The Role of Metabolism in Chemical-Induced

Pulmonary Toxicity* 1 BRIAN R. SMITH

AND

WILLIAM R. BRIAN

Department of Drug Metabolism, Smithkline Beecham Pharmaceuticals, P.O. Box 1539, King of Prussia, Pennsylvania 19406 ABSTRACT The lung is a target organ for the toxic effects of several chemical agents, including natural products, industrial chemicals, pesticides, environmental agents, and occasionally, drugs. Factors that establish the lung as a target organ include selective tissue exposure, high tissue oxygenation, and the presence of bioactivating systems that can generate toxic products from initially innocuous substances. Selective pulmonary exposure most often results from the fact that the lung serves as the major portal of entry for most airborne substances, but in some cases, selective exposure is the consequence of accumulation of agents, such as certain basic amines, from the circulation. Lung tumor development following long-term exposure to cigarette smoke or diesel engine exhaust is an example of pulmonary toxicity resulting from selective external exposure. Selective internal exposure, on the other hand, is exemplified by the pulmonary uptake of the herbicide paraquat from the circulation which is in part responsible for its lung-toxic effects. Although the lung contains drug metabolizing enzymes similar to those found in the liver, the enzymatic systems in the lung are sometimes highly concentrated in specific cell types. In the rabbit, for example, the lung-selective toxicity of the natural product ipomeanol is the consequence of relatively large amounts of cytochromes P450 2B1and 4B1in nonciliated bronchiolar epithelial cells (Clara cells) of the terminal airways. These P450 enzymes are highly proficient in vitro at converting ipomeanol to reactive products. Lung tissue contains other enzymic systems which are capable of catalyzing phase I biostransformation pathways (e.g., flavin-containing amine monooxygenase, amine oxidase, and prostaglandin synthase). Examples, however, where pulmonary metabolism by these pathways results in lung toxicity are less numerous than with P450 mediated reactions. Pulmonary prostaglandin H-synthase mediated cooxygenation has been shown to activate procarcinogens such as benzo(a)pyrene 7,8-dihydrodiol, aflatoxin B1, and monosubstituted hydrazines. The activities of pulmonary phase II (conjugation) pathways may also contribute to lung toxicity. Low glutathione transferase activity (relative to P450 mediated aryl hydrocarbon hydroxylase activity) in lung tissue has been suggested to correlate with elevated risk of lung cancer in smokers. Other examples of lung-specific toxic agents and possible causative roles of biotransformation are also discussed.

Key words. Toxicity; cytochrome P450 environment. The

INTRODUCTION

lung may also be extensively ex-

posed internally agents in the systemic circulaas the tion, pulmonary capillary bed receives the to

The lung is a target organ for many chemical substances including pesticides, environmental agents, and in some cases, drugs. Many factors contribute to chemical-induced pulmonary toxicity, including selective exposure. The lung can be extensively exposed to airborne substances such as gases, vapors, and smoke condensates, and this exposure may be greatly intensified by personal habits or the work

entire cardiac output.

Contributing to the extensive to pulmonary exposure circulating agents is the ability of the lung to extract from the plasma and consome basic amines. For exselective ample, uptake of circulating paraquat with the extensive oxygenation of pul(coupled is monary tissue) thought to play an important role in the tissue-selective toxicity of this herbicide. Many toxic chemicals owe their undesirable characteristics to highly reactive metabolic products which are generated within the target organ, a process known as bioactivation. Rarely, chemical bioactivation occurs within the liver and the toxic products are sub-

centrate, agents such as

* Address correspondence to: Dr. Brian R. Smith, Department of Drug Metabolism, SmithKline Beecham Pharmaceuticals, P.O. Box 1539, King of Prussia, Pennsylvania 19406. 1 Presented at the Tenth International Symposium of the Society of Toxicologic Pathologists, &dquo;Pulmonary Toxicologic Pathology,&dquo; June 2-6, 1991 in Monterey, California.

470 Downloaded from tpx.sagepub.com by guest on August 22, 2015

471

sequently transported

to the

target organ via the

species, lung tissue contains chemical-activating enzymic systems which are unique to, or more active in this organ compared with other tissues. There are, for example, pulmonary cytochrome P450-dependent monooxygenases

circulation. In

some

which are known to be effective in vitro in the bioactivation of lung-specific toxic agents. Furthermore, some lung cells such as the nonciliated bronchiolar epithelial cells in the terminal airways (Clara cells) contain relatively large amounts of these P450 dependent monooxygenases. Studies of the pulmonary bioactivation systems have revealed that lung specific cytochrome P450-dependent monooxygenases do in fact play a causative role in establishing the lung as the target organ for several toxic agents. Other enzymatic systems such as the flavin-containing amine monooxygenase system and prosta-

glandin H-synthase-dependent cooxygenation pathways have been implicated in the bioactivation of chemicals, but the direct contribution of these systems to lung toxicity is less clearly established. PULMONARY DRUG METABOLIZING PATHWAYS

Many foreign chemicals or xenobiotics enter the body in inert but unexcretable forms. For example, large polycyclic hydrocarbon molecules such as those found in smoke condensates have little if any biological activity but are very water insoluble, and therefore, the body cannot directly eliminate them. The first step in the metabolic conversion of such agents to excretable forms involves the introduction of a polar functionality such as a hydroxyl group into the chemical structure. The hydroxyl group is subsequently coupled (conjugated) with an endogenous cofactor which usually contains a functional group that is ionized at physiological pH. The ionic functional group facilitates active secretion of the metabolite into the hepatobiliary and/or urinary systems. Introduction of the hydroxyl group is generally referred to as phase I metabolism, while the conjugation step is known as phase II metabolism.

Phase I Metabolism I

Lung tissue is known to contain most of the phase pathways found in the liver, although the enzymes

themselves may be present in amounts different from those in the liver. The lung phase I enzymes often exhibit unique catalytic specificity relative to those in hepatic tissue. Important phase I systems in the lung include the cytochrome P450-dependent monooxygenases, flavin-containing monooxygenases (FMO), the prostaglandin H-synthase (PG Hsynthase) pathway, and epoxide hydrolase. Phase I metabolism is most often catalyzed by the cytochrome P450-dependent monooxygenase system.

