Substrate specificities and functions of the P450 cytochromes

Life Sciences, Vol. 47, pp. 2385-2394 Printed in the U.S.A.

Pergamon Press

MINIREVIEW SUBSTRATE SPECIFICITIES AND FUNCTIONS OF THE P450 CYTOCHROMES

M.R. Juchau Dcpamnent of Pharmacology School of Medicine SJ-30 University of Washington Seattle, WA 98195 (Received in final form October 25, 1990) Summary Currently,the major recognized biochemical functionsof members of the large superfamlly of P450 hemoproteins (referredto commonly as the cytochromes P450) include catalysesof the monooxygenations of a wide varietyof endogenous and exogenous lipophilicchemicals. Substrates that have attracted the greatest attention thus far are steroids, fatty acids, eicosanoids, rctinoids,other endogenous lipids,therapeutic agents, pesticides/herbicides, chemical carcinogens,industrialchemicals and otherenvironmental contaminants and toxic xcnobiotic organics of low molecular weight. Commonly, monooxygenation of such substrates results in the generation of metabolites capable of producing biological effects that arc profoundly different (qualitatively as well as quantitatively) from those elicitable by the parent chemical per se. P450,XIX-dependent conversion of testosterone to estradiol- 178 provides a dramatic example. Thus, these hemoproteins serve as extremely important but, as yet, largely unpredictable re2ulators of the biological effects producible by endobiotics as well as by xenobiotics. Current focus is on the identification and acquisition of sequence information on hereto unidentified and/or uncharacterized P450 isoforms and ascertainment of the specific functions of svecific, ~ isoforms. The regulation of quantities and activities of such isoforms in-specific species/tissues, understandably, is also of great current interest. This interest has been further intensified by recent results indicating that substrate specificity associated with one P450 may not be the same as the corresponding isoform derived from a different animal species. Recent technological advances promise to greatly hasten the acquisition of knowledge concerning the functions of these important hemoproteins. Since the initial, independent reports in 1958 of the discovery of the P450 cytochromes by Klingenberg (1) and Garfinkel (2), scientifc interest in these unique hemoproteins (initially referred to as "CO-binding p~gment") has continued to increase. In 1990, approximately 100 abstracts containing the term "P450" were published in The FASEB Journal alone. Although initially regarded as a single cytochrome (3,4) capable of catalyzing hydroxylating reactions (5), it is now recognized that several hundred separate P450 lsofonns exist in nature. As of this writing, over 100 distinct isoforms have been sequenced, with sequences derived from cDNAs, proteins or genomic clones (Gonzalez, personal communication), and many are currently undergoing this process. The literature also contains considerable evidence for the existence of many quantitatively minor forms (e.g., in extrahepatic and embryonic tissues) and, expectedly, acquisition or sequence information for such isoforms will be slower. The advent of techniques involving PCR (polymerase chain reaction), site-directed mutagenesis, mosaic gene construction and vector expression systems, however, provide considerable optimism for rigorous studies of even these isoforms It is probably fair to state that comprehensive biologic roles/functions have not been defined precisely for any of the currently recognized P450 lsoforms even though huge strides have been made in recent years toward an understanding of P450 functions. We now understand many of the activities that indiwdual isoforms can and cannot perform but a complete understanding of the ~ to organisms of various catalytic activities in terms of selective advantages and to the preservation of the species remains, for the greater part, to be achieved. Discussions thereof also remain largely speculative in nature. Research along several lines, however, has provided intriguing clues to more global definitions of biologic function. Investigations of specific isoforms with respect to their substrate specificities, tissue/cellular distribution, distribution among species, genetic variabtlity within species, ontogeny, modes/mechanisms of intrinsic as 0024-3205/90 $3.00 + .00 Copyright (c) 1990 Pergamon Press plc

2386

Functions of the P450 Cytochromes

Vol. 47, No. 26, 1990

well as exogenous/environmental regulation (induction, repression, inhibition, activation, etc.), interrelationships with endocrine/hormonal and nutritional systems, "nonmonooxygenase" (e g , as peroxidases, catalases, reductases, dehydrogenases, hydmlases, etc.) acUvities, interrelationships and interactions with functionally related enzyme systems (e.g., glucuronyl transferases, sulphotransferases, epoxide hydrolases, glutathione transferases, etc.) all have provided and will continue to provide clues to global functions of individual P450 isoforrns as well as the P450 supeffamily as a whole A number of excellent recent reviews has dealt with several of the above-menuoned lines of investigation These include reviews of nomenclature (6,7), molecular genetics (7), induction (8), purification procedures (9), evolution (7, 10), roles m carcinogenesis (I 1, 12), in embryotoxicity (13) and in various other toxic effects (14), catalytic mechanisms(15), molecular biology (16-18) and in P450 structure and mechanlsuc enzymology (19). These reviews also contain numerous references to earher reviews as well as to important papers in the field. The purpose of this brief review is to focus more directly on our current understanding of the functional roles of P450 isoforms in biological systems. This wdl be discussed with reference to the 14 currently renognized families (6,7) of P450 cytoclmames. It should be emphasized that several additional families will undoubtedly be added in the near future Family I Currently, only two members of this family have been characterized - P450s IA1 and IA2 -- and have been sometimes referred to collectively as "P448". Both isoforms are inducible by 3-methylcholanthrene (MC) and "MC-type" inducing agents (polycyclic aromatic hydrocarbons, naphthoflavones, dioxins, dibenzofurans, planar polyhalogenated biphenyls, methylated xanthines, etc.). Members of this group of inducers also are recognized substrates, inhibitors and/or activators of ~ese isofo..rms,suggesting ~at one major function of IAI and IA2 might be to protect the organism against ~ potenuauy aeletenous erfecKsoF such environmentally ubiquitous chemicals. The observations that IA1 is usually expressed only after exposure to such inducers, that it has a very broad tissue and species distribution, and that endogenous steroids and other lipids are not good substrates for IA1 provide further credence for this as a proposed major function, Studies of the substrate specificity of IAI (20,21) thus far indi hc~ a preference for rigid, planar compounds with fused aromatic or heteroa.mmaticrin.gs: wi9 (.area/de~) ratios usually greate,r than 4 and with length/width ratios usually less than z. ~ucn chemicals are irequenuy rouna among me products of combustion. Thus, it would seem advantageous to the organism if the IA1 isoform were expressed primarily in response to exposure to these kinds of chemicals and if IAl-catalyzed monooxygenation of the same chemicals led to ~eir rapid elimination and/or inactivation. However,

