Glutathione Adduct Formation Research Articles

Improving the stability of thiol-maleimide bioconjugates via the formation of a thiazine structure.

Linker stability is critically important for the efficacy and safety of peptide and protein conjugates used for biological applications. One common conjugation strategy, thiol-maleimide coupling, generates a succinimidyl thioether linker with limited stability under physiological conditions. We have shown in previous work that when a peptide with an N-terminal cysteine is conjugated to a maleimide reagent, a thiazine structure is formed via a chemical rearrangement. Our preliminary work indicated that the thiazine linker has favorable stability. Here, we report the evaluation of a thiazine linker as an alternative to the widely used succinimidyl thioether linker for thiol-maleimide bioconjugation. The stability of the thiazine conjugate in comparison to the thioether conjugate was assessed across a broad pH range. Additionally, the propensity for retro-Michael reaction and cross-reactivity with other thiols was evaluated by treating conjugates in the presence of glutathione. The studies indicated that the thiazine linker degrades markedly slower than the thioether conjugate. In addition, the thiazine linker is over 20 times less susceptible to glutathione adduct formation. The NMR study of the thiazine structure confirmed that the formation of the thiazine linker is a stereoselective process that yields a single diastereomer. In summary, we propose the use of the thiazine linker obtained by conjugation of maleimide-containing reagents with peptides or proteins presenting an N-terminal cysteine as a novel approach for bioconjugation. The advantages of this approach are the formation of a linker with a well-defined stereochemical configuration, increased stability at physiological pH, and a strongly reduced propensity for thiol exchange.

Improved production of class I lanthipeptides in Escherichia coli.

Lanthipeptides are ribosomally synthesised and post-translationally modified peptides containing lanthionine (Lan) and methyllanthionine (MeLan) residues that are formed by dehydration of Ser/Thr residues followed by conjugate addition of Cys to the resulting dehydroamino acids. Class I lanthipeptide dehydratases utilize glutamyl-tRNAGlu as a co-substrate to glutamylate Ser/Thr followed by glutamate elimination. Here we report a new system to heterologously express class I lanthipeptides in Escherichia coli through co-expression of the producing organism's glutamyl-tRNA synthetase (GluRS) and tRNAGlu pair in the vector pEVOL. In contrast to the results in the absence of the pEVOL system, we observed the production of fully-dehydrated peptides, including epilancin 15X, and peptides from the Bacteroidota Chryseobacterium and Runella. A second common obstacle to production of lanthipeptides in E. coli is the formation of glutathione adducts. LanC-like (LanCL) enzymes were previously reported to add glutathione to dehydroamino-acid-containing proteins in Eukarya. Herein, we demonstrate that the LanCL enzymes can remove GSH adducts from C-glutathionylated peptides with dl- or ll-lanthionine stereochemistry. These two advances will aid synthetic biology-driven genome mining efforts to discover new lanthipeptides.

Discovery of a pyrazolo[1,5-a]pyrimidine derivative (MT-3014) as a highly selective PDE10A inhibitor via core structure transformation from the stilbene moiety

We have developed a new class of PDE10A inhibitor, a pyrazolo[1,5-a]pyrimidine derivative MT-3014 (1). A previous compound introduced was deprioritized due to concerns for E/Z-isomerization and glutathione-adduct formation at the core stilbene structure. We discovered pyrazolo [1,5-a] pyrimidine as a new lead scaffold by structure-based drug design utilizing a co-crystal structure with PDE10A. The lead compound was optimized for in vitro activity, solubility, and selectivity against human ether-á-go-go related gene cardiac channel binding. We observed that MT-3014 shows excellent efficacy in rat conditioned avoidance response test and suitable pharmacokinetic properties in rats, especially high brain penetration.

