Food and Chemical Toxicology 48 (2010) 29342944

Contents lists available at ScienceDirect

Food and Chemical Toxicology

journal homepage: www.elsevier.com/locate/foodchemtox

Cytogenetic evaluation and DNA interaction studies of the food colorants

amaranth, erythrosine and tartrazine
Panagiotis Mpountoukas a, Anastasia Pantazaki b,*, Efterpi Kostareli b, Pantelitsa Christodoulou b,
Dimitra Kareli a, Stamatia Poliliou a, Costas Mourelatos a, Vasso Lambropoulou a, Theodore Lialiaris a,**
a
b

Department of Genetics, Medical School, Demokrition University of Thrace, Alexandroupolis, Greece

Lab. of Biochemistry, Department of Chemistry, Aristotle University of Thessaloniki, Greece

a r t i c l e

i n f o

Article history:
Received 11 May 2010
Accepted 20 July 2010

Keywords:
Amaranth
Erythrosine
Tartrazine
Sister Chromatid Exchanges
Spectroscopic titration
Binding study

a b s t r a c t
Food coloring agents, amaranth, erythrosine and tartrazine have been tested at 0.028 mM in human
peripheral blood cells in vitro, in order to investigate their genotoxic, cytotoxic and cytostatic potential.
Amaranth at the highest concentration (8 mM) demonstrates high genotoxicity, cytostaticity and cytotoxicity. The frequency of SCEs/cell was increased 1.7 times over the control level. Additionally, erythrosine at 8, 4 and 2 mM shows a high cytotoxicity and cytostaticity. Finally, tartrazine seems to be toxic at 8
and 4 mM. No signs of genotoxicity were observed. Reversely, tartrazine showed cytotoxicity at 1 and
2 mM. Furthermore, spectroscopic titration studies for the interaction of these food additives with
DNA showed that these dyes bind to calf thymus DNA and distinct isosbestic points are observed clearly
suggesting binding of the dyes to DNA. Additionally DNA electrophoretic mobility experiments showed
that these colorants are obviously capable for strong binding to linear dsDNA causing its degradation.
PCR amplication of all DNA fragments (which previously were pre-treated with three different concentrations of the colorants, extracted from agarose gel after separation and then puried), seems to be
attenuated with a manner dye concentration-dependent reecting in a delayed electrophoretic mobility
due to the possible binding of some molecules of the dyes. Evaluation of the data and curves were
obtained after quantitative and qualitative analysis of the lanes of the gel by an analyzer computer program. Our results indicate that these food colorants had a toxic potential to human lymphocytes in vitro
and it seems that they bind directly to DNA.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
Color additives have long been a part of human culture. Archaeologists date cosmetics colors as far as 5000 B.C. Ancient Egyptian
writings mention the use of drug colorants and historians estimate
that food colors likely emerged around 1500 B.C. Color is an essential criterion for food choice. The food industry uses synthetic colorants in a way to improve the esthetic quality of a food product.
The total world colorant production is estimated to be 80,00,000
tons per year (Revankar and Lele, 2007).

Abbreviations: SCEs, Sister Chromatid Exchanges; PRI, Proliferation Rate Index;

MI, Mitotic Index.
* Corresponding author. Address: Lab. of Biochemistry, Department of Chemistry,
Aristotle University of Thessaloniki, Thessaloniki 54124, Greece. Tel.: +30 2310
997838; fax: +30 2310 997689.
** Corresponding author. Address: Department of Genetics, Faculty of Medicine,
Demokrition University of Thrace, Alexandroupolis 68100, Greece. Tel./fax: +30
25510 30522.
E-mail addresses: natasa@chem.auth.gr (A. Pantazaki), lialiari@med.duth.gr
(T. Lialiaris).
0278-6915/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fct.2010.07.030

Synthetic colorants are divided into ve classes: the azo compounds (such as amaranth and tartrazine), the chinophthalon
derivatives of Quinoline Yellow, the triarylmethane group, xanthenes (such as erythrosine) and the indigo colorants (Minioti
et al., 2007).
Lipid oxidation during food processing is one of the most significant agents of food deterioration. It was demonstrated that food
colorants with a xanthene skeleton have higher potential as photo
sensitizers (Pan et al., 2005). So, erythrosine shows an accelerated
oxidation under light exposure, a phenomenon that was not observed in the two azo dyes that tested in this study.
The red dye amaranth (FD & C Red No. 2, E123) (Fig. 1) is used in
foods with a reddish or brownish color, including soft drinks, icecreams, cake mixes, wines, tinned fruit pie llings, soups, prawns,
cereals, salad dressings, chewing gums, jams, chocolates and coffee
as well as a variety of drugs and cosmetics. Erythrosine (FD & C Red
No. 3, E127) (Fig. 2) is widely used as a coloring agent in food,
drugs and cosmetics, such as sweets, candies, cake decorating gel
and dental plaque disclosing agent. Finally, the food drug and cosmetics coloring tartrazine (FD & C Yellow No. 5, T102) (Fig. 3) is a

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

Fig. 1. Chemical structure of amaranth.

Fig. 2. Chemical structure of erythrosine.

