This article was downloaded by: [University of Arizona] On: 20 April 2015, At: 08:14 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Toxicology and Environmental Health: Current Issues Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uteh19
Toxic agents resulting from the oxidative metabolism of steroid hormones and drugs a
b
E.C. Horning , J.‐P. Thenot & E. D. Helton
c
a
Baylor College of Medicine , Institute for Lipid Research , Houston, Texas, 77030
b
Baylor College of Medicine , Institute for Lipid Research , Houston, Texas
c
National Center for Toxicological Research , Jefferson, Arkansas Published online: 19 Oct 2009.
To cite this article: E.C. Horning , J.‐P. Thenot & E. D. Helton (1978) Toxic agents resulting from the oxidative metabolism of steroid hormones and drugs, Journal of Toxicology and Environmental Health: Current Issues, 4:2-3, 341-361, DOI: 10.1080/15287397809529665 To link to this article: http://dx.doi.org/10.1080/15287397809529665
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Session: STEROIDS: MECHANISMS, METABOLISM, AND RESULTING TOXICITIES
TOXIC AGENTS RESULTING FROM THE OXIDATIVE METABOLISM OF STEROID HORMONES AND DRUGS E. C. Horning, J.-P. Thenot Institute for Lipid Research, Baylor College of Medicine, Houston, Texas E. D. Helton
Downloaded by [University of Arizona] at 08:14 20 April 2015
National Center for Toxicological Research, Jefferson, Arkansas
The oxidative metabolism of some exogenous compounds, and possibly some endogenous compounds as well, can lead to the formation of reactive metabolites. These intermediates react as electrophiles, and they lead in some instances to cell death or cell transformation. Three routes (other routes are also known) of toxicity are discussed. These are the epoxide/dihydrodiol pathway, the catechol/o-quinone pathway, and the alkylation pathway. The possible formation of electrophiles from diethylstilbestrol, from natural estrogens, and from ethynylestradiol is discussed in terms of protein binding. Protein binding is presumptive evidence of electrophile formation, but it does not necessarily indicate that the parent compound is highly cytotoxic, mutagenic, or carcinogenic. Mutagenic and carcinogenic activity is presumed to require reaction of an electrophile with nuclear material. There is evidence for protein binding for these estrogens (diethylstilbestrol, natural estrogens, ethynylestradiol) as a consequence of oxidative metabolism.
INTRODUCTION The fact that numerous exogenous organic chemical compounds, and possibly some endogenous compounds as well, can be converted to cytotoxic, teratogenic, mutagenic, or carcinogenic substances by normal enzymatic reactions is now generally accepted. This was not always true. Pioneering studies were carried out by E. C. Miller and J. A. Miller and by a few other organic chemists and biochemists who were concerned with problems of chemical carcinogenesis and chemical toxicity, and in most of this work it was necessary to postulate the existence of toxic metabolites whose structures were (and in many instances still are) uncertain. Metabolism studies carried out by pharmacologists who were concerned with The work in Houston (E. C. Horning and J.-P. Thenot) was supported by grant GM13901 of the National Institute of General Medical Sciences. Requests for reprints should be sent to E. C. Horning, Institute for Lipid Research, Baylor College of Medicine, Houston, Texas 77030. 341 Journal of Toxicology and Environmental Health, 4:341 -361, 1978 Copyright © 1978 by Hemisphere Publishing Corporation 0098-4108/78/0402-0341 $ 2.25
Downloaded by [University of Arizona] at 08:14 20 April 2015
342
E.C. HORNING ETAL.
