Advertisement
Research ArticleArticle

Identification of 2-Aminothiazolobenzazepine Metabolites in Human, Rat, Dog, and Monkey Microsomes by Ion-Molecule Reactions in Linear Quadrupole Ion Trap Mass Spectrometry

Minli Zhang, Ryan Eismin, Hilkka Kenttämaa, Hui Xiong, Ye Wu, Doug Burdette and Rebecca Urbanek
Drug Metabolism and Disposition March 2015, 43 (3) 358-366; DOI: https://doi.org/10.1124/dmd.114.061978

Abstract

2-Aminothiazolobenzazepine (2-ATBA), 7-[(1-methyl-1H-pyrazol-4-yl)methyl]-6,7,8,9-tetrahydro-5H-[1,3]thiazolo[4,5-h][3]benzazepin-2-amine, is a D2 partial agonist that has demonstrated antipsychotic effects in a rodent in vivo efficacy model. The metabolite profile showed that 2-ATBA is mainly metabolized by oxidation. However, identification of the oxidation site(s) in the 2-aminothiazole group presents a challenge for the traditional metabolite identification methods such as liquid chromatography/mass spectrometry and NMR due to the lack of unique tandem mass spectrometry fragmentation patterns for ions with the 2-aminothiazole group oxidized at different sites and the lack of stability for purification or reference standard synthesis. We describe the characterization of the oxidized heteroatoms of the 2-aminothiazole group via gas-phase ion-molecule reactions (GPIMR) in a modified linear quadrupole ion trap mass spectrometer. The GPIMR reagents used were dimethyl disulfide, tert-butyl peroxide, and tri(dimethylamino)borane. Each reagent was introduced into the ion trap through the helium line and was allowed to react with the protonated metabolites. The ionic ion-molecule reaction products and their fragmentation profiles were compared with the profiles of the ionic ion-molecule reaction products of protonated reference compounds that had specific heteroatom functionalities. The oxidized 2-aminothiazole metabolite of 2-ATBA showed a similar GPIMR profile to that of the reference compounds with a tertiary N-oxide functionality and distinct from the profiles of the reference compounds with N-aryl hydroxylamine, nitroso, or pyridine N-oxide functionalities. This study demonstrates the feasibility of fingerprinting the chemical nature of oxidized nitrogen functional groups via GPIMR profiling for metabolite structure elucidation.

Introduction

Dopamine D2 partial agonists are effective in regulating dopamine levels in the neurons at hyperdopaminergic or hypodopaminergic state without significant impact on other normal dopaminergic pathways, thus demonstrating antipsychotic efficacy in schizophrenia with fewer side effects (Kikuchi et al., 1995; Davies et al., 2004; Kinghorn and McEvoy, 2005; Miller et al., 2007). After the discovery of AZD9821 [(R)-N6-ethyl-6,7-dihydro-5H-indeno[5,6-d]thiazole-2,6-diamine] as a D2 partial agonist candidate, a new benzazepine-based 2-aminothiazole series was developed to improve the D2 partial agonist activity and selectivity (Urbanek et al., 2013). Among the compounds in the series, the active 2-aminothiazolobenzazepine compound 2-ATBA [2-aminothiazolobenzazepine, 7-[(1-methyl-1H-pyrazol-4-yl)methyl]-6,7,8,9-tetrahydro-5H-[1,3]thiazolo[4,5-h][3]benzazepin-2-amine] was identified (Fig. 1) with good in vitro and in vivo absorption, distribution, metabolism, and excretion profiles that demonstrated in vivo efficacy in the rat amphetamine-induced locomotor activity model (Urbanek et al., 2013).

Fig. 1.
Fig. 1.

Chemical structures. (1) Nitrosobenzene. (2) Pyridine N-oxide. (3) Trimethylamine N-oxide. (4) 4-Tert-butyl-N-aryl hydroxylamine.

In this study, we investigated the in vitro metabolic stability and metabolite profile of 2-ATBA in liver microsomes and hepatocytes from humans, rats, dogs, and cynomolgus monkeys. Because there are three heteroatoms in the aminothiazole moiety, pinpointing the site of oxidation has proved to be very challenging due to the lack of a unique tandem mass spectrometry (MS/MS) fragmentation pattern from the molecule. Furthermore, one of the oxidized 2-ATBA metabolites was labile, which prevented the purification or chemical synthesis of the metabolite reference standard. Therefore, an alternative approach was necessary for the structural characterization.