produced by the same chemicals (14). Clearly, tbeexposure/pathology relationships cc.~.be l~gi~ly.con~..plcx and appear to depend on a large number of factors in addition to IA1. For example, Nebcrt (14) l~asnstco 15 non-P450 proteins induced by "MC-type" inducers, many of which are known to act as important determinants of the toxicity of foreign organic chemicals. Thus, the ~ functions o f l A 1 may.l~. explicable primarily only in conjunction with those of functionally related proteins (e.g., Mt;-mauc~ole Wamferases) as well as a number of other determinants. Although P450IA2 is also strongly inducible by "MC-type" inducers and attacks many of the same substrates a~acked by IA1, the two members of this family differ in several ways. IA2 is largely confined to hepatic tissues, is expressed constitutively, is preferentially inducible by methylenedioxypbenyt compounds (e.g., isosafrole) and attacks certain endogenous substrates quite efficiently; e.g., the aromatic hydroxylation of estrogens (22,23). Several other differences in substrate spccincity ana regulation nave also been noted. For example, rat IAI attacks 2-acetylaminoflnorene more selectively than IA2 to yield ring-hydroxylated products whereas rat IA2 attacks the same substrate more preferentially .at the amide nitrogen to yield N-hydroxylated products. The functional significance of these anu omer omerences m substrate/position selectivity is currently under investigation, Recent reports have served to demonstrate graphically that P450IA2 from humans, mice, rats (and presumably other species) can differ significantly in terms of.substrate .s~.eiffel.ties (24~.5)_ RsepoT~ of analogous species differences for P450IA1 have also appeareu in me publi~ed literature (zo:za) us, even though many striking similarities exist across species lines, extrapolations from experimental animals to humans are still not possible, even when isoforms bear the same systematic names. This concept appears to apply not only to members of the IA subfamily but to all P450 isoforms bearing .~e same systemauc name among various species as well as among various .str.aln.sOf ~ same species: tins.Is not un.cxpcc~. in view of recent observations that very minor amino acia su~atutions can result m s ~ n c a n t caanges m P450 substrate specificity (29,30). It has in fact been found (31) that a ~ amino acid substitution (phenylalanine to leucine at position 209) can produce profound differences in substrate specificity and

Vol. 47, No. 26, 1990

Functions of the P450 Cytochromes

2387

function. The genetic heterogeneity of human polmlallons renders this a particularly formidable problem m terms of pwAicting biological responses to chemicals and the implications of the finding are enoromous.

F_emax~ Family II is now recognized as the largest and most diverse of the P450 families, consisting of eight subfamilies (IIA, IIB, IIC, IID, liE, IIF, IIG and IIH) as of the lastpublished listings(6,7). Subfamily IIC is particularlylarge with at least 16 isoforms currentlyknown (32). Polymorphisms are also relatively c o m m o n within this family. Because substratespecificityand function can vary considerably from one isoform to another,it is difficultto provide concise generalizationswith respectto a functionalrole(s)for family II. Ithas been speculatedthatthisfamily evolved as the resultof environmental selectionpressures ("animal-plant warfare") involving,primarily,the naturallyoccurring plant toxins important to plants for theirprotectionagainstplantpredators (10). Credence for thisconcept arisesfrom observationsthatfamily II isoforms are capable of catalyzing the monooxygenation of a very wide variety of foreign organic chemicals that include various plant toxins as well as numerous structurallydiverse therapeuticagents and pesticides. Rigorous studies of differences in the complement of family II (especiallyIIC) isoforrnsin carnivorous vs. herbivorous species willhe of greatinterestin futureresearch Also of interest is the fact that phenobarbital (PB) and 'TB-type" inducers can profoundly regulate levels of various family II isoforms, especially some (IIB1, IIB2, IIC1, IIC2, IIC4, IIC6 and IIC16), but not all, IIB and IIC isoforms. In general, the regulation appears to occur primarily via increased rates of transcription of the corresponding genes. P450 genes of land vertebrates respond to phenobarbital (exceptions include placental mammals during their embryonic and early fetal development) but fish, reptiles and amphibians are not known to be capable of responding to the P450-inducing effects of phenobarbital (33). Interestingly, however, phenobarbital is a potent P450 inducer in certain plants, in Drosophila and in Bacdlus megaterium (8). These observations again suggest that selection pressures produced by environmental chemicals have been a major factor in the evolution of certain family H isoforms and their diverse regulatory mechanisms. Survival of plant-eating land animals would tend to be favored by responsiveness to mononxygenase induction by xenobiotics. Conversely, more facile nonbiotransformational elimination of lipophilic toxins in an aqueous environment would tend to lessen selection pressures for xenobiotic induction of mononxygenases in fish, reptiles and amphibia. These ideas are simplistic and generalized and tend to neglect the observations that various members of the IIC subfamily (e.g., rat IIC11) are profoundly ~Dressed following exposure of certain mammalian vertebrates to xenobiotic inducing agents and that responsiveness to inducers of family I isoforms is very high in the same aquatic species. Now recognized also is the fact that altered physiologic states can have a profound effect on certain isofonns of family II. For example, diabetes or fasting results in severalfold increases in levels of P450IIE (which displays selectivity for small solvent molecules) and the increases are reversible by treatment with insulin. These observations suggest that the primary role/function of this subfamily may be related to the oxidation of endogenous substrates functional in energy metabolism, i.e., conversion of fatty acid oxidation products to intermediates in gluconengenesis. A logical, function-based rationale for the striking differences in tissue-specific regulation of the P450II isoforms also has not yet been promulgated. Thus, it seems clear and should he emphasized that much remains to he learned concermng the functional role(s) of family II P450s. Familv HI Heavy involvement of family III isoforms in the biotransformation of steroid hormones, striking sex differences for IIIA(s) in rats and the profound responsiveness of certain P450III isoforms to the inducing effects of glucocorticoids and anti-giucocorticoids (dexamethasone and pregnenolone-16a-carbonitrile as respective prototype inducers) all lead one to suspect that major functional roles of the P450HI family center around oxidative steroid biowansfonnation. Research on substrate preferences has been comparatively slow because the catalyticactivityof P4501II isoforms tends to be lostduring purificationprocedures. In spiteof this, progress has been reasonably rapid. It is remarkable that P450III isoforms exifibit a high degree of regio/stereoselectivityfor such substrates as testosterone and warfarin, displaying marked selectivity for the B face and for the 10 position of the R enantiomer, respectively. Yet, they appear capable of accommodating extremely large substrate molecules such as triacetyloleandomycin, erythromycin and cyclosporine (8). The implications of this seemingly paradoxical situation for the ~ n ~ o n a l role(s) of the IliA isoforms are not understood at present, but the application of vector expression systems and sitedirected mutagenesis promise to greatly aid in the delineation of function for this family. Added impetus is given to the study of this family by the presence of relatively large quantifies of IIIA isoforms in human tissues, including the human fetal liver (34,35). Rifampicin, an established Inducer of IHA, has the reputation as one of the most effective inducers of P450-dependent monooxygenases in humans (36,37) Recent investigations have also shown that benzo(a)pyrene and 2-acetylaminofluorene, 2 frequently investigated, bioac'tivatable mutagen/carcinogens are good substrates for at least one (IIIA1) of the IIIA isoforms (38). These observations provide yet additional interest m this intriguing family. It now seems

