Naturally occurring carbonyl compounds are mutagens Salmonella tester strain TA104

Mutation Research, 148 (1985) 25-34

25

Elsevier MTR 03966

Naturally occurring carbonyl compounds are mutagens in Salmonella tester strain TA104 L a w r e n c e J. M a r n e t t *, H o l l y K. H u r d , M o n i c a C. Hollstein, D a v i d E. Levin, H e r m a n n E s t e r b a u e r ** a n d B r u c e N. A m e s *** Department of Biochemistry, Universityof California, Berkeley, CA 94720 (U.S.A.) (Received 12 March 1984) (Revision received 6 August 1984) (Accepted 26 August 1984)

Summary Strains of Salmonella typhimurium that carry a nonsense mutation at the site of reversion detect a variety of naturally occurring and synthetic carbonyl compounds as direct-acting mutagens. TA104 is reverted efficiently by formaldehyde, a,fl-unsaturated aldehydes (enals), and dicarbonyl compounds, such as diacetyl and glutaraldehyde. This strain is much more sensitive to carbonyl mutagenesis than is TA100, a strain previously reported to detect aldehydes as mutagens, or any other characterized strains of Salmonella. Long-chain enals are very toxic to TA104, but addition of a reduced glutathione chase following an incubation period decreases this toxicity, thus enabling the detection of 4-hydroxy-pentenal, a homolog of the lipid peroxidation product, 4-hydroxy-nonenal, as a mutagen. This is the first report of the mutagenicity of a hydroxy-enal, a class of enals produced by lipid peroxidation. Testing conducted with strains that carry the nonsense mutation in different repair backgrounds indicates that the presence of pKM101 and the deletion of the uvrB gene facilitate the detection of enals and dicarbonyls, but not malondialdehyde, as mutagens. Since carbonyl compounds are widely distributed in foods, are generated during cellular metabolism, and are present in body fluids, they may make a significant contribution to the risk of human cancer.

Carbonyl compounds are ubiquitously distributed in the environment. In addition to their use in the chemical industry (National Research Council, 1981), aldehydes and ketones are components of food (Hsieh et al., 1981; Stone et al., 1975; Schreier et al., 1981; deLumen et al., 1978; Reindl and Stan, 1982), intermediates in metabolism (Schauenstein et al., 1977), and end products of lipid * Present address: Department of Chemistry, Wayne State University, Detroit, MI 48202, U.S.A. ** lnstitut ~ Biochemie der Karl-Franzens-Universif/tt Graz, A-8010 Graz, Austria. *** To whom reprint requests should be sent.

peroxidation (Reindl and Stan, 1982; Schauenstein et al., 1977; Esterbauer et al., 1982). They have been detected in urine, plasma, and expired air of normal individuals and their concentrations in humans can be substantial (Krotoszynski et al., 1977; Rhodes et al., 1982; Zlatkis et al., 1980, 1981). Saturated and unsaturated aldehydes and ketones and dicarbonyl compounds have been tested for mutagenicity in Salmonella typhimurium (Sasaki and Endo, 1978; Eder et al., 1982; Rosen et al., 1980; McCann et al., 1975a; Lrfroth, 1978; Lijinsky and Andrews, 1980; Lutz et al., 1982; Bjeldanes and Chew, 1979; Yamaguchi, 1982; Kasai et al., 1982; Mukai and Goldstein, 1976; Basu and

002%5107/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

26

Marnett, 1983, 1984; Marnett and Tuttle, 1980; Levin et al., 1982). Formaldehyde, halo aldehydes, 2-butene-3-one, and some enais and halo enals are mutagenic, but saturated aldehydes and ketones are nonmutagenic in TA100 (Sasaki and Endo, 1978; Eder et al., 1982; Rosen et al., 1980; McCann et al., 1975a; L/3froth, 1978; Lijinsky and Andrews, 1980; Lutz et al., 1982). Certain a-diketones are also positive in TA100 and methyl glyoxal exhibits potent activity (Bjeldanes and Chew, 1979; Yamaguchi, 1982; Kasai et al., 1982). Although most aldehydes appear only to revert the base-pair substitution strain TA100, malondialdehyde and a series of structurally related/t-substituted acroleins are active in hisD3052 (Mukai and Goldstein, 1976; Basu and Marnett, 1983, 1984; Marnett and Turtle, 1980), a strain that detects frameshift mutagens. We report here the use of two new base substitution strains, TA104 and TA102 (Levin et al., 1982). to reassess the mutagenic potency of carbonyl compounds by testing 28 naturally occurring compounds in the Salmonella test. TA104 carries a nonsense mutation (-TAA-) at the site of reversion that is present in single copy on the chromosome, and TA102 contains the same histidine mutation on a multicopy plasmid. Both strains were originally developed to detect peroxides and other oxidants. Our results indicate that TA104 also detects carbonyl mutagens efficiently and this strain is considerably more sensitive to carbonyls than is TA100. In addition, with TA104 we demonstrate the mutagenic activity of several naturally occurring compounds not previously reported positive in Salmonella and define the genetic requirements for the detection of carbonyl compounds as mutagens in this strain. Materials and methods