2935

cal agents, mutagens, antimutagenic agents, antioxidants (Lialiaris

et al., 1987, 1990, 1992, 2008, 2009a; Karapidaki et al., 2009, 2010)
and food preservatives (Mpountoukas et al., 2008). The induction
of SCEs is a rapid and sensitive end-point for clastogenicity
tests and chromosome instability (Lialiaris et al., 2009b,c, 2010;
Papachristou et al., 2006, 2008). This cytogenetic test may be useful to indicate possible genotoxic results of low doses of different
chemicals.
Binding or interaction of a compound with DNA causes adsorption spectral changes that can be used to detect the possible manner of binding of the compound leading to its DNA intercalation or
degradation. Some compounds may interact with DNA either covalently or non-covalently. In a covalent binding the reactive part of
the compounds interacts with a nitrogen base of DNA such as
guanine N7 e.g. cisplatin. On the other hand, the non-covalent
DNA interactions include intercalative, electrostatic and groove
binding of cationic metal complexes along outside of DNA helix,
the major or minor groove. Intercalation involves the partial insertion of aromatic heterocyclic rings between the DNA base pairs
(Polyanichko et al., 2004) and it is a general observation that the
binding of an intercalative molecule to DNA is accompanied by a
large hypochromism (decrease of absorbance) and signicant
bathochromism or red-shift (migration of the characteristic peak
to longer wavelengths as a consequence of binding with other substance) in the absorption spectra due to strong stacking interaction
between the aromatic chromophore of the ligand and DNA base
pairs. The extent of spectral changes is related to the strength of
binding and the spectra for intercalators are more perturbated than
those for groove binders (Kelly et al., 1985; Chow and Barton,
1992).
The binding or cleavage reaction of DNA with compounds such
as amaranth, erythrosine and tartrazine might monitored using
agarose gel electrophoresis. The DNA binding efciency of the
complexes was estimated by determining the mobility reected
in an up-shift migration of a DNA fragment to higher molecular
weight DNA products, or in a down-shift migration to lower ones
suggesting the DNA degradation.
In this study, effects of amaranth, erythrosine and tartrazine on
the frequency of Sister Chromatid Exchanges (SCEs), the Proliferating Rate Index (PRI) and the Mitotic Index (MI) were investigated.
These indices have been described as criteria of possible genotoxicity, cytostaticity and cytotoxicity of some chemical factors
(Pantazaki et al., 1999; Stanimirovic et al., 2005; Bakopoulou
et al., 2008). Furthermore, we examined the binding efciency of
these dyes to calf thymus DNA (CT-DNA), as a prerequisite to affect
the DNA integrity both by agarose gel electrophoresis and by spectroscopic titration studies, since binding or interaction of a compound with DNA might lead to its DNA intercalation or degradation.
2. Materials and methods
2.1. Materials and cell culture

Fig. 3. Chemical structure of tartrazine.

synthetic bright orangeyellow powder that is used to color soft

drinks, chips, cereals, mustards, ice-creams, hand lotions, drug capsules and many other products.
The acceptable daily intake (ADI) levels recommended by the
Joint FAO/WHO Expert Committee on Food Additives (JECFA)
for amaranth, erythrosine and tartrazine is 01.5, 00.05 and
07.5 mg/kg bw, respectively.
Sister Chromatid Exchanges (SCEs) analysis is a well established
method aiming at evaluating human exposure to different chemi-

Heparinized blood samples were obtained and cultured from nine healthy individuals (aged 2530). None of them was a smoker or was receiving drugs for medical or other reasons. Eleven drops of whole blood was added to 5 ml of
chromosome medium B (Biochrom 0303H) with phytohemagglutinin in universal
containers. In order to visualize the results of the SCE test, 5 lg/ml 5-bromodeoxyuridine (BrdU, CAS No. 59-14-3, EC No. 200-415-9, P99.0% pure) were added at the
beginning of the culture period. The solutions of amaranth (CAS No. 915-67-3, EEC
No. 213-022-2, approx. 90% pure), erythrosine (CAS No. 16423-68-0, EEC No. 240474-8, approx. 90% pure) or tartrazine (CAS No. 1934-21-0, EEC No. 217-699-5, approx. 90% pure) were added at various concentrations. All chemicals were dissolved
in distilled water and the same amount of distilled water was introduced in the
control cultures. The cultures were incubated at 37 C for 72 h throughout all cultures were maintained in the dark to minimize photolysis of BrdU. At 70 h colchicine (CAS No. 64-86-8, EC No. 200-598-5, P97.0% pure) at 0.3 lg/ml was injected
to each culture and, at the end of the incubation period, cultures harvested with
hypotonic KCl solution and xed in methanol:acetic acid (3:1, v/v).