Traditional methods for metabolite structural elucidation such as NMR, Fourier transform infrared spectroscopy, and x-ray crystallography are powerful but require a large quantity of highly purified sample. In the past decade, gas-phase ion-molecule reactions (GPIMR) have been effectively used for the identification of functional groups (Brodbelt, 1997; Watkins et al., 2004, 2005; Eberlin, 2006; Fu et al., 2012; Osburn and Ryzhov, 2013). Several neutral reagents for GPIMR such as dimethyl disulfide (DMDS), tert-butyl peroxide (TBP), and tri(dimethylamino)borane (TDMAB) have been developed to identify oxidized heteroatom functionalities (Watkins et al., 2005; Duan et al., 2008, 2009; Fu et al., 2012). These GPIMR reagents react with functional groups of the protonated analyte molecules in the ion trap and generate unique products that can be further fragmented by collision-induced dissociation (CID) to generate characteristic product ions or neutral molecules specific to the functional groups. Because each reagent can react with different oxidized nitrogen functionalities and generate different reaction products or fragmentation patterns, it is hard to use a single reagent to identify a specific functionality. Therefore, our focus was on the use of multiple reagents to develop fingerprints specific to individual oxidized nitrogen functionalities. In this study, we reported the use of these three neutral reagents together with a variety of reference compounds to develop GPIMR profiles specific to oxidized nitrogen functionalities of the reference compounds that could relate to the functionalities of the unknown metabolite. The goal was to determine the structural and functional characteristics of the 2-ATBA oxidized metabolite.

Materials and Methods

The 2-ATBA compound was synthesized by AstraZeneca (Wilmington, DE) (Urbanek et al., 2013). The 2-ATBA azepine N-oxide metabolite (M2) reference compound was prepared by oxidation of the tertiary amino precursor (2-ATBA) with hydrogen peroxide in methanol. The structure was confirmed by NMR and liquid chromatography with mass spectrometry (LC-MS). The purity was determined by liquid chromatography with UV detection.

The GPIMR reagents DMDS, TBP, and TDMAB; the reference compounds (1) nitrosobenzene, (2) pyridine N-oxide, and (3) trimethylamine N-oxide; the chemicals hydrogen peroxide, potassium phosphate, dimethylsulfoxide, TiCl3 in hydrochloric acid solution, LC-MS-grade formic acid, ammonium formate, NADPH, magnesium chloride, and other chemicals were purchased from Sigma-Aldrich (St. Louis, MO) and were used as received. The reference compound 4-tert-butyl-N-aryl hydroxylamine (4) was obtained from the AstraZeneca compound collection. High-pressure liquid chromatography (HPLC) water, methanol, potassium carbonate, and acetonitrile were purchased from Thermo Fisher Scientific (Pittsburg, PA). The liver microsomes from human (HLM), rat (RLM), dog (DLM), and cynomolgus monkey (CLM), as well as the cDNA expressed CYP450 and flavin-containing monooxygenase (FMO) isozymes were purchased from Corning Life Sciences (Woburn, MA). The fresh hepatocytes from male Sprague-Dawley rats and beagle dogs were prepared at AstraZeneca using standard techniques. Cryopreserved human (from three donors) and cynomolgus monkey hepatocytes were purchased from Life Technologies (Grand Island, NY).

Microsome and Recombinant Isozyme Incubations.

The incubation solution contained 1 μM 2-ATBA, 0.5 mg/ml of liver microsomes, and 1 mM NADPH in 100 mM potassium phosphate buffer, pH 7.4. For metabolite characterization, 10 µM 2-ATBA was used in the incubation. For metabolic stability assays in recombinant CYP450 isozymes, 50 pM of CYP450 isozymes were used in the incubation. For metabolic stability assays using FMO, 125 µg/ml of FMO isozyme in 100 mM glycine buffer, pH 9.5, were used in the incubation. The mixture was preincubated at 37°C for 3 minutes. The reaction was initiated by the addition of NADPH. Aliquots of the incubate were removed at 0, 5, 10, 15, 20, and 25 minutes and were added to equal volumes of acetonitrile/methanol (1:1) containing 0.1% formic acid (v/v) to stop the enzymatic reaction. After centrifugation for 10 minutes at 3500 rpm, the supernatant was stored at −80°C until the analyses.

Hepatocytes Metabolic Stability Assay.

The incubation solution contained 1 μM 2-ATBA and 1 × 106 cells/ml of hepatocytes. The incubations were performed at 37°C in Williams E medium, containing 1% GIBCO insulin transferrin selenium solution, 2 mM l-glutamine, and 25 mM HEPES. Aliquots of the suspension were removed at 0, 5, 10, 15, 20, 25, and 30 minutes and were added to three volumes of ice-cold acetonitrile to stop the metabolic reactions. After 10 minutes of centrifugation at 3500 rpm, the supernatant was stored at −80°C until analyses.

LC-MS Analysis.

