TOXICOLOGICAL SCIENCES, 150(2), 2016, 301–311
doi: 10.1093/toxsci/kfv322
Advance Access Publication Date: December 29, 2015
Research Article
Structure-Activity Relationships for DNA Damage by
Alkenylbenzenes in Turkey Egg Fetal Liver
Tetyana Kobets,*,1 Jian-Dong Duan,* Klaus D. Brunnemann,* Sylvain Etter,†
Benjamin Smith,‡ and Gary M. Williams*
*Department of Pathology, New York Medical College, Valhalla, New York 10595 and †Corporate Compliance
Division, Firmenich SA, Rue de la Bergere 7, CH - 1217 Meyrin 2, Switzerland, ‡Corporate Compliance Division,
Firmenich Incorporated, 250 Plainsboro road, Plainsboro, NJ 08536, USA.
1
To whom correspondence should be addressed at 40 Sunshine Cottage Road, BSB # 413, Valhalla, NY 10595. Fax: (914) 594-4163. E-mail:
[email protected].
ABSTRACT
Certain alkenylbenzenes (AB), flavoring chemicals naturally occurring in spices and herbs, are established to be cytotoxic
and hepatocarcinogenic in rodents. The purpose of the present study was to determine the DNA damaging potential of key
representatives of this class using the Turkey Egg Genotoxicity Assay. Medium white turkey eggs with 22- to 24-day-old
fetuses received three injections of nine AB with different carcinogenic potentials: safrole (1, 2 mg/egg), methyl eugenol (2,
4 mg/egg), estragole (20, 40 mg/egg), myristicin (25, 50 mg/egg), elemicin (20, 50 mg/egg), anethole (5, 10 mg/egg), methyl
isoeugenol (40, 80 mg/egg), eugenol (1, 2.5 mg/egg), and isoeugenol (1, 4 mg/egg). Three hours after the last injection, fetal
livers were harvested for measurement of DNA strand breaks, using the comet assay and DNA adducts formation, using the
nucleotide 32P-postlabeling assay. Estragole, myristicin, and elemicin induced DNA stand breaks. These compounds as well
as safrole, methyl eugenol and anethole, at the highest doses tested, induced DNA adduct formation. Methyl isoeugenol,
eugenol, and isoeugenol did not induce genotoxicity. The genotoxic AB all had the structural features of either a double
bond in the alkenyl side chain at the terminal 20 ,30 -position, favorable to formation of proximate carcinogenic 10 hydroxymetabolite or terminal epoxide, or the absence of a free phenolic hydroxyl group crucial for formation of a nontoxic
glucuronide conjugate. In contrast, methyl isoeugenol, eugenol and isoeugenol, which were nongenotoxic, possessed
chemical features, unfavorable to activation.
Key words: alkenylbenzenes (AB); carcinogenicity; chemical structure; genotoxicity; Turkey Egg Genotoxicity Assay (TEGA).
Alkenylbenzenes (AB) belong to a class of food-borne compounds
naturally occurring in spices and herbs, including sassafras, nutmeg, cinnamon, anise, basil, and tarragon. Several compounds
from this class (ie, safrole, methyl eugenol, and estragole) have
been found to be cytotoxic and hepatocarcinogenic in rodent
assays, causing increased incidences of hepatocellular adenoma
and carcinoma, and cholangiosarcoma in rats and mice of both
sexes (Bristol, 2011; International Agency for Research on Cancer
[IARC], 1976; National Toxicology Program [NTP], 2000; Williams
and Mattia, 2009). Based on data from high-dose animal studies,
safrole and methyl eugenol were judged to be “possible human
carcinogens (group 2B)” by the IARC (1976) and “reasonably
anticipated to be a human carcinogens” by the NTP (2000). These
findings together with the wide consumption of AB-containing
foods by humans led to awareness of the possible carcinogenic
potential of other structurally related representatives of this
class, such as myristicin, anethole, elemicin, and eugenol.
However, the results of safety evaluations of these compounds
were equivocal (Hasheminejad and Caldwell, 1994; NTP, 1983;
Sekizawa and Shibamoto, 1982).
The chemical structures of AB share several common features: the presence of a double bond in the alkenyl side chain and
the presence of a free and/or methoxylated hydroxyl group(s) on
the benzene core ring (Figure 1). The major hypothesis is that
minor structural variations such as the position of the alkenyl
double bond or the substitution of the alkoxy group on the
C The Author 2015. Published by Oxford University Press on behalf of the Society of Toxicology.
V
All rights reserved. For Permissions, please e-mail: [email protected]
301
302
|
TOXICOLOGICAL SCIENCES, 2016, Vol. 150, No. 2
FIG. 1. Chemical structures of tested alkenylbenzenes.
benzene ring by a hydroxyl group can determine the carcinogenic
potential of these compounds by allowing different routes of
metabolism (Figure 2) (Howes et al., 1990; Miller et al., 1979, 1983;
Rietjens et al., 2014; Smith et al., 2002). In general, at low doses the
predominant metabolic pathway for AB is O-demethylation that
results in detoxification of compounds through glucuronidation
(Figure 2). However, at higher doses, the O-demethylation pathway becomes saturated and metabolism is shifted to 10 -hydroxylation and epoxidation of the allyl side chain. Such hydroxyl
metabolites, predominantly in rodents, undergo sulfonation and
form metabolites that readily bind to DNA, whereas the epoxidation route leads to the formation of 20 ,30 -epoxide on the allyl double bond that also is capable of DNA binding, as has been shown
in vitro. In vivo epoxides, however, are more readily detoxified
(Figure 2) (Anthony et al., 1987; Drinkwater et al., 1976; Guenthner
and Luo, 2001; Howes et al., 1990; Rietjens et al., 2005). The data
from mechanistic studies confirm that such metabolic activation
of safrole, methyl eugenol, and estragole underlie their ability to
induce liver neoplasia by production of DNA damage (Burkey
et al., 2000; Paini et al., 2011; Phillips et al., 1984; Randerath et al.,
1984; Rietjens et al., 2005; Williams et al., 2013; Wislocki et al.,
1977). Meanwhile, compounds with a free phenolic hydroxyl
group, such as eugenol and isoeugenol, more readily undergo
conjugation with glucuronic acid and are excreted. Thus,
although still being cytotoxic to the liver, they do not share the
genotoxicity of other AB (NTP, 1983, 2010).
Importantly, the formation of DNA adducts from exposure to
AB has been observed in humans (Herrmann et al., 2013; Stening
et al., 1997; Zhou et al., 2007) and even associated with carcinogenicity. For example, safrole-DNA adducts were found in oral tissues of individuals who were betel quid chewers and are thought
to contribute to oral squamous cell carcinoma and esophageal
cancer pathogenesis (Chen et al., 1999; Lee et al., 2005).