Chemicals Acetaldehyde, p-tolualdehyde, acrolein, crotonaldehyde, furfural, 2-hexenal, 2,4-hexadienal, 2,4nonadienal, nonanal, 3-buten-2-one, 2-butanone, 3-penten-2-one, and acrylonitrile were from Aldrich. Methacrolein, diacetyl, methyl glyoxal, glyoxal, glutathione, and t-butyl-hydroperoxide were from Sigma. Hexanal and 2-nonenal were from Pflatz and Bauer. 2-Heptenal and 2-octenal

were from Alfa Products. Glutaraldehyde (Baker), formaldehyde (Mallinckrodt), acetol (ICN Pharmaceuticals), kethoxal (US Biochemicals), and cinnamaldehyde (BDH Chemicals Ltd.) were obtained from the indicated sources. 4-Hydroxypentenal, 4-hydroxy-octenal, 4-hydroxy-nonenal, and malondialdehyde were synthesized according to literature procedures (Basu and Marnett, 1983; Esterbauer and Weger, 1967). Chemicals from commercial sources were of the highest purity available and were used without additional purification. Mutagenicity assays Compounds were tested in strains of Salmonella using a liquid preincubation procedure (Maron and Ames, 1983). 0.02 M sodium phosphate buffer (pH 7.4) was added to sterile 13 mm × 100 mm capped culture tubes followed by the test compound (in DMSO or water) and then 0.1 ml of an overnight culture of the bacterial strain to a final volume of 0.5 ml. The tubes were incubated with shaking at 37°C for 20 min. In experiments with glutathione, 50 /~1 of a 0.1 M solution (in 0.2 M sodium phosphate buffer pH 7.4) was added to the reaction tubes at the end of the preincubation period. Following incubation, 2 ml of molten top agar containing histidine and biotin was added to each tube, the mixture was plated on minimal glucose, and the revertants were scored after 48 h. Experiments were performed in duplicate and tbutyl-hydroperoxide was used as a positive control (Levin et al., 1982). Each compound was tested to its toxic limit. Results

28 carbonyl compounds were tested in Salmonella typhimurium TA104 by liquid preincubation methods (Table 1); 13 compounds were mutagenic as judged by a dose-response curve showing at least a 2-fold increase over the spontaneous reversion frequency. Compounds positive in TA104 were either less active or not mutagenic in the standard tester strains TA100, TA97, TA98, and TA102. The liquid preincubation procedure was employed for testing because it enhances the reversion of TA100 in the presence of certain carbonyl compounds (Neudecker et al., 1981).

27 TABLE 1 C A R B O N Y L C O M P O U N D S TESTED FOR M U T A G E N I C I T Y IN TA104 Compound

Detection

M a x i m u m non-toxic dose (# moles) a

Mutagenicity TA 104 (revertants//t mole) ~

Serum Serum, Serum, Serum, Serum. Food -

3 > 114 8 0.4 > 3 > 68 > 111

1 790

Saturated Formaldehyde Acetaldehyde Hexanal Nonanal Butanone Acetol Dihydroxyacetone

urine, expired air lipid peroxidation lipid peroxidation expired air

250 5

Unsaturated Acrolein Crotonaldehyde Methacrolein 2-Hexenal 2-Heptenal 2-Octenal 2-Nonenal 4-Hydroxy-pentenal 4-Hydroxy-octenal 4-Hydroxy-nonenal 2,4-Hexadienal 2,4-Nonadienal 3-Penten-2-one 3-Buten-2-one p-Tolualdehyde Cinnamaldehyde Furfural

