Toxicology Letters 180 (2008) 131–136

Contents lists available at ScienceDirect

Toxicology Letters
journal homepage: www.elsevier.com/locate/toxlet

Risk assessment of botanicals and botanical preparations intended

for use in food and food supplements: Emerging issues
Ivonne M.C.M. Rietjens a,∗ , Wout Slob b,e , Corrado Galli c , Vittorio Silano d
a
Division of Toxicology, Wageningen University, Tuinlaan 5, 6703HE Wageningen, The Netherlands
b
National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands
c
Laboratory of Toxicology, Department of Pharmacological Sciences, University of Milan, Italy
d
National Institute for the Promotion Migrant’s Health and the Control of Poverty-related Diseases, Rome, Italy
e
Institute of Risk Assessment Sciences (IRAS), University of Utrecht, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: At present there is a growing interest for use of botanicals and botanical ingredients in medicines, for
Available online 17 June 2008 teas or in foods and in food supplements. In addition, a number of plant-derived food items form an
integral part of regular human diets. Currently, there is an increasing awareness among safety experts
Keywords: and regulators of risks associated with the use of botanicals and botanical ingredients in food including
Botanicals food supplements. It is becoming clear that “natural” does not equal “safe” and that, in modern society,
Risk assessment
adverse health effects can occur as a result of (mis)use. With the growing awareness of these issues efforts
MOE approach
to ensure safety of botanicals and botanical ingredients are also increasing. Several guidance documents
Methyleugenol
Allylalkoxybenzenes
on safety assessment of botanicals and botanical preparations to be used as ingredients in food and food
supplements have been published, although, at present, relevant legislative frameworks and guidances for
risk assessment are not established yet. Furthermore, when defining possible guidance documents for risk
assessment of botanicals, several issues emerge that need to be developed beyond the present state-of-
the-art. The present paper describes some of the issues to be considered and developed to a further extent
to improve risk assessment of botanicals and botanical preparations, illustrated by examples based on
some allylalkoxybenzenes. It is concluded that, for an improved and more accurate future risk assessment
of botanicals, it is necessary to further develop and validate: (i) the use of the margin of exposure (MOE)
concept for compounds that are both genotoxic and carcinogenic; (ii) new ways to quantify and incorporate
matrix effects into risk assessment strategies; (iii) the use of analytical chemistry approaches, enabling
complete chemical characterisation of complex mixtures. Defining new approaches in risk assessment
would be in line with the inspiring attitude of the late Professor Robert Kroes, who, for example by
supporting the threshold of toxicological concern (TTC) concept, was a pioneer for development and
implementation of new paradigms in the field of risk assessment and food safety.
© 2008 Elsevier Ireland Ltd. All rights reserved.

1. Introduction In a discussion paper the Scientific Committee of EFSA expressed

Many consumers equate “natural” with “safe” when considering
plant-based food supplements or drug preparations. Unfortunately,
the assumption that natural products are by definition safe is false.
In fact, in spite of a long history of safe use of many botanical or
herb-based food items, some botanicals and botanical preparations
contain individual ingredients known to be toxic and even geno-
toxic and carcinogenic, which may become reasons of concern at
specific exposure levels.
∗ Corresponding author. Tel.: +31 317 483971; fax: +31 317 484931.
E-mail address: ivonne.rietjens@wur.nl (I.M.C.M. Rietjens).
concerns about quality and safety issues of botanicals and botanical
preparations (EFSA, 2004a). As the market volumes and the variety
of products expand, so does the need for a better characterisation
of the range of botanicals and botanical preparations on the market,
and for harmonising the risk assessment for these products.
For several botanical products regulators are already aware of
the problems encountered and have taken or are considering appro-
priate regulatory actions to protect the public. Such regulatory
actions vary (Rietjens et al., 2005) from (i) setting tolerable daily
intakes (TDIs) (such as for example for coumarin) (EFSA, 2004b);
(ii) applying limits for the presence of a compound in foods and
beverages (such as for HCN, thujone and glycoalkaloids); (iii) devel-
oping safe upper limits (such as for beta-carotene); (iv) informing