Enhancing Throughput of Glutathione Adduct Formation Studies and Structural Identification Using a Software-Assisted Workflow Based on High-Resolution Mass Spectrometry (HRMS) Data

Enhancing Throughput of Glutathione Adduct Formation Studies and Structural Identification Using a Software-Assisted Workflow Based on High-Resolution Mass Spectrometry (HRMS) Data

Minimizing DILI risk in drug discovery — A screening tool for drug candidates

Drug-induced liver injury (DILI) is a leading cause of acute hepatic failure and a major reason for market withdrawal of drugs. Idiosyncratic DILI is multifactorial, with unclear dose-dependency and poor predictability since the underlying patient-related susceptibilities are not sufficiently understood. Because of these limitations, a pharmaceutical research option would be to reduce the compound-related risk factors in the drug-discovery process.Here we describe the development and validation of a methodology for the assessment of DILI risk of drug candidates. As a training set, 81 marketed or withdrawn compounds with differing DILI rates – according to the FDA categorization – were tested in a combination of assays covering different mechanisms and endpoints contributing to human DILI. These include the generation of reactive metabolites (CYP3A4 time-dependent inhibition and glutathione adduct formation), inhibition of the human bile salt export pump (BSEP), mitochondrial toxicity and cytotoxicity (fibroblasts and human hepatocytes). Different approaches for dose- and exposure-based calibrations were assessed and the same parameters applied to a test set of 39 different compounds. We achieved a similar performance to the training set with an overall accuracy of 79% correctly predicted, a sensitivity of 76% and a specificity of 82%. This test system may be applied in a prospective manner to reduce the risk of idiosyncratic DILI of drug candidates.

Phase I and phase II reductive metabolism simulation of nitro aromatic xenobiotics with electrochemistry coupled with high resolution mass spectrometry.

Electrochemistry combined with (liquid chromatography) high resolution mass spectrometry was used to simulate the general reductive metabolism of three biologically important nitro aromatic molecules: 3-trifluoromethyl-4-nitrophenol (TFM), niclosamide, and nilutamide. TFM is a pesticide used in the Laurential Great Lakes while niclosamide and nilutamide are used in cancer therapy. At first, a flow-through electrochemical cell was directly connected to a high resolution mass spectrometer to evaluate the ability of electrochemistry to produce the main reduction metabolites of nitro aromatic, nitroso, hydroxylamine, and amine functional groups. Electrochemical experiments were then carried out at a constant potential of -2.5V before analysis of the reduction products by LC-HRMS, which confirmed the presence of the nitroso, hydroxylamine, and amine species as well as dimers. Dimer identification illustrates the reactivity of the nitroso species with amine and hydroxylamine species. To investigate xenobiotic metabolism,the reactivity of nitroso species to biomolecules was also examined. Binding of the nitroso metabolite to glutathione was demonstrated by the observation of adducts by LC-ESI(+)-HRMS and the characteristics of their MSMS fragmentation. In conclusion, electrochemistry produces the main reductive metabolites of nitro aromatics and supports the observation of nitroso reactivity through dimer or glutathione adduct formation.

Cell toxicity of methacrylate monomers—The role of glutathione adduct formation

Polymer-based dental restorative materials are designed to polymerize in situ. However, the conversion of methacrylate monomer to polymer is never complete, and leakage of the monomer occurs. It has been shown that these monomers are toxic in vitro; hence concerns regarding exposure of patients and dental personnel have been raised. Different monomer methacrylates are thought to cause toxicity through similar mechanisms, and the sequestration of cellular glutathione (GSH) may be a key event. In this study we examined the commonly used monomer methacrylates, 2-hydroxyethylmethacrylate (HEMA), triethylenglycol-dimethacrylate (TEGDMA), bisphenol-A-glycidyl-dimethacrylate (BisGMA), glycerol-dimethacrylate (GDMA) and methyl-methacrylate (MMA). The study aimed to establish monomers' ability to complex with GSH, and relate this to cellular toxicity endpoints. Except for BisGMA, all the monomer methacrylates decreased the GSH levels both in cells and in a cell-free system. The spontaneous formation of methacrylate-GSH adducts were observed for all methacrylate monomers except BisGMA. However, we were not able to correlate GSH depletion and toxic response measured as SDH activity and changes in cell growth pattern. Together, the current study indicates mechanisms other than GSH-binding to be involved in the toxicity of methacrylate monomers.

