Xenobiotic acyl-CoA formation: evidence of kinetically distinct hepatic microsomal long-chain fatty acid and nafenopin-CoA ligases

Metabolism of xenobiotic carboxylic acids: focus on coenzyme A conjugation, reactivity, and interference with lipid metabolism

While xenobiotic carboxylic acids (XCAs) have been studied extensively with respect to their enzymatic conversion to potentially reactive acyl glucuronides with implications to drug induced hepatotoxicity, the formation of xenobiotic-S-acyl-CoA thioesters (xenobiotic-CoAs) have been much less studied in spite of data indicating that such conjugates may be equally or more reactive than the corresponding acyl glucuronides. This review addresses enzymes and cell organelles involved in the formation of xenobiotic-CoAs, the reactivity of such conjugates toward biological macromolecules, and in vitro and in vivo methodology to assess consequences of such reactivity. Further, the propensity of xenobiotic-CoAs to interfere with endogenous lipid metabolism, e.g., inhibition of β-oxidation or depletion of the CoA or carnitine pools, adds to the complexity of the potential contribution of XCAs to hepatotoxicity by a number of mechanisms in addition to those in common with the corresponding acyl glucuronides. On the basis of our review of the literature on xenobiotic-CoA conjugates, there appear to be a number of gaps in our understanding of the bioactivation of XCA both with respect to the mechanisms involved and the experimental approaches to distinguish between the role of acyl glucuronides and xenobiotic-CoA conjugates. These aspects are focused upon and described in detail in this review.

Techniques for measuring the activity of carboxylic acid:CoA ligase and acyl-CoA:amino acid N-acyltransferase: the amino acid conjugation pathway

A wide variety of xenobiotic carboxylic acids are metabolized to their amino acid conjugates via a pathway that exists primarily in liver and kidney. This conjugation occurs in a two-step pathway catalyzed by two distinct types of enzymes, ligases and transferases. Measurements of acyl-CoA ligase activity include monitoring the rate of appearance of AMP or PPi, or the CoA adduct. N-acyltransferases catalyze formation of an amino acid conjugate from the CoA-activated intermediate, releasing CoA. This reaction is monitored by following the release of free CoA or the disappearance of the acyl-CoA adduct.

Hepatic enzymatic synthesis and hydrolysis of CoA esters of solvent-derived oxa acids

Many ethylene glycol-derived solvents are oxidized to xenobiotic alkoxyacetic acids (3-oxa acids) by hepatic enzymes. The toxicity of these ubiquitous solvents has been associated with their oxa acid metabolites. For many xenobiotic carboxylic acids, the toxicity is associated with the CoA ester of the acid. In this study, related alkoxyacetic acids were evaluated as potential substrates for acyl-CoA synthetases found in mitochondrial, peroxisomal, and microsomal fractions isolated from rat liver. Likewise, chemically synthesized oxa acyl-CoAs were used as substrates for acyl-CoA hydrolases associated with the same rat liver fractions. Activities of the xenobiotic oxygen-substituted substrates were compared with analogous physiologic aliphatic substrates by UV-vis spectrophotometric methods. All of the solvent-derived oxa acids were reasonable substrates for the acyl-CoA synthetases, although their activity was usually less than the corresponding physiologic acid. Acyl-CoA hydrolase activities were decreased compared with acyl-CoA synthetase activities for all substrates, especially for the oxa acyl-CoAs. These studies suggest that these xenobiotic carboxylic acids may be converted to reactive acyl-CoA moieties which will persist in areas of the cell proximal to lipid synthesis, beta-oxidation, protein acylation, and amino acid conjugation. The interaction of these xenobiotic acyl-CoAs with those processes may be important to their toxicity and/or detoxification.

