ARCHIVES

Vol.

OF BIOCHEMISTRY

277, No. 1, February

AND

BIOPHYSICS

15, pp. 8-16,199O

Selective Induction of Cytochrome P450 lsozymes in Rat Liver by 4-n-Alkyl-methylenedioxybenzenes Craig B. Marcus,“* Neil M. Wilson,? and Curtis J. OmiecinskiJ-

Colin

R. Jefcoate,$

Christopher

F. Wilkinson,$

*Department of Pharmacology and Toxicology, School of Pharmacy and Pharmacal Sciences, Purdue University, West Lafayette, Indiana 47907; TDepartment of Environmental Health, School of Public Health and Community Medicine, University of Washington, Seattle, Washington 98195; *Department of Pharmacology, Medical School, University of Wisconsin, Madison, Wisconsin 53706; and 8 Versar Inc., 6850 Versar Circle, Springfield, Virginia 22151

Received

June

5,1989,

and in revised

form

September

29,1989

To examine the structural requirements of cytochrome P450 induction by 4-n-alkyl-substituted methylenedioxybenzenes (MDBs), rats were treated in uiuo with a series of MDBs that differed in alkyl carbon sidechain length (0, 1, 2, 3, 4, 5, 6, or 8). Expression patterns of specific P450 isozymes were evaluated with Western and Northern blotting, enzymatic assays, and solution hybridization assays. As determined by carbon monoxide difference spectroscopy, maximal hepatic induction of total P450 content occurred when rats were treated with MDB derivatives with alkyl chain lengths of five or six carbons. However, maximum induction of the specific P450s-P450IA1, P450IIB1, and P450IIB2-occurred with n-hexylMDB. In contrast to effects observed with phenobarbital, treatment with MDBs resulted in higher levels of P450IIB2 than of P450IIBl in rat hepatic microsomes. Western blot quantitation of MDB-induced hepatic P450IIBl and P450IIB2 apoenzymes did not correlate to measured levels of the corresponding P450 mRNAs. In fact, P450IIBl and P450IIB2 apoenzyme levels were consistently lower than expected based on Northern blot and solution hybridization measures of the respective mRNAs. These data suggest that the n-alkyl-MDBs effect increases in levels of hepatic P450 in a complex manner, producing accumulation of P450 mRNAs concomitant with alterations in processes regulating steady-state levels of P450 apoenzyme. 0 1990 Academic Press, Inc.

Cytochromes P450 are a superfamily of genes whose products catalyze a wide array of oxidative reactions in 1 To whom 8

correspondence

should

be addressed.

nearly all eukaryotic organisms (1, 2). Individual isozymes are characterized in general by exhibiting broad yet overlapping substrate specificities. The mechanisms by which many xenobiotics selectively induce cytochrome P450 isozymes remain unclear. Cytochrome P450 isozymes from the same subfamily often appear to respond coordinately to xenobiotic inducers although distinctive expression patterns are also observed (3, 4). Rat P450IA cytochromes, IA1 and IA2, are frequently induced in tandem in the liver by polycyclic aromatic hydrocarbons (PAHs) ,’ while the P450IIB cytochromes, IIBl and IIBS, are coordinately induced in liver by phenobarbital (PB) and other compounds (5). Proportionally however, the resulting contents of each isozyme may vary considerably. The ratio of P450IIBl:P450IIB2 proteins in the rat liver following induction by PB and other compounds is considerably greater than unity (6-g), yet the reverse is true in untreated rat liver (9). Methylenedioxybenzenes (MDBs) have been extensively studied as inhibitors and inducers of cytochrome P450 isozymes (10, 11). Pretreatment of animals in uiuo with MDBs results in a biphasic response of cytochrome P450 activity. An initial, rapid inhibition of certain P450-mediated metabolic activities (12-14) is due to formation of an MDB-metabolite:P450 complex that ’ Abbreviations used: MDB, methylenedioxybenzene; PB, phenobarbital; BNF, P-naphthoflavone; MC, 3-methylcholanthrene; PAH, polycyclic aromatic hydrocarbon; SDS-PAGE, sodium dodecyl sulfateepolyacrylamide gel electrophoresis; t-but, 4.tertiary butyl-MDB; C4-MDB, 4-n-butyl-MDB; CG-MDB, 4-n-hexyl-MDB; C8-MDB, 4-noctyl-MDB; AHH, aryl hydrocarbon hydroxylase; PRD, pentoxyresorufin 0-deethylase; PAP, peroxidase-antiperoxidase; Ah, aryl hydrocarbon; PCB, polychlorinated biphenyl; FTIR, Fourier transform infrared spectroscopy; FTNMR, Fourier transform nuclear magnetic resonance spectroscopy; P450IIB1, P450b; P450IIB2, P450e; P450IA1, P45Oc; P450IA2, P450d (1,2). 0003.9861/90

Copyright All

rights

$3.00

0 1990 by Academic of reproduction

in any

Press, Inc. form

reserved.