To further evaluate the structure activity of AB, we have
studied the in ovo DNA damage in turkey fetal livers by a set of
nine compounds differing in structural features believed to be
critical to carcinogenicity. The susceptibility of turkeys to environmental chemical toxins led to the discovery of a major class of
potent carcinogens, the aflatoxins (Lancaster et al., 1961). The ability of turkey livers to respond to environmental chemicals
resulted in the successful use of the In Ovo Carcinogenicity Assay,
a model which assesses the potential of the tested substances to
produce preneoplastic lesions in turkey fetal livers (Enzmann
et al., 1992, 2013). The related Turkey Egg Genotoxicity Assay
(TEGA) developed in this laboratory (Williams et al., 2011) is a
novel, intact nonanimal model which uses the in ovo approach for
the assessment of the potential of chemicals to induce DNA damage in the fetal turkey liver. The sensitivity of turkey fetal liver to
DNA damaging activity of aflatoxin B1 was established in this
model (Williams et al., 2011). The model is intermediate between
in vitro and in vivo assays and does not require an exogenous
enzyme source for bioactivation. The endpoints of the TEGA
include the alkaline single cell gel electrophoresis (comet) assay
which detects DNA strand breaks and the nucleotide 32P-postlabeling (NPL) assay which detects DNA adducts. This model is similar to the Chicken Egg Genotoxicity Assay which has been
successfully used for the evaluation of chemical-induced genotoxic events and proven to be a reliable tool, capable of detection
of the genotoxicity of diverse DNA-reactive carcinogens, while
not yielding false-positive results for noncarcinogens (Williams
et al., 2014). The detection of various DNA-reactive carcinogens
that require bioactivation is possible in chicken and turkey fetal
KOBETS ET AL.
|
303
FIG. 2. General metabolic pathways for alkenylbenzenes. At low doses O-demetylation (CYP-catalyzed oxidation and the cleavage of O-methylene) is a predominant
pathway, yielding metabolites that are readily excreted as either free or as sulfate or glucuronic acid conjugates. At high doses, O-demethylation becomes saturated,
and 10 -hydroxylation and epoxidation of the allyl side-chain become major routs of metabolism. In rodents hydroxyl metabolites predominantly undergo sulfonation,
whereas in humans glucuronidation or oxidation of hydroxyl metabolites is more common. A, Alkenylbenzenes with double bond at the 20 ,30 -position.
B, Alkenylbenzenes with double bond at the 10 ,20 -position.
304
|
TOXICOLOGICAL SCIENCES, 2016, Vol. 150, No. 2
livers due to the presence of major phase I and phase II biotransformation enzymes (Hamilton et al., 1983; Perrone et al., 2004; Rifkind
et al., 1979). Moreover, the activity of the enzymes in embryonic turkey liver was shown to be similar to that in rats (Perrone et al.,
2004). The model has numerous advantages, such as: utilization of
intact organisms which are terminated several days before hatching and thus are not considered to be animals; an aseptic condition
with rigorous environmental control that prevents several types of
bias; organisms do not possess a microbiome and hence release of
endotoxin capable of enhancing the exposure effects is not a factor;
provides mechanistic data that elucidates the mechanisms
involved; time efficiency and lesser cost. A disadvantage of the
model is a limitation on tested doses of low solubility compounds,
which, however, was not a problem in the present experiments.
The present study assessed the responses of TEGA to nine
structurally diverse AB in relationship to their reported genotoxicity and carcinogenicity. The results provide a basis for
structure activity analysis.
MATERIALS AND METHODS
Chemicals. Safrole (CAS: 94-59-7, 97% pure as reported by the
supplier), estragole (CAS: 140-67-0, 98%), anethole (trans-anethole) (CAS: 4180-23-8, 99%), and eugenol (CAS: 97-53-0, 99%)
were obtained from Sigma-Aldrich (St Louis, Missouri). Methyl
eugenol (CAS: 93-15-2, 98%) was purchased from Vigon
International (East Stroudsburg, Pennsylvania), myristicin (CAS:
607-91-0, > 97%) was obtained from Abcam Inc. (Cambridge,
Massachusetts), elemicin (CAS: 487-11-6, 98%) was obtained
from Combi-Blocks, Inc. (San Diego, California), and isoeugenol
(CAS: 97-54-1, 99%) was purchased from Thermo Fisher
Scientific Inc. (Waltham, Massachusetts). Methyl isoeugenol
(CAS: 93-16-3, > 99%) was obtained from Firmenich SA (Meyrin,
Switzerland). A 20% aqueous solution of Kolliphor HS15 (20%
HS15), a gift from BASF SE (Ludwigshafen, Germany), was used
as a vehicle. The structures of test substances are shown in
Figure 1. The range of total doses is given in Table 1.
Experimental design. Fertilized broad-breasted white turkey
(Meleagris gallopavo) eggs of undetermined sex were purchased
from Aviagen Turkeys, Inc. (Lewisburg, West Virginia). Prior to
incubation, eggs were randomly divided into control and dosed
groups and stored at room temperature for no more than 3
days. The day on which incubation commenced was designated
as day 0. Eggs were incubated in 2GIF Styrofoam incubator
(Murray McMurray Hatchery, Webster City, Iowa) at 37 6 0.5 C
and 60 6 5% humidity and manually turned 180 twice per day.
Viability was assessed on day 19 by transillumination. Eggs that
did not develop were eliminated. Control and dosed (n ¼ 10 eggs
per group) eggs were separated to avoid any possible airborne
cross contamination. Vehicle and a dose range of each test substance: safrole, methyl eugenol, estragole, myristicin, elemicin,
anethole, methyl isoeugenol, eugenol, and isoeugenol were
administered in a total volume of 0.6 ml/egg through three daily
injections into the air sac on days 22 through 24 of incubation.
The dose ranges selected (Table 1) were based on available toxicity data (oral LD50 in rats, which for the tested set of AB ranged
from 810 to 4260 mg/kg). For each compound, the cumulative 3day dose was expressed as mg/egg total. At termination, 3 h
after the last injection, egg shells were opened, fetuses were
removed and decapitated. Fetal livers were harvested; half of
the collected samples were processed immediately for the
comet assay, the other half was frozen at 80 C for a subsequent NPL assay. Fetal body and liver weights as well as viability
were recorded. Groups chosen for analysis were required to
have viability levels higher than 50% in order to avoid falsepositive results due to cytotoxicity. For each procedure, appropriate New York Medical College-Chemical Safety Laboratory
Standard Operating Procedures were followed.