drug toxicity were, until recently, dominated by an opposite viewpoint: that the metabolism of drugs and other exogenous compounds led, with few exceptions, to compounds that lacked biologic activity and were essentially excretion products. The title of the classic monograph on drug metabolism by Williams (1947) was Detoxication Mechanisms. Instances of lethal organ damage by a drug were unexplained events. The studies of Mitchell and co-workers (Mitchell and Jollow, 1975; Mitchell et al., 1976, 1977; Nelson et al., 1977) have now thrown much light on metabolic pathways leading to toxicity. There are several reasons for the relative slowness of the biomedical research community to recognize that normal biologic processes can result in the biosynthesis of cytotoxic or carcinogenic substances from apparently innocuous precursors. The chief experimental difficulty in studying these problems is that in many instances the toxic metabolite is a reactive electrophile that does not leave the cell in which it is formed and does not survive in intact form when attempts are made to isolate it. In these instances the structures of the reactive metabolites must be inferred, rather than demonstrated by direct isolation. A second experimental difficulty, now largely removed, was the relatively poor state of organic chemical analytical methodology for studying biologic problems. In early work it was often difficult or impossible to find key intermediates present in low concentration in a biologic matrix. Conceptual difficulties were also present. The nature and site of cellular enzymes responsible for the oxidation of lipid-soluble exogenous compounds were not well defined until recently, and there were few unifying concepts or principles that provided a rational explanation for observed processes of cell destruction or cell transformation resulting from entirely normal oxidative metabolic reactions. Toxicity related to oxidative metabolism is usually explainable on the basis of generation of a reactive electrophile. This metabolite (E in Fig. 1) may react with water, alcohols or phenols, amines, and thiols; in general, these reactions occur with increasing ease, depending on the functional nucleophilic atom, in the order O < N < S . An observed rate of reaction may be a result of the rate of carbocation formation for the parent molecule. The ionic intermediate may be formed by protonation, dehydration, or bond shifts and bond cleavage aided by solvation. For example, epoxides may react through a transient carbocation. In these instances, the structure of the ionic intermediate must be inferred, but the parent molecule (E) may be sufficiently stable that it can be isolated as a biologic reaction product. The protective effect of glutathione, which can usually be demonstrated as a threshold effect, results from cellular reaction of the electrophile with glutathione (as in Fig. 1). Another way in which reactive electrophiles of a different kind may be generated is by neutral radical formation. The carcinogenicity of ester and halide alkylating agents is paralleled by their reaction rates in SN~\
TOXIC AGENTS FROM OXIDATIVE METABOLISM
343
RS-E RSH|
RNH2
RNH-E
ROHJor HoO
Downloaded by [University of Arizona] at 08:14 20 April 2015
RO-E FIGURE 1. Reactive electrophiles (E) generated by oxidative metabolism will react with cellular nucleophiles. Reactions with macromolecules are believed to occur through thiol and amino groups. Enzymatic and nonenzymatic reaction with glutathione provides a protective mechanism against toxicity because of nonenzymatic reaction with cellular macromolecules.
reactions (Lawley, 1976). The same order of rates should hold for neutral alkyl radical formation by reaction of the alkyiating agent with superoxide ions. This mechanism, however, has not been considered for cellular alkylations by such compounds as halides, sulfates, and sulfonates. Addition reactions of o-quinones are known to occur under nonbiologic conditions by reactions shown in Fig. 1, but ft has also been suggested that biologic addition reactions of o-quinones involve the neutral or charged semiquinone radical as an intermediate. In this brief review, the emphasis is on structures and modes of formation and reaction of reactive metabolites, for compounds that are known carcinogens or require study because of their association with cancer in animal models or in humans, and that are related to estrogenic hormones. The types of structures and reactions that are included are those arising from the epoxide/dihydrodiol pathway, with an epoxide as the reactive metabolite and with a postulated ionic reaction mechanism for reaction with cell components; the catechol/o-quinone pathway, with an o-quinone as the reactive metabolite, and with either an ionic or a neutral semiquinone radical reaction mechanism; and alkylation reactions, where a radical may be the intermediate species. A discussion of protein binding of estrogens is also included. A demonstration that covalent protein binding occurs for an exogenous or an endogenous compound during oxidative metabolism in vitro does not implicate the compound as a mutagen, or necessarily indicate that it might be cytotoxic in vivo. Protein binding, however, is presumptive evidence of electrophiie formation in vitro or in vivo. In all discussions of relationships between estrogenic hormones and transformed estrogen-sensitive tissue, it is necessary to differentiate mechanisms involving hormonal stimulation of cellular metabolism,
E.C. HORNING ET AL.
344
without the formation of a reactive toxic substance from the hormone, and mechanisms in which a double effect is postulated—that is, where the hormone may provide a toxic metabolite. If all observed associative effects are results only of hormonal effects on cellular metabolism, it is not likely that an increased risk of cancer should result from any specific estrogen, in excess of that related to the hormonal activity alone. If, however, some synthetic or naturally occurring estrogens are metabolized to toxic intermediates, increased risk and specific toxic effects might be associated with the compounds in question. It is difficult to reach conclusions on the basis of current knowledge, but the questions that require investigation are now better defined and more open to study than a few years ago.
