[CANCER RESEARCH 41, 4391-4398, November 1981]0008-5472/81 /0041-OOOOS02.00

Roles of 2-Haloethylene Oxides and 2-Haloacetaldehydes Derived from

Vinyl Bromide and Vinyl Chloride in Irreversible Binding toProtein and DMA1

F. Peter Guengerich,2 Patricia S. Mason, William T. Stott, Tony R. Fox, and Philip G. Watanabe

Department of Biochemistry and Center in Environmental Toxicology, Vanderbilt University School of Medicine, Nashville, Tennessee 37232 [F. P. G., P. S. M.¡,and Toxicology Laboratory, Dow Chemical Company, Midland, Michigan 48640 ¡W.T. S., T. R. F., P. G. W.]

ABSTRACT

The metabolism of [1,2-14C]vinyl bromide (VBR) to products

irreversibly bound to DNA and protein was examined in rat livermicrosomes, reconstituted cytochrome P-450 systems, andisolated hepatocytes. A role for cytochrome P-450 was con

firmed using inhibition and reconstitution experiments. Themajor form of cytochrome P-450 involved in VBR metabolism

does not appear to be either of the major isozymes induced byphénobarbital or ß-naphthoflavone, as determined by induction, reconstitution, and antibody inhibition studies. 2-Brom-oethylene oxide and 2-bromoacetaldehyde, suspected metab

olites of VBR, were synthesized and found to be substrates forrat liver epoxide hydrolase and equine liver alcohol dehydro-

genase, respectively. These enzymes were used to probe theroles of the two possible metabolites in the irreversible bindingof products of VBR to protein and DNA. Alcohol dehydrogenasewas more effective than epoxide hydrolase in inhibiting thebinding of VBR metabolites to protein in microsomal incubations. Epoxide hydrolase was effective in inhibiting the bindingof VBR or vinyl chloride metabolites to calf thymus DNA addedto such systems, but alcohol dehydrogenase was not. Similarresults were obtained for binding of VBR metabolites to DNA ina reconstituted enzyme system. Reduced glutathione blockednonenzymatic binding of 2-bromo[1,2-14C]acetaldehyde to pro

tein but not DNA. Binding of vinyl chloride and VBR metabolitesto protein was blocked by reduced glutathione, but binding toDNA was not. These results are consistent with the view that 2-

haloethylene oxides are the major alkylating agents bound toDNA, and 2-haloacetaldehydes are the major alkylating agents

bound to protein in these experimental systems. Studies withlabeled 2-bromoacetaldehyde indicate that the slow kinetics of

DNA binding by this compound is responsible in part for thisphenomenon. Studies with isolated rat hepatocytes suggestthat a significant portion of the total and reactive metabolitesare able to leave these cells. In these systems, binding ofmetabolites of vinyl chloride to DNA outside the hepatocytescould be partially blocked by epoxide hydrolase or by alcoholdehydrogenase, implying that, as targets farther away fromsources of reactive species are considered, the stabilities ofthese species become more important for reaction with nucleo-

philic sites.

' Supported in part by Grants ES 02205, ES 00267, and ES 01590 from the

National Institute of Environmental Health Sciences.2 Recipient of Research Career Development Award ES 00041 from the

National Institute of Environmental Health Sciences. To whom requests forreprints should be addressed.

Received April 20, 1981 ; accepted August 6, 1981.

INTRODUCTION

VC3 exposure has been shown to produce liver tumors in

experimental animals (24) and has been linked to human hepatic hemangiosarcoma in epidemiological studies (7, 9, 21).The heavy industrial use of this monomer, currently about 4x 109 kg/year in the United States (2), justifies studies on the

mechanism by which tumors are induced. VBR, used to alesser degree in industry, also appears to be carcinogenic inanimals (1).

Metabolic activation of VC is required for irreversible bindingof radioactive labels of VC to tissue macromolecules (17, 20,28) and for mutagenesis, at least in some bacterial strains (23).The most commonly postulated reactive metabolite has been2-chloroethylene oxide (4, 5, 14, 25, 28, 36). This epoxiderearranges to 2-chloroacetaldehyde, a reactive a-halocarbonyl

compound which can also react with tissue nucleophiles andlead to mutation of bacteria (14, 25). VBR has been studiedless than VC but has also been shown to require metabolicactivation to generate reactive compounds (4, 6, 29, 30).

Previous studies in our laboratories indicated that 2-chlo

roacetaldehyde was the metabolite of VC involved to the greatest extent in irreversible binding to microsomal protein underin vitro conditions when purified dehydrogenases and epoxidehydrolase were utilized to study the roles of these transientintermediates (14). In this report, we have extended this workto studies with VBR and utilized both DNA and protein astargets for irreversible binding of metabolites. The ability ofreactive intermediates to leave hepatocytes was also examinedin light of the fact that hepatic hemangiosarcoma is a tumor ofendothelial cell origin as opposed to hepatocyte origin.

MATERIALS AND METHODS

Chemicals and Radiochemicals. [1,2-"C|VC was prepared as described elsewhere (32). [1,2-MC]VBR was prepared in a similar manner: 2 jil of 1,2-dibromo[1,2-'4C]ethane (specific activity, 14.6 mCi/

mmol; Amersham/Searle Corp., Arlington Heights, III.) were injectedonto a GC column (glass, 2.5-m x 3-mm outer diameter). The first 9 cmof the column were filled with glass beads (100-,»m diameter) andmaintained at 450° with the use of heating tape. The latter 2.4 m of the

column were packed with 5% DC-4 on 80 to 90 mesh Chromosorb Wand maintained at 25°.The injection port temperature was 200°, and

the carrier gas helium was used at a flow rate of 20 ml/min. The majorproducts detected using a thermal conductivity monitor were acetylene,VBR, and unreacted 1,2-dibromoethane with retention times of 1, 2,

and 15 min, respectively. The VBR was collected in a glass syringe

3 The abbreviations used are: VC. vinyl chloride; VBR, vinyl bromide; GC. gas

chromatography; GSH. reduced glutathione.