Identification of the amino acids of human serum albumin involved in the reaction with the naproxen acyl coenzyme A thioester using liquid chromatography combined with fluorescence and mass spectrometric detection

Xenobiotic carboxylic acids, that via their metabolites covalently modify proteins, have been associated with serious side effects in man. Such reactive metabolites may be acyl glucuronides or alternatively, the corresponding acyl-CoA thioesters. In this study, the reaction of a model xenobiotic acyl-CoA, the naproxen-CoA, with human serum albumin (HSA), was characterized by high-performance liquid chromatography employing fluorescence and mass spectrometric detection. One mM naproxen-CoA was incubated for 6h with HSA (0.45 mM) at 37 degrees C in a 0.1M phosphate buffer (pH 7.4). The tryptic digest of the reduced and alkylated protein was analyzed in order to identify the amino acids in the sequence that were covalently modified with naproxen. Fluorescent peptides, that represented naproxen-modified peptides, were characterized using HPLC-MS-MS and HPLC-MS in zoom scan mode, which provided information on the structure and the charge of the modified peptides. The naproxen-CoA reacted predominantly with lysine 199, lysine 541, and lysine 351, which was in agreement with the binding pattern that has previously been reported for the reactive acyl glucuronides and their reaction with HSA.

Chemical reactivity of the naproxen acyl glucuronide and the naproxen coenzyme A thioester towards bionucleophiles

Drugs may be metabolised to reactive electrophilic species that spontaneously react with proteins. The presence of such drug-protein adducts has been associated with drug toxicity. In this study, the reactivity of the major metabolite of naproxen--the 1-beta-O-glucuronide (Nap-GlcU)--was compared to the corresponding naproxen coenzyme A (Nap-CoA) thioester. The reactivity of the two metabolites was assessed in vitro in a phosphate buffer (pH 7.4; 0.1 M) at 37 degrees C towards the model bionucleophiles glutathione and human serum albumin (HSA). The reaction between the electrophilic species (Nap-GlcU and Nap-CoA) and glutathione forming the Nap-glutathione conjugate was monitored using LC-MS-MS and LC-UV, respectively. It was shown that Nap-CoA resulted in an approximate 100-fold higher formation of Nap-glutathione conjugate than Nap-GlcU. The presence of Nap-CoA also resulted in acylated HSA with a rate and a yield that was significantly higher than reported for Nap-GlcU. In summary, the data suggest that CoA metabolites may be more reactive species than acyl glucuronides that previously have been associated with severe drug related side effects in vivo.

Nafenopin-, ciprofibroyl-, and palmitoyl-CoA conjugation in vitro: kinetic and molecular characterization of marmoset liver microsomes and expressed MLCL1

Acyl-CoA conjugation of xenobiotic carboxylic acids is catalyzed by hepatic microsomal long-chain fatty acid CoA ligases (LCL, EC 6.2.1.3). Marmosets (Callithrix jacchus) are considered genetically closer to humans than rodents and are used in pharmacological and toxicological studies. We have demonstrated that marmoset liver microsomes catalyze nafenopin-, ciprofibroyl-, and palmitoyl-CoA conjugation and that only palmitoyl-CoA conjugation is significantly upregulated (1.7-fold, P < 0.02) by a high fat diet. Additionally, the apparent C(50) values for nafenopin-, ciprofibroyl-, and palmitoyl-CoA conjugation of 149.7, 413.4, and 3.4 microM were comparable to those reported for human liver microsomes viz, 213.7, 379.8, and 3.4 microM, respectively. Comparison with human data was enabled by the cloning of a full-length marmoset cDNA (MLCL1) that encoded a 698-amino-acid protein sharing 83% similarity with rat liver acyl-CoA synthetase (ACS1) and 93 and 90% similarity with human liver LCL1 and LCL2, respectively. MLCL1 transiently expressed in COS-7 cells activated nafenopin (C(50) 192.9 microM), ciprofibrate (C(50) 168.7 microM), and palmitic acid (C(50) 4.5 microM) to their respective CoA conjugates. This study also demonstrated that the sigmoidal kinetics observed for nafenopin- and ciprofibroyl-CoA conjugation were not unique to human liver microsomes but were also characteristic of marmoset liver microsomes and recombinant MLCL1. More extensive characterization of the substrate specificity of marmoset LCL isoforms will aid in determining further the suitability of marmosets as a model for human xenobiotic metabolism via acyl-CoA conjugation.

