Carcinogenesis vol.19 no.2 pp.359–363, 1998
Characterization of the carcinogen 2-amino-3,8dimethylimidazo[4,5-f]quinoxaline in cooking aerosols under
domestic conditions
Chung-Chie Yang, Shinn Nen Jenq and Huei Lee1
Institute of Biochemistry and Toxicology, Chung Shan Medical and Dental
College, No. 110, Sec. 1, Chan-Gau North Rd., Taichung, Taiwan, ROC
1To
whom correspondence should be addressed
Email: [email protected]
Lung cancer in women is the leading cause of cancer death
in Taiwan. Most Chinese women are non-smokers and 60%
of female lung cancer patients have adenocarcinomas.
Epidemiological data indicate that the incidence of lung
cancer among Chinese women may be correlated with
cooking fumes. However, the carcinogenic compound(s) in
cooking fume aerosols is not defined. In the present study,
the cooking aerosols from Chinese stir-frying of fish were
prepared under domestic conditions. To determine the
mutagenic compounds in the cooking aerosol, mutagens
were purified by two steps of high-performance liquid
chromatography (HPLC), and their mutagenicity was
monitored with Salmonella typhimurium TA98. The
mutagen was eluted as a single peak. The chemical
structure of the mutagenic fraction of cooking aerosol
from frying of fish was characterized by UV spectra and
electrospray mass spectrometry. The bacterial indirectacting mutagenic compound in the cooking aerosol extract
was determined to be 2-amino-3,8-dimethylimidazo[4,5f]quinoxaline (MeIQx). An amount of 0.25 ng MeIQx/g of
meat per min was estimated based on the mutagenic
response. These data indicated that significant amounts of
MeIQx (268.1 ng/Chinese dish of frying fish) were present
in cooking aerosol in a short time. Chinese women spend
~1 h preparing meals everyday, thus, they may be exposed
to significant amounts of MeIQx from cooking aerosols in
the kitchen.
Introduction
Lung cancer in women is the leading cause of cancer death in
Taiwan. Most Chinese women are non-smokers and 60% of
female lung cancer patients suffer from adenocarcinomas (1,2);
adenocarcinoma is the most common form of lung cancer in
women who are non-smokers. This percentage closely agrees
with the 64.5% of non-smokers reported in a summary of 16
studies from six countries (1). Epidemiological studies of lung
cancer in Chinese women indicate that factors other than
cigarette smoking are related to lung cancer risk (1,2,4). Much
interest has been focused on identifying other potential risk
factors of lung cancer, including passive smoking, incense
burning, mosquito coil smoke, domestic coal fires and cooking
oil emissions (5–9). Multiple conditional logistic regression
analyses of risk factors of lung cancer in Taiwan reveal that
cooking is a significant risk factor associated with adeno*Abbreviations: HPLC, high-performance liquid chromatography; LC-MS,
liquid chromatography-mass spectrometry; DMSO, dimethylsulfoxide; ES,
electrospray; dbaha, dibenzo[a,h]anthracene; bap, benzo[a]pyrene.
© Oxford University Press
carcinoma of the lung (4). A possible explanation is the
high exposure to volatile mutagenic/carcinogenic components
formed during cooking (10). The mutagenic compounds in
cooking aerosols have still not been characterized, although
the mutagenicity of smoke formed during pan-broiling and
frying of lean pork has been demonstrated (11,12). In the
present study, the mutagenicity of cooking fumes from stirfrying fish was examined by a Salmonella/microsomal test.
To determine the major mutagenic compounds in this
cooking aerosol, mutagens were purified by two steps of
high-performance liquid chromatography (HPLC*), and their
mutagenicity was monitored with Salmonella typhimurium
TA98. The chemical structure of the major mutagenic fraction
of cooking aerosol from frying of fish was characterized by
UV spectra and liquid chromatography-mass spectrometry
(LC-MS).
