Inhibition of cytochrome P-450 2E1 by diallyl sulfide and its

Nov 1, 1991 - John F. Brady, Hiroyuki Ishizaki, Jon M. Fukuto, Marie C. Lin, Addi Fadel, ... Ute M. Kent, Elizabeth S. Roberts, Jarin Chun, Kimberly H...
0 downloads 0 Views 777KB Size
Chem. Res. Toxicol. 1991, 4 , 642-647

642

Art i d e s Inhibition of Cytochrome P-450 2E1 by Diallyl Sulfide and Its Metabolites John F. Brady,tJ Hiroyuki Ishizaki,t Jon M. Fukuto,§ Marie C. Lin,t Addi Fadel,' Jeanne M. Gapac,i and Chung S. Yang*it Laboratory for Cancer Research, College of Pharmacy, Rutgers University, Piscataway, New Jersey 08855, School of Pharmacy, University of the Pacific, Stockton, California 95211, and Department of Pharmacology, Center for the Health Sciences, University of California at Los Angeles, Los Angeles, California 90024 Received May 6, 1991 Diallyl sulfide, a major flavor ingredient from garlic, was previously shown to inhibit chemically induced carcinogenesis and cytotoxicity in animal model systems. It modulated cytochrome P-450 compositions by inactivating P-450 2E1 and inducing P-450 2B1. The present studies examined the inhibition of P-450 2E1 mediated p-nitrophenol hydroxylase activity by diallyl sulfide and its putative metabolites diallyl sulfoxide and diallyl sulfone (DASOJ. Each compound displayed competitive inhibition of p-nitrophenol hydroxylase activity in incubations using liver microsomes from acetone-pretreated male Sprague-Dawley rats. Preincubation of the microsomes with DAS02 inactivated p-nitrophenol hydroxylase activity in a process that was time- and NADPH-dependent and saturable, exhibited pseudo-first-order kinetics, was protected by alternate substrate, was accompanied by a loss of microsomal P-450-CO binding spectrum, and was unaffected by exogenous nucleophile. The Ki value for DAS02 was 188 pM and the maximal rate of inactivation was 0.32 min-'. DASOz was ineffective in the inactivation of ethoxyresorufin dealkylase, pentoxyresorufin dealkylase, or benzphetamine demethylase activity. Purified P-450 2E1 in a reconstituted system was inactivated in a time- and NADPH-dependent manner by DAS02. The metabolic conversion of diallyl sulfide to the sulfoxide and sulfone was observed in vivo and in vitro. The results suggest that diallyl sulfide inhibits the metabolism of P-450 2E1 substrates by competitive inhibition mechanisms and by inactivating P-450 2E1 via a suicide-inhibitory action of DAS02.

Introduction Diallyl sulfide (DAS),' a flavor component of garlic (Allium sativum), has been shown to protect against chemically induced toxicity and carcinogenesis in animal model systems (1-5). A possible mechanism for such protective action is the inhibition of the initial bioactivation of these toxic compounds. Alternative mechanisms such as the induction of glutathione S-transferase and the trapping of reactive electrophiles by DAS have also been suggested (6). Previous work from this laboratory demonstrated that oral treatment of rats with DAS resulted in a time- and dose-dependent decrease of hepatic microsomal cytochrome P-450 2E1 activity, induction of P-450 2B1 and pentoxyresorufin dealkylase activity, and moderate induction of ethoxyresorufindealkylase activity. Treatment of rats with diallyl sulfoxide (DASO) and diallyl sulfone (DAS0.J led to similar modulations in monooxygenase activities, but the decrease of P-450 2E1 activity by the

sulfone occurred more rapidly (5). When added to a microsomal incubation mixture, DAS was a competitive inhibitor of N-nitrosodimethylamine (NDMA) demethylase, an activity displayed by P-450 2E1. In the presence of NADPH, diallyl sulfone caused a metabolism-dependent inactivation of microsomal NDMA demethylase activity, but such inactivation was not appreciable in similar experiments with DAS or DASO. These observations suggest that the inactivation and inhibition of P-450 2E1 are likely mechanisms for the aforementioned protective action, especially against compounds such as 1,2-dimethylhydrazine, NDMA, and CC14, which are known P-450 2E1 substrates (7). The present work further investigated the mode of inhibition of P-450 2E1 by DAS, DASO, and DASOz as well as the metabolic conversion of DAS to DASO and DAS02. p-Nitrophenol (PNP) hydroxylase activity was used to measure the activity of P-450 2E1 in microsomal and reconstituted enzyme systems.