In vitro metabolism of acitretin by human liver microsomes: evidence of an acitretinoyl-coenzyme A thioester conjugate in the transesterification to etretinate

The aromatic retinoid acitretin is the primary active metabolite of etretinate, and in this study we investigated the ethyl esterification of acitretin to etretinate using [(14)C]acitretin and human liver microsomes. Samples were analysed by TLC, HPLC, and LC-MS. Essential requirements for the transesterification reaction were identified and included viable microsomal protein, ATP, CoASH, and ethanol. Human liver microsomes catalysed formation of acitretinoyl-CoA at the rate of 0.08 +/- 0.02 nmol/min/mg (mean +/- SD, N = 10). Acitretinoyl-CoA was pivotal for the transesterification to etretinate and in the presence of methanol, ethanol, n-propanol, n-butanol, and hexanol, the corresponding esters, namely methyl-, ethyl (etretinate)-, propyl-, butyl-, and hexyl-acitretinate, were formed. On average, 1.7% of the acitretin present in the incubation was converted to etretinate in the presence of ethanol. In the absence of ethanol, transesterification did not proceed. Inhibition of the ester hydrolysis of etretinate by bis-p-nitrophenylphosphate (BNPP, 1 mM) prevented futile cycling of etretinate via acitretinoyl-CoA. An additional finding was that acitretin (15-30 microM) activated significantly human liver microsomal long-chain fatty acid-CoA ligase (E.C.6.2.1.3, LCL), resulting in enhanced formation of palmitoyl-CoA. This study demonstrated that in the presence of ethanol the ethyl esterification of acitretin to etretinate proceeds via formation of acitretinoyl-CoA. Predicting clearance of acitretin in vivo via this unique metabolic pathway will be a challenge, as the intracellular concentration of ethanol could never be predicted with any degree of accuracy in humans.

In vitro covalent binding of nafenopin-CoA to human liver proteins

Endogenous fatty acyl-CoAs play an important role in the acylation of proteins. A number of xenobiotic carboxylic acids are able to mimic fatty acids, forming CoA conjugates and acting as substrates in pathways of lipid metabolism. In this study nafenopin, a substrate for human hepatic fatty acid-CoA ligases, was chosen as a model compound to study xenobiotic acylation of human liver proteins. (3)H-nafenopin (+/- unlabeled palmitate) or (14)C-palmitate (+/- unlabeled nafenopin) were incubated for up to 120 min at 37 degrees C with ATP, CoA, and homogenate protein (1 mg/ml) from four individual human livers. Nafenopin covalently bound to proteins was detectable in all human livers and increased with time. Nafenopin adduct formation was directly proportional to nafenopin-CoA formation (r = 0.985, p < 0.05). Attachment of nafenopin to proteins involved both thioester and amide linkages with 76 and 24% of adducts formed with proteins > 100 and 50-100 kDa, respectively. Protein acylation by palmitate was also demonstrated. Palmitate significantly inhibited nafenopin-CoA formation by 29% but had no effect on nafenopin-CoA-mediated protein acylation. In contrast, nafenopin significantly inhibited protein palmitoylation by palmitoyl-CoA. This is the first study to demonstrate a direct relationship between xenobiotic-CoA formation, acylation of human liver proteins, and inhibition of endogenous palmitoylation. The ability of xenobiotics to acylate tissue proteins may have important biological consequences including perturbation of endogenous regulation of protein localization and function.

Effect of chlorophenoxyacetic acid herbicides on glycine conjugation of benzoic acid

1. 2,4-Dichlorophenoxyacetic acid (2,4-D) and 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) (0.1-0.5 mmol/kg i.p.) delayed the disappearance of injected benzoate from blood and diminished the urinary excretion of the formed benzoylglycine, but elevated the blood levels of benzoylglycine in rat, suggesting that these herbicides interfere with both the formation and the renal transport of benzoylglycine. 2. Inhibition of the renal excretion of benzoylglycine by 2,4-D or 2,4,5-T (0.5 mmol/kg i.p.) was directly demonstrated in rat injected with benzoylglycine. 3. Inhibition of benzoylglycine formation from benzoic acid by 2,4-D or 2,4,5-T (0.5 mmol/kg i.p.) was directly demonstrated in renal pedicles-ligated rats injected with benzoate. 4. Neither 2,4-D nor 2,4,5-T influenced the hepatic concentrations of ATP, coenzyme A (CoA) or glycine; therefore, it is unlikely that they inhibit glycine conjugation of benzoic acid by diminishing the availability of co-substrates. 5. Although the chlorophenoxyacetic acids did not appear to be a substrate for the mitochondrial acyl-CoA synthetases, both 2,4-D and 2,4,5-T diminished the activity of benzoyl-CoA synthetase (but not that of benzoyl-CoA:glycine N-acyltransferase) in solubilized hepatic mitochondria. These findings suggest that 2,4-D and 2,4,5-T impair benzoylglycine formation in rat by inhibiting benzoyl-CoA synthetase.