Some

nucleophilic chemical structures, however,

may also undergo phase I metabolism via the FMO system. The FMO system, however, has not yet been

clearly shown to be responsible for lung-specific toxicity. The prostaglandin H-synthase pathway has been implicated in the phase I metabolism of several carcinogens in extrahepatic tissues, and this pathway is known to be active in the lung. Epoxide hydrolase, which catalyzes the conversion of epoxides to trans dihydrodiols, can be considered as a phase I reaction, and is involved in the activation of some carcinogens and is present in lungs of most species. Phase II Metabolism Phase II metabolism usually involves conjugation of the phase I products with glutathione, glucuronic acid, sulfuric acid, or in some cases, with amino acids. Lung tissue contains the enzymatic systems which catalyze most of these pathways and the requisite cofactors, but these reactions within the lung itself are not clearly associated with the production of pulmonary toxicity. A possible exception to this generality involves the pulmonary glutathione S-transferases. There is evidence that individuals with low levels of glutathione S-transferase form 1A, which is particularly effective against polycyclic aromatic hydrocarbon epoxides, have enhanced risk to smoke-related pulmonary carcinoma (50). These various phase I and phase II systems are not evenly distributed among the some 40 different lung cell types within the lung, and the contribution of high or low enzyme concentrations in various cell types to lung-specific toxicity has been only partially resolved.

CYTOCHROME P450-DEPENDENT PATHWAYS

Cytochrome P450 mediated metabolism occurs a cyclic monooxygenase pathway which is depicted in Fig. 1. This system resides predominantly within endoplasmic reticulum, but is also present via

to some extent in the nuclear and other membranes.

The first step in the pathway involves binding of the chemical to the enzyme substrate binding site. This site is hydrophobic, and confers some substrate specificity to the enzyme. The next step involves the addition of an electron, supplied by NADPH via the enzyme NADPH-cytochrome P450 reductase (also termed cytochrome C reductase). The reduced complex then combines with molecular oxygen and receives a second electron from P450 reductase, or in some cases from cytochrome bs. The resulting high energy oxygenated/reduced complex rearranges and decomposes to release the oxygenated substrate (usually hydroxylated, sometimes epoxidated)

and water.

Downloaded from tpx.sagepub.com by guest on August 22, 2015

472

11:0

of the cytochrome P450-dependent monooxygenase system. Two electrons are furnished by NADPH cytochrome P450 reductase. In some cases, the second electron is supplied by NADH cytochrome b,. The components of the monooxygenase system are membrane bound, and reside largely in the endoplasmic reticulum. (Redrawn from Guengerich and McDonald (28).) FIG. 1.-The

cycle

nature

Enzymology The major

P450 enzymes involved in chemical bioactivation have molecular weights ranging from 45,000 to 60,000 daltons. The enzymes are anchored in the membrane by a hydrophobic N-terminal &dquo;tail,&dquo; with the remainder of the enzyme situated in the cytoplasm (43). The P450s contain a heme group in the active site, bound through a cysteinyl sulfur group located in the C-terminal portion of the enzyme (49). This cysteine is surrounded by a highly conserved amino acid sequence, containing approximately 25 amino acids, found in all P450 enzymes (23). The P450 enzymes

are encoded by a superfamily (meaning that they share some structural similarity) (40), and it is estimated that 60-200 dif-

of genes

ferent P450 enzymes are present within most mammalian species (24). To date, many of the enzyme forms studied for their involvement in chemical activation have been isolated from liver. Of the extrahepatic P450 sources, the lung has been one of the most extensively studied. Related enzymes isolated from lung and liver within a species appear to be highly similar in structural characteristics such as relative mobility during sodium dodecyl sulfate polyacrylamide gel electrophoresis, immunochemical reactivity, and in most cases catalytic specificity (27). However, the regulation of enzyme expression varies considerably between lung and liver (27). When considered collectively as a group of enzymes, the P450s exhibit broad substrate specificity and are capable of metabolizing many structurally

diverse compounds (24). Individually, P450 enzymes often exhibit unique substrate specificities, and can sometimes be identified based on the stereoand regio-selective reactions they catalyze with certain compounds (see reference 58 for examples). There are, however, many other examples where 2 or more enzymes show overlapping abilities to metabolize the same compound to the same product(s). P450 enzymes are regulated at the transcriptional and translational levels (24), and regulation is often tissue and even cell-type specific. The cell-specific regulation is especially important in the lung with its diversity of cell types. Three populations of lung cells, Clara, type II, and macrophages, have been studied extensively for P450 content, and have been found to differ in the quantity and types of P450 enzymes they contain (1, 18). Regulation of lung P450 expression between species is also variable. Some of the P450 enzymes are induced (i.e., levels of protein are increased) by the administration of chemicals to animals. Phenobarbital, 3-methylcholanthrene (3MC), and 2,3,7,8-tetrachlorodibenzodioxin (TCDD) are well characterized inducers of hepatic P450s in animals, and also exhibit the ability to induce some lung P450s, although differences exist between the tissues (12, 22).

P450 Nomenclature

Initially, P450 enzymes were named based on the wavelength of the absorption maxima of the reduced enzyme in the presence of carbon monoxide (450 nm), together with whether they were constitutive (present normally in the animal) or produced in response to pretreatment with 3MC. The constitu-

tive enzyme, which could be increased several-fold

by treatment of the animals with phenobarbital, was named P450, while the enzyme induced by 3MC treatment was named P,-450 (54, and references therein). 3MC treatment also caused the absorption maxima of the carbon monoxide spectrum to shift consequently, this enzyme became widely known as P448. During the decade following these observations, protein purification techniques resulted in the isolation and characterization of many unique cytochromes P450. These studies revealed that liver from most species contained several P450 enzymes, and they were named based on their electrophoretic motility, substrate specificity, immunologic characteristics, or on the inducing agent responsible for their synthesis. As many laboratories isolated and characterized P450 enzymes, several arrays of names appeared in the literature. By the mid 1980s, effective communication between different laboratories was becoming a challenging endeavor. In an effort to reduce the confusion, a systematic nomenclature was introduced in 1987 based to 448 nm, and

Downloaded from tpx.sagepub.com by guest on August 22, 2015

473 TABLE

FIG. 2.-Cytochrome P450 nomenclature based on gene structure. The first number designates the gene family, the capital letter designates the subfamily, and the second number identifies the specific protein. Family and subfamily members must, respectively, show at least 40% or 59% amino acid sequence similarity. This figure illustrates the name for the major human liver enzyme catalyzing nifedipine oxidation and steroid 6#-hydroxylation.

