Kaltiala, T.K.I. Larmi, N.T. Karki, Cytochrome P 450 linked monooxygenase system and drug induced spectral interactions in human liver microsomes, Chem. Biol. Interact. 9 (1974) 205–216. [30] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding, Anal. Biochem. 72 (1976) 248–254. [31] J.H. Kim, C. Chang, J. Moon, H. Lee, J. Lee, H. Park, B. Park, E. Hwang, H. Lee, K. Liu, Comparative metabolism of insecticide carbosulfan by human, mouse and rat liver microsomes, Toxicol. Lett. 172 (2007) S211–S1211. View publication stats 219 [32] K.A. Usmani, E. Hodgson, R.L. Rose, In vitro metabolism of carbofuran by human, mouse, and rat cytochrome P450 and interactions with chlorpyrifos, testosterone, and estradiol, Chem. Biol. Interact. 150 (2004) 221– 232. [33] C. Soler, B. Hamilton, A. Furey, K.J. James, J. Manes, Y. Pico, Liquid chromatography quadrupole time-of-flight mass spectrometry analysis of carbosulfan, carbofuran, 3-hydroxycarbofuran, and other metabolites in food, Anal. Chem. 79 (2007) 1492–1501. [34] Y. Kinoshita, T.R. Fukuto, Insecticidal properties of N-sulfonyl derivatives of propoxur and carbofuran, J. Agric. Food Chem. 28 (1980) 1325– 1327. [35] T.R. Roberts, D.H. Hutson, P.J. Jewess, Metabolic Pathways of Agrochemicals. Part 2: Insecticides and Fungicides, Royal Society of Chemistry, Cambridge, 1999, p. 1475. [36] M. Turpeinen, C. Ghiciuc, M. Opritoui, L. Tursas, O. Pelkonen, M. Pasanen, Predictive value of animal models for human cytochrome P450 (CYP)-mediated metabolism: a comparative study in vitro, Xenobiotica 37 (2007) 1367– 1377. [37] D. Lang, D. Criegee, A. Grothusen, R.W. Saalfrank, R.H. Bocker, In vitro metabolism of atrazine, terbuthylazine, ametryne, and terbutryne in rats, pigs, and humans, Drug Metab. Dispos. 24 (1996) 859–865. [38] A.G. Renwick, N.R. Lazarus, Human variability and noncancer risk assessment—an analysis of the default uncertainty factor, Regul. Toxicol. Pharm. 27 (1998) 3–20. [39] B. Meek, A. Renwick, C. Sonich-Mullin, Practical application of kinetic data in risk assessment—an IPCS initiative, Toxicol. Lett. 138 (2003) 151– 160. [40] K. Walton, J.L. Dorne, A.G. Renwick, Uncertainty factors for chemical risk assessment: interspecies differences in glucuronidation, Food Chem. Toxicol. 39 (2001) 1175–1190.