2936

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

2.2. In vitro SCE assay

The chromosomes were stained using a modied Fluorescence Plus Giemsa
(FPG) technique (Goto et al., 1978; Ioannidou et al., 1989; Lialiaris et al., 1988,
1989, 1992). For each culture, 40 well spread second division metaphases were analyzed. Preparations were scored for cells in their rst mitosis (both chromatids dark
staining), second mitosis (one chromatid of each chromosome dark staining) and
third and subsequent divisions (a portion of chromosomes with both chromatids
light staining). Slides had been previously coded and scoring was done blindly.
2.3. Assessment of Proliferation Rate Index and Mitotic Index
Cell cycle kinetics were estimated from the Proliferation Rate Index (PRI) scored
on at least 300 metaphases. PRI calculated according to the formula
PRI = (1M1 + 2M2 + 3M3+)/N, where M1 is the percentage of cells in the rst division,
M2 in the second and M3+ in the third and higher divisions and N is the total number
of metaphases scored for each culture. Mitotic Indices (MIs) were calculated by
counting the number of metaphases for 5000 activated lymphocytes for all cultures
(Goto et al., 1978; Lialiaris et al., 1988, 1992, 2007).
2.4. DNA mobility shift experiments in agarose gel electrophoresis
The binding or cleavage reaction of DNA with compounds was monitored using
agarose gel electrophoresis. The DNA interaction efciency of a compound reects
to the electrophoretic mobility and it is dependent from the form of the DNA substrate used. Retardation of the electrophoretic mobility of a linear DNA could be
attributed to the binding of certain molecules of the compound able to increase
its molecular weight, while precession of the electrophoretic mobility could be
attributed to the degradation (or damage) of the initial DNA substrate mirrored
to decrease of its molecular weight. DNA molecules with lower molecular weight
from this of the initial DNA molecule migrate faster, while DNA molecules with
higher molecular weight derived from DNA breakage effect delay compared with
the mobility of the initial DNA (Afrati et al., 2008, 2010).
Reactions, which contained aliquots of 3 lg of nucleic acid (linear DNA), was
incubated at a constant temperature of 37 C for 30 min (or as indicated in the legends) in the presence of various concentrations of these dyes in a buffer A to a nal
volume of 20 ll. It was terminated by the addition of 5 ll loading buffer consisting
of 0.25% bromophenol blue, 0.25% xylene cyanol FF and 30% glycerol in water. The
products resulting from interactions of these compounds with DNA were separated
by electrophoresis on agarose gels (1% w/v), which contained 1 lg/ml ethidium bromide in 40  10 3 M Trisacetate, pH 7.5, 2  10 2 M sodium acetate, 2  10 3 M
Na2EDTA, at 5 V/cm. Agarose gel electrophoresis was performed in a horizontal
gel apparatus (Mini-Sub DNA Cell, BioRad) for about 2 h. The gels were observed
after staining with the uorescence intercalated dye ethidium bromide under a UV
illuminator (Afrati et al., 2008, 2010).
2.5. DNA binding studies by spectroscopic titration
The binding effect of the dyes with DNA was studied following the absorption
titration technique in the visible and UV regions and accomplished by gradually
adding a certain volume of DNA solution (step by step) into a solution of dye. Then,
the inuence of the DNA addition into the dye solution was obtained by recording
the variation of spectra compared to the initial spectrum of the dye (control) after
successive additions of DNA (Afrati et al., 2008, 2010).
Adsorption UVvis spectra for DNA binding studies by spectroscopic titration
were recorded with a Shimadzu-160A dual beam spectrophotometer. Buffer A
(50 mM Tris ((hydroxymethyl) aminomethane)HCl buffer (pH 7.5)) was used for
absorption spectral experiments. The DNA stock solution (1 mg/ml) was prepared
at 04 C by dissolving the commercially purchased calf thymus DNA in buffer A.
Native DNA (dsDNA) (CT-DNA) type I, highly polymerized from calf thymus gland
was purchased from Sigma (D-1501). DNA concentrations were determined from
UV adsorption at 260 nm using a molar extinction coefcient, 6600 M 1 cm 1. A
stock solution of each dye was prepared at a nal concentration of 10 5 M by dissolving the dyes in water.
2.6. DNA samples preparation
Human genomic DNA derived from total blood sample and extraction was performed by QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany) according to the manufacturers protocol. PCR amplication was performed using sequence-specic
primers to obtain a short double stranded DNA. PCR protocol was followed by using
Taq platinum (Invitrogen), associated buffer 10 (Invitrogen), dNTPs (Invitrogen),
VgCl2 (Invitrogen). The amplied gene segment was a linear DNA fragment of
300 bp. The PCR product was cleaned up with PureLink PCR Purication Kit
(Invitrogen).
Puried linear DNA fragment was treated with three concentrations of each of
the colorants under the same conditions and the products were separated by agarose gel electrophoresis. Each pre-treated DNA band was isolated from the gel
and subsequently DNA gel extraction was performed by QIAquick Gel Extraction

Kit (QIAGEN, Hilden, Germany). All puried DNA fragments (corresponded to these
which previously were pre-treated with three different concentrations of the colorants), were again amplied by PCR using gene-specic primers for the utilized linear DNA fragment. Agarose was purchased from BRL. Intercalative dye ethidium
bromide (EthBr) was purchased from Sigma.
2.7. PCR amplication
The 300 bp DNA fragment was prepared and cleaned as described above. The
PCR mixture in a total volume of 102.5 ll containing 5 ll of DNA fragment, 10 ll
buffer 10, 3 ll of 50 mM MgCl2 (2 mM nal concentration), 2 ll of 10 mM dNTPs
(200 lV nal concentration), 3 ll of primer-reverse, 3 ll of primer-forward, 2.5 ll
(12.5 units) of Taq platinum polymerase (5 units/ll) and 74 ll of DEPC water, was
conducted for PCR amplication according the protocol: (1) denaturation at 94 C
for 5 min, (2) 94 C for 1 min, (3) annealing at 59 C for 1 min, (4) extension at
72 C for 1.5 min, (5) repeat of steps 24 for 35 cycles and (6) nal extension step
at 72 C, 10 min.
2.8. Statistical analysis
Evaluation of MI and PRI was based on the v2-test. For the SCE frequencies the
one-way analysis of variance (ANOVA) and the Duncan test were used to compare
various treatments (Lialiaris et al., 1992, 2007). Correlations between SCEs, MIs and
PRIs were calculated (Lialiaris et al., 1988; Maskaleris et al., 1998). Evaluation of
data was performed after quantitative and qualitative analysis of the lanes of the
gel by the Gelpro Analyzer V.3 computer program and the corresponding curves
were obtained.