L-Pulmonary cytochrome

~

~

~

P450

~~~

isozymes.

--

originated by Guengerich and co-workers (27). originated by Philpot and co-workers (11, 19, 22). Cytochrome not detected in lung of this species.

nomenclature b Nomenclature c

and/or substrate specificity (26, 34). As a result, nearly identical enzymes may have quite disparate catalytic properties; cytochrome P450s 4B 1 from rat and rabbit lung N-hydroxylate 2-aminofluorene, while human lung 4B 1 apparently does not (41, 56). It is also evident that levels of very similar P450s are regulated differently in lung and liver (12, 22).

the gene structure coding for the P450 proteins (40). Further refinements were made in 1989 and 1991 (41, 42). This approach was facilitated by the implementation of molecular biology techniques whereby the protein amino acid sequence can be

Lung P450 Enzymes

deduced from the cDNA sequence. All P450s contain a conserved primary amino acid sequence localized around the heme binding site, indicating that they belong to a superfamily of enzymes. This superfamily is divided into several different families, with proteins sharing ~:40% amino acid sequence similarity considered to belong to the same family. Families are further divided into subfamilies, with members sharing a 59% sequence similarity. Families are designated by numbers, and subfamilies are identified by a capital letter. Individual proteins are also assigned a number. An example of the utility of the gene product-based nomenclature system can be provided by the major human liver enzyme catalyzing nifedipine oxidation and steroid 6 ~3-hydroxylation. This enzyme has in the past been called P450NF, P450 human-1, and hPCNI, as well as other names (see reference 7 for examples), but with the new nomenclature is referred to as P450 3A4 (see Fig. 2). Previously, roman numerals were used to designate the gene family, but such a large number of gene families have been identified that less cumbersome arabic numbers are now used (e.g., 3A4 was previously IIIA4). The development of this systematic nomenclature has assisted greatly the comparisons of enzymes isolated in different laboratories. A summary of pulmonary cytochromes P450 and their alternate names are summarized in Table I. There are, however, some disadvantages associated with the gene product-based nomenclature. The names are not descriptive of properties such as substrate specificity, inducibility, or tissue location. Subtle differences in amino acid sequence, for example, can have dramatic effect on catalytic activity

The characterization of pulmonary P450 enzymes and their roles in xenobiotic metabolism have been complicated by the diversity of cell types within the lung and hampered by the relatively low abundance of P450 protein in this tissue. Studies based on microsomal enzyme activities and on enzyme purification suggest that lung contains fewer P450 isozymes than does liver. A summary of P450 cytochromes isolated from rat, rabbit, and human lung and their more salient features are summarized in Table II. Several lines of evidence indicate that the average concentration of P450 in lung tissue is low (estimated to be about 10% of the amount found in liver on a per gram of microsomal protein basis: 27), but importantly, some P450 enzymes appear to be highly concentrated in certain lung cell types. Localization of P450 enzymes and associated activities in lung tissue have been studied in various ways, including isolation and characterization of individual P450 enzymes (27), immunohistochemical localization of enzymes in lung tissue (1), and work with isolated pulmonary cells in vitro ( 13). The results from these studies indicate that the high cytochrome concentration in pulmonary Clara and Type II epithelial cells may play an important role in establishing the lung as a target organ for many lung-specific agents. Rabbit Lung. Lung tissue from untreated rabbits contains higher concentrations of P450 enzymes compared to other species, and therefore, rabbit pulmonary P450s have been extensively studied. Three P450 enzymes have been isolated from lung of uninduced rabbits, and are commonly referred to as forms 2, 5, and 6, or 2B 1, 4B 1, and 1 A 1, respectively. Cytochromes 2B 1 (Fonm 2) and 4B 1 (form

on

Downloaded from tpx.sagepub.com by guest on August 22, 2015

474 TABLE IL-Properties of P450 cytochromes in rat, rabbit, and human lung.

3-Methylcholanthrene, benzo(a)pyrene, #-naphthoflavone, 2,3,7,8-tetrachlordibenzo-p-dioxin. a

5) each constitute between 30 and 40% of total lung P450 (16). Cytochrome 2B1 (form 2) catalyzes

d-benzphetamine N-demethylase (45, 25), p-nitroanisole O-demethylation (11), and 3-methylindole bioactivation (62, 30). Cytochrome 4B 1 (Form 5) catalyzes the N-hydroxylation of 2-aminofluorene (47) and other aromatic amines. Cytochromes 2B i and 4B 1 (forms 2 and 5) are responsible for the metabolism of 4-ipomeanol (61). P450s related to 2B 1 and 4B 1 have been identified in uninduced rabbit liver, but are present at much lower levels than in lung (27). Cytochrome IAI (form 6) in rabbit lung is the major enzyme catalyzing ethoxyresorufin O-deethylation, and the metabolism of benzo(a)pyrene and other polycyclic hydrocarbons (19). This enzyme accounts for only 1-3% of the total pulmonary P450 in untreated rabbit, but is increased about 20-fold in lungs from TCDD-induced rabbits (18). Another P450, form 4, has been identified as a minor form in lung of TCDD-induced rabbits, but it has not yet been characterized (12). Rat Lung. Four cytochrome P450 enzymes have been identified in rat lung (2B 1, 4B 1, 1 A 1, and 3A2). The rat ortholog to rabbit form 2, 2B 1 (trivial name PB-B) is structurally similar to the major phenobarbital-inducible P450 in rat liver, although the lung enzyme is not inducible by phenobarbital. This form catalyzes the metabolism of 4-ipomeanol in