Materials and methods
Collection and preparation of cooking fumes from stir-frying of fish
The cooking aerosol from Chinese frying of fish was prepared under domestic
conditions. Soybean oil (30 ml) was first heated in a wok until the oil
temperature reached to 180°C. Shredded fish (pomfret, 150 g) was quickly
placed in the wok and stir-fried for 5 min. The cooking aerosol was collected
for 5 min when the fish was cooked in the wok. The aerosol from heating
soybean oil (30 ml) alone was also prepared according to the conditions of
frying fish. The temperature changes during the frying of fish and heating
soybean oil alone were detected by Portable Digital Thermometer (JENCO
Electronics Ltd, Model 7000CH, Taiwan). The temperature changes are shown
in Figure 1. The aerosol of cooking fumes was collected with a high volume
air sampler through the hood at a flow rate of 1 m3/min. The cooking aerosol
was filtered through glass filters (Whatman, EPM1000) and extracted with
acetone in a shaker as described previously (13). The acetone extracts (1 mg)
were dissolved in 4 ml of redistilled water and were reacted with 10 mg of
blue cotton for 30 min. The aromatic compounds in acetone extracts adsorbed
by the blue cotton were then eluted with methanol/ammonium (50:1, v/v)
three times (14). The extract was dissolved in methanol and stored at 280°C
for mutagenicity test and HPLC analysis.
HPLC purification
The purification procedures for the major mutagen(s) from blue cotton
extracts were as described previously. Briefly, the blue cotton extracts of
cooking aerosol from fish-frying were first injected into a semi-preparative
Nucleosil column (10 µm particle, 103250 mm) and eluted at a flow rate of
2.5 ml/min with a gradient of acetonitrile in 10 mM phosphate/sodium
hydroxide (pH 7.2) using the following concentrations of acetonitrile: for
0–5 min a linear gradient of 10–25%; for 5–10 min a linear gradient of
25–35%; for 10–20 min a linear gradient of 35–55%; for 20–25 min a
linear gradient of 55–60%. Fractions were collected at 1-min intervals for
mutagenicity testing. The mutagenic fractions were pooled and loaded onto
a Nucleosil CN column (5 µm particle, 4.63250 mm). The mobile phase
was acetonitrile/water/diethylamine (12:88:0.1) at a flow rate of 1 ml/min.
Finally, the fractions showing mutagenicity from the second HPLC
purification were injected into a Nucleosil C18 column (5 µm particle,
4.63250 mm). The mobile phase was the same as the second HPLC
analysis. The peak of the sample with retention time corresponding to
standard mutagenic compound was monitored with a photodiode array
detector to compare them with the UV spectrum using standard MeIQx.
The authentic MeIQx was purchased from Toronto Research Chemicals
Inc. (Ontario, Canada).
Mutagenicity assay
The samples were tested for mutagenicity by the plate-incorporation assay
as described by Maron and Ames (15). An aliquot of 2 ml of molten top
359
C.-C.Yang, S.N.Jenq and H.Lee
Fig. 1. Changes of temperatures of heating soybean oil alone (d) and
frying fish (j) during the 5-min cooking processing. Bars among the
points show the standard deviation from the mean.
Table I. The amounts of cooking aerosol and acetone extracts collected
from cooking and processing of fried fish
Frying fish
Cooking
aerosol
Acetone
extracts
Blue cotton
extracts
Total amounts (mg)
g of meat/min (µg )
Percent of aerosol (%)
39.8
53.0
–
34.5
46.0
86.8
29.0
38.6
72.8
Fish (150 g) was fried with 30 ml soybean oil for 5 min. The meat
was added into the wok when the oil temperature reached 180°C.
The cooking aerosol, acetone extracts and blue cotton extracts obtained
from heating 30 ml of soybean oil alone for 5 min were 3.8 mg,
2.3 mg and 0.97 mg respectively.
agar was added to various concentrations of samples in 50 µl of
dimethylsulfoxide (DMSO) and 100 µl of each overnight culture of TA98.
