- Email: [email protected]
Cytogenetic response to 1,2-dicarbonyls and hydrogen peroxide in Chinese hamster ovary AUXB1 cells and human peripheral lymphocytes
Mutation Research, 224 (1989) 269-279 Elsevier
269
MUTGEN 01478
Cytogenetic response to 1,2-dicarbonyls and hydrogen peroxide in Chinese hamster ovary AUXB1 cells and human peripheral lymphocytes James D. Tucker, Robert T. Taylor, Mari L. Christensen, Cheryl L. Strout, M. Leslie Hanna and Anthony V. Carrano Lawrence Livermore National Laboratory, Biomedical Sciences Division, University of California, P.O. Box 5507, L-452, Livermore, CA 94550 (U.S.A.) (Received 9 December 1988) (Revision received 24 April 1989) (Accepted 8 May 1989)
Keywords: 1,2-Dicarbonyls; Hydrogen peroxide; Coffee; Chinese hamster ovary AUXB1 cells; Human peripheral lymphocytes
Summary Mutagenic 1,2-dicarbonyls have been reported to occur in coffee and other beverages and in various foods. We have measured the induction of sister-chromatid exchanges (SCEs) and endoreduplicated cells (ERCs) to determine the genotoxicity of various 1,2-dicarbonyl compounds in Chinese hamster ovary (CHO) AUXB1 cells and human peripheral lymphocytes. The 1,2-dicarbonyls glyoxal, methylglyoxal and kethoxal each induced highly significant increases in both SCEs and ERCs in AUXB1 cells. Glyoxal and kethoxal induced SCEs but not ERCs in human peripheral lymphocytes. In addition, hydrogen peroxide induced highly significant levels of SCEs and ERCs in AUXB1 cells. Bisulfite, which reacts with carbonyl groups to form addition products, significantly reduced the frequency of SCEs and the proportion of ERCs when glyoxal, methylglyoxal, kethoxal and diacetyl were administered to AUXB1 ceils. In addition, bisulfite blocked the formation of ERCs, but not SCEs, induced by hydrogen peroxide. These in vitro results suggest that 1,2-dicarbonyls may play an important role in the genotoxicity of some foods and beverages.
1,2-Dicarbonyls have been shown to exist in coffee, numerous other beverages, and in many processed foods (Palo and Ilkova, 1970; Hayashi and Shibamoto, 1985; Nagao et al., 1986). Methylglyoxal is the most prevalent 1,2-dicarbonyl in brewed and instant coffee, and in bourbon whiskey (Nagao et al., 1986). Glyoxal and ethylglyoxal are
Correspondence: Dr. James D. Tucker, Lawrence Livermore National Laboratory, Biomedical Sciences Division, University of California, P.O. Box 5507, L-452, Livermore, CA 94550
(U.S.A.).
also present in these and other foods and beverages (Nagao et al., 1986). Methylglyoxal is a product of cellular metabolism (Cooper and Anderson, 1970; Sato et al., 1980), and has been shown to induce mutations in V79 cells (Cajelli et al., 1987), SCEs in Chinese hamster ovary (CHO) cells (Faggin et al., 1985), and fibrosarcomas in rats (Nagao et al., 1986). The structures of the 1,2-dicarbonyls used in this study are given in Table 1. Glyoxal is the parent compound, with methylglyoxal and diacetyl differing by the addition of one and two methyl groups, respectively. Kethoxal is an
0165-1218/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
270 TABLE 1 CHEMICAL STRUCTURES Chemical
Structure
Glyoxal
CHO-CHO
Methylglyoxal
C(CH3)O-CHO
Diacetyl
C(CH3)O-C(CH3)O
Kethoxal
CHO-CO I C H 3 - C H - O - C H 2- c n 3
various coffees in these same systems is reported elsewhere (Tucker et al., 1989). The results presented in this paper indicate that each of the 1,2-dicarbonyls tested induced SCEs and ERCs, and that bisulfite, which forms complexes with carbonyl groups, blocked a significant proportion of the 1,2-dicarbonyl genotoxicity. Bisulfite was also found to block the formation of ERCs, but not SCEs, induced by hydrogen peroxide. Materials and methods
Chemicals
ethoxy-substituted 2-ketobutyraldehyde that was originally investigated for its carcinostatic properties (French and Freedlander, 1958). Kethoxal has not been shown to be present in food or beverages, but was included as a model compound for comparative purposes. Genotoxic compounds other than 1,2-dicarbonyls are also found in coffee. Hydrogen peroxide (H202) is generated in coffee beginning soon after preparing the beverage, and its production continues for at least 24 h (Fujita et al., 1985; Nagao et al., 1986; Rinkus and Taylor, 1989). It has been shown to induce mutations (Fujita et al., 1985; Speit, 1986), SCEs (Speit et al., 1982; Estervig and Wang, 1984; Mehnert et al., 1984; Speit, 1986) and possibly chromosome aberrations (Sofuni and Ishidate, 1984; Ishidate et al., 1988), and was found to have increased Salmonella mutagenic activity in the presence of methylglyoxal (Fujita et al., 1985). The genotoxicity of active oxygen species in mammalian cells recently has been reviewed (Imlay and Linn, 1988; Meneghini, 1988). Due to the presence of 1,2-dicarbonyl compounds in both food and beverages, and to hydrogen peroxide in coffee, we have undertaken an investigation of two cytogenetic effects of these compounds in mammalian cells. This paper reports the results of our studies with four different 1,2-dicarbonyls in CHO AUXB1 cells using SCEs and endoreduplicated cells (ERCs) as the endpoints. Two of the 1,2-dicarbonyls were also studied in human peripheral lymphocytes. Work done in parallel with this study on the effects of
Glyoxal (CAS No. 107-22-2) was purchased from Aldrich, methylglyoxal (CAS No. 78-98-8) from Sigma, kethoxal (CAS No. 3688-37-8) from U.S. Biochemical Corporation, diacetyl (CAS No. 431-03-8) from Eastman, hydrogen peroxide (CAS No. 7722-84-1) from Mallinckrodt, sodium bisulfite (CAS No. 7631-90-5) from Baker, and mitomycin C (MMC, CAS No. 50-07-7) from CalBiochem. Hydrogen peroxide was dissolved in phosphate-buffered saline, kethoxal was initially dissolved in a small amount of ethanol prior to being dissolved in distilled water, sodium bisulfite was dissolved in water and adjusted to pH 7.2, and the remaining chemicals were dissolved in distilled water. Chinese hamster ovary (CHO) AUXB1 cells, triple-auxotrophic for glycine, adenosine and thymidine (Taylor et al., 1977) were grown as described previously (Taylor et al., 1985a, b), with the following modifications. Exposure times differed according to the endpoint being examined, and were 15, and 20-22 hours, for ERCs and SCEs, respectively. This difference was necessary at the lower doses because of the need to obtain predominantly first-division cells for ERCs and second-division cells for SCEs. Cells were exposed to the test chemicals in foil-covered 25 × 150 mm glass culture tubes in 40 ml of medium containing 0.67 mM glycine and 37/~M adenosine and were incubated on a roller wheel at 37 o C. For sisterchromatid differentiation, the amount of thymidine was reduced to 10 /xM and bromodeoxyuridine (BrdUrd) was incorporated at a concentration of 30 #M. When cells were exposed to more than one chemical, for example bisulfite and the test agent, the former was incorporated into the
271 cultures and the test compound was added immediately thereafter. Following exposure, the cells were collected by centrifugation, resuspended into fresh glycine-adenosine-supplemented medium containing 30/~M BrdUrd plus 10/~M thymidine, but lacking the test agent(s). They were then plated for cell survival (Taylor, 1985a) and aliquoted in 12-ml volumes into foil-covered Coming 16 x 125 mm sterile disposable culture tubes for continued incubation at 3 7 ° C on a roller wheel. Harvest times were determined from previous experiments designed to measure growth kinetics following exposure to each dose. Harvest times varied from 25 to 55 h following the beginning of exposure to capture second-division cells for SCEs, and from 20 to 45 h to capture first-division cells for endoreduplication. 5 h prior to harvest, colcemid was added at a final concentration of 0.1/~g/ml. The cells were then centrifuged, the supernatant removed, and 0.075 M KC1 was added for 30 min. Freshly prepared fixative (methanol:acetic acid, 3 : 1 ) was added, the cells centrifuged again, and then fixed an additional 3 times. Cells were dropped onto clean dry slides and stained by the fluorescence plus Giemsa method as described (Perry and Wolff, 1974; Minkler et al., 1978). For SCEs, 50 second-division cells (25 from each of two consecutive harvest times) were scored unless otherwise noted. For endoreduplicated cells, 1000 first-division cells from a single harvest time were scored.
Human peripheral lymphocytes Peripheral blood from a healthy, nonsmoking female was drawn into a heparinized tube and cultured in RPMI-1640 with 25 mM Hepes supplemented with 16% fetal calf serum, L-glutamine (final concentration 2 mM), 1% penicillin-streptomycin (Gibco), 1% phytohemagglutinin (M form, Gibco), and 25 /~M BrdUrd. The cultures were grown in the dark in 5% CO 2 at 3 7 ° C for 52 h (ERCs) or 72 h (SCEs). Two cultures for each chemical, dose, and harvest time were employed. Colcemid (0.1 # g / m l ) was present for the final 4 h. The slides were prepared and stained as described for the AUXB1 cells. The scoring for both cell types was done on coded slides. For SCEs, 25 cells were scored from each replicate culture, for a
total of 50 cells from each dose point. For ERCs, at least 500 cells were scored from a single culture.
Statistical methods The SCEs were normalized to 21 or 46 chromosomes per cell for the C H O AUXB1 and human cells, respectively. Student's t-test and Fisher's exact test were used to compare the SCE and ERC frequencies, respectively. One-tailed comparisons were made in each case. Results
The SCE and ERC results for glyoxal, methylglyoxal, kethoxal and hydrogen peroxide in AUXB1 cells are presented :in Table 2. A first-division endoreduplicated AUXB1 cell is illustrated in Fig. 1. Each of these chemicals produced highly significant increases in both endpoints. For the SCEs, treatment with each agent was dose-dependent. Kethoxal and glyoxal produced a doubling and a tripling of the baseline frequency, respectively, while methylglyoxal and hydrogen peroxide nearly doubled the baseline. Glyoxal produced the largest increase in SCEs while hydrogen peroxide induced a significant increase at the lowest dose. The relative percent cell survivals are also given. Significant cell cycle delays were obtained in every case. The ERC response to these chemicals is also highly significant. Dose responses were obtained with each agent, although toxicity was apparent at the highest doses in some cases. The weakest response was produced by hydrogen peroxide, and the strongest response was produced by kethoxal. Kethoxal is the most potent in terms of the induced ERC response, being active at 75 #M. The lowest significant doses for glyoxal, methylglyoxal and hydrogen peroxide were 200/tM, 100/~M and 160/tM, resp. To determine whether these responses were typical of mammalian cells in general, or were unique to AUXB1 cells, we tested glyoxal and kethoxal in human peripheral lymphocytes. These results are presented in Table 3. Glyoxal produced small but significant SCE increases at every dose tested but did not yield a dose response. However, the results did produce a positive trend test ( p < 0.05). Kethoxal produced a linear SCE response
0 100' 200 300 400 500
0 200 400 600 800 1000 1 200 1 400 1 600
Glyoxal
Methylglyoxal
Dose (/tM)
Chemical
12.2 14.4 17.9 17.8 20.9 21.6
10.6 15.8 19.0 19.4 23.7 24.8 25.4 28.0 30.9
Mean/ cell
1.01 1.22 1.58 1.64 1.57 1.75
0.35 0.57 0.77 0.68 0.90 0.92 0.75 0.80 1.28
Std. Err.