Alkaline single cell gel electrophoresis (comet) assay. The standard
comet assay as previously reported by Tice and Strauss (1995)
was performed to assess DNA strand breaks according to OECD
guidelines (OECD, 2014). The protocol previously described in
detail (Williams et al., 2014) was used. DNA migration was analyzed by fluorescence microscopy using a Nikon OPTIPHOT
microscope. The percentage of DNA-in-tail in samples (n ¼ 3 per
group) was determined using the Comet Score software v 1.5
(TriTek Corp, Sumerduck, Virginia), counting > 150 cells per sample. Median % tail DNA for each slide was determined and the
mean of the median values was calculated for each fetus, group
mean was calculated as an average of individual fetal means.
The obtained mean comet value from each of the dose groups
was then compared with the corresponding control (vehicle)
group, and a ratio value which reflects the fold change from the
control value was obtained. As recommended by van der Leede
et al. (2014), the evaluation criteria for determining positive
results include the following: a statistically significant increase in
the percentage of DNA-in-tail in dose group(s) compared with
corresponding vehicle control group; a dose-related increase in
DNA damage determined by linear regression analysis; the percentage of DNA-in-tail in dose group(s) should be outside of the
historical vehicle control range established in the laboratory
based on over 20 experiments (for 20% HS, a 95% CI is 9.73-11.68).
NPL assay. DNA adducts were assessed using the NPL method of
Randerath et al. (1981), as previously described in detail (Williams
et al., 2014). NPL analysis was conducted only for the highest
tested doses of AB, as they are most likely to reveal the presence
of adducts, which, according to Phillips and colleagues, is
adequate for identifying a positive response (Phillips et al., 2000).
Statistical analyses. Results are presented as mean 6 SD.
Statistical analyses were performed using SigmaStat software
version 3.11.0 (Systat Software Inc., Chicago, Illinois), with the
most appropriate statistical analysis of the data sets, which
included the one-way analysis of variance with the pairwise
multiple comparison being made by Tukey’s method. Linear
regression analysis was used to determine dose-related trends.
P values < 0.05 were considered significant.
RESULTS
The viability of turkey eggs in control groups was greater or
equal to 80%, with an average value of 90%. The results are summarized in Tables 1 and 2.
DNA strand breaks. Comet assay revealed statistically significant
dose-related increase in DNA strand breaks in the livers of turkey fetuses that were dosed with 20 and 40 mg/egg of estragole
(3.2- and 3.5-fold increase, respectively), whereas mortality
rates were slightly increased (6% and 19% respectively), compared with the control group (Figure 3). Lower doses of estragole
(5 and 10 mg/egg) were also tested in the comet assay, yielding
positive results, ie, 1.9- and 2.5-fold increases in the percentage
of tail-DNA, respectively (data not shown). The highest tested
dose of myristicin (50 mg/egg) also induced a significant (1.8fold) increase in the percentage of tail-DNA, while reducing
KOBETS ET AL.
|
305
TABLE 1. Total Results of AB Testing in TEGA
Compound
Dose, mg/egga
Vehicle, 20% HS15
Safrole
Safrole
Methyl eugenol
Methyl eugenol
Estragole
Estragole
Myristicin
Myristicin
Elemicin
Elemicin
Anethole
Anethole
Methyl isoeugenol
Methyl isoeugenol
Methyl isoeugenol
Eugenol
Eugenol
Eugenol
Isoeugenol
Isoeugenol
Isoeugenol
0
1
2
2
4
20
40
25
50
20
50
5
10
40
80
120
1
2.5
5
1
4
8
Dose, mM/egg
Viability, %
Comet
NPL
Rodent
Carcinogenicity Data
6.16
12.32
11.22
22.44
134.95
269.90
130.06
260.12
96.35
240.09
33.73
67.46
224.43
448.86
673.29
6.09
15.22
30.45
6.09
24.36
48.72
90 b
67
92
67
88
94
81
100
50
100
78
100
91
70
70
20
78
90
27
78
88
25
–
–
–
–
–
þ
þ
–
þ
þ
þ
–
–
–
–
N/M
–
–
N/M
–
–
N/M
–
N/M
þ
N/M
þ
N/M
þ
N/M
þ
N/M
þ
N/M
þ
N/M
–
N/M
N/M
–
N/M
N/M
–
N/M
–
þ
þ
þ
N/A
–
6
N/A
–
6
, negative; 6, equivocal; þ, positive; N/A, not available; N/M, not measured.
a
Administered in 3 daily injections.
b
Values presented as average.
TABLE 2. Normalization of DNA Adducts Levels to the Dose of AB
Tested in NPL
Compound
Dose,
Dose, m DNA Adducts
mg/egga M/egg 108ntsb
Adducts/
mM
Vehicle, 20% HS15
Safrole
Methyl eugenol
Estragole
Myristicin
Elemicin
Anethole
Methyl isoeugenol
Eugenol
Isoeugenol
0
2
4
40
50
50
10
80
2.5
4
0
0.25
8.25
6.99
2.90
45.92
0.11
0
0
0
0
12.32
22.44
269.90
260.12
240.09
67.46
448.86
15.22
24.36
0
3.07 6 0.47
185.06 6 34.88
1886.62 6 445.99
755.62 6 158.27
11 025.40 6 1907.47
7.5 6 1.05
0
0
0
nts, nucleotides.
a
b
Administered in three daily injections.
Values are expressed as the mean 6 SD.
viability by 50% (Figure 3). Likewise, elemicin at 20 and 50 mg/
egg also yielded positive statistically significant comet results,
however, not in a dose-related manner (1.65- and 1.5-fold
increase, respectively). Only the highest total dose induced an
increase in mortality rate by 22% (Figure 3).
Figure 3 also shows that dosing of turkey eggs with safrole,
methyl eugenol, anethole, methyl isoeugenol, eugenol, and isoeugenol had no effect on the percentage tail-DNA in comet
assay. Methyl eugenol, methyl isoeugenol, and isoeugenol
reduced survival of eggs (Table 1).
DNA adducts. NPL chromatograms (Figs. 4 and 5) show that
safrole, methyl eugenol, estragole, myristicin, elemicin, and
anethole at the highest doses tested caused formation of DNA
adducts in the livers of turkey fetuses. The highest levels of
adducts were detected with 50 mg/egg of elemicin (11 025 in 108
of normal nucleotides) followed by 40 mg/egg estragole (1886 in
108 of normal nucleotides), 50 mg/egg myristicin (755 in 108 of
normal nucleotides), and 4 mg/egg methyl eugenol (185 in 108 of
normal nucleotides) (Figure 5). Anethole at 10 mg/egg showed a
low level of DNA-binding (7.5 in 108 of normal nucleotides) and
safrole showed the lowest, but still statistically significant, levels of DNA-binding with dosing of 2 mg/egg (3 in 108 of normal
nucleotides) (Figure 5).
