No abstract

The hepatotoxicity of thioacetamide (TA) has been known since 1948. In rats, single doses cause centrilobular necrosis accompanied by increases in plasma transaminases and bilirubin. To elicit these effects TA requires oxidative bioactivation leading first to its S-oxide (TASO) and then to its chemically reactive S,S-dioxide (TASO2) which ultimately modifies amine-lipids and proteins. To generate a suite of liver proteins adducted by TA metabolites for proteomic analysis, and to reduce the need for both animals and labeled compounds, we treated isolated hepatocytes directly with TA. Surprisingly, TA was not toxic at concentrations up to 50 mM for 40 hr. On the other hand, TASO was highly toxic to isolated hepatocytes as indicated by LDH release, cellular morphology and vital staining with Hoechst 33342/propidium iodide. TASO toxicity was partially blocked by the CYP2E1 inhibitors diallyl sulfide and 4-methylpyrazole, and was strongly inhibited by TA. Significantly, we found that hepatocytes produce TA from TASO relatively efficiently by back-reduction. The covalent binding of [14C]-TASO is inhibited by unlabeled TA which acts as a “cold-trap” for [14C]-TA and prevents its re-oxidation to [14C]-TASO. This in turn increases the net consumption of [14C]-TASO despite the fact that its oxidation to TASO2 is inhibited. The potent inhibition of TASO oxidation by TA, coupled with the back-reduction of TASO and its futile redox cycling with TA may help explain phenomena previously interpreted as “saturation toxicokinetics” in the in vivo metabolism and toxicity of TA and TASO. The improved understanding of the metabolism and covalent binding of TA and TASO facilitates the use of hepatocytes to prepare protein adducts for target protein identification.

Thiobenzamide (TB) is a potent hepatotoxin in rats, causing dose-dependent hyperbilirubinemia, steatosis, and centrolobular necrosis. These effects arise subsequent to and appear to result from the covalent binding of the iminosulfinic acid metabolite of TB to cellular proteins and phosphatidylethanolamine lipids [ Ji et al. ( 2007) Chem. Res. Toxicol. 20, 701- 708 ]. To better understand the relationship between the protein covalent binding and the toxicity of TB, we investigated the chemistry of the adduction process and the identity of the target proteins. Cytosolic and microsomal proteins isolated from the livers of rats treated with a hepatotoxic dose of [ carboxyl- (14)C]TB contained high levels of covalently bound radioactivity (25.6 and 36.8 nmol equiv/mg protein, respectively). These proteins were fractionated by two-dimensional gel electrophoresis, and radioactive spots (154 cytosolic and 118 microsomal) were located by phosphorimaging. Corresponding spots from animals treated with a 1:1 mixture of TB and TB- d 5 were similarly separated, the spots were excised, and the proteins were digested in gel with trypsin. Peptide mass mapping identified 42 cytosolic and 24 microsomal proteins, many of which appeared in more than one spot on the gel; however, only a few spots contained more than one identifiable protein. Eighty-six peptides carrying either a benzoyl or a benzimidoyl adduct on a lysine side chain were clearly recognized by their d 0/ d 5 isotopic signature (sometimes both in the same digest). Because model studies showed that benzoyl adducts do not arise by hydrolysis of benzimidoyl adducts, it was proposed that TB undergoes S-oxidation twice to form iminosulfinic acid 4 [PhC(NH)SO 2H], which either benzimidoylates a lysine side chain or undergoes hydrolysis to 9 [PhC(O)SO 2H] and then benzoylates a lysine side chain. The proteins modified by TB metabolites serve a range of biological functions and form a set that overlaps partly with the sets of proteins known to be modified by several other metabolically activated hepatotoxins. The relationship of the adduction of these target proteins to the cytotoxicity of reactive metabolites is discussed in terms of three currently popular mechanisms of toxicity: inhibition of enzymes important to the maintenance of cellular energy and homeostasis, the unfolded protein response, and interference with kinase-based signaling pathways that affect cell survival.