Cytochrome P450 Architecture and Cysteine Nucleophile Placement Impact Raloxifene-Mediated Mechanism-Based Inactivation

The propensity for cytochrome P450 (P450) enzymes to bioactivate xenobiotics is governed by the inherent chemistry of the xenobiotic itself and the active site architecture of the P450 enzyme(s). Accessible nucleophiles in the active site or egress channels of the P450 enzyme have the potential of sequestering reactive metabolites through covalent modification, thereby limiting their exposure to other proteins. Raloxifene, a drug known to undergo CYP3A-mediated reactive metabolite formation and time-dependent inhibition in vitro, was used to explore the potential for bioactivation and enzyme inactivation of additional P450 enzymes (CYP1A2, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A5). Every P450 tested except CYP2E1 was capable of raloxifene bioactivation, based on glutathione adduct formation. However, raloxifene-mediated time-dependent inhibition only occurred in CYP2C8 and CYP3A4. Comparable inactivation kinetics were achieved with K(I) and k(inact) values of 0.26 μM and 0.10 min(-1) and 0.81 μM and 0.20 min(-1) for CYP2C8 and CYP3A4, respectively. Proteolytic digests of CYP2C8 and CYP3A4 Supersomes revealed adducts to Cys225 and Cys239 for CYP2C8 and CYP3A4, respectively. For each P450 enzyme, proposed substrate/metabolite access channels were mapped and active site cysteines were identified, which revealed that only CYP2C8 and CYP3A4 possess accessible cysteine residues near the active site cavities, a result consistent with the observed kinetics. The combined data suggest that the extent of bioactivation across P450 enzymes does not correlate with P450 inactivation. In addition, multiple factors contribute to the ability of reactive metabolites to form apo-adducts with P450 enzymes.

Determination of S-containing drug metabolites from in vitro and in vivo metabolism studies by using LC-ICP/MS

Recently, liquid chromatography coupled to inductively coupled plasma mass spectrometry (LC-ICP/MS) has been introduced to deal with some applications in the field of pharmaceutical, biomedical, and clinical analysis. In the case of drug research, the number of drugs and their metabolites containing detectable elements is quite limited. In this paper, LC-ICP/MS has been demonstrated to be suitable for the determination of S-containing drugs and their metabolites. In order to minimize the interference of polyatomic oxygen (m/z 32), the indirect detection of S, by means of the SO(+) ion (m/z 48), was optimized. For quantification purposes, it has been encountered that the percentage of organic solvent in the mobile phase strongly affects the sensitivity. Here, corrective strategies based on calibration curves established at different solvent concentrations (solvent-zone quantification) and post-column gradient compensation have been proposed to circumvent sensitivity variations. Results obtained have shown that suitable calibration models have been built for any compound regardless of the solvent percentage at which it is eluted from the chromatographic column. To prove the applicability of this methodology, the metabolism of ethacrynic acid and tiotropium bromide has been studied in vitro and in vivo. In the first case, ethacrynic acid does not contain S in its structure, however, the major route of metabolism for this compound consists of the formation of glutathione adduct and its further degradation. In the second case, tiotropium bromide contains two S atoms in its structure.