0378-4274/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.toxlet.2008.05.024
132 I.M.C.M. Rietjens et al. / Toxicology Letters 180 (2008) 131–136
Fig. 1. Structural formulas of the six allylalkoxybenzenes to be evaluated by JECFA.
the public to be cautious and aware of possible adverse side effects
(as for St. John’s wort, and glycyrrhizinic acid); or (v) taking specific
plant varieties and/or their ingredients away from the market (such
as for aristolochic acids, pyrrolizidine alkaloids and kava-kava).
In addition, several agencies and other organizations have con-
ducted activities related to the possible legislative framework and
guidance for risk assessment of botanicals and botanical prepara-
tions. These include, for example, the European Medicines Agency
(EMEA), in particular the Committee on Herbal Medicinal Products
(EMEA, 2006), the Council of Europe (Council of Europe, 2005),
and ILSI Europe (Schilter et al., 2003). Several of these activities
have resulted in proposals for guidance for safety assessment of
botanicals and botanical preparations to be used as ingredients in
food and food supplements (EFSA, 2007). Generally, these guidance
documents list topics to be considered including requirements for
(i) adequate specification and characterisation of the ingredient,
including identification of the starting material and processes
applied to it;
(ii) description of any history of safe use;
(iii) information on intended use and use levels and estimated
dietary exposure resulting from these uses;
(iv) hazard identification including data from toxicity and genotox-
icity testing including any available human data.
In spite of these proposed guidelines for risk assessment of
botanicals and botanical preparations, several issues in the field of
Fig. 2. Summary of the results from an NTP study with methyleugenol, revealing an increase in the incidence of hepatocellular carcinomas upon gavage dosing in (A) rats
and (B) mice (NTP, 2000). For actual incidences in the rat data see Table 1.
risk assessment remain to be discussed and developed beyond the
present state-of-the-art. The present paper presents some issues to
be considered and developed to a further extent in risk assessment
of botanicals, illustrated by examples based on the allylalkoxyben-
zenes.
2. Naturally occurring compounds that are both genotoxic
and carcinogenic
A major issue is the question on how to deal with botanicals
and botanical ingredients that contain chemicals that are both
genotoxic and carcinogenic. Such compounds include, for exam-
ple, the allylalkoxybenzenes estragole, methyleugenol, elemicin,
tetramethoxyalkylbenzene, safrole, myristicin and apiole (Fig. 1).
The list of substances scheduled for evaluation or re-evaluation
at the 69th meeting of the Joint FAO/WHO Expert Committee on
Food Additives (JECFA) contains six of these allylalkoxybenzenes.
Several of these allylalkoxybenzenes have been shown to be geno-
toxic and carcinogenic (SCF, 2001a,b,c and references therein). As
an example, Fig. 2 and Table 1 present the results from a 2-year
NTP study (NTP, 2000) in which methyleugenol was administered
by gavage to rats and mice, revealing a dose-dependent increase
in formation of hepatocellular carcinomas, especially in rats. The
SCF has previously launched scientific evaluations on three of these
allylalkoxybenzenes including safrole (SCF, 2001a), estragole (SCF,
2001b), and methyleugenol (SCF, 2001c). The SCF concluded that
safrole, estragole and methyleugenol are genotoxic and carcino-
genic, and recommended restrictions in use of the compounds. On
the other hand, these compounds are well known constituents of a
wide range of botanicals and botanical ingredients in the food chain
including herbs like nutmeg, cinnamon, anise star, tarragon, sweet
basil, sweet fennel and anise vert and products derived from them
like pesto and essential oils (SCF, 2001a,b,c) resulting in exposure
to the respective compounds.
3. Margin of exposure (MOE) approach
Risk assessment strategies need to be developed beyond the
present state-of-the-art to define adequate and widely accepted
methods for risk assessment of botanical compounds that are both
genotoxic and carcinogenic.
The margin of exposure approach was developed by EFSA as a
harmonised approach for risk assessment of substances which are
 I.M.C.M. Rietjens et al. / Toxicology Letters 180 (2008) 131–136 133