*Address correspondence to this author at the Laboratory for Cancer Research, Department of Chemical Biology and Pharmacognosy, College of Pharmacy, Rutgers UniversityiPiscataway, NJ 08855-0789. Rutgers University. *University of the Pacific. 'University of California at Los Angeles.

Materials. DAS with a purity of 97% was purchased from Aldrich Chemical C O . (Milwaukee, WI). DASOz was Purchased

Materlals and Methods

'

Abbreviations: DAS, diallyl sulfide;DASO,diallyl sulfoxide; DASO1, diallyl sulfone; P-450 2E1, cytochrome P-450 2E1 (other cytochromes P-450 are abbreviated similarly); NDMA, N-nitrosodimethylamine;PNP, p-nitrophenol; GC, gas chromatography; MS, mass spectrometry.

OS93-228~/91/2704-0642$02.50/0 0 1991 American Chemical Society

Inhibition of P-450 2E1 by Diallyl Sulfide from Parish Chemical Co. (Orem, UT) and purified by vacuum distillation and column chromatography on silica gel with CHzClz as the elution solvent. The purity of the DASOz was found to be 98% as determined by integration of the proton nuclear magnetic resonance spectrum and by gas chromatographic (GC) analysis. DASO was synthesized according to a modification of the method of Leonard and Johnson (8) as described previously (5). P N P was obtained from Kodak Chemical Co. (Rochester, NY). Ascorbic acid and catalase were purchased from Sigma Chemical Co. (St.Louis, MO). Benzphetamine hydrochloride was a gift from the Upjohn Co. (Kalamazoo, MI). N-Benzylcyclopropylamine hydrochloride was synthesized by the method of Bumgardner et al. (9). Other reagents were obtained commercially as the reagent grade products, and the sources of other chemicals were indicated previously (5). Microsomal Preparation. Liver microsomes were prepared (10)from male Sprague-Dawley rats (90-100 g) that were pretreated intragastrically with a single dose of acetone, 10 mL of a 50% aqueous solution per kilogram body weight, 20 h prior to killing. The rate of P N P hydroxylase activity of the acetoneinduced microsomes, as measured by the method of Koop (II), was elevated 3.3-fold compared to that of control microsomes. Competitive Inhibition of P N P Hydroxylase. The P N P hydroxylase assay mixture (0.5 mL) contained acetone-induced microsomes (0.1 mg of protein), P N P (0.01-0.05 mM), an NADPH-generating system (0.4 mM NADP+, 10 mM glucose 6-phosphate, 0.2 unit of glucose-6-phosphate dehydrogenase), 1.0 mM ascorbate, and 0.1 M potassium phosphate buffer, pH 6.8. After incubation for 10 min a t 37 "C, the reaction was terminated by the addition of perchloric acid and the catechol concentration determined spectrophotometrically (11). DAS (0.125 mM), DASO (0.25 mM), and DASOz (0.25 mM) were used as the inhibitors. Apparent K , values were determined by nonlinear regression analysis using the mean substrate concentrations (EnzFitter; Elsevier Biosoft, Cambridge, U.K.). Apparent Ki values were where calculated by using the equation Ki = K,[I]/(K', - Km), K , and K',,, were obtained in the absence and the presence of inhibitor, respectively,and [I] is the concentration of the inhibitor. Metabolism-Dependent Inactivation of P-4502E1. In order to examine the time-dependent inactivation of P N P hydroxylase activity, a two-stage incubation procedure was used. The first stage contained 0.1 M potassium phosphate buffer, pH 7.4, acetone-induced microsomes (4 mg of protein/mL), various concentrations of DASOz or other inhibitors added as an aqueous solution, and 1mM NADPH. At various times of the incubation (37 "C), an aliquot of the first-stage incubation mixture was diluted 20-fold into a second-stage incubation (37 "C) containing 0.1 M potassium phosphate buffer, pH 6.8,l mM ascorbic acid, 0.1 mM PNP, and 1 mM NADPH. The P N P hydroxylase assay was carried out as described in the preceding section. The inactivation of P-450 2E1 in a reconstituted system was studied under similar conditions, except that 0.1 nmol of P-450 2E1,2700 units of NADPH-cytochrome P-450 reductase (12),0.1 nmol of cytochrome b5, 40 pg of dilauroylphosphatidylcholine, and 20 units of catalase were used in place of microsomes in the first-stage incubation with a volume of 0.1 mL. An NADPHgenerating system was used in both stages of incubation. Identification of Metabolites of DAS. Three male SpragueDawley rats (approximate body weight 80 g) were treated with DAS (200 mg/kg, PO). At 18 h, liver, blood, and urine samples were collected for the analysis of DAS and metabolites. The livers were combined and homogenized in 30 mL of 1.15% KCl on ice. An aliquot totaling 0.5 volume of ice-cold 7.5% perchloric acid was added with mixing, followed by centrifugation. The supernatant was added to 2 volumes of CH2Clzand 100 mg of NaCl and shaken for 10 min. The blood sample was similarly acidified and extracted. Urine was extracted without acidification. After centrifugation, the organic layers were evaporated to a volume of 200 pL under nitrogen and subjected to GC/MS analysis. A Hewlett-Packard 5890 GC equipped with a flame ionization detector and a fused silica 25 m X 0.2 mm HP-5,5% methylphenyl silicone capillary column was used. Initial temperature was 55 "C for 5 min, and then the temperature was increased to 180 "C a t a rate of 10 "C/min. The temperature was finally increased to 300 "C a t a rate of 20 "C/min. The column flow was 0.7 mL of hydrogen/min. The injection temperature was 190 "C, and