NOVEMBER 1981 4391

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

F. P. Guengerich et al.

and stored in polyethylene bags. VBR concentrations were determinedby GC using 2-m columns of Porapak QS (150°). [1,2-14C]VBR was

diluted with carrier VBR to give specific activities of 0.5 to 2 mCi/mmol. Radiochemical purity of the synthesized [1,2-'4C]VBR was de

termined by mass spectrometry-radio-GC analysis to be >97%, basedupon recovery, and [14C]acetylene was not present.

2-Bromoethylene oxide was prepared using the general proceduredescribed by Walling and Fredricks (33). fert-Butylhypobromite was

prepared by the method of Walling and Padwa (34); 4 ml of the distilledmaterial were mixed with 50 ml of ethylene oxide and illuminated (125-watt flood lamp at a distance of 15 cm) for 60 min at 0°under an N2

atmosphere. During this time, the deep-brown solution became color

less. The volume was reduced to about 4 ml under a stream of N2 at20°.Analysis using p-nitrobenzylpyridine reagent (4,14) indicated that

about 10% of the solution consisted of 2-bromoethylene oxide. (The

characteristic purple color formed without the need for heating. It hasbeen demonstrated previously that the epoxide derivatives of VC (4,14) and 1,1,2-trichloroethylene (5) react with the reagent under such

conditions but that hypohalites, ethylene oxide, and rearrangement orhydrolysis products of the epoxides do not. Subsequent spectral measurements indicated that the remaining material consisted of acetone,ferf-butyl alcohol, and 2-propanol.) Mass spectrometry of the materialusing a reservoir leak (25°; electron impact, 70 eV; Finnigan 3200

instrument) showed the following peaks at the indicated m/e values(relative intensities and postulated fragments in parentheses): 124 (31,M+); 122 (32, M +); 96 (100, CH3Br+); 94 (100, CH3Br+); 81 (16, Br+);and 79(16, Br+).

2-Bromoacetaldehyde was prepared by hydrolysis of 2-bromoac-

etaldehyde diethyl acetal (Aldrich Chemical Co., Milwaukee, Wis.) inthe manner described by Secrist era/. (31) for 2-chloroacetaldehyde.

2-Bromo[1,2-14C]acetaldehyde was prepared by bromination of [1,2-4C]acetaldehyde (paraldehyde; New England Nuclear, Boston, Mass.)

as described by Levene (22). The paraldehyde (0.2 mmol, 1 mCi) wasdissolved in 1 ml of 0.05 N HjSCu and mixed with 0.2 mmol of Br2. Thereaction mixture became colorless after 40 min at 70°.The pH of the

solution was adjusted to 4.5 with 6 N KOH, and the product wasrecovered by distillation in quantitative yield. More than 95% of theradioactivity eluted in a peak with the same retention time as didunlabeled 2-bromoacetaldehyde when an aliquot of the radioactivematerial was analyzed by GC using a 2-m column of Porapak QS(180°), and residual [14C]acetaldehyde was not detected (<0.1%).Small aliquots of the aqueous solution of 2-bromo[l ,2-14C]acetalde-hyde were stored under N2 at -20°.

Enzymes and Enzyme Preparations. Male rats of Sprague-Dawley

origin (200 to 250 g) were obtained from HaríanIndustries, Indianapolis, Ind. Treatments with phénobarbital, 3-methylcholanthrene, and/}-naphthoflavone (5,6-benzoflavone) were as described elsewhere(15). Rats were treated with 2-acetylaminofluorene in the manner

described by Cameron et al. (8). Subcellular fractions were preparedby differential centrifugation (11) and stored at —¿�70°in mw Tris-

acetate (pH 7.4) buffer containing 1 ITIM EDTA, 20% (v/v) glycerol,and 0.4 mM phenylmethylsulfonyl fluoride.

Cytochrome P-450, NADPH-cytochrome P-450 reductase, and

epoxide hydrolase were purified using procedures described elsewhere (12, 15, 16). The "B2" cytochrome P-450 fraction (15) and the"A" epoxide hydrolase fraction (16) derived from phenobarbital-

treated rats were used in these experiments. All 3 enzymes wereapparently homogeneous as judged by sodium dodecyl sulfaterpoly-

acrylamide gel electrophoresis, and other analytical data were typicalof previous preparations (12, 15, 16).

Horse liver alcohol dehydrogenase and yeast aldehyde dehydrogen-

ase were purchased from Boehringer Mannheim (Indianapolis, Ind.),dialyzed 3 times versus 100 volumes of 50 mM potassium phosphatebuffer (pH 8.0) at 4°for 12 hr, and centrifuged for 10 min at 104 x g

to remove inactive material.Epoxide hydrolase was inactivated by treatment with excess p-

nitrophenacyl bromide for 60 min at 37° (10). Residual reagent was

removed by gel filtration using a Sephadex G-25 column. The level ofresidual styrene-7,8-oxide hydrolase activity was <3%. Alcohol dehydrogenase was inactivated by heating for 20 min at 100°; the level of

residual activity was <1%.Rat liver hepatocytes were prepared from untreated male Fischer

344 rats (250 g) using collagenase perfusion as described elsewhere(35). Cells were suspended in 0.5% (w/v) fetal calf serum in WilliamsMedium E (Grand Island Biological Co., Grand Island, N. Y.) and usedimmediately in incubations.

Incubation Conditions. Typical microsomal incubations containedin a final liquid volume of 1.0 ml: 1 mg microsomal protein per ml; 0.1M potassium phosphate buffer (pH 7.7); 1.0 ID glucose-6-phosphatedehydrogenase per ml; 0.5 mM NADP*; 10 mw glucose 6-phosphate;

and, when appropriate, 2 mg calf thymus DNA per ml. The head space(2.5 ml) contained VBR (10" ppm) in air, and the gas phase was

routinely checked to monitor VBR. At this level of VBR in the gasphase, the equilibrium concentration measured by GC in the aqueousphase under these conditions was 14 J^M.Vials were sealed with Teflonliners and incubated at 37°with shaking for 30 min, after which the

vials were chilled on ice and swept with N2 to remove most of theresidual substrate. The incubates were washed 3 times with 5 volumesof hexane to remove residual substrate. (2-Bromoacetaldehyde, 2-

bromoethanol, and glycolaldehyde were not extracted into hexaneunder such conditions.) Metabolites irreversibly bound to protein (11)and DNA (19) were determined as described elsewhere. Experimentsestablished that production of non-hexane-extractable and DNA-bound

metabolites was proportional to protein concentration and linear withtime under such conditions, and metabolism was saturated at this VBRconcentration.