lung (25). Rat 4B1 is 85% structurally similar to rabbit 4B 1 (form 5), and catalyzes the N-hydroxylation of 2-aminofluorene (56, 22). The enzyme is present in lung at low levels in untreated, phenobarbital, and TCDD rats (56). Rat lung I A 1 (trivial name l3NF-B), the ortholog to rabbit 1 A (form 6) is present in very low amounts in untreated rats but is induced dramatically by polycyclic hydrocarbons, PCBs, and TCDD (27). At least one P450 enzyme belonging to the 3A family has been identified in rat lung, but only by immunochemical means (57). Orthologs of the 3A enzymes in rat and human liver are involved in the metabolism of a diverse array of drugs, steroids, and procarcinogens (7, 39) so it is of interest to determine the catalytic activity of this enzyme in lung. Human lung. Human lung has been shown to contain only very low levels of cytochrome P450 enzymes (38) compared with other species. Immunoblot analysis of lung microsomes using antibodies cross reactive with human liver P450s 1A2, 2C, 2D, and 3A showed no evidence of these proteins in lung (27). Using a monoclonal antibody, Wheeler et al. (59) were able to identify P450 1A1 in human lung microsomes of 19 subjects. The amounts of lAl protein correlated well with the ethoxyresorufin O-deethylase activities in these lung samples. Interestingly, however, there was no apparent correlation between the amounts of 1 A 1 protein and cigarette smoking. A cDNA clone has been isolated from a human lung library that corresponds to the rabbit 4B family (form 5). However, when this clone was inserted into a vaccinia virus expression system, the expressed enzyme did not catalyze 2-aminofluorene N-hydroxylation or benzo(a)pyrene metabolism (44). This observation is not, however, definitive evidence that the native enzyme is incapable of catalyzing these reactions. rat

isolated Cells

Pulmonary metabolism has also been studied using isolated lung cells. Much interest in Clara cells was generated early on, when it was observed that this cell type was specifically damaged by some systemically administered chemical agents (5, 62). Methods exist for the isolation of enriched populations of viable Clara and type II epithelial cells, and aleveolar macrophages; all 3 cell types have been shown to contain monooxygenase activity and are capable of xenobiotic metabolism (13). Immunochemical analysis of rabbit pulmonary cells (18) showed that Clara and type II cells contain P450 2B and 4B l, with concentrations 2-3-fold higher in Clara cells than in whole tissue. These P450 forms were found in alveolar macrophages as well, but at levels of 5-10%, or less, of those in Clara cells. Cy-

Downloaded from tpx.sagepub.com by guest on August 22, 2015

475

tochrome P450 reductase was also observed in these 3 cell populations, with activity in Clara cells approximately 2- and 5-fold higher than in type II cells or alveolar macrophages, respectively. Clara and type II cells from untreated rabbit lung were found to contain P450 1 A 1, but this form was below the limit of detection in alveolar macrophages. TCDD administration to rabbits resulted in a 20-fold increase in I A 1 in Clara and type II cells, and an apparent 90-fold increase in 1A1 in macrophages (13, 18). Cytochrome lAl appears to be distributed differently than 2B 1 and 4B 1 between the cell types, with most of the pulmonary 1 A residing in cell populations other than Clara cells, type II cells, or macrophages. The amount of enzyme activity found within each cell type, and the distribution of enzyme activities between the 3 cell types was found to be

species-dependent (13).

Examples of P450-Mediated Bioactivation of Pulmonary Toxicants Consumption of contaminated sweet potatoes by cattle or laboratory animals has long been known to induce a hemorrhagic, pulmonary edema which be lethal. This observation led to the isolation and characterization of several terpenoid natural products that are produced by sweet potatoes in response to infection with Fusarium solani mold. The principal pulmonary toxicant was identified as a furanosesquiterpene that was named ipomeanol (4). It was further determined that the Clara cells of the terminal airways were the specific target for the sweet potato toxin (5), that radiolabelled impomeanol was extensively covalently bound to these cells (5), and that P450-dependent bioactivation was required for the covalent binding (6) (see Fig. 3 for the proposed bioactivation mechanism). More than decade of work by several investigators has provided evidence that ipomeanol is activated to a reactive dialdehyde (Fig. 3) (46) by cytochrome P450 forms 2B1 and 4B2, which are major forms in the lung (61), and present in high concentrations in Clara cells (13). As Clara cells are a potential origin of bronchogenic carcinomas, ipomeanol is under clinical investigation as an antineoplastic agent against can

lung cancer (10). The relocation of cattle or other grazing animals from poor to lush pastureland is also known to occasionally induce a pulmonary distress syndrome in these animals. The likely causative agent for this syndrome is 3-methylindole, which is produced from tryptophan by the ruminal Lactobacilus bacteria which bloom in response to the improved feed quality (29). Bioactivation of 3-methylindole by the P450 system is required for toxicity, and Clara cells and alveolar epithelial cells are the specific target of the reactive metabolite(s) (references 8 and 2; see Fig.

FIG. 3.-Cytochrome P450-dependent bioactivation of Ipomeanol. The pulmonary toxin 4-Ipomeanol is a furanosesquiterpene isolated from moldy sweet potatoes. The lung specific toxicity of this agent is likely the result of reactive epoxide or dialdehyde formation by cytochromes P450 2B and/or 4B in Clara cells.

4 for the

proposed bioactivation mechanism). Re-

cent evidence

suggests that P450 2B

or

the ruminal