2388

Functions of the P450 Cytoehromes

Vol. 47, No. 26, 1990

clear that members of the P450HIA subfamily as a whole are capable of catalyzing the monooxygenation of a wide variety of steroids as well as xenobiotics. Capabilities of individual members of the subfamily are, however, less clear and wtll require much further investigatiorL Of increasing interest with respect to the-P450-dependent hydroxylauons of endogenous steroids m reactions heretofore regarded as "degmdative" is the possibtiity that hydroxylation at specific positions on steroidal molecules may result in the formation of metabolites with intrinsic hormonal or other biological acuwties currently unrecognized and undefined (39). This idea provides an attractive teleological explanation for the high degree of regio~elecivity exhibtted by many P450 isoforms (including members of the I l i a subfamily) for steroid molecules. Systematic investigations of the brologic effects of hydroxylated stermd molecules (except catechol estrogens) appear not to have been reported in the open literature although papers dealing with this issue are beginning to appear (e.g., ref 39). The profound biologic effects of catechol estrogens may suggest to us that this should be much more intensively investigated. Certainly, a thorough understanding of the brologic effects of hydroxylated steroids would greatly assist our understanding of the functional roles of those P450 cytochromes known to catalyze steroid hydroxylation reactions. As one example, rattos for rates of P450IIIAl-catalyzed 28-hydroxylation vs. P450IIC11-catalyzed 2a-hydroxylation of testosterone in rats are markedly increased following exposure of the rats to giucocomcold or anti-giucocorticold inducing agents. Levels of P450IIIA 1 are profoundly increased whereas levels of P450IIC11 are greatly decreased. Is the organism benefitted by tlus response? How? Knowledge of the intrinsic biological effects of the two hydroxylated metabolites would probably help to answer these questions and better define the physrologic roles of each of the two P450 isoforms. Fanulv IV Currently it is widely felt that 1'450 families I, II and HI are responsthle for nearly all of the P450-dependent xenobiotic monooxygenation that oecors in vertebrate species. Other P450 families are believed to exhibit a high degree of selectivity for endogenous chemicals. One exception to this generalization is catalysis of the metabolism of 2-aminofluorene, carbon tetrachloride, 4-ipomeanol, aflatoxin and other xenobioucs by P450IVB isoforms. Interestingly, human and rabbit IVB1 counterparts appear not to catalyze the same reactions (7). Several additional exceptions are known (vide infra) and the dmcovery of very many more is highly likely. P450IVA isoforms, in general, (IVA4 may be an exception) appear to function primarily in the ~-oxadations of medium and long-chain fatty acids (e.g., laurate, palmitate and arachidonate) and eicosanoids (prostaglandins, prostacyclins, thromboxanes and leukotrienes). Products of these reactions sometimes exhibit very potent biologic effects and this has become a very important area of research. It now appears that the ¢o-1- and e~-2-oxidatious of the same fatty acids are catalyzed primarily by tsoforms from other P450 families. For example, Holm et al. (40) have provided evidence that the ¢o-2 hydroxylation of prostagiandins is effectively catalyzed by P450IA1 and/or P450 IA2, with rates profoundly increased following exposure to methylcholanthrene. The results of experiments by these and other investigators have shown that ¢o-1 hydroxylations can be catalyzed by a number of P450 isoforms (e.g, IA1, IIB1, etc) not belonging to family IV. Studies by CaJacob et al. (41) suggest that a highly structured IVA active site can actually suppress co-1 hydroxylation sterically in order to deliver the oxygen to the thermodynamically disfavored terminal (co) carbon. Unfortunately, the physiologic role of o)-oxidation is poorly understood at present. It has been suggested that co-oxidation permits B-oxidation to proceed from both ends of the fatty acid molecules and results in a more efficient utilization of fatty acids for intermediary metabolism. Such a function would be of particular value during starvation, diabetes, ketosis and acyl CoA dehydrogenase deficiencies. A better understanding of this will be needed for accurate definition of the role(s) of P450IV family isoforms. In addition, substrate specificities of the individual family IV isoforms are not yet well delineated although studies are now beginning to appear in the literatme (42,43). Of p~rticular interest in recent years has been the P450-dependent conversion of arachidonate to a variety of highly active metabohtes. This was reviewed recently by Fitzpatrick and Murphy (44) and the phenomenon is now commonly referred to as the "epoxygenase" pathway. This pathway results in the generation of c~epoxyeicusatrienoic acids (EETs) and dihydmxyeicosatrienoic acids (DIHETEs) which have exhibited a variety of potent biological effects including stimulus-respouse coupling in endocnne, renal, ocular and secretory cells. However, these P450-depondent conversions involve the generation of epoxides across carbon-carbon double bonds rather than hydroxylation at the terminal carbon, rendering it doubtful that family IV iso_foml,s are involved. The syntheses, of prostacyclin ( ~ . I2) and of thromboxane A 2 (TXA2) Item p~sm.g[andm enooperoxme ( ~ H ~ are also catalyzed by at least two/450 hemoproteins (45), but me iamily(ies) to which these isoforms belong also remains unknown at this writing. The unique oxene transferase activity" of these eicosanoid synthesizing P450s renders it unlikely that they would be placed in family IV.