3. Results
3.1. In vitro SCE assay
Addition of amaranth into the cultures at all concentrations
tested induced statistically signicant (p < 0.01) increase in SCEs/
cell in relation to control, while statistically signicant (p < 0.01)
cell division delays were observed at 8 mM of amaranth. Additionally, there is statistically signicant (p < 0.01) difference between
the concentration of 8 mM and all the others concerning on the
genotoxicity of the dye. Finally, the presence of amaranth at 4
and 8 mM (cultures 7 and 8, respectively) gave statistically significant (p < 0.01) decrease in MI compared to control. A statistically
signicant (p < 0.01) difference was presented among the cultures
2, 3, 4, 5, 6 and 7 (0.02, 0.2, 0.5, 1, 2 and 4 mM, respectively) and
the highest concentration concerning to PRI (Table 1). There was
a strong positive correlation between PRI and MI (r = 0.970;
p < 0.01) in Table 1.
3.2. Assessment of Proliferation Rate Index and Mitotic Index
Presence of erythrosine into the cultures at concentrations 8, 4
and 2 mM, respectively showed high cytotoxic and cytostatic action, so no metaphases were presented in cultures. As a result,
the level of SCEs/cell, PRI and MI could not be computed at these
concentrations. At the other concentrations, no statistically significant results were observed for SCEs/cell. A statistically signicant
(p < 0.01) decrease of MI at 1 mM of erythrosine was observed, in
comparison to the control culture. Furthermore, the other concentrations showed statistically signicant (p < 0.01) increase of MI
compared to the culture 5 (erythrosine at 1 mM). All the concentrations caused statistically signicant cell division delays in comparison to the control (Table 2). Therefore, there was a positive
correlation between PRI and MI (r = 0.956; p < 0.05) in Table 2.
Presence of tartrazine into the cultures at the highest concentrations (8 and 4 mM) affected the quality of chromosomes in a
way that differentiation between sister chromatids did not occur.
Consequently, the levels of SCEs/cell and PRI did not be computed
at these concentrations. Tartrazine at 4 and 8 mM (cultures 7 and
8, respectively) showed statistically signicant (p < 0.01) decrease
of MI compared to control. Furthermore, statistically signicant
(p < 0.05) differences of MI were observed at 1, 0.5 and 0.2 mM

2937

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

Table 1
Cytogenetic effects of amaranth (AM) in cultured human lymphocytes.
Treatment

MI ()

Mean SCEs/cell SEM (range)

(1) Control
(2) FM 0.02 mM
(3) FM 0.2 mM
4) FM 0.5 mM
(5) FM 1 mM
(6) FM 2 mM
(7) FM 4 mM
(8) FM 8 mM

47.0
46.4b,c
42.6b,c
42.2b,c
40.6b,c
39.4b,d
30.0a,b
7.0a

6.09 0.46 (013)

8.63 0.46a (317)
8.71 0.52a (117)
8.74 0.60a (216)
8.85 0.59a (115)
9.11 0.64a (220)
9.96 0.49a (419)
10.12 0.65a (319)

Percent of cells in 1st, 2nd and subsequent (3rd+) divisions

1st

2nd

3rd+

18.0
13.0
14.0
15.0
14.0
17.0
24.0
32.0

28.0
30.0
29.0
30.0
32.0
28.0
26.0
43.0

54.0
57.0
57.0
55.0
54.0
55.0
50.0
25.0

PRI

2.36
2.44b,c
2.43b,c
2.40b,c
2.40b,c
2.38b
2.26b
1.93a

The SCE frequency was based on 40 second division cells.

Results were based on three experiments with the same protocol.
A total of 300 metaphases were scored for the calculation of PRIs and 5000 activated lymphocytes were scored for MIs.
The suppression of PRIs was positively correlated with the reduction of MIs (r = 0.970; p < 0.01).
a
p < 0.01 vs. line 1.
b
p < 0.01 vs. line 8.
c
p < 0.01 vs. line 7.
d
p < 0.05 vs. line 7.

Table 2
Cytogenetic effects of erythrosine (ER) in cultured human lymphocytes.
Treatment

MI ()

Mean SCEs/cell SEM (range)

(1)
(2)
(3)
(4)
(5)

39.0
38.0b
40.0b
36.0b
8.0a

6.15 0.50
5.92 0.37
6.39 0.39
6.52 0.32
6.80 0.41

Control
TR 0.02 mM
TR 0.2 mM
TR 0.5 mM
TR 1 mM

(112)
(29)
(312)
(211)
(213)

Percent of cells in 1st, 2nd and subsequent (3rd+) divisions

1st

2nd

3rd+

12.0
9.0
8.0
14.0
28.0

18.0
29.0
34.0
32.0
32.0

70.0
62.0
58.0
54.0
40.0

PRI

2.58
2.53a,b
2.50a,b
2.40a,b
2.12a

The SCE frequency was based on 40 second division cells.

Results were based on three experiments with the same protocol.
A total of 300 metaphases were scored for the calculation of PRIs and 5000 activated lymphocytes were scored for MIs.
ER at the concentrations of 2, 4 and 8 mM had toxic effects (no dividing cells).
A correlation was observed between the magnitude of PRIs suppression and the MIs reduction (r = 0.956; p < 0.05).
a
p < 0.01 vs. line 1.
b
p < 0.01 vs. line 5.

Table 3
Cytogenetic effects of tartrazine (TAR) in cultured human lymphocytes.
Treatment

MI ()

Mean SCEs/cell SEM (range)

(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)

38.8
35.0c
33.8d
33.0d
32.4d
31.4
29.0a
25.6a

5.90 0.46
6.17 0.30
6.37 0.44
6.68 0.42
6.80 0.42
6.97 0.44

Control
NAR 0.02 mM
NAR 0.2 mM
NAR 0.5 mM
NAR 1 mM
NAR 2 mM
NAR 4 mM
NAR 8 mM

(112)
(29)
(113)
(111)
(210)
(212)

Percent of cells in 1st, 2nd and subsequent (3rd+) divisions

1st

2nd

3rd+

16.0
23.0
19.0
20.0
22.0
22.0

29.0
21.0
31.0
30.0
30.0
32.0

55.0
56.0
50.0
50.0
48.0
46.0

PRI

2.39
2.33b,e,f
2.31
2.30
2.26a
2.24a

The SCE frequency was based on 40 second division cells.