Table 1
Summary of the data of the NTP study on incidence of hepatocellular carcinomas in male and female rats exposed to increasing dose of methyleugenol for 5 days a week by
gavage for 105 weeks (NTP, 2000)

Dose No. of animals Incidence hepatocellular carcinomas Gender No. of animals Incidence hepatocellular carcinomas Gender

0 50 2 Male 50 0 Female
37 50 3 Male 50 0 Female
75 50 14 Male 49 4 Female
150 50 25 Male 49 8 Female
300 50 36 Male 50 22 Female

both genotoxic and carcinogenic (EFSA, 2005). It is presently often
assumed that the exposure to such compounds is undesirable and
that exposure should be as low as reasonable achievable (ALARA).
The EFSA opinion (EFSA, 2005) stated that substances which are
both genotoxic and carcinogenic should not be approved for delib-
erate addition to foods or for any use in the food chain that could
result in their presence in food. The MOE approach was developed
to be applied in cases where there is a need for an assessment and
guidance on the risks to those who are, or have been, unintention-
ally exposed, through food, to substances that are both genotoxic
and carcinogenic.
The MOE approach uses a reference point, usually taken from
data from an animal experiment that represents a dose causing
a low but measurable cancer response. It can be for example the
BMDL10 , the lower confidence bound of the benchmark dose that
gives 10% (extra) cancer incidence (BMD10 ). The MOE is defined as
the ratio between this reference point, the BMDL10 , and the esti-
mated dietary intake (EDI) in humans.
EFSA has already clarified that the reference point is to be
compared to various dietary intake estimates in humans, taking
into account different consumption patterns (EFSA, 2005). This
provides the opportunity to calculate the MOE for intake of an
allylalkoxybenzene from various dietary sources providing a frame-
work to evaluate whether intake of the genotoxic and carcinogenic
ingredients from a particular botanical use or use level would be
considered a priority for risk management as compared to its intake
from other sources. Use of the MOE would even allow to evalu-
ate the priority for risk management of the botanical ingredient
versus other priorities in the field of genotoxic and carcinogenic
compounds in food.
To further illustrate this, Table 1 presents the data describing
the incidence of hepatocellular carcinomas in the NTP study with
male and female rats that were exposed by gavage for 5 days per
week for 105 weeks to methyleugenol at dose levels of 0 mg/kg bw,
37 mg/kg bw, 75 mg/kg bw, 150 mg/kg bw and 300 mg/kg bw (NTP,
2000). The study also described increased incidence of liver ade-
nomas but since these are benign neoplastic effects, the present
MOE evaluation focuses on the malignant hepatocellular carci-
nomas. Table 2 describes the results of a BMD analysis of these
data using BMDS version 1.4.1c and/or PROAST software. From
this it is concluded that the BMDL10 values vary between about
22–32 mg/kg bw for male and about 77–79 mg/kg bw for female
rats.
The SCF (SCF, 2001c) presented an intake estimate of
methyleugenol based on maximum use levels of methyleugenol
obtained from source materials, i.e. essential oils and considering
the following food categories: non-alcoholic beverages (includ-
ing all soft drinks and fruit juices), alcoholic beverages (liqueurs
only), ices (including ice cream and ice lollies), candy (excluding
chocolate), baked goods, gelatine-based desserts, meat products
and condiments and relishes (including sauces and spreads). Using
these assumptions, the average intake (for consumers only) was
reported to amount to 0.19 mg/(kg bw day) and the 97.5th per-
centile was 0.53 mg/(kg bw day), respectively.
Comparison of these BMDL10 values to the average and to
the 97.5th percentile intake estimates made by the SCF for
methyleugenol, leads to MOE values of 116–168 for males and
405–416 for females in the case of mean intake and to 42–60
for males and 145–149 for females in the case of the 97.5th per-
centile intake, which is considered as a high priority for risk
management. When the MOE would be calculated based on
the intake of methyleugenol from consumption of spices alone,
which would be between 0.5 ␮g/(kg bw day) and 5.0 ␮g/(kg bw day)
(Smith et al., 2002), the MOE would amount to 4400–64,000 for
males and 15,400–158,000 for females, indicating that the use of
methyleugenol containing spices should be considered as a low pri-
ority for risk management. This example illustrates that the MOE
approach may be of help for setting priorities in risk management
of botanicals and botanical ingredients containing compounds that
are both genotoxic and carcinogenic. The example also illustrates
that it is important to consider the combined intake from different
sources as well as the relative contribution of each source.