RESULTS

Incubation and Inhibition of VBR Metabolism in Microsomal Systems. Metabolism of VBR to hexane-insoluble andDNA-bound products was almost completely dependent upon

NADPH, although the NADPH requirement could be partiallyfulfilled by NADH (Table 1, Experiment A). Other experiments(not shown) did not show an additive effect of the 2 reducedpyridine nucleotides. The values obtained in the absence ofNADPH were viewed as adequate controls and subtracted insubsequent experiments. Metabolism was essentially abolished when CO was added to the atmosphere. VBR metabolismwas strongly inhibited by diethylaminoethyl-2,2-diphenylvaler-ate but not by metyrapone or by antibody raised to the majorform of cytochrome P-450 isolated from phenobarbital-treatedrats. The epoxide hydrolase inhibitor 3,3,3-trichloropropyleneoxide did not affect the metabolism or binding of VBR.

Metabolism and binding were not significantly enhanced bypretreatment of rats with phénobarbital (Table 1, ExperimentB). Pretreatment of the rats with /?-naphthoflavone led to anincrease of about 50% in both hexane-insoluble and DNA-bound metabolites. A similar increase in hexane-insoluble butnot DNA-bound metabolites was observed after rats weretreated with 3-methylcholanthrene, Aroclor 1254, or 2-acetyl

aminofluorene. However, enzyme activity was not significantlyinhibited by an antibody raised to the major form of cytochromeP-450 isolated from /î-naphthoflavone-treated rats (12).

When the metabolism of VC and VBR were compared underconditions of saturating substrate concentrations with the sameenzyme preparations, the metabolism of VBR was found toproceed at a rate about 50% faster than in the case of VC withmicrosomes prepared from phenobarbital-treated rats (Table2). The ratios of DNA-bound to hexane-insoluble metabolites

4392 CANCER RESEARCH VOL. 41

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Activation of Vinyl Halides

were similar for VC and VBR, but the ratio of protein-bound tohexane-insoluble metabolites was higher in the case of VBR

than VC. The addition of GSH produced a stimulation of therate of hexane-insoluble metabolite production in this study;

the magnitude of this stimulation was somewhat variable. Theratio of DMA-bound to hexane-insoluble metabolites was not

affected by GSH, but GSH blocked 50 to 90% of the irreversiblebinding of label from VBR or VC to protein.

Metabolism of 2-Bromoethylene Oxide and 2-Bromoac-

etaldehyde by Epoxide Hydrolase and Dehydrogenases.2-Bromoethylene oxide had a fi/z of about 45 sec at 23° in

Tris-HCI buffer at pH 8.7 (Chart 1). The f1/2 was not affected

by albumin but was reduced by purified epoxide hydrolase.Kinetic parameters were determined for the destruction of 2-bromo-ethylene oxide and 2-bromoacetaldehyde by the en

zymes to be used in inhibition experiments (Table 3). Alcoholdehydrogenase was found to be capable of blocking the irreversible binding of 2-bromo[1,214C]acetaldehyde to DNA or

microsomal protein (Chart 2). Aldehyde dehydrogenase was aseffective as alcohol dehydrogenase in blocking protein bindingbut only blocked 60% of the binding of bromoacetaldehyde toDNA.

Table 1Effects of inducers and inhibitors on microsomal metabolism of VBR

General incubation conditions are described under "Materials and Methods."

Metabolites (nmol/mg protein/30 min)

Experiment IncubationconditionsA

Complete microsomalsystem,phenobarbital-treatedratsMinusNADPHMinus

NADPH, plus NADH(1mw)With

C0:02 atmosphere(4:1,v/v)Plus

anti-phenobarbital ratcy-tochromeP-450 (12mg)Plus

metyrapone (20/IM)Plusdiethylaminoethyl-2,2-di-phenylvalerate

(1HIM)Plus3,3,3-trichloropropyleneoxide

(1HIM)B6

Complete microsomalsystem.untreated

ratsPhenobarbital-treatedratsy8-Naphthoflavone-treatedrats3-Methylcholanthrene-treatedratsAroclor

1254-treatedrats2-Acetylaminofluorene-treated

ratsHexane

insoluble31.9

±2.5a0.8

±0.17.5±0.50.6

±0.124.5

±2.629.5

±1.05.0±1.724.0

±3.128.9

±2.738.0

±5.749.6±4.646.9±2.547.6

±2.448.4±3.8Irreversiblybound

toDNA2.20

±0.350.11

±0.030.48±0.040.08

±0.021

.95 ±0.651.93

±0.510.34±0.052.02

±0.182.39

±0.182.06

±0.323.73±0.231.96±0.541.49

±0.381.89 ±0.69a

Mean ±S.D.

as described under "Materials and Methods."

Kinetics of Binding of 2-Bromoethylene Oxide and 2-Bro-moacetaldehyde. The nonenzymatic, irreversible binding of 2-bromo[1,2-14C]aldehyde to microsomal protein occurred in a

biphasic manner. The first phase had a ti/2 of about 1 min at37° and was complete within 10 min; approximately 50% of

the available 2-bromoacetaldehyde was bound in this time.

After 6 hr, an additional 30 to 40% of the compound wasbound. Binding to DNA also occurred in a biphasic manner.The first phase had a r1/2 of about 1 min, but only 2% of thelabel was bound during this time. About 120 hr were requiredfor the reaction to reach 90% of completion. These data are

0.75

0.50

0.25 •¿�

EPOXIDE HYDROLASE, MG ML"'

Chart 1. Effect of epoxide hydrolase on the decomposition of 2-bromoeth-ylene oxide. All incubations were carried out in Tris-HCI buffer (pH 8.7) at 23°.2-Bromoethylene oxide (0.8 mM) was added to each incubation, and aliquotswere withdrawn at the indicated times and analyzed for residual oxide immediately (4, 14). The A56oat zero time was 1.17. The incubation devoid of protein(•)and an incubation containing 2 mg bovine serum albumin per ml (O) weredone in triplicate, and results are expressed as means ±S.D. for each time point.Other incubations contained epoxide hydrolase at concentrations of 0.18 mg/ml(A). 0.36 mg/ml (•),0.72 mg/ml O, 1.44 mg/ml (A), and 3.3 mg/ml (C). Dataobtained were from single determinations. Inset, apparent («>of 2-bromoethyleneoxide in the presence of varying concentrations of epoxide hydrolase.