Vol. 47, No. 26, 1990

Functions of the P450 Cytochromes

2389

At least one isoform of this family (IVA1) is induced by a group of chemicals referred to as "pemxisome proliferators". This group includes clofibrate, together with structurally related hypollpidemic drugs, and phihalates -- used in plasticizers. These agents produce very large increases in liver weights and a dramatic proliferation of peroxisomes with resultant parallel increases in peroxisomal enzymes such as catalase They are also hepatocarcinogens. It is therefore tempting to speculate that P450IVA isoforms may have functions related to these biological effects. In the rabbit lung, pregnancy or treatment of the animals with progestational agents results in dramatic increases in certain isoforms (46) of family IV. The functional significance ofllns up-regulation is likewise not understood at present Isoforms of the P450IV family are very widely distributed in nature, suggesting that the catalytic activines discussed above are of unportance and probably provide survival value to the orgarusms in which they are present and functional Nevertheless, it is clear that our understanding of their overall biologic role is just now beginning to take shape. Family VI Currently, only one member of P450 family VI has been characterized sufcienfly to be included in the most recent listings (6,7). This is isoform VIA1 from the abdomens of phenobarbital-induced houseflies (47) and Is thus far the only insect P450 to be included. It is certain, however, that a great many other insect P450s exist in nature and will ultimately be included in such lists. Interest in insect P450s has been given great impetus by observations that resistance of insects to insecticides is often assoctated with marked increases in rates of P450-dependent biotransformation of insecticides in resistant species/strains (48). Such observations graphically illustrate the survival value of P450s to species following exposures to toxic organic chemicals. Hodgson (49) has pointed to addmonal roles of insect P450s in terms of biotransformation of endo/~enous substrates and involvement in normal metabolic processes such as hormone and pheromone blosysnthesis. Multiple P450s in Drosophila (50) provide particularly good opportunities to investigate genetic aspects of P450 regulation as well as mechanisms of resistance and other aspects of P450-dependent activities in insects. Familv XI Two isoforms of the P450XI family are included in the most recent listings (6,7). These are P450XIA1, which also bears 1)450118 as a trivial name, and P450XIB1, which also bears P450scc as a trivial name. Both isoforms are present in mitochondrial inner membranes rather than in the endoplasmic reUculum, accept elctrons from a non-home iron protein (adrenodoxin) rather than from a flavoprotein (P450 reductase) and are found predominantly in vertebrate tissues involved in the synthesis of steroid hormones rather than in the liver. XIA1 is localized almost exclusively in the adrenal cortex and appears to function primarily in the biosynthesis of gluco- and mineralocorticoid hormones via hydroxylatiun at the 1 lfi position of steroid molecules. XIB2, on the other hand, is found in several steroid-synthesizing organs (adrenals, testes, ovaries, placenta) and its primary function appears to be in the conversion of cholesterol to pregnenolone, the initiat and rate-limiting step in the conve~nn of cholesterol to all other steroid hormones. The two mitochondrial isoforms (-37-38% sequence homology) provide a good example illustrating that P450 isoforms are not invariably assigned to families strictly according to amino acid sequence. Normally, > 40% homology is required for assignment to the same family. Considerable controversy has surrounded the function of P450XIA1. While generally recognized as critically functional in the catalysis of highly important 1ll~-hydroxylafions of several steroid substrates (deoxycordcosterone, denxycortisol, etc.), involvements in the biosysyntheses of aldosterone and steroidal estrogens -- via catalysis of hydroxylation reactions at steroid carbons 18 and 19, respectively -- have also been ascribed to XIA1. It must be conceded that XIA1 will catalyze hydroxylafion at the 19 carbon of C-19 steroids in aromatase/10-demethylase reactions, but the physiologic significance of this catalytic activity xs currently unknown (51). Insofar as is now known, P450XIXA1 is the most important catalyst for aromatization of C-19 steroids to estrogens (vide infra). Suhara et al. (51) have suggested the possibility that XIAl-catalyzed conversion of androstenedionc to intermediary 19-norandrostenedione (a potent mineralocorticoid secreted from the adrenal cortex) could serve an important physiologic function. In terms of the biosymthesis of aldostemne via 18-hydroxylation, it now seems likely that a separate, but closely related 1)450 isoform is involved (52,53). An isoform isolated from zona glomerulosal mitocbondria could catalyze the three successive hydroxylation reactions of deoxycorticosteronc to yield aldosterone. The isoform isolated from zonae fasciculata/reticularis, in contrast, catalyzed only 11B- and 18-hydroxylations of deoxycorticosterone, i.e., single step reactions. The major physiologic function of P450XIB1 (scc) appears to be well establisbed. This mitochondrial isoform catalyzes the conversion of free cholesterol, via 22R-hydmxycholestero1 and 20ct, 22Rdihydmxycholesteml, to pregnenolone in the first and rate-limiting step (side-chain cleavage) in the biosynthesis of all steroidal hormones (androgens, estrogens, progestins, giucocorticoids and mineraloconicoids). Clearly, this is a highly important function and its importance for survival of the

2390

Functions of the P450 Cytochromes

Vol. 47, No. 26, 1990

species is well recognized. Not clear at present is whether XIB1 may serve other functions of lesser impoRance. Investigations of the extent to which XIB1 would catalyze the oxidation of non-cbolesterol subswates or even of cholesterol at carbons other than 20 or 22 would provide insights into this question of function. It is known that 25-hydroxycbolesterol serves as a substrate for the side-chain cleavage reaction (54) and that multiple forms of XIB1 exist (55,56) but the functional significance of ~ . s e observations is not understood. Detection of this or a closely relatect isofrom in the brain (57) may have important functional significance.

