Vol.

169,

No.

June

29,

1990

3, 1990

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS Pages

1094-l

098

SYNTHESIS AND PROPERTlES OF VINYL CARBAMA TE EPOXIDE, A POSSIBLE ULTIMATE ELfWl’ROPHILIC AND CABCINOGEMC METABOLITE OF VINYL CARBAMA TE AND ETHYL CARBAMATE Kwang-Kyun

Park, Young-Joon

McArdle

Received

May 8,

Surh, Bradley

C. Stewart

and James A. Miller

Laboratory for Cancer Research, School of Medicine, University of Wisconsin, Madison, WI 53706 1990

SUMMARX Vinyl carbamate reacted with dimethyldioxirane in dry acetone to give a high yield of pure crystalline vinyl carbamate epoxide. This epoxide was characterized by its NMR and MS spectra and elementary analysis. It is unstable at room temperature and has a half-life in water solution of approximately 32 minutes. It reacts with adenosine to form l,@-ethenoadenosine and more of this etheno nucleoside was found in hydrolysates of hepatic RNA of male mice injected i.p. with the epoxide than with vinyl carbamate. Tests with Salmonella typhimurium TA1535 showed that this epoxide is a strong direct mutagen. It is also more toxic in the mouse than vinyl carbamate. Studies on the Q1990Academic Press, Inc. carcinogenicity of this epoxide are in progress.

Ethyl carbamate (EC; CH3-CH,-O-CO-NH,) has long been known as a versatile carcinogen in rodents, especially in the mouse (1). It occurs naturally in small amounts in ethanolic fermentations (2). Unlike all other structural variants of EC that have been tested (1,3), vinyl carbamate (VC; CH, = CH-O-CO-NH,) is 10 to 50 times more carcinogenic than EC in the mouse (4,5). These carbamates induce identical spectra of tumors in the mouse and rat (5). Likewise, in the mouse liver in uiuo these carbamates produce the same DNA adduct [7-(2’-oxoethyl)deoxyguanosinel

and RNA

adducts

(l,i@-ethenoadenosine

and 3,N*-ethenocytidine)

(6-9). However, VC forms far more of these adducts than does EC (8,9). The same hepatic nucleic acid adducts are found in rats exposed to the related carcinogen, vinyl chloride (10-13). Vinyl chloride epoxide is an ultimate electrophilic and carcinogenic metabolite of vinyl chloride (14-16). By analogy, vinyl carbamate

/Q

epoxide (VCO; H,C-CH-O-CO-NH,) electrophilic and carcinogenic metabolite

has been proposed as an ultimate of VC and EC (4,5,9) and considerable

Abbreviations: EC, ethyl carbamate; VC, vinyl carbamate; VCO, vinyl carbamate epoxide; DMDO, dimethyldioxirane; DMSO, dimethyl sulfoxide. 0006-291x/90 Copyright All rights

$1.50

0 1990 by Academic Press, Inc. of reproduction in any form resewed.

1094

Vol.

169, No. 3, 1990

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

recent evidence supports but does not prove this concept (9). Further study of the metabolic activation of VC and EC is needed and may depend on the availability of VCO. The present report concerns the first synthesis of pure crystalline VCO and some of its chemical, biochemical and biological properties. MATERIATS