INDUCTION

OF

HEPATIC

P45Os

blocks carbon monoxide binding and catalytic activity (12,13,15). Subsequently, induction of associated catalytic activities (16, 17) appears to be a direct result of increased synthesis of P450 holoenzyme (2, 18). However, MDB-metabolite-P450 complex formation also has been associated with decreased turnover and subsequent accumulation of distinct isozymes (19). MDBs can be characterized as mixed inducers of several families of hepatic microsomal cytochromes P450. MDBs are perhaps best known for their ability to induce members of the P450IA family, isozymes also inducible by PAHs, dioxins, and PCBs (3, 10, 20, 21). However, MDBs also induce PB-responsive P45Os, primarily P450IIBl and P450IIB2 (3,22,23). Differential binding of MDBs with P450 isozymes, coupled with the parallel induction of specific P450 genes (10, 11, 24, 25), results in selective modulation of P450 isozyme pools. In rats, safrole, isosafrole, dihydrosafrole, and 4-n-alkoxyMDBs increase primarily P450IA isozymes (3, 10, 17, 26), while 5-t-butyl-MDB exclusively increases P450IIB isozymes (23). In mice, however, these same MDBs all induce primarily P450IIB isozymes (15). Extensive studies have demonstrated that interaction of PAHs and other substances with the Ah receptor mediates induction of P450IAl and P450IIA2 (2, 18), and it is clear that in uiuo exposure to MDBs results in the accumulation of P450IA mRNA and protein (10, 27). However, there is substantial question as to whether the Ah receptor is involved in the induction of P450IAl and P450IA2 by MDBs. Bigelow and Nebert (28) initially reported that MDBs are capable of binding to the Ah receptor, and that isosafrole could displace bound TCDD, but subsequent studies have been unable to confirm this (29) (Marcus, Denison, and Wilkinson, unpublished results). Likewise, MDBs are equally effective inducers of P450IA cytochromes in strains of mice that are PAH responsive as well as nonresponsive strains with altered Ah receptors (22,30-33). Although control of P450 levels clearly can be mediated by alterations in transcriptional events (18), other factors may also be of considerable importance in determining steady-state concentrations of these cytochromes (2). Formation of the P450-MDB-metabolite complex itself appears to stabilize P450 protein and may also trigger P450 induction (15, 22, 26). Noncoordinate induction of P450IAl and P450IA2 by xenobiotics (including MDBs) has been reported both in uiuo (27, 3335) and in vitro (19). Furthermore, recent evidence suggests that MDBs may actually elevate levels of cytosolic Ah receptor (29). Therefore, the mechanisms by which MDBs augment levels of P450IA cytochromes remain largely undefined. Even less is known regarding the molecular mechanisms of induction of P450IIB isozymes by MDBs and other xenobiotics (2, 18). While planar congeners of polychlorinated biphenyls selectively induce P450IA iso-

BY

METHYLENEDIOXYBENZENES

9

zymes, nonplanar congeners selectively induce P450IIB isozymes (36, 37), suggesting that this stimulation requires a related yet different receptor. Similarly, 5-t-butyl-MDB induces exclusively P450IIB isozymes (23), while 4-n-alkoxy-MDBs induce mainly P450IA isozymes (26). This study describes the effects of a series of 4-n-alkyl-MDBs on rat hepatic cytochrome P450 levels. Specifically, we show that selectivity between enhancement of P450IA cytochromes and P450IIB cytochromes depends on the chain length of the 4-n-alkyl substituent. We demonstrate that while 4-n-alkyl-MDBs are equally effective as PB in increasing P450IIB mRNA levels, the selectivity for increases in P450IIB protein correlate poorly to measured levels of the respective mRNAs. MATERIALS

AND METHODS

Chemicals.Methylenedioxybenzene, 4-n-methyl-MDB, 4-n-propyl-MDB, 4-acetophenone-MDB, 4-n-butane-Z-one-MDB, and 4-npentene-3-one-MDB were purchased from Frinton Laboratories (Vineland, NJ). Benzo[a]pyrene, resorufin, catechol, hexanoyl and octanoyl chloride, cesium fluoride, carbon disulfide, and dimethyl formamide were purchased from Aldrich Chemical Inc. (Milwaukee, WI). 4-t-Butyl-MDB was a generous gift from Dr. E. Hodgson, North Carolina State University. Ethoxy and pentoxyresorufin were purchased from Molecular Probes Inc. (Junction City, OR). Chemicals for SDS PAGE and nitrocellulose paper were purchased from Bio-Rad (Richmond, CA). Immunoreagents were purchased from ICN Immunobiologicals (Lisle, IL). GeneScreen Plus membranes and [y-32P]ATP (>3000 Ci/mmol) were from New England Nuclear/DuPont (Boston, MA). All other chemicals, solvents, and reagents were of the highest purity commercially available. Synthesis of MDBs. 4-n-Ethyl-MDB and 4-n-butyl-MDB were prepared from 4-n-acetophenone and 4-n-butane-2-one, respectively, via Clemmensen reduction in concentrated hydrochloric acid with zinc amalgam as catalyst. 4-n-Propyl-MDB was obtained by catalytic hydrogenation of 4-n-allyl-MDB (safrole) as described previously (17). 4-n-Pentyl-MDB was obtained by Clemmensen reduction of 4-rz-pentene-3-one-MDB followed by catalytic hydrogenation of the unsaturated carbon-carbon bond in the presence of platinum oxide as catalyst. 4-n-Hexyland 4-n-octyl-MDB were synthesized by Friedell Crafts acylation of 4-n-alkyl-catechols and the appropriate alkanoic anhydride or chloride, and subsequent closure of the methylene bridge via methylation with dichloromethane catalyzed by cesium fluoride in N,N-dimethylformamide (38). Synthetic MDBs were purified by vacuum distillation and/or silica gel column chromatography with benzene. Purity was in excess of 98% as judged by thin-layer chromatography or by FTIR and FTNMR. Animal treatments and tissue preparation. Male SpragueeDawley rats were obtained from Harlan Sprague-Dawley (Madison, WI) or Blue Spruce Farms (Altemont, NY). Fisher F344 male rats were obtained from Harlan Sprague-Dawley. Rats were housed in suspended wire cages for several days prior to treatment. Animals treated with nalkyl-MDBs were injected ip with 0.5 ml corn oil containing a single MDB compound to achieve a final dose of 1 mmol/kg/day for 3 days. Although dose-response determinations for induction by each MDB were not conducted, the doses utilized (1 mmol/kg/day for 3 days) have been shown to elicit maximal responses (with respect to P450IAl) for other MDB analogs. Phenobarbital was administered similarly except in 0.5 ml saline to give a final dose of 100 mg/kg/day. Control animals received corn oil only. Microsomes were prepared 24 h following the final treatment. Food (Purina Lab Chow) and water were provided ad libitum, except that food was withdrawn 24 h prior