Results were based on three experiments with the same protocol.
A total of 300 metaphases were scored for the calculation of PRIs and 5000 activated lymphocytes were scored for MIs.
A correlation was observed between the magnitude of SCE response and the PRI (r = 0.969; p < 0.01) or MI (r = 0.949; p < 0.01) depression. Correlation was also observed
between the magnitude of PRIs suppression and the MIs depression (r = 0.988; p < 0.01).
a
p < 0.01 vs. line 1.
b
p < 0.05 vs. line 1.
c
p < 0.01 vs. line 8.
d
p < 0.05 vs. line 8.
e
p < 0.01 vs. line 6.
f
p < 0.01 vs. line 6.

of tartrazine in comparison to the highest concentration of tartrazine (Table 3). Tartrazine at 1 and 2 mM (cultures 5 and 6, respectively) showed statistically signicant increase (p < 0.01) of cell
division delays compared to the control (Table 3). Finally, there

was a statistically signicant negative correlation (r = 0.969;

p < 0.01) between SCEs and PRIs (the higher the SCE frequency
the lower the PRI value), and statistically signicant negative correlation (r = 0.949; p < 0.01) between SCEs and MIs (the higher

2938

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

the SCE frequency the lower the MI value). The positive correlation
between PRI and MI (r = 0.988; p < 0.01) conrms the other two
negative correlations.

3.3. Binding studies of dyes by spectroscopic titration

3.3.1. Binding of erythrosine to CT-DNA
The erythrosine binding to the DNA base pairs was obtained
by monitoring the changes in the UVvis region and especially
the absorbance at 260 and 527 nm with increasing concentrations of DNA (Fig. 4). Fig. 4 illustrates spectral changes of erythrosine in aqueous solution observed during the titration upon
addition of increasing amounts of CT-DNA. The initial spectrum
of a DNA solution exhibited a maximum at 260 nm (upper line
at 260 nm). The spectrum referred to a xed concentration
(10 5 M) of the free erythrosine in the absence of CT-DNA (control), exhibited a maximum at 527 nm (upper line at 527 nm).
All the following spectra were recorded after successive addition
of CT-DNA in proportion ranging from 1.4 to 9.3 lM of CT-DNA
to the assay mixture containing the xed concentration of the
free erythrosine (the ratio R = [erythrosine]/[DNA] ranges from
0.01 to 0.07) (Fig. 4, from the upper line to the descending direction at 527 nm).
Even though no appreciable change was observed in the position of the erythrosine absorption centred at 527 nm by addition
of CT-DNA, the intensity of this band has been found to decrease
substantially in the presence of DNA resulting in a hypochromism. The binding strength of the dye with CT-DNA is mirrored
in the percentage of hypochromism. Addition of DNA resulted in
hypochromism at 527 nm and a hyperchromism at 260 nm and
blue shift of 23 nm. The hyperchromism at 260 nm might be
evidently attributed to the gradual DNA addition while the percentage of hypochromism in the band at 527 nm for the erythrosine in the presence of CT-DNA at saturation reaches 23%.
Furthermore, a distinct isosbestic point in the spectrum at
297 nm accompanies the hypochromism, as DNA concentration
is increased, clearly suggesting binding of CT-DNA to the erythrosine dye and formation of a new product among the dye and
DNA. An isosbestic point is usually observed in overlaid spectra
when a chromophoric precursor is converted to a product with
a different spectrum, so that it is often assumed that an isosbestic point occurs only when the precursor is quantitatively
converted to a single product.

3.3.2. Binding of amaranth to CT-DNA

In a similar experiment the amaranth binding to the DNA base
pairs was obtained by monitoring the changes in the UVvis and
especially in the absorbance at 260 and 523 nm with increasing concentrations of DNA. Fig. 5 illustrates the initial spectrum which refers
to the spectrum of a xed concentration (10 5 M) of the free amaranth in the absence of CT-DNA (control), which exhibited maximal
absorbencies at 523, 333 and 242 nm (upper line). All the following
spectra were recorded after successive addition of CT-DNA in proportion ranging from 2.7 to 6 lM CT-DNA to the assay mixture containing the xed concentration of the free amaranth (the ratio
R = [amaranth]/[DNA] ranges from 0.016 to 0.036) (Fig. 5, from the
upper line to the descending direction at 523 nm).
Even though no appreciable change was observed in the position of the amaranth absorption centred at 523 and 333 nm as well
by addition of CT-DNA, the intensity of the this band has been
found to decrease substantially in the presence of DNA resulting
in a hypochromism while a 6 nm red-shift was observed and a
strong hyperchromism was appeared in the initial absorption spectra of amaranth in the position of 242 nm. Furthermore, a distinct
isosbestic point at 297 nm approximately is observed, (as similarly
obtained with CT-DNA addition), clearly suggesting binding of the
amaranth to the DNA.
3.3.3. Binding of tartrazine to CT-DNA
To elucidate a possible binding of tartrazine to the DNA base
pairs a similar experiment was performed by monitoring the spectroscopic changes in the UVvis (and especially followed in the
absorbancies at 260 and 430 nm), either by adding increasing
amounts of DNA to a xed concentration of tartrazine (10 5 M)
(Fig. 6A, upper line at 430 nm) or by adding increasing amounts
of tartrazine to a xed concentration of DNA (3  10 3 M) in the
absence of tartrazine (Fig. 6B, upper line at 260 nm). Fig. 6A mirrored the spectroscopic changes of the tartrazine spectrum upon
successive addition of increasing CT-DNA amounts resulting in a
hypochromism and a clear isosbestic point at 307 nm conrming
again the binding of the tartrazine with DNA. All the spectra recorded and followed (from the upper line to the descending direction at 430 nm) upon increasing amounts of DNA added into the
solution) at concentrations 1.4, 3.9, 6 and 7.7 lM of dsDNA,
respectively.
Fig. 6B mirrored the spectroscopic changes of the CT-DNA spectrum upon successive addition of increasing amounts of tartrazine
ranging from 50 to 3 lV, respectively. All the spectra recorded and

Fig. 4. Spectroscopic changes of erythrosine spectrum in water, monitored in the UVvis region upon increasing addition of DNA. The initial spectrum of a xed concentration
of DNA was 10 5 M (upper line at 260 nm). The initial spectrum of a xed concentration of erythrosine was 10 5 M (upper line at 527 nm) and all the spectra recorded and
followed (from the upper line to the descending direction at 527 nm) upon increasing amounts of DNA added into the solution) at concentration 1.4, 3.9, 6, 7.77, 9.3 lM of
dsDNA, respectively. Inset shows the typical isosbestic point formatted at 297 nm.