Chem. Res. Toxicol., Vol. 4, No. 6, 1991 643

01 0

I 0.02

0.04

0.06

[PNPI-l

0.08

0.10

0.12

0.14

ph4-l

Figure 1. Lineweaver-Burk plot showing the inhibition of P N P hydroxylaseactivity by DAS02. The incubation mixture contained acetone-induced microsomes (0.10 mg of protein), an NADPHgenerating system, and P N P (0.01-0.05 mM) either with).( or without (a) 0.25 mM DASOz in a final volume of 0.5 mL. The reaction time was 10 min a t 37 "C. Each data point represents the average from duplicate incubations. The units of the rate ( u ) are nmol of product/(min.mg of protein). Table I. Kinetic Parameters of Inhibition of Microsomal P N P Hydroxylase Activity by DAS, DASO, and DAS020 inhibitor, rrM V,. nmol/(min.md KL,rrM K;.UM none 5.5 f 0.4 13.8 f 2.6 DAS (125) 5.1 f 0.1 23.3 f 2.2 188 48 DASO (250) 5.7 f 0.4 24.2 f 0.5 390 f 18 DASOz (250) 5.7 0.6 48.9 f 5.2 118 f 19

*

'The apparent V,, and K6p values were obtained by nonlinear regression analysis. The results are expressed as mean i SD for three experiments. The concentrations of the inhibitors are shown in parentheses. the detector temperature was 250 "C. DAS was eluted a t 7.96 min, DASO a t 14.5 min, and DASOz a t 15.2 min. An internal standard, acetophenone, was eluted at 12.9 min.

Results Competitive Inhibition of PNP Hydroxylase Activity. DAS is a potent competitive inhibitor of NDMA demethylase activity (IO). DAS and its putative metabolites, DASO and DAS02, were found to be competitive inhibitors of microsomal P N P hydroxylase activity, another activity displayed by P-450 2E1. A representative set of results is shown in Figure 1, and the overall results are summarized in Table I. Apparent Ki values of 188, 390, and 118 pM were observed with DAS, DASO, and DAS02,2respectively, indicating DAS02 was a more effective inhibitor than its precursors. Suicide Inhibition of Microsomal P-4502E1. The time-dependent inhibition of microsomal NDMA demethylase activity by DASOz in vitro has been described previously (5). The inhibition microsomal P N P hydroxylase activity by DASO, was found to be time- and NADPH-dependent (Figure 2) and saturable (Figure 3). No inactivation of P-450 2E1 was detectable in the absence of either DAS02 or NADPH. By use of the notation of Main (13) for mechanism-based inactivation, the Ki value for DAS02 was calculated to be 188 f 34 pM and the kz value was 0.32 f 0.08 min-' (mean f SD, n = 3). A slow and slight metabolism-dependent inactivation of P-450 2E1 was also observed in prolonged incubation with DAS (data not shown). The inactivation of the first-stage incubation was inhibited by various concentrations of acetone, an alternate The suicide inhibition of P-450 2E1 by DASOp affected the Ki value observed in this experiment.

Brady et al.

644 Chem. Res. Toxicol., Vol. 4, No. 6,1991

x 0

2.0

0

1

2

3

4

5

6

7

10

0

8

Time (min)

20

Time (min)

Figure 2. Effect of concentration of DASOz and time on the inactivation of microsomal P N P hydroxylase activity. First-stage incubations were conducted as described under Materials and Methods for the times indicated and contained 0 (0) , 2 5 (A),50 (A),100 (o),200 ( 0 )pM DAS02. The units of the rate ( u ) are nmol of product/(min.mg of protein).