Table 2Comparison of microsomal metabolism of VBRand VC in the presence and absence of GSH

Phenobarbital-treated rats served as the source of enzyme fractions. The incubation time was 30 min.

Metabolites (nmol/mg protein)

Hexane insoluble Irreversibly bound to DNA Irreversibly bound to proteinIncubation con

ditions VBR VC VBR VC VBR VCComplete system 15.3 ±0.6a 9.6 ±0.5 1.02 ±0.10 0.70 ±0.05 9.91 ±0.38 3.36 ±0.21

Plus GSH (5 mM) 28.7 ±2.8 15.9 ±1.0 2.28 ±0.48 1.06 ±0.48 0.57 ±0.22 1.48 ±0.03" Mean ±S.D.

NOVEMBER 1981 4393

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

F. P. Guengerich et al.

Table 3

Kinetic parameters for metabolism of suspected VBR products by purified enzymesRates of 2-bromoethylene oxide hydrolase activity were estimated by quantitation of substrate remaining 30 sec after initiation of reactions

in 50 rriM Tris-HCI (pH 8.0) at 23°.The epoxide hydrolase concentration was 0.22 mg/ml, and the concentration of 2-bromoethylene oxidevaried from 0.12 to 0.72 mM. Rates of 2-bromoacetaldehyde reduction were estimated by observing NADH oxidation at 340 nm (37°).

Incubations included 1.0 /ig alcohol dehydrogenase per ml, 0.1 M potassium phosphate buffer (pH 7.7). 0.2 mM NADH, and 0.01 to 0.25 mM2-bromoacetaldehyde. Rates of 2-bromoacetaldehyde oxidation were estimated by observing NADH oxidation at 340 nm (37°). Incubationsincluded 10 fig aldehyde dehydrogenase per ml, 0.1 M potassium phosphate buffer (pH 7.7), 0.2 mM NAD*, and 0.01 to 0.25 mM 2-bromoacetaldehyde. the parameters were calculated from Hanes' plots (S/v versus S) using linear regression analysis.

Substrate EnzymeKm

(mM) (/unol/min/mg) (/imol/min/mg/mM)

2-Bromoethylene oxide2-Bromoacetaldehyde NADH2-Bromoacetaldehyde, NAD4

Epoxide hydrolaseAlcohol dehydrogenaseAldehyde dehydrogenase

0.150.220.13

1.2815.60.95

8.570.9

7.3

2002.8 -

DEHYDROGENASE, MG ML-1

Chart 2. Inhibition of 2-bromo[1,2-'4C]acetaldehyde binding to DNA and protein by alcohol dehydrogenase. In A. 16 nmol 2-bromo[1,2-'*C]acetaldehyde

were incubated with 2 mg calf thymus DNA, 1 mg rat liver microsomal protein,0.1 mmol potassium phosphate (pH 7.7), the indicated amount of dehydrogenase,and 1 fimol NADH in a total volume of 1.0 ml for 2 hr at 37°. DNA adducts (i.e..

metabolites irreversibly bound to DNA) were determined as described elsewhere(19). In B, 40 nmol 2-bromo{1,2-"C]acetaldehyde were incubated with 4 mg rat

liver microsomal protein, 0.1 mmol potassium phosphate (pH 7.7), the indicatedamount of dehydrogenase, and 1 ,»molNADH in a total volume of 2.0 ml for 2 hrat 37°. Metabolites irreversibly bound to protein were determined as described

elsewhere (11). •¿�.incubations with active dehydrogenase; O, incubations withboiled dehydrogenase. Results are presented as means of duplicate incubations.Bars, S.D.

qualitatively similar to those presented by Banerjee ef a/. (3),although the total extent of alkylation was greater in this study,and the previous report did not examine binding at intervalsshort enough to segregate the initial rapid phases.

The reactions of 2-haloethylene oxides and 2-haloacetalde-hydes with adenosine to form 1,6N-ethenoadenosine (4, 17,

30) were also studied as models for DNA binding. With boththe bromo and chloro compounds, initial formation of thefluorescent base occurred rapidly with the haloethylene oxides.As the oxiranes rearrange to yield the haloacetaldehydes (14),

60 80TIME, HOURS

I20

Chart 3. Kinetics of 1 .''N-ethenoadenosme formation from adenosine and 2-

haloethylene oxides or 2-haloacetaldehydes. In each case, 10 /imol of adenosinewere incubated in 1.0 ml of 0.1 M potassium phosphate buffer (pH 7.7) at 37°in

Teflon-sealed vials (total volume, 3.5 ml) with 5 (imol of 2-chloroethylene oxide(•)2-chloroacetaldehyde O, 2-bromoethylene oxide (•),or 2-bromoacetaldehyde (O). At various times, 1,6N-ethenoadenosine formation was assayed fluo-

rimetrically (duplicate readings, excitation at 306 nm. emission at 409 nm) (23).

similar levels of 1,6N-ethenoadenosine were formed with both

types of compounds after a prolonged period (Chart 3).Effects of Epoxide Hydrolase and Alcohol Dehydrogenase

on Irreversible Binding of VBR Metabolites to Protein. Neitherepoxide hydrolase nor alcohol dehydrogenase affected thelevel of hexane-insoluble VBR metabolites produced in microsomal incubations. Epoxide hydrolase was capable of blockinga maximum of about 35% of the irreversible binding to protein(Chart 4). Alcohol dehydrogenase, in the presence of NADH,blocked as much as 75% of the irreversible binding of VBRmetabolites. In both cases, no inhibition was observed withinactivated enzymes.