As can be seen in the NPL chromatograms (Figure 4), the patterns of adducts were specific for each compound.
Methyl isoeugenol, eugenol, and isoeugenol did not yield
any detectable DNA adducts (Figs. 4 and 5) at the highest doses
tested and thus were negative in the NPL assay (Table 1).
To compare DNA-binding between positive compounds
under the conditions of exposure, the total level of adducts at
the highest tested dose was normalized to molecular weight
(Table 2), with the caveat that adduct formation may not be linear. Anethole showed the lowest DNA-binding activity (0.11
adducts/mM), activity of safrole was almost two times higher
compared with anethole (0.25 adducts/mM), myristicin binding
activity was 2.90 adducts/mM, DNA-binding levels of estragole
and methyl eugenol were similar (6.99 and 8.25 adducts/mM,
respectively), elemicin still was the most active, inducing formation of 45.92 adducts/mM.
DISCUSSION
The present study assessed DNA damage by nine AB using
TEGA. Compounds known to be DNA-reactive carcinogens
(safrole, methyl eugenol, and estragole), compounds weakly
genotoxic in rodent bioassays (myristicin and elemicin) and an
306
|
TOXICOLOGICAL SCIENCES, 2016, Vol. 150, No. 2
FIG. 3. The percentage of DNA-in-tail in the livers of turkey egg fetuses exposed to alkenylbenzenes. The percentage of DNA-in-tail was determined using comet assay. The
results are presented as an average fold change in the percentage of tail-DNA in the turkey fetal livers from the groups dosed with AB relative to that in the corresponding
control group, which was assigned a value of 1. Values are mean 6 SD (n ¼ 3). *Significant (P < .05) difference from the control group. †Significant (P < .05) trend.
equivocal rodent carcinogen (anethole) yielded positive results
in at least one of the assays (Tables 1 and 2). This indicates their
ability to induce DNA strand breaks and DNA adduct formation
in the turkey fetal livers. Similar DNA-binding activity for most
compounds was observed in rats and mice. However, in the
livers of mice, DNA binding of myristicin, elemicin, and anethole was lower compared with that in turkey fetal livers and
not persistent (Phillips et al., 1984; Randerath et al., 1984). These
KOBETS ET AL.
|
307
FIG. 4. NPL chromatograms of DNA adducts in the livers of turkey egg fetuses exposed to alkenylbenzenes. NPL analysis of DNA adducts from the liver of turkey fetuses
3 h after the last injection with alkenylbenzenes obtained with NP1 enrichment. Adducts were resolved in the second and third directions of chromatography and are
indicated by arrows. Corresponding spots are absent in the livers of turkey in the control group.
findings demonstrate that turkey fetal liver can carry out
the two-step bioactivation of AB, ie, epoxidation and hydroxylation followed by sulfonation. In contrast, nongenotoxic
AB (methyl isoeugenol, eugenol, and isoeugenol) were negative
in TEGA (Tables 1 and 2), demonstrating the specificity of the
model response in distinguishing between genotoxic and nongenotoxic compounds with similar chemical structures.
Assays used in the current assessment, comet and NPL, are
widely used highly sensitive techniques for detection of chemical-induced DNA damage (Kawaguchi et al., 2010; Phillips et al.,
2000). With regard to current recommendations, the assays
were undertaken 3 h after the last injection in order to detect
DNA damage at the highest exposure to the test chemical or its
metabolites. Comet assay detects alkali-labile sites and sites of
DNA repair (Garaj-Vrhovac and Kopjar, 2003). Thus, compounds
positive in NPL assay, estragole, myristicin and elemicin,
yielded positive results in comet as well. In the case of the negative findings in comet assay for safrole, methyl eugenol and
anethole, the low levels of DNA adducts formed by these compounds (Table 2) suggest that possibly there is a threshold for
the comet assay detection of breaks resulting from DNA
adducts, ie, comet assay does not detect low levels of DNA
adducts. Thus, based on the current findings for methyl eugenol, the threshold would be 185 per 108 nucleotides. Another
explanation can be that the DNA adducts formed after dosing
with these compounds are not alkali-labile or due to the low
amount are more rapidly repaired compared with those induced
by estragole, myristicin, and elemicin dosing. In any event for
the studied AB, the NPL assay showed higher sensitivity compared with comet assay.
AB are established to be extensively metabolized by rodents
and humans (Anthony et al., 1987; Burkey et al., 2000; Gardner
et al., 1997; Howes et al., 1990; Paini et al., 2011; Phillips et al., 1984;
Randerath et al., 1984; Rietjens et al., 2005; Sangster et al., 1987;
Stening et al., 1997; Williams and Mattia, 2009; Wislocki et al.,
1977). The carcinogenicity of safrole, methyl eugenol and estragole is largely attributed to their bioactivation by cytochrome
P450 enzymes, leading to the formation of their proximate carcinogenic 10 -hydroxy metabolites followed by sulfonation to the
ultimate carcinogenic 10 -sulfoxy metabolites (Figure 2) (Anthony
et al., 1987; Borchert et al., 1973; Drinkwater et al., 1976; Gardner
et al., 1997; Wislocki et al., 1977). These unstable metabolites have
carcinogenic potential due to their high DNA-binding capacity,
which results in DNA adduct formation in rodents (Paini et al.,
2011; Phillips et al., 1984; Randerath et al., 1984; Williams et al.,
2013). Our findings, congruent with the data obtained from rodent
assays, demonstrate the genotoxic potential of safrole, methyl
eugenol and estragole, as they produced increases in the levels of
DNA adducts in turkey fetal livers. However, although in rodents
safrole has a very strong binding ability (Phillips et al., 1984;
Randerath et al., 1984), in TEGA this compound showed relatively
low DNA-binding levels. Importantly, the patterns of adducts
caused by the exposure to methyl eugenol and estragole were
similar to the patterns of adducts obtained from the in vivo studies of the effects of these compounds in the livers of F344 rats
(Ding et al., 2015; Williams et al., 2013). The presence of methyl
eugenol DNA adducts was also detected in the livers of patients
who underwent liver surgery due to various reasons, including
cholangiocellular carcinoma, hepatocellular carcinoma, and liver
metastases (Herrmann et al., 2013). Moreover, dosing with
308
|
TOXICOLOGICAL SCIENCES, 2016, Vol. 150, No. 2
FIG. 5. Total DNA adducts levels in the livers of turkey egg fetuses exposed to alkenylbenzenes. The formation of DNA adducts in turkey fetal livers was assessed using
the nucleotide
32
P-postlabeling (NPL) assay. The results are presented as a mean value of DNA adducts in 108 of normal nucleotides. nts, nucleotides. *Significant
(P < .05) difference from the control group.