Role of hepatic fatty acid:coenzyme A ligases in the metabolism of xenobiotic carboxylic acids

1. Formation of acyl-coenzymes (Co)A occurs as an obligatory step in the metabolism of a variety of endogenous substrates, including fatty acids. The reaction is catalysed by ATP-dependent acid:CoA ligases (EC 6.2.1.1-2.1.3; AMP forming), classified on the basis of their ability to conjugate saturated fatty acids of differing chain lengths, short (C2-C4), medium (C4-C12) and long (C10-C22). The enzymes are located in various cell compartments (cytosol, smooth endoplasmic reticulum, mitochondria and peroxisomes) and exhibit wide tissue distribution, with highest activity associated with liver and adipose tissue. 2. Formation of acyl-CoA is not unique to endogenous substrates, but also occurs as an obligatory step in the metabolism of some xenobiotic carboxylic acids. The mitochondrial medium-chain CoA ligase is principally associated with metabolism via amino acid conjugation and activates substrates such as benzoic and salicylic acids. Although amino acid conjugation was previously considered an a priori route of metabolism for xenobiotic-CoA, it is now recognized that these highly reactive and potentially toxic intermediates function as alternative substrates in pathways of intermediary metabolism, particularly those associated with lipid biosyntheses. 3. In addition to a role in fatty acid metabolism, the hepatic microsomal and peroxisomal long-chain-CoA-ligases have been implicated in the formation of the acyl-CoA thioesters of a variety of hypolipidaemic and peroxisome proliferating agents (e.g. clofibric acid) and of the R(-)-enantiomers of the commonly used 2-arylpropionic acid non-steroidal anti-inflammatory drugs (e.g. ibuprofen). In vitro kinetic studies using rat hepatic microsomes and peroxisomes have alluded to the possibility of xenobiotic-CoA ligase multiplicity. Although cDNA encoding a long-chain ligase have been isolated from rat and human liver, there is currently no molecular evidence of multiple isoforms. The gene has been localized to chromosome 4 and homology searches have revealed a significant similarity with enzymes of the luciferase family. 4. Increasing recognition that formation of a CoA conjugate increases chemical reactivity of xenobiotic carboxylic acids has led to an awareness that the relative activity, substrate specificity and intracellular location of the xenobiotic-CoA ligases may explain differences in toxicity. 5. Continued characterization of the human xenobiotic-CoA ligases in terms of substrate/inhibitor profiles and regulation, will allow a greater understanding of the role of these enzymes in the metabolism of carboxylic acids.

Mechanistic studies on metabolic chiral inversion of 4-(4-methylphenyl)-2-methylthiomethyl-4-oxobutanoic acid (KE-748), an active metabolite of the new anti-rheumatic agent 2-acetylthiomethyl-4-(4-methylphenyl)-4-oxobutanoic acid (KE-298), in rats