orthologs are the primary enzymes involved in 3-methylindole activation in rabbits and goats (30, 62). Pulmonary lesions and pulmonary hypertension are produced in animals grazing on various species of plants belonging to the Crotalaria and Sencicio genera. The pulmonary toxicity is due to pyrrolizidine alkaloids which undergo bioactivation to alkylating agents. Studies in rats, however, indicate that the bioactivation is mediated by cytochrome 3A2 which is present predominantly in liver (60). Data reported by Roth and co-workers strongly support the hypothesis that monocrotaline (a widely studied pyrrolizidine alkaloid) is bioactivated to a pyrrole in the liver and subsequently transported via the circulation to the lungs where damage to the vascular endothelium occurs (48). Other studies suggest that hepatic formation of a reactive glutathione conjugate from monocrotaline might also play

Downloaded from tpx.sagepub.com by guest on August 22, 2015

476

polycyclic aromatic hydrocarbons, but exposure of most species to such agents is known to induce the activity of this enzyme several-fold. In summary, it appears that the predominant pulmonary cytochromes ( 1 A 1, 2B2, and 4B2) are, individually or collectively, important in bioactivating lung-specific toxic agents. Other lung selective toxic agents in which P450-mediated bioactivation may be involved include naphthalene, O,O,S-tri-

methyl phosphorodithioate, trichloroethylene, alkyl butylated hydroxy toluene.

N-nitrosoamines and

FLAVIN-CONTA1NING MONOOXYGENASE-DEPENDENT PATHWAYS

Flavin-containing monooxygenase (FMO) is another enzyme(s) that also catalyzes the oxidation and possibly the bioactivation of chemicals (63). Like the P450-dependent monooxygenases, this system is localized predominantly in the endoplasmic reticulum, is NADPH dependent, and involves a cyclic, multistep process (Fig. 6). In contrast the P450-dependent system, however, an activated peroxy-flavin enzyme intermediate is generated before

FIG. 4. - Bioactivation of the pulmonary toxin 3-methIndoleacetic acid the product of oxidative deamination of the amino acid tryptophan by the ruminal flora. Lactobacillus species convert the indoleacetic acid to 3-methylindole which is subsequently converted to the reactive methylene imine by orthologs of rat P450 2B1 in Clara and Type II epithelial cells. (Redrawn from ref-

ylindole.

erence

53.)

substrate binding occurs. In order for the oxygenation to take place, the substrate must be adequately nucleophilic to react with the peroxy-flavin intermediate (32). This requirement limits FMO substrates largely to basic amines, divalent sulfur groups, and trivalent phosphorus compounds (32). Also in contrast to P450-dependent monooxygenases, there are no known inducers of the FMO system. This system is located in both liver and lung, and it appears that in most laboratory animals the pulmonary enzymes differ from the hepatic enzymes (35, 55). Although the pulmonary FMO system is known to actively metabolize several drugs and other chemicals, there are few examples where bioactivation by this system has been shown to play a critical role in establishing the lung as a specific target for toxic agents. The pneumotoxic effects of thioamides, and possible thioureas, may be related to bioactivation in the lung by this pathway (9). PROSTAGLANDIN H-SYNTHASE-DEPENDENT

role in the pulmonary toxicity of pyrrolizidine alkaloids (31). Activation of polycyclic aromatic hydrocarbons such as benzo(a)pyrene to carcinogenic forms involves biotransformation by both cytochrome P450 1 A 1-dependent monooxygenases and epoxide hydrolase (Fig. 5). Although selective exposure of the lung to smoke condensates plays an important role in establishing the lung as a primary target for such carcinogens, lung tissue from most animals contains more 1 A 1 cytochrome than does liver or other tissues. Not only is this cytochrome adept at activating

PATHWAYS

a

Although the importance of P450-mediated bioactivation can hardly be overemphasized, the fact that tissues such as skin which contain almost unmeasurable P450 levels are targets for chemical carcinogens which require bioactivation is intriguing. This apparent discrepancy was at least partially resolved in 1975 by a report by Marnett et al (36). This group reported that during the oxidation of arachidonic acid (AA) by the prostaglandin H-synthase (PG H-synthase) pathway, chemicals such as benzo(a)pyrene that were included in the incubation

Downloaded from tpx.sagepub.com by guest on August 22, 2015

477

FIG. 6.-The

cyclic nature of the flavin-containing (FMO) system. The FMO system, and its components are bound to the endoplasmic reticulum. monooxygenase

The enzyme is activated to a reactive flavin hydroperoxide prior to substrate binding. In order for a particular substrate to be oxidized by this system, it must be capable of binding to the enzyme active site, and must also be sufficiently nucleophilic to react with the flavin hydroperoxide. Most substrates are basic amines, phosphines, and divalent sulfur compounds. (Redrawn from reference

63.)

prostacyclin, and prostaglandins E2 and F2a (Fig. 7). The first or cyclooxygenase step catalyzes the addition of oxygen to AA to form the cyclic endoperoxide hydroperoxide known as PGG2. The peroxidase activity of PG H-synthase subsequently reduces the hydroperoxide to generate PGH,. The reductase step has an electron requirement which can be fulfilled by a wide variety oflipophilic agents with relatively low oxidation potentials. It is this requirement which confers PG H-synthase with cooxygenase activity. Aromatic amines and phe-

anes,

FIG. 5.-Bioactivation of benzo(a)pyrene by the cytochrome P450 dependent monooxygenase system or by the

prostaglandin H-synthase pathway. Benzo(a)pyrene 7,8dihydrodiol 9,10-epoxide is a potent carcinogen generated by a multi-step pathway requiring both cytochrome P450 and epoxide hydrolase. The formation of the 9,10-epoxide is also catalyzed by peroxyl radicals in the presence of hematin or hematin-containing proteins such as PHS. Prostaglandin H-synthase may play an additional role in this process by generating lipid hydroperoxides which are precursors to the peroxyl radicals (see Fig. 8). Only one of the two possible diastereomers of benzo(a)pyrene 7,8dihydrodiol is illustrated above, and the stereochemistry of the oxirane ring is not addressed in the figure. See reference 14 for stereochemical details.