10

MARCUS

to microsome preparation. Liver microsomes were prepared as previously described (39), rapidly frozen, and stored in liquid nitrogen. In some cases, portions of freshly removed hepatic tissues were individually quick frozen in liquid nitrogen, then stored at -80°C for subsequent RNA preparation. Analytical methods. Cytochrome P450 content of rat liver microsomes was quantitated from the reduced carbon monoxide difference spectrum (40) UtiliZing an extinction coefficient of 91 mMm’ cm-’ for the wavelength pair 490-450 nm. Cytochrome P450-MDB-metabolite complexes were measured spectrally from oxidized/reduced difference spectra but using NADH (0.16 mM) as the reductant (26). Protein was measured according to Lowry et al. (41), utilizing bovine serum albumin as the standard. Electrophoresis and Western blotting. Sodium dodecyl sulfatepolyacrylamide electrophoresis and Western blotting were conducted as previously described (9). Immunoreactive protein bands were quantitated by the PAP-diaminobenzidine reaction and a scanning transmission densitometer utilizing a tungsten light source. Peak areas were quantitated by microcomputer-assisted integration. Standard curves with purified P450 isozymes were run on each gel. Samples and standards were in the range of l-100 ng P450 protein per lane (l-100 pg of total microsomal protein per lane). Replicate samples were run at several protein concentrations, and results averaged. Standard deviations were typically less than 10% between samples. Preparation of purified proteins and antibodies. Cytochromes P450IA1, P450IIB1, and P450IIB2 were purified from male SpragueDawley rat livers as previously described (2539). Antibodies to these purified enzymes were generated in New Zealand White rabbits (42) and purified IgG fractions obtained from Protein A-Sepharose (9,39). RNA preparation. For each treatment group, livers were excised, divided in half, and then pooled. One-half of each pooled sample was used to prepare total RNA as previously described (43,44) and modified to include the use of lithium chloride precipitation (45,46). Synthetic oligonucleotides. Two 18-mer synthetic oligodeoxyribonucleotides, ,5-d(GGTTGGTAGCCGGTGTGA)-3’ and 5’-d(GGAcomplementary to the mRNAs for TGGTGGCCTGTGAGA))3’, P450IIBl and P450IIB2, respectively, and two 20.mers, 5’.d(TCTGGTGAGCATCCAGGACA)3’ and 5’-d(GGAAAAGGAACAAGGGTGGC)-3’, complementary to the P450IAl and P450IA2 mRNAs, respectively, were synthesized and end-labeled with 32P and T4 polynucleotide kinase as described previously (43, 47). These oligomer probes were utilized for all Northern blotting and solution hybridization experiments. Solution conducted P450IIB2

Solution hybridizations hybridization experiments. with the synthetic oligonucleotides to P450IIBl with rat hepatic RNA as previously described (43).

were and

Northern blotting. Isolated rat hepatic RNA was electrophoresed in 1.25% denaturing formaldehyde-agarose gels, blotted to nylon (GeneScreenPlus) membranes, and employed for subsequent hybridizations as previously described (47). Metabolism assays. Cytochrome P450-mediated microsomal benzo[a]pyrene hydroxylase activity (AHH) and pentoxyresorufin Odeethylase activity (PRD) were determined fluorometrically by published procedures (2548).

RESULTS

In Vitro Metabolism

Studies

For the initial experiment Sprague-Dawley male rats were treated with one of nine 4-n-alkyl-MDBs (n = O8) or 4-t-butyl-MDB. Animals received 1 mmol/kg/day for 3 consecutive days and hepatic microsomes were prepared 24 h following the final treatment. The resulting

ET

AL.

levels of carbon monoxide-binding P450, AHH and PRD activities are shown in Fig. 1. While the magnitude of the reduced carbon monoxide difference spectrum is generally increased by all MDBs (n = O-8), the effect of the MDB side-chain length appears biphasic. Stimulation of carbon monoxide binding peaked at both n = 2 and n = 5 and was considerably lower for either the shortest (n = 0, 1) or longest side chains (n = 8). AHH and PRD activities are functional markers for cytochromes P450IA and P450IIB, respectively (25, 48). These two activities exhibited very different dependencies on side-chain length. AHH activity showed a maximum for n = 4,5,6, yet PRD was maximal with shorter side-chain substituents (n = 2, 3). In general, peak AHH and PRD activity also corresponded to peaks in the amount of carbon monoxide-binding cytochrome. t-Butyl-MDB more closely resembled the shorter-chain MDBs (n = 2,3), since PRD enhancement was greater (5.6-fold) relative to AHH increases (1.7fold). Increasing the steric hindrance of the side chain therefore decreased inductive effectiveness toward P450IA activity in favor of P450IIB activity. These results suggest that these MDBs can modulate two distinct mechanisms of cytochrome P450 enhancement that regulate P450IIB and P450IA isozymes. Unsubstituted MDB (n = 0) failed to increase these activities yet clearly elevated total carbon monoxide-binding cytochrome P450. Total cytochrome P450 measured by carbon monoxide binding and P450 isozymes measured by specific catalytic activities can be significantly underestimated due to residual MDB-metabolite complex formation (10,26, 49). Additional optical difference spectroscopy (26) indicated formation of such complexes in hepatic microsomes prepared from rats treated in uiuo with n-alkylMDBs (average of 0.44 or 0.71 nmol/mg protein in liver microsomes from C4-MDBand CG-MDB-treated rats, respectively). However, this technique cannot identify the proportions of specific isozymes which are complexed. Results from these experiments, as with most previous studies (10, 26, 49), must therefore be interpreted with caution, since isozyme specificities for complex formation and inhibitory potencies are unknown. More accurate quantitation of the effects of MDBs on specific P450 isozymes was therefore effected by immunological methods. Immunoquantitation of Cytochrome P450 Apoenzyme levels of individual isozymes of cytochrome P450 were quantitated by Western blotting for selected microsomes utilized for the in vitro metabolism studies. These results, shown in Fig. 2, also support the conclusion that alkyl-MDBs are “mixed” inducers of both the P450IA and P450IIB families of cytochrome P450 with selectivities dependent upon the MDB substituent.

INDUCTION

OF

HEPATIC

0.