Only one isoform is currently listed (as P450XVIIA1) in family XVII. This isoform is also commonly known as P45017(x because of its known function in the catalysis of the 17c¢-hydroxylation of steroid molecules. It can exhibit both 17cx-hydroxylase and 17,20-desmolase (cleavage of the 17-20 carbon-carbon bond) activities (58)* and is known to be essential for the biosynthesis of sex steroids and cortisol. Interesting species and tissue differences in both regulatory control and catalytic activity point toward species/tissue differences in physiologic function (59). For example, it has been shown (59) that an XVIIA1 isolated from rat testes will catalyze the conversion of both 17cx-hydroxypregnenolone and 17~hydroxyprogestcroue to dehydruepiandrosteronc and androsteuedion¢ respectively. Human and bowne isoforms failed to catalyze the latter reaction at significant rates. Human, but not rat P450XVII effectively catalyzed the 16a-hydroxylaton of progesterone. Other species-specific and tissue-specific differences have been observed but the biologic significance of these biochemical differences are not fully understood at present. The capacity of XVII to catalyze xenobiotic monooxygemtion is evidenced from the fact that spironolactone win serve as a suicide substrate for this isoform, at least for certain species and tissues (60) Systematic investigations of the capacity of XVII to catalyze other xenobiotic monooxygenation reactions, however, appear not to have been undertaken. Family XIX A single member (P450XIXA1) of this family is currently listed. Even though the published literature contains numerous pieces of evidence that multiple XIX isoforms exist (e.g., ref. 61), human and chicken eDNA sequences exhibit extensive similarities (62). P450XIXA1, commonly referred to as aromatase, appears to be quite ubiquitous in vertebrate systems, occurring not only in the reproducuve tissues of females (ovaries, primate placentas) but also in such diverse tissues as testes, adipose tissues, brain and skin. Again, the ~ function of this family appears to be well established. XIXA1 catalyzes the conversion of C-19 steroids to estrogens via hydroxylation of the angular 19-methyl group to yield, successively, the 19-hydroxy and 19-oxo intermediates with subsequent loss of the C-19 carbon and c~s elimination of the 18 and 28 hydrogens, possibly via 28-hydmxylaton (63), to yield a corresponding aromatized A ring steroid and formic acid. Three P450-dependent monooxygenaton reactions are involved in the conversion process and each, including the rate-limiting 19-hydroxylaton reaction, appears to be catalyzed by the same P450 isoform. This aromatization reaction is the last in a series of steps in the biusynthesis of estrogens from cholesterol. The reaction exhibits a very high degree of substrate specificity (64,65) but 17o~-eth~. yl-19-nors~roids such as the synthetic progestins present in oral contraceptive preparations are sui~de substrates for XIX P450(s) (66). Other 19-norsteroids are convertible to estrogens. Androstene~lione and testosterone are excellent substrates and are regarded as the natural substrates for the enzyme. Nonsteroidal xenobioti~ would be expected to be very poor substrates for P450XIX (65,67) but this remains to be exhaustively investigated because only a limited number nf chemicals have been offered as substrates in purified, reconstituted or vector-expressed systems. The very limited P450-dependent xenobiotic oxidizing activity of human placental microsomes (68,69) (aside from P450IAl-~atalyzed reactions in placentas of women exposed to "MC4ype inducers, vide supra), which are ¥~ry rich in P450XIX, strongly speaks against a xenobiotic-oxidizing function for P450XIX that would be pharmacologically or toxicologically significant. Nevertheless, the possibility remains that certain nonsteroidal xenobioti~ may serve as effective substrates.

Eamll.v.2~ Isoforms of this family function in the catalysis of the 21-hydroxylation of C-21 steroid molecules and are essential in the biosynthesis of glucocorticoid and mineralocorticoid hormones. Genetic deficiencies (congenital adrenal hypcrplasia) of XXI arc relatively common in humans and are regarded as the most common autosomal recessive metabofic disease in man (70). Two genes (CYP21A1, CYP21A2) and two pscudogenes (CYP21A1P, CYP21A2P) have been characterized (6,7). No other function for this family has been described to my knowledge and ali information concerning P450XXI (e.g., exclusive presence m the adrenal cortex) indicates a high specificity for C21-hydroxylation of progesterone and 17¢hydroxyprogesteroue in pathways leading to the synthesis of aldosterone and cortisol, respectively. Recently, Lorcnce et al. (71) investigated the substrate specificity after expressing two full-length eDNA clones encoding P450XXI in eukaryotic COS1 cells. Of 18 steroids tested, only three -- progesterone,

Vol. 47, No. 26, 1990

Functions of the P450 Cytochromes

2391

17a-hydroxyprogesterone and 118,17a-dihydroxyprogesterone -- were reported to serve as substrates. Neither pregnenolone nor 17a-hydroxypregnenolone could serve as substrates. The data were consistent with previous studies in which other techniques were used to explore substrate specificities and strongly indicated that catalysis of the monooxygenation of xenobiotic substrates by P450XXI(s) would be negligible. Significant participmion in other functions also appears unlikely at this point in time. Familv XXVI A very recent addition to the growing list (6,7) of P450 families is P450XXVI which, at present, has only one isoform as a member. This isoform catalyzes the 26-hydroxylation of 5g-cholestane-3a,7a,12a-triol to yield the corespondlng tetml as the first step in the oxidation of the side chain of sterol intermediates m the biosynthesis of bile acids. The isoform has been purified from the hepaticmitochondria of rabbits (72) and, like other mitochondrial I'450 isoforms, accepts electrons from ferredoxin. Of functional interest is the capacity of this isoform to catalyze the 26-hydroxylation of cholesterol, 5-cholestene-3g, 7a-diol, 7ahydroxy-4-cholesten-3-one and 513-cholestan-3a,7o~-dioL 26-Hydrooxycholesterol and other oxygenated sterols are powerful suppressors of the transcription of genes involved in cholesterol metabolism (72). Thus, it seems possible or even likely that this P450 isoform may serve as an important regulatory function in the modulation of cholesterol metabolism and cholesterol levels. The extent to which it may subserve other functions is not known at present. It seems somewhat improbable that it would be significantly involved in xenobiotic biotransformations.