ANDmODS

Chemicals VC was prepared as described previously (9). Potassium peroxymonosulfate (OXONE), dry acetone (ACS reagent), molecular sieves (4A, activated powder) and DMSO-d6 were obtained from Aldrich Chemical Co. (Milwaukee, WI). Adenosine and l,@-ethenoadenosine were products of Sigma Chemical Co. (St. Louis, MO). All other chemicals were of ACS reagent grade. Synthesis of VCO Dimethyldioxirane (DMDO) was prepared and quantitated iodometrically as described previously (23). The synthesis of VCO was based on the previously reported method for the preparation of aflatoxin 8,9-epoxide (21) except that equimolar amounts of DMDO and VCO were reacted for 30 min. Instrumentation and Procedures lH-NMR spectra were determined with a Bruker AM-500 spectrometer and the mass spectral analyses were carried out with a Kratos MS50TC system. Reverse-phase HPLC on a C18 column was performed with a Waters system and a Schoeffel fluorescence detector as described previously (9). TLC was performed with plastic sheets pre-coated with 0.25 mm silica gel without gypsum (Polygram Sil G, Brinkmann instruments, Westbury, NY) with ethyl ether-hexane (6:1, v/v) as the solvent. The spots were visualized with sprays of an acidic ethanol solution of p-dimethylaminobenzaldehyde (yellow spots), fresh Tollen’s reagent (dark brown spots; prepared from equal volumes of 10% AgNO,and 10% NaOH followed by dropwise addition of cont. NH40H until the Ag20 precipitate dissolves), and cont. H2S04 followed by heating at 120°C (dark spots). Huffman Laboratories (Golden, previously described (9).

The elementary analyses were performed CO). The mutagenicity tests were carried

at the out as

Animals Female C57BL/6J and male C3HIHeJ mice from the Jackson Laboratory (Bar Harbor, ME) were bred in our laboratory to give B6C3FI offspring. RESULTS

AND DISCUSSION

Previous attempts to synthesize and isolate VCO in this laboratory by the oxidation of VC with 3chloroperbenzoic acid were unsuccessful. However, the use of this oxidant to form VCO in. situ succeeded in a reaction in which VC and the peracid dissolved in methylene dichloride were shaken vigorously for 12 hours with a solution of adenosine in water. Yields of 5-10% of l,I@-ethenoadenosine were obtained (9). It is likely that much of the VCO formed in this procedure and in attempts to isolate VCO from reactions of VC with the peracid was lost in reactions of VCO with the nucleophiles, water and 3-chlorobenzoic acid. The same problem was apparently the case in the unsuccessful attempts to prepare the 8,9-epoxide of aflatoxin B, (17-20). Recently, the successful isolation of this 1095

Vol.

169, No. 3, 1990

BIOCHEMICAL

AND BIOPHYSICAL

H

H

‘c-c’ H ’ ‘0’

‘c=c H’

vc m

0

RESEARCH COMMUNICATIONS

NH2

y

‘H

+ d

w

8 C ’ ‘CH,

vco The reaction of dimethyldioxirane

with VC to form VCO.

reactive epoxide was achieved (21) with the relatively new and versatile oxidant, DMDO (22). This reagent epoxidizes carbon-carbon double bonds with the formation of acetone as an unreactive product (22,23) (Fig. 1). We were pleased to find that DMDO readily oxidizes VC to form VCO in a smooth high-yield reaction. This oxidant was generated from acetone and a preparation of potassium peroxymonosulfate (OXONE) by the small-scale procedure of Adam et al. (23). This procedure yields about 10 ml of a faintly yellow solution of DMDO in wet acetone at a concentration of about 0.1 M. The exact concentration was determined iodometrically (23). The solution of DMDO was dried at room temperature by stirring it with 1 g of activated 4A molecular sieves for 1 hour and then filtering it by gentle vacuum through a fine glass frit funnel into a small reaction flask. Approximately the same volume of dry acetone containing a mole equivalent of VC was added and the reaction allowed to proceed for 30 minutes at room temperature (the faint yellow color of DMDO disappeared). The reaction mixture was then cooled to O°C and the acetone was removed in uucuo on a rotary evaporator while the flask was surrounded by ice. Crystallization of the VCO occurred as the acetone was removed and when all of the residual solvent was distilled off, a white mass of transparent crystals of VCO remained in yields of at least 90%. The preparation was pure without further recrystallization and a solution in acetone gave only one spot on TLC with an Rf of 0.27 (VC, 0.49; EC, 0.37). The crystals were stable at room temperature for only 1 to 2 days when kept dry in the dark. They appear stable at -7OOC. The crystals decomposed on heating without melting to give a dark mass. The crystals were characterized as VCO by elementary analysis and NMR spectroscopy. Elementary analysis gave values for the percentages of C,H,N and 0 as 34.92, 4.88, 13.53 and 46.90, respectively, as compared to 34.96 (C), 4.89 (H), 13.59 (N) and 46.56 (0) for C,H,N03 The 500 MHz IH-NMR spectrum of the crystals in DMSO-d6 exhibited a modified ABX pattern (24) for the carbon-bound protons in which the more usual pair of doublets for each proton appeared as a singlet for the C-H proton (X, 5.40 ppm) and as a doublet and triplet, respectively, for the CH2 protons (A, centered at 2.84 ppm and B, centered at 2.75 ppm). The NH, protons appeared as nearly separate and relatively sharp equal peaks at 6.98 ppm and 6.85 ppm. The integral accounted for each of the five protons. The ABX pattern in the corresponding spectrum for VC was considerably shifted downfield 1096