estragole resulted in dose-dependent increase of DNA strand
breaks (Table 1). Similar results were observed in the studies of
in vivo estragole toxicity in rodents (Ding et al., 2015). These data
suggest that DNA adduct formation and, in the case of estragole,
DNA breakage may be causative factors in the hepatocarcinogenicity of these compounds. Additionally, positive outcomes for
these compounds in at least one of the assays serve as a proof
that turkey fetal livers are capable of performing xenobiotic metabolic activation and possess phase I enzymes that belong to cytochrome family and phase II enzymes that are crucial for
sulfonation. Interestingly, previously published data provide evidence of DNA strand break induced by safrole in liver-derived cell
culture systems (Bradley, 1985; Chiang et al., 2011); however,
in vivo safrole often yielded negative comet outcomes in liver tissue (Ding et al., 2015). Similarly, in TEGA safrole was negative in
the comet assay (Table 1).
The genotoxic and carcinogenic potential of myristicin is not
yet well studied. In a 12-month mouse assay conducted by
Miller et al. (1983), the compound did not show any activity for
the initiation of hepatic tumors. However, myristicin showed
DNA-binding activity in B6C3F1 mice (Phillips et al., 1984) and
has been predicted to be weakly hepatocarcinogenic in rodent
bioassay (Auerbach et al., 2010). The findings in the current
study, ie, positive results in comet and NPL (Table 1), support
the possibility of myristicin genotoxicity and carcinogenicity.
The chemical structure of elemicin possesses features that
suggest potential genotoxicity. Specifically, the compound has a
terminal (20 ,30 -position) double bond and does not possess a
phenolic hydroxyl group for conjugation (Figure 1). In male
mice, administration of elemicin did not result in a significant
increase in the formation of hepatic tumors; however, it was
found to be positive in DNA-binding assays with lower potency
compared with structurally related carcinogenic compounds
such as estragole (Hasheminejad and Caldwell, 1994; Miller
et al., 1983; Phillips et al., 1984; Randerath et al., 1984). In TEGA,
elemicin yielded positive outcomes in both comet and NPL
assays, and it appeared to have the greatest genotoxic potential
in turkey fetal livers among all AB tested in TEGA (Tables 1 and
2). Although a dose-related response in the comet assay was
not observed, the increase in the percentage of DNA-in-tail was
statistically significant compared with corresponding control
group and the viability at the dose 50 mg/egg was decreased. We
hypothesize that the slight reduction in the percentage of DNAin-tail at 50 mg/egg can be caused by decreased ability of hepatocytes to repair DNA damage due to higher toxicity of elemicin.
Thus, we provide evidence of elemicin genotoxicity that should
be considered in its safety evaluation, although, based on
kinetic models it has been suggested that elimicin poses a lower
priority for risk management (van den Berg et al., 2012).
Due to the position of the double bond in the side chain, the
bioactivation of anethole leads to the formation of 30 -hydroxymetabolite (Figure 2) that does not possess the metabolic potential of
the proximate carcinogenic 10 -hydroxymetabolite and more readily undergoes detoxification and elimination in the urine
(Caldwell and Sutton, 1988; Sangster et al., 1987). The epoxidation
of the side chain, the major metabolic route for anethole, was
believed to lead to cytotoxicity but not genotoxicity (Marshall and
Caldwell, 1991). Previously, anethole was not considered to be genotoxic (Gorelick, 1995; Howes et al., 1990). However, in our study
anethole induced formation of DNA adducts, showing that it has
genotoxic potential (Table 1), although when compared with
other compounds tested, its DNA-binding activity was the lowest
KOBETS ET AL.
(Table 2). Importantly, anethole has been shown to produce low
levels of DNA adducts in newborn and adult mice (Phillips et al.,
1984; Randerath et al., 1984) and to cause a very small increase in
the incidences of hepatocellular carcinoma in female rats
(Newberne et al., 1999; Truhaut et al., 1989). The synthetic metabolite trans-anethole oxide (or anethole epoxide) was mutagenic in
bacteria and increased the incidences of hepatomas in male
B6C3F1 mice and skin papillomas in female CD-1 mice (Kim et al.,
1999). Investigation of the metabolism of beta-asarone, a known
rodent hepatocarcinogenic AB with a similar chemical structure to
anethole, revealed that epoxidation of the side chain double bond
was the major metabolic route for the compound, resulting in the
formation of unstable and highly reactive epoxides which are ultimate carcinogens (Cartus et al., 2015). These data in combination
with the findings from the current study suggest that epoxidation
of anethole might lead to the formation of reactive genotoxic
metabolites.
Methyl isoeugenol also possesses the double bond at the
10 ,20 -position. Despite a lack of definitive evidence, methyl isoeugenol is claimed to be nongenotoxic (Hasheminejad and
Caldwell, 1994) and is approved by U.S. Food and Drug
Administration for human consumption (USFDA, 2010). The
administration of the compound in diet to Sprague Dawley
male and female rats at up to 300 mg/kg body weight/day did
not induce any adverse effects (Purchase et al., 1992). The data
from the published studies that investigated the metabolism of
methyl isoeugenol show that its major metabolite was a nongenotoxic 30 -hydroxy metabolite (Cartus et al., 2011), whereas
epoxidation of the side chain plays a minor role (Solheim and
Scheline, 1976). These peculiarities of methyl isoeugenol metabolism caused by chemical structure can explain the difference
in genotoxic potential of methyl isoeugenol and structurally
similar methyl eugenol, the latter showed neoplasm-initiating
activity in the livers of male F344 rats as a result of DNA-binding
(Williams et al., 2013). The findings in TEGA support the hypothesis for the lack of the genotoxicity of methyl isoeugenol (Table
1). When comparing these findings with the results of anethole
testing in TEGA, it can be concluded that despite the position of
alkenyl double bond in anethole, it is capable of forming reactive metabolites, possibly due to more active epoxidation, a
metabolic route which does not prevail in methyl isoeugenol
biotransformation.