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

2939

Fig. 5. Spectroscopic changes of amaranth spectrum in water, monitored in the UVvis region upon increasing addition of DNA. The initial spectrum of a xed concentration
of amaranth was 10 5 M (upper line at 523 nm) and all the spectra recorded and followed (from the upper line to the descending direction at 523 nm) upon increasing
amounts of DNA added into the solution) at concentration 2.7, 5 and 6 lM of dsDNA, respectively. Inset shows the typical isosbestic point formatted at 297 nm.

Fig. 6. (A) Spectroscopic changes of tartrazine spectrum in water monitored in the UVvis region upon increasing addition of DNA. The initial spectrum of a xed
concentration of tartrazine was 10 5 M (upper line at 423 nm) and all the spectra recorded and followed (from the upper line to the descending direction at 423 nm) upon
increasing amounts of DNA added into the solution at concentration 1.4, 3.9, 6 and 7.7 lM of dsDNA, respectively. Inset shows the typical isosbestic point. (B) Spectroscopic
changes of CT-DNA spectrum in water monitored in the UVvis region (upper line at 260 nm) and all the spectra recorded and followed (from the upper line to the descending
direction at 260 nm) upon increasing addition of tartrazine. The initial concentration of CT-DNA was 3  10 3 M and the amounts of tartrazine added ranged from 50 to 3 lV,
respectively. Inset shows the typical isosbestic point at 307 nm.

followed after addition of tartrazine in proportion ranging from 50

to 3 lM of tartrazine to the assay mixture containing the xed concentration of the DNA (the ratio R = [tartrazine]/[DNA] ranges from
16.6  10 6 to 10 3 or the ratio R = [DNA]/[tartrazine] ranges from

60,000 to 1000) (Fig. 6B, from the upper line to the descending
direction at 260 nm).
Even though no appreciable change (barely 1 nm) by addition of
CT-DNA was observed in the position of the tartrazine absorption

2940

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

Fig. 7. Agarose (1% w/v) gel electrophoretic pattern of EthBr stained linear DNA (PCR product) after 2 h duration of electrophoresis. Each sample containing 5 lg of linear DNA
incubated with the dyes erythrosine (E), tartrazine (T) and amaranth (A) at 37 C for 30 min (upper) and for 1 h (down), respectively. Lane c: control, linear DNA without
treatment. Lanes 13: linear DNA treated with 13 mM of each dye, respectively. Lane M: DNA ladder 100 bp plus. (Inset) Agarose (1% w/v) gel electrophoretic pattern of
EthBr stained linear DNA (PCR product) after 2 h duration of electrophoresis. Each sample containing 5 lg of linear DNA incubated with the dye tartrazine at 37 C for 30 min.
Lane c: control, linear DNA without treatment. Lanes 1 and 2: linear DNA treated with 1 and 2 mM of tartrazine, respectively. (BPB: bromophenol blue).

band of vis region centred at 427 nm, the intensity of this band has
been found to decrease substantially in the presence of DNA resulting in a hypochromism (Fig. 6). In addition a strong hyperchromism in the initial absorption spectra of tartrazine at the
position of the peaks corresponded to tartrazine centred at 257
and 212 nm occurred, which accompanied with a simultaneous
small red-shift of 1 and 3 nm, respectively. Furthermore, a distinct
isosbestic point at 307 nm approximately is observed, as similarly
obtained with CT-DNA addition, clearly suggesting binding of the
tartrazine to the DNA.
3.4. DNA mobility shift experiments in agarose gel electrophoresis
The linear dsDNA obtained as a PCR product (as described in
Section 2.6) was incubated with 1.0, 2.0 and 3.0 mM of each of
the dyes, erythrosine, tartrazine and amaranth. Fig. 7 (upper)
shows a pronounced up-shift of the initial DNA band, which is
caused during it incubation with dyes erythrosine (lanes E1E3)
or tartrazine (lanes T1T3), respectively. This up-shift of the initial
DNA band is concentration dye-dependent resulting in its migration to higher molecular weight (delayed electrophoretic mobility).
This result supports the binding of some molecules of each of dye
to DNA, which is able and sufcient to cause this retardation compared to the untreated DNA molecule (control). The Rf value varies
from 0.95 to 0.92 for the higher concentration as it was estimated
with Gelpro Analyzer V.3 computer program. In contrast amaranth
does not caused a similar strong effect under the same conditions
(Fig. 7, upper, lanes A1A3).
When the incubation time was longer (Fig. 7 down), interesting
phenomena occurred. At one side, a pronounced up-shift of the initial DNA band was observed in some cases, (Fig. 7 down, lanes T1
T3 and A2A3), which reected to migrate as a band corresponding
to higher molecular weight. This result supports the binding of
some molecules of the dyes to DNA enabling to cause this retardation compared to the untreated DNA molecule (control). At the
other side, a diminution of the initial DNA band reected to migrate as bands corresponding to smaller molecular weights and
supporting a degradative effect due to the dyes effect. This effect
is obvious especially in the cases of tartrazine and amaranth
(Fig. 7 down, lanes T2, T3 and A2 and A3, respectively), where