Vol.

169, No. 3, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

VCO (nmollplate) E&u-e

2, The strong TA1535.

direct

mutagenicity

of VCO with Salmonella

typhimurium

from that of VCO. Each of the protons in the vinyl group appeared as doublets (X, centered at 7.08 ppm; A, centered, at 4.69 ppm; B, centered at 4.40 ppm). The NH, protons appeared as equal and nearly separate sharp peaks at 7.05 ppm and 6.85 ppm. The integral accounted for each of the five protons. The mass spectral analysis of VCO did not show a molecular ion at 70, 25, or 18 electron volts in the EI spectra.

Under these conditions the spectrum of VC showed a molecular ion.

VCO and VC gave very similar fragmentation m/e of 44,63 and 78. The half-life of VCO approximately 32 minutes nitrobenzyl)pyridine

(25).

l,@-ethenoadenosine an authentic

peaks at

in water solution at room temperature was as determined by the alkylation of 4-(p-

A reaction of VCO with adenosine in water yielded

as shown by fluorometric

sample of the etheno derivative.

1.3% and this may be a reflection of the lability 20 pg of VCO or VC in trioctanoin mice.

patterns with prominent

At 8 hours the hepatic

cochromatography

on HPLC with

However, the yield was only about of VCO in water. Suspensions of

were injected into 12-day-old male B6C3F1

RNAs of these mice were analyzed

for 1,N6-

ethenoadenosine (9) and found to contain 135 pmol/mg RNA for the animals injected with VCO as compared to 90 pmol/mg RNA for those injected with VC. Intraperitoneal

injections of VC are acutely toxic in mice as compared with

EC (4) and VCO, in turn, appears to be more acutely toxic than VC in 12-day-old male B6C3F, mice. VC is only mutagenic for Salmonella typhimurium TA1535 when it is activated by liver microsomes fortified with NADPH and oxygen (5,6,9). However, VCO is a strong direct mutagen as shown in Fig. 2. The carcinogenicity and other properties of VCO are under investigation Recently

Gupta

and Dani (26) reported

in our laboratory.

the metabolic

formation

of the

“epoxy derivative of ethyl carbamate” upon the incubation of EC with rat lung microsomes. In this study the metabolites were analyzed by TLC on silica gel with an aqueous solvent system of butanol-propanol-2 N ammonia (2:1:1, v/v). The “epoxy” spot moved with an Rf value higher than that of EC. No standard of 1097

Vol. 169, No. 3, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

VCO and no characterization of the “epoxy” spot were reported. reproduce this TLC profile with EC and chemically synthesized doubt if this water-labile epoxide was observed by these investigators.

We could not VCO and we

ACKNOWLEDGMEN’IS Supported by grants CA-07175 and CA-22484, National Cancer Institute, U.S. Department of Health and Human Services. We are indebted to Judy Blomquist and Amy Liem for the mutagenicity assays and animal studies, respectively.