For the in vitro metabolic stability assays, 2-ATBA in the incubation medium was quantified by LC-MS. The separation was performed using a Shimadzu VP HPLC system (Columbia, MD) with a Synergi Max-RP column (3.5 µm, 3 × 50 mm; Phenomenex, Torrance, CA) coupled to a Micromass Ultima triple quadrupole mass spectrometer (Waters, Milford, MA). The compounds were ionized in positive electrospray ionization mode and analyzed using multiple reaction monitoring. The mobile phases used were 0.1% formic acid in HPLC water (A) and 10% (v/v) methanol in acetonitrile (B). The gradient started at 1.5 ml/min with 100% A for 0.3 minutes, was linearly increased to 95% B in 1.2 minutes, and held at 95% B for 0.1 minutes before returning to 100% A.

Metabolite Profile Analysis.

The in vitro metabolite profile of 2-ATBA was analyzed using an orbitrap mass spectrometer (Thermo Scientific, San Jose, CA) under positive electrospray ionization mode with source voltage at 4.5 kV, capillary temperature at 350°C, and sheath gas flow rates at 40 arbitrary units. The orbitrap resolution was set at 7500 with a scan range of 50–850 m/z (mass to charge ratio). Data-dependent MS2 and MS3 fragmentation experiments were performed in the ion trap where the normalized collision energy was set at 25% with 0.25 activation q and 30-millisecond activation time. The metabolites were separated by UPLC (Waters, Milford, MA) at 0.2 ml/min in a BEH C18 column (2.1 × 100, 1.7 µm). The eluting solvent consisted of 10 mM ammonium formate and 5% acetonitrile in water (A1) and 0.1% formic acid in acetonitrile (B1). The gradient started with 100% A1, then B1 was increased from 0% to 10% in 5 minutes, from 10% to 40% in 4 minutes, from 40% to 95% in 3 minutes, and kept at 95% for 1 minute before equilibration at 100% A1 for 3 minutes.

TiCl3 Reduction.

Aliquots (100 µl) of DLM incubates or the synthetic M2 standard reference (1 µM in 10% acetonitrile, v/v) were allowed to react with 3 µl of TiCl3 solution at 5°C or 50°C for 3 hours. At the end of the reaction, the samples were directly analyzed by LC-MS to assess the reduction of the oxidized 2-ATBA products (M1 and M2).

Modification of Linear Quadrupole Ion Trap Ion Trap for GPIMR.

The LTQ XL linear quadrupole ion trap mass spectrometer with Xcalibur 2.1 software (Thermo Scientific) coupled with the Accela Ultrapressure LC System (Thermo Scientific) was modified as previously described elsewhere (Habicht et al., 2008). Briefly, a gas line was diverted from the helium gas source with a three-way check valve. A syringe pump for the introduction of the GPIMR reagent was connected to this line with a tee. The other end of this GPIMR reagent line was connected back to the main helium line with a three-way check valve before entering the ion trap (Fig. 2). The diverted line was heated to facilitate the evaporation of the GPIMR reagents, which were then carried into the ion trap by the helium gas. The flow of GPIMR reagents into the ion trap was constant and was equilibrated for at least 1 hour before the GPIMR experiment.

Fig. 2.
Fig. 2.

Schematic illustration of the modified linear quadrupole ion trap used for the GPIMR experiments.

The GPIMR experiment was conducted under the same LC conditions as those used for metabolite profiling. Multiple events were used in setting up the GPIMR experiment in the Xcalibur software where the event 1 was set for the full scan. In events 2, 3, and 4, the protonated ion of interest was isolated for collision with a reagent in the trap. The collision energy for the reaction was set at 0%, q at 0.25, and the reaction time at 30, 100, and 1000 milliseconds for events 2, 3, and 4, respectively. When needed, an additional event was added to increase the reaction time to 5000 milliseconds. In the last experiment, an additional event was set up where the ion-molecule reaction product ion of interest was isolated for CID.

Results

In Vitro Metabolism of 2-ATBA.

The intrinsic clearance of 2-ATBA in the microsomes and hepatocytes was low in HLM, RLM, and CLM, and moderate in DLM (Table 1). When scaled to in vivo, the predicted hepatic clearances were around or below 30% liver blood flow, except in the dog. The predicted in vivo clearance from liver microsome or hepatocyte data was similar, suggesting that the clearance was mainly through metabolism by phase I enzymes such as CYP450.

View this table:
TABLE 1

Intrinsic clearance of 2-ATBA in the liver microsomes and hepatocytes

Profile of 2-ATBA Metabolite.

Metabolite profile analyses showed that the major peak corresponded to the parent compound in liver microsome incubates, with two small metabolite peaks M1 and M2 at 5.97 minutes and 6.48 minutes, respectively (Fig. 3). Significantly more M2 metabolite was found in HLM and RLM whereas a larger M1 proportion was found in DLM and CLM (Fig. 4).