,o P-450

A-A

AHH

q

- -0

P45Os

BY

11

METHYLENEDIOXYBENZENES

PRD

2.5

.;

2.0

5 FL

1.5 1.0

i? 2 10 t d Q)

/ 0.

-,

Con

CO

0.5

\

d

-0.

’ q- ’

’ ‘O-

- .o

Y

I-

Cl

C2

C3

C4

C5

z c

C6

0.0

C8

Treatment FIG. 1. Structure-activity relationships for the induction of rat liver microsomal metabolism and cytochrome P450 by alkyl-methylenedioxybenzenes. Carbon monoxide-binding cytochrome P450, aryl hydrocarbon hydroxylase (AHH) (benzo[a]pyrene oxidation) activity, and pentoxyresorufin 0-deethylase (PRD) activities were measured in rat hepatic microsomes. Male Sprague-Dawley rats (200 g) were treated for 3 days with 1 mmol/kg/day methylenedioxybenzene (MDB) substituted at the 4-position with no alkyl groups (CO), methyl (Cl), ethyl (C2), n-propyl (C3), n-butyl (C4), tertiary butyl (t-butyl), n-pentyl (C5), n-hexyl (C6), or n-octyl (C8) alkyl groups. Results are means of triplicate assays of pooled microsomes from three animals. Standard deviations are shown for t-butyl analogs for each assay and are representative of all samples.

In agreement with in vitro activity measurements, the side-chain substituent dependence for P450IIB cytochrome activity was different from the dependence for P450IA. Thus enhancement of P450IA was highest for C4-MDB and CG-MDB and lowest for t-butyl-MDB,

while P450IIB enhancement was highest for t-butylMDB. For all isozymes, C%MDB was the least effective inducer. Most surprisingly, the structure-activity dependence for P450IIBl was very different from that for P450IIB2. Enhancement of P450IIBl levels was far

EQB P45OllBl 0 P4501182 EE4 P450lAl

Control

tBut-MDB

C4-MDB

CG-MDB

C8-MDB

Treatment FIG. 2. Effect of in uiuo administration of alkyl-methylenedioxyhenzenes on cytochrome P450 isozymes in rat hepatic microsomes. P450 isozymes were quantitated by Western blotting. Immunoreactive bands were visualized by the PAP-diaminobenzidine reaction and quantitated by densitometry. Male Sprague-Dawley rats (100 g) were treated for 3 days with 1 mmol/kg/day methylenedioxybenzene (MDB) substituted at the 4-position with none (control), tertiary hutyl (tBut), n-butyl (C4), n-hexyl (C6), or n-octyl (C8) alkyl groups. Results shown are means + SD of triplicate measurements of pooled microsomes from three animals.

12

MARCUS TABLE

I

Selective Effects of n-Alkyl-methylenedioxybenzenes on Rat Hepatic Microsomal Cytochromes P450IA1, P450IIB1, and P450IIB2 Cytochrome Treatment” None (control) PB MC C4-MDB CG-MDB C8-MDB

P450

(pmol/mg

P450IIBl

P450IIB2

92 7 974 f 174 2+ 1 60? 20 59k 6 13+ 6

61Ifr 20 473 f 170 12f 5 194 + 62 279 k 27 49+ 21

microsomal @l/B21 (0.15)

(2.0) (0.16) (0.31) (0.21) (0.26)

protein)* P450IAl 5+ 2 8+6 894 f 150 186 f 56 356 + 67 132 + 24

’ Male Sprague-Dawley rats (100 g) were treated for 3 days with 1 mmol/kg/day methylenedioxybenzene (MDB) substituted at the 4position with n-butyl (CXMDB), n-hexyl (CG-MDB), n-octyl (C8MDB), MC (40 mg/kg/day), or PB (100 mg/kg/day). Injections were 0.5 ml corn oil ip, except for PB in saline. Control animals received corn oil only. b Results shown are means f SD of P450 apoenzyme per milligram of total microsomal protein from replicate assays of each of at least six individual rats. Immunoreactive protein bands were visualized by the PAP-diaminobenzidine reaction and quantitated by densitometry.

more sensitive to MDB side-chain substitution structure than P450IIB2. Increases in P450IIBl were maximal for t-butyl-MDB but absent for C8-MDB. P450IIB2 levels, however, were enhanced at least 4-fold by each of the four MDBs examined, with a maximum 12-fold increase for C4-MDB. The levels of P450IIB2 protein were greater than the levels of P450IIBl in all MDB-treated animals except those treated with t-butyl-MDB, for which the ratio of P450IIBl relative to P450IIB2 remained 2-3, a value typical of PB-induced liver. The rank order of enhancement of P4501IB cytochromes (tbutyl > C4 > C6 > C8) paralleled PRD activities while the rank order for P450IAl (C4 > C6 > C8 > t-butyl) paralleled AHH activities. P450IA2 levels were also substantially increased in microsomes prepared from MDB-treated rats. Compared to untreated control animals, C4-MDB, CG-MDB, C8-MDB, and t-butyl-MDB resulted in increases of 20-, 15-, 12-, and 1.7-fold, respectively, measured by the relative immunological staining intensity of the bands per milligram of microsomal protein. Enhancement of P450IA2 apoenzyme by MDBs thus closely paralleled effects of MDBs on P450IAl. In a second experiment with younger 100-g rats (Table I) CG-MDB was again the most effective inducer of P450IIB1, P450IIB2, and P450IAl of the MDBs tested. As in the initial experiments, in vivo administration of n-alkyl-MDBs resulted in levels of P450IIB2 apoenzyme in rat hepatic microsomes that exceeded those of P450IIBl. Ratios of P450IIBl/P450IIB2 were approximately 1:4, the reverse of the 2-3:l ratio usually obtained

ET

AL.