Table 2
Results of a BMD analysis of the NTP data on incidence of hepatocellular carcinomas in male and female rats exposed to increasing dose of methyleugenol for 5 days a week
by gavage for 105 weeks (Table 1, NTP, 2000) using BMDS version 1.4.1c and/or PROAST software and the default settings of extra risk, a Benchmark Response (BMR) of 10%
and a 95% confidence limit

Rats gender Model No. of parameters Log likelihood Accepted BMD10 (mg/(kg bw day)) BMDL10 (mg/(kg bw day))

Males Null 1 −156.72

Males Full 5 −113.70
Males One-stage 2 −116.34 Yes 26.9 22.0
Males Gamma 3 −115.25 Yes 42.1 24.5
Males Log-logistic 3 −114.73 Yes 44.7 28.9
Males Log-probit 3 −114.46 Yes 46.7 31.5
Males Weibull 3 −115.44 Yes 39.4 23.9

Females Null 1 −69.96

Females Full 5 −99.12
Females Two-stage 3 −71.27 Yes 113 77.9
Females Gamma 2 −71.03 Yes 107.5 78.5
Females log-logistic 2 −71.04 Yes 106.9 78.0
Females Log-probit 2 −70.86 Yes 102.9 77.2
Females Weibull 2 −71.11 Yes 108.7 78.0
134 I.M.C.M. Rietjens et al. / Toxicology Letters 180 (2008) 131–136
Fig. 3. Bioactivation pathway of estragole, also relevant for the other allylalkoxybenzenes.
The analysis carried out above may help establishing whether
the risk assessment of (new) uses of botanicals and/or botanical
ingredients or even of new botanicals could be based on the fact
that a newly proposed use or use levels would result in a negligible
increase of the intake of a genotoxic carcinogen as compared to
already established intakes from accepted sources.
4. Matrix effects
Another important aspect that should be taken into account
when assessing the risk of naturally occurring toxic or carcinogenic
substances is whether results from long-term animal studies with
pure compounds dosed by gavage without the presence of a normal
food matrix, represent a good starting point for the risk assess-
ment and the MOE approach. For example, a slow or incomplete
release of the ingredient from the matrix and/or inhibition of spe-
cific intestinal carriers involved in active uptake of an ingredient by
compounds present in the matrix, may result in reduced bioavail-
ability of a compound as compared to the bioavailability of the same
compound when dosed as a pure compound by gavage. There seems
to be agreement that elucidation of possible matrix effects on the
bioavailability and biological activity of key constituents in botani-
cals may play an important role in the risk assessment of a botanical
or botanical ingredient. Schilter et al. (2003) already concluded that
such a matrix interaction would raise serious questions about the
use of toxicity data of the pure compound for risk assessment of the
compound within the complex food matrix. However, actual data
or paradigms that describe how this should be taken into account
need to be further defined in the near future.
In addition to interactions at the level of bioavailability, inter-
actions may also occur at other levels. Such an example can be
found in recent data on a possible matrix effects on the bioac-
tivation of the allylalkoxybenzenes that need bioactivation to
become genotoxic. Fig. 3 presents an overview of the bioactiva-
tion pathway of estragole, also representative for the bioactivation
of the other allylalkoxybenzenes. The 1 -hydroxymetabolites of
the allylalkoxybenzenes are the supposed proximate carcinogens;
1 -hydroxestragole has been found in urine of men dosed with
1 ␮g estragole/kg bw (Sangster et al., 1987). Bioactivation to the
ultimate carcinogen requires the involvement of sulfotransferases
(SULTs) converting the 1 -hydroxymetabolite to a 1 -sulfooxy
metabolite that is unstable and decomposes to the carbocation that
is the ultimate electrophilic and carcinogenic metabolite (Wiseman
et al., 1987) (Fig. 3).
Recently, Jeurissen et al. (2008) demonstrated that the level