Serum, food Serum Serum Serum, lipid peroxidation, food Lipid peroxidation Lipid peroxidation Lipid peroxidation Lipid peroxidation Lipid peroxidation Lipid peroxidation Lipid peroxidation Serum, lipid peroxidation Serum, urine Expired air Food Food, plasma, urine

>

>

> >

0.9 (1.8) 3 (3) 1.4 ( > 3) 2 ( > 5) 0.9 (4.4) 0.8 (4) 0.007 2 (5) 1.4 ( > 4) 0.5 ( > 1.3) 1 ( > 5) 0.4 2 0.7 0.8 0.8 1

1080 940 610 460 165 b 960 -

Dicarbonyls Glutaraldehyde Glyoxal Methyl glyoxal Kethoxal Diacetyl Malondialdehyde

Food Food Food Lipid peroxidation

> 0.5 > 2 7 > 1 6 > 111

4150 2250 18 250 3 200 340 5

a Numbers in parentheses refer to m a x i m u m nontoxic dose in the presence of 10 m M glutathione as described in Materials and Methods. b Value obtained for test when glutathione was added during post-incubation. The number of induced revertants//tmole were calculated from the linear portion of dose-response curves from which an average spontaneous reversion value of 304 was subtracted.

Saturated carbonyl compounds Formaldehyde, acetaldehyde, hexanal, nonanal and butanone were tested at concentrations up to 1 mg/plate in TA102 and TA104. Of these only formaldehyde is mutagenic, and its activity in the new strains is compared with the standard tester strain TA97, TA98, and TA100 in the dose-response curves in Fig. 1. The new strains are more

sensitive to formaldehyde-induced mutagenesis than is TA100 (Levin et al., 1982), the strain previously reported to demonstrate the mutagenicity of this compound (Sasaki and Endo, 1978).

Unsaturated carbonyl compounds Fig. 2 displays the dose-response curves for a series of a, fl-unsaturated aldehydes (enals) that

28 1

1

1

I

I

I

I

2,000 2,000

TAI02 1,500 Crotonaldehyde

1,500

x

0Q ¢

-

t,O00

HexlldleTl~

r oleln'-

2 c

¢=

"2

1,000

re 500 .c.°,.,+

5OO

0

0

05

I /xmoles/plat

0

0.5

I

Formaldehyde

15 2 (p, moles/plate)

Fig. 1. Reversion of standard tester strains by formaldehyde. The number of spontaneous revertants per plate of 76, 28, 146, 330 and 308 for TA97, TA98, TAIO0, TAI02 and TAI04, respectively, were subtracted.

are mutagenic in TA104; no mutagenicity was observed with TA102. Most of the compounds appear to exhibit similar mutagenic potency at low concentrations, though significant differences in toxicity affect the maximum number of revertants/plate. These observations suggest that toxicity is an important factor in the detection of enals as mutagens. For example, 2-nonenal and 2,4-nonadienal may be too toxic (Table 1) to enable their testing at concentrations predicted to give a positive response based on the data in Fig. 2. In order to reduce toxicity, we retested both the mutagenic and nonmutagenic enals in the presence of a glutathione chase. Excess enai molecules still present during post-incubation can be inactivated by reacting with glutathione, thereby preventing further damage to protein sulfhydryls. Reaction

15

e

Fig. 2. Mutagenicity of unsaturated aldehydes in TA104. The number of spontaneous revertants per plate of 371 for acrolein and crotonaldehyde, 291 for methacrolein, and 210 for 2,4hexadienal and 2-hexenal were subtracted.

with sulfhydryl groups has been proposed as the cause of enal toxicity (Schauenstein and Esterbauer, 1979). As shown in Table 1, the glutathione chase reduced toxicity of all the enals tested; however, it did not reduce their mutagenic potency. This indicates that mutagenicity and toxicity of enals are independent events. In addition, 4-hydroxy-pentenal (nonmutagenic in the standard assay) shows significant mutagenic activity when tested at higher doses in the presence of a glutathione chase (Fig. 3). Malondialdehyde, which exists as fl-hydroxyacrolein in aqueous solution, is the single enal that was mutagenic in TA102. The dose-response shown in Fig. 4 indicates that malondialdehyde is actually slightly more mutagenic in TA102 than in TA104. This implies that the mechanism of mutagenesis of malondialdehyde is different than that of the other enals. Although malondialdehyde could also be considered a dicarbonyl compound,

29

"iiI

I

1

1

I

I

l

500

400

+GSH

Booy ® 700 a~

E ~o o.