The only currently listed member of this family, which also bears the trivial names of P450DM and P45014DM, has been purified from rat liver, from Saccharomyces cerevisiae, and from Candida albicans (7,73). It catalyzes the complete oxidative demethylation of 24,25-dihydrolanosterol to 4,4-dimethyl-5acholesta-8,14-dien-3g-ol and formic acid via three monooxygenation steps. It also catalyzes the C-14 demethylation of lanosterol (4,4,14a-trimethyl-5a-cholesta-8,24-dien-38-ol)to 4,4-dimethyl-5a-cholesta8,14,24-trien-38-ol. Oxidative removal of the 14a-methyl group from lanosternl represents the first step in the biosynthesis of ergosterol in yeast and of cholesterol in mammals. The antifungal activities of a number of currently prescribed azole antibiotics is attributed to inhibition of this isoform. A number of mechanistic studies (74-77) have provided evidence that a stogie isoform catalyzes each oftbe steps in the demethylation reaction. While clearly subserving an extremely important biological function, it m not yet clear whether this or other closely related isoforms may play additional important roles in biology. Furore research must answer this question. F~nilv LH

An aikane-inducible P450 isoform (a common trivial name is P450alk) from Candida tropwalts has been isolated and sequenced (78) and is thus far the only listed (as of July, 1989) representative of the LII family. It catalyzes the terminal (¢o)hydroxylation of alkanes and fatty acids, the first and rate-limiting step in the assimilation of these substances by yeasts. The hydroxyladon also represents an important enzymanc step in the production of fatty acid alcohols and dicarboxyiic acids by alkane assimilating yeasts. This isoform (LItAI) is a membrane-associated protein that is localized in the endoplasmic reticulum and, in terms of sequence similarity to other isoforms was found to be most closely related to a member of the P450III family (78) although, functionally it is most closely related to the P450IVA subfamily discussed above. For example, P450s IVA 1 and LIIA1 hoth catalyze the o)-oxidalion of lauric acid yet display very little sequence similarities. This serves as yet another example of the fact that P450 isoforms with very different sequence similarities may have very similar catalytic activities whereas conversely, isoforms with highly similar sequence similarities can have widely divergent enzymatic activities and substrate specificities. Familv CI Again, only one isoform has been assigned to this family. It is commonly referred to by a trivial name, P450cam, and currently its structure is the most extensively investigated and best understood of all P450 isoforms (7,8). It has been isolated, purified and sequenced from Pseudomonas puuda, a bacterial organism which expresses the isoform only after having been grown on the substrate camphor or a structurally related chemical as the sole carbon source. A principal advantage of the use of this isoform for detailed studies is that it is not membrane-hound, thus permitting investigators to more readily aace~m the crystal structure of both substrate-free and substrate-hound forms. It provides the best-known example of the concept that all P450s are not membrane-hound. Interestingly, it behaves functionally like the mitochondrial eukaryotic isoforms in families XI and XXVI (discussed above) in that it accepts electrons from a non-berne iron prrrtein -- in lids case, lmtidaredoxin.

2392

F u n c t i o n s of t h e P450 Cytochromes

Vol. 47, No. 26, 1990

Recent studies of substrate specificities by Raag and Poulos (79) have provided additional insights into the functional capabilities of this iso_form. No.reamphor, a subslrate lacking the 8-, 9- and 1G-methyl groups of campnor, was ooservea to bind about 0.9 A further than camphor from the oxygen binding site, permitting the endogenous ligand to remain bound at the sixth llganding position. Adamantanone, on the other hand, binds even more closely than camphor, leaving the home iron pentacoordInate. All three chemicals are substrates for the CI isoform but camphor and adamantanone are hydroxylated with very high regiospecificity whereas norcamphor is nonstereoselectively hydroxylated at three positions on the molecule. The looser fit of norcamphor in the active-site pocket appeared responsible for a less specific pattern of hydroxylation. The studies demonstrated the importance of the control of specificity by the substrate via stefic crowding of the sixth liganding position (degree of substrate displacement of the endogenous sixth ligand) and modulation of aqua ligand protonation. Thus, these factors should be added e.n~tion o f the ~.bst~ate. with ~ to the iron-linked, activated oxygen atom, substrate mobility and cat reacuvmes oi mmwauat suostrate atoms as 0eterminants of whether or not any individual molecule will be efficiently hydroxylated at a given position. Further studies of this nature will prowde important insights into the fimctionality and potential functionality of individual P450 isoforms.

~

Familv CII Hydroxylations of long-chain fatty acids and their respective amides and alcohols m a carbon monoxidesensitive, NADPH-dependent enzyme system in Bacillus megaterium have been shown to be catalyzed by a single, soluble P450 isoform of a molecular weight of 119,000 (80). This interesting isoform has two functional domains on the same polypeptide chain -- a P450 reductase containing both FAD and FMN as prosthetic groups and also a heine-containing P450 domain. The gene for this unique, bifunctional protein has been cloned, sequenced and expressed in E. Coh. This isofonn (CIIA1, with the trivial name of P450!~M.3) has both sequence and functional similarities to P450IVA1. It catalyzes the monooxygenation o nora saturated and unsaturated fatty acids and their corresponding alcohols and amides. Recent studies reported by Boddupalli et aL (81) suggested that CIIA1 would catalyze the conversion of fatty acids to co-, ¢o-1-, e~-2- and ¢e-3-hydroxy acids. Determinations of the relative rates of these reactions will be of great interest because of the selectivity of P450IVA1 for the co-oxidation reaction (vide supra). In contrast to IVA1, the CII isoform is soluble, is inducible by phenobarbital and other barbiturates but is not known to be inducible by clofibrate and other peroxisome prohferators. The relationship(s) of these biochemical differences tophysiologic functionality is not clear at present. Future comparisons of IVA1, LIIA1 and CIIA1 (each of which catalyzes co-oxidations of fatty acids) will be of high interest

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1I. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