Effects of Epoxide Hydrolase and Alcohol Dehydrogenaseon Irreversible Binding of VBR and VC Metabolites to DNA.Epoxide hydrolase was highly effective in reducing irreversiblebinding of VBR metabolites to DNA in rat liver microsomalsystems (Chart 5). However, this effect was not completelyspecific, as the inactivated enzyme also lowered DNA binding.The decrease due to actual hydrolase activity of the enzyme(i.e., difference between the 2 curves) was significant, and, atlow levels of the hydrolase, the binding was reduced to lessthan 50% of the level observed with the inactive enzyme. Onthe other hand, alcohol dehydrogenase (active or inactive) didnot inhibit irreversible binding of VBR metabolites to DNA.

Similar results were obtained when irreversible binding of VCmetabolites to DNA was studied in a microsomal system (Chart

4394 CANCER RESEARCH VOL. 41

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Activation of Vinyl Halides

50

40

30

20

10

TOTALMETABOLITES

ADDUCTS

TOTALMETABOLITES

AODUCTS /INACTIVE]( ENZYME/

ADDUCTS

0 0.2 0.4 0.6 0.8 1.0

EPOXIDE HYDROLASE , MG

0 0.2 0.4 0.6 0.8 1.0

ALCOHOL DEHYDROGENASE,MG

Chart 4. Effects of epoxide hydrolase and alcohol dehydrogenase on metabolism of [1.2- aC|VBR to metabolites irreversibly bound to protein by rat liver

microsomes. In A, varying amounts of epoxide hydrolase (•,•¿�)or inactiveepoxide hydrolase (O) were present. In B, 1 RIMNADH and the indicated amountsof alcohol dehydrogenase (•,•¿�)or inactivated alcohol dehydrogenase (O) werepresent. Adducts are metabolites irreversibly bound to protein, determined asdescribed elsewhere (11). Points, means of duplicate determinations.

0.2. 0.4 0.6 1.0EPOXIDE HYDROLASE

ORALCOHOL DEHYDROGENASE, MGML

Chart 5. Effects of epoxide hydrolase and alcohol dehydrogenase on themetabolism of [1,2-14C]VBR by rat liver microsomes to products irreversibly

bound to DNA. Varying amounts of epoxide hydrolase (•.•¿�),inactive epoxidehydrolase (O), or alcohol dehydrogenase (A) were present. Results are expressedas means ±S.D. of triplicate determinations when variations are noted.

6) or when binding of VBR metabolites to DNA was studied ina reconstituted cytochrome P-450 system (Chart 7). The binding of VBR metabolites to DNA (in a microsomal system) wasnot blocked by aldehyde dehydrogenase and NAD* (data not

shown).VC Metabolism in Hepatocytes. The cells used in these

experiments readily metabolized VC in the absence of added

cofactors (Table 4). When calf thymus DNA was placed in themedium, a considerable portion of the DNA adducts formedoutside of the cells, suggesting that reactive intermediatesderived from VC can penetrate the cell membrane. Furtherexperiments showed that binding to the exogenous DNA couldbe significantly inhibited by epoxide hydrolase or by alcoholdehydrogenase in the presence of NADH.

DISCUSSION

Rat liver microsomes were found to metabolize VBR to hex-ane-insoluble metabolites and products which became irre-

1.2

. 0.8

0.4

0 ,.0.2 0.4 06 0.8 LOEPOXIDE HYDROLASE OR ALCOHOL DEHYDROGENASE, MG ML"1

Chart 6. Effects of epoxide hydrolase and alcohol dehydrogenase on themetabolism of [1,2-'"C]VC to products irreversibly bound to DNA by rat livermicrosomes. [1,2-'4C]VC replaced [1,2-"C]VBR at the same concentration, and

the incubation time was 45 min. Varying amounts of epoxide hydrolase (•).inactive epoxide hydrolase (O), or alcohol dehydrogenase (A, with 1 HIM NADH)were present. Points, means of duplicate incubations. Inactive alcohol dehydrogenase was also without effect on the level of binding (data not shown).

0.8

O2 O4 0.6 03 1.0EPOXIDE HYDROLASE OR ALCOHOL DEHYDROGENASE, MG ML'1

Chart 7. Effects of epoxide hydrolase and alcohol dehydrogenase on themetabolism of [1,2-"C]VBR to products irreversibly bound to DNA by a recon

stituted cytochrome P-450 system. Incubations were carried out using thegeneral method, with the exception that the liver microsomes were replaced withcytochrome P-450 (2 ¡IM)and NADPH-cytochrome P-450 reductase (2 UM)purified from phenobarbital-treated rats, 45 /IM L-a-dilauroylglyceryl-3-phospho-rylcholine, and 0.01 /<g catatase per ml. Varying amounts of epoxide hydrolase(•),Inactive epoxide hydrolase (O), or alcohol dehydrogenase (A; 2 separateexperiments which included 1 mM NADH) were present. Results are expressedas means ±S.D. of triplicate determinations in the case of epoxide hydrolaseand as means of duplicate determinations in the case of alcohol dehydrogenase.No inhibition of binding was observed in the presence of inactive alcohol dehydrogenase (data not shown).

NOVEMBER 1981 4395

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

F. P. Guengerich et al.

Table 4

Metabolism of VC by isolated hepatocytesHepatocyes were isolated from male Fisher 344 rats as described under "Materials and Methods," without the

addition of antibiotics and antimycotics. After the collagenase perfusion, the hepatocytes were suspended in WilliamsMedium E plus 0.5% fetal calf serum and buffered with 50 mM N-hydroxyethyl-N'-2-piperazineethanesulfonate (pH7.4). Incubations contained 1.7 x 10°viable cells/ml, with the indicated additions as in the table plus 2 mg calf thymusDNA per ml. The head space in the vials was made up to 10* ppm [1,2-'4C]VC (specific activity, 1.22 mCi/mmol). Cellswere incubated at 37°for 30 min, with occasional gentle mixing by hand. The cells were precipitated by centrifugationat 2 x 103 xg for 10 min, and the supernatants were analyzed for metabolites.