The mixture was gently mixed and poured onto minimal glucose agar
plates. The plates were incubated at 37°C for 48 h and revertant colonies
were counted. All experiments were performed at least twice with triplet
plates for each dose. The linear regression equation from the dose–
response curve of mutagenicity was used to calculate the mutagenic activity
of 1 mg of cooking aerosol extract from fish-frying
LC-MS analysis
The mass spectrum of the major mutagenic fraction of cooking aerosol
from the final HPLC purification was performed by an electrospray (ES)
MS (VG Platform II, Fisons, UK). The sample was loaded onto a Nucleosil
C18 column (5 µm particle, 4.63250 mm) with a delivery system of
acetonitrile/water/formic acid (50/50/0.1) at a flow rate of 10 µl/min. This
makeup flow is necessary to obtain stable ES conditions. The ES efficacy
was the determining factor in ascertaining the moment when the mobile
phase reached the probe tip. Positive ES ionization (MH1) was performed
using an ionization voltage of –4.5 KV at the probe tip. The cone voltage
was set at 30 V, and the source temperature was kept constant at 60°C.
The mass spectrum was acquired over a 600 atomic mass unit (amu)
range at a 0.5 s/scan while the peak was eluted.
Results
The amounts of cooking aerosol, acetone extracts and blue
cotton extracts from 150 g of frying fish for 5 min are
shown in Table I. An amount of 39.8 mg of cooking aerosol
was collected from frying 150 g of fish for 5 min. The
amount of aerosol collected from 30 ml of cooking oil
alone was 3.8 mg in 5 min. The amount of aerosol from
fish-frying was 10.5-fold greater than that of oil-cooking
alone. The cooking aerosol from fish-frying was recovered
360
Fig. 2. The dose–response mutagenicity of the blue cotton extracts of
cooking aerosol in S.typhimurium TA98 with (m) and without (d) S9
mix. The linear regression equation from the dose–response curves of
indirect mutagenicity is shown in the top of the figure. Bars indicate SD
of the mean.
as acetone extracts with a 86.8% yield and as blue cotton
extract with a 72.8% yield. The mutagenicity of blue cotton
extracts of cooking aerosol from fish-frying and oil cooking
alone, were examined with S.typhimurium TA98 in the
presence and absence of S9 mix. The metabolic-acting
mutagenicity of cooking aerosols extracted from frying was
higher than that of direct-acting mutagenicity (Figure 2).
No mutagenic response was detected in the extracts of
soybean oil aerosol in the same dose ranges (data not shown).
To characterize the mutagenic compounds in cooking
aerosol extracts, the mutagenic fractions purified from semipreparative HPLC and an analytical Nucleosil CN column
were further loaded onto a Nucleosil C18 column. The most
potent mutagenic fraction was found at a retention time of
16 min, which corresponded to the retention time of
authentic MeIQx (Figure 3). Comparison of the authentic
MeIQx and the active mutagenic peaks collected from a
C18 column using a photodiode array detector and ES mass
spectrometry, confirmed the presence of MeIQx. The UV
spectra (Figure 4) and mass spectra (Figure 5) of the
mutagenic fraction with a retention time of 16 min
corresponded to the authentic MeIQx. The amount of MeIQx
in the cooking aerosol extracts from fish-frying was calculated
from the total revertants of mutagenic fractions (Figure 3)
based on the dose–response mutagenicity of authentic MeIQx
(Figure 6) . One milligram of blue cotton extract of cooking
aerosol was estimated to contain 9.26 ng MeIQx (Table II).
Therefore, 0.25 ng MeIQx/g of meat per min was calculated,
based on the data presented in Table I.
Discussion
MeIQx has been shown to be carcinogenic in rodents by
oral ingestion causing leukemia, and tumors of the liver,
clitoral gland, zymbal gland and lung (16,17). In the light
of previous reports (18,19), it is conceivable that dietary
exposure to MeIQx may have significant human health
implications. Layton et al. (20) indicated that MeIQx
2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline in cooking aerosols
Fig. 3. The HPLC chromatogram and mutagenic profile of the blue
cotton extracts of cooking aerosol. The mutagenicity of 1-min fractions
were tested with S.typhimurium TA98 in the presence of the S9 mix.