Eugenol and isoeugenol both possess a free phenolic
hydroxyl group which is readily conjugated with glucuronic
acid or sulfate to form stable metabolites which are excreted in
the urine (Fischer et al., 1990). In rodents, 2-year carcinogenicity
studies of eugenol showed no significant increase in the incidences of malignant liver tumors and equivocal evidence of carcinogenic activity of isoeugenol in male rats and mice (NTP,
1983, 2010). It is possible that weak carcinogenicity of isoeugenol can be caused by other mechanisms, eg, oxidative DNA
damage. In rodents, eugenol and isoeugenol did not yield any
detectable DNA adducts (Phillips et al., 1984; Randerath et al.,
1984). Similarly, in TEGA both compounds did not induce DNA
damage, thus, strengthening the evidence for their lack of genotoxicity (Table 1).
In summary, our findings congruently with the data obtained
from previous studies confirm in a newly tested species the genotoxic potential of safrole, methyl eugenol, and estragole, supporting the interpretation that their toxicity is caused by chemical
specific DNA damage. Interestingly, myristicin, elemicin, and
anethole also yielded positive results in at least one of the assays
providing evidence of their genotoxicity. Lack of genotoxicity
of methyl isoeugenol, eugenol, and isoeugenol supports the
|
309
hypothesis that the genotoxicity of AB is determined by their
chemical structure, specifically, 10 ,20 -position of the double bond
in the side chain which is not favorable to formation of proximate
carcinogenic 10 -hydroxymetabolite or terminal epoxide, or the
presence of a free phenolic hydroxyl group, which leads to the
formation of a nontoxic glucuronide conjugate. These findings
in TEGA extend the range of compounds demonstrated to elicit
genotoxicity (Perrone et al., 2004; Williams et al., 2011).
FUNDING
This work was supported by Firmenich Inc.
ACKNOWLEDGMENT
The authors would like to thank Sean Taylor for advice on
structure of compounds and metabolites.
REFERENCES
Anthony, A., Caldwell, J., Hutt, A. J., and Smith, R. L. (1987).
Metabolism of estragole in rat and mouse and influence of
dose size on excretion of the proximate carcinogen 10 hydroxyestragole. Food Chem. Toxicol. 25, 799–806.
Auerbach, S. S., Shah, R. R., Mav, D., Smith, C. S., Walker, N. J.,
Vallant, M. K., Boorman, G. A., and Irwin, R. D. (2010).
Predicting the hepatocarcinogenic potential of alkenylbenzene flavoring agents using toxicogenomics and machine
learning. Toxicol. Appl. Pharmacol. 243, 300–314.
Borchert, P., Miller, J. A., Miller, E. C., and Shires, T. K. (1973). 10 Hydroxysafrole, a proximate carcinogenic metabolite of safrole in the rat and mouse. Cancer Res. 33, 590–600.
Bradley, M. O. (1985). Measurement of DNA single-strand breaks
by alkaline elution in rat hepatocytes. In Progress in Mutation
Research. Evaluation of Short-Term Tests for Carcinogens
(J. Ashby, F. J. de Serres, M. Draper, M. Ishidate Jr, B. H.
Margolin, B. E. Matter, and M. D. Shelby, Eds.), Vol. 5,
pp. 353–357. Elsevier, New York, NY.
Bristol, D. W. (2011). NTP 3-month toxicity studies of estragole
(CAS No. 140-67-0) administered by gavage to F344/N
rats and B6C3F1 mice. National Toxicology Program
Toxicity Report Series Number 82. NIH Publication No. 085966. pp. 1–111.
Burkey, J. L., Sauer, J. M., McQueen, C. A., and Sipes, I. G. (2000).
Cytotoxicity and genotoxicity of methyleugenol and related
congeners––A mechanism of activation for methyleugenol.
Mutat. Res. 453, 25–33.
Caldwell, J., and Sutton, J. D. (1988). Influence of dose size on the
disposition of trans-[methoxy-14C]anethole in human volunteers. Food Chem. Toxicol. 26, 87–91.
Cartus, A. T., Merz, K. H., and Schrenk, D. (2011). Metabolism of
methylisoeugenol in liver microsomes of human, rat, and
bovine origin. Drug Metab. Dispos. 39, 1727–1733.
Cartus, A. T., Stegmüller, S., Simson, N., Wahl, A., Neef, S., Kelm,
H., and Schrenk, D. (2015). Hepatic metabolism of carcinogenic beta-asarone. Chem. Res. Toxicol. 28, 1760–1773.
Chen, C. L., Chi, C. W., Chang, K. W., and Liu, T. Y. (1999). Safrolelike DNA adducts in oral tissue from oral cancer patients
with a betel quid chewing history. Carcinogenesis 20, 2331–
2334.
Chiang, S. Y., Lee, P. Y., Lai, M. T., Shen, L. C., Chung, W. S.,
Huang, H. F., Wu, K. Y., and Wu, H. C. (2011). Safrole-20 ,30 -oxide induces cytotoxic and genotoxic effects in HepG2 cells
and in mice. Mutat. Res. 726, 234–241.
310
|
TOXICOLOGICAL SCIENCES, 2016, Vol. 150, No. 2
Ding, W., Levy, D. D., Bishop, M. E., Pearce, M. G., Davis, K. J.,
Jeffrey, A. M., Duan, J. D., Williams, G. M., White, G. A., LynCook, L. E., et al. (2015). In vivo genotoxicity of estragole in
male F344 rats. Environ. Mol. Mutagen. 56, 356–365.
Drinkwater, N. R., Miller, E. C., Miller, J. A., and Pitot, H. C. (1976).
Hepatocarcinogenicity of estragole (1-allyl-4-methoxybenzene) and 10 -hydroxyestragole in the mouse and mutagenicity of 10 -acetoxyestragole in bacteria. J. Natl Cancer Inst. 57,
1323–1331.
Enzmann, H., Brunnemann, K., Iatropoulos, M., Shpyleva, S.,
Lukanova, N., Todor, I., Moore, M., Spicher, K., Chekhun, V.,
Tsuda, H., et al. (2013). Inter-laboratory comparison of turkey
in ovo carcinogenicity assessment (IOCA) of hepatocarcinogens. Exp. Toxicol. Pathol. 65, 729–735.
Enzmann, H., Kaliner, G., Watta-Gebert, B., and Loser, E. (1992).
Foci of altered hepatocytes induced in embryonal turkey
liver. Carcinogenesis 13, 943–946.
Fischer, I. U., von Unruh, G. E., and Dengler, H. J. (1990). The metabolism of eugenol in man. Xenobiotica 20, 209–222.
Garaj-Vrhovac, V., and Kopjar, N. (2003). The alkaline Comet assay as biomarker in assessment of DNA damage in medical
personnel occupationally exposed to ionizing radiation.