Fig. 3.
Fig. 3.

Orbitrap total ion chromatograms of protonated 2-ATBA metabolites from the incubations of (A) HLM, (B) RLM, (C) DLM, and (D) CLM. Mass defect filters around the parent and its metabolite mass range were applied in generating the total ion chromatograms.

Fig. 4.
Fig. 4.

Orbitrap selected ion chromatograms of protonated 2-ATBA oxidized metabolites. The selected ion chromatograms were generated from ions of m/z 314.1428 for 2-ATBA reference standard and m/z 330.1375 for the metabolites M1 and M2, respectively, with mass tolerance set at 10 ppm. (A) 2-ATBA standard. (B) M2 synthetic standard. (C) HLM incubates. (D) RLM incubates. (E) DLM incubates. (F) CLM incubates.

CYP450 and FMO Isozyme Involvement.

The recombinant CYP450 and FMO3 incubation results showed that CYP2D6 and CYP3A4 were the major contributors to 2-ATBA metabolism (Fig. 5). CYP2D6 was more efficient than CYP3A4 in metabolizing 2-ATBA. When we used the extrapolating method by Rodrigues (1999), CYP3A4 was estimated to contribute 67% of 2-ATBA metabolism and CYP 2D6 33%. Only M2 was detectable in the CYP3A4 and CYP2D6 incubates. Although the contribution of recombinant FMO3 was not quantifiable, a minor peak corresponding to M2 was detectable in incubates.

Fig. 5.
Fig. 5.

Metabolic clearance of 2-ATBA by individual isozymes. The intrinsic clearance (CLint) for CYP450 isozymes is expressed as µl/min/pmol, and the CLint for FMO is expressed as µL/min/mg protein.

Characterization of M1 and M2.

Both metabolites M1 and M2 generated a protonated molecule of m/z 330.1389 which is 16 atomic mass units higher than that of the protonated parent compound, suggesting the addition of an extra oxygen atom (Table 2). Upon CID, both protonated M1 and M2 generated a fragment ion of m/z 95.0597 (Fig. 6), which is characteristic of the N-methylpyrazolylmethylene group. Both metabolites also generated fragment ions of m/z 234.0693 and m/z 248.0854 as the result of the loss of methylpyrazol and methylpyrazolylmethylene groups, respectively, suggesting that the oxidation occurred in the aminothiazolobenzazepine core.

View this table:
TABLE 2

Metabolite ion and fragmentation summary

Fig. 6.
Fig. 6.

Fragmentation patterns of protonated metabolites (A) M1 and (B) M2. The high-resolution data were collected at 7500 resolution for the fragmentation of protonated M1 and M2 ions of m/z 330 at 25% CID.

The fragmentation of protonated M1 also generated a fragment ion of m/z 313.1389, which is consistent with the loss of a hydroxyl group (HO•), indicating that there are no hydrogen atoms near the eliminated hydroxyl group for formation of a water molecule as the leaving group. Therefore, the oxidation to generate M1 likely occurred in the aminothiazole moiety. Based on the fragmentation patterns, however, we could not unambiguously assign the site of oxidation in the thiazole nitrogen or amino nitrogen atom. The CID fragmentation of protonated M2 generated an ion of m/z 312.1279, consistent with the loss of H2O, which suggests the availability of a hydrogen atom near the cleaving hydroxyl group. We thus assigned the oxidation site in M2 at the azepine nitrogen. The structure of this azepine N-oxide metabolite (M2) was further confirmed by comparing its retention time and the fragmentation patterns to those of the synthetic reference compound. The CID fragment ion assignments for protonated M1 and M2 are summarized in Fig. 6.

The TiCl3 reduction products of M1 and M2 were analyzed by LC-MS in the orbitrap. After the reaction on ice, M1 and M2 from DLM as well as the M2 synthetic reference compound were found to be reduced by TiCl3. Under this reduction condition, only N-oxidation product was expected to be reduced, and sulfur oxidized product should be stable. These results suggest that the sites of M1 oxidations are in the nitrogen rather than on the sulfur atom.

Identification of the Oxidation Site in M1 by GPIMR.

As the TiCl3 reduction results had suggested the oxidation of the nitrogen atom, our focus was then on the nitrogen atom being oxidized in M1. Three GPIMR reagents were used to investigate the ion-molecule reaction products of the protonated M1. The GPIMR reaction products and their fragmentation profiles were compared with those obtained for reference compounds that have unique 1) nitroso, 2) pyridine N-oxide, 3) tertiary amine N-oxide, or 4) N-aryl hydroxylamine functionalities. The observed fragmentation patterns for these different oxidized nitrogen functionalities and the proposed mechanisms of GPIMR product formations are summarized in Tables 3 and 4, respectively.