M. KLINGENBERG, Arch. Biochem. Biophys. 75 376-86 (1958). D. GARFINKEL, Arch. Biocbem. Biophys. 77 493-509 (1958). T. OMURA AND R. SATO, J. Biol. Chem. 239 2370-8 (1964). T. OMURA AND R. SATO, J. Biol. Chem. 239 2379-85 (1964). R.W. ESTABROOK, D.Y. COOPER AND O. ROSENTHAL, Biochem. Z. 338 741-9 (1963). D.W. NEBERT, D.R. NELSON, M. ADESNIK, M.J. COON, R.W. ESTABROOK, F.J. GONZALEZ, F.P. GUENGERICH, I.C. GUNSALUS, E.F. JOHNSON, B. KEMPER, W LEVIN, I.R. PI-IK~IPS, R. SATO AND M.R. WATERMAN, DNA 8 1-13 (1989). F.J. GONZALEZ, F'harm. Ther. 45 1-38 (1990). A.B. OKEY, Pharm. Ther. 45 341-98 (1990). D.E. RYAN AND W. LEVIN, Pharm. Ther. 45 153-240 (1990). D.W. NEBERT, D.R. NELSON AND R. FEYEREISEN, Xenobiotica 19 1149-60 (1989). F.P. GUENGERICH, Cancer Res. 48 2946-54 (1988). F.F. KADLUBAR AND G.J. HAMMONS, The role of cytochrome P450 in the metabolism of chemical carcinogens, in: Mammalian Cvtochrome P450, Ed. F.P. Guengerich, pp. 81-130, CRC Press, Boca Raton, Florida, (1987). M.R. JUCHAU, Ann. Rev. Pharmacol. Toxicol. 29 165-87 (1989). D.W. NEBERT, Crit. Rev. Toxicol. 20 137-52 (1989). F.P. GUENGERICH AND T.L. MACDONALD, FASEB J. 4 2453-2459 (1990). D.W. NEBERT AND F.J. GONZALEZ, Ann. Rev. Biocbem. 56 945-93 (1987). O. GOTOH, Y. TAGASHIRA, T. IIZUKA AND Y. FUJII-KURIYAMA, J. Biocbem. 93 807-17 (1983). F.J. GONZALEZ, Pharmac. Rev. 40 243-88 (1988). P.R. ORTIZ DE MONTELLANO, ed.: Cvtochrome P450: Structure. Mechanism and Biocbemi#try, Plenum Publishing Corp., New York, 1986. D.F.V. LEWIS, C. IOANNIDES AND D.V. PARKE, Biochem. Pharmacol. 35 2179-86 (1986). D.F.V. LEWIS, C. IOANNIDES AND D.V. PARKE, Chem. Biol. Interact. 64 39-60 (1987). G.A. DAN'NAN, D.J. PORUBEK, S.D. NELSON, D.J. WAXMAN AND F.P. GUENGERICH, Endocrinology 118 1152-60 (1986).

Vol. 47, No. 26, 1990

23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61.

Functions of the P450 Cytochromes

2393

M.J. NAMKUNG, D.H. PORUBEK, S.D. NELSON AND M.R. ~UCHAU, J. Steroid Biochem. 22 563-7 (1985). T. AOYAMA, F.J. GONZALEZ AND H.V. GELBOIN, Molec. Cardnogen. 2 192-8 (1989). T. SHIMADA, M. IWASAKI, M.V. MARTIN AND F.P. GUENGERICH, Cancer Res. 49 321828 (1989). D.S. RIDDICK, J.E. MACKIE, T.E. MASSEY AND G.S. MARKS, Can. J. Physiol. Pharmacol. 68 370-4 (1989). A. PUGA AND D.W. NEBERT, Biochem. Soc. Trans. 18 7-10 (1989). L.S. KAMINSKY, G.A. DANNAN AND F.P. GUENGERICH, Eur. J. Biochem. 141 141-8 (1984). T. AOYAMA, K. KORZEKWA, K. NAGATA, M. ADESNIK, A. REISS, D.P. LAPENSON, J. GILLETrE, H.V. GELBOIN, D.J. WAXMAN AND F.J. GONZALEZ, J. Biol. Chem. 2~4 21327-33 (1989). H. FURUYA, T. SHIMIZU, K. HIRANO, M. HATANO AND Y.FUJII-KURIYAMA, Biochem 28 6848-57 (1989). R.L.P. LINDBERG AND M. NEGISHI, Nature 339 632-634 (1989). C. HASSET AND C.J. OMIECINSKI, Nucl. Acids Res. lg 1429-34 (1990). R. SCHWENN AND G.J. MANNERING, Comp. Biochem. Physiol. 71B 445-53 (1982). M. KOMORI, K. NISHIO, M. KITADA, K. SHIRAMATSU, K MUROYA, M. SOMA, K. NAGASHIMA AND T. KAMATAKI, Biochem. 29 4430°34 (1990). S.A. WRIGHTON AND M. VANDENBRADEN, Arch. Biochem. Biophys. 268 144-51 (1989) T.S. BARNES, M.D. BURKE AND W.T. MELVIN, Biochem J. 260 635-40 (1989). P.B. WATKINS, S.A. MURRAY, L.G. WINKLEMAN, D.M. HEUMAN, S A WRIGHTON AND P.S. GUZELIAN, J. Clin. Invest. 83688-97 (1989). M.J. NAMKUNG, H.L. YANG, J.E. HULLA AND M.R. JUCHAU, Molec. Pharmaoal. 34 62837 (1988). W.M. GROGAN, V.M. PHILLIPS, E.G. SCHUETZ, P.S. GUZELIAN AND C.O. WATLINGTON, Am. J. Physiol. 258 C480-88 (1990). K.A. HOLM, S.S. PARK, H.V. GELBOIN AND D. KUPFER, Arch. Blochem. Biophys. 269 664-77 (1989). C.A. CAJACOB, W.K. CHAN, E. SHEPHARD AND P.R. ORTIZ DE MONTELLANO, J. Biol. Chem. 263 18640-9 (1988). L.J. ROMAN, B.S.S. MASTERS, J.E CLARK, K.J. GRIFFIN AND E.F JOHNSON, FASEB Jrd. 4(7) A2238 (1990). R.T. OKITA, J.M. BERGHOLTE, F. LAIB AND J.R. OKITA, FASEB Jnl. 4(3) A598 (1990). F.A. FITZPATRICK AND R.C. MURPHY, Pharmacol. Rev. 40 229-41 (1989). M. HECKER AND V ULLRICIt, J. BioL Chem 264 141-50 (1989). S. MATSUBARA, S. YAMAMOTO, K. SOGAWA0 N. YOKOTANI, Y. FUJII-KURIYAMA, M. HANIU, J.E. SHIVELY, O, GOTOH AND E KUSONOSE, J. Biol. Chem. 262 13366-71 (1987). R. FEYEREISEN, J.F. KOENER, D.E. FARNSWORTH AND D W. NEBERT, Proc Soc. Natl. Acad. Sci. USA 86 1465-9 (1989). J.G. SCOTT, S.S.T. LEE AND T. SHONO, Pest. Biochem. Physiol. 36 127-34 (1990) E. HODGSON, Insect Biochem. 13 237-46 (1983). S.S. SUNDSETH, C.E. NIX AND L.C. WATERS, Biochem. J. 265 213-7 (1990). K. SUHARA, K. OHASHI, K. TAKAHASHI, AND M. KATAGIRI, Arch Biochem. Biophys 267 31-7 (1988). S. KIRITA, K. MOROHASHI, T. HASHIMOTO, H YOSHIOKA, Y FUJII-KURIYAMA AND T. OMURA, J. Biochem. 104 683-6 (1988). T-. OGISHIMA, F. M1TANI AND Y. ISHIMURA, J. Biol. Chem. 264 10935-8 (1989). J.D. LAMBETH, S.E. KITCHEN, A.A. FAROOQUI, R. TUCKEY AND H KAMIN, J. Biol Chem. 257 1876-84 (1982). M. TSUBAKI, H. OHKUBO, Y. TSUNEOKA, S. TOMITA, A HIWATASHI AND Y ICHIKAWA, Biochim. Biophys. Acta 914 246-58 (1987). S. SUGANO, M. OKAMOTO, H. IKEDA, N. TAKIZAWA AND S. HORIE, Biochim. Biophys Acta 994 235-45 (1989). B. WALTHER, J. GHERSI-EGEA, A. MINN AND G. SIEST, Arch. Biochem. Blophys 254592-6 (1987). J.S.D. WINTER, R.M. COUCH, J. MULLER, Y.S. PERRY, P. FERREIRA, L. BAYDALA AND C.H.L. SHACKLETON, J. Clin. Endocrinol. Metab. 68 309-16 (1989). H.R. FEVOLD, M.C. LORENCE, J.L. MCCARTHY, J.M. TRANT, M. KAGIMOTO, M.R. WATERMAN AND J.I. MASON, Mole~. Endocrinol. 3 968-75 (1989). J.H. SHERRY, P.B. JOHNSON AND H.D. COLBY, Biochem. Pharmacol. 37 355-7 (1988) Y. OSAWA, T. HIGASHIYAMA, M. FRONCKOWIAK, N. YOSHIDA AND C. YARBOROUGH, J. Steroid Btochem. 27 781-9 (1987).