Mutagenesis 18, 265–271.
Gardner, I., Wakazono, H., Bergin, P., de Waziers, I., Beaune, P.,
Kenna, J. G., and Caldwell, J. (1997). Cytochrome P450 mediated bioactivation of methyleugenol to 10 -hydroxymethyleugenol in Fischer 344 rat and human liver microsomes.
Carcinogenesis 18, 1775–1783.
Gorelick, N. J. (1995). Genotoxicity of trans-anethole in vitro.
Mutat. Res. 326, 199–209.
Guenthner, T. M., and Luo, G. (2001). Investigation of the role of the
20 ,30 -epoxidation pathway in the bioactivation and genotoxicity of dietary allylbenzene analogs. Toxicology 160, 47–58.
Hamilton, J. W., Denison, M. S., and Bloom, S. E. (1983).
Development of basal and induced aryl hydrocarbon (benzo[a]pyrene) hydroxylase activity in the chicken embryo in
ovo. Proc. Natl. Acad. Sci. U.S.A. 80, 3372–3376.
Hasheminejad, G., and Caldwell, J. (1994). Genotoxicity of the
alkenylbenzenes alpha- and beta-asarone, myristicin and
elimicin as determined by the UDS assay in cultured rat hepatocytes. Food Chem. Toxicol. 32, 223–231.
Herrmann, K., Schumacher, F., Engst, W., Appel, K. E., Klein, K.,
Zanger, U. M., and Glatt, H. (2013). Abundance of DNA adducts of methyleugenol, a rodent hepatocarcinogen, in human liver samples. Carcinogenesis 34, 1025–1030.
Howes, A. J., Chan, V. S., and Caldwell, J. (1990). Structurespecificity of the genotoxicity of some naturally occurring
alkenylbenzenes determined by the unscheduled DNA synthesis assay in rat hepatocytes. Food Chem. Toxicol. 28, 537–542.
International Agency for Research on Cancer (IARC) (1976).
Safrole, isosafrole and dihydrosafrole. In Some Naturally
Occurring Substances, Vol. 10, pp. 231–244. International
Agency for Research on Cancer, Lyon, France. IARC
Monographs on the Evaluation of Carcinogenic Risk of
Chemicals to Humans.
Kawaguchi, S., Nakamura, T., Yamamoto, A., Honda, G., and
Sasaki Y. F. (2010). Is the comet assay a sensitive procedure
for detecting genotoxicity? J. Nucleic Acids 2010, 541050.
Kim, S. G., Liem, A., Stewart, B. C., and Miller, J. A. (1999). New
studies on trans-anethole oxide and trans-asarone oxide.
Carcinogenesis 20, 1303–1307.
Lancaster, M. C., Jenkins, F. P., and Philp, J. (1961). Toxicity associated with certain samples of groundnuts. Nature 192, 1095–
1096.
Lee, J. M., Liu, T. Y., Wu, D. C., Tang, H. C., Leh, J., Wu, M. T., Hsu,
H. H., Huang, P. M., Chen, J. S., Lee, C. J., et al. (2005). SafroleDNA adducts in tissues from esophageal cancer patients:
Clues to areca-related esophageal carcinogenesis. Mutat. Res.
565, 121–128.
Marshall, A. D., and Caldwell, J. (1991). Modulation of epoxide
metabolism enhances the cytotoxicity but not the genotoxicity of the food flavour anethole. Hum. Exp. Toxicol. 10, 486–
487.
Miller, E. C., Swanson, A. B., Phillips, D. H., Fletcher, T. L., Liem,
A., and Miller, J. A. (1983). Structure-activity studies of the
carcinogenicities in the mouse and rat of some naturally occurring and synthetic alkenylbenzene derivatives related to
safrole and estragole. Cancer Res. 43, 1124–1134.
Miller, J. A., Swanson, A. B., and Miller, E. C. (1979). The metabolic
activation of safrole and related naturally occurring alkenylbenzenes in relation to their carcinogenesis by these agents. In
Naturally Occuring Carcinogens—Mutagens and Modulators of
Carcinogenesis. (E. C. Miller, Ed.), pp. 111–125. Japanese Scientific
Societies Press, Tokyo/University Park Press, Baltimore.
National Toxicology Program (NTP) (1983). Carcinogenesis studies
of eugenol (CAS No. 97-53-0) in F344/N rats and B6C3F1 mice
(feed studies). National Toxicology Program Toxicity Report
Series Number 223, NIH publication No. 84-1779. pp. 1–159.
National Toxicology Program (NTP) (2000). NTP toxicology and
carcinogenesis studies of methyleugenol (CAS NO. 93-15-2)
in F344/N rats and B6C3F1 mice (gavage studies). National
Toxicology Program Toxicity Report Series Number 491, NIH
publication No. 00-3950. pp. 1–412.
National Toxicology Program (NTP) (2010). Toxicology and carcinogenesis studies of isoeugenol (CAS No. 97-54-1) in F344/N
rats and B6C3F1 mice (gavage studies). National Toxicology
Program Toxicity Report Series Number 551, NIH Publication
No. 10-5892. pp. 1–178.
Newberne, P., Smith, R. L., Doull, J., Goodman, J. I., Munro, I. C.,
Portoghese, P. S., Wagner, B. M., Weil, C. S., Woods, L. A.,
Adams, T. B., et al. (1999). The FEMA GRAS assessment of
trans-anethole used as a flavor substance. Flavour and Extract
Manufacturer’s Association. Food Chem. Toxicol. 37, 789–811.
OECD (2014). Test No. 489: In Vivo Mammalian Alkaline Comet
Assay. OECD Guidelines for the Testing of Chemicals. Section
4. OECD Publishing, Paris.
Paini, A., Scholz, G., Marin-Kuan, M., Schilter, B., O’Brien, J., van
Bladeren, P. J., and Rietjens, I. M. (2011). Quantitative comparison between in vivo DNA adduct formation from exposure
to selected DNA-reactive carcinogens, natural background
levels of DNA adduct formation and tumor incidence in rodent bioassays. Mutagenesis 26, 605–618.
Perrone, C. E., Ahr, H. J., Duan, J. D., Jeffrey, A. M., Schmidt, U.,
Williams, G. M., and Enzmann, H. H. (2004). Embryonic turkey
liver: Activities of biotransformation enzymes and activation
of DNA-reactive carcinogens. Arch. Toxicol. 78, 589–598.
Phillips, D. H., Farmer, P. B., Beland, F. A., Nath, R. G., Poirier, M. C.,
Reddy, M. V., and Turteltaub, K. W. (2000). Methods of DNA adduct determination and their application to testing compounds for genotoxicity. Environ. Mol. Mutagen. 35, 222–233.