The hepatotoxicity of bromobenzene is strongly correlated with the covalent binding of chemically reactive metabolites to cellular proteins, but up to now relatively few hepatic protein targets of these reactive metabolites have been identified. To identify additional hepatic protein targets we injected an hepatotoxic dose of [14C]bromobenzene to phenobarbital-pretreated male Sprague-Dawley rats ip. After 4 h, their livers were removed and homogenized, and the homogenates fractionated by differential ultracentrifugation. The highest specific radiolabeling (6.1 nmol equiv 14C/mg of protein) was observed in a particulate fraction (P25) sedimented at 25000g from a 6000g supernatant fraction. Proteins in this fraction were separated by two-dimensional electrophoresis and, after transblotting, analyzed for radioactivity by phosphorimaging. More than 20 radiolabeled protein spots were observed in the blots. For 17 of these spots, peptide mass maps were obtained using in-gel digestion with trypsin, followed by MALDI-TOF mass spectrometric analysis of the resulting peptide mixtures. By searching genomic databases, the 17 sets of MS-derived peptide masses were found to match predicted tryptic fragments of just 7 proteins. Spots 1-4 matched with 78 kDa glucose regulated protein (GRP78), protein disulfide isomerase isozyme A1 (PDIA1), endoplasmic reticulum protein ERp29, and PDIA6, respectively. Spots 5 and 6, 7-11, and 12-17 presented as apparent "charge trains" of spots, each of which gave peptide mixtures closely similar to those of other spots within the train. The proteins present in these sets of spots were identified as transthyretin, serum albumin precursor and PDIA3, respectively. The possible relationship of the adduction of these proteins to the toxicological outcome is discussed.

The role of single electron transfer (SET) in P450-catalyzed N-dealkylation reactions has been studied using the probe substrates N-cyclopropyl-N-methylaniline (2a) and N-(1'-methylcyclopropyl)-N-methylaniline (2b). In earlier work, we showed that SET oxidation of 2a by horseadish peroxidase leads exclusively to products arising via fragmentation of the cyclopropane ring [Shaffer, C. L.; Morton, M. D.; Hanzlik, R. P. J. Am. Chem. Soc. 2001, 123, 8502-8508]. In the present study, we found that liver microsomes from phenobarbital pretreated rats (which contain CYP2B1 as the predominant isozyme) oxidize [1'-(13)C, 1'-(14)C]-2a efficiently (80% consumption in 90 min). Disappearance of 2a follows first-order kinetics throughout, indicating a lack of P450 inactivation by 2a. HPLC examination of incubation mixtures revealed three UV-absorbing metabolites: N-methylaniline (4), N-cyclopropylaniline (6a), and a metabolite (M1) tentatively identified as p-hydroxy-2a, in a 2:5:2 mole ratio, respectively. 2,4-Dinitrophenylhydrazine trapping indicated formation of formaldehyde equimolar with 6a; 3-hydroxypropionaldehyde and acrolein were not detected. Examination of incubations of 2a by (13)C NMR revealed four (13)C-enriched signals, three of which were identified by comparison to authentic standards as N-cyclopropylaniline (6a, 33.6 ppm), cyclopropanone hydrate (11, 79.2 ppm), and propionic acid (12, 179.9 ppm); the fourth signal (42.2 ppm) was tentatively determined to be p-hydroxy-2a. Incubation of 2a with purified reconstituted CYP2B1 also afforded 4, 6a, and M1 in a 2:5:2 mole ratio (by HPLC), indicating that all metabolites are formed at a single active site. Incubation of 2b with PB microsomes resulted in p-hydroxylation and N-demethylation only; no loss or ring-opening of the cyclopropyl group occurred. These results effectively rule out the participation of a SET mechanism in the P450-catalyzed N-dealkylation of cyclopropylamines 2a and 2b, and argue strongly for the N-dealkylation of 2a via a carbinolamine intermediate formed by a conventional C-hydroxylation mechanism.