subsequent to induction with PB, the prototypical inducer of the P4501IB family. Maximal levels of P450IIB1, P450IIB2, and P450IAl protein induced by the MDBs were generally less than one-half those obtained with PB or BNF. Treatment with n-alkyl-MDBs did not alter P450IIB1, P4501IB2, P450IA1, or P450IA2 contents in microsomes from rat kidney, adrenal, or small intestine (data not shown). P450 mRNA and Immunoreactive Protein Quantitation To further investigate the effects of n-alkyl-MDBs on the expression of P450 isozymes, additional experiments were conducted to measure levels of P450 mRNAs in C4MDB- and CG-MDB-treated rat livers, and these levels were compared to those in control (untreated) and PBtreated livers. One half of the pooled liver tissue from each group of rats was utilized for microsome preparation and Western blotting; the other half for RNA preparation. Northern blotting. Northern blots of total RNA isolated from livers from each treatment group were carried out with 32P-labeled oligonucleotide probes specific for P450IIBl or P450IIB2 mRNA (43, 47). This technique was used to investigate the effects of n-alkyl-MDBs on steady-state levels of P450IIBl and P450IIB2 mRNA in rat liver. The two 18-mer cDNA probes (differing by four bases and utilized under conditions of high stringency) distinguish between mRNAs for P450IIBl and P450IIB2 which are 98% similar in nucleotide sequence. Autoradiographs of the resulting blots are shown in Fig. 3. The single bands on each blot confirm the apparent selectivity of these probes. Consistent with the metabolism and Western blot assays, both C4-MDB and C6MDB produced substantial increases in the levels of P450IIBl and P450IIB2 mRNA. However, in contrast to the Western blot (immunoreactive protein) data, P450IIBl and P450IIB2 mRNAs in MDB-treated animals accumulated to nearly equal levels, and were similar to those in PB-treated animals. Treatment with both MDB derivatives also resulted in marked increases in hepatic P450IAl and P450IA2 mRNA levels as determined by Northern blotting with two 20-mer oligomers selective for P450IAl and P450IA2 mRNA. Solution hybridization. These experiments more precisely quantitated effects of MDBs on the levels of P450IIBl and P450IIB2 mRNA. The results of these determinations are shown in Table II. Administration of C4- and CG-substituted MDBs produced increases in steady-state levels of rat hepatic P450IIBl and P450IIB2 mRNA that were approximately 70 and 90%, respectively, of those observed in the PB-induced samples. These data agreed well with those from the Northern blot experiments (Fig. 3). As expected, PB markedly elevated both P450IIBl and P450IIB2 apoenzyme and mRNA levels, particularly P450IIBl.

INDUCTION

P450

IlBl

OF

HEPATIC

P450

P45Os

BY

13

METHYLENEDIOXYBENZENES

P450

llB2

IA1

P450

IA2

FIG. 3. Effect of alkyl-methylenedioxybenzenes on cytochrome P450 mRNA levels in rat livers. Autoradiogram of Northern blot of 10 pg total liver RNA per lane. RNA was prepared from male Sprague-Dawley rats (100 g) treated for 3 days with 1 mmol/kg/day methylenedioxybenzene (MDB) substituted at the 4-position with none (control, Co) n-butyl (C4), n-hexyl (C6); or PB (100 mg/kg/day). Hybridization probes were 32P-labeled synthetic oligomers specific for each mRNA (43,44).

Treatment with MDBs produced different selectivities with immunoreactive protein and mRNA. Thus C6MDB or C4-MDB each produced approximately threeto fourfold more P450IIBl than P450IIB2 mRNA in MDB-treated rat livers. However, this was in contrast to the Western blot results (Table II) which indicated that three- to fivefold more P450IIB2 than P450IIBl apoenzyme is present in MDB-treated rat livers. While MDB-induced P450IIBl mRNA levels were 67% of those found in PB-treated rat livers, P450IIBl upoenzyme levels were only 6% of those in PB-treated rat livers. The concentrations of apoenzyme and mRNA was much closer for P450IIB2, 71 and 50%, respectively, compared to PB-induced levels. These results strongly suggest that MDBs exert marked but selective post-

transcriptional effects, potentially tion of P450IIBl mRNA. DISCUSSION

Previous studies have shown that MDBs augment levels of individual P450 cytochromes with a selectivity that depends on the nature of benzene ring substitution. Thus, 4-t-butyl-MDB increases P450IIB cytochromes (15,23) but not P450IA cytochromes, while the opposite is true of 4-n-alkoxy-MDBs and isosafrole, the latter agent being particularly selective for P450IA2 (10, 17, 26). In this study we show, from examination of a series of 4-n-alkyl-MDBs (n = O-8), that selectivity for stimulation of P450IIB and P450IA cytochromes is highly de-

TABLE Effects

of Alkyl-methylenedioxybenzenes

on Rat mRNA-solution (fmol X lO’/pg

Hepatic

at the level of transla-

II

Cytochrome

P450IIBl

and

P450IIB2

hybridization total RNAb)

Apoenzymes

Apoenzyme-Western (pmol/mg microsomal

Treatment”

P450IIBl

P450IIB2

IIBl/IIBZ

P450IIBl

Untreated (control) Phenobarbital CI-MDB CG-MDB

NDd 5.9 3.9 4.2

0.07 1.4 1.3 1.3

4.1 3.1 3.3

8f 850 ?I 150 f 70 Ik

and blotting protein’)

P450IIB2

2 75 20 10

a Animal treatments as described under Materials and Methods. * Values represent means of two experiments, each conducted in duplicate; variations < 10%. Five hybridized to 4000 cpm of the respective a2P-oligomer, as described under Materials and Methods. ’ P450 apoenzyme was quantitated by Western blotting (pmol P450 immunoreactive protein/mg total Materials and Methods. Results are means * SD for n = 3. d Not detectable within sensitivity limits of this assay (~15 cpm above background).

mRNAs

IIBl/IIB2

15. 2 670 ? 45 310 f 50 320 f 50

micrograms microsomal

of total protein)

0.5 1.3 0.5 0.2

hepatic

RNA

as described

was under

14

MARCUS

ET AL.