50 50 50 50 50 50
100 50 50 50 50 50 50 50 50
Nu mbe r of cells scored
Sister-chromatid exchanges
0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
N.S. e 0.002 0.002 < 0.0001 < 0.0001
< < < < < < < <
Probability
29, 30, 30, 35, 40, 48,
25, 30, 30, 30, 35, 35, 40, 40, 50, 34 35 35 40 45 53
30 35 35 35 40 40 45 45 55
Harvest times (b)
a
2.09 1.94 1.94 1.76 1.90 1.66 1.71 1.27 1.56 2.28 2.01 1.99 1.93 1.82 1.64
CKI
1.90 1.51 1.80 1.40 1.45 1.09 1.20 1.03 1.37 2.04, 2.00, 1.64, 1.48, 1.61, 1.41,
Std. Dev. 1.7 4.0 5.0 5.3 6.8 5.0 5.9 5.7
2.2 4.2 3.8 4.8 4.9 6.4
ERCs/ 1000 cells 1.5 b 16 26 29 48 26 36 34 _ c 5 18 15 24 25 43
0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
0.005 0.02 0.0003 0.0002 < 0.0001
< < < < < < <
Probability
Endoreduplicated cells
20 20 25 30 35 40
25 30 30 30 35 35 40 40 50
Harvest time (h)
1.11 1.02 1.00 1.00 1.03 1.09
1.90 1.51 1.80 1.40 1.45 1.09 1.20 1.03 1.37
CK I a
100+/--3 97 + / -- 3 92 + / -- 3 83 + / -- 3 61 + / -- 2 31 + / -- 1
100+/--4 d 100+/--4 100+/--4 100+/--5 91 + / -- 3 50 + / -- 4 28 + / -- 2 9.4 + / -- 0.5 0.9+/--0.2
Relative percent survival
S I S T E R - C H R O M A T I D E X C H A N G E S A N D E N D O R E D U P L I C A T E D CELLS I N D U C E D BY G L Y O X A L , K E T H O X A L , M E T H Y L G L Y O X A L A N D H Y D R O G E N P E R O X I D E 1N C H O AUXB1 CELLS
TABLE 2
bJ
b c a e f g
a
0 40 80 120 160 200 240
50 75 100 125 150 175 200 225 250 275 300
11.0 14.1 14.9 14.9 17.3 17.5 19.6
0 11.5 12.7 14.8 15.2 16.4 17.7 20.8 _ t 24.9 _ t 19.5 25
1.75 50 50 50 50 50 50 50
25
0.98
0.44 0.85 0.75 0.73 0.87 1.15 1.56
0.52 50 50 50 50 25 50 25
11.2 0.54 0.59 0.59 0.78 0.72 0.66 0.75
< < < < <
0.001 0.0001 0.0001 0.0001 0.0001 0.0001
< 0.0001
< 0.0001
50 N.S. 0.03 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001
25, 30 25, 30 30, 35 35, 40 35, 40 40, 45 40, 45
25, 30 25, 30 30, 35 30, 35 30, 35 35, 40 35, 40 35, 40 40, 45 40, 45 45, 50 1.77, 1.76, 1.98, 1.98, 1.98, 1.97, 1.94,
2.00 2.00 2.08 2.46 2.13 2.61 2.36
25, 30 1.77, 2.00 1.39, 1.82 1.45, 1.95 1.16, 1.74 1.04, 1.37 1.07, 1.52 1.00, 1.11 1.01, 1.08 1.01, 1.06 1.00, 1.01 1.00, 1.02 1 6 2 5 16 24 39
1.78, 1.97 9 32 22 42 49 52 85 81 g 60 g 98 g 101 g 1.0 2.4 1.4 2.2 4.0 4.8 6.1
6 3.0 5.6 4.6 6.3 6.8 7.0 8.8 12.2 10.6 13.3 13.5
N.S. N.S. N.S. 0.0002 < 0.0001 < 0.0001
2.4 N.S. 0.0002 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001
CKI, cell kinetics index = {number of M1 cells+2(number of M2 cells)+3(number of M3 cells))/total cells scored. 2000 cells scored. Insufficient first-division cells. Ranges represent the larger of the standard errors of the mean, or the counting error (Taylor et al., 1985). N.S., not significant. Insufficient second-division cells. Mean of 2 separate Expts. 1000 cells scored from each.
Hydrogen peroxide
Kethoxal
20 20 25 30 30 35 35
25 25 30 30 30 35 35 35, 35 45, 40 45,45 50, 50 1.24 1.10 1.49 1.93 1.81 1.92 1.85
100+/-4 96+/-4 88 + / 67 + / 51 + / 42 + / 23 + / -
4 3 2 3 1
25 1.78100+/-4 1.77 95 + / - 4 1.39 100+/-4 1.45 100+/-7 1.16 78+/-4 1.04 57 + / - 4 1.07 65 + / - 5 1.00 41 + / - 3 1.01, 1.02 44 + / - 3 1.06, 1.02 35 + / - 2 1.01,1.03 1 2 + / - 1 1.02, 1.14 7.4+/-0.5
274
t
experiment because sodium bisulfite is also a nucleophile for H202 (Ruggero and Edwards, 1970). The SCE and ERC results, along with the relative percent cell survivals, are presented in Table 4. Bisulfite (1.0 mM) blocked much, but not all, of the 1,2-dicarbonyl genotoxicity as measured by SCEs and ERCs. It also consistently enhanced the cell survival of cytotoxic concentrations of both the 1,2-dicarbonyls and H202. The number of SCEs was reduced dramatically in cultures treated with each of the four 1,2-dicarbonyls, although the frequencies remained elevated compared to the controls in most cases. For 1,2-dicarbonyls, the average percentage of ERCs blocked by bisulfite was greater than the mean percentage of SCEs blocked (89% vs. 69%). Bisulfite, however, did not prevent the formation of SCEs induced by H202, but did block a significant fraction of the ERCs at the higher dose tested. This may simply reflect the higher levels of H202 that are needed to
Fig. 1. Endoreduplicated AUXB1 cell with 21 diplochromosomes induced by 400/~ M methylglyoxal treatment. TABLE 3
but did not achieve a doubling of the baseline. Neither of these agents induced ERCs at any of the doses tested (data not shown). 1,2-Dicarbonyls are known to react reversibly with bisulfite to form addition products (Shapiro, 1977). These reactions eliminate the carbonyl moiety, as shown in the following reaction equilibria:
SISTER-CHROMATID EXCHANGES INDUCED BY G L Y O X A L A N D K E T H O X A L IN H U M A N P E R I P H E R A L LYMPHOCYTES a Chemical
Dose (/~ M)
SCEs/cell
Std. Err.
Probability
Glyoxal
0 400 700 1 000 1300 1600 1900 2200 2500 2800
9.3 10.9 12.8 11.9 12.7 12.5 13.7 11.4 12.0 14.3
0.69 0.64 0.67 0.72 0.77 0.71 0.83 0.68 0.63 0.86
0.05 < 0.0001 0.005 0.003 0.001 < 0.0001 0.02 0.002 < 0.0001
0 1 000 1 400 1 800 2200 2600 3000 3 400
10.7 11.1 12.7 13.0 13.7 15.4 16.9 _ o
0.62 0.70 0.79 0.66 0.70 0.58 0.72
N.S. c 0.03 0.01 0.001 < 0.0001 < 0.0001
R1CO-R2CO + NaHSO 3 R1CO-R2C(OH)SO3Na R I C O - R 2 C ( O H ) S O 3 N a + NaHSO 3 R ~ C ( O H ) S O 3 N a - R 2C(OH)SO3Na We made use of these reactions to determine if the 1,2-dicarbonyl functional groups are indeed responsible for the observed genotoxicity. AUXB1 cells were exposed to each of 4 different 1,2-dicarbonyls (glyoxal, methylglyoxal, kethoxal or diacetyl) and hydrogen peroxide at 2 different doses in the presence or absence of 1 mM sodium bisulfite. Hydrogen peroxide was included in this
Kethoxal
CKI b 2.33 1.89 1.80 1.66 1.45 1.37 1.66 1.68 1.33 1.41 2.07 1.85 1.87 1.55 1.36 1.22 1.10 1.02
50 cells scored per dose point. b CKI, cell kinetics index = {number of M1 c e l l s + 2 ( n u m b e r of M2 cells)+ 3(number of M3 c e l l s ) } / t o t a l cells scored. c N.S., not significant. d Insufficient second-division cells for analysis.
+ +
+ +
+ +
250 250 500 500
K e t h o x a l f 125 125 250 250
125 125 250 250
0 0 100 100 200 200
Methylglyoxal
Diacetyl
Hydrogen peroxide
c c a e r g
a
+ +
600 600 1200 1200
Glyoxal
11.60 10.35 13.74 13.46 16.29 18.00
14.8 10.1 16.6 13.3
13.3 10.7 20.8 9.0
17.3 11.6 23.0 e 13.8
18.3 11.9 23.8 12.7
10.1 9.3 N.S. 'j
-
bisulf.
Culture without
0.51 0.42 N.S. 0.59 < 0 . 0 0 0 1 0.77 0.0006 0.60 < 0.0001 0.86 < 0.0001
64.8
68.3
75.3
68.0
N.S. N.S. 0
0
-
39.2
82.7
20, 20, 25, 20, 30, 25,
30, 25, 50, 25,
30, 25, 40, 25,
25 25 30 25 35 30
35 30 55 30
35 30 45 30
30, 35 25, 30 50, 55 25,30
30, 35 25, 30 40, 45 25,30
25, 30 25, 30
1.71, 1.78, 1.94, 1.92, 1.94, 1.93,
1.31, 1.46, 1.95, 1.07,
1.06, 1.95, 1.00, 1.97,
1.99 1.99 1.99 1.99 2.00 1.87
1.79 1.92 1.99 1.88
1.86 2.00 1.03 2.00
1.04, 1.17 1.82, 2.00 1.53, 1.58 1.07,1.58
1.20, 1.73 1.73, 1.97 1.12, 1.99 1.21,1.48
1.78, 2.01 1.72, 1.92
Harvest CK1 b time
SCEs (h) blocked a
0.0002 5 7 . 0 0.0001 100.0
0.78 < 0.0001 0.47 N.S. < 0.0001 0.89 < 0.0001 0.61 < 0.0001 0.001
0.55 < 0.0001 0.46 0.03 1.31 < 0.0001 0.37 N.S. <
0.73 < 0.0001 0.44 0.001 < 0.0001 1.01 < 0.0001 0.69 < 0 . 0 0 0 1 < 0 . 0 0 0 1
0.68 < 0.0001 0.59 0.001 < 0.0001 0.85 < 0.0001 0.64 < 0 . 0 0 0 1 < 0 . 0 0 0 1
0.49 0.58
Control
Percent induced
10 1 15 10 38 11
7 1 0 1
36 10 44 2
36 4 77 18
15 3 62 6
0 1
cells
ERCs/ 1000
3.1 1.0 3.8 3.1 6.0 3.3
2.6 1.0 0.0 1.0
5.9 3.1 6.5 1.4
5.9 2.0 1.7 4.2
3.8 1.7 7.6 2.4
0.0 1.0
1000 1000 1000 1000 1000 1000
1000 1000 1000 1000
1000 1000 1000 1000
1000 1000 39 1000
1000 1000 1000 1000
1000 1 000
64.0
0
1.01
1.00
20
25
1.04
1.01
20
20
1.30 1.24
1.07
1.95
20 20
50
1.31 1.46
30 25 25
1.03
1.83
20
40
1.05
1.14
25
20
N.S.