The chiral inversion properties of 4-(4-methylphenyl)-2-methylthiomethyl-4-oxobutanoic acid (KE-748), an active metabolite of 2-acetylthiomethyl-4-(4-methylphenyl)-4-oxobutanoic acid (KE-298), were compared with those of ibuprofen in rats. After administration of R(-)-[2 alpha-2H]KE-748, S(+)-KE-748 was present in the rat plasma, and the deuterium atoms of the S(+)-enantiomer were almost all replaced by hydrogen atoms. After administration of S(+)-[2 alpha-2H]KE-748, the deuterium content of S(+)-KE-748 in the plasma remained intact. In the in vitro study, using a cell-free system and rat liver homogenates, the chiral inversion of ibuprofen was apparent when both CoA and ATP were present; however, KE-748 was not inverted. In the study on isolated rat hepatocytes, the unidirectional chiral inversion from R(-)-to S(+)-enantiomer was observed for both ibuprofen and KE-748. When R(-)-ibuprofen was incubated with medium and long chain fatty acids (carbon chain length C6 to C16), using isolated hepatocytes, the chiral inversion decreased significantly. On the other hand, when R(-)-KE-748 was incubated with short and medium chain fatty acids (carbon chain length C3 to C8), chiral inversion was inhibited markedly. To induce hepatic microsomal long chain fatty acid CoA ligase, rats were treated with clofibric acid (CF rats). In both in vitro and in vivo experiments on CF rats, chiral inversion from R(-)-to S(+)-ibuprofen was enhanced significantly compared with that in controls, whereas the enhancement was not observed in the case of R(-)-KE-748. There was no influence of benzoic acid, a typical substrate on medium chain fatty acid CoA ligase in the mitochondrial matrix, on chiral inversion of R(-)-ibuprofen, using, isolated hepatocytes. In contrast, the chiral inversion from R(-)-to S(+)-KE-748 was strongly inhibited in the presence of benzoic acid. These results indicate that chiral inversion of R(-)-KE-748 may proceed via formation of the CoA-thioester intermediate with loss of the 2 alpha-methine proton, in a manner similar to that seem with R(-)-ibuprofen. However, the enzymes needed to form CoA-thioester of R(-)-KE-748 differ from those for R(-)-ibuprofen.

Species differences in the intracellular distribution of ciprofibroyl-CoA hydrolase. Implications for peroxisome proliferation

Peroxisomal proliferators (HPP), such as ciprofibrate and clofibric acid, are species-specific drugs. Since HPP-coenzyme A derivatives might be involved in their action, we studied the subcellular distribution of liver ciprofibroyl-CoA hydrolase in rat and in two HPP-unresponsive species, humans and guinea pig. Total activity was similar in the three species and was not induced by clofibric acid treatment. In guinea pig, as in humans, the enzyme is localized in the mitochondrial and soluble fractions and no changes are observed after drug treatment. In the rat, the enzyme has a microsomal localization, but upon clofibric acid treatment it changes to a mitochondrial and soluble distribution, as in unresponsive species. These results raise the possibility that drug-induced hydrolases in rats might be normally expressed in humans and guinea pigs.

Kinetic characteristics of rat liver peroxisomal nafenopin-CoA ligase

In this study we have demonstrated that rat hepatic peroxisomes catalyse the formation of nafenopin-CoA. The process is mediated by apparent high affinity (Km 6.7 microM), low capacity (Vmax 0.31 nmol/mg/min) and low affinity, high capacity isoforms. Palmitic acid (Ki 1.1 microM), R(-) ibuprofen (Ki 7.9 microM), ciprofibrate (Ki 60.2 microM) and clofibric acid (Ki 86.8 microM) competitively inhibited nafenopin-CoA formation catalysed by the apparent high affinity isoform. An antibody raised against the microsomal palmitoyl-CoA ligase inhibited the equivalent peroxisomal enzyme significantly (P < 0.001) but did not inhibit peroxisomal nafenopin-CoA ligase activity. These data suggest that nafenopin-CoA formation is catalysed by a peroxisomal CoA ligase which differs from the peroxisomal long chain fatty acid-CoA ligase in relation to its xenobiotic/antibody inhibitor profile and kinetic characteristics.

Differential induction of rat hepatic microsomal and peroxisomal long-chain and nafenopin-CoA ligases by clofibric acid and di-(2-ethylhexyl)phthalate

1. Activity of rat hepatic microsomal and peroxisomal long-chain (palmitoyl) and nafenopin-CoA ligases were studied following administration of either clofibric acid, di-(2-ethylhexyl)phthalate (DEHP) or phenobarbitone. 2. Clofibric acid significantly induced the peroxisomal palmitoyl and nafenopin-CoA ligases, whilst no induction of the equivalent enzymes was observed in the microsomal fraction. 3. DEHP induced only palmitoyl-CoA formation in peroxisomes, whilst all enzymes were refractory to phenobarbitone treatment. 4. The enzyme-specific patterns of inductions both intra- and inter-organelle suggest that the palmitoyl and nafenopin-CoA ligases are under different regulatory control. 5. Modulation of both the rate and extent of nafenopin- and palmitoyl-CoA formation was both agent and organelle specific. 6. This study highlights the difficulty in delineating the individual roles of both fatty acyl-CoAs and xenobiotic-CoAs in peroxisome proliferation.