The mutagenic profile of the upper figure showed the most potent
mutagenic fraction with a retention time of 16 min, which corresponded
to the retention time of authentic meiqx.
Fig. 4. UV spectra of the active fractions of the blue cotton extracts of
cooking aerosol from an Nucleosil C18 column (A) and the authentic
MeIQx (B) monitored with a photodiode array by HPLC.
accounts for 27% of total incremental cancer risk among
the five principal heterocyclic amines found in cooked foods.
The highest cancer risk results from the ingestion of beef
products and fish (20). In the present study, we report for
the first time that MeIQx was the major metabolic-acting
mutagenic compound in cooking aerosol from fish-frying.
Berg et al. (12) indicated that the contribution of smoke
from broiled minced meat to total mutagenicity is influenced
by heating temperatures by 2.5–10.4%. However, the
mutagens in cooking aerosol were not identified by them.
Berg et al. (12) have suggested that the nature of the
mutagens present in the smoke, such as IQ-type mutagens,
are common in the crust of pan residue and the aerosol
fraction of smoke. Our data indicate that significant amounts
of MeIQx (268.1 ng/Chinese dish of fish-frying) were
present in cooking aerosols in a short time. Chinese women
spend ~1 h preparing meals every day.
Our results indicate that much of the mutagenic compounds,
which are acetone extractable and blue cotton adsorbable,
were generated in cooking aerosols during preparation of a
Chinese dish (Table I). In addition, greater amounts of
cooking aerosols were generated from stir-frying pork using
a small amount of cooking oil in a wok at 100–180°C than
that from deep frying with large amounts of cooking oil at
170–200°C (unpublished data). In Taiwan, to prepare
delicious and tender meats, women generally use the stirfrying method to cook julienne of meats with soybean oil
at high temperature. Thus, they may be exposed to significant
amounts of MeIQx from stir-frying cooking aerosols in
the kitchen.
The mortality rate of lung cancer in Chinese women is
among the highest in the world (21). It seems possible that
risk factors other than active cigarette smoking are involved
in the development of adenocarcinoma (1,2). In a casecontrol study (4), both active and passive cigarette smoking
are significantly associated with the development of three
pathological types of lung cancer, i.e. epidermoid carcinoma,
small-cell carcinoma and adenocarcinoma. While two types
of indoor air pollutants, burning incense and cooking fuels,
were not associated with the development of lung cancer.
The association of exposure to volatile emissions of rapeseed
cooking-oil with lung cancer risk in Chinese women has
been reported (2). Shields et al. (22) demonstrated that
condensates of the volatile emissions from heating Chinese
rapeseed oil to 275–280°C contain different amounts of
1,3-butadiene (504 ng/l) and benzene (2391 ng/l), formaldehyde (71.2 µg/l), acetaldehyde (306.9 µg/l) and acrolein
(391.8 µg/l). Among the five kinds of volatile organic
compounds, the Chinese rapeseed oil yielded a higher
emission rate of 1,3-butadiene than the peanut, soybean,
lard, sesame and Canola oils. Pellizzari et al. (23) also
reported that heating cooking oil to 265°C emitted higher
amounts of 1,3-butadiene, benzene and other organic acids.
In Shields’ experiment, soybean oil was heated at a
designated temperature (260–265°C). However in our study,
only 30 ml of soybean oil was heated in a wok and the
temperature was controlled at 225–240°C (Figure 1). The
cooking temperature is one of the critical factors for the
mutagenicity and emission of cooking oils (12,22,23). This
may cause the conflicting result on the mutagenicity of
extracts of cooking soybean oil. Nevertheless, it is possible
that the cooking aerosol from fish-frying may also contain
detectable amounts of these volatile mutagens/carcinogens.
On the other hand, Li et al. (24) indicated that soybean
cooking oil fumes prepared at 265°C contained polycyclic
aromatic hydrocarbons, especially dibenzo[a,h]anthracene
(DBahA) (3725 ng/g) and benzo[a]pyrene (BaP) (341 ng/
g). We also determined the amounts of DBahA and BaP in
oil cooking using high-performance liquid chromatography.