100.0
97.7
75.0
1.07
1.53
50
25
1.82
1.04
25
30
1.20 1.73 1.12 1.21
30 25 40 25
-
0.04
-
0.006 N.S. N.S. N.S. < 0.0001 < 0.0001 N.S.
0.008 N.S. N.S. N.S.
< 0.0001 < 0.0001 0.02 < 0.0001 <0.0001 N.S.
77.9
91.7
91.9
< 0.0001
-
86.7
1.78 1.72
25 25
2
5
100+/-4 106+/-4 73 + / - 2 88 + / - 3 20 + / - 1 47 + / - 1
117+/-8 97 + / - 4 42 + / - 2 75 + / - 4
107 + / 95+/-4 61 + / 95+/-4
98 + / - 4 20 + / - 2 120+/-4
102 + / - 5
103+/-4
31+/-2
93 + / - 5 115+/-5
100+/-4 103+/-4
Percent Harvest CKI b Relative induced time percent ERCs (h) survival blocked a
0.004
-
Culture without bisulf.
< 0.0001 < 0.0001 N.S. < 0.0001 < 0 . 0 0 0 1 < 0.0001
< 0.0001 N.S. < 0.0001 N.S.
N.S.
Number Control of cells scored
Std. P r o b a b i l i t y Dev. c o m p a r e d to:
IN AUXB1 CELLS
g
Percent of induced SCEs or E R C s b l o c k e d by bisulfite = { ( n u m b e r induced by c h e m i c a l - n u m b e r i n d u c e d b y c h e m i c a l a n d bisulfite t o g e t h e r ) / n u m b e r i n d u c e d b y chemical)} × 100. C K I , cell kinetics index = { n u m b e r of M1 cells + 2 ( n u m b e r of M 2 c e l l s ) + 3 ( n u m b e r of M3 c e l l s ) } / t o t a l cells scored. C o n t r o l s for glyoxal, methylglyoxal, k e t h o x a l a n d diacetyl data. N.S., not significant. First harvest time had only 8 scoreable cells. C o n t r o l values for E R C d a t a are 7 (no t r e a t m e n t ) a n d 2 (1 m M bisulfite), with 1000 cells scored in each case. R a n g e s represent the larger of the s t a n d a r d errors of the mean, or the c o u n t i n g error ( T a y l o r et al., 1985).
+ + +
+
(1 (mM)
Dose Bisul- S C E s / Std. P r o b a b i l i t y ( # M ) rite cell Err. c o m p a r e d to:
None c (control)
Chemical
E F F E C T S O F S O D I U M B I S U L F 1 T E O N SCE A N D E R C I N D U C T I O N
TABLE 4
%n
276 TABLE 5 SCEs A N D ERCs IN AUXB1 CELLS T R E A T E D W I T H M I T O M Y C I N C Chemical
Dose
Sister-chromatid exchanges
Endoreduplicated cells
Relative percent survival
Mean/cell
Std. error
Number of cells scored
ERCs/ 1000
Std. Deviation
N u m b e r of cells scored
0 40 nM
10.6 46.0
0.35 1.30
100 50
1.5 2
1.7 1.4
2000 1000
1004/-5 ~ 57 + / - 4
0 80 nM
11.8 58.8
0.35 1.64
100 50
2 1
2.0 1.0
2000 1000
100 + / - 5 42 + / - 3
Expt. 1 MMC
Expt. 2 MMC
a Ranges represent the larger of the standard errors of the mean, or the counting error (Taylor et al., 1985).
induce ERCs, as opposed to SCEs in AUXB1 cells (Table 2) and the inability of 1.0 mM bisulfite to d e s t r o y H 2 0 2 quantitatively and rapidly when both are together in a complex culture medium. It still seemed remotely possible that the formation of SCEs and ERCs might be tightly coupled in AUXB1 cells, such that agents inducing one endpoint would always also induce the other. Alternatively, ERCs might only be formed by a subset of those chemicals that are capable of inducing SCEs in these cells, e.g., 1,2-dicarbonyls. To determine if other chemicals that generate SCEs also induce ERCs in AUXB1 cells, we tested mitomycin C (MMC), a potent inducer of SCEs (Latt et al., 1981), that lacks a 1,2-dicarbonyl functional group. The results are presented in Table 5. It is clear that MMC induced many SCEs but no ERCs in these cells, indicating that the induction of ERCs is not an SCE-coupled general phenomenon in AUXB1 cells. This finding is in agreement with the results seen in the human peripheral lymphocytes, where SCEs but not ERCs were induced by 1,2-dicarbonyls. Moreover, as noted in Table 4, bisulfite blocked the formation of ERCs, but not SCEs, induced by H202, lending further support to the idea that these endpoints are not necessarily coupled in this cultured hamster cell line. Discussion
The results presented here indicate that the 1,2-dicarbonyls tested in these studies are potent
inducers of SCEs in both C H O AUXB1 cells and in human peripheral lymphocytes, and that they also induce high levels of ERCs in AUXB1 cells. These findings are of interest because 1,2-dicarbonyls, particularly glyoxal, methylglyoxal and diacetyl, are found in coffee and other foodstuffs. 1 g of instant coffee has been reported to contain 100 /ag of glyoxal, 100-210 /ag of methylglyoxal, and 100/ag of diacetyl, while 8 g of coffee beans have been reported to yield 470-730/ag of methylglyoxal per 100 ml of brewed coffee (Sugimura and Sato, 1983; Nagao et al., 1983; Kasai et al., 1982). The latter translates into 65 /aM methylglyoxal, while 200/aM and 100 laM are the lowest concentrations producing a significant increase of SCEs and ERCs, respectively, in this in vitro study. These observations suggest that the amount of methylglyoxal present in Japanese prepared coffees is too low to be responsible for our measurable increases in genotoxicity. This conclusion is also supported by other work in our laboratory (Tucker et al., 1989) where several types of coffee were observed to induce SCEs at concentrations ranging from approximately 1% to 10% of the full strength preparation. If we assume that our coffee is similar to that prepared by the Japanese, then the concentration of methylglyoxal would range from 0.65 /aM to 6.5 /xM, which is considerably below the lowest level found to induce SCEs (Table 2). Irrespective, the data in Tables 2 4 stress the importance of understanding the role of 1,2dicarbonyls in a variety of foods and beverages, including coffees. Our results, along with the above
277 and other published data, suggest that people may be exposed to significant amounts of genotoxic 1,2-dicarbonyls through their diet. Nakasato et al. (1984) demonstrated that methylglyoxal induces mutations in cultured Chinese hamster lung cells, but the influence of $9 was not examined. $9 virtually eliminates the mutagenicity of coffee in vitro (Nagao et al., 1984; Aeschbacher et al., 1985; Friederich et al., 1985). However, methylglyoxal retains slightly more than half its genetic activity in Salmonella strain TA100 when preincubation is carried out in the presence of rat-liver $9 (Friederich et al., 1985). Bronzetti et al. (1977) showed that $9 had little effect upon the mutagenicity of methylglyoxal in Salmonella strain TA97. However, these authors did observe a decrease due to $9 in the amount of gene conversion and in the number of mutations induced in Saccharomyces cerevisiae. Shane et al. (1988) examined the effects of glyoxal, methylglyoxal and diacetyl in 3 different Salmonella strains with and without $9. In some cases it was observed to decrease the mutagenic potency, while in other situations, no effect, or even increases were seen. The relative genotoxicity of the 1,2-dicarbonyls, as measured with SCEs and ERCs and calculated with linear regression analysis using the linear portion of the dose-response curves, is kethoxal > methylglyoxal > glyoxal, with hydrogen peroxide intermediate to kethoxal and methylglyoxal. In the Salmonella assay, methylglyoxal was also more potent than glyoxal (Sugimura and Sato, 1983). This agrees with our findings, although the differences between glyoxal and methylglyoxal in the SCE and ERC assays were not large. The concentration of hydrogen peroxide in coffee increases with time, and interacts synergistically with methylglyoxal to produce mutations in Salmonella strain TA100 (Fujita et al., 1985). Thus, the role of hydrogen peroxide in the genotoxicity of coffee probably depends upon several variables. The data presented here show that hydrogen peroxide by itself induces both SCEs and ERCs in AUXB1 cells. The mechanism for SCE induction by hydrogen peroxide was recently elucidated by Larramendy et al. (1987) who showed that the hydroxyl radical O H ' i s the intermediate responsible for this response. The reaction producing the OH" requires Fe 2+ as a catalyst. This indicates
that the concentration of Fe 2÷ in culture medium a n d / o r a beverage could influence the observed SCE frequency. Consequently, the concentration of Fe 2+ in coffee may at least partially determine the SCE response to coffee. Bisulfite prevents the formation of the majority of SCEs and ERCs induced by 1,2-dicarbonyls, but does not block SCE formation by H202 (Table 4). This argues in favor of the dicarbonyl group as being the structure responsible for the observed genotoxicity. Although cultures treated with both bisulfite and a 1,2-dicarbonyl still exhibited elevated SCEs and ERCs, the level was always significantly reduced compared to the effects of the 1,2-dicarbonyl alone. One would not necessarily expect the SCE or ERC frequency to be completely reduced to baseline levels due to competing reactions of certain culture medium components (e.g. pyruvate) for N a H S O 3 and the reversibility of NaHSO 3 addition reactions with carbonyl groups. In the experiments described here, we added kethoxal, methylglyoxal, or diacetyl to cultures containing a 2-8-fold molar excess of pre-added bisulfite. For cultures containing glyoxal, bisulfite was not added in excess because concentrations above 1.0 mlVL markedly inhibit AUXB1 suspension cell growth. However, cultures containing 0.6 or 1.2 mM glyoxal plus 0.1 mM NaHSO 3 showed comparable amounts of blockage of SCEs and ERCs. Another possibility for the incomplete blockage of genotoxicity rests on the observation that aldehydes induce SCEs in mammalian cells (Obe and Ristow, 1977; Ristow and Obe, 1978; Jansson, 1982). 1,2-Dicarbonyls with a single carbonyl group blocked are essentially aldehydes or ketones, and only a single carbonyl functional group may be required for at least some of their observed genotoxicity. Suwa et al. (1982) obtained nearly complete inactivation of the mutagenic activity of glyoxal and diacetyl in Salmonella TA100 with 3 molar equivalents of sodium sulfite, but these assays are performed in much simpler media. The evidence presented here indicates that additional factors in the genotoxicity of 1,2-dicarbonyls remain to be elucidated. Schwarzacher and Schnedl (1965) proposed that endoreduplication occurs when cells proceed through G2 directly to a second S phase without going through M and G1. This results in diplo-
278
chromosomes, and the cells will be tetraploid following another mitotic division. Huang et al. (1983) suggest that endoreduplication may result from the arrest of normal DNA replication. Sutou and Arai (1975) have suggested that alterations in cell surfaces, the cytoskeleton, or both, may be responsible for ERC formation. Whether 1,2-dicarbonyls or other agents known to induce ERCs interact with either the cell surface or the cytoskeleton is unknown. Although the precise mechanism of ERC formation is unknown, the chemical induction of endoreduplication has been explored in some detail. ERCs can form in the absence of chemical exposure, and numerous agents have been observed to increase this frequency (Sutou and Arai, 1975, and refs. therein). Our results demonstrate that 1,2-dicarbonyls can induce high levels of endoreduplication, at least in AUXB1 cells. Significant induction of ERCs by 1,2-dicarbonyls was also observed in CHO-S cells, the parental line (data not shown). The frequencies of ERCs reported here are similar to those observed for other chemicals (Sutou, 1981), with kethoxal appearing to be the most potent inducer of this endpoint. In light of our negative results with MMC in AUXB1 cells, and our negative results with 1,2-dicarbonyls in human peripheral lymphocytes, it is interesting that Takanari and Izutsu (1983) observed increases in the ERC frequency in human tonsillar lymphocytes following MMC treatment. The reason for this difference is not clear. The results presented in this paper indicate that 1,2-dicarbonyls and hydrogen peroxide, both constituents in coffee, induce significant levels of SCEs and ERCs in vitro. Whether significant genotoxic effects can be induced by these agents in vivo when they are administered either individually or in combination remains to be seen. Further work is needed to answer this question.