Analysis of Arabidopsis glutathione-transferases in yeast

The genome of Arabidopsis thaliana encodes 54 functional glutathione transferases (GSTs), classified in seven clades. Although plant GSTs have been implicated in the detoxification of xenobiotics, such as herbicides, extensive redundancy within this large gene family impedes a functional analysis in planta. In this study, a GST-deficient yeast strain was established as a system for analyzing plant GSTs that allows screening for GST substrates and identifying substrate preferences within the plant GST family. To this end, five yeast genes encoding GSTs and GST-related proteins were simultaneously disrupted. The resulting yeast quintuple mutant showed a strongly reduced conjugation of the GST substrates 1-chloro-2,4-dinitrobenzene (CDNB) and 4-chloro-7-nitro-2,1,3-benzoxadiazole (NBD-Cl). Consistently, the quintuple mutant was hypersensitive to CDNB, and this phenotype was complemented by the inducible expression of Arabidopsis GSTs. The conjugating activity of the plant GSTs was assessed by in vitro enzymatic assays and via analysis of exposed yeast cells. The formation of glutathione adducts with dinitrobenzene was unequivocally verified by stable isotope labeling and subsequent accurate ultrahigh-resolution mass spectrometry (ICR-FTMS). Analysis of Arabidopsis GSTs encompassing six clades and 42 members demonstrated functional expression in yeast by using CDNB and NBD-Cl as model substrates. Subsequently, the established yeast system was explored for its potential to screen the Arabidopsis GST family for conjugation of the fungicide anilazine. Thirty Arabidopsis GSTs were identified that conferred increased levels of glutathionylated anilazine. Efficient anilazine conjugation was observed in the presence of the phi, tau, and theta clade GSTs including AtGSTF2, AtGSTF4, AtGSTF6, AtGSTF8, AtGSTF10, and AtGSTT2, none of which had previously been known to contribute to fungicide detoxification. ICR-FTMS analysis of yeast extracts allowed the simultaneous detection and semiquantification of anilazine conjugates as well as catabolites.

Quantifying intrinsic chemical reactivity of molecular structural features for protein binding and reactive toxicity, using the MOSES chemoinformatics system

Covalent binding of xenobiotic compounds to endogenous biomolecular sites, e.g. protein residues, leads to potentially irreversible toxic effects such as enhanced acute toxicity or skin sensitization [1]. This mechanistic knowledge provides the basis for the in silico prediction of these toxicities, as required by the EU REACH legislation and the EU Cosmetics Directive. A general toxicity prediction can be based on three consecutive steps [2]: (1.) Identification of a potential reactive protein binding mechanism via a set of molecular structural features. Those structural features can be encoded by the Chemical Subgraph Representation Markup Language (CSRML), that supports a flexible annotation of meta information, including physicochemical properties as annotations. (2.) Confirmation and quantification of (bio)chemical reactivity. The potential for a chemical to be reactive can be captured by mechanistically based QSAR models. This intrinsic reactivity is calculated rapidly by descriptors of the chemoinformatics platform Molecular Structure Encoding System (MOSES) [3]. It represents electronic, steric and resonance effects in chemical structures. The performances obtained by these reactivity models are close to time-consuming quantum chemical reactivity calculations, e.g., se = 0.44 versus 0.40 for glutathione adduct formation via Michael addition, comparing predicted values to an experimental reactivity dataset [1]. (3.) Establishing a relationship between calculated reactivity and toxicity. The predicted intrinsic reactivity values were linked to the computational prediction for different modes of toxic action, with good correlations between predicted and in vitro toxicity (up to r2=0.86). The combined use of structural information and computed reactivity could assist in the non-animal based risk assessment of chemicals for regulatory purposes and in the application of integrated testing strategies (ITS). The research has received funding from the EU FP7 COSMOS Project (grant agreement n° 266835) and financing from COLIPA.

In vitro evaluation of the potential for drug-induced toxicity based on 35S-labeled glutathione adduct formation and daily dose

Drug-induced toxicity such as idiosyncratic drug toxicity is believed to be reduced when either reactive metabolite formation or exposure to a drug is minimized. The objective of the present study was therefore to clarify the relationship between the daily doses, the formation rates of reactive metabolite adduct with (35)S-glutathione (RM-GS) and the safety profiles of compounds. The RM-GS formation rates for 113 test compounds were determined by incubation with human liver microsomes, and the test compounds were classified into three categories of safe, warning and withdrawn/black box warning. A total of 23 out of 28 withdrawn/black box warning drugs showed both a RM-GS formation rate of over 1 pmol/30 min/mg protein and a dose of over 100 mg. These results suggest that when compounds are plotted in this region, the compounds would have a relatively high idiosyncratic drug toxicity potential.

2,7-Pyrrolo[2,1-f][1,2,4]triazines as JAK2 inhibitors: Modification of target structure to minimize reactive metabolite formation

The JAK2/STAT pathway has important roles in hematopoiesis. With the discovery of the JAK2 V617F mutation and its presence in many patients with myeloproliferative neoplasms, research in the JAK2 inhibitor arena has dramatically increased. We report a novel series of potent JAK2 inhibitors containing a 2,7-pyrrolotriazine core. To minimize potential drug-induced toxicity, targets were analyzed for the ability to form a glutathione adduct. Glutathione adduct formation was decreased by modification of the aniline substituent at C2.