TAI02

ID

600 t~

500 > rr

/

I

i

400

500

/~~

TAI04

50O

>

200

~r 200

I00

I

2

4-Hydr¢,xypentenal

3

I 5

4

(/~moles/plat

ioo

e)

Fig. 3. Reversion of TA104 by 4-hydroxy-pentenal in the presence and absence of glutathione added after incubation. Spontaneous revertants per plate of 215 (plus GSH) and 235 (minus GSH) were subtracted.

the strain specificity that it exhibits differs from that of several dicarbonyl compounds tested (see below). The aromatic enals p-tolualdehyde, cinnamaldehyde, and furfural were nonmutagenic in TA102 and TA104, as were the enones, 3-buten-2one and 3-penten-2-one, and the a,fl-unsaturated nitrile, acrylonitrile.

Dicarbonyl compounds All 6 dicarbonyl compounds tested are mutagenic in TA104 (Fig. 5). The substituted glyoxals, methyl glyoxal and kethoxal, are the most potent of the carbonyl compounds tested in this study. These dicarbonyls are also mutagenic in TA102 (data not shown), and others have demonstrated the mutagenicity of this class of compounds in TA100 (Sasaki and Endo, 1978; Kasai et al., 1982). Fig. 6 shows a comparison of revertants using the standard tester strains in the presence of the pro-

o 0

50

Malondialdehyde

I00 (/~moles/plat

e)

Fig. 4. Mutagenicityof malondialdehyde in TA102 and TA104. The number of spontaneous revertants per plate of 292 for TA102 and 341 for TA104 were subtracted.

totypic dicarbonyl glyoxal and indicates that TA104 is the most sensitive. Figs. 2 and 5 show that the mutagenic activity of dicarbonyl compounds is greater than that of enals in TA104. The toxicity of the two classes of compounds occurs in a similar concentration range, which appears to account for the ability of TA102 to detect dicarbonyl compounds but not enals. This provides additional evidence that toxicity and mutagenicity in these strains are independent events triggered by different chemical reactions.

Genetic requirements for mutagenicity TA104 was constructed from the histidine auxotroph hisG428 by deletion of the uurB gene, alteration of the cell wall by insertion of a deep-rough mutation, and incorporation of the R factor,

30

t~,O00

1

i

5,000

Methyl Glyoxel

I

1

TAI04 IO,O00

4,000

8,000]E ~

6,000

3,000

e~ I

I/

/o,,o:., Kethoxlll

2 2,000 er

2,000

I,O00

0

05

I

15

bcrnofes/plat e

Fig. 5. Mutagenicity of dicarbonyls in TAI04. The number of spontaneous revertants per plate of 385 for glutaraldehyde, 210 for methyl glyoxal and diacetyl, and 436 for kethoxal and glyoxal were subtracted. Kethoxal. glyoxal and methyl glyoxal were dissolved in water whereas glutaraldehyde and diacetyl were dissolved in DMSO.

pKMI01, which carries a gene for an error-prone repair enzyme of the SOS system (Levin et al., 1982). The uvrB deletion and R factor dramatically increase the sensitivity of Salmonella strains to mutagens and, in fact, pKM101 is essential for the detection of certain classes of compounds (Maron and Ames, 1983; Ames et al.. 1973; McCann et al., 1975a, b). To investigate the effect of the uvrB deletion on carbonyl mutagenicity, several enals and methyl glyoxai were assayed in TA2638 (Levin et al., 1982), a strain that contains the deep-rough mutation and pKM101, but is uorB +. The mutagenicity of methyl glyoxal in TA2638 and TA104 are compared in Fig. 7. Results from experiments with acrolein, crotonaldehyde, and methacrolein are identical to those obtained with methyl glyoxal (data not shown). TA2638 is approximately 10-fold less sensitive than TA104 to

_• 0

0

~n ~

TAIO0 e / ~ 0.5 Glyoxal

~';rAio2 ~

•

~1

L,TA98 -~a'TAg.7 1.5 2

;

(/~ moles/plate)

Fig. 6. Reversion of standard tester strains by glyoxal. The number of spontaneous revertants per plate of 76, 28. 146, 117 and 436 for TA97, TA98, TAIO0, TA102 and TAI04, respectively, were subtracted.