Metabolic activation and protein covalent binding are early and apparently obligatory events in the cytotoxicity of many simple organic chemicals including drugs and natural products. Although much has been learned about the chemistry of reactive metabolite formation and reactivity toward protein nucleophiles, progress in identifying specific protein targets for reactive metabolites of various protoxins has been much slower. We previously reported nine microsomal and three cytosolic proteins as targets for reactive metabolites of bromobenzene in rat liver. These results, and contemporary work by others, indicate that protein covalent binding is not totally random in cells. Moreover, as protein targets for other protoxins were identified, little commonality of target proteins became apparent. In the present work, we used two-dimensional gel electrophoresis to separate liver cytosolic proteins from rats treated with 14C-bromobenzene; 110 of the 836 observed spots contained measurable radioactivity that varied over a 600-fold range of adduct density. Of these 110 spots, in-gel digestion coupled with mass spectrometry identified apparently single proteins in 57 spots. A few other spots clearly contained more than one identifiable protein, and in several cases, the same protein was identified in several spots having different apparent molecular masses and/or pI. Altogether, 33 unique new protein targets for bromobenzene metabolites were identified and compared to those known for acetaminophen, naphthalene, butylated hydroxytoluene, benzene, thiobenzamide, and halothane via a target protein database available at http://tpdb.medchem.ku.edu:8080/protein_database/. With increasing numbers of target proteins becoming known, more commonality in targeting by reactive metabolites from diverse chemical agents may be seen. Such commonality may help to separate toxicologically significant covalent binding events from a background of covalent binding that is toxicologically inconsequential.

Thioacetamide (TA) has long been known as a hepatotoxicant whose bioactivation requires S-oxidation to thioacetamide S-oxide (TASO) and then to the very reactive S,S-dioxide (TASO2). The latter can tautomerize to form acylating species capable of covalently modifying cellular nucleophiles including phosphatidylethanolamine (PE) lipids and protein lysine side chains. Isolated hepatocytes efficiently oxidize TA to TASO but experience little covalent binding or cytotoxicity because TA is a very potent inhibitor of the oxidation of TASO to TASO2. On the other hand hepatocytes treated with TASO show extensive covalent binding to both lipids and proteins accompanied by extensive cytotoxicity. In this work, we treated rat hepatocytes with [14C]-TASO and submitted the mitochondrial, microsomal and cytosolic fractions to 2DGE which revealed a total of 321 radioactive protein spots. To facilitate the identification of target proteins and adducted peptides we also treated cells with a mixture of TASO/[13C2D3]-TASO. Using a combination of 1DGE- and 2DGE-based proteomic approaches, we identified 187 modified peptides (174 acetylated, 50 acetimidoylated and 37 in both forms) from a total of 88 non-redundant target proteins. Among the latter, 57 are also known targets of at least one other hepatotoxin. The formation of both amide- and amidine-type adducts to protein lysine side chains is in contrast to the exclusive formation of amidine-type adducts with PE phospholipids. Thiobenzamide (TB) undergoes the same two-step oxidative bioactivation as TA, and it also gives rise to both amide and amidine adducts on protein lysine side chains but only amidine adducts to PE lipids. Despite their similarity in functional group chemical reactivity, only 38 of 62 known TB target proteins are found among the 88 known targets of TASO. The potential roles of protein modification by TASO in triggering cytotoxicity are discussed in terms of enzyme inhibition, protein folding and chaperone function, and the emerging role of protein acetylation in intracellular signaling and the regulation of biochemical pathways.

Thiobenzamide (TB) is hepatotoxic in rats causing centrolobular necrosis, steatosis, cholestasis, and hyperbilirubinemia. It serves as a model compound for a number of thiocarbonyl compounds that undergo oxidative bioactivation to chemically reactive metabolites. The hepatotoxicity of TB is strongly dependent on the electronic character of substituents in the meta- and para-positions, with Hammett rho values ranging from -4 to -2. On the other hand, ortho substituents that hinder nucleophilic addition to the benzylic carbon of S-oxidized TB metabolites abrogate the toxicity and protein covalent binding of TB. This strong linkage between the chemistry of TB and its metabolites and their toxicity suggests that this model is a good one for probing the overall mechanism of chemically induced biological responses. While investigating the protein covalent binding of TB metabolites, we noticed an unusually large amount of radioactivity associated with the lipid fraction of rat liver microsomes. Thin-layer chromatography showed that most of the radioactivity was contained in a single spot more polar than the neutral lipids but less polar than the phospholipid fractions. Mass spectral analyses aided by the use of synthetic standards identified the material as N-benzimidoyl derivatives of typical microsomal phosphatidylethanolamine (PE) lipids. Quantitative analysis indicated that up to 25% of total microsomal PE became modified within 5 h after a hepatotoxic dose of TB. Further studies will be required to determine the contribution of lipid modification to the hepatotoxicity of TB.