of DNA binding of the proximate carcinogenic metabolite 1 -
hydroxyestragole to DNA in vitro but also to DNA in intact HepG2
human hepatoma cells could be inhibited by a methanolic basil
extract (Fig. 4). A similar effect could be obtained by adding the SULT
inhibitor pentachlorophenol to the incubation medium (Fig. 4).
Because the inhibition by basil extract resembles the inhibition
by the SULT inhibitor pentachlorophenol and because the inhibi-
tion was not observed in incubations with the direct electrophile
1 -acetoxyestragole, it was concluded that the inhibition by the
basil extract occurs at the level of the SULT-mediated bioactiva-
tion of 1 -hydroxyestragole to 1 -sulfooxyestragole (Jeurissen et al.,
2008). Although it remains to be established whether a similar
inhibition will occur in the in vivo situation, the inhibition of SULT-
mediated bioactivation of 1 -hydroxyestragole by basil ingredients
suggests that the possibilities for bioactivation and subsequent
adverse effects may be lower when estragole is dosed in a matrix of
other basil ingredients than what would be expected on the basis
of experiments dosing estragole as a single compound.
In contrast to the findings with this methanolic basil extract,
Müller et al. (1994) showed that the genotoxic potential of estragole
is not masked by ingredients of basil oil. The genotoxic potentials
of basil oil and estragole were compared in the unscheduled DNA
Fig. 4. Effect of basil extract on binding of 1 -hydroxyestragole to DNA of HepG2
human hepatoma cells (Jeurissen et al., 2008). PCP stands for pentachlorophenol a
well-known sulfotransferase inhibitor.
 I.M.C.M. Rietjens et al. / Toxicology Letters 180 (2008) 131–136
synthesis (UDS) test, using basil oil with an estragole content of 88%,
and it was concluded that basil oil induced UDS in the same dose
range as estragole (Müller et al., 1994). This lack of protective effect
in basil oil may be related to the high ratio between estragole (88%)
and other herbal oil ingredients (12%) in the essential oil fraction
of basil compared to the ratio between estragole and other herbal
ingredients in the total herb.
From these examples it becomes clear that when a matrix effect
is advocated to support the safety of a botanical or a botanical ingre-
dient, experimental and/or other data need to be provided that
support the occurrence of the matrix effect in vivo at relevant lev-
els of intake and with the botanicals or botanical preparation of
interest. Clearly this raises a practical problem because one cannot
imagine that for each estragole-containing botanical preparation in
vivo studies characterising and quantifying the matrix effect of that
specific preparation can be performed. Therefore research should
focus on developing and validating methods that allow judgement
of the consequences of matrix effects for risk assessment of botani-
cals and botanical ingredients based on analytical characterisation
of the major compounds responsible for the effect. One could fore-
see, for example, to identify the actual compound present in the
basil extract that is responsible for the inhibition of the in vitro DNA
binding of 1 -hydroxyestragole followed by validation of the rele-
vance of the interaction of this SULT inhibitor and estragole in an in
vivo study. The safety of estragole-containing botanicals could sub-
sequently be judged and possibly regulated taking the relative ratio
between estragole and the respective SULT inhibitor into account.
Accepting and using such approaches in risk assessment requires
further development of the present state-of-the-art. Some ongo-
ing studies at the Wageningen University (The Netherlands) may
be helpful in this context.
5. Use of analytical chemistry as a basis for risk assessment