View this table:
TABLE 3

Comparison of GPIMR adducts and fragments of the protonated reference compounds and the M2 metabolite

View this table:
TABLE 4

Summary of proposed functional group-specific GPIMR mechanisms

  1. GPIMR in the presence of DMDS. Upon reaction with DMDS in the ion trap, both protonated nitrobenzene (1) and N-aryl hydroxylamine (4) generated a [MH+31]+ ion-molecule reaction product (Fig. 7, A and B). The proposed mechanisms for [MH+31]+ product formation are summarized in schemes 1 and 2 (Table 4). This [MH+31]+ product was not detected for reactions of DMDS with protonated pyridine N-oxide (2) or tertiary N-oxides (3). Neither protonated M2 nor M1 generated [MH+31]+ product (Fig. 7, C and D, respectively), suggesting that the oxidized nitrogen functionality in M1 is different from that of nitroso or N-aryl hydroxylamine.

    Fig. 7.
    Fig. 7.

    MS analyses of products from ion-molecule reaction between DMDS and (A) reference compound 1, (B) compound 4, (C) M2, and (D) M1, respectively.

  2. GPIMR in the presence of TBP. Upon reaction with TBP, the protonated reference compound (1), which has a nitroso functionality, generated a unique [MH+56]+ product (Fig. 8A), which was not observed for the TBP reaction with other protonated oxidized nitrogen atoms tested. The proposed mechanism for the formation of this [MH+56]+ product includes an initial nucleophilic attack of the peroxide moiety of the TBP to the nitroso nitrogen followed by a rearrangement to form the [MH+56]+ product (Table 4, scheme 3). All tested compounds with an oxidized nitrogen within the aromatic ring or attached to the aromatic ring formed [MH+145]+ or [MH+146]+ product (Fig. 8, A–C) when allowed to react with TBP (Table 3). However, these products were not detected for the TBP reaction with aliphatic tertiary nitrogen oxides (3 and M2). The protonated N-aryl hydroxylamine also generated a [MH+220]+ product (Fig. 8B), which was not observed for the TBP reaction with the other protonated reference compounds or M1 metabolite tested. Elucidation of the mechanisms for formation of this [MH+220]+ product (scheme 5) will require further investigation. Although the formation of the [MH+145]+ product in the TBP reaction with protonated M1 (Fig. 8D) is consistent with the ion-molecule reaction products observed for the reference compounds 2 and 4 in which the oxidized nitrogen is within or attached to an aromatic ring (Table 4, schemes 4), the lack of [MH+56]+ or [MH+220]+ products for M1 demonstrates that the functionality of its oxidized nitrogen atom is different from those of nitroso (1) or N-aryl hydroxylamine (4) functionalities.

    Fig. 8.
    Fig. 8.

    MS analyses of products from ion-molecule reaction between TBP and (A) reference compound 1, (B) compound 2, (C) compound 4, and (D) M1, respectively.

  3. GPIMR in the presence of TDMAB. TDMAB formed a [MH+98]+ ion-molecule reaction product with the all protonated reference compounds tested (Table 3) except nitrosobenezene (1). The [MH+98]+ product was likely formed as the result of proton transfer from the protonated oxidized nitrogen to one of the dimethylamino functionalities in TDMAB followed by nucleophilic attack of the oxidized nitrogen at the boron and the elimination of a dimethylamine (Table 4, schemes 6 and 7). The fragmentation of the [MH+98]+ product varied depending on the type of oxidized nitrogen atom. The [MH+98]+ product of N-aryl hydroxylamine (4) was stable in CID fragmentation, and no fragments were detected. In pyridine N-oxide (2), the [MH+98]+ product gave a unique m/z 115 fragment ion (Fig. 9A) as the result of nitrogen-oxygen bond cleavage and charge migration to the bis(dimethylamino)hydroxylborane moiety to form a N-[(dimethylamino)(hydroxy)boranyl]-N-dimethyl iminium ion (Table 4, scheme 6). In protonated aliphatic tertiary N-oxides (3 and M2), CID fragmentation of [MH+98]+ product led to the cleavage of the nitrogen-oxygen bond with the charge retained in the nitrogen atom and a neutral loss of 116 atomic mass units (NL116), which corresponds to the bis(dimethylamino)hydroxylborane group (Fig. 9, B and C; Table 4, scheme 7). Upon reaction with TDMAB, the protonated metabolite M1 also formed the [MH+98]+ ion-molecule reaction product, which generated a NL116 upon CID (Fig. 9D), suggesting that the charge was retained upon a nitrogen-oxygen bond cleavage, consistent with the fragmentation of the [MH+98]+ ion-molecule reaction product of an aliphatic tertiary N-oxide (3) with TDMAB.

Fig. 9.
Fig. 9.