Acknowledgments The expert technical assistance of J. Lamerdin is gratefully appreciated. Work performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under contract number W-7405-ENG-48, with sup-
port from PHS grant No. CA40816 awarded to R.T.T. by the National Cancer Institute, DHHS.
References Aeschbacher, H.U., E. Ruch, H. Meier, H.O. Wurzner and R. Munoz-Box (1985) Instant and brewed coffees in the in vitro human lymphocyte mutagenicity test, Fd. Chem. Toxicol., 23, 747-752. Bronzetti, G., C. Corsi, D. Del Chiaro, P. Boccardo, R. Vellosi, F. Rossi, M. Paolini and G. Cantelli Forti (1987) Methylglyoxal: genotoxicity studies and its effect in vivo on the hepatic microsomal mono-oxygenase system of the mouse, Mutagenesis, 2, 275-277. Cajelli, E., R. Canonero, A. Martelli and G. Brambilla (1987) Methylglyoxal-induced mutation to 6-thioguanine resistance in V79 cells, Mutation Res., 190, 47-50. Cooper, R.A., and A. Anderson (1970) The formation and catabolism of methylglyoxal during glycolysis in E. coli, FEBS Lett., 11,273-276. Estervig, D., and R.J. Wang (1984) Sister chromatid exchanges and chromosome aberrations in human cells induced by H202 and other photoproducts generated in fluorescent light-exposed medium, Photochem. Photobiol., 40, 333-336. Faggin, P., A.M. Bassi, R. Finollo and G. Brambilla (1985) Induction of sister-chromatid exchanges in Chinese hamster ovary cells by the biotic ketoaldehyde methylglyoxal, Mutation Res., 144, 189-191. French, F.A., and B.L. Freedlander (1958) Carcinostatic action of polycarbonyl compounds and their derivatives, I. 3Ethoxy-2-ketobutyraldehyde and related compounds, Cancer Res., 18, 172-175. Friederich, U., D. Hann, S. Albertini, C. Schlatter and F.E. Wiirgler (1985) Mutagenicity studies on coffee, The influence of different factors on the mutagenic activity in the Salmonella/mammalian microsome assay, Mutation Res., 156, 39-52. Fujita, Y., K. Wakabayashi, M. Nagao and T. Sugimura (1985) Implication of hydrogen peroxide in the mutagenicity of coffee, Mutation Res., 144, 227-230. Hayashi, T., and T. Shibamoto (1985) Analysis of methyl glyoxal in foods and beverages, J. Agric. Food Chem., 33, 1090-1093. Huang, Y., C.-c. Chang and J.E. Trosko (1983) Aphidocolininduced endoreduplication in Chinese hamster cells, Cancer Res., 43, 1361-1364. Imlay, J.A., and S. Linn (1988) DNA damage and oxygen radical toxicity, Science, 240, 1302-1309. Ishidate Jr., M., M.C. Harnois and T. Sofuni (1988) A comparative analysis of data on the clastogenicity at 951 chemical substances tested in mammalian cell cultures, Mutation Res., 195, 151-213. Jansson, T. (1982) The frequency of sister chromatid exchanges in human lymphocytes treated with ethanol and acetaldehyde, Hereditas, 97, 301-303. Kasai, H., K. Kumeno, Z. Yamaizumi, S. Nishimura, M. Nagao, Y. Fujita, T. Sugimura, H. Nukaya and K. Takuo
279 (1982) Mutagenicity of methylglyoxal in coffee, Gann, 73, 681-683. Larramendy, M., A.C. Mello-Filho, E.A. Leme Martins and R. Meneghini (1987) Iron-mediated induction of sister-chromatid exchanges by hydrogen peroxide and superoxide anion, Mutation Res., 178, 57-63. Latt, S.A., J. Allen, S.E. Bloom, A.V. Carrano, E. Falke, D. Kram, E. Schneider, R. Schreck, R. Tice, B. Whitfield and S. Wolff (1981) Sister-chromatid exchanges: A report of the Gene-tox program, Mutation Res., 87, 17-62. Mehnert, K., W. Vogel, R. Benz and G. Speit (1984) Different effects of-mutagens on sister chromatid exchange induction in three Chinese hamster cell lines, Environ. Mutagen., 6, 573-583. Meneghini, R. (1988) Genotoxicity of active oxygen species in mammalian cells, Mutation Res., 195, 215-230. Minkler, J., D. Stetka Jr. and A.V. Carrano (1978) An ultraviolet light source for consistent differential stainng of sister chromatids, Stain Technol., 53, 359-360. Nagao, M., S. Sato and T. Sugimura (1983) in: G.R. Waller and M.S. Frather (Eds.), The Maillard Reaction in Foods and Nutrition, ACS Symp. Ser. 215, Ch. 28, pp. 521-536, American Chemical Society, Washington. Nagao, M., Y. Suwa, H. Yoshizumi and T. Sugimura (1984) Mutagens in coffee, Banbury Report, 17, 69-77. Nagao, M., Y. Fujita, K. Wakabayashi, H. Nukaya, T. Kosuge and T. Sugimura (1986) Mutagens in coffee and other beverages, Environ. Hlth. Persp., 67, 89-91. Nakasato, F., M. Nakayasu, Y. Fujita, M. Nagao, M. Terada and T. Sugimura (1984) Mutagenicity of instant coffee on cultured Chinese hamster lung cells, Mutation Res., 141, 109-112. Obe, G., and H. Ristow (1977) Acetaldehyde, but not ethanol induces sister chromatid exchanges in Chinese hamster cells in vitro, Mutation Res., 56, 211-213. Palo, V., and H. llkova (1970) Direct gas chromatographic estimation of lower alcohols, acetaldehyde, acetone and diacetyl in milk products, J. Chromatog., 53, 363-367. Perry, P., and S. Wolff (1974) New Giemsa method for the differential staining of sister chromatids, Nature (London), 251, 156-158. Rinkus, S.J., and R.T. Taylor (1989) Analysis of hydrogen peroxide in freshly prepared coffees, in preparation. Ristow, H., and G. Obe (1978) Acetaldehyde induces crosslinks in DNA and causes sister chromatid exchanges in human cells, Mutation Res., 58, 115-119. Ruggero, C., and J.U. Edwards (1970) Peroxide reaction mechanisms - - polar, in: D. Swern (Ed.), Organ Peroxides, Vol. 1, Wiley-Interscience, New York, Ch. 4, pp. 199-264. Sato, J., W. Yeu-Ming and J. Van Tys (1980) Methylglyoxal formation in rat liver cells, J. Biol. Chem., 255, 2046-2050.
Schwarzacher, H.G., and W. Schnedl (1965) Endoreduplication in human fibroblast cultures, Cytogenetics, 4, 1-18. Shane, B.S., A.M. Troxclair, D.J. McMillin and C.B. Henry (1988) Comparative mutagenicity of nine brands of coffee to Salmonella typhimurium TA100, TA102, and TA104, Environ. Mol. Mutagen., 11, 195-206. Shapiro, R. (1977) Genetic effects of bisulfite (sulfur dioxide), Mutation Res., 39, 149-176. Sofuni, T., and M. Ishidate Jr. (1984) Induction of chromosomal aberrations in cultured Chinese hamster cells in a superoxide-generating system, Mutation Res., 140, 27-31. Speit, G. (1986) The relationship between the induction of SCEs and mutations in Chinese hamster cells, I. Experiments with hydrogen peroxide and caffeine, Mutation Res., 174, 21-26. Speit, G., W. Vogel and M. Wolf (1982) Characterization of sister chromatid exchange induction by hydrogen peroxide, Environ. Mutagen., 4, 135-142. Sugimura, T., and S. Sato (1983) Mutagens-Carcinogens in Foods, Cancer Res., 2415s-2421s. Sutou, S. (1981) The mechanisms of endoreduplication, Cancer Genet. Cytogenet., 3, 317-325. Sutou, S., and Y. Arai (1975) Possible mechanisms of endoreduplication induction, Exp. Cell Res., 92, 15-22. Suwa, Y., M. Nagao, A. Kosugi and T. Sugimura (1982) Sulfite suppresses the mutagenic property of coffee, Mutation Res., 102, 383-391. Takanari, H., and K. Izutsu (1983) Studies on endoreduplication, III. Endoreduplication induced by mitomycin C in PHA-stimulated tonsillar lymphocyte cultures, Mutation Res., 107, 297-306. Taylor, R.T., and M.L. Hanna (1977) Folate-dependent enzymes in cultured Chinese hamster cells: Folylpolyglutamate synthetase and its absence in mutants auxotrophic for glycine + adenosine + thymidine, Arch. Biochem. Biophys., 181, 331-334. Taylor, R.T., R. Wu and M.L. Hanna (1985a) Induced reversion of a Chinese hamster triple auxotroph, Validation of the system with several mutagens, Mutation Res., 151, 293-308. Taylor, R.T., M.L Hanna, S.A. Stewart, D.K. Hendron, J.L. Minkler and A.V. Carrano (1985b) Measurement of induced reversion and sister chromatid exchange in a Chinese hamster ovary triple auxotroph: assays of mitomycin C and methylglyoxal (a coffee mutagen), Environ. Mutagen., 7, Suppl. 3, 7. Tucker, J.D., R.T. Taylor, M.L. Christensen, C.L. Strout and M.L. Hanna (1989) Cytogenetic response to coffee in Chinese hamster ovary cells and human peripheral lymphocytes, Mutagenesis, in press.