pendent upon alkyl chain length. P450 concentrations chromes do not simply parallel hydrophobicity (sidedetermined from the carbon monoxide complex in- chain length) but are instead optimal for a range of alkyl creased with an apparent biphasic dependence on chain chain lengths (n = 4-6) and conformations. However, length, suggesting the possibility of two distinct en- since these techniques measure only “free” cytochrome, hancement processes. the formation of MDB-metabolite-P450 complexes conQuantitation of P450IIB-specific activity (pentoxyfounds interpretation of this type of experiment. resorufin dealkylation) indicates that this subfamily of Although P450IIBl and P450IIB2 are greater than isozymes is selectively increased by short-chain n-alkyl97% homologous (51, 52), constitutive expression of MDB substituents (IL = 2, 3). In contrast, the P450IA these P450IIB genes exhibits specificity between liver family (indicated both by AHH activity and immuno(predominantly P450IIB2) and lung (predominantly quantitation), is selectively increased with much longer P450IIBl). Treatment with PB induces P450IIBl in 4’-substituents (n = 5-8). Although the intermediatepreference to P450IIB2 in a ratio of approximately 2-3: length substituents (n = 4) were more effective as 1(6,8,9). This study (Table I) provides the first example P450IA inducers, they exhibited considerable “mixed” where chemical treatment in viuo selectively increases stimulatory behavior, and significantly elevated P450IIB P450IIB2 protein relative to P450IIBl. Proceeding from levels. Like other “mixed-type” inducers such as isosaf- 4-t-butyl-MDB to longer (and potentially more planar) role and Arochlor 1254 (3,23), n-alkyl-MDBs do not in- 4-n-alkyl-MDBs, selectivity progressively shifts toward crease these activities as well as the prototypical inducP450IIB2. This shift is apparently accounted for by a ers PB and BNF. much greater decline (loo-fold for CG-MDB) in the stimComparisons between catalytic activities (Fig. 1) and ulation of P450IIBl as compared to P450IIB2. P450IIB2 P450 isozyme apoenzyme levels (Fig. 2) indicate that in- levels in C8-MDB-induced microsomes, like P450IIB1, creased PRD activity is highly correlated with P450IIBl were not markedly different from controls (even though content (r2 = 0.99), while increased AHH activity is P450IAl was substantially induced by this compound), highly correlated with P450IAl content (r2 = 0.93). The in agreement with in vitro metabolism studies (Fig. 1). correlation of PRD activity with P450IIB2 or total Previous work suggests that MDBs increase levels of P450IIB (P450IIBl + P450IIB2) content was substan- specific P450 cytochromes through both induction (entially less (r2 = 0.14 and 0.80, respectively), consistent hanced synthesis) and decreased degradation of the prowith the greater than tenfold lower activity of P450IIB2 tein. Isosafrole increases steady-state levels of mRNA for this substrate (50). The correlation between AHH for both P450IA and P450IIB cytochromes (23), as well activity and P450IAl content is somewhat lower in part as decreasing the turnover of P450IA2 protein (19). In this study quantitation of total mRNA for each P450IIB because of the significant contributions to AHH activity arising from P450IIB and other cytochromes induced by correlated poorly with immunoreactive protein levels. Northern blot and solution hybridization experiments t-butyl-MDB and short-chain substituted n-alkylMDBs (39). These results and previous results with 4-n- with discriminatory oligomer probes indicate that both C4-MDB and CG-MDB were very effective inducers of alkoxy-MDBs (26) indicate that replacement of oxygen by methylene at the 4-position does not substantially al- both P450IIBl and P450IIB2 mRNAs. Both compounds ter corresponding activity profiles. In both series of induced P450IIBl and P450IIB2 mRNA levels to 70 and MDBs, peak P450IAl-like activity is achieved for side 90%, respectively, of those measured in PB-treated anichains n = 6, and P450IIB-like activities are greatest at mals. With MDB as well as with PB treatment three- to fourfold more P450IIBl mRNA was present in the liver n = 2. Replacement of n-butyl by t-butyl changes the selec- than P450IIB2 mRNA (Table II). The reversal of apoenzyme ratios in n-alkyl-MDBtivity of the enhancement effect on apoenzyme levels by P450IIBl:P450IIB2 increasing total P450IIB levels while removing most of treated rat livers (compared to PB and other inducers) the capacity to increase P450IA, in agreement with pre- therefore cannot be accounted for by the respective mRNA levels but would be consistent with an MDB-invious reports (23). The preference of C3-MDB (dihydrosafrole) for P450IIBl over P450IAl also indicated a duced alteration in translational efficiency or in protein similar response to that previously reported for turnover. Levels of hepatic P450IIB apoenzyme lower than preisosafrole (23). dicted from measured mRNA levels (as compared with Together, these results suggest that side-chain length PB, Table II) also suggest that MDBs regulate P450 levof the alkyl-MDBs is the prime determinant for strucels by altering events distinct from gene transcription. ture-activity considerations selecting between P450IIB and P450IA enhancement mechanisms. Chains longer However, another possible explanation for the apparent noncoordinate expression of P450IIBl/IIB2 mRNA verthan n = 3 become progressively less effective as enhancsus apoenzyme may be that MDBs induce another memers of P450IIB1, although inducing P450IA1, while shorter chains, including t-butyl-MDB, show preference ber of the P450IIB family. If the mRNAs for other IIB isozymes also hybridize with the oligomer probes, then for increasing P450IIBl. Increases in P450IA cyto-