Metabolites

Hexane insoluble (nmol) Irreversibly bound to ONA (pmol)

Experiment Addition to medium Cells Media Cells Media

Epoxlde hydrolase (0.2mg/ml)

Inactive epoxide hydrolase(0.2 mg/ml)

Alcohol dehydrogenase(0.2 mg/ml) plus 1 mMNADH

Inactive alcohol dehydrogenase (0.2 mg/ml) plus1 mM NADPH

1.75 ±0.30* 20.8*2.9 2.02 ±0.16 12.2 ±0.7

1.90 ±0.10 17.3 ±2.5 2.03 ±0.51 16.2 ±1.6

2.00 ±0.11 26.4 ±1.6 2.65 ±0.61 18.4 ±1.3

1.59 ±0.27 27.6 ±4.1 2.45 ±0.64 29.1 ±4.1

8 Mean ±S.D.

versibly bound to protein and DMA. As much as 50% of themetabolites became bound to protein in such experiments(Table 2). The incubation system differed somewhat from thatused by Bolt era/. (6), who reported lower rates of metabolism.The role of cytochrome P-450 was implicated in inhibitionexperiments with classic inhibitors (CO, diethylaminoethyl-2,2-

diphenylvalerate) and by the observation that a purified cytochrome P-450 preparation was capable of metabolizing VBR

in a reconstituted system. However, the rate of VBR metabolismby this form of cytochrome P-450 was lower than expected on

the basis of the microsomal turnover number, and an antibodyraised to this isozyme did not inhibit metabolism (Table 1). Theextent of induction of metabolism of VBR by most compoundstested was slight (Table 1; cf. Réf.15), and antibody raised tothe major form of cytochrome P-450 isolated from /î-naphtho-flavone-treated rats was not inhibitory. Thus, the major form(s)of cytochrome P-450 involved in VBR metabolism does not

appear to be one of the major isozymes induced by phénobarbital or /f-naphthoflavone.

2-Bromoethylene oxide and 2-bromoacetaldehyde are pos

sible reactive metabolites that could be derived from VBR. In atime frame of min, synthetic 2-bromoacetaldehyde was found

to bind extensively to protein, but DNA binding was minimal.GSH blocked the binding to protein but not DNA. 2-Bromo

ethylene oxide was bound more rapidly to adenosine (to form1,6N-ethenoadenosine) than was 2-bromoacetaldehyde. Alco

hol dehydrogenase was found to block binding of 2-bromoace

taldehyde to DNA or protein in the presence of NADH. Sincethe binding of label from [1,2-14C]VBR to protein was selectively

blocked by alcohol dehydrogenase in microsomal systems andsince binding to DNA was selectively inhibited by epoxidehydrolase, we conclude that 2-bromoethylene oxide is responsible for the bulk of DNA binding and that 2-bromoacetaldehyde

is responsible for the bulk of protein binding under the conditions used in the microsomal incubations (Chart 8). This viewis consistent with the results of the kinetic experiments involving binding of 2-bromo[1,2-14C]acetaldehyde, as well as theobservations that GSH blocked binding of 2-bromo[1,2-14C]-

acetaldehyde binding to protein but not DNA and that GSH

PROTEIN-7-ADDUCTS

HEME ONADESTRUCTION ADDUCTS

GLUTATHIONECONJUGATES

EXTRACELLULARADDUCTS

Chart 8. Postulated scheme for metabolism of vinyl halides. See text fordiscussion. The heavier arrows indicate the principal reactions on the basis ofthe results presented here and elsewhere.

blocked binding of metabolites derived from [1,2-14C]VBR to

protein but not DNA (Table 2). Similar results O.e., selectiveinhibition of DNA binding with epoxide hydrolase and proteinbinding with alcohol dehydrogenase) were found for the metabolism of [1,2-14C]VBR in a reconstituted system and thebinding of [1,2-14C]VC in microsomal systems (Charts 6 and

7). Moreover, aldehyde dehydrogenase did not inhibit bindingof [1,2-14C]VBR metabolites to DNA in the presence of NAD+.

Inactivated epoxide hydrolase, but not inactivated alcoholdehydrogenase, was consistently found to block some of thebinding of VBR and VC metabolites to DNA, but the inhibitionseen in the presence of active enzyme was significantly greater.Thus, the hydrolase can specifically block at least 50% of thebinding at the lower concentrations of the enzyme used bycatalyzing the hydration of 2-haloethylene oxides. The level of

endogenous epoxide hydrolase in microsomes is probably notsufficient to play a role in VBR metabolism under the cell-free

conditions used in these experiments. Binding did not increasein the presence of 1 mw concentration of 3,3,3-trichloropro-

pylene oxide, an epoxide hydrolase inhibitor. If one assumesthat about 2% of the microsomal protein is epoxide hydrolase,based upon the fold purification of epoxide hydrolase frommicrosomes to homogeneity (16), and uses the ratio of VmaxtoKm (Table 4) to estimate a first-order rate constant (k = 0.16/min) for 2-bromoethylene oxide hydrolysis (which is the effec

tive rate constant to be used for low concentrations of the

4396 CANCER RESEARCH VOL. 41

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Activation of Vinyl Halides

epoxide), a fi/2 of 4 min can be calculated when the concentration of microsomal protein is 1 mg/ml. Since the f1/2 of theepoxide is less than 1 min, a major role of epoxide hydrolasein the hydrolysis of 2-bromoethylene oxide would not be anticipated under the experimental conditions involving micro-

somes.2-Haloacetaldehydes have also been considered as reactive

metabolites generated from 1,2-dihaloalkanes (3, 13, 18). Zaj-dela ef al. (36) found that s.c. injection of 2-chloroethyleneoxide induced local tumors in mice but that injection of 2-

chloroacetaldehyde did not, although the latter compound wasrather toxic. These findings are consistent with the view thattumorigenesis is the result of alkylation of nucleic acids by 2-haloethylene oxides and that toxicity is the result of proteinalkylation by 2-chloroacetaldehyde. Nevertheless, the situation

in which extraparenchymal and extrahepatic tumors are produced by administration of vinyl halides may be more complex,with the possibility of migration of both metabolites throughoutthe body. While the experiments presented here might suggestthat epoxide hydrolase does not play a major role in thedetoxification of 2-haloethylene oxides generated from VC and