MS and MS2 analyses of products from ion-molecule reaction between TDMAB and reference (A) compound 1, (B)compound 3, (C) M2, and (D) M1, respectively.

The overall GPIMR profile results indicate that the oxidized functionality of the 2-aminothiazole oxide in M1 is more likely to be a tertiary alkyl N-oxide or pyridine N-oxide than nitrobenzene or N-aryl hydroxylamine. Therefore, the site of oxidation in M1 was assigned to the 3-nitrogen atom. The proposed in vitro oxidative metabolic pathway of 2-ATBA is summarized in Fig. 10.

Fig. 10.
Fig. 10.

Proposed 2-ATBA oxidation pathway in the liver microsome incubations

Discussion

Results from this study show that 2-ATBA is relatively stable in human, rat, and monkey liver microsome and hepatocyte incubations. The intrinsic clearance in dog microsome and hepatocytes was moderate. Oxidation in the azepine and aminothiazole nitrogen atoms appeared to be the major metabolic pathway. The M1 metabolite was more predominant in dog and cynomolgus monkey microsome incubates whereas significantly more M2 was detected in human and rat microsome incubates. CYP3A and CYP2D6 appeared to be the major enzymes responsible for these oxidations even though only M2 was detected in these CYP incubations. The isozymes responsible for the formation of M1 remain to be identified.

The site of oxidation in M2 was assigned by the MS/MS fragmentation patterns of the protonated molecule in comparison with the patterns of the protonated synthetic N-oxide reference compound. MS fragmentation patterns observed for protonated M1 suggest that the oxidation had occurred in the aminothiazole. However, the MS/MS fragmentation analysis was unable to differentiate the oxidation of the three heteroatoms in the aminothiazole of M1. The instability of M1 made it challenging to identify the site of oxidation by traditional metabolite purification or standard reference syntheses. This report describes the use of LC-MS combined with diagnostic GPIMR for the structure elucidation.

Because the reduction of an N-oxidized compound by TiCl3 in solution can be achieved at 5°C whereas the reduction of sulfoxide requires heating, adjusting the reaction temperature enables the control of TiCl3 reduction in the sulfoxide or the N-oxidation groups (Kulanthaivel et al., 2004). The observation of M1 being reduced by TiCl3 at 5°C suggested that the site of oxidation was on one of the nitrogen atoms, but this reaction could not identify which nitrogen in the aminothiazole was oxidized.

The 2-N-hydroxyl aminothiazole resulting from oxidation at the 2-amino group of the 2-aminothiozole moiety in 2-ATBA could form an electrophilic intermediate through the β-elimination of water (Kalgutkar et al., 2007). This hydroxylation would be similar to N-hydroxyl aryl amino functionalities, some of which have been reported to be positive in Ames testing (McCarren et al., 2011). Some N-aryl hydroxylamine have also been shown to be liable for toxicities associated to their bioactivation to generate the electrophilic intermediates that could react with cellular thiols to form sulfinamide adducts (Harrison and Jollow, 1986; Liu et al., 2008; Siraki, 2013). Thus, identifying which nitrogen is the site of oxidation in M1 is also important in assessing the potential for genotoxicity risk.

To determine which of the nitrogen atoms in M1 was oxidized, three different GPIMR reagents were used to compare the reaction profiles of M1 with those of the reference compounds (1–4). The protonated compounds with oxidized nitrogen groups formed unique ion-molecule products upon reactions with specific GPIMR reagents, and the reaction products generated diagnostic fragmentation patterns upon CID. The oxidized nitrogen in metabolite M1 is different from that in nitroso (1) and N-aryl hydroxylamine (4) because the [MH+31]+ product was not formed by the protonated M1 upon reaction with DMDS. Furthermore, the protonated M1 did not form the [MH+56]+ or [MH+220]+ ion-molecule reaction product with TBP whereas the reference compounds 1 and 4 did. The GPIMR profiles of protonated M1 were more consistent with the profiles of the protonated pyridine N-oxide (2) and aliphatic tertiary N-oxide (3). Further confirmation was provided by the fragmentation of the M1-TDMAB reaction product [MH+98]+ that generated NL116, which is characteristic of a product of TDMAB reaction with protonated tertiary N-oxides (3 and M2). Therefore, the site of oxidation in M1 was assigned to the thiazole-nitrogen.