mixtures a

were

also oxidized. Arachidonic acid

was

requisite component of this pathway, and this pro-

became known as &dquo;cooxidation&dquo; of &dquo;cosubstrates.&dquo; The PG H-synthase pathway has also been shown to be active in lung tissue (52), and at least 2 mechanisms are involved in AA-dependent coox-

cess

ygenation. Peroxidase-Mediated Mechanisms PG H-synthase has 2 distinct catalytic functions in the production of PGH2, the immediate precursor to a variety of prostanoids including thrombox-

nolic compounds are excellent electron donors for the peroxidase step, with these agents being oxidized during this process. The carcinogen 2-aminofluorene is an example of an electron donor which is converted to DNA-reactive (33) and highly mutagenic (3) derivatives during the cooxygenation process. Activation of 2-aminofluorene by PG H-synthase-mediated cooxygenations, however, is probably more relevant to bladder tissue than to lung. On the other hand, the lung toxin 3-methylindole is converted to electrophilic product(s) by arachidonic acid-mediated cooxygenation by goat lung microsomes (21 ).

Hydroperoxide-Mediated Oxidation Several agents that are poor electron donors are known to undergo AA-dependent cooxygenation via PG H-synthase. For example, benzo(a)pyrene 7,8dihydrodiol is oxidized to the corresponding 7,8dihydrodiol 9,10 epoxide by PG H-synthase in the presence of AA or lipid hydroperoxides (Fig. 5) (37,

Downloaded from tpx.sagepub.com by guest on August 22, 2015

478

FIG. 7.-Generation of prostaglandin H2 from arachidonic acid by the prostaglandin H-synthase (PHS) pathway. Prostaglandin H-synthase carries out both the cyclooxygenase and peroxidase steps, which are two distinct catalytic processes. The peroxidase step requires two electrons which can be supplied by any of a variety of easily oxidized, lipophilic agents. Phenols and aromatic amines are examples of excellent electron donors in this reaction, and these substances are oxidized in the process. This pathway, which may play an important role in the bioactivation of carcinogens in extrahepatic tissues, is active in lung. Lipid hydroperoxides generated by PHS may also play a role in the bioactivation of carcinogens (see Fig. 5). (Redrawn from reference 20.)

51 ). Studies carried out by Dix and Mamett (14, 16, 17) indicated that this reaction involves a rather complex mechanism in which the epoxide is generated from a chemical reaction between the 7,8dihydrodiol and peroxyl radicals. The peroxyl radicals are produced by oxidation oflipid hydroperoxides by the hematin prosthetic group of PG H-synthase. This reaction, which is catalyzed equally well by hematin in place of PG H-synthase, is summarized in Fig. 8 ( 14). The AA-mediated epoxidation shows PG H-synthase dependence as AA must first be converted to a lipid hydroperoxide which requires PG H-synthase (see Fig. 7). Although the conversion of benzo(a)pyrene to the 7,8-dihydrodiol requires both cytochrome P 450 ( 1 A 1 ) and epoxide hydrolase, hematin/peroxyl radical-mediated generation of the highly carcinogenic diol epox-

FIG. 8.

-

Lipoxygenase-dependent generation of peroxyl

radicals. Fatty acids containing 2 double bonds separated by 1 carbon atom may be converted to lipid hydroperoxides by a variety of lipoxygenase enzymes (including PG H-synthase). Lipid hydroperoxides, in the presence of hematin or hematin containing proteins (such as PG H-synthase) can rearrange to peroxyl radicals capable of converting carbon-carbon double bonds to epoxides. This reaction sequence has been shown to generate the potent

carcinogen benzo(a) pyrene 7,8-dihydrodiol 9,10-epoxide from the 7,8-dihydrodiol (36, 51). (Redrawn from reference 20.) ide may have important biological implications in extrahepatic tissues such as lung. This importance is underscored by the observation that activation of

benzo(a)pyrene 7,8-dihydrodiol to the carcinogenic diol-epoxide occurs at an accelerated rate during lipid peroxidation of rat liver microsomes (15). SUMMARY

lung is clearly a target organ for many toxic agents. The lung specific toxicity is related to a variety of factors which include, among other things, metabolism of the agent within the lung itself. Although the lung seems equipped to carry out most biotransformation reactions, two factors appear prominent in establishing the lung as a target organ: The

Downloaded from tpx.sagepub.com by guest on August 22, 2015

479

1) lung contains 2 lung-specific cytochromes (2B 1 and 4B 1 ) which are very efficient in activating some chemical forms; 2) these cytochromes appear to be highly concentrated in the noncilliated bronchiolar epithelial (Clara) cells of the terminal airways. Other factors, such as sequestration of detoxication enzymes or required cofactors away from chemical activation sites may eventually be shown to play a role in lung-specific toxicity. ACKNOWLEDGMENTS We would like to thank the following individuals for very helpful and informative discussions: James Bond, Michael Boyd, Theodora Devereux, Thomas Eling, Ryan Huxtable, Christobal Miranda, Richard Philpot, Robert Roth, Cosette Serabjit-Singh, and Gerald Yost. REFERENCES 1. Baron J and Voigt JM (1990). Localization, distribution, and induction of xenobiotic-metabolizing enzymes and aryl hydrocarbon hydroxylase activity within lung. Pharmac. Ther. 47: 419-445. 2. Becker GM, Breeze RG, and Carlson JR (1984). Autoradiographic evidence of 3-methylindole covalent binding to pulmonary epithelial cells in the goat. Toxicology 31: 109-121. 3. Boyd JA and Eling TE (1987). Prostaglandin H synthase-catalyzed metabolism and DNA binding of 2-naphthylamine. Cancer Res. 47: 4007-4014. 4. Boyd MR and Wilson BJ (1972). Isolation and characterization of 4-ipomeanol, a lung-toxic furanoter-