Downloaded by [University of Arizona] at 08:14 20 April 2015
EPOXIDE/DIHYDRODIOL METABOLIC PATHWAY Most lipid-soluble exogenous aromatic compounds, and most compounds with olefinic unsaturation, are metabolized in humans and in animals through epoxide formation. The precise chemical mechanism by which this occurs is not known, but the cellular site of oxidation is the endoplasmic reticulum, and the enzymes that are responsible for the oxidation belong to the cytochrome P-450 class. The oxygen is derived from oxygen of the air. Some metal-catalyzed nonenzymatic oxidations that are believed to involve radicals mimic these enzymatic oxidations. It is possible that the biologic reaction is essentially due to the superoxide ion (O2~) or the corresponding neutral radical ( H O 2 ) :
M + HO2"-> [MO 2 H]"-^MO + HO" or for benzene:
OOH
:o + Her
Transition states of this kind are believed to be involved in some gas phase reactions of superoxide ions; the usual gas phase product is a phenoxide ion (Dzidic et al., 1975). Epoxides show a wide range of chemical reactivity. Some are highly stable substances that can be synthesized without difficulty and can be isolated as metabolites in in vitro or in vivo studies. Others are highly reactive and are difficult to isolate in intact form under any circumstance. It is not likely that a chemically highly reactive metabolite would be found as a microsomal oxidation product or as a urinary, fecal, or biliary
Downloaded by [University of Arizona] at 08:14 20 April 2015
TOXIC AGENTS FROM OXIDATIVE METABOLISM
345
metabolite, and this makes it difficult in some instances to secure direct proof of epoxide formation. The detection of a diol of appropriate structure, however, is now acceptable evidence for an epoxide precursor. Epoxide hydratases (Oesch, 1972) lead to frw75-diols; c/s-diols have been observed as animal metabolites, but these are believed to be of bacterial origin and derived from peroxides rather than from epoxides. Phenols are observed as metabolic products from aromatic hydrocarbons; these are derived from the corresponding epoxides by enzymatic or nonenzymatic reaction (they are also formed as artifacts by acidic treatment of body fluids containing dihydrodiols; acid hydrolysis conditions should not be used). Premercapturic acids, which become the familiar mercapturic acids, are derived from epoxides by enzymatic or nonenzymatic reaction with glutathione. A dihydrodiol from phenytoin (diphenylhydantoin) was described by Chang et al. (1970) and by Horning et al. (1971). This was the first dihydrodiol to be found as a metabolite of a commonly used drug, and since that time dihydrodiols of other drugs and many foreign compounds have been found as human and animal metabolites (Horning et al., 1976). Very few attempts have been made to isolate or to work directly with benzenoid epoxides, but epoxides from naphthalene and higher polycyclic hydrocarbons are more stable. A frans-diepoxide of naphthalene was recently described by Ishikawa and Griffin (1977) as a direct chemical epoxidation product of naphthalene. Epoxides and dihydrodiols of aromatic compounds may be more subject to epoxidation than the original compound. For example, naphthalene can be converted to a diepoxide with no direct evidence of monoepoxide formation; the rate of the reaction leading to the introduction of the second oxygen is greater than the rate of introduction of the first oxygen. Similarly, dihydrodiols can be converted by both biologic and nonbiologic oxidation to dihydrodiol epoxides. These reactions are outlined in schematic form in Fig. 2. Route A leads to a dihydrodiol epoxide by way of a trans-d\epoxide, while route B leads to a dihydrodiol epoxide by way of a dihydrodiol. The two products may be the same or different, but a trans relationship is always present for the diol pair. The study of compounds of this kind as biologic products is of relatively recent origin. When epoxidation was accepted as a route of metabolism for aromatic compounds, it was believed that K-region epoxides (the 9,10oxide of phenanthrene is an example of a K-region epoxide) were the "reactive epoxides" sought as intermediates in the metabolism of polycyclic aromatic hydrocarbons. It was found, however, that K-region epoxides of carcinogenic hydrocarbons were less carcinogenic than the parent compounds, and this hypothesis was abandoned. It is now believed that the reactive epoxides are dihydrodiol epoxides. For benzo[
Journal of Toxicology and Environmental Health: Current Issues Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uteh19
Toxic agents resulting from the oxidative metabolism of steroid hormones and drugs a
b
E.C. Horning , J.‐P. Thenot & E. D. Helton
c
a
Baylor College of Medicine , Institute for Lipid Research , Houston, Texas, 77030
b
Baylor College of Medicine , Institute for Lipid Research , Houston, Texas
c
National Center for Toxicological Research , Jefferson, Arkansas Published online: 19 Oct 2009.