-.-

Figure 5. Effect of preincubation with DASOz on spectrally detectable P-450 and P N P hydroxylase activity. First-stage incubations were conducted with 200 pM DASOp as described in Figure 2. At the indicated times, portions of the incubations were and P N P hydroxylase activity ( 0 ) . The assayed for P-450 (0) 100% values were 1.31 nmol of P-450/mg and 7.68 nmol of product/ (min-mg of protein), respectively.

E

E

L

2 2 .-

v N

>

L

60

50

Y

2

40

M

1001

cn 90

Figure 3. Saturability of the rate of inactivation of P N P hydroxylase activity by DAS02 The t l l Zvalues were obtained from the data in Figure 2. Nonlinear regression analysis of plots of 0.693/(t1/2)vs [DAS02]gave values for Ki and maximal tllPwhich were used to draw the straight line.

.c .-

80

$

70 60

2

50

2

40

E

.+ cn

100

.-cC

.

be

.-

E

\

30 0

80

2

4

6

8

10

Q)

LY h c

Time (min)

.-> .-c

2 50 I

0

1

2

3

4

Time (min) Figure 4. Inhibition of the DASOp-dependent inactivation of PNP hydroxylase activity by acetone. Incubations were conducted in the presence of 200 pM DASOz as described in Figure 2, but contained 0 (a),0.2 (A),or 1 (m)mM acetone. The rates a t 0.5 min were 7.13,6.70, and 6.50 nmol of product/(min.mg of protein), respectively. substrate for P-450 2E1 (Figure 4). T h e inclusion of 1m M cysteine in t h e first-stage incubation had n o effect on t h e inactivation (data n o t shown). T h e DAS02-dependent inactivation of PNP hydroxylase was also accompanied by a decrease in t h e reduced P450-CO binding spectrum (Figure 5). When t h e first-stage incubation was allowed t o proceed for u p t o 20 min, t h e remaining P N P hydroxylase activity decreased t o 20% of t h e initial rate, and spectrally detectable P-450 decreased t o 75% of t h e initial value. No measurable effect on t h e

Figure 6. (A) Effect of preincubation with DASOz and time on microsomal pentoxyresorufin dealkylase activity. The first-stage no DAS02 and 1mM NADPH; ( 0 )500 incubation contained (0) pM DASOz and no NADPH; ( 0 ) 100 pM DASOz and 1 mM NADPH; or (A)500 pM DASOz and 1 mM NADPH. The second-stage incubation contained 0.05 M Tris buffer, pH 7.5,0.025 M MgC12, 10 pM pentoxyresorufin, and 4 mM NADPH. The 100% activity value was 1.0 pmol of resorufin/(min.mg). (B) Effect of preincubation with DAS02or N-benzylcyclopropylamhe on microsomal benzphetamine N-demethylase activity. First-stage incubations containing 1mM NADPH and 500 pM DASOp (0) or N-benzylcyclopropylamine ( 0 )were conducted as described in Figure 2. The second-stage incubation (10 min) contained 0.1 M potassium phosphate buffer, pH 7.4, 500 pM benzphetamine, and 1 mM NADPH. The 100% activity value was 9.8 nmol of HCHO/(min.mg). contents of immunodetectable P-450 2 E 1 or P-450 2B1 were seen during t h i s 20-min period (data n o t shown).

Selectivity in the Inactivation of Microsomal Enzymes by DAS02. T h e possible metabolism-dependent inactivation of other microsomal monooxygenase activities by DASOz was also studied. A metabolism-dependent inactivation of pentoxyresorufin dealkylase activity i n acetone-induced microsomes was not observed with 0.1 m M DASOz in t h e first-stage incubation, b u t a 25% in-

Inhibition of P-450 2El by Diallyl Sulfide

Chem. Res. Toxicol., Vol. 4, No. 6, 1991 645

II

z I c 2

v

>

0

,

, 2

,

,

.

4

,

6

,

B

I

,I 0

Time (min)