VBR in situ, high localized concentrations of the enzyme in vivoand possible coupling of the enzyme to cytochrome P-450 (27)

may be important.When rat liver hepatocytes were incubated with VC, more

than 90% of the hexane-insoluble metabolites were found to

leave the cells. DNA and protein were added to the incubationmedium, and more than 70% of the total irreversibly boundspecies were found outside the cell. Since viability did notsubstantially decrease during the time frame of the experiment,we conclude that the majority of the reactive metabolites canleave the intact hepatocyte. The binding to the exogenous DNAcould be partially blocked with epoxide hydrolase (25% inhibition) or alcohol dehydrogenase (40% inhibition), implying thatboth 2-chloroaceta!dehyde and, perhaps to a lesser extent, 2-

chloroethylene oxide can leave the hepatocyte. This observation is of significance because it is recognized that hemangio-

sarcomas are tumors of endothelial cells and not parenchymalcells. The endothelial cells are low in capacity to metabolizeforeign compounds (26), including VC (29). A possible schememight involve production of vinyl halide metabolites in hepatocytes, migration of reactive metabolites to the target endothelialcells, and subsequent binding to DNA and protein in thosecells. The data presented here indicate that reactive metabolites of VC are probably stable enough for this process tooccur. As one considers targets that are further removed fromthe activation source, stability of reactive metabolites becomesincreasingly important in determining if these metabolites canreach the target. The results presented here are consistentwith this view, as some alcohol dehydrogenase inhibition of thebinding of metabolites to DNA was observed when DNA wasplaced outside of hepatocytes.

Extensive differences between the metabolism of VC andVBR were not apparent in these experiments. At saturatingconcentrations of substrate, VBR was metabolized about 50%faster than VC. Bolt ef al. (6) have reported that the effectiveKmfor VBR metabolism is about one order of magnitude smallerthan that for VC. Thus, at low substrate concentrations, onewould expect VBR to be metabolized roughly an order ofmagnitude faster than VC. The ratio of DNA adducts to hexane-insoluble metabolites was similar for VC and VBR in the micro

somal experiments, but the ratio of protein adducts to hexane-

insoluble metabolites was higher in the case of VBR (approximately 0.5) than VC (approximately 0.1). 2-Bromoethyleneoxide had a shorter i,/2 than did 2-chloroethylene oxide (15),and 2-bromoethylene oxide and 2-bromoacetaldehyde reactedmore rapidly with adenine to form 1,6N-ethenoadenosine than

did the chlorine analogs (Chart 3). These observations maysuggest that VBR would be a more potent toxin and carcinogenthan VC, although comparative in vivo studies have not beenreported.

ACKNOWLEDGMENTS

We thank M. B. Mitchell, M. V. Martin, S. T. Wright, and J. Y. Domoradzki fortheir technical assistance and Dr. B. A. Schweiz and Dr. P. J. Gehring forreviewing the manuscript.

REFERENCES

1. Anonymous. Vinyl bromide tests show increase in cancer in rats. Chem.Week, Nov. 16, p. 40, 1977.

2. Anonymous. Key chemicals: vinyl chloride. Chem. Eng. News, 58: July 7, 9,1980.

3. Banerjee, S., Van Duuren, B. L., and Kline, S. A. Interaction of potentialmetabolites of the carcinogen ethylene dibromide with protein and DNA invitro. Biochem. Biophys. Res. Commun., 90. 1214-1220, 1979.

4. Barbin, A., Brésil,H., Croisy, A., Jacquignon, P.. Malaveille, C., Montesano,R., and Bartsch, H. Liver microsome-mediated formation of alkylating agentsfrom vinyl bromide and vinyl chloride. Biochem. Biophys. Res. Commun.,67: 596-603. 1975.

5. Bartsch, H., Malaveille, C., Barbin, A., and Planche, G. Mutagenic andalkylating metabolites of nalo-ethylenes, chlorobutadienes, and dichlorobu-tenes produced by rodent or human liver tissues: evidence for oxiraneformation by P450-linked microsomal mono-oxygenases. Arch. Toxicol., 41 :249-277, 1979.

6. Bolt, H. M., Filser, J. G., and Hinderer, R. K. Rat liver microsomal uptakeand irreversible protein binding of [1,2-"C]vinyl bromide. Toxicol. Appi.

Pharmacol.. 44: 481-489, 1978.7. Büffler,P. A., Wood, S., Eifler, C., Suarez, L., and Kilian, D. J. Mortality

experience of workers in a vinyl chloride monomer production plant. J.Occup. Med., 21: 195-203. 1979.

8. Cameron, R., Sharma, R. N., Sweeney, G. D., Farber, E., and Murray, R. K.A novel electrophoretic pattern of induction of rat liver microsomal membranepolypeptides by 2-acerylaminofluorene, nitrosamine, and azo dye administration. Biochem. Biophys. Res. Commun., 71. 1054-1061, 1976.

9. Creech. J. L., and Johnson, M. N. Angiosarcoma of the liver in the manufacture of polyvinyl chloride. J. Occup. Med., 16: 150-151, 1974.

10. DuBois, G. C.. Apella, E., Levin, W., Lu, A. Y. H., and Jerina, D. M. Hepaticmicrosomal epoxide hydrase: involvement of a histidine at the active sitesuggests a nucleophilic mechanism. J. Biol. Chem., 253:2932-2939,1978.

11. Guengerich, F. P. Studies on the activation of a model turan compound:toxicity and covalent binding of 2-<fV-ethylcarbamoylhydroxymethyl)furan.Biochem. Pharmacol., 26: 1909-1915, 1977.

12. Guengerich, F. P. Separation and purification of multiple forms of microsomalcytochrome P-450: partial characterization of three apparently homogeneous cytochromes P-450 prepared from livers of phénobarbital- and 3-methyl-cholanthrene-treated rats. J. Biol. Chem., 253. 7931-7939, 1978.

13. Guengerich, F. P., Crawford. W. M., Jr., Domoradzki, J. Y.. Macdonald, T.L.. and Watanabe, P. G. In vitro activation of 1,2-dichloroethane by microsomal and cytosolic enzymes. Toxicol. Appi. Pharmacol., 55: 304-317.

1980.14. Guengerich. F. P., Crawford, W. M., Jr., and Watanabe, P. G. Activation of

vinyl chloride to covalently bound metabolites: roles of 2-chloroethyleneoxide and 2-chloroacetaldehyde. Biochemistry, 18: 5177-5182, 1979.