mutagenesis induced by enals and methyl glyoxal, which indicates that these agents cause DNA damage that is repaired relatively efficiently in a uvrB-dependent process. The requirement for pKM101 was investigated by assaying a series of enals and methyl glyoxal in TA2659, a strain that contains a deep-rough mutation and a uvrB deletion but does not carry pKM101 (Levin et al., 1982). TA2659 detects methyl glyoxal as a mutagen but is nearly 100-fold less sensitive than TA104 to mutagenesis by methyl glyoxal (Fig. 7). This dramatic decrease in sensitivity, also seen with the enals tested in TA2659 (data not shown), indicates that the muc genes, carried on pKM101 (Perry and Walker, 1982), an essential to the efficient detection of carbonyl compounds as mutagens in Salmonella.

31 I

1

12,000

4

IO, O 0 0

~, 8 , 0 0 0

¢&

6,ooo

4.000

05

Mel tlyl O l y o x a l

I

15

Q/.molo s / p l a t e }

Fig. 7. Reversion of TA2659, TA2638 and TA104 by methyl glyoxal. The number of spontaneous revertants per plate of 9, 42 and 210 for TA2659, TA2638 and TA104, respectively, were subtracted.

Discussion

The present work indicates that a number of naturally occurring aldehydes and ketones, some of which are also important in industry, are mutagenic without metabolic activation in TA104 (Table 1), a strain of Salmonella that contains a nonsense mutation at the site of reversion (Levin et al., 1982). This new tester strain is considerably more sensitive than other strains for identifying this class of compounds as mutagens. The analogous strain, TAI02, in which the same histidine mutation is present on a multicopy plasmid (Levin et al., 1982), also detects some of these mutagens, though not as effectively as TA104. The greater reversion of TA104 over TA102 has also been seen with peroxides, although TA104 is more sensitive to killing which limits the concentration range of

the compounds that can be used for screening (Levin et al., 1982). The toxicity of enals to TA104 is a major limitation in detecting them as mutagens. The lower mutagenicity of acrolein in the present assay relative to that reported previously may be a reflection of its marked toxicity in TA104. The fact that acrolein is detectable at all is probably due to its high mutagenicity relative to other enals. Enals appear to be toxic to cells because they react with sulfhydryl groups of proteins (Schauenstein and Esterbauer. 1979), whereas their mutagenicity is due to addition to nucleic acid bases, particularly deoxyguanosine (Chung and Hecht, 1983). By adding glutathione to the reaction mix as a chase prior to plating we were able to reduce the toxicity of the enals up to 5-fold, thereby increasing the concentration range over which we could test for mutagenic activity. This enabled us to show that 4-hydroxy-pentenal, a homolog of the lipid peroxidation product, 4-hydroxy-nonenal is mutagenic in Salmonella. Glutathione also reduced the toxicity of the enal mutagens acrolein, methacrolein, and crotonaldehyde without affecting their mutagenic activity. This is further evidence that toxicity and mutagenicity for a series of enals result from different chemical reactions and are, therefore, independent events. Although toxicity limits the detection of high molecular weight enals as mutagens, TA104 appears to be highly sensitive for screening carbonyl compounds for mutagenicity. Important factors in the sensitivity of TA104 are the deletion of the uvrB gene and the incorporation of pKM101. The uvrB gene codes for an error-free DNA excisionrepair system (Ames et al., 1973), and the insertion of the pKM101 plasmid into Salmonella tester strains amplifies their sensitivity to many classes of mutagens by providing a gene product that appears to function in an error-prone repair pathway (Perry and Walker, 1982). The uvrB deletion and pKM101 amplify sensitivity to enals and dicarbonyl compounds. In contrast, malondialdehyde, a dicarbonyl compound that exists as an enal (B-hydroxy-acrolein) in aqueous solution, exhibits comparable mutagenicity in TA104 and hisG428. * Since the latter strain does not contain " L.J. Marnett and D.E. Levin, unpublished result.