361
C.-C.Yang, S.N.Jenq and H.Lee
Fig. 5. Mass spectra of authentic MeIQx (A) and the HPLC purified MeIQx (B) that were present in the blue cotton extracts of cooking aerosols
from frying fish.
Table II. The mutagenicity and the amounts of MeIQx in the blue
cotton extracts of cooking aerosol from frying fish
Cooking aerosol
from frying fish
Mutagenicity
(revertants)
Amount of
MeIQx (ng)
Cooking fumes aerosols/mg
Blue cotton extracts/mg
Total 150 g fish/min
ND
635
4290
6.54
9.26
268.10
The number of revertants of blue cotton extracts/mg was calculated from
the dose–response curve of the blue cotton extracts from cooking
aerosols as shown in Figure 2.
The total net number of revertants of four mutagenic fractions: 1605
revertants/2.1 mg of the blue cotton extracts in Figure 3 that
corresponded to authentic MeIQx, were used to estimate the amounts of
MeIQx in blue cotton extracts from the dose–response curve of authentic
MeIQx as shown in Figure 6. The amounts of MeIQx in the blue cotton
extracts were further corrected and based on the recovery of authentic
MeIQx from HPLC processes.
ND: No mutagenic response was detected.
Fig. 6. The linear dose–response mutagenicity of authentic MeIQx in
S.typhimurium TA98 with a S9 mix. The linear regression equation is
shown at the top of figure.
362
The amount of DBahA and BaP in aerosol from heating
soybean oil alone at 180–252°C (,1 ng/g for DBahA,
68 ng/g for BaP) and frying fish at 98–162°C (,1 ng/g
2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline in cooking aerosols
for DBahA and BaP) were much lower than those of Li
et al.’s data (24).
In this study, we demonstrated that cooking aerosol
contains significant amounts of MeIQx. The association
between these mutagens/carcinogens from cooking emissions
and human lung cancer is not clear at present. Further
studies are needed to clarify the relationship between cooking
aerosol, oil vapors and other environmental exposures, with
genetic susceptibilities to lung cancer in Chinese populations.
Acknowledgements
This work was supported by grants from the Department of Health
(DOH83-TD-057 and DOH84-TD-034), the Executive Yuan of ROC For
their financial support. The authors wish to thank Dr M.-L.Hu for helpful
criticisms and for editing the manuscript.
(1995) Cancer risk of heterocyclic amines in cooked foods: An analysis
and implications for research. Carcinogenesis, 16, 39–52.
21. IARC (1987) Cancer Incidence in Five Continents, Vol 5. IARC
Scientific Publishing, Lyons, France.
22. Shields,P.G., Xu,G.X., Blot,W.J., Fraumenui,J.F.Jr, Trivers,G.E.,
Pellizzari,E.D., Qu,Y.H., Gao,Y.T. and Harris,C.C. (1995) Mutagens
from heated Chinese and US cooking oils. J. Natl Cancer Inst., 87,
836–841.
23. Pellizzari,E.D., Michael,L. C., Thomas,K.W., Shields,P.G. and Harris,C.
(1995) Identification of 1,3-butadiene, benzene, and other volatile
organics from wok oil emissions. J. Exp. Anal. Environ. Epidemiol.,
5(1), 77–87.
24. Li,S., Pan,D. and Wang,G. (1994) Analysis of polycyclic aromatic
hydrocarbons in cooking oil fumes. Arch. Environ. Health, 49(2),
119–122.
Received on April 2, 1997; revised on October 6, 1997; accepted on
October 7, 1997
References
1. Koo,L.C. and Ho,J.H.C. (1990) Worldwide epidemiological patterns of
lung cancer in nonsmokers. Int. J. Epidemiol., 19, S14–S23.
2. Gao,Y.T., Blot,W.J., Ershow,A.G., Hsu,C.W., Levin,L.I., Fraumeni,J. and
Zheng,W. (1987) Lung cancer among Chinese women. Int. J. Cancer,
40, 604–609.