To cite this article: E.C. Horning , J.‐P. Thenot & E. D. Helton (1978) Toxic agents resulting from the oxidative metabolism of steroid hormones and drugs, Journal of Toxicology and Environmental Health: Current Issues, 4:2-3, 341-361, DOI: 10.1080/15287397809529665 To link to this article: http://dx.doi.org/10.1080/15287397809529665
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Session: STEROIDS: MECHANISMS, METABOLISM, AND RESULTING TOXICITIES
TOXIC AGENTS RESULTING FROM THE OXIDATIVE METABOLISM OF STEROID HORMONES AND DRUGS E. C. Horning, J.-P. Thenot Institute for Lipid Research, Baylor College of Medicine, Houston, Texas E. D. Helton
Downloaded by [University of Arizona] at 08:14 20 April 2015
National Center for Toxicological Research, Jefferson, Arkansas
The oxidative metabolism of some exogenous compounds, and possibly some endogenous compounds as well, can lead to the formation of reactive metabolites. These intermediates react as electrophiles, and they lead in some instances to cell death or cell transformation. Three routes (other routes are also known) of toxicity are discussed. These are the epoxide/dihydrodiol pathway, the catechol/o-quinone pathway, and the alkylation pathway. The possible formation of electrophiles from diethylstilbestrol, from natural estrogens, and from ethynylestradiol is discussed in terms of protein binding. Protein binding is presumptive evidence of electrophile formation, but it does not necessarily indicate that the parent compound is highly cytotoxic, mutagenic, or carcinogenic. Mutagenic and carcinogenic activity is presumed to require reaction of an electrophile with nuclear material. There is evidence for protein binding for these estrogens (diethylstilbestrol, natural estrogens, ethynylestradiol) as a consequence of oxidative metabolism.
INTRODUCTION The fact that numerous exogenous organic chemical compounds, and possibly some endogenous compounds as well, can be converted to cytotoxic, teratogenic, mutagenic, or carcinogenic substances by normal enzymatic reactions is now generally accepted. This was not always true. Pioneering studies were carried out by E. C. Miller and J. A. Miller and by a few other organic chemists and biochemists who were concerned with problems of chemical carcinogenesis and chemical toxicity, and in most of this work it was necessary to postulate the existence of toxic metabolites whose structures were (and in many instances still are) uncertain. Metabolism studies carried out by pharmacologists who were concerned with The work in Houston (E. C. Horning and J.-P. Thenot) was supported by grant GM13901 of the National Institute of General Medical Sciences. Requests for reprints should be sent to E. C. Horning, Institute for Lipid Research, Baylor College of Medicine, Houston, Texas 77030. 341 Journal of Toxicology and Environmental Health, 4:341 -361, 1978 Copyright © 1978 by Hemisphere Publishing Corporation 0098-4108/78/0402-0341 $ 2.25
Downloaded by [University of Arizona] at 08:14 20 April 2015
342
E.C. HORNING ETAL.