269
MUTGEN 01478
Cytogenetic response to 1,2-dicarbonyls and hydrogen peroxide in Chinese hamster ovary AUXB1 cells and human peripheral lymphocytes James D. Tucker, Robert T. Taylor, Mari L. Christensen, Cheryl L. Strout, M. Leslie Hanna and Anthony V. Carrano Lawrence Livermore National Laboratory, Biomedical Sciences Division, University of California, P.O. Box 5507, L-452, Livermore, CA 94550 (U.S.A.) (Received 9 December 1988) (Revision received 24 April 1989) (Accepted 8 May 1989)
Keywords: 1,2-Dicarbonyls; Hydrogen peroxide; Coffee; Chinese hamster ovary AUXB1 cells; Human peripheral lymphocytes
Summary Mutagenic 1,2-dicarbonyls have been reported to occur in coffee and other beverages and in various foods. We have measured the induction of sister-chromatid exchanges (SCEs) and endoreduplicated cells (ERCs) to determine the genotoxicity of various 1,2-dicarbonyl compounds in Chinese hamster ovary (CHO) AUXB1 cells and human peripheral lymphocytes. The 1,2-dicarbonyls glyoxal, methylglyoxal and kethoxal each induced highly significant increases in both SCEs and ERCs in AUXB1 cells. Glyoxal and kethoxal induced SCEs but not ERCs in human peripheral lymphocytes. In addition, hydrogen peroxide induced highly significant levels of SCEs and ERCs in AUXB1 cells. Bisulfite, which reacts with carbonyl groups to form addition products, significantly reduced the frequency of SCEs and the proportion of ERCs when glyoxal, methylglyoxal, kethoxal and diacetyl were administered to AUXB1 ceils. In addition, bisulfite blocked the formation of ERCs, but not SCEs, induced by hydrogen peroxide. These in vitro results suggest that 1,2-dicarbonyls may play an important role in the genotoxicity of some foods and beverages.
1,2-Dicarbonyls have been shown to exist in coffee, numerous other beverages, and in many processed foods (Palo and Ilkova, 1970; Hayashi and Shibamoto, 1985; Nagao et al., 1986). Methylglyoxal is the most prevalent 1,2-dicarbonyl in brewed and instant coffee, and in bourbon whiskey (Nagao et al., 1986). Glyoxal and ethylglyoxal are
Correspondence: Dr. James D. Tucker, Lawrence Livermore National Laboratory, Biomedical Sciences Division, University of California, P.O. Box 5507, L-452, Livermore, CA 94550
(U.S.A.).
also present in these and other foods and beverages (Nagao et al., 1986). Methylglyoxal is a product of cellular metabolism (Cooper and Anderson, 1970; Sato et al., 1980), and has been shown to induce mutations in V79 cells (Cajelli et al., 1987), SCEs in Chinese hamster ovary (CHO) cells (Faggin et al., 1985), and fibrosarcomas in rats (Nagao et al., 1986). The structures of the 1,2-dicarbonyls used in this study are given in Table 1. Glyoxal is the parent compound, with methylglyoxal and diacetyl differing by the addition of one and two methyl groups, respectively. Kethoxal is an
0165-1218/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
270 TABLE 1 CHEMICAL STRUCTURES Chemical
Structure
Glyoxal
CHO-CHO
Methylglyoxal
C(CH3)O-CHO
Diacetyl
C(CH3)O-C(CH3)O
Kethoxal
CHO-CO I C H 3 - C H - O - C H 2- c n 3
various coffees in these same systems is reported elsewhere (Tucker et al., 1989). The results presented in this paper indicate that each of the 1,2-dicarbonyls tested induced SCEs and ERCs, and that bisulfite, which forms complexes with carbonyl groups, blocked a significant proportion of the 1,2-dicarbonyl genotoxicity. Bisulfite was also found to block the formation of ERCs, but not SCEs, induced by hydrogen peroxide. Materials and methods
Chemicals
ethoxy-substituted 2-ketobutyraldehyde that was originally investigated for its carcinostatic properties (French and Freedlander, 1958). Kethoxal has not been shown to be present in food or beverages, but was included as a model compound for comparative purposes. Genotoxic compounds other than 1,2-dicarbonyls are also found in coffee. Hydrogen peroxide (H202) is generated in coffee beginning soon after preparing the beverage, and its production continues for at least 24 h (Fujita et al., 1985; Nagao et al., 1986; Rinkus and Taylor, 1989). It has been shown to induce mutations (Fujita et al., 1985; Speit, 1986), SCEs (Speit et al., 1982; Estervig and Wang, 1984; Mehnert et al., 1984; Speit, 1986) and possibly chromosome aberrations (Sofuni and Ishidate, 1984; Ishidate et al., 1988), and was found to have increased Salmonella mutagenic activity in the presence of methylglyoxal (Fujita et al., 1985). The genotoxicity of active oxygen species in mammalian cells recently has been reviewed (Imlay and Linn, 1988; Meneghini, 1988). Due to the presence of 1,2-dicarbonyl compounds in both food and beverages, and to hydrogen peroxide in coffee, we have undertaken an investigation of two cytogenetic effects of these compounds in mammalian cells. This paper reports the results of our studies with four different 1,2-dicarbonyls in CHO AUXB1 cells using SCEs and endoreduplicated cells (ERCs) as the endpoints. Two of the 1,2-dicarbonyls were also studied in human peripheral lymphocytes. Work done in parallel with this study on the effects of
Glyoxal (CAS No. 107-22-2) was purchased from Aldrich, methylglyoxal (CAS No. 78-98-8) from Sigma, kethoxal (CAS No. 3688-37-8) from U.S. Biochemical Corporation, diacetyl (CAS No. 431-03-8) from Eastman, hydrogen peroxide (CAS No. 7722-84-1) from Mallinckrodt, sodium bisulfite (CAS No. 7631-90-5) from Baker, and mitomycin C (MMC, CAS No. 50-07-7) from CalBiochem. Hydrogen peroxide was dissolved in phosphate-buffered saline, kethoxal was initially dissolved in a small amount of ethanol prior to being dissolved in distilled water, sodium bisulfite was dissolved in water and adjusted to pH 7.2, and the remaining chemicals were dissolved in distilled water. Chinese hamster ovary (CHO) AUXB1 cells, triple-auxotrophic for glycine, adenosine and thymidine (Taylor et al., 1977) were grown as described previously (Taylor et al., 1985a, b), with the following modifications. Exposure times differed according to the endpoint being examined, and were 15, and 20-22 hours, for ERCs and SCEs, respectively. This difference was necessary at the lower doses because of the need to obtain predominantly first-division cells for ERCs and second-division cells for SCEs. Cells were exposed to the test chemicals in foil-covered 25 × 150 mm glass culture tubes in 40 ml of medium containing 0.67 mM glycine and 37/~M adenosine and were incubated on a roller wheel at 37 o C. For sisterchromatid differentiation, the amount of thymidine was reduced to 10 /xM and bromodeoxyuridine (BrdUrd) was incorporated at a concentration of 30 #M. When cells were exposed to more than one chemical, for example bisulfite and the test agent, the former was incorporated into the
271 cultures and the test compound was added immediately thereafter. Following exposure, the cells were collected by centrifugation, resuspended into fresh glycine-adenosine-supplemented medium containing 30/~M BrdUrd plus 10/~M thymidine, but lacking the test agent(s). They were then plated for cell survival (Taylor, 1985a) and aliquoted in 12-ml volumes into foil-covered Coming 16 x 125 mm sterile disposable culture tubes for continued incubation at 3 7 ° C on a roller wheel. Harvest times were determined from previous experiments designed to measure growth kinetics following exposure to each dose. Harvest times varied from 25 to 55 h following the beginning of exposure to capture second-division cells for SCEs, and from 20 to 45 h to capture first-division cells for endoreduplication. 5 h prior to harvest, colcemid was added at a final concentration of 0.1/~g/ml. The cells were then centrifuged, the supernatant removed, and 0.075 M KC1 was added for 30 min. Freshly prepared fixative (methanol:acetic acid, 3 : 1 ) was added, the cells centrifuged again, and then fixed an additional 3 times. Cells were dropped onto clean dry slides and stained by the fluorescence plus Giemsa method as described (Perry and Wolff, 1974; Minkler et al., 1978). For SCEs, 50 second-division cells (25 from each of two consecutive harvest times) were scored unless otherwise noted. For endoreduplicated cells, 1000 first-division cells from a single harvest time were scored.
Human peripheral lymphocytes Peripheral blood from a healthy, nonsmoking female was drawn into a heparinized tube and cultured in RPMI-1640 with 25 mM Hepes supplemented with 16% fetal calf serum, L-glutamine (final concentration 2 mM), 1% penicillin-streptomycin (Gibco), 1% phytohemagglutinin (M form, Gibco), and 25 /~M BrdUrd. The cultures were grown in the dark in 5% CO 2 at 3 7 ° C for 52 h (ERCs) or 72 h (SCEs). Two cultures for each chemical, dose, and harvest time were employed. Colcemid (0.1 # g / m l ) was present for the final 4 h. The slides were prepared and stained as described for the AUXB1 cells. The scoring for both cell types was done on coded slides. For SCEs, 25 cells were scored from each replicate culture, for a
total of 50 cells from each dose point. For ERCs, at least 500 cells were scored from a single culture.
Statistical methods The SCEs were normalized to 21 or 46 chromosomes per cell for the C H O AUXB1 and human cells, respectively. Student's t-test and Fisher's exact test were used to compare the SCE and ERC frequencies, respectively. One-tailed comparisons were made in each case. Results
The SCE and ERC results for glyoxal, methylglyoxal, kethoxal and hydrogen peroxide in AUXB1 cells are presented :in Table 2. A first-division endoreduplicated AUXB1 cell is illustrated in Fig. 1. Each of these chemicals produced highly significant increases in both endpoints. For the SCEs, treatment with each agent was dose-dependent. Kethoxal and glyoxal produced a doubling and a tripling of the baseline frequency, respectively, while methylglyoxal and hydrogen peroxide nearly doubled the baseline. Glyoxal produced the largest increase in SCEs while hydrogen peroxide induced a significant increase at the lowest dose. The relative percent cell survivals are also given. Significant cell cycle delays were obtained in every case. The ERC response to these chemicals is also highly significant. Dose responses were obtained with each agent, although toxicity was apparent at the highest doses in some cases. The weakest response was produced by hydrogen peroxide, and the strongest response was produced by kethoxal. Kethoxal is the most potent in terms of the induced ERC response, being active at 75 #M. The lowest significant doses for glyoxal, methylglyoxal and hydrogen peroxide were 200/tM, 100/~M and 160/tM, resp. To determine whether these responses were typical of mammalian cells in general, or were unique to AUXB1 cells, we tested glyoxal and kethoxal in human peripheral lymphocytes. These results are presented in Table 3. Glyoxal produced small but significant SCE increases at every dose tested but did not yield a dose response. However, the results did produce a positive trend test ( p < 0.05). Kethoxal produced a linear SCE response
0 100' 200 300 400 500
0 200 400 600 800 1000 1 200 1 400 1 600
Glyoxal
Methylglyoxal
Dose (/tM)
Chemical
12.2 14.4 17.9 17.8 20.9 21.6
10.6 15.8 19.0 19.4 23.7 24.8 25.4 28.0 30.9
Mean/ cell
1.01 1.22 1.58 1.64 1.57 1.75
0.35 0.57 0.77 0.68 0.90 0.92 0.75 0.80 1.28
Std. Err.