Conclusion

This study demonstrates the potential and feasibility of fingerprinting the chemical nature of an oxidized nitrogen functional group via multireagent GPIMR profiling. Reacting with a series of diagnostic reagents, each type of oxidized nitrogen functional group tested generated diagnostic GPIMR patterns. Although no single GPIMR could differentiate between different types of oxidized nitrogen functionalities, using these three reagents together could uniquely fingerprint these functional groups. The GPIMR profile of the 2-aminothiazole oxidation metabolite (M1) was more consistent with profile of the aliphatic N-oxide and was clearly distinct from N-aryl hydroxylamine, suggesting that the chemical properties and potentially the biologically reactive profiles of M1 are more aligned to the aliphatic tertiary nitrogen. Our preliminary results also showed that GPIMR has the ability to augment traditional metabolite ID methods to quickly identify hard-to-characterize metabolites. Further work with a broader set of neutral reagents and functional groups could potentially expand the utility of this technique in metabolite structure elucidations.

Acknowledgments

The authors thank Cindy Shen, Abdul Conteh, Bernard Lanoue, and Christina Resuello for technical assistance with the in vitro assays.

Authorship Contributions

Participated in research design: Zhang, Eismin, Kenttämaa.

Conducted experiments: Zhang, Eismin.

Contributed new reagents or analytic tools: Xiong, Wu, Urbanek.

Performed data analysis: Zhang, Eismin, Kenttämaa, Burdette.

Wrote or contributed to the writing of the manuscript: Zhang, Eismin, Kenttämaa, Wu, Burdette, Urbanek.

Footnotes

Abbreviations

2-ATBA
2-aminothiazolobezazepine, 7-[(1-methyl-1H-pyrazol-4-yl)methyl]-6,7,8,9-tetrahydro-5H-[1,3]thiazolo[4,5-h][3]benzazepin-2-amine
AZD9821
[(R)-N6-ethyl-6,7-dihydro-5H-indeno[5,6-d]thiazole-2,6-diamine]
CID
collision-induced dissociation
CLM
cynomolgus monkey liver microsome
DLM
dog liver microsome
DMDS
dimethyl disulfide
FMO
flavin-containing monooxygenase
GPIMR
gas-phase ion-molecule reactions
HLM
human liver microsomes
HPLC
high-pressure liquid chromatography
LC-MS
liquid chromatography with mass spectrometry
MS/MS
tandem mass spectrometry
m/z
mass to charge ratio
NL116
neutral loss of 116 atomic mass units
RLM
rat liver microsomes
TBP
tert-butyl peroxide
TDMAB
tri(dimethylamino)borane
UPLC
ultraperformance liquid chromatography