drug toxicity were, until recently, dominated by an opposite viewpoint: that the metabolism of drugs and other exogenous compounds led, with few exceptions, to compounds that lacked biologic activity and were essentially excretion products. The title of the classic monograph on drug metabolism by Williams (1947) was Detoxication Mechanisms. Instances of lethal organ damage by a drug were unexplained events. The studies of Mitchell and co-workers (Mitchell and Jollow, 1975; Mitchell et al., 1976, 1977; Nelson et al., 1977) have now thrown much light on metabolic pathways leading to toxicity. There are several reasons for the relative slowness of the biomedical research community to recognize that normal biologic processes can result in the biosynthesis of cytotoxic or carcinogenic substances from apparently innocuous precursors. The chief experimental difficulty in studying these problems is that in many instances the toxic metabolite is a reactive electrophile that does not leave the cell in which it is formed and does not survive in intact form when attempts are made to isolate it. In these instances the structures of the reactive metabolites must be inferred, rather than demonstrated by direct isolation. A second experimental difficulty, now largely removed, was the relatively poor state of organic chemical analytical methodology for studying biologic problems. In early work it was often difficult or impossible to find key intermediates present in low concentration in a biologic matrix. Conceptual difficulties were also present. The nature and site of cellular enzymes responsible for the oxidation of lipid-soluble exogenous compounds were not well defined until recently, and there were few unifying concepts or principles that provided a rational explanation for observed processes of cell destruction or cell transformation resulting from entirely normal oxidative metabolic reactions. Toxicity related to oxidative metabolism is usually explainable on the basis of generation of a reactive electrophile. This metabolite (E in Fig. 1) may react with water, alcohols or phenols, amines, and thiols; in general, these reactions occur with increasing ease, depending on the functional nucleophilic atom, in the order O < N < S . An observed rate of reaction may be a result of the rate of carbocation formation for the parent molecule. The ionic intermediate may be formed by protonation, dehydration, or bond shifts and bond cleavage aided by solvation. For example, epoxides may react through a transient carbocation. In these instances, the structure of the ionic intermediate must be inferred, but the parent molecule (E) may be sufficiently stable that it can be isolated as a biologic reaction product. The protective effect of glutathione, which can usually be demonstrated as a threshold effect, results from cellular reaction of the electrophile with glutathione (as in Fig. 1). Another way in which reactive electrophiles of a different kind may be generated is by neutral radical formation. The carcinogenicity of ester and halide alkylating agents is paralleled by their reaction rates in SN~\
TOXIC AGENTS FROM OXIDATIVE METABOLISM
343
RS-E RSH|
RNH2
RNH-E
ROHJor HoO
Downloaded by [University of Arizona] at 08:14 20 April 2015
RO-E FIGURE 1. Reactive electrophiles (E) generated by oxidative metabolism will react with cellular nucleophiles. Reactions with macromolecules are believed to occur through thiol and amino groups. Enzymatic and nonenzymatic reaction with glutathione provides a protective mechanism against toxicity because of nonenzymatic reaction with cellular macromolecules.
reactions (Lawley, 1976). The same order of rates should hold for neutral alkyl radical formation by reaction of the alkyiating agent with superoxide ions. This mechanism, however, has not been considered for cellular alkylations by such compounds as halides, sulfates, and sulfonates. Addition reactions of o-quinones are known to occur under nonbiologic conditions by reactions shown in Fig. 1, but ft has also been suggested that biologic addition reactions of o-quinones involve the neutral or charged semiquinone radical as an intermediate. In this brief review, the emphasis is on structures and modes of formation and reaction of reactive metabolites, for compounds that are known carcinogens or require study because of their association with cancer in animal models or in humans, and that are related to estrogenic hormones. The types of structures and reactions that are included are those arising from the epoxide/dihydrodiol pathway, with an epoxide as the reactive metabolite and with a postulated ionic reaction mechanism for reaction with cell components; the catechol/o-quinone pathway, with an o-quinone as the reactive metabolite, and with either an ionic or a neutral semiquinone radical reaction mechanism; and alkylation reactions, where a radical may be the intermediate species. A discussion of protein binding of estrogens is also included. A demonstration that covalent protein binding occurs for an exogenous or an endogenous compound during oxidative metabolism in vitro does not implicate the compound as a mutagen, or necessarily indicate that it might be cytotoxic in vivo. Protein binding, however, is presumptive evidence of electrophiie formation in vitro or in vivo. In all discussions of relationships between estrogenic hormones and transformed estrogen-sensitive tissue, it is necessary to differentiate mechanisms involving hormonal stimulation of cellular metabolism,
E.C. HORNING ET AL.
344
without the formation of a reactive toxic substance from the hormone, and mechanisms in which a double effect is postulated—that is, where the hormone may provide a toxic metabolite. If all observed associative effects are results only of hormonal effects on cellular metabolism, it is not likely that an increased risk of cancer should result from any specific estrogen, in excess of that related to the hormonal activity alone. If, however, some synthetic or naturally occurring estrogens are metabolized to toxic intermediates, increased risk and specific toxic effects might be associated with the compounds in question. It is difficult to reach conclusions on the basis of current knowledge, but the questions that require investigation are now better defined and more open to study than a few years ago.