Figure 7. Time-dependent inactivation of purified P-450 2E1 by DAS02. The first-stage incubations contained NADPH-generating system but no DASOz (u),DAS02 (0.5 mM) but no NADPH-generating system (O),or both NADPH-generating system and DAS02(0).Aliquota (0.02 mL) were removed at the indicated times and added to 0.23 mL of complete PNP hydroxylase mixtures for activity measurement. activation was observed with 0.5 mM DASOz (Figure 6A). Under these conditions, inactivation of microsomal ethoxyresorufin dealkylase activity was not observed with either concentration of DAS02. The pentoxyresorufin dealkylase activity in phenobarbital-induced microsomes was not inactivated by 0.5 mM DASOz under conditions similar to those described for Figure 6A (data not shown). In another set of experiments to determine the selectivity in the action of DAS02, benzphetamine was used as the substrate in the second stage after a first-stage incubation with 0.1 mM DAS02 for 5 or 10 min. DASOz had minimal effect on benzphetamine demethylase activity, but when N-benzylcyclopropylamine(14) replaced the sulfone in the first stage, marked inhibition of the demethylase activity was seen (Figure 6B). Inactivation of P-450 2E1 in a Reconstituted System. Inactivation of purified P-450 2E1 in a reconstitution system by 0.5 mM DASO, was also found to be time- and NADPH-dependent (Figure 7). Without NADPH or DAS02, there was no inhibition of PNP hydroxylase activities. The results indicated that P-450 2E1 can catalyze the activation of DASOz to a reactive species which inactivates the enzyme itself. During an &min incubation with 0.5 mM DAS02, about 56% of the P N P hydroxylase activity associated with P-450 2E1 was inhibited (Figure 7), whereas about 86% of the activity was inhibited in a similar experiment with microsomes (data not shown). Other experiments also indicated that higher concentrations of DAS02 were required to produce the same extent of inhibition in the reconstituted system as in microsomes (data not shown). This disparity may be related to the difference in apparent K , values of P-450 2E1 catalyzed reactions in microsomes and in reconstituted systems. For example, the K, for NDMA demethylase is 15-20 pM in microsomes (15) and 350 pM in the reconstituted system, possibly due to the presence of glycerol and other inhibitors in the latter system (12). Detection of DASO and DASOz as Metabolites of DAS. Liver, blood and urine samples from DAS-pretreated rats (see Materials and Methods) were extracted and analyzed by capillary GC (Figure 8). Peaks corresponding in retention time to standards for DASO and DAS02 were observed in each of these extracts. The identities of the peaks in the liver extracts were verified by mass spectral analysis. No peak for DAS was observed. The liver was found to contain approximately 10% of the initial dose of DAS as the sum of DASO and DAS02. In

0

5

10

15

20

Minutes

Figure 8. Capillary gas chromatographyof CHzClzextracts from the liver of rats pretreated at 18 h with DAS (200 mg/kg). The standard was acetophenone. The peaks were confirmed by mass spectral analysis as DASO (A) and DAS02 (B). Mass spectral analysis was performed on a Hewlett-Packard 5971A GC/MS equipped with a Hewlett-Packard HP-1 fused silica capillary column, cross-linkedmethyl silicon, 12.5 m, 0.2 mm i.d., 0.33-rL film thickness. Spectra were obtained at 70 eV. The mass spectrum of DASO showed major ions at m / z = 41,68, and 81 and gave a molecular ion at m/z = 130. The mass spectrum of DASOzshowed major ions at m/z = 41,54,67, and 105. Comparison of spectra from the incubation products revealed identical fragmentation patterns and identical chromatographicretention times with the standards. preliminary experiments with microsomes and supernatant from the 9OOOg centrifugation of liver homogenate (S-9 fraction), NADPH-dependent formations of DASO from DAS and of DAS02 from DASO were also detected (data not shown).

Discussion The previously described chemoprotective properties of DAS may be largely attributed to the inhibition of P-450 2E1 mediated bioactivation of the challenging toxic agents. In comparison to DAS and DASO, DASOz more rapidly inactivated NDMA demethylase activity after in vivo treatment ( 5 ) . DASOz inactivated P-450 2E1 mediated NDMA demethylase activity in vitro in a time- and NADPH-dependent manner, but it was unknown whether the sulfone was actually produced from the sulfide in vivo. Evidence for this metabolic conversion has been obtained in the present study by using the whole animal, and each of the two oxidative steps have been observed in cell-free incubations. It is likely that the flavin-containing monooxygenase participates in the initial oxidation of DAS (16) and that the formation of sulfone is catalyzed by P-450 (17). However, the roles of these enzymes in the metabolism of DAS require further investigation. By using another representative assay of P-450 2E1, PNP hydroxylase, we herein confirmed the metabolismdependent inactivation of this enzyme by DAS02 and further characterized this process. Our results showed that DASOz displayed characteristics of a mechanism-based irreversible inhibitor: the inhibition was time- and NADPH-dependent, was saturable, exhibited pseudofirst-order kinetics, and coincided with the loss of CO binding spectrum of P-450 2E1. The 25% loss of spectrally detectable P-450 corresponds to most of the P-450 2E1 content in microsomes from acetone-induced rats (12). The residual microsomal PNP hydroxylase activity could be due to P-450 2E1 that was not denatured and other P-450 isozymes. The inactivated protein apparently persisted in the membrane since no change in the immunodetectable content of P-450 2E1 was observed. The

Brady et al.