32

pKM101, the plasmid does not increase sensitivity to malondiaidehyde. Similar results have been reported with a series of strains derived from the hisD3052 mutation (Mukai and Goldstein, 1976; Basu and Marnett, 1984). This implies that the mechanism of mutagenicity by malondialdehyde is different from that of enals and a-dicarbonyl compounds. This is also suggested by the observation that malondialdehyde is as mutagenic in TA102 as in TA104. The adducts formed to deoxyguanosine by acrolein, methyl glyoxal, and malondialdehyde are chemically related but not structurally identical (Chung and Hecht, 1983; Shapiro et al., 1970; Moschel and Leonard, 1976; Galliani and Pantarotto, 1983). Carbonyl compounds constitute a potentially important source of human exposure to mutagens (Table 1). In addition to the use of certain carbonyl compounds as intermediates in manufacturing, these compounds are nearly ubiquitous in food and animal tissues. Methyl glyoxai, glyoxal, and diacetyl were the major mutagenic carbonyl compounds extracted from ground-roasted coffee beans (Kasai et al., 1982). Methyl glyoxal is present at levels of approximately 0.5 mg per cup of freshly brewed coffee and accounts for 50% of the mutagenicity of coffee (Kasai et al., 1982). We show here that acetol, another major component of coffee (Kasai et al., 1982), is mutagenic in TA102 and TA104 as are the 3 other carbonyl compounds in coffee. Methyl glyoxal can also occur as a product of cellular metabolism (Sato et al., 1980). Diacetyl is the natural aroma of butter and is also present in many foods such as wine (Jay, 1982). Cigarette smoke contains glyoxal, and diacetyl is present at levels of hundreds of p,g per cigarette (Moree-Testa and Saint-Jaim, 1981). trans-2-Hexenal is widely distributed in fruits and vegetables as the isomerization product of a flavor and aroma component, cis-3-hexenal; approximately 4 mg are present in a cup of apple juice (Schreier and Lorenz, 1982; Durr et al., 1981). The induction of lipid peroxidation in rat liver microsomes results in the generation of numerous carbonyl compounds, corresponding to 1 ptmole/g liver, among which 4-hy° droxy-nonenal is a major component (Esterbauer et al., 1982). This compound has been identified as the major toxic species during lipid peroxidation (Benedetti et al., 1980). Dihydroxyacetone is the

active ingredient (up to 7.5%) in some sun-tanning lotions (Pham et al,, 1979). Glutaraldehyde, which is quite mutagenic in TA104, is widely used in both medicine and manufacturing and has previously been negative in a number of mutagenicity tests (Slesinski et al., 1983). Considering their widespread occurrence, it is perhaps not surprising that carbonyl compounds are present in body fluids of normal, healthy individuals (Krotoszynski et al.. 1977; Rhodes et al., 1982; Zlatkis et al., 1980, 1981). The presence of a variety of inducible detoxifying enzymes active on aldehydes and ketones also suggests that they could be of importance in cell damage (Weiner and Wermuth, 1982). Thus both endogenous and exogenous aldehydes and ketones may contribute to cancer in humans. The extent of this contribution remains to be determined.

Acknowledgements This work was supported by National Institutes of Health Research Grant CA22206 to L.J.M., Department of Energy Contract DE-AT0376EV70156 to B.N.A., and by National Institute of Environmental Health Sciences Center Grant ES01896. D.E.L. was supported by National Institute of Environmental Health Sciences Training Grant ES07075. L.J.M. is a recipient of a Faculty Research Award from the American Cancer Society (FRA243). The work of H. Esterbauer was supported by the National Foundation for Cancer Research, Bethesda, MD.

References Ames, B.N., F.D. Lee and W.E. Durston (1973) An improved bacterial test system for the detection and classification of mutagens and carcinogens, Proc. Natl. Acad. Sci. (U.S.A.), 70, 782-786. Basu, A., and L.J. Marnett (1983) Unequivocal demonstration that malondialdehyde is a mutagen, Carcinogenesis, 4, 331-333. Basu, A., and L.J. Marnett (1984) Molecular requirements for the mutagenicity of malondialdehyde and related acroleins, Cancer Res., 44, 2848-2854. Benedetti, A., M. Comporti and H. Esterbauer (1980) Identification of 4-hydroxynonenal as a cytotoxic product originating from the peroxidation of liver microsomal lipids, Biochim. Biophys. Acta, 620, 281-296. Bjeldanes, L.F., and H. Chew (1979) Mutagenicity of 1,2-dicarbonyl compounds: maltol, kojic acid, diacetyl and related substances, Mutation Res,, 67, 367-371.