3. Brownson,R.C., Loy,T.S., Ingram,E., Myers,J.L., Alavanja, M.C.R.,
Sharp,D.J. and Chang,J.C. (1995) Lung cancer in nonsmoking women.
Cancer, 75, 29–33.
4.Chen,C.-J., Wu,H.-Y., Chuang,Y.-C. et al. (1990) Epidemiologic
characteristics and multiple risk factors of lung cancer in Taiwan.
Anticancer Res., 10, 971–976.
5.Stockwell,H.G., Goldman,A.L., Lyman,G.H., Noss,C.I., Armstrong,A.W.,
Pinkham,P.A., Candelora,E.C. and Brusa,M.R. (1992) Environmental
tobacco smoke and lung cancer risk in nonsmoking women. J. Natl
Cancer Inst., 84, 1417–1422.
6. Chen,C.J. and Lee,H. (1996) Gentoxicity and DNA adduct formation
of incense smoke condensates: comparison with environmental tobacco
smoke condensates. Mutat. Res., 367, 105–114.
7. Lofroth,G., Stensman,C. and Brandhorst-Satzkorn,M. (1991) Indoor
sources of mutagenic aerosol particulate matter: Smoking, cooking, and
incense burning. Mutat. Res., 261, 21–28.
8. Mumford,J.L., He,X.Z., Chapman,R.S. et al. (1987) Lung cancer and
indoor air pollution in Xuan Wei, China. Science, 235, 217–220.
9. Qu,Y.H., Xu,G.X., Zhou,J.Z., Chen,T.D., Zhu, TL.F., Shields,P.G.,
Wang,H.W. and Gao,Y.T. (1992) Genotoxicity of heated cooking oil
vapors. Mutat. Res., 298, 105–111.
10. Lofroth,G. (1994) Airborne mutagens and carcinogens from cooking
and other food preparation processes. Toxicol. Lett., 72, 83–86.
11. Rappaport,S.M., McCartney,M.C. and Wet,E.T. (1979) Volatiliztion of
mutagens from beef during cooking. Cancer Lett., 8, 139–145.
12. Berg,I., Overvik,E., Nord,C.-E. and Gustafsson,J.A. (1988) Mutagenic
activity in smoke formed during broiling of lean pork at 200, 250 and
300°C. Mutat. Res., 207, 199–204.
13. Lee,H., Lin,T.-L., Shieh,R.-L. and Bian,S.-S. (1994) Mutagenicity of
airborne particulates from combustion of electric cables in a waste
metal retrieval area. Mutat. Res., 324, 77–84.
14. Lee,H., Lin,M.Y. and Lin,S.T. (1994) Characterization of the mutagen 2amino-3-methylimidazo[4,5-f]quinoline prepared from 2-methylpyridine/
creatinine/acetylformaldehyde model system. Mutagenesis, 9, 157–162.
15. Maron,D.M. and Ames,B.N. (1983) Revised methods for the Salmonella
mutagenicity test. Mutat. Res., 113, 173–215.
16. Kato,T., Ohgaki,H., Hasegawa,H., Sato,S., Takayama,S. and Sugimura,T.
(1988) Carcinogenicity in rats of a mutagenic compound, 2-amino-3,8dimethylimidazo[4,5-f]quinoxaline. Carcinogenesis, 9, 71–74.
17. Ohgaki,H., Hasegawa,H., Suenaga,M., Sato,S., Takayama,S. and
Sugimura,T. (1987) Carcinogenicity in mice of a mutagenic compound,
2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) from cooked
foods. Carcinogenesis, 8, 665–668.
18. Nagao,M. and Sugimura,T. (1993) Carcinogenic factors in food with
relevance to colon cancer development. Mutat. Res., 290, 43–51.
19. Goldbohm,R.A., van den Brandt,P.A., van’t Veep,P., Brants,H.A.M.,
Dorant,E., Sturmans,F. and Hermus,R.J.J. (1994) A prospective cohort
study on the relation between meat consumption and the risk of colon
cancer. Cancer Res., 54, 718–723.
20. Layton,D.W., Bogen,K.T., Knize,M.G., Hatch,F.T. and Johnson,V.M.
363