646 Chem. Res. Toxicol., Vol. 4, No. 6, 1991

N \/\\

1

Diallyl sulfide

0

II

’-

Diallyl sulfoxide

I

0

II

Niw

Diallyl

sulfone

0

Reactitive Intermediate p45Q

suicide inactivation

Figure 9. A proposed scheme for the metabolism of DAS and suicide inactivation of P-450 2E1 by DAS02.

action of DAS02 may be selective toward P-450 2E1 since the inactivation of PNP hydroxylase was inhibited by the alternate substrate acetone, and DAS02 did not inactivate NADPH-P-450 reductase (5),ethoxyresorufin dealkylase, or benzphetamine demethylase activity. Additional studies with other inhibitors and activities will be required to further characterize the specificity. The lack of an effect of cysteine suggested that the inactivation was not due to release of a reactive electrophilic product from the active site of the enzyme. A high affinity of DAS, DASO, and DASOz toward P-450 2E1 was suggested by the low apparent Ki values and is in accord with the affinity of this isozyme toward other low molecular weight organic compounds. While the kinetics of the inactivation are consistent with a suicidal inactivation process, the mechanism of action of DAS02 cannot be determined from the present data. Three mechanisms are possible, which include protein alkylation (18,19),heme alkylation as observed with other allyl-containing substrates (20), or covalent binding of a modified heme product to the protein (21,22). Preliminary results indicate that acidic-acetone-extractable modified heme products, which were separable from the native heme by HPLC, were formed. The structures of these products require detailed characterization. The allyl group is apparently necessary for the inactivation since a structural analogue, dipropyl sulfide, is not chemoprotective against 1,2-dimethylhydrazine (23, 24) and did not inhibit microsomal NDMA demethylase activity after re treatment.^ The oxidation state of the sulfur atom is also important since little or no inactivation by DAS or DASO was detected under the conditions described for Figure 2. It is probable that the rate of conversion of these compounds was too slow t o generate sufficient sulfone for the inactivation under the conditions used. A proposed scheme for the metabolism of DAS and suicide inactivation of P-450 2E1 by DASOz is shown in Figure 9. Another sulfurcontaining compound, spironolactone, inactivates P-450 selectively and also requires oxidation of the sulfur (25, 26). Other inhibitors of P-450 2E1 such as 3-amino-1,2,4triazole, dihydrocapsaicin and disulfiram have been described (27-30),but they may be less selective and more toxic than DASOB. Since DAS is relatively nontoxic, with an LDa of about 3.5 g/kg for rats? it may be expected that J. F. Brady and C. S. Yang, unpublished results.

DAS02, a major metabolite, would also have low toxicity. The rapid in vivo suppressive effects of DAS02, along with its minimal odor as compared to DAS, suggest that it may be useful as a preventive or antidotal agent for minimizing the toxicity of exposure to substrates of P-450 2E1.

Acknowledgment. We thank Ms. C. H. Nguyen for expert technical assistance in the GC/MS analysis and Ms. Dorothy Wong for excellent secretarial assistance in the preparation of the manuscript. This work was supported by NIH Grant ES03938, Grant 88B18 from the American Institute for Cancer Research, and NIEHS Center Grant ES05022. Registry No. DAS, 592-88-1; DASO, 14180-63-3; DAS02, 16841-48-8; P-450, 9035-51-2; PNP hydroxylase, 126341-87-5; ethoxyresorufin dealkylase, 59793-97-4; pentoxyresorufin dealkylase, 9695-04-9; benzphetamine demethylase, 37237-40-4.

References (1) Wargovich, M. J. (1987) Diallyl sulfide, a flavor component of

garlic (Allium satiuum), inhibits dimethylhydrazine-induced colon cancer. Carcinogenesis (London) 8, 487-489. (2) Wargovich, M. J., and Goldberg, M. T. (1985) Diallyl sulfide: a naturally occurring thioether that inhibits carcinogen-induced nuclear damage to colon epithelial cells in vivo. Mutat. Res. 143, 127-129. (3) Wargovich, M. J., Woods, C., Eng. V. W. S., Stephens, L. C., and Gray, K. (1988) Chemoprevention of N-nitrosomethylbenzyl-

amine-induced esophageal cancer in rats by the naturally occurring thioether, diallyl sulfide. Cancer Res. 48, 6872-6875. (4) Wattenberg, L. W., Lam, L. K. T., Fladmoe, A. V., and Borchert, P. (1977) Inhibitors of colon carcinogenesis. Cancer 40, 2432-2435. (5) Brady, J. F., Wang, M.-H., Hong, J.-Y., Xiao, F., Li, Y., Yoo, J.-S. H., Ning, S. M., Fukuto, J. M., Gapac, J. M., and Yang, C. S. (1991) Modulation of rat hepatic microsomal monooxygenase activities and cytotoxicity by diallyl sulfide. Toxicol. Appl. Pharmacol. 108, 342-354. (6) Wattenberg, L. W., Spamins, V. L., and Barany, G. (1989) In-