33 Chung, F.L., and S.S. Hecht (1983) Formation of cyclic I,N 2adducts by reaction of deoxyguanosine with a-acetoxy-Nnitrosopyrrolidine, 4-(carbethoxynitrosamino)butanal, or crotonaldehyde, Cancer Res., 43, 1230-1235. deLumen, B.O., E.J. Stone, S.J. Kazeniac and R.H. Forsythe (1978) Formation of volatile flavor compounds in green beans from linoleic and linolenic acids, J. Food Sci., 43, 698-702. Durr, P., U. Schobinger and M. Zellweger (1981) Das Aroma von Apfelmaische bei deren Verflussigung durch Pektinasen und Zellulasen, Lebensm.-Wiss. Technol., 14(5), 268-272. Eder, E., D. Henschler and T. Neudecker (1982) Mutagenic properties of allylic and a.,O-unsaturated compounds: consideration of alkylating mechanisms. Xenobiotica, 12, 831-848. Esterbauer, H., and W. Weger (1967) Uber die Wirkung von Aldehyden auf gesund unde maligne Zellen, 3. Mitteilung: Synthese von homologen 4-Hydroxy-2-alkenalen, Monatsh. Chem., 98, 1994-2000. Esterbauer, H., K.H. Cheeseman, M.U. Dianzani, G. Poli and T.F. Slater (1982) Separation and characterization of the aldehydic products of lipid peroxidation stimulated by ADP-Fe 2~ in rat liver microsomes, Biochem. J., 208, 129-140. Galliani, G., and C. Pantarotto (1983) The reaction of gnanosine and 2'-deoxyguanosine with acrolein, Tetrahedron Lett., 24, 4491-4492. Hsieh, O.A.L., A.S. Huang and S.S. Chang (1981) Isolation and identification of objectionable volatile flavor compounds in defatted soybean flour, J. Food Sei., 47, 16-18. Jay, J.M. (1982) Antimicrobial properties of diacetyl, Appl. Environ. Microbiol., 44, 525-532. Kasai, H., K. Kumeno, Z. Yamaizumi, S. Nishimura, M. Nagao, Y. Fujita, T. Sugimura, H. Nukaya and T. Kosuge (1982) Mutagenicity of methylglyoxal in coffee, Gann, 73, 681-683. Krotoszynski, B., G. Gabriel and H. O'Neill (1977) Characterization of human expired air: a promising investigative and diagnostic technique, J. Chrom. SCi., 15, 239-244. Levin, D.E., M. Hollstein, M.F. Christman, E.A. Schwiers and B.N. Ames (1982) A new Salmonella tester strain (TA102) with A:T base pairs at the site of mutation detects oxidative mutagens, Proc. Natl. Acad. Sci. (U.S.A.), 79, 7445-7449. Lijinsky, W., and A.W. Andrews (1980) Mutagenicity of vinyl compounds in Salmonella typhimurium, Teratogen. Carcinogen. Mutagen., 1,259--267. LOfroth, G. (1978) Mutagenicity of dichloroacetaldehyde, Z. Natarforsch., 33c, 783-785. Lutz, D., E. Eder, T. Neudecker and D. Henschler (1982) Structure-mutagenicity relationship in a,,O-unsaturated carbonylic compounds and their corresponding allylic alcohols, Mutation Res., 93, 305-315. Marnett, L.J., and M.A. Tuttle (1980) Comparison of the mutagenicities of malondialdehyde and side products formed during its chemical synthesis, Cancer Res., 40, 276-282. Maron, D.B., and B.N. Ames (1983) Revised methods for the Salmonella mutagenicity test, Mutation Res., 113, 173-215. McCann, J., V. Simmon, D. Streitwieser and B.N. Ames (1975a) Mutagerticity of chloroacetaldehyde, a possible metabolic