INDUCTION

OF

HEPATIC

P45Os

the estimations of P450IIBl or P450IIB2 mRNA levels may have been inflated. Although additional genes are in fact recognized by the IIBl and IIB2 probes employed in this study (47), it is not yet clear how many of these IIB subfamily genes are expressed. Only single bands were resolved on Northern blots of RNA from MDBtreated livers with our oligo probes. Indeed, recent evidence suggests that two other IIB subfamily member mRNAs are expressed, but at levels substantially lower than that observed for induced levels of IIBl or IIB2 (53). In separate Northern blot experiments (data not presented), no evidence for an MDB-induced IIB3 or IIB4 mRNA product was derived. However, an anomalous band was observed on Western blots with antiP45OIIBl/B2 IgG from rat hepatic microsomes from C4MDB, CG-MDB, and C8-MDB treatments. This band had an electrophoretic mobility slightly slower than that of P450IIB2, and the relative intensity of this band was closely correlated with that of P450IIB2 (rather than P450IIBl). Additional work is in progress to characterize this protein. Discrepancies between the relative levels of P450 apoenzyme and mRNA levels have also been reported by others (3, 23, 27, 54-56) and there is some doubt as to whether MDBs require a specific receptor to exert their effects on cytochrome P450 systems. MDBs do not appear to interact directly with the Ah receptor for P450IA induction (29), and a specific receptor for P450IIB induction has yet to be discovered. Several reports indicate that strong complex formation with individual P450 isozymes may result in selective increases of the complexed isozyme (19, 20,55,66). Strong complex formation may elevate levels of cytochrome P450 either by providing stabilization against degradation or by increasing the capacity of the cytochrome to compete for a limiting pool of free heme. This effect may be especially important for P450IA2, the isozyme that is most effectively induced by MDBs and forms the most stable complex (10). Nonetheless, data from studies with P450IAl and P450IA2 indicate that these isozymes do in fact share some common mechanisms of control (such as the Ah locus), but that each isozyme is also regulated independently, depending upon the inducer (3,4,18,29,57-59). These data, together with those reported here, suggest that mechanisms other than direct transcriptional modulation can be significant effecters of P450 activity in uiuo. Presumably, the structures of the different MDB molecules correspond to the structures of the relevant protein binding site(s) that interact with different regulatory pathways controlling P450 expression. Therefore both structural conformation and hydrophobicity appear to be important parameters for dictating molecular selectivity and efficacy of P450 inducers. Although the exact mechanisms accounting for the differential effects of n-alkyl-MDBs on P450IAl and P450IA2, and P450IIBl and P450IIB2, remain to be elu-

BY

15

METHYLENEDIOXYBENZENES

cidated, it is clear from the hybridization data that nalkyl-MDBs are effective inducers of the steady-state levels of these P450 mRNAs. Consequently, the failure of C4-MDB and CG-MDB to increase P450IIBl apoenzyme is evidently not due to a lack of transcriptional activation or enhanced rate of mRNA turnover. Rather, it appears probable that the MDBs may exert selective negative effects on translational efficiencies and/or specifically modulate turnover kinetics of the corresponding proteins. Due to the apparently complex interactions of MDBs with components regulating the activity of individual P45Os, additional studies are in progress to fully detail the effects of these compounds on the P450 system. ACKNOWLEDGMENTS This work was supported by National Institutes of Health Grants GM-32281 and ES-04696 (to C.J.O.) and CA-16265 (to C.R.J.), and in part by American Cancer Society Institutional Grant IN-17 (to C.B.M.). N.M.W. was supported in part by National Institutes of Environmental Health Sciences Training Grant ES-07015 The authors thank Karen Rudd for her excellent technical assistance in portions of this work, and David Presser and Mattson Instruments for the FTIR analyses of synthetic methylenedioxybenzenes. REFERENCES 1. Nebert, D. W., Adesnik, M., Coon, M. J., Estabrook, R. W., Gonzalez, F. J., Guengerich, F. P., Gunsalus, I. C., Johnson, E. F., Kemper, B., Levin, W., Phillips, I. R., Sato, R., and Waterman, M. (1987) DNA 6, l-11. 2. Nebert, D. W., and Gonzalez, 56,945-993.

F. J. (1987)

Arznu.

Reu. Biochen.

3. Thomas, P. E., Reik, L. M., Ryan, D. E., and Levin, W. (1983) J. Biol. Chem. 258,4590-4598. 4 Parkinson, A., Safe, S. H., Robertson, L. W., Thomas, P. E., Ryan, D. E., Reik, L. M., and Levin, W. (1983) J. Biol. Chem. 258,59675976. 5. Negishi, 11023.

M., and Nebert,

D. W. (1979)

6. Rampersaud, A., and Walz, USA 80,6542-6546.

J. Biol. Chem.

F. G., Jr. (1983)

Proc.

254,11015-

Natl.

Acad.

Sci.

7. Vlasuk, G. P., Ghyrayeb, J., Ryan, D. E., Reik, L., Thomas, P. E., Levin, W., and Walz, F. G., Jr. (1982) Biochemistry 21, 789-798. 8. Vlasuk, G. P., Ryan, D. E., Thomas, P. E., Levin, W., and Walz, F. G., Jr. (1982b) Biochemistry 21,6288-6292. 9. Wilson, N. M., them. Biophys.

Christou, 256(2),

10. Ryan, D. E., Thomas, 255,7941&7955.

M., and Jefcoate, 407-420. P. E., and Levin,

11. Wilkinson, C. F., Murray, them. Toxicol. 6,27-63. 12. Franklin,

M. R. (1971)

M.,

Xenobiotica

and

C. R. (1987) W. (1980)

Marcus,

Arch.

J. Biol.

C. (1984)

Rev.

BioChem. Bio-

1,581-591.

13. Philpot, R. M., and Hodgson, E. (1971) Life Sci. lO(II), 503-512. 14. Hodgson, E., and Philpot, R. M. (1974) Drug. Metab. Reu. 3,231301. 15. Cook, J. C., and Hodgson, E. (1983) Toxicol. Appl. Pharmacol. 68, 131-139. 16. Kamienski, F. X., and Murphy, macol. 18,883-894.

S. D. (1971)

Toricol.

Appl.

Phar-

16

MARCUS

17. Murray, M., Wilkinson, Pharmacol. 68,66-76.

C. F., and Dube,

C. (1983)

Tonicol.

Appl.

18. Whitlock, J. P. (1986) Annu. Reu. Pharmacol. Toricol. 26, 333369. 19. Steward, A. R., Wrighton, S. A., Pasco, D. S., Fagan, J. B., Li, D., and Guzelian, P. S. (1985) Arch. Biochem. Biophys. 241,494-508. 20. Hardwick, J. P., Linko, macol. 27,676-682.