Downloaded by [University of Arizona] at 08:14 20 April 2015
EPOXIDE/DIHYDRODIOL METABOLIC PATHWAY Most lipid-soluble exogenous aromatic compounds, and most compounds with olefinic unsaturation, are metabolized in humans and in animals through epoxide formation. The precise chemical mechanism by which this occurs is not known, but the cellular site of oxidation is the endoplasmic reticulum, and the enzymes that are responsible for the oxidation belong to the cytochrome P-450 class. The oxygen is derived from oxygen of the air. Some metal-catalyzed nonenzymatic oxidations that are believed to involve radicals mimic these enzymatic oxidations. It is possible that the biologic reaction is essentially due to the superoxide ion (O2~) or the corresponding neutral radical ( H O 2 ) :
M + HO2"-> [MO 2 H]"-^MO + HO" or for benzene:
OOH
:o + Her
Transition states of this kind are believed to be involved in some gas phase reactions of superoxide ions; the usual gas phase product is a phenoxide ion (Dzidic et al., 1975). Epoxides show a wide range of chemical reactivity. Some are highly stable substances that can be synthesized without difficulty and can be isolated as metabolites in in vitro or in vivo studies. Others are highly reactive and are difficult to isolate in intact form under any circumstance. It is not likely that a chemically highly reactive metabolite would be found as a microsomal oxidation product or as a urinary, fecal, or biliary
Downloaded by [University of Arizona] at 08:14 20 April 2015
TOXIC AGENTS FROM OXIDATIVE METABOLISM
345
metabolite, and this makes it difficult in some instances to secure direct proof of epoxide formation. The detection of a diol of appropriate structure, however, is now acceptable evidence for an epoxide precursor. Epoxide hydratases (Oesch, 1972) lead to frw75-diols; c/s-diols have been observed as animal metabolites, but these are believed to be of bacterial origin and derived from peroxides rather than from epoxides. Phenols are observed as metabolic products from aromatic hydrocarbons; these are derived from the corresponding epoxides by enzymatic or nonenzymatic reaction (they are also formed as artifacts by acidic treatment of body fluids containing dihydrodiols; acid hydrolysis conditions should not be used). Premercapturic acids, which become the familiar mercapturic acids, are derived from epoxides by enzymatic or nonenzymatic reaction with glutathione. A dihydrodiol from phenytoin (diphenylhydantoin) was described by Chang et al. (1970) and by Horning et al. (1971). This was the first dihydrodiol to be found as a metabolite of a commonly used drug, and since that time dihydrodiols of other drugs and many foreign compounds have been found as human and animal metabolites (Horning et al., 1976). Very few attempts have been made to isolate or to work directly with benzenoid epoxides, but epoxides from naphthalene and higher polycyclic hydrocarbons are more stable. A frans-diepoxide of naphthalene was recently described by Ishikawa and Griffin (1977) as a direct chemical epoxidation product of naphthalene. Epoxides and dihydrodiols of aromatic compounds may be more subject to epoxidation than the original compound. For example, naphthalene can be converted to a diepoxide with no direct evidence of monoepoxide formation; the rate of the reaction leading to the introduction of the second oxygen is greater than the rate of introduction of the first oxygen. Similarly, dihydrodiols can be converted by both biologic and nonbiologic oxidation to dihydrodiol epoxides. These reactions are outlined in schematic form in Fig. 2. Route A leads to a dihydrodiol epoxide by way of a trans-d\epoxide, while route B leads to a dihydrodiol epoxide by way of a dihydrodiol. The two products may be the same or different, but a trans relationship is always present for the diol pair. The study of compounds of this kind as biologic products is of relatively recent origin. When epoxidation was accepted as a route of metabolism for aromatic compounds, it was believed that K-region epoxides (the 9,10oxide of phenanthrene is an example of a K-region epoxide) were the "reactive epoxides" sought as intermediates in the metabolism of polycyclic aromatic hydrocarbons. It was found, however, that K-region epoxides of carcinogenic hydrocarbons were less carcinogenic than the parent compounds, and this hypothesis was abandoned. It is now believed that the reactive epoxides are dihydrodiol epoxides. For benzo[