product of 1,2-dichloroethane (ethylene dichloride), chloroethanol (ethylene chlorohydrin), vinyl chloride, and cyclophosphamide, Proc. Natl. Acad. Sci. (U.S.A.), 72. 3190-3193. McCann, J., N.E. Spingarn, J. Kobori and B.N. Ames (1975b) Detection of carcinogens as mutagens: Bacterial tester strains with R factor plasmids, Proc. Natl. Acad. Sci. (U.S.A.), 72. 979-983. Moree-Testa, P.. and Y. Saint-Jalm (1981) Determination of a-dicarbonyl compounds in cigarette smoke, J. Chromatog., 217, 197-208. Moschel, R.C., and N.J. Leonard (1976) Fluorescent modification of guanine reaction with substituted malondialdehydes, J. Org. Chem., 41, 294-300. Mukai, F.H., and B.D. Goldstein (1976) Mutagenicity of malonaldehyde, a decomposition product of peroxidized polyunsaturated fatty acids, Science, 191,868-869. National Research Council, Committee on Aldehydes, Board on Toxicology and Environmental Health Hazards, Assembly of Life Sciences (1981) Formaldehyde and other Aldehydes, National Academy Press, Washington, DC. Neudecker, T., D. Lutz, E. Eder and D. Henschler (1981) Crotonaldehyde is mutagenic in a modified Salmonella typhimurium mutagenicity testing system, Mutation Res., 91, 27-31. Perry, K.L., and G.C. Walker (1982) Identification of plasmid (pKM101) coded proteins involved in mutagenesis and UV resistance, Nature (London), 300, 278-279. Pham, H.N., D.M. DeMarini and H.E. Brockman (1979) Mutagenicity of skin tanning lotions, J. Environ. Pathol. Toxicol., 3, 227-231. Reindl, B., and H.-J. Stan (1982) Determination of volatile aldehydes in meat as 2,4-dinitrophenylhydrazones using reversed phase HPLC, J. Agric. Food Chem., 30, 849-854. Rhodes, G., M.L. Holland, D. Wiesler, M. Novotny, S.A. Moore, R.G. Peterson and D.L. Fehon (1982) Structural relationships between the endogenous volatile urinary metabolites of experimentally diabetic rats and certain neurotoxins, Experientia, 38, 75-77. Rosen, J.D., Y. Segall and J.E. Casida (1980) Mutagenic potency of haloacroleins and related compounds, Mutation Res., 78, 113-119. Sasaki, Y., and R. Endo (1978) Mutagenicity of aldehydes in Salmonella, Mutation Res., 54, 251-252. Sato, J., Y. Wang and J. van Eys (1980) Methylglyoxal formation in rat liver cells, J. Biol. Chem., 255, 2046-2050. Schauenstein, E., and H. Esterbauer (1979) Formation and properties of reactive aldehydes, Reprinted from 'Submolecular Biology and Cancer', Ciba Foundation Series 67, Excerpta Medica, Amsterdam, pp. 225-244. Schauenstein, E., H. Esterbauer and H. Zollner (Eds.) (1977) Aldehydes in Biological Systems, Pion, New York. Schreier, P., and G. Lorenz (1982) Separation, partial purification and characterization of a fatty acid hydroperoxide cleaving enzyme from apple and tomato fruits, Z. Naturforsch., 37c, 165-173. Schreier, P., F. Drawert and I. Heindze (1981) The quantitative composition of natural and technologically changed aromas of plants, VIII. Volatile constituent of fresh dill herb,

34

Anethum graveolens, L. (Umbelliferae), Lebensm.-Wiss. Technol., 14, 150-152. Shapiro, R., B.I. Cohen and D.C. Clagett (1970) Specific acylation of the guanine residues of ribonucleic acid, J. Biol. Chem., 245, 2633-2639. Slesinski, R.S., W.C. Hengler, P.J. Guzzie and K.J. Wagner (1983) Mutagenicity evaluation of glutaraldehyde in a battery of in vitro bacterial and mammalian test systems, Food Chem. Toxicol., 21,621-629. Stone, E.J., R.M. Hall and S.J. Kazeniac (1975) Formation of aldehydes and alcohols in tomato fruits from U-~4C-labelled linolenic and linoleic acids, J. Food Sci., 40, 1138-1141. Weiner. H., and B. Wermuth (Eds.) (1982) Enzymology of

Carbonyl Metabolism: Aldehyde Dehydrogenase and Aldo/Keto Reductase, Progress in Clinical and Biological Research, Vol. 114, Liss, New York. Yamaguchi, T. (1982) Mutagenicity of trioses and methylglyoxal on Salmonella typhimurium, Agric. Biol. Chem.. 46, 849-851. Zlatkis, A., C.F. Poole, R. Brazeli, D.A. Bafus and P.S. Spencer (1980) Volatile metabolites in sera of normal and diabetic patients, J. Chromatog., 182, 137-145. Zlatkis, A., R.S. Brazeli and C.F. Poole (1981) The role of organic volatile profiles in clinical diagnosis, Clin. Chem.. 27. 789-797.