new bands appeared with lower molecular weights. In the case

of erythrosine the mobility of the DNA band is faster than that of
the initial DNA band in all cases and its mobility is concentration
dye-dependent (Fig. 7, upper, lanes E1E3). These results direct
to the suggestion that when the duration of incubation time with
the dyes was shorter (30 min) assure just as an initial effect the
dyes binding to DNA, while when it was longer the DNA degradative effect followed as a second effect resulting from the extensive
incubation and contact with the dyes.
Furthermore, it is notable that tartrazine binds to the DNA fragment so strongly such as the dyes characteristic color is located
overlapping the uorescent DNA band (obviously visible at the initial stage of electrophoresis), despite that it was expected that dye
might migrated with the dye bromophenol blue, according its
molecular weight (Fig. 7, inset). The Rf value changes from 0.62 (control) and 1 mM erythrosine (lanes coincident) to 0.60 and 0.56 for 2
and 3 mM of erythrosine, respectively. The corresponding Rf values
for tartrazine are 0.6 and 0.58 for 2 and 3 mM, respectively.
3.5. Effect of dyes binding to the PCR amplication
To monitor and conrm whether a possible binding of the dyes
to DNA inhibited the PCR amplication, DNA fragments pre-treated
with various concentrations of the colorants were separated on the
gel, extracted and then were subjected to PCR amplication and assessed by agarose electrophoresis. Fig. 8 illustrates the agarose gel
electrophoretic pattern of the linear DNA fragment pre-treated
with three different concentrations (1, 2 and 3 mM) of the colorant
erythrosine (lanes E1E3) and tartrazine (lanes T1T3), respectively. The PCR amplication products obtained when the DNA
fragment of 300 bp was pre-treated with the colorants exhibited
a delay of their electrophoretic mobility. In addition the PCR amplication efciency seems to be attenuated with a manner concentration-dependent (increase of the colorant concentration used
for the pretreatment increase the attenuation of the PCR amplication). This result was expected and conrms the suggestion that
some molecules of colorants might bind at a certain extent to the
DNA fragment during pretreatment with them. The same conclusion was deduced also from the evaluation of the data after quantitative and qualitative analysis of the lanes of the gel by the Gelpro

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

2941

Fig. 8. Agarose (2%) gel electrophoretic pattern of linear 300 bp DNA fragment treated with three different concentrations of erythrosine and tartrazine separated in agarose
gel electrophoresis. Subsequently, the DNA product of treatment was subjected to DNA gel extraction, then amplied by a new PCR and separated in agarose gel
electrophoresis again. The staining was performed with EtBr. Each sample containing 5 lg of linear DNA incubated with 13 mM of erythrosine and tartrazine, respectively at
37 C for 30 min. Lane C (+): DNA fragment without treatment and amplied. Lanes 13: linear DNA treated with 13 mM of erythrosine and tartrazine, respectively and then
amplied. Lane C ( ): DNA fragment without treatment and not amplied. Lane M: DNA ladder 100 bp plus. (Evaluation of data was performed after quantitative and
qualitative analysis of the lanes of the gel by the Gelpro Analyzer V.3 computer program and curves were obtained (B) for erythrosine and (C) for tartrazine.)

Analyzer V.3 computer program from and curves obtained (Fig. 8A)
for erythrosine and (Fig. 8B) for tartrazine and the obvious changes
of the Rf of the peaks.

aqueous solution of each dye with native random-sequence DNA

(from calf thymus CT-DNA), was performed since recently, it was
applied for other compounds (Chaviara et al., 2008; Dimitrakopoulou et al., 2008).

4. Discussion
4.1. Cytogenetic studies of amaranth
Numerous and varied food additives, such as colors, preservatives, sweeteners and antioxidants are consumed in a typical daily
diet. One approach of their safety is to assay the clastogenicity of
these additives using cultured human peripheral lymphocytes
(Lialiaris et al., 1987, 2007; Mpountoukas et al., 2008). A change
in the frequency of SCEs is an essential marker of genotoxicity in
in vitro (Rogers et al., 1988; Lialiaris et al., 1990) or in in vivo studies (Mourelatos et al., 1988; Zijno et al., 1994; Digkas et al., 2010).
SCEs methodology is more sensitive than other cytogenetic methods, like chromosome aberrations (CAs) and micronuclei test. The
failure of repair mechanisms to achieve recovery of damages
caused by different agents, leads to double strand breaks to DNA,
so an increase in the frequency of SCEs/cell was observed (Sonoda
et al., 1999; Johnson and Jasin, 2000).
In an attempt to provide evidence for the possibility of binding
of the dyes to double stranded DNA, spectroscopic titration of a

The genotoxicity of amaranth is a matter of debate because of

the controversial current results. Its genotoxic effects have been
tested in Ames test (Auletta et al., 1977; Brown et al., 1978;
Al-Mossawi, 1983; Ishidate et al., 1984; Izbirak et al., 1990), in
the mice dominant lethal assay (Arnold et al., 1976), in cell transformation assay of Syrian hamster embryo cells (Heidelberg et al.,
1983), in Saccharomyces cerevisiae (Parry, 1977; Sankaranarayanan
and Murthy, 1979) in the somatic and germ line cells of Drosophila
melanogaster (Tripathy et al., 1995) and in SpragueDawley rats
(Munzner, 1979) with negative results.
Reversely, positive genotoxic effects of amaranth were presented in a chromosomal aberration test in vitro using a Chinese
hamster broblast cell line (Ishidate et al., 1984) and in application
of the standard plate assay to ether extracts of aqueous solution of
amaranth (Prival et al., 1988).