A Dedicated Glutathione S-Transferase Mediates Carbon–Sulfur Bond Formation in Gliotoxin Biosynthesis

Gliotoxin is a virulence factor of the human pathogen Aspergillus fumigatus , the leading cause of invasive aspergillosis. Its toxicity is mediated by the unusual transannular disulfide bridge of the epidithiodiketopiperazine (ETP) scaffold. Here we disclose the critical role of a specialized glutathione S-transferase (GST), GliG, in enzymatic sulfurization. Furthermore, we show that bishydroxylation of the diketopiperazine by the oxygenase GliC is a prerequisite for glutathione adduct formation. This is the first report of the involvement of a GST in enzymatic C-S bond formation in microbial secondary metabolism.

Regulation of Benzo[a]pyrene-Mediated DNA- and Glutathione-Adduct Formation by 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Human Lung Cells

Environmental carcinogens, such as polycyclic aromatic hydrocarbons (PAHs), require metabolic activation to DNA-reactive metabolites in order to exert their tumorigenic effects. Benzo[a]pyrene (B[a]P), a prototypic PAH, is metabolized by cytochrome P450 (P450) 1A1/1B1 and epoxide hydrolase to (−)-B[a]P-7,8-dihydro-7,8-diol (B[a]P-7,8-dihydrodiol). B[a]P-7,8-dihydrodiol then undergoes further P4501A1/1B1-mediated metabolism to the ultimate carcinogen, (+)-anti-7,8-dihydroxy-9,10-epoxy-7,8,9,10-tetrahydro-B[a]P (B[a]PDE), which forms DNA-adducts primarily with 2′-deoxyguanosine (dGuo) to form (+)-anti-trans-B[a]PDE-N2-dGuo (B[a]PDE-dGuo) in DNA. Pretreatment of cells with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is known to induce P4501A1/1B1 mRNA expression through the aryl hydrocarbon receptor (AhR) pathway. This causes increased B[a]PDE-dGuo formation in liver cells. In contrast, TCDD induction of H358 lung cells surprisingly caused a decrease in (−)-B[a]P-7,8-dihydrodiol-mediated (+)-B[a]PDE-dGuo adduct formation when compared with the non-TCDD-induced cells. Furthermore, treatment of the TCDD-induced cells with (±)-B[a]PDE also resulted in decreased (+)-B[a]PDE-dGuo adduct formation when compared with the non-TCDD-induced cells. These data suggested that it was a detoxification pathway that had been up-regulated rather than an activation pathway that had been down-regulated. LC-MS was used to analyze B[a]PDE-dGuo and B[a]PDE-GSH-adducts in H358 lung and HepG2 liver cells. There was a significant increase in the (−)-B[a]PDE-GSH-adduct with high enantiomeric excess after treatment of the TCDD-induced H358 cells with (±)-B[a]PDE when compared with the noninduced cells. This could explain why increased activation of (−)-B[a]P-7,8-dihydrodiol through TCDD up-regulation of P4501A1/1B1 did not lead to increased (+)-B[a]PDE-dGuo adducts in the H358 lung cells. In addition, TCDD did not induce B[a]PDE-GSH-adduct formation in HepG2 liver cells. (±)-B[a]PDE-GSH-adducts were formed at much lower levels in both TCDD-induced and noninduced HepG2 cells when compared with (−)-B[a]PDE-GSH-adducts in the H358 lung cells. Therefore, our study has revealed that there is a subtle balance between activation and detoxification of B[a]P in lung-derived compared with liver-derived cells and that this determines how much DNA damage occurs.