2394

62. 63. 64. 65. 66. 67 68. 69. 70. 71. 72. 73. 74 75 76. 77. 78. 79. 80. 81.

Functions of the P450 Cytochromes

Vol. 47, No. 26, 1990

G.J. HICKEY, J.S. KRASNOW, W G BEATrIE AND J.S. RICHARDS, Molec. Endocnnol 4_ 3-12 (1990). E.F. HAHN AND J. FISHMAN, J. Biol. Chem. 259 1689-94 (1984). D.D. BEUSEN, H.L. CARRELL AND D.F. COVEY, Biochem. 26 7833-41 (1987). N. HARADA, J. Biochem. 103 106-13 (1988). OSAWA, YOSHIO, OSAWA YOICHI, YARBOROUGH, C. AND BORZYNSKI, L , Biochem Soc. Trans. 11 656-9 (1983). M.R. JUCHAU, M.J. NAMKUNG AND A.E. RETTIE, Trophoblast Res. 2 235-63 (1987) M.R. JUCHAU, Pharmacol. Ther. 8 501-24 (1980). M. PASANEN AND O. PELKONEN, Drug Mctab. Rev. 21 427-62 (1990). A.N. AL-OTHMAN, K. DOCHERTY, M.W. MAKGOBA, M.C. SHEPPARD AND D.R. LONDON, J. Mol. Endocrinol. 1 157-64 (1988). M.C. LORENCE, J.M. TRANT, J.I. MASON, C.R. BHASKER, Y. FUJII-KURIYAMA, R W ESTABROOK AND M.R. WATERMAN, Arch. Biochem. Biophys. 273 79-88 (1989) S. ANDERSSON, D.L. DAVIS, H DAHLBACK, H. JORNVALL AND D W RUSSELL, J Biol. Chem. 264 8222-9 (1989). C.A. HITCHCOCK, K. DICKINSON, S.B. BROWN, E.G.V. EVANS AND D.J. ADAMS, Biochem. J. 266 475-80 (1990). Y. A O Y A M A , Y. YOSHIDA, Y. S O N O D A A N D Y SATO, Biocl'am.B1ophys Acta 1006 209-13 (1989). R.T. FISCHER, S.H. STAM, P.R. JOHNSON, S.S. KO, R.L. M A G O L D A , J L G A Y L O R A N D J.M. TRZASKOS, J. Lipid Res. 30 1621-32 (1989). C.A. HITCHCOCK, S.B. B R O W N , E.G.V. E V A N S A N D D.L A D A M S , B1ochem J. 260 54956 (1989). Y. A O Y A M A , Y. YOSHIDA, Y. S O N O D A A N D Y. SATO, J.B1ol Chem. 264 18502-5 (1989) D. S A N G L A R D A N D J.C.LOPER, Genc 76 121-36 (1989). R. R A A G A N D T.L. POULOS, Biochem. 28 917-22 (1989). L.O.N A R H I A N D A J. FULCO, J. Biol.Chem 261 7160-9 (1986). S.S. BODDUPALLI, R.W. E S T A B R O O K A N D J.A. PETERSON, J. Biol. Chem 165 4233-9 (1990).