P., and Goldstein,

J. A. (1985)

21. Dickens, M., Bridges, J. W., Elcombe, (1978) Biochem. Biophys. Res. Commun.

C. R., and 80,89-96.

Mol. Netter,

PharK. J.

M., Zaluzny,

L., and Farrel,

G. C. (1986)

Arch.

Biochem.

251,471-478. C., Murray,

M., and Wilkinson,

C. F. (1985)

26. Murray, M., Wilkinson, C. F., Marcus, Mol. Pharmacol. 24,129-136.

C., and Dube,

28. Bigelow, 112.

S. W., and Nebert,

29. Cook, J. C., and Hodgson, 299-315. 30. Cook, J. C., and Hodgson, 233-240. 31. Ohyama, T., Nebert, 259,2675-2682.

D. W. (1982)

42. Thomas, P. E., Lu, A. Y. H., Ryan, D. E., Kawalek, W. (1976) J. Biol. Chem. 251,1385-1391. 43. Omiecinski, C. J., Walz, Chem. 260,3247-3250.

F. G., Jr., and Vlasuk,

46. Cathala, Martial,

Ku-

10, 109-

Lett.

C. M., and Omiecinski,

C. J. (1987)

50. Wolf, Kelly,

E. (1986)

Chem.-Biol.

Interact.

58,

51. Yuan, P. M., Ryan, D. E., Levin, W., Proc. Natl. Acad. Sci. USA 80,1169-1173.

J. Biol.

Chem.

M. (1984)

33. Ikeda, T., Altieri, M., Chen, Y. T., Nakamura, M., Tukey, and Nebert, D. W. (1983) Eur. J. Biochem. 134,13-18.

R. H.,

S. C., Gozurka, Arch. Biochem.

E., GuenBiophys.

C. R., Miles, K., and Smith,

Pharmacol. J. M. (1976)

25,

185-

J. D., and

13,924-938. Tetrahedron

Mol.

M.,

31,

Pharmacol.

J., Haaper34,3337-

S. S., Burke, W. E. (1988)

M. D., Rospendowski, B. N., Biochemistry 27,1597-1603. and Shively,

J. E. (1983)

52. Fujii-Kuriyama, Y., Mizukami, Y., Kawajiri, K., Sogowa, K., and Muramatasu, M. (1982) Proc. N&l. Acad. Sci. USA 79, 27932797. 53. Giachelli, C. M., Lin-Jones, J. L., and Omiecinski, C. J. (1989) J. Biol. Chem. 264,7046-7053. 54. Simmons, D. L., McQuiddy, Chem. 262,326-332.

P., and Kasper,

55. Schuetz, E. G., Wrighton, S. A., Safe, (1986) Biochemistry25,1124-1133. 56. Pershing,

35. Goldstein, J. A., and Linko, P. (1984) Mol. Pharmacol. 191. 36. Goldstein, J. A., Hickman, P., Bergman, H., McKinney, Walker, M. P. (1977) Chem.-Biol. Interact. 17,69-87. H. L., and Miller,

J. Biol.

49. Elcombe, C. R., Bridges, J. W., Gray, T. J. B., Nimmo-Smith, R. H., and Netter, K. J. (1975) Biochem. Pharmacol. 24, 14271433.

54,

Mol.

G. P. (1985)

G., Savouret, J-F., Mendez, B., West, B. L., Karin, J. A., and Baxter, J. D. (1983) DNA 2,329-335.

Interact.

E. (1977)

J., and Levin,

C. M., and Omiecinski, C. J. (1986) J. Biol. Chem. 1359-1363. 45. Frazier, M. L., Mars, W., Florine, D. L., Montagna, R. A., and Saunders, G. F. (1983) Mol. Cell. Biochem. 56,113-122.

Chem.-Biol.

34. Fagan, J. B., Pastewka, J. V., Chalberg, gerich, F. P., and Gelboin, H. V. (1986) 244,261-272.

R. J.

261(3),

E. (1985)

D. W., and Negishi,

N. L., and Randall,

48. Burke, D. M., Thompson, S., Elcombe, C. R., Halpert, anta, T., and Mayer, R. T. (1985) Biochem. Pharmacol. 3345.

D. R.,

38. Clark, J. H., Holland, Lett. 38.3361-3364.

0. H., Rosebrough, N. J., Farr, J. Biol. Chem. 193,265-275.

C. E. (1983)

D. E., Koop, 235,239-253.

A., and Glover,

41. Lowry, (1951)

47. Giachelli, 4777484.

K., and Fujii-

Toxicol.

32. Thomas, P. E., Reidy, J., Reik, L. M., Ryan, and Levin, W. (1984) Arch. Biochem. Biophys.

37. Poland,

39. Wilson, N. M., Christou, M., Turner, C. R., Wrighton, S. A., and Jefcoate, C. R. (1984) Carcinogenesis 5,1475-1483. 40. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239,2370-2378.

Xenobiotica

15(4),351-362.

27. Kawajiri, K., Gotoh, O., Tagashira, Y., Sogawa, riyama, Y. (1984) J. Biol. Chem. 259,10145-10149.

AL.

44. Giachelli,

22. Fennel, T. R., Sweatman, B. C., and Bridges, J. W. (1980) Chem. Biol. Interact. 31,189-201. 23. Yeowell, H. N., Linko, P., Hodgson, E., and Goldstein, J. A. (1985) Arch. Biochem. Biophys. 243(2), 408-419. 24. Murray, Biophys. 25. Marcus,

ET

L. K., and Franklin,

S.

M. R. (1982)

C. B. (1987)

J. Biol.

H., and Guzelian,

P. S.

12,687-

Xenobiotica

699. 57. Dannap, G. A., Guengerich, F. P., Kaminsky, (1983) J. Biol. Chem. 258,1282-1288. 58. Linko, (1986)

P., Yeowell, J. Biochem.

H. N., Gasiewicz, Toxicol. 1,96-107.

59. Fennel, T. R., Dickins, Pharmacol. 28,1427-1429.

M.,

L. S., and Aust,

T. A., and Goldstein,

and Bridges,

J. W. (1979)

S. D. J. A.

Biochem.