hibition of N-nitrosodiethylamine carcinogenesis in mice by naturally occurring organosulfur compounds and monoterpenes. Cancer Res. 49, 2689-2692. (7) Yang, C. S., Yoo, J.-S. H., Ishizaki, H., and Hong, J.-Y. (1990) Cytochrome P450IIE1: roles in nitrosamine metabolism and mechanisms of regulation. Drug Metab. Rev. 22, 147-160. (8) Leonard, N. J., and Johnson, C. R. (1962) Periodate oxidation of sulfides to sulfoxides: scope of the reaction. J. Org. Chem. 27, 282-284. (9) Bumgardner, C. L., Lawton, E. L., and Carver, J. G. (1972)

Hydride reduction of N-cyclopropylimines. J. Org. Chem. 37,

407-409. (10) Brady, J. F., Li, D., Ishizaki, H., and Yang, C. S. (1988) Effect of diallyl sulfide on rat liver microsomal nitrosamine metabolism and other monooxygenase activities. Cancer Res. 48,5937-5940. (11) Koop, D. R. (1986) Hydroxylation of p-nitrophenol by rabbit ethanol-inducible cytochrome P-450 isozyme 3a. Mol. Pharmacol. 29, 399-404. (12) Patten, C. J., Ning, S. M., Lu, A. Y. H., and Yang, C. S. (1986) Acetone-inducible cytochrome P-450: purification, catalytic ac-

tivity and interaction with cytochrome bb. Arch. Biochem. Biophys. 251, 629-638.

(13) Main, A. R. (1973) Kinetics of active-site-directed irreversible inhibition. Essays Toricol. 4, 59-105. (14) Macdonald, T. L., Zirvi, K., Burka, L. T., Peyman, P., and Guengerich, F. P. (1987) Mechanism of cytochrome P450 inhibition by cyclopropylamines. J,Am. Chem. SOC.104, 2050-2052. (15) Yoo, J.-S. H., Ishizaki, H., and Yang, C. S. (1990) Roles of cytochrome P450IIE1 in the dealkylation and denitrosation of

N-nitrosodimethylamine and N-nitrosodiethylamine in rat liver microsomes. Carcinogenesis (London) 11, 2239-2243. (16) Ziegler, D. M. (1984) Metabolic oxygenation of organic nitrogen and sulfur compounds. In Drug Metabolism and Drug Toxicity (Mitchell, J. R., and Horning, M. G., Eds.) pp 33-53, Raven Press, New York. H. Sumiyoshi and M. J. Wargovich, personal communication.

Chem. Res. Toxicol. 1991, 4,641-654 (17) Renwick, A. G. (1989) Sulphoxides and sulphones. In Sulphur-Containing Drugs and Related Organic Compounds (Damani, L. A., Ed.) Part B, pp 133-154, Ellis Horwood, Chichester. (18) Halpert, J. R., Balfour, C., Miller, N. E., and Kaminsky, L. S. (1986) Dichloromethyl compounds as mechanism-based inactivators of rat liver cytochromes P-450 in oitro. Mol. Pharmacol. 30, 19-24. (19) Halpert, J. R., Jaw, J.-Y., and Balfour, C. (1989) Specific inactivation by l7b-substituted steroids of rabbit and liver cytochromes P-450 responsible for progesterone 21-hydroxylation. Mol. Pharmacol. 34, 139-147. (20) Ortiz de Montellano, P. R. (1986) Oxygen activation and transfer. In Cytochrome P-450, Structure, Mechanism and Biochemistry (Ortiz de Montellano, P. R., Ed.) pp 217-271, Plenum Publishing, New York. (21) Osawa, Y., and Pohl, L. R. (1989) Covalent bonding of the prosthetic heme to protein: A potential mechanism for the suicide inactivation or activation of hemoproteins. Chem. Res. Toxicol. 2, 131-141. (22) Correia, M. A., Decker, C., Sugiyama, K., Caldera, P., Bornheim, L., Wrighton, S. A., Rettie, A. E., and Trager, W. F. (1987) Degradation of rat hepatic cytochrome P-450 heme by 3,5-dicarbethoxy-2,6-dimethyl-4-ethyl-1,4-dihydropyridine to irreversibly bound protein adducts. Arch. Biochem. Biophys. 258,435-451. (23) Zemaitis, M. A., and Greene, F. E. (1976) Impairment of hepatic microsomal drug metabolism in the rat during daily disulfiram