50 50 50 50 50 50
100 50 50 50 50 50 50 50 50
Nu mbe r of cells scored
Sister-chromatid exchanges
0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
N.S. e 0.002 0.002 < 0.0001 < 0.0001
< < < < < < < <
Probability
29, 30, 30, 35, 40, 48,
25, 30, 30, 30, 35, 35, 40, 40, 50, 34 35 35 40 45 53
30 35 35 35 40 40 45 45 55
Harvest times (b)
a
2.09 1.94 1.94 1.76 1.90 1.66 1.71 1.27 1.56 2.28 2.01 1.99 1.93 1.82 1.64
CKI
1.90 1.51 1.80 1.40 1.45 1.09 1.20 1.03 1.37 2.04, 2.00, 1.64, 1.48, 1.61, 1.41,
Std. Dev. 1.7 4.0 5.0 5.3 6.8 5.0 5.9 5.7
2.2 4.2 3.8 4.8 4.9 6.4
ERCs/ 1000 cells 1.5 b 16 26 29 48 26 36 34 _ c 5 18 15 24 25 43
0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
0.005 0.02 0.0003 0.0002 < 0.0001
< < < < < < <
Probability
Endoreduplicated cells
20 20 25 30 35 40
25 30 30 30 35 35 40 40 50
Harvest time (h)
1.11 1.02 1.00 1.00 1.03 1.09
1.90 1.51 1.80 1.40 1.45 1.09 1.20 1.03 1.37
CK I a
100+/--3 97 + / -- 3 92 + / -- 3 83 + / -- 3 61 + / -- 2 31 + / -- 1
100+/--4 d 100+/--4 100+/--4 100+/--5 91 + / -- 3 50 + / -- 4 28 + / -- 2 9.4 + / -- 0.5 0.9+/--0.2
Relative percent survival
S I S T E R - C H R O M A T I D E X C H A N G E S A N D E N D O R E D U P L I C A T E D CELLS I N D U C E D BY G L Y O X A L , K E T H O X A L , M E T H Y L G L Y O X A L A N D H Y D R O G E N P E R O X I D E 1N C H O AUXB1 CELLS
TABLE 2
bJ
b c a e f g
a
0 40 80 120 160 200 240
50 75 100 125 150 175 200 225 250 275 300
11.0 14.1 14.9 14.9 17.3 17.5 19.6
0 11.5 12.7 14.8 15.2 16.4 17.7 20.8 _ t 24.9 _ t 19.5 25
1.75 50 50 50 50 50 50 50
25
0.98
0.44 0.85 0.75 0.73 0.87 1.15 1.56
0.52 50 50 50 50 25 50 25
11.2 0.54 0.59 0.59 0.78 0.72 0.66 0.75
< < < < <
0.001 0.0001 0.0001 0.0001 0.0001 0.0001
< 0.0001
< 0.0001
50 N.S. 0.03 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001
25, 30 25, 30 30, 35 35, 40 35, 40 40, 45 40, 45
25, 30 25, 30 30, 35 30, 35 30, 35 35, 40 35, 40 35, 40 40, 45 40, 45 45, 50 1.77, 1.76, 1.98, 1.98, 1.98, 1.97, 1.94,
2.00 2.00 2.08 2.46 2.13 2.61 2.36
25, 30 1.77, 2.00 1.39, 1.82 1.45, 1.95 1.16, 1.74 1.04, 1.37 1.07, 1.52 1.00, 1.11 1.01, 1.08 1.01, 1.06 1.00, 1.01 1.00, 1.02 1 6 2 5 16 24 39
1.78, 1.97 9 32 22 42 49 52 85 81 g 60 g 98 g 101 g 1.0 2.4 1.4 2.2 4.0 4.8 6.1
6 3.0 5.6 4.6 6.3 6.8 7.0 8.8 12.2 10.6 13.3 13.5
N.S. N.S. N.S. 0.0002 < 0.0001 < 0.0001
2.4 N.S. 0.0002 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001
CKI, cell kinetics index = {number of M1 cells+2(number of M2 cells)+3(number of M3 cells))/total cells scored. 2000 cells scored. Insufficient first-division cells. Ranges represent the larger of the standard errors of the mean, or the counting error (Taylor et al., 1985). N.S., not significant. Insufficient second-division cells. Mean of 2 separate Expts. 1000 cells scored from each.
Hydrogen peroxide
Kethoxal
20 20 25 30 30 35 35
25 25 30 30 30 35 35 35, 35 45, 40 45,45 50, 50 1.24 1.10 1.49 1.93 1.81 1.92 1.85
100+/-4 96+/-4 88 + / 67 + / 51 + / 42 + / 23 + / -
4 3 2 3 1
25 1.78100+/-4 1.77 95 + / - 4 1.39 100+/-4 1.45 100+/-7 1.16 78+/-4 1.04 57 + / - 4 1.07 65 + / - 5 1.00 41 + / - 3 1.01, 1.02 44 + / - 3 1.06, 1.02 35 + / - 2 1.01,1.03 1 2 + / - 1 1.02, 1.14 7.4+/-0.5
274
t
experiment because sodium bisulfite is also a nucleophile for H202 (Ruggero and Edwards, 1970). The SCE and ERC results, along with the relative percent cell survivals, are presented in Table 4. Bisulfite (1.0 mM) blocked much, but not all, of the 1,2-dicarbonyl genotoxicity as measured by SCEs and ERCs. It also consistently enhanced the cell survival of cytotoxic concentrations of both the 1,2-dicarbonyls and H202. The number of SCEs was reduced dramatically in cultures treated with each of the four 1,2-dicarbonyls, although the frequencies remained elevated compared to the controls in most cases. For 1,2-dicarbonyls, the average percentage of ERCs blocked by bisulfite was greater than the mean percentage of SCEs blocked (89% vs. 69%). Bisulfite, however, did not prevent the formation of SCEs induced by H202, but did block a significant fraction of the ERCs at the higher dose tested. This may simply reflect the higher levels of H202 that are needed to
Fig. 1. Endoreduplicated AUXB1 cell with 21 diplochromosomes induced by 400/~ M methylglyoxal treatment. TABLE 3
but did not achieve a doubling of the baseline. Neither of these agents induced ERCs at any of the doses tested (data not shown). 1,2-Dicarbonyls are known to react reversibly with bisulfite to form addition products (Shapiro, 1977). These reactions eliminate the carbonyl moiety, as shown in the following reaction equilibria:
SISTER-CHROMATID EXCHANGES INDUCED BY G L Y O X A L A N D K E T H O X A L IN H U M A N P E R I P H E R A L LYMPHOCYTES a Chemical
Dose (/~ M)
SCEs/cell
Std. Err.
Probability
Glyoxal
0 400 700 1 000 1300 1600 1900 2200 2500 2800
9.3 10.9 12.8 11.9 12.7 12.5 13.7 11.4 12.0 14.3
0.69 0.64 0.67 0.72 0.77 0.71 0.83 0.68 0.63 0.86
0.05 < 0.0001 0.005 0.003 0.001 < 0.0001 0.02 0.002 < 0.0001
0 1 000 1 400 1 800 2200 2600 3000 3 400
10.7 11.1 12.7 13.0 13.7 15.4 16.9 _ o
0.62 0.70 0.79 0.66 0.70 0.58 0.72
N.S. c 0.03 0.01 0.001 < 0.0001 < 0.0001
R1CO-R2CO + NaHSO 3 R1CO-R2C(OH)SO3Na R I C O - R 2 C ( O H ) S O 3 N a + NaHSO 3 R ~ C ( O H ) S O 3 N a - R 2C(OH)SO3Na We made use of these reactions to determine if the 1,2-dicarbonyl functional groups are indeed responsible for the observed genotoxicity. AUXB1 cells were exposed to each of 4 different 1,2-dicarbonyls (glyoxal, methylglyoxal, kethoxal or diacetyl) and hydrogen peroxide at 2 different doses in the presence or absence of 1 mM sodium bisulfite. Hydrogen peroxide was included in this
Kethoxal
CKI b 2.33 1.89 1.80 1.66 1.45 1.37 1.66 1.68 1.33 1.41 2.07 1.85 1.87 1.55 1.36 1.22 1.10 1.02
50 cells scored per dose point. b CKI, cell kinetics index = {number of M1 c e l l s + 2 ( n u m b e r of M2 cells)+ 3(number of M3 c e l l s ) } / t o t a l cells scored. c N.S., not significant. d Insufficient second-division cells for analysis.
+ +
+ +
+ +
250 250 500 500
K e t h o x a l f 125 125 250 250
125 125 250 250
0 0 100 100 200 200
Methylglyoxal
Diacetyl
Hydrogen peroxide
c c a e r g
a
+ +
600 600 1200 1200
Glyoxal
11.60 10.35 13.74 13.46 16.29 18.00
14.8 10.1 16.6 13.3
13.3 10.7 20.8 9.0
17.3 11.6 23.0 e 13.8
18.3 11.9 23.8 12.7
10.1 9.3 N.S. 'j
-
bisulf.
Culture without
0.51 0.42 N.S. 0.59 < 0 . 0 0 0 1 0.77 0.0006 0.60 < 0.0001 0.86 < 0.0001
64.8
68.3
75.3
68.0
N.S. N.S. 0
0
-
39.2
82.7
20, 20, 25, 20, 30, 25,
30, 25, 50, 25,
30, 25, 40, 25,
25 25 30 25 35 30
35 30 55 30
35 30 45 30
30, 35 25, 30 50, 55 25,30
30, 35 25, 30 40, 45 25,30
25, 30 25, 30
1.71, 1.78, 1.94, 1.92, 1.94, 1.93,
1.31, 1.46, 1.95, 1.07,
1.06, 1.95, 1.00, 1.97,
1.99 1.99 1.99 1.99 2.00 1.87
1.79 1.92 1.99 1.88
1.86 2.00 1.03 2.00
1.04, 1.17 1.82, 2.00 1.53, 1.58 1.07,1.58
1.20, 1.73 1.73, 1.97 1.12, 1.99 1.21,1.48
1.78, 2.01 1.72, 1.92
Harvest CK1 b time
SCEs (h) blocked a
0.0002 5 7 . 0 0.0001 100.0
0.78 < 0.0001 0.47 N.S. < 0.0001 0.89 < 0.0001 0.61 < 0.0001 0.001
0.55 < 0.0001 0.46 0.03 1.31 < 0.0001 0.37 N.S. <
0.73 < 0.0001 0.44 0.001 < 0.0001 1.01 < 0.0001 0.69 < 0 . 0 0 0 1 < 0 . 0 0 0 1
0.68 < 0.0001 0.59 0.001 < 0.0001 0.85 < 0.0001 0.64 < 0 . 0 0 0 1 < 0 . 0 0 0 1
0.49 0.58
Control
Percent induced
10 1 15 10 38 11
7 1 0 1
36 10 44 2
36 4 77 18
15 3 62 6
0 1
cells
ERCs/ 1000
3.1 1.0 3.8 3.1 6.0 3.3
2.6 1.0 0.0 1.0
5.9 3.1 6.5 1.4
5.9 2.0 1.7 4.2
3.8 1.7 7.6 2.4
0.0 1.0
1000 1000 1000 1000 1000 1000
1000 1000 1000 1000
1000 1000 1000 1000
1000 1000 39 1000
1000 1000 1000 1000
1000 1 000
64.0
0
1.01
1.00
20
25
1.04
1.01
20
20
1.30 1.24
1.07
1.95
20 20
50
1.31 1.46
30 25 25
1.03
1.83
20
40
1.05
1.14
25
20
N.S.