Identification Strategy Using Combined Mass Spectrometric Techniques for Elucidation of Phase I and Phase II in Vitro Metabolites of Lipophilic Marine Biotoxins

Combining mass spectrometric tools, a total of 47 in vitro metabolites of okadaic acid (OA), dinophysistoxins 1 and 2 (DTX1 and DTX2), yessotoxin (YTX), azaspiracid1 (AZA1), and pectenotoxin 2 (PTX2) could be detected and confirmed after an incubation with rat liver S9-mix. In a first step, liquid chromatography (LC) combined with tandem mass spectrometry (MS/MS) was used as a screening tool for the identification of in vitro metabolites of lipophilic marine biotoxins. Metabolic phase I and phase II reactions were screened for metabolites by calculating and subsequently monitoring theoretical MS transitions. In a second step, metabolites were confirmed by determination of accurate masses using high resolution MS provided by Orbitrap technology. Subsequently, product ion spectra, precursor ion spectra, and MS3 spectra were recorded for structure elucidation of metabolites. While all investigated toxins were found to form various oxygenated metabolites during the oxidative phase I metabolism, those metabolites varied in the number of added oxygen atoms and in the number of individual isomers. No hints were obtained concerning the formation of glutathione adducts, and a conjugation with glucuronic acid was detected for AZA1 only.

An integrated reactive metabolite evaluation approach to assess and reduce safety risk during drug discovery and development

Metabolic bioactivation is widely considered an undesirable event and a likely prerequisite step in the expression of drug-induced hepatotoxicity and hypersensitivity. Reducing bioactivation risk early in drug discovery, therefore, may help reduce compound attrition and provide safer drug therapies. In vitro bioactivation data and clinical dose for a large set of marketed drugs were analysed for their concordance with clinical hepatotoxicity and the data used to develop an early reactive metabolite strategy. A contingency table analysis of cytochrome P450 metabolism-dependent inhibition (CYP MDI), glutathione trapping data, and dose for >200 marketed drugs with or without a clinical hepatotoxic signal; and microsomal covalent binding data and dose for ∼60 marketed compounds obtained from literature publications was performed to assess concordance with hepatotoxicity. Clinical daily dose ≥100 mg or glutathione adduct formation was strongly associated with hepatotoxicity ( p < 0.0001, p = 0.003, respectively). A trend towards clinical hepatotoxicity was observed with marked CYP MDI or metabolism-dependent covalent binding ≥200 pmol/mg. The percentage of hepatotoxic drugs identified by high dose (67%) increased significantly when bioactivation data were combined with dose (80–100%). As CYP MDI and glutathione adduct assays do not require the synthesis of radiolabelled compound and are relatively easy to conduct, they may be of particular value for early assessment in programs with lower risk tolerance. Such information together with an overall understanding of the metabolic properties of the compound and risk/benefit considerations may trigger further assessment. Additionally hepatic transcriptomic data (e.g., Nrf2-activated gene expression) from rat toxicity studies can provide evidence of in vivo consequences of bioactivation. As attenuation of a metabolic bioactivation risk early in drug discovery could reduce compound attrition and provide safer drug therapies, we have developed a decision-based early reactive metabolite strategy that can be tailored to the needs of individual programs.

Characterization of Dasatinib and Its Structural Analogs as CYP3A4 Mechanism-Based Inactivators and the Proposed Bioactivation Pathways

Dasatinib was approved in 2006 for the treatment of imatinib-resistant chronic myelogenous leukemia and functions primarily through the inhibition of BCR-ABL and Src kinase. Dasatinib is extensively metabolized in humans by CYP3A4. In this study, we report that the bioactivation of dasatinib by CYP3A4 proceeds through a reactive intermediate that leads to CYP3A4 inactivation with K(I) = 6.3 microM and k(inact) = 0.034 min(-1). The major mechanism of inactivation proceeds through hydroxylation at the para-position of the 2-chloro-6-methylphenyl ring followed by further oxidation, forming a reactive quinone-imine, similar to the reactive intermediates formed by acetaminophen and diclofenac. Formation of a reactive imine-methide was also detected but appears to be a minor pathway. When glutathione was added to human liver microsomal incubations, dasatinib-glutathione adducts were detected. Numerous dasatinib analogs were synthesized in an effort to understand what modifications would block the formation of reactive intermediates during dasatinib metabolism. It is interesting to note that blocking the site of hydroxylation with a methyl group was not effective because a reactive imine-methide was formed, nor was blocking the site with fluorine because the fluorine was removed through an oxidative defluorination mechanism and the reactive quinone-imine was still formed. Numerous analogs are presented that did effectively block the formation of glutathione adducts and prevent the inactivation of CYP3A4.