References

  1. ↵
    1. Brodbelt JS
    (1997) Analytical applications of ion-molecule reactions. Mass Spectrom Rev 16:91–110 DOI: 10.1002/(SICI)1098-2787(1997)16:2<91::AID-MAS3>3.0.CO;2-4.
    OpenUrlCrossRef
  2. ↵
    1. Davies MA,
    2. Sheffler DJ, and
    3. Roth BL
    (2004) Aripiprazole: a novel atypical antipsychotic drug with a uniquely robust pharmacology. CNS Drug Rev 10:317–336.
    OpenUrlPubMed
  3. ↵
    1. Duan P,
    2. Gillespie TA,
    3. Winger BE, and
    4. Kenttämaa HI
    (2008) Identification of the aromatic tertiary N-oxide functionality in protonated analytes via ion/molecule reactions in mass spectrometers. J Org Chem 73:4888–4894.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Duan P,
    2. Fu M,
    3. Gillespie TA,
    4. Winger BE, and
    5. Kenttämaa HI
    (2009) Identification of aliphatic and aromatic tertiary N-oxide functionalities in protonated analytes via ion/molecule and dissociation reactions in an FT-ICR mass spectrometer. J Org Chem 74:1114–1123.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Eberlin MN
    (2006) Structurally diagnostic ion/molecule reactions: class and functional-group identification by mass spectrometry. J Mass Spectrom 41:141–156.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Fu M,
    2. Duan P,
    3. Gao J, and
    4. Kenttämaa HI
    (2012) Ion-molecule reactions for the differentiation of primary, secondary and tertiary hydroxyl functionalities in protonated analytes in a tandem mass spectrometer. Analyst (Lond) 137:5720–5722.
    OpenUrlCrossRef
  7. ↵
    1. Habicht SC,
    2. Vinueza NR,
    3. Archibold EF,
    4. Duan P, and
    5. Kenttämaa HI
    (2008) Identification of the carboxylic acid functionality by using electrospray ionization and ion-molecule reactions in a modified linear quadrupole ion trap mass spectrometer. Anal Chem 80:3416–3421.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Harrison JH Jr. and
    2. Jollow DJ
    (1986) Role of aniline metabolites in aniline-induced hemolytic anemia. J Pharmacol Exp Ther 238:1045–1054.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Kalgutkar AS,
    2. Driscoll J,
    3. Zhao SX,
    4. Walker GS,
    5. Shepard RM,
    6. Soglia JR,
    7. Atherton J,
    8. Yu L,
    9. Mutlib AE,
    10. Munchhof MJ,
    11. et al.
    (2007) A rational chemical intervention strategy to circumvent bioactivation liabilities associated with a nonpeptidyl thrombopoietin receptor agonist containing a 2-amino-4-arylthiazole motif. Chem Res Toxicol 20:1954–1965.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Kikuchi T,
    2. Tottori K,
    3. Uwahodo Y,
    4. Hirose T,
    5. Miwa T,
    6. Oshiro Y, and
    7. Morita S
    (1995) 7-(4-[4-(2,3-Dichlorophenyl)-1-piperazinyl]butyloxy)-3,4-dihydro-2(1H)-quinolinone (OPC-14597), a new putative antipsychotic drug with both presynaptic dopamine autoreceptor agonistic activity and postsynaptic D2 receptor antagonistic activity. J Pharmacol Exp Ther 274:329–336.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Kinghorn WA and
    2. McEvoy JP
    (2005) Aripiprazole: pharmacology, efficacy, safety and tolerability. Expert Rev Neurother 5:297–307.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Kulanthaivel P,
    2. Barbuch RJ,
    3. Davidson RS,
    4. Yi P,
    5. Rener GA,
    6. Mattiuz EL,
    7. Hadden CE,
    8. Goodwin LA, and
    9. Ehlhardt WJ
    (2004) Selective reduction of N-oxides to amines: application to drug metabolism. Drug Metab Dispos 32:966–972.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Liu L,
    2. Wagner CR, and
    3. Hanna PE
    (2008) Human arylamine N-acetyltransferase 1: in vitro and intracellular inactivation by nitrosoarene metabolites of toxic and carcinogenic arylamines. Chem Res Toxicol 21:2005–2016.
    OpenUrlCrossRefPubMed
  14. ↵
    1. McCarren P,
    2. Bebernitz GR,
    3. Gedeck P,
    4. Glowienke S,
    5. Grondine MS,
    6. Kirman LC,
    7. Klickstein J,
    8. Schuster HF, and
    9. Whitehead L
    (2011) Avoidance of the Ames test liability for aryl-amines via computation. Bioorg Med Chem 19:3173–3182.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Miller DD,
    2. Eudicone JM,
    3. Pikalov A, and
    4. Kim E
    (2007) Comparative assessment of the incidence and severity of tardive dyskinesia in patients receiving aripiprazole or haloperidol for the treatment of schizophrenia: a post hoc analysis. J Clin Psychiatry 68:1901–1906.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Osburn S and
    2. Ryzhov V
    (2013) Ion-molecule reactions: analytical and structural tool. Anal Chem 85:769–778.
    OpenUrlCrossRef
  17. ↵
    1. Rodrigues AD
    (1999) Integrated cytochrome P450 reaction phenotyping: attempting to bridge the gap between cDNA-expressed cytochromes P450 and native human liver microsomes. Biochem Pharmacol 57:465–480.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Fishbein JC and
    2. Heilman JM
    1. Siraki AG
    (2013) Free radical metabolites in arylamine toxicity, in Advances in Molecular Toxicology (Fishbein JC and Heilman JM eds) vol. 7, pp 39–82, Elsevier, Amsterdam, DOI: 10.1016/B978-0-444-62645-5.00002-X.
    OpenUrlCrossRef
  19. ↵
    1. Urbanek RA,
    2. Xiong H,
    3. Wu Y,
    4. Blackwell W,
    5. Steelman G,
    6. Rosamond J,
    7. Wesolowski SS,
    8. Campbell JB,
    9. Zhang M,
    10. Brockel B,
    11. et al.
    (2013) Synthesis and SAR of aminothiazole fused benzazepines as selective dopamine D2 partial agonists. Bioorg Med Chem Lett 23:543–547.
    OpenUrlCrossRef
  20. ↵
    1. Watkins MA,
    2. Price JM,
    3. Winger BE, and
    4. Kenttämaa HI
    (2004) Ion-molecule reactions for mass spectrometric identification of functional groups in protonated oxygen-containing monofunctional compounds. Anal Chem 76:964–976.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Watkins MA,
    2. WeWora DV,
    3. Li S,
    4. Winger BE, and
    5. Kenttämaa HI
    (2005) Compound screening for the presence of the primary N-oxide functionality via ion-molecule reactions in a mass spectrometer. Anal Chem 77:5311–5316.
    OpenUrlCrossRefPubMed
View Abstract
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools
Research ArticleArticle

Structure Elucidation by Gas Phase Ion-Molecule Reaction

Minli Zhang, Ryan Eismin, Hilkka Kenttämaa, Hui Xiong, Ye Wu, Doug Burdette and Rebecca Urbanek
Drug Metabolism and Disposition March 1, 2015, 43 (3) 358-366; DOI: https://doi.org/10.1124/dmd.114.061978
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section