647

administration. Biochem. Pharmacol. 25, 1355-1360. (24) Goldberg, M. T. (1987) Inhibition of genotoxicity by diallyl sulfide and structural analogues. In Anticarcinogenesis and Radiation Protection (Cerutti, P. A,, Nygaard, 0. F., and Simic, M. G., Eds.) pp 905-912, Plenum Publishing, New York. (25) Decker, C., Sugiyama, K., Underwood, M., and Correia, M. A. (1986) Inactivation of rat hepatic cytochrome P-450 by spironolactone. Biochem. Biophys. Res. Commun. 136, 1162-1169. (26) Decker, C. J., Rashed, M. S., Baillie, T. A., Maltby, D., and Correia, M. A. (1989) Oxidative metabolism of spironolactone: evidence for the involvement of electrophilic thiosteroid species in drug-mediated destruction of rat hepatic cytochrome P450. Biochemistry 28, 5128-5136. (27) Koop, D. R. (1990) Inhibition of ethanol-inducible cytochrome P450IIE1 by 3-amino-1,2,4-triazole. Chem. Res. Toxicol. 3, 377-383. (28) Gannett, P. M., Iversen, P., and Lawson, T. (1990) The mechanism of inhibition of cytochrome P450IIE1 by dihydrocapsaicin. Bioorg. Chem. 18, 185-198. (29) Brady, J. F., Xiao, F., Wang, M.-H., Li, Y., Ning, S. M., Gapac, J. M., and Yang, C. S. (1991) Effects of disulfiram on hepatic P450IIE1, other microsomal enzymes, and hepatotoxicity in rats. Toxicol. Appl. Pharmacol. 108, 366-373. (30) Guengerich, F. P., Kim, D.-H., and Iwasaki, M. (1991) Role of human cytochrome P-450 IIEl in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168-179.

Positional Effects on the Structure and Stability of Abbreviated H-ras DNA Sequences Containing 0 ‘-Methylguanine Residues at Codon 12 Ronald E. Bishop and Robert C. Moschel* Chemistry of Carcinogenesis Laboratory, ABL-Basic Research Program, NCI-Frederick Cancer Research and Development Center, P.O. Box B , Frederick, Maryland 21 702 Received December 26, 1990 Activation of the H-ras protooncogene in rats by methylating carcinogens results from a G-to-A transition mutation a t the second position of codon 1 2 (GGA), presumably due to formation of an @-methylguanine (m6G) a t this position. A similar transition at the fist position of codon 12 appears not to occur in vivo. T o study the possible structural basis for this bias in mutation, we synthesized a series of 11-base H-ras sequences [e.g., 5’-d(CGCTG*G*AGGCG)-3’ and two complementary strands] containing an m6G a t the first, second, or both positions of codon 12 (i.e., G* = m6G). The results of solution chemical studies indicated that the individual strands formed stable hairpin structures among which that containing m6G a t the second position of codon 1 2 was most stable. Further, the DNA duplex with m6G a t the second position was significantly more stable than that with m6G a t the first position, and under certain conditions, it was more stable than the unmodified duplex as well. I t is possible that such a difference in stability might lead to more ready recognition of an m6G a t the first position by repair proteins, and this could contribute to the apparent site specificity of mutation by methylating carcinogens a t codon 1 2 of the H-ras gene,

Introduction The production of 06-methylguanine residues by methylating carcinogens at codon 12 (GGA) of the rat H-ras gene is presumed responsible for inducing the G-to-A transition mutations that convert the normal gene to a highly transforming oncogene (1-4). Interestingly, only the second guanine of codon 12 is mutated. It is unclear if this results from a preferential reaction of the second guanine with the methylating carcinogen or if repair of an Os-methylguanineresidue at the first position of codon 12 is more efficient than repair at the second position. Knowledge of the DNA structural perturbations produced by Os-methylguanine (m6G) residues (Figure 1) should further our understanding of how these adducts are rec-

ognized by repair proteins specific for this damage (5-7) as well as how these modified guanines code for thymine incorporation during DNA replication (8-14). To model the possible effects of 06-methylguanine substitution on the structure of the H-ras gene, we have synthesized four undecamer (11-mer) oligodeoxyribonucleotides (oligonucleotides) of type 1 (Figure l ) , which are abbreviated versions of the H-ras sequence encompassing codon 12 (GGA) and spanning the third base of codon 10 through the first base of codon 14. We have compared the solution properties of the unmodified single-strand oligonucleotide 1 with those of the analogues containing an 06-methylguanine residue at the first or second position, or both the first and second positions, of the modeled codon 12 [i.e.,

0893-228~/91/2704-0647$02.50/0 0 1991 American Chemical Society