Chemical Reactivity of Methoxy 4-O-Aryl Quinolines: Identification of Glutathione Displacement Products in Vitro and in Vivo

AMG 458 {1-(2-hydroxy-2-methylpropyl)-N-[5-(7-methoxyquinolin-4-yloxy)pyridin-2-yl]-5-methyl-3-oxo-2-phenyl-2,3-dihydro-1H-pyrazole-4-carboxamide} is a potent, selective inhibitor of c-Met, a receptor tyrosine kinase that is often deregulated in cancer. AMG 458 was observed to bind covalently to liver microsomal proteins from rats and humans in the absence of NADPH. When [(14)C]AMG 458 was incubated with liver microsomes in the presence of glutathione and N-acetyl cysteine, thioether adducts were detected by radiochromatography and LC/MS/MS analysis. These adducts were also formed upon incubation of AMG 458 with glutathione and N-acetyl cysteine in buffers at pH 7.4. In vivo, the thioether adducts were detected in bile and urine of bile duct-cannulated rats dosed with [(14)C]AMG 458. The two adducts were isolated, and their structures were determined by MS/MS and NMR analysis. The identified structures resulted from a thiol displacement reaction to yield a quinoline thioether structure and the corresponding hydroxyaryl moiety. The insights gained from elucidating the mechanism of adduct formation led to the design of AMG 458 analogues that exhibited eliminated or reduced glutathione adduct formation in vitro and in vivo.

Enzymatic Reduction and Glutathione Conjugation of Benzoquinone Ansamycin Heat Shock Protein 90 Inhibitors: Relevance for Toxicity and Mechanism of Action

Two-electron reduction of benzoquinone ansamycin (BA) heat shock protein (Hsp) 90 inhibitors by NAD(P)H:quinone oxidoreductase 1 (NQO1) to hydroquinone ansamycins (BAH2s) leads to greater Hsp90 inhibitory activity. BAs can also be metabolized by one-electron reductases and can interact with glutathione, reactions that have been associated with toxicity. Using a series of BAs, we investigated the stability of the BAH2s generated by NQO1, the ability of BAs to be metabolized by one-electron reductases, and their conjugation with glutathione. The BAs used were geldanamycin (GM), 17-(allylamino)-17-demethoxygeldanamycin (17AAG), 17-demethoxy-17-[[2-(dimethyl amino)ethyl]amino]-geldanamycin (17DMAG), 17-(amino)-17-demethoxygeldanamycin (17AG), and 17-demethoxy-17-[[2-(pyrrolidin-1-yl)ethyl]amino]-geldanamycin (17AEP-GA). The relative stabilities of BAH2s at pH 7.4 were GM hydroquinone>17AAG hydroquinone>17DMAG hydroquinone>17AG hydroquinone and 17AEP-GA hydroquinone. Using human and mouse liver microsomes and either NADPH or NADH as cofactors, 17AAG had the lowest rate of one-electron reduction, whereas GM had the highest rate. 17DMAG demonstrated the greatest rate of redox cycling catalyzed by purified human cytochrome P450 reductase, whereas 17AAG again had the slowest rate. GM formed a glutathione adduct most readily followed by 17DMAG. The formation of glutathione adducts of 17AAG and 17AG were relatively slow in comparison. These data demonstrate that GM, the most hepatotoxic BAs in the series had a greater propensity to undergo redox cycling reactions catalyzed by hepatic one-electron reductases and markedly greater reactivity with thiols when compared with the least hepatotoxic analog 17AAG. Minimizing the propensity of BA derivatives to undergo one-electron reduction and glutathione conjugation while maximizing their two-electron reduction to stable Hsp90 inhibitory hydroquinones may be a useful strategy for optimizing the therapeutic index of BAs.