The further development of the use of detailed chemical anal-
ysis in safety assessment of botanicals could be based on a recent
proposal for a procedure for the safety evaluation of natural flavour
complexes used as ingredients in food, focussing on essential oils
(Smith et al., 2005). The approach is an extension of the cur-
rent procedure used by several authorities for the evaluation of
chemically defined flavouring substances. It relies on the almost
complete chemical characterisation of the natural flavour complex,
and the grouping of all the chemicals into 36 chemical groups
of compounds. The safety of the intake of each congeneric group
resulting from consumption of the essential oil is then evalu-
ated in the context of the data on absorption, metabolism, and
toxicology of members of the group. The intake of the group of
unidentified constituents is evaluated using a conservative thresh-
old of toxicological concern. This approach is based on the concept
that the types of chemicals in natural flavour complexes are not
infinite in structural variation, because chemical constituents in
botanicals originate from a limited number of biosynthetic path-
ways. If more than 95% of the essential oil constituents can be
adequately characterised and assigned to well defined congeneric
groups, the safety evaluation of this natural flavour complex could
be based on a safety evaluation of each congeneric group. Clearly
such an approach would rely more on adequate analytic char-
acterisation than on new toxicity testing of the natural flavour
complexes. Given the rapid improvement but reducing costs of
analytical tools, in comparison to the ever-increasing costs for
traditional toxicity testing, it would be of interest to further
develop such approaches, that rely on analytical chemistry and the
use of existing toxicological data bases for safety assessment of
botanicals.
135
6. Conclusions
The present paper describes some of the issues to be considered
and developed to a further extent in risk assessment of botani-
cals and botanical preparations intended for use in food and food
supplements, illustrated by examples based on the allylalkoxyben-
zenes. In order to further improve risk assessment of botanicals and
botanical preparations it is concluded that it would be important
to investigate, develop and validate beyond the present state-of-
the-art: (i) the use of the MOE concept for compounds that are
genotoxic and carcinogenic; (ii) new ways to quantify and incor-
porate the matrix effects into risk assessment strategies; (iii) the
use of analytical chemistry approaches, enabling almost complete
chemical characterisation of complex mixtures.
Acknowledgements
The authors would like to memorise and thank the late professor
Robert Kroes for his ever inspiring leadership and great friendship
over the years. We all miss him sincerely.
References
Council of Europe, 2005. Guidelines on the Safety of Plant-based Food Supplements
and Consumer Interest in those Supplements. Public health Committee, Com-
mittee of Experts on Nutrition, Food Safety and Consumer Health, Ad-hoc Group
on Food Supplements.
EFSA, 2004a. Discussion paper on Botanicals and botanical preparations widely
used as food supplements and related products: coherent and comprehensive
risk assessment and consumer information approaches. Scientific Committee of
the European Food Safety Authority, Brussel, pp. 1–6. http://www.efsa.eu.int/
science/sc commitee/sc documents/616/scdoc advice03 botanicals en1.pdf.
EFSA, 2004b. Opinion of the Scientific Panel on Food Additives, Flavour-
ings, Processing Aids and Materials in Contacts with Food (AFC) on a
Request from the Commission Related to Coumarin. Question num-
ber EFSA-Q-2003-EFSA Journal 104, 1–36. http://www.efsa.europa.eu/