15. Guengerich, F. P., and Martin, M. V. Purification of cytochrome P-450,NADPH-cytochrome P-450 reducíase, and epoxide hydratase from a singlepreparation of rat liver microsomes. Arch. Biochem. Biophys., 205: 365-

379, 1980.16. Guengerich, F. P., Wang. P., Mitchell, M. B.. and Mason, P. S. Rat and

human liver epoxide hydratase: purification and evidence for the existenceof multiple forms. J. Biol. Chem., 254: 12248-12254, 1979.

17. Guengerich, F. P., and Watanabe, P. G. Metabolism of "C- and 36CI-labeledvinyl chloride in vivo and in vitro. Biochem. Pharmacol., 28:589-596, 1979.

18. Hill. D. L., Shih, T-W.. Johnston, T. P., and Struck, R. F. Macromolecularbinding and metabolism of the carcinogen 1,2-dibromoethane. Cancer Res..38: 2438-2442, 1978.

19. Kadlubar, F. F.. Miller, J. A., and Miller. E. C. Microsomal N-oxidation of the

NOVEMBER 1981 4397

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

F. P. Guengerich et al.

hepatocarcinogen N-methyl-4-aminoazobenzene and the reactivity of N-hydroxy-fV-methyl-4-aminoazobenzene. Cancer Res., 36. 1196-1206,

1976.20. Kappus, H., Boit, H. M., Buchter, A., and Boit, H. M. Rat liver microsomes

catalyse covalent binding of "C-vinyl chloride to macromolecules. Nature

(Lond.), 257. 134-135. 1975.21. Lee, F. I., and Harry, D. S. Angiosarcoma of the liver in a vinyl-chloride

worker. Lancet, /. 1316-1319. 1974.22. Levene, P. A. Bromoacetone. Org. Syn., 2. 88-89, 1943.23. Malaveille. C.. Bartsch, H., Barbin, A., Camuss, A. M.. Montesano, R.,

Croisy. A., and Jacquignon, P. Mutagenicity of vinyl chloride, chloroethyl-eneoxide, chloroacetaldehyde, and chloroethanol. Biochem. Biophys. Res.Commun., 63: 363-370, 1975.

24. Maltoni, C., and Lefemine. G. Carcinogenicity bioassays of vinyl chloride. I.Research plan and early results. Environ. Res., 7: 387-405, 1974.

25. McCann, J . Simmon, V., Streitwieser, D., and Ames, B. H. Mutagenicity ofchloroacetaldehyde, a possible metabolic product of 1,2-dichloroethane(ethylene dichloride), chloroethanol (ethylene chlorohydrin), vinyl chloride,and cyclophosphamide. Proc. Nati. Acad. Sei. U. S. A., 72. 3190-3193,1975.

26. Morland, J., and Olsen, H. Metabolism of sulfadimidine, sulfanilamide, p-aminobenzoic acid, and isoniazid in suspensions of parenchyma! and non-parenchymal rat liver cells. Drug Metab. Dispos.. 5 511-517, 1977.

27. Oesch, F., and Daly, J. Conversion of naphthalene to (rans-naphthalenedihydrodiol: evidence for the presence of a coupled aryl monooxygenase-epoxide hydrase system in hepatic microsomes. Biochem. Biophys. Res.Commun.. 46: 1713-1720, 1972.

28. Osterman-Golkar, S . Hultmark. D., Segerback, D., Calleman, C. J., Gothe,

R., Ehrenberg, L., and Wachtmeister. C. A. Alkylation of DNA and proteinsin mice exposed to vinyl chloride. Biochem. Biophys. Res. Commun.. 76.259-266, 1977.

29. Ottenwälder, H., and Bolt, H. M. Metabolic activation of vinyl chloride andvinyl bromide by isolated hepatocytes and hepatic sinusoidal cells. J. Environ. Pathol. Toxico!., 4: 411-417, 1980.

30. Ottenwälder. H.. Laib, R.J., and Bolt, H.W. Alkylation of RNA by vinylbromide metabolites in vitro and in vivo. Arch. Toxicol., 41:279-286.1979.

31. Secnst, J. A., Ill, Barrio, J. R., Leonard, N. J., and Weber. G. Fluorescentmodification of adenosme-containing coenzymes: biological activities andspectroscopic properties. Biochemistry. 11: 3499-3506, 1972.

32. Wagner, E. R., Muelder, W. M., Watanabe, P. G., Hefner, R. E., Jr., Braun,W. H., and Gehring, P. J. A gas-chromatographic method for the preparationof '4C-labeled vinyl chloride. J. Labeled Comp., 11: 535-542. 1975.

33. Walling, C., and Fredricks, P. S. Positive halogen compounds. IV. Radicalreactions of chlorine and f-butyl hypochlorite with some small ring compounds. J. Am. Chem. Soc., 84: 3326-3331, 1962.

34. Walling, C., and Padwa, A. Positive halogen compounds. V. f-Butyl hypo-bromite and two new techniques for hydrocarbon bromination. J. Org.Chem., 27. 2976-2977, 1962.

35. Williams, G. M. Detection of chemical carcinogens by unscheduled DNAsynthesis in rat liver primary cell cultures. Cancer Res., 37: 1845-1851,1977.

36. Zajdela, F., Croisy, A., Barbin, A., Malaveille, C., Tomatis, L.. and Bartsch,H. Carcinogenicity of chloroethylene oxide, an ultimate reactive metaboliteof vinyl chloride, and bis(chloromethyl)ether after subcutaneous administration and in initiation-promotion experiments in mice. Cancer Res., 40:352-

356, 1980.

4398 CANCER RESEARCH VOL. 41

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1981;41:4391-4398. Cancer Res   F. Peter Guengerich, Patricia S. Mason, William T. Stott, et al.   Binding to Protein and DNADerived from Vinyl Bromide and Vinyl Chloride in Irreversible Roles of 2-Haloethylene Oxides and 2-Haloacetaldehydes

  Updated version

  http://cancerres.aacrjournals.org/content/41/11_Part_1/4391

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/41/11_Part_1/4391To request permission to re-use all or part of this article, use this link

on July 16, 2020. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from