Phillips, D. H., Reddy, M. V., and Randerath, K. (1984). 32P-postlabelling analysis of DNA adducts formed in the livers of
animals treated with safrole, estragole and other naturallyoccurring alkenylbenzenes. II. Newborn male B6C3F1 mice.
Carcinogenesis 5, 1623–1628.
Purchase, R., Ford, G. P., Creasy, D. M., Brantom, P. G., and
Gangolli, S. D. (1992). A 28-day feeding study with methyl isoeugenol in rats. Food Chem. Toxicol. 30, 475–481.
KOBETS ET AL.
Randerath, K., Haglund, R. E., Phillips, D. H., and Reddy, M. V.
(1984). 32P-postlabeling analysis of DNA adducts formed in
the livers of animals treated with safrole, estragole and other
naturally-occurring alkenylbenzenes. I. Adult female CD-1
mice. Carcinogenesis 5, 1613–1622.
Randerath, K., Reddy, M. V., and Gupta, R. C. (1981). 32P-Labeling
test for DNA damage. Proc. Natl. Acad. Sci. U.S.A. 78, 6126–
6129.
Rietjens, I. M., Boersma, M. G., van der Woude, H., Jeurissen, S. M.
F., Schutte, M. E., and Alink, G. M. (2005). Flavonoids and alkenylbenzenes: mechanisms of mutagenic action and carcinogenic risk. Mutat. Res. 574, 124–138.
Rietjens, I. M., Cohen, S. M., Fukushima, S., Gooderham, N. J.,
Hecht, S., Marnett, L. J., Smith, R. L., Adams, T. B., Bastaki, M.,
Harman, C. G., et al. (2014). Impact of structural and metabolic variations on the toxicity and carcinogenicity of hydroxy- and alkoxy-substituted allyl- and propenylbenzenes.
Chem. Res. Toxicol. 27, 1092–1103.
Rifkind, A. B., Troeger, M., and Petschke, T. (1979). Equality of the
rates of mixed function oxidation in livers of male and female chick embryos. Biochem. Pharmacol. 28, 1681–1683.
Sangster, S. A., Caldwell, J., Anthony, A., Hutt, A. J., and Smith, R.
L. (1987). The metabolic disposition of [methoxy-14C]-labeled
trans-anethole, estragole and p-proylanisole in human volunteers. Xenobiotica 17, 1223–1232.
Sekizawa, J., and Shibamoto, T. (1982). Genotoxicity of safrolerelated chemicals in microbial test systems. Mutat. Res. 101,
127–140.
Smith, R. L., Adams, T. B., Doull, J., Feron, V. J., Goodman, J. I.,
Marnett, L. J., Portoghese, P. S., Waddell, W. J., Wagner, B. M.,
Rogers, A. E., et al. (2002). Safety assessment of allylalkoxybenzene derivatives used as flavouring substances––Methyl
eugenol and estragole. Food Chem. Toxicol. 40, 851–870.
Solheim, E., and Scheline, R. R. (1976). Metabolism of alkenebenzene derivatives in the rat. II. Eugenol and isoeugenol methyl
ethers. Xenobiotica 6, 137–150.
Stening, P., Gardner, I., Kenna, J. G., Coughtric, M. W. H., and
Caldwell, J. (1997). Formation of alkenylbenzene macromolecular adducts in human fibroblast V79 cells transfected
with human sulfotransferases. Hum. Exp. Toxicol. 16, 62.
Tice, R. R., and Strauss, G. H. (1995). The single cell gel electrophoresis/comet assay: a potential tool for detecting radiation-induced
DNA damage in humans. Stem Cells 13, 207–214.
|
311
Truhaut, R., Le Bourhis, B., Attia, M., Glomot, R., Newman, J., and
Caldwell, J. (1989). Chronic toxicity/carcinogenicity study of
trans-anethole in rats. Food Chem. Toxicol. 27, 11–20.
U.S. Food and Drug Administration (USFDA) (2010). Code of
Federal Regulations. Title 21, Vol. 3, 21CFR172.515. Available at:
https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/
CFRSearch.cfm?fr=172.515.
van den Berg, S. J., Punt, A., Soffers, A. E., Vervoort, J., Ngeleja, S.,
Spenkelink, B., and Rietjens, I. M. (2012). Physiologically
based kinetic models for the alkenylbenzene elemicin in rat
and human and possible implications for risk assessment.
Chem. Res. Toxicol. 25, 2352–2367.
van der Leede, B. J., Doherty, A., Guérard, M., Howe, J.,
O’Donovan, M., Plappert-Helbig, U., and Thybaud, V. (2014).
Performance and data interpretation of the in vivo comet assay
in
pharmaceutical
industry:
EFPIA
survey
results. Mutat. Res. Genet. Toxicol. Environ. Mutagen. 775–776,
81–88.
Williams, G. M., and Mattia, A. (2009). Alkoxy-substituted allylbenzenes present in foods and essential oils and used as flavouring agents. In Data Safety Evaluation of Certain Food
Additives/Prepared by the Sixty-Ninth Meeting of the Joint
FAO/WHO Expert Committee on Food Additives (JECFA).
WHO food additives series 60. pp. 352–480.
Williams, G. M., Duan, J. D., Brunnemann, K. D., Iatropoulos, M.
J., Vock, E., and Deschl, U. (2014). Chicken fetal liver DNA
damage and adduct formation by activation-dependent
DNA-reactive carcinogens and related compounds of several
structural classes. Toxicol. Sci. 141, 18–28.
Williams, G. M., Iatropoulos, M. J., Jeffrey, A. M., and Duan, J. D.
(2013). Methyleugenol hepatocellular cancer initiating effects
in rat liver. Food Chem. Toxicol. 53, 187–196.
Williams, J. G., Deschl, U., and Williams, G. M. (2011). DNA damage in fetal liver cells of turkey and chicken eggs dosed with
aflatoxin B1. Arch. Toxicol. 85, 1167–1172.
Wislocki, P. G., Miller, E. C., Miller, J. A., McCoy, E. C., and
Rosenkranz, H. S. (1977). Carcinogenic and mutagenic activities of safrole, 10 -hydroxysafrole, and some known or possible metabolites. Cancer Res. 37, 1883–1891.
Zhou, G. D., Moorthy, B., Bi, J., Donnelly, K. C., and Randerath, K.
(2007). DNA adducts from alkoxyallylbenzene herb and spice
constituents in cultured human (HepG2) cells. Environ. Mol.
Mutagen. 48, 715–721.