1. Mirvish, S.S. (1968) Adv. Cancer Res., u l-42. Ough, C.S. (1976) J. Agric. Food Chem., a, 323-328. I., Ben-Ishai, D., Haran-Ghera, N., Lapidot, A., Simon, E. and 32: Berenblum, Trainin, N. (1959) Biochem. Pharmacol., 2,168-176. 4. Dal-d, G.A., Miller, J.A. and Miller, E.C. (1978) Cancer Res., 3,3793-3804. Dahl, G.A., Miller, E.C. and Miller, J.A. (1980) Cancer Res., a, 1194-1203. E: Scherer, E., Winterwerp, H. and Emmelot, P. (1986) In The role of Cyclic Nucleic Acid Adducts in Carcinogenesis and Mutagenesis (B. Singer and H. Bartsch, eds.), pp. 109-125, IARC Scientific Publ. No. 70, Lyon. L.G. (1982) 7. Ribovich, M.L., Miller, J.A., Miller, E.C. and Timmins, Carcinogenesis, 3,539-546. 8. Miller, J.A. and Miller, E.C. (1983) Br. J. Cancer, &&l-15. 9. Leithauser, M.T., Liem, A., Stewart, B.C., Miller, E.C. and Miller, J.A. (1990) Carcinogenesis, u 463-473. 10. Laib, R.J. and Bolt, H.M. (1977) Toxicology, &185-195. 11. Laib, R.J. and Bolt, H.M. (1978) Arch. Toxicol.,& 235240. H. (1980) Arch. Toxicol., 12. Bolt, H.M., Filser, J.G., Laib, R.J. and Ottenwalder, Suppl. 3,129142. 13. Laib, R.J., Gwinner, L.M. and Bolt, H.M. (1981) Chem.-Biol. Intertact.,X, 219-231. 14. Bartsch, H., Maleveille, C., Barbin, A. and Planche, G. (1979) Arch. Toxicol., a, 249-277. 15. Bolt, H.M., Laib, R.J. and Filser, J.G. (1982) Biochem. Pharmacol., a, l-4. 16. Zajdela, F., Croisy, A., Barbin, A., Maleveille, C., Tomatis, L. and Bartsch, H. (1980) Cancer Res., a, 352-356. 17. Gorst-Allman, C.P., Steyn, P.S. and Wessels, P.L. (1977) J. Chem. Sot., Perkin Trans., 1,1360-1364. 18. Coles, B.F., Lindsay Smith, J.R. and Garner, R.C. (1979) J. Chem. Sot., Perkin Trans., 1,26642671. 19. Garner, R.C., Martin, C.N., Lindsay Smith, J.R., Coles, B.F. and Tolson, M.R. (1979) Chem.-Biol. Interact., Z,57-73. Miller, J.A. (1970) Cancer Res., a, 559-576. 2 Baertschi, S.W., Raney, K.D., Stone, M.P. and Harris, T.M. (1988) J. Am. Chem. Sot., m 7929-7931. 22. Adam, W., Curci, R. and Edwards, J.O. (1989) Act. Chem. Res.,a, 205-211. 23. Adam, W., Chan, Y.-Y., Cremer, D., Gauss, J., Scheutzow, D. and Schindler, M. (1987) J. Org. Chem., a, 2800-2803. 24. Silverstein, R.M., Bassler, G.C. and Merrill, T.C. (1974) Spectrophotometric Identification of Organic Compounds. 3rd ed., p. 340, John Wiley, New York. 25. Barbin, A., Bre’sil, H., Croisy, A., Jacquignon, P., Malaveille, C., Montesano, R. and Bartsch, H. (1975) Biochem. Biophys. Res. Commun., a, 596-603. 26. Gupta, R. and Dani, H.M. (1989) Toxicology Letters, &, 49-53. 1098