penoid produced by sweet potatoes ( Ipomoea bata). J. Agric. Food Chem. 20: 428-430. tas 5. Boyd MR (1977). Evidence for the Clara cell as a site of cytochrome P450-dependent mixed-function oxidase activity in lung. Nature 269: 713-715. 6. Boyd MR, Burka LT, Wilson BJ, and Sasame HA (1978). In vitro studies on the metabolic activation of the pulmonary toxin, 4-ipomeanol, by rat lung and liver microsomes. J. Pharmacol. Exp. Ther. 207: 677686. 7. Brian WR, Sari M-A, Iwasaki M, Shimada T, Kaminsky LS, and Guengerich FP (1990). Catalytic activities of human liver cytochrome P-450 IIIA4 expressed in Saccharomyces cerevisiae. Biochemistry 29: 11280-11292. 8. Carlson JR and Yost GS (1989). 3-Methylindole-induced acute lung injury resulting from ruminal fermentation of tryptophan. In: Toxicants of Plant Origin, PR Cheeks (ed). CRC Press, Bota Raton, FL, 9.

pp. 108-123. Cashman JR, Traiger GJ, and Hanzlik

RP

(1982).

Pneumotoxic effects of thiobenzamide derivates.

Toxicology 23: 85-93. 10. Christian MC, Wittes RE, Leyland-Jones B, McLemore TL, Smith AC, Grieshaber CK, Chabner BA, and Boyd MR (1989). 4-Ipomeanol: A novel investigational new drug for lung cancer. J. Natl. Cancer Instit. 81: 1133-1143.

11. Croft JE, Harrelson WF, Parandoosh Z, and Philpot RM (1986). Unique properties of NADPH- and

NADH-dependent metabolism nitroanisole of p- catalyzed by cytochrome P-450 isozyme 2 in pulmonary and hepatic microsomal preparations from rabbits. Chem.-Biol. Interact. 57: 143-160. 12. Dees JH, Masters BSS, Muller-Eberhard U, and Johnson EF (1982). Effect of 2,3,7,8-tetra-chlorodibenzo-p-dioxin and phenobarbital on the occurrence and distribution of four cytochrome P-450 isozymes in rabbit kidney, lung, and liver. Cancer Res. 42: 1423-1432. 13. Devereux TR, Domin BA, and Philpot RM (1989). Xenobiotic metabolism by isolated pulmonary cells. Pharmacol. Ther. 41: 243-256. 14. Dix TA and Mamett LJ (1981). Free radical epoxidation of 7,8-dihydroxy-7,8-dihydrobenzo(a)pyrene by hematin and polyunsaturated fatty acid hydroperoxides. J. Am. Chem. Soc. 103: 6744-6746. 15. Dix TA and Mamett LJ (1983). Metabolism of polycyclic aromatic hydrocarbon derivatives to ultimate carcinogens during lipid peroxidation. Science 221: 77-79. 16. Dix TA and Mamett LJ (1985). Conversion of linoleic acid hydroperoxide to hydroxy, keto, epoxyhydroxy, and trihydroxy fatty acids by hematin. J. Biol. Chem. 260: 5351-5357. 17. Dix TA and Mamett LJ (1985). Hematin-catalyzed epoxidation of 7,8-dihydroxy-7,8-dihydrobenzo(a)pyrene by polyunsaturated fatty acid hydroperoxides. J. Biol. Chem. 260: 5358-5365. 18. Domin BA, Devereux TR, and Philpot RM (1986). The cytochrome P-450 monooxygenase system of rabbit lung: Enzyme components, activities, and induction in the noncilliated bronchiolar epithelial (Clara) cell, Alveolar type II cell, and alveolar macrophage. Mol. Pharmacol. 30: 296-303. 19. Domin BA and Philpot RM (1986). The effect of substrate on the expression of activity catalyzed by cytochrome P-450: Metabolism mediated by rabbit isozyme 6 in pulmonary microsomal and reconstituted monooxygenase systems. Arch. Biochem. Biophys. 246: 128-142. 20. Eling TE, Thompson DC, Foureman GL, Curtis JF, and Hughes MF (1990). Prostaglandin H synthase and xenobiotic oxidation. Annu. Rev. Pharmacol. Toxicol. 30: 1-45. 21. Formosa PJ and Bray TM (1988). Evidence for metabolism of 3-methylindole by prostaglandin H synthase and mixed-function oxidases in goat lung and liver microsomes. Biochem Pharmacol. 37: 43594366. 22. Gasser R and

Philpot RM (1989). Primary structures

of cytochrome P-450 isozyme 5 from rabbit and rat and regulation of species-dependent expression and induction in lung and liver: Identification of cytochrome P-450 gene subfamily IVB. Mol. Pharmacol. 35: 617-625. 23. Gonzalez FJ (1989). The molecular biology of cytochrome P450s. Pharmacol. Rev. 40: 243-288. 24. Gonzalez FJ and Nebert DW (1990). Evolution of

Downloaded from tpx.sagepub.com by guest on August 22, 2015

480 the P450 gene superfamily: Animal-plant ’warfare,’ molecular drive and human genetic differences in drug oxidation. Trends Genet. 6: 182-186. 25. Guengerich FP (1977). Preparation and properties of highly purified cytochrome P-450 and NADPH-cytochrome P-450 reductase from pulmonary microsomes of untreated rabbits. Mol. Pharmacol. 13: 911923. 26. Guengerich FP (1987). Enzymology of rat liver cytochrome P-450. In: Mammalian cytochromes P-450, FP Guengerich (ed). CRC Press, Boca Raton, FL, pp. 1-54. 27. Guengerich FP (1990). Purification and characterization of xenobiotic-metabolizing enzymes from lung tissue. Pharmacol. Ther. 45: 299-308. 28. Guengerich FP and MacDonald TL (1990). Mechanisms of cytochrome P-450 catalysis. FASEB J. 4: 2453-2459. 29. Hammond AC, Bradley BJ, Yokoyama MT, Carlson JR, and Dickinson EO (1979). 3-Methylindole and naturally occurring acute bovine pulmonary edema and emphysema. Am. J. Vet. Res. 40: 1398-1404. 30. Huijzer JC, Adams JD Jr., Jaw J-Y, and Yost GS (1989). Inhibition of 3-methylindole bioactivation by the cytochrome P-450 suicide substrate 1-aminobenzotriazole and α-methylbenzylaminobenzotriazole. Drug Metab. Dispos. 17: 37-42. 31. Huxtable RJ, Bowers R, Mattocks AR, and Michnicka M (1990). Sulfur conjugates as putative pneumotoxic metabolites of the pyrrolizidine alkaloid, monocrotaline. In: Biological Reactive Intermediates IV, CM Witmer et al (eds). Plenum Press, NY, pp. 605-612. 32. Jakoby WB and Zeigler DM (1990). The enzymes of detoxication. J. Biol. Chem. 265: 20715-20718. 33. Krauss RS, Angerman-Stewart J, Lookey KL, Kadlubar FF, and Eling TE (1989). The formation of 2-aminofluorene-DNA in vivo: Evidence for peroxidase-mediated activation. Biochem. Toxicol. 4: 111117. 34. Kronbach T, Larabee TM, and Johnson EF (1989). Hybrid cytochromes P-450 identify a substrate binding domain in P-450IIC5 and P-450IIC4. Proc. Natl. Acad. Sci. USA 86: 8262-8265. 35. Lawton MP, Gasser R, Tynes RE, Hodgson E, and Philpot RM (1990). The flavin-containing monooxygenase enzymes expressed in rabbit liver and lung are products of related but distinctly different genes. J. Biol. Chem. 265: 5855-5861. 36. Mamett LJ, Wlodawer P, and Samuelson B (1975). Co-oxygenation of organic substrates by the prostaglandin synthetase of sheep vesicular gland. J. Biol. Chem. 250: 8510-8517. 37. Mamett LJ, Johnson JT, and Bienkowski MJ (1979). Arachidonic acid-dependent metabolism of 7,8-dihydroxy-7,8-dihydrobenzo(a)pyrene by ram seminal vesicles. FEBS Lett. 106: 17-20. 38. McManus ME, Boobis AR, Pacifici GM, Frempong RY, Brodie MJ, Kahn GC, Whyte C, and Davies DS (1979). Xenobiotic metabolism in human lung. Life Sci. 26: 481-489.