100.0
97.7
75.0
1.07
1.53
50
25
1.82
1.04
25
30
1.20 1.73 1.12 1.21
30 25 40 25
-
0.04
-
0.006 N.S. N.S. N.S. < 0.0001 < 0.0001 N.S.
0.008 N.S. N.S. N.S.
< 0.0001 < 0.0001 0.02 < 0.0001 <0.0001 N.S.
77.9
91.7
91.9
< 0.0001
-
86.7
1.78 1.72
25 25
2
5
100+/-4 106+/-4 73 + / - 2 88 + / - 3 20 + / - 1 47 + / - 1
117+/-8 97 + / - 4 42 + / - 2 75 + / - 4
107 + / 95+/-4 61 + / 95+/-4
98 + / - 4 20 + / - 2 120+/-4
102 + / - 5
103+/-4
31+/-2
93 + / - 5 115+/-5
100+/-4 103+/-4
Percent Harvest CKI b Relative induced time percent ERCs (h) survival blocked a
0.004
-
Culture without bisulf.
< 0.0001 < 0.0001 N.S. < 0.0001 < 0 . 0 0 0 1 < 0.0001
< 0.0001 N.S. < 0.0001 N.S.
N.S.
Number Control of cells scored
Std. P r o b a b i l i t y Dev. c o m p a r e d to:
IN AUXB1 CELLS
g
Percent of induced SCEs or E R C s b l o c k e d by bisulfite = { ( n u m b e r induced by c h e m i c a l - n u m b e r i n d u c e d b y c h e m i c a l a n d bisulfite t o g e t h e r ) / n u m b e r i n d u c e d b y chemical)} × 100. C K I , cell kinetics index = { n u m b e r of M1 cells + 2 ( n u m b e r of M 2 c e l l s ) + 3 ( n u m b e r of M3 c e l l s ) } / t o t a l cells scored. C o n t r o l s for glyoxal, methylglyoxal, k e t h o x a l a n d diacetyl data. N.S., not significant. First harvest time had only 8 scoreable cells. C o n t r o l values for E R C d a t a are 7 (no t r e a t m e n t ) a n d 2 (1 m M bisulfite), with 1000 cells scored in each case. R a n g e s represent the larger of the s t a n d a r d errors of the mean, or the c o u n t i n g error ( T a y l o r et al., 1985).
+ + +
+
(1 (mM)
Dose Bisul- S C E s / Std. P r o b a b i l i t y ( # M ) rite cell Err. c o m p a r e d to:
None c (control)
Chemical
E F F E C T S O F S O D I U M B I S U L F 1 T E O N SCE A N D E R C I N D U C T I O N
TABLE 4
%n
276 TABLE 5 SCEs A N D ERCs IN AUXB1 CELLS T R E A T E D W I T H M I T O M Y C I N C Chemical
Dose
Sister-chromatid exchanges
Endoreduplicated cells
Relative percent survival
Mean/cell
Std. error
Number of cells scored
ERCs/ 1000
Std. Deviation
N u m b e r of cells scored
0 40 nM
10.6 46.0
0.35 1.30
100 50
1.5 2
1.7 1.4
2000 1000
1004/-5 ~ 57 + / - 4
0 80 nM
11.8 58.8
0.35 1.64
100 50
2 1
2.0 1.0
2000 1000
100 + / - 5 42 + / - 3
Expt. 1 MMC
Expt. 2 MMC
a Ranges represent the larger of the standard errors of the mean, or the counting error (Taylor et al., 1985).
induce ERCs, as opposed to SCEs in AUXB1 cells (Table 2) and the inability of 1.0 mM bisulfite to d e s t r o y H 2 0 2 quantitatively and rapidly when both are together in a complex culture medium. It still seemed remotely possible that the formation of SCEs and ERCs might be tightly coupled in AUXB1 cells, such that agents inducing one endpoint would always also induce the other. Alternatively, ERCs might only be formed by a subset of those chemicals that are capable of inducing SCEs in these cells, e.g., 1,2-dicarbonyls. To determine if other chemicals that generate SCEs also induce ERCs in AUXB1 cells, we tested mitomycin C (MMC), a potent inducer of SCEs (Latt et al., 1981), that lacks a 1,2-dicarbonyl functional group. The results are presented in Table 5. It is clear that MMC induced many SCEs but no ERCs in these cells, indicating that the induction of ERCs is not an SCE-coupled general phenomenon in AUXB1 cells. This finding is in agreement with the results seen in the human peripheral lymphocytes, where SCEs but not ERCs were induced by 1,2-dicarbonyls. Moreover, as noted in Table 4, bisulfite blocked the formation of ERCs, but not SCEs, induced by H202, lending further support to the idea that these endpoints are not necessarily coupled in this cultured hamster cell line. Discussion
The results presented here indicate that the 1,2-dicarbonyls tested in these studies are potent
inducers of SCEs in both C H O AUXB1 cells and in human peripheral lymphocytes, and that they also induce high levels of ERCs in AUXB1 cells. These findings are of interest because 1,2-dicarbonyls, particularly glyoxal, methylglyoxal and diacetyl, are found in coffee and other foodstuffs. 1 g of instant coffee has been reported to contain 100 /ag of glyoxal, 100-210 /ag of methylglyoxal, and 100/ag of diacetyl, while 8 g of coffee beans have been reported to yield 470-730/ag of methylglyoxal per 100 ml of brewed coffee (Sugimura and Sato, 1983; Nagao et al., 1983; Kasai et al., 1982). The latter translates into 65 /aM methylglyoxal, while 200/aM and 100 laM are the lowest concentrations producing a significant increase of SCEs and ERCs, respectively, in this in vitro study. These observations suggest that the amount of methylglyoxal present in Japanese prepared coffees is too low to be responsible for our measurable increases in genotoxicity. This conclusion is also supported by other work in our laboratory (Tucker et al., 1989) where several types of coffee were observed to induce SCEs at concentrations ranging from approximately 1% to 10% of the full strength preparation. If we assume that our coffee is similar to that prepared by the Japanese, then the concentration of methylglyoxal would range from 0.65 /aM to 6.5 /xM, which is considerably below the lowest level found to induce SCEs (Table 2). Irrespective, the data in Tables 2 4 stress the importance of understanding the role of 1,2dicarbonyls in a variety of foods and beverages, including coffees. Our results, along with the above
277 and other published data, suggest that people may be exposed to significant amounts of genotoxic 1,2-dicarbonyls through their diet. Nakasato et al. (1984) demonstrated that methylglyoxal induces mutations in cultured Chinese hamster lung cells, but the influence of $9 was not examined. $9 virtually eliminates the mutagenicity of coffee in vitro (Nagao et al., 1984; Aeschbacher et al., 1985; Friederich et al., 1985). However, methylglyoxal retains slightly more than half its genetic activity in Salmonella strain TA100 when preincubation is carried out in the presence of rat-liver $9 (Friederich et al., 1985). Bronzetti et al. (1977) showed that $9 had little effect upon the mutagenicity of methylglyoxal in Salmonella strain TA97. However, these authors did observe a decrease due to $9 in the amount of gene conversion and in the number of mutations induced in Saccharomyces cerevisiae. Shane et al. (1988) examined the effects of glyoxal, methylglyoxal and diacetyl in 3 different Salmonella strains with and without $9. In some cases it was observed to decrease the mutagenic potency, while in other situations, no effect, or even increases were seen. The relative genotoxicity of the 1,2-dicarbonyls, as measured with SCEs and ERCs and calculated with linear regression analysis using the linear portion of the dose-response curves, is kethoxal > methylglyoxal > glyoxal, with hydrogen peroxide intermediate to kethoxal and methylglyoxal. In the Salmonella assay, methylglyoxal was also more potent than glyoxal (Sugimura and Sato, 1983). This agrees with our findings, although the differences between glyoxal and methylglyoxal in the SCE and ERC assays were not large. The concentration of hydrogen peroxide in coffee increases with time, and interacts synergistically with methylglyoxal to produce mutations in Salmonella strain TA100 (Fujita et al., 1985). Thus, the role of hydrogen peroxide in the genotoxicity of coffee probably depends upon several variables. The data presented here show that hydrogen peroxide by itself induces both SCEs and ERCs in AUXB1 cells. The mechanism for SCE induction by hydrogen peroxide was recently elucidated by Larramendy et al. (1987) who showed that the hydroxyl radical O H ' i s the intermediate responsible for this response. The reaction producing the OH" requires Fe 2+ as a catalyst. This indicates
that the concentration of Fe 2÷ in culture medium a n d / o r a beverage could influence the observed SCE frequency. Consequently, the concentration of Fe 2+ in coffee may at least partially determine the SCE response to coffee. Bisulfite prevents the formation of the majority of SCEs and ERCs induced by 1,2-dicarbonyls, but does not block SCE formation by H202 (Table 4). This argues in favor of the dicarbonyl group as being the structure responsible for the observed genotoxicity. Although cultures treated with both bisulfite and a 1,2-dicarbonyl still exhibited elevated SCEs and ERCs, the level was always significantly reduced compared to the effects of the 1,2-dicarbonyl alone. One would not necessarily expect the SCE or ERC frequency to be completely reduced to baseline levels due to competing reactions of certain culture medium components (e.g. pyruvate) for N a H S O 3 and the reversibility of NaHSO 3 addition reactions with carbonyl groups. In the experiments described here, we added kethoxal, methylglyoxal, or diacetyl to cultures containing a 2-8-fold molar excess of pre-added bisulfite. For cultures containing glyoxal, bisulfite was not added in excess because concentrations above 1.0 mlVL markedly inhibit AUXB1 suspension cell growth. However, cultures containing 0.6 or 1.2 mM glyoxal plus 0.1 mM NaHSO 3 showed comparable amounts of blockage of SCEs and ERCs. Another possibility for the incomplete blockage of genotoxicity rests on the observation that aldehydes induce SCEs in mammalian cells (Obe and Ristow, 1977; Ristow and Obe, 1978; Jansson, 1982). 1,2-Dicarbonyls with a single carbonyl group blocked are essentially aldehydes or ketones, and only a single carbonyl functional group may be required for at least some of their observed genotoxicity. Suwa et al. (1982) obtained nearly complete inactivation of the mutagenic activity of glyoxal and diacetyl in Salmonella TA100 with 3 molar equivalents of sodium sulfite, but these assays are performed in much simpler media. The evidence presented here indicates that additional factors in the genotoxicity of 1,2-dicarbonyls remain to be elucidated. Schwarzacher and Schnedl (1965) proposed that endoreduplication occurs when cells proceed through G2 directly to a second S phase without going through M and G1. This results in diplo-
278
chromosomes, and the cells will be tetraploid following another mitotic division. Huang et al. (1983) suggest that endoreduplication may result from the arrest of normal DNA replication. Sutou and Arai (1975) have suggested that alterations in cell surfaces, the cytoskeleton, or both, may be responsible for ERC formation. Whether 1,2-dicarbonyls or other agents known to induce ERCs interact with either the cell surface or the cytoskeleton is unknown. Although the precise mechanism of ERC formation is unknown, the chemical induction of endoreduplication has been explored in some detail. ERCs can form in the absence of chemical exposure, and numerous agents have been observed to increase this frequency (Sutou and Arai, 1975, and refs. therein). Our results demonstrate that 1,2-dicarbonyls can induce high levels of endoreduplication, at least in AUXB1 cells. Significant induction of ERCs by 1,2-dicarbonyls was also observed in CHO-S cells, the parental line (data not shown). The frequencies of ERCs reported here are similar to those observed for other chemicals (Sutou, 1981), with kethoxal appearing to be the most potent inducer of this endpoint. In light of our negative results with MMC in AUXB1 cells, and our negative results with 1,2-dicarbonyls in human peripheral lymphocytes, it is interesting that Takanari and Izutsu (1983) observed increases in the ERC frequency in human tonsillar lymphocytes following MMC treatment. The reason for this difference is not clear. The results presented in this paper indicate that 1,2-dicarbonyls and hydrogen peroxide, both constituents in coffee, induce significant levels of SCEs and ERCs in vitro. Whether significant genotoxic effects can be induced by these agents in vivo when they are administered either individually or in combination remains to be seen. Further work is needed to answer this question.