EFSA/Scientific Opinion/afc opinion26 ej104 coumarin en1,3.pdf.
EFSA, 2005. Opinion of the scientific committee on a request from EFSA
related to a harmonised approach for risk assessment of substances
which are both genotoxic and carcinogenic. http://www. efsa.eu.int/science/
sc commitee/sc opinions/1201 en.html.
EFSA, 2007. Safety assessment of botanicals and botanical preparations intended for
use as food supplements. Draft guidance document of the scientific commit-
tee. Intermediate document adopted for public consultation on 19 November,
2007. http://www.efsa.europa.eu/EFSA/DocumentSet/sc guidance bota public
consultation,0.pdf.
EMEA, 2006. Guideline on the Assessment of Genotoxic Constituents in Herbal Sub-
stances/Herbal Preparations. EMEA Committee on Herbal Medicinal Products
(HMPC), European Medicines Agency, EMEA/HMPC/104681/207.
Jeurissen, S.M.F., Punt, A., Delatour, T., Rietjens, I.M.C.M., 2008. Basil extract inhibits
the sulfotransferase mediated formation of DNA adducts of the procarcinogen 1 -
hydroxyestragole by rat and human liver S9 homogenates and in HepG2 human
hepatoma cells. Food Chem. Toxicol. 46, 2296–2302.
Müller, L., Kasper, P., Müller-Tegethoof, K., Petr, T., 1994. The genotoxic potential in
vitro and in vivo of the allyl benzene etheric oils estragole, basil oil and trans-
anethole. Mutat. Res. 325, 129–136.
National Toxicology Program (NTP), 2000. NTP technical report on the toxicology
and carcinogenesis studies of methyleugenol (CAS NO. 93-15-2) in F344/N rats
and B6C3F1 mice (gavage studies). NIH Publication No. 00-3950.
Rietjens, I.M.C.M., Martena, M., Boersma, M.G., Spiegelenberg, W., Alink, G.M., 2005.
Molecular mechanisms of toxicity of some food borne phytotoxins. Mol. Nutr.
Food Res. 49, 131–158.
Sangster, S.A., Caldwell, J., Hutt, A.J., Anthony, A., Smith, R.L., 1987. The
metabolic disposition of [methoxy-14C]-labelled trans-anethole, estragole and
p-propylanisole in human volunteers. Xenobiotica 17, 1223–1232.
SCF, 2001a. Opinion of the scientific committee on food on the safety of
the presence of safrole (1-allyl-3,4-methylene dioxybenzene) in flavourings
and other food ingredients with flavouring properties. http://europa.eu.int/
comm/food/fs/sc/scf/out116 en.pdf.
SCF, 2001b. Opinion of the scientific committee on food on estragole (1-allyl-4-
methoxybenzene). http://europa.eu.int/comm/food/fs/sc/scf/out104 en.pdf.
SCF, 2001c. Opinion of the scientific committee on food on methyleugenol
(4-allyl-1,2-dimethoxybenzene). http://europa.eu.int/comm/food/fs/sc/scf/
out102 en.pdf.
Schilter, B., Andersson, C., Anton, R., Constable, A., Kleiner, J., O’Brien, J., Renwick,
A.G., Korver, O., Smit, F., Walker, R., 2003. Guidance for the safety assessment
of botanicals and botanical preparation for use in food and food supplements.
Food Chem. Toxicol. 41, 1625–1649.
136 I.M.C.M. Rietjens et al. / Toxicology Letters 180 (2008) 131–136

Smith, R.L., Adams, T.B., Doull, J., Feron, V.J., Goodman, J.I., Marnett, L.J., Portogh-
ese, P.S., Waddell, W.J., Wagner, B.M., Rogers, A.E., Caldwell, J., Sipes, I.G.,
2002. Safety assessment of allylalkoxybenzene derivatives used as flavouring
substances—methyl eugenol and estragole. Food Chem. Toxicol. 40, 851–870.
Smith, R.L., Cohen, S.M., Doull, J., Feron, V.J., Goodman, J.I., Marnett, L.J., Portoghese,
P.S., Waddell, W.J., Wagner, B.M., Hall, R.L., Higley, N.A., Lucas-Gavin, C., Adams,
T.B., 2005. A procedure for the safety evaluation of natural flavor complexes used
as ingredients in food: essential oils. Food Chem. Toxicol. 43, 345–363.
Wiseman, R.W., Miller, E.C., Miller, J.A., Liem, A., 1987. Structure-activity studies of
the hepatocarcinogenicities of alkenylbenzene derivatives related to estragole
and safrole on administration to preweanling male C57BL/6J × C3H/HeJ F1 mice.
Cancer Res. 47, 2275–2283.