39.

Namkung MJ, Yang HL, Hulla JE, and Juchau MR (1988). On the substrate specificity of cytochrome

P450IIIA1. Mol. Pharmacol. 34: 628-637. 40. Nebert DW, Adesnik M, Coon MJ, Estabrook RW, Gonzalez FJ, Guengerich FP, Gunsalus IC, Johnson EF, Kemper B, Levin W, Phillips IR, Sato R, and Waterman MR (1987). The P450 gene superfamily: Recommended nomenclature. DNA 6: 1-11. 41. Nebert DW, Nelson DR, Adesnik M, Coon MJ, Estabrook RW, Gonzalez FJ, Guengerich FP, Gunsalus IC, Johnson EF, Kemper B, Levin W, Phillips IR, Sato R, and Waterman MR (1989). The P450 superfamily: Updated listing of all genes and recommended nomenclature for the chromosomal loci. DNA 8:1-13. 42. Nebert DW, Nelson DR, Coon MJ, Estabrook RW, Feyereisen R, Fujii-Kuriyama Y, Gonzalez FJ, Guengerich FP, Gunsalus IC, Johnson EF, Loper JC, Sato R, Waterman MR, and Waxman DJ (1991). The P450 superfamily: Update on new sequences, gene mapping, and recommended nomenclature. DNA and Cell Biol. 10: 1-14. 43. Nelson DR and Strobel HW (1988). On the membrane topology of vertebrate cytochrome P-450 proteins. J. Biol. Chem. 263: 6038-6050. 44. Nhamburo PT, Gonzalez FJ, McBride OW, Gelboin HV, and Kimura S (1989). Identification of a new P450 expressed in human lung: Complete cDNA sequence, cDNA-directed expression, and chromosome mapping. Biochemistry 28: 8060-8066. 45. Philpot RM and Arinc E (1976). Separation and purification of two forms of hepatic cytochrome P-450 from untreated rabbits. Mol. Pharmacol. 12: 483493. 46. Ravindranath V, Burka LT, and Boyd MR (1984). Reactive metabolites from the bioactivation of toxic methylfurans. Science 224: 884-886. 47. Robertson IGC, Philpot RM, Zeiger E, and Wolf CR (1981). Specificity of rabbit pulmonary cytochrome P-450 isozymes in the activation of several aromatic amines and aflatoxin B 1to mutagenic products. Mol. Pharmacol. 20: 662-668. 48. Roth RA and Reindel JF (1990). Lung vascular injury from monocrotaline pyrrole, a putative hepatic metabolite. In: Biological Reactive Intermediates IV, CM Witmer et al (eds). Plenum Press, NY, pp. 477-487. 49. Shimizu T, Hirano K, Takahashi M, Hatano M, and Fujii-Kuriyama Y (1988). Site-directed mutageneses of rat liver cytochrome P-450d: Axial ligand and heme incorporation. Biochemistry 27: 4138-4141. 50. Siedegard J, Pero RW, Markowitz MM, Roush G, Miller DG, and Beatie EJ (1990). Isozyme of glutathione S-transferase as a marker for the susceptibility to lung cancer: A follow up study. Carcinogenesis 11: 33-36. 51. Sivarajah K, Mukhtar H, and Eling T (1979). Arachidonic acid-dependent metabolism of (±)-trans-7,8-

dihydroxy-7,8-dihydrobenzo(a)pyrene (BP-7,8-diol) 52.

to 7,10/8,9 tetrols. FEBS Lett. 106: 17-20. Sivarajah K, Lasker JM, and Eling TE (1981). Prostaglandin synthetase-dependent cooxidation of (±)-

Downloaded from tpx.sagepub.com by guest on August 22, 2015

481

benzo(a)pyrene-7,8-dihydrodiol by human lung and other mammalian tissues. Cancer Res. 41: 1834-1839. 53. Skiles GL, Smith DJ, Appleton ML, Carlson JR, and Yost GS (1991). Isolation of a mercapturate adduct

produced subsequent to glutahtione conjugation of bioactivated 3-methylindole. Toxicol. Appl. Pharmacol. 108: 531-537. 54. Sladek NE and Mannering GJ (1966). Evidence for a new P-450 hemoprotein in hepatic microsomes from methylcholanthrene treated rats. Biochem. Biophys. Res. Commun. 24: 668-674. 55. Tynes RE and Hodgson E (1984). The measurement of FAD-containing mono-oxygenase activity in microsomes containing cytochrome P-450. Xenobiotica 14: 515-520. 56. Vanderslice RR, Domin BA, Carver GT, and Philpot RM (1987). Species-dependent expression and induction of homologues of rabbit cytochrome P-450 isozyme 5 in liver and lung. Mol. Pharmacol. 31: 320325. 57. Voigt JM, Kawabata TT, Burke JP, Martin MV, Guengerich FP, and Baron J (1990). In situ localization and distribution of xenobiotic-activating enzymes and aryl hydrocarbon hydroylase activity in lungs of untreated rats. Mol. Pharmacol. 37: 182191. 58. Waxman DJ (1988). Interactions of hepatic cyto-

chromes P-450 with steroid hormones. Biochem. Pharmacol. 37: 71-84. 59. Wheeler CW, Park SS, and Guenthner TM (1990). Immunochemical analysis of cytochrome P-450IA1 homologue in human lung microsomes. Mol. Pharmacol. 38: 634-643. 60. Williams DE, Reed RL, Kedzierski G, Guengerich FP, and Buhler DR (1989). Bioactivation and detoxication of the pyrrolizidine alkaloid senecionine by cytochrome P-450 enzymes in rat liver. Drug Metab. Dispos. 17: 387-392. 61. Wolf CR, Statham CN, McMenamin MG, Bend JR, Boyd MR, and Philpot RM (1982). The relationship between the catalytic activities of rabbit pulmonary cytochrome P-450 isozymes and the lung-specific toxicity of the furan derivative, 4-ipomeanol. Mol. Pharmacol. 22: 738-744. 62. Yost GS, Buckpitt AR, Roth RA, and McLemore TL (1989). Mechanism of lung injury by systemically administered chemicals. Toxicol. Appl. Pharmacol. 101: 179-195. 63. Ziegler DM (1990). Microsomal flavin-containing monooxygenases: Oxygenation of nucleophile nitrogen and sulfur compounds. In: Enzmatic Basis of Detoxication I, WB Jakoby (ed). Academic Press, New

York, pp. 201-227.

Downloaded from tpx.sagepub.com by guest on August 22, 2015