Acknowledgments The expert technical assistance of J. Lamerdin is gratefully appreciated. Work performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under contract number W-7405-ENG-48, with sup-
port from PHS grant No. CA40816 awarded to R.T.T. by the National Cancer Institute, DHHS.
References Aeschbacher, H.U., E. Ruch, H. Meier, H.O. Wurzner and R. Munoz-Box (1985) Instant and brewed coffees in the in vitro human lymphocyte mutagenicity test, Fd. Chem. Toxicol., 23, 747-752. Bronzetti, G., C. Corsi, D. Del Chiaro, P. Boccardo, R. Vellosi, F. Rossi, M. Paolini and G. Cantelli Forti (1987) Methylglyoxal: genotoxicity studies and its effect in vivo on the hepatic microsomal mono-oxygenase system of the mouse, Mutagenesis, 2, 275-277. Cajelli, E., R. Canonero, A. Martelli and G. Brambilla (1987) Methylglyoxal-induced mutation to 6-thioguanine resistance in V79 cells, Mutation Res., 190, 47-50. Cooper, R.A., and A. Anderson (1970) The formation and catabolism of methylglyoxal during glycolysis in E. coli, FEBS Lett., 11,273-276. Estervig, D., and R.J. Wang (1984) Sister chromatid exchanges and chromosome aberrations in human cells induced by H202 and other photoproducts generated in fluorescent light-exposed medium, Photochem. Photobiol., 40, 333-336. Faggin, P., A.M. Bassi, R. Finollo and G. Brambilla (1985) Induction of sister-chromatid exchanges in Chinese hamster ovary cells by the biotic ketoaldehyde methylglyoxal, Mutation Res., 144, 189-191. French, F.A., and B.L. Freedlander (1958) Carcinostatic action of polycarbonyl compounds and their derivatives, I. 3Ethoxy-2-ketobutyraldehyde and related compounds, Cancer Res., 18, 172-175. Friederich, U., D. Hann, S. Albertini, C. Schlatter and F.E. Wiirgler (1985) Mutagenicity studies on coffee, The influence of different factors on the mutagenic activity in the Salmonella/mammalian microsome assay, Mutation Res., 156, 39-52. Fujita, Y., K. Wakabayashi, M. Nagao and T. Sugimura (1985) Implication of hydrogen peroxide in the mutagenicity of coffee, Mutation Res., 144, 227-230. Hayashi, T., and T. Shibamoto (1985) Analysis of methyl glyoxal in foods and beverages, J. Agric. Food Chem., 33, 1090-1093. Huang, Y., C.-c. Chang and J.E. Trosko (1983) Aphidocolininduced endoreduplication in Chinese hamster cells, Cancer Res., 43, 1361-1364. Imlay, J.A., and S. Linn (1988) DNA damage and oxygen radical toxicity, Science, 240, 1302-1309. Ishidate Jr., M., M.C. Harnois and T. Sofuni (1988) A comparative analysis of data on the clastogenicity at 951 chemical substances tested in mammalian cell cultures, Mutation Res., 195, 151-213. Jansson, T. (1982) The frequency of sister chromatid exchanges in human lymphocytes treated with ethanol and acetaldehyde, Hereditas, 97, 301-303. Kasai, H., K. Kumeno, Z. Yamaizumi, S. Nishimura, M. Nagao, Y. Fujita, T. Sugimura, H. Nukaya and K. Takuo
279 (1982) Mutagenicity of methylglyoxal in coffee, Gann, 73, 681-683. Larramendy, M., A.C. Mello-Filho, E.A. Leme Martins and R. Meneghini (1987) Iron-mediated induction of sister-chromatid exchanges by hydrogen peroxide and superoxide anion, Mutation Res., 178, 57-63. Latt, S.A., J. Allen, S.E. Bloom, A.V. Carrano, E. Falke, D. Kram, E. Schneider, R. Schreck, R. Tice, B. Whitfield and S. Wolff (1981) Sister-chromatid exchanges: A report of the Gene-tox program, Mutation Res., 87, 17-62. Mehnert, K., W. Vogel, R. Benz and G. Speit (1984) Different effects of-mutagens on sister chromatid exchange induction in three Chinese hamster cell lines, Environ. Mutagen., 6, 573-583. Meneghini, R. (1988) Genotoxicity of active oxygen species in mammalian cells, Mutation Res., 195, 215-230. Minkler, J., D. Stetka Jr. and A.V. Carrano (1978) An ultraviolet light source for consistent differential stainng of sister chromatids, Stain Technol., 53, 359-360. Nagao, M., S. Sato and T. Sugimura (1983) in: G.R. Waller and M.S. Frather (Eds.), The Maillard Reaction in Foods and Nutrition, ACS Symp. Ser. 215, Ch. 28, pp. 521-536, American Chemical Society, Washington. Nagao, M., Y. Suwa, H. Yoshizumi and T. Sugimura (1984) Mutagens in coffee, Banbury Report, 17, 69-77. Nagao, M., Y. Fujita, K. Wakabayashi, H. Nukaya, T. Kosuge and T. Sugimura (1986) Mutagens in coffee and other beverages, Environ. Hlth. Persp., 67, 89-91. Nakasato, F., M. Nakayasu, Y. Fujita, M. Nagao, M. Terada and T. Sugimura (1984) Mutagenicity of instant coffee on cultured Chinese hamster lung cells, Mutation Res., 141, 109-112. Obe, G., and H. Ristow (1977) Acetaldehyde, but not ethanol induces sister chromatid exchanges in Chinese hamster cells in vitro, Mutation Res., 56, 211-213. Palo, V., and H. llkova (1970) Direct gas chromatographic estimation of lower alcohols, acetaldehyde, acetone and diacetyl in milk products, J. Chromatog., 53, 363-367. Perry, P., and S. Wolff (1974) New Giemsa method for the differential staining of sister chromatids, Nature (London), 251, 156-158. Rinkus, S.J., and R.T. Taylor (1989) Analysis of hydrogen peroxide in freshly prepared coffees, in preparation. Ristow, H., and G. Obe (1978) Acetaldehyde induces crosslinks in DNA and causes sister chromatid exchanges in human cells, Mutation Res., 58, 115-119. Ruggero, C., and J.U. Edwards (1970) Peroxide reaction mechanisms - - polar, in: D. Swern (Ed.), Organ Peroxides, Vol. 1, Wiley-Interscience, New York, Ch. 4, pp. 199-264. Sato, J., W. Yeu-Ming and J. Van Tys (1980) Methylglyoxal formation in rat liver cells, J. Biol. Chem., 255, 2046-2050.
Schwarzacher, H.G., and W. Schnedl (1965) Endoreduplication in human fibroblast cultures, Cytogenetics, 4, 1-18. Shane, B.S., A.M. Troxclair, D.J. McMillin and C.B. Henry (1988) Comparative mutagenicity of nine brands of coffee to Salmonella typhimurium TA100, TA102, and TA104, Environ. Mol. Mutagen., 11, 195-206. Shapiro, R. (1977) Genetic effects of bisulfite (sulfur dioxide), Mutation Res., 39, 149-176. Sofuni, T., and M. Ishidate Jr. (1984) Induction of chromosomal aberrations in cultured Chinese hamster cells in a superoxide-generating system, Mutation Res., 140, 27-31. Speit, G. (1986) The relationship between the induction of SCEs and mutations in Chinese hamster cells, I. Experiments with hydrogen peroxide and caffeine, Mutation Res., 174, 21-26. Speit, G., W. Vogel and M. Wolf (1982) Characterization of sister chromatid exchange induction by hydrogen peroxide, Environ. Mutagen., 4, 135-142. Sugimura, T., and S. Sato (1983) Mutagens-Carcinogens in Foods, Cancer Res., 2415s-2421s. Sutou, S. (1981) The mechanisms of endoreduplication, Cancer Genet. Cytogenet., 3, 317-325. Sutou, S., and Y. Arai (1975) Possible mechanisms of endoreduplication induction, Exp. Cell Res., 92, 15-22. Suwa, Y., M. Nagao, A. Kosugi and T. Sugimura (1982) Sulfite suppresses the mutagenic property of coffee, Mutation Res., 102, 383-391. Takanari, H., and K. Izutsu (1983) Studies on endoreduplication, III. Endoreduplication induced by mitomycin C in PHA-stimulated tonsillar lymphocyte cultures, Mutation Res., 107, 297-306. Taylor, R.T., and M.L. Hanna (1977) Folate-dependent enzymes in cultured Chinese hamster cells: Folylpolyglutamate synthetase and its absence in mutants auxotrophic for glycine + adenosine + thymidine, Arch. Biochem. Biophys., 181, 331-334. Taylor, R.T., R. Wu and M.L. Hanna (1985a) Induced reversion of a Chinese hamster triple auxotroph, Validation of the system with several mutagens, Mutation Res., 151, 293-308. Taylor, R.T., M.L Hanna, S.A. Stewart, D.K. Hendron, J.L. Minkler and A.V. Carrano (1985b) Measurement of induced reversion and sister chromatid exchange in a Chinese hamster ovary triple auxotroph: assays of mitomycin C and methylglyoxal (a coffee mutagen), Environ. Mutagen., 7, Suppl. 3, 7. Tucker, J.D., R.T. Taylor, M.L. Christensen, C.L. Strout and M.L. Hanna (1989) Cytogenetic response to coffee in Chinese hamster ovary cells and human peripheral lymphocytes, Mutagenesis, in press.