2942

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

Among 39 currently used food additives the dyes amaranth, tartrazine and erythrosine induced dose-related DNA damage in the
glandular stomach, colon, and/or urinary bladder and in the gastrointestinal organs at a low dose (10 or 100 mg/kg). In addition amaranth, and tartrazine induced DNA damage in the colon at close to
the acceptable daily intakes (ADIs) (Sasaki et al., 2002).
In this study amaranth had a genotoxic effect at all concentrations tested as it is indicated from the statistically signicant increase of the frequency of SCEs/cell. The level of SCEs/cell at
8 mM increased 1.7 times over the control level. These results
probably might be attributed to the two rings that are presented
in the chemical structure of amaranth (Fig. 1). These two symmetrical rings possibly act as intercalating agents to the double strand
of DNA. This may result to breaks on chromosomes and on high
genotoxicity (Table 1).
Cytostaticity of amaranth was observed at the highest concentrations, 4 and 8 mM. Additionally, amaranth presents a high cytotoxic action at 4 and 8 mM. These results were conrmed by the
strong positive correlation between PRI and MI. It is the rst time
in bibliography, that one study includes simultaneous analysis of
genotoxicity, cytostaticity and cytotoxicity of amaranth in cultured
human lymphocytes.
4.2. Cytogenetic studies of erythrosine
Erythrosines genotoxicity and mutagenicity are under discussion provided by some equivocal results in some different cytogenetic tests. In bacterial reversion assays (Auletta et al., 1977;
Brown et al., 1978; Cameron et al., 1987; Haveland-Smith et al.,
1981; Lin and Brusick, 1986) were negative results, whereas in
strains D7 and XV185-14C of S. cerevisiae, gene conversions and reverse mutation were positive (Matula and Downie, 1984). In Ames
test erythrosine gave negative results in a range of 2 mg/plate
(Lakdawalla and Netrawali, 1988) to 10 mg/plate (Lin and Brusick,
1986).
Erythrosine was positive in the chromosome aberrations test
in vitro using a Chinese hamster broblast cell line (Ishidate et al.,
1984). Similarly, erythrosine is inactive in the SCEs test in peripheral
blood lymphocytes, in the micronuclei assay in peripheral blood reticulocytes and in bone marrow polychromatic erythrocytes (Zijno
et al., 1994). Additionally, chromosome aberrations were observed
using Syrian hamster embryo (SHE) cells when treated in the presence of exogenous metabolic activation (Hagiwara et al., 2006). In
the same cells, erythrosine at concentrations between 10 and
110 lM had no effect on SCE induction (Miyachi and Tsutsui, 2005).
Studies have been published in reproduction effects of erythrosine but with negative results (Burnett et al., 1974; Borzelleca and
Hallagan, 1990; Collins et al., 1993a,b; Tanaka, 2001). These data
demonstrate that erythrosine was neither foetotoxic nor teratogenic. Erythrosine is not the only dye with no adverse effects on
reproductive indices. Amaranth (Collins and McLaughlin, 1972;
Larsson, 1975; Piersma et al., 1995) and tartrazine (Borzelleca
and Hallagan, 1988; Collins et al., 1990, 1992) have been observed
to cause no foetal malformations or teratogenic effects in different
experimental animals.
In this study erythrosine at 1, 2, 4 and 8 mM presents a high
cytotoxicity and cytostaticity (Table 2). These results are in concurrence with previous studies (Maskaleris et al., 1998). No signs of
genotoxicity were observed.
4.3. Cytogenetic studies of tartrazine
It was proved that the administration of tartrazine up to ADI
does not create cytogenetic damages. But, at higher concentrations
has reverse effects (Renner, 1984; Giri et al., 1990). Data pertaining
to the genotoxicity or carcinogenicity of tartrazine in various sys-

tems with negative results are available (Kada et al., 1972; Brown
et al., 1978; Ishidate et al., 1984; Price et al., 1978; Haveland-Smith
and Combes, 1980; Chung et al., 1981; Patterson and Butler, 1982;
Chung, 1983; Brown and Dietrich, 1983; Combes, 1986).
On the other hand, tartrazine has been shown to induce chromosomal aberrations in broblast cells of Muntiacus muntjac
(Chung et al., 1981), on bone marrow cells of mice and rats (Giri
et al., 1990) and on chromosomes of Allium cepa (Roychoudhury
and Giri, 1989).
In this study two higher concentrations of tartrazine (4 and
8 mM) have a signicant toxic effect on the quality of chromosomes, so the differentiation stain between two chromatids was
not clear. One possible explanation of this phenomenon is that tartrazine was toxic at the condensation of chromosomes in mitosis.
At the other concentrations no signs of genotoxicity were observed. Reversely, the azo dye had cytotoxic effects at 4 and
8 mM, as published before (Patterson and Butler, 1982).

5. Conclusion
Our results indicate that the food colorants, that had a positive
effect in the present study, are potentially genotoxic to mammalian
cells. The cytogenetic analysis of peripheral human lymphocytes
used in this study can be regarded as a valuable tool for the assessment of the in vitro genotoxicity of food colors such as erythrosine,
amaranth and tartrazine. The binding studies of the above dyes by
spectroscopic titration showed distinct isosbestic points in spectrum at 297, 242 and 307 nm, respectively and that means clear
binding of these dyes to DNA, producing each of them a single
product with the molecule of DNA, and this is of great clinical
importance. Combinations of these food colors in in vitro and
in vivo studies with or without metabolic activation are our future
projects.

Conict of Interest
The authors declare that there are no conicts of interest.

References
Afrati, T., Dendrinou-Samara, C., Raptopoulou, C., Terzis, A., Tangoulis, V., Tsipis, A.,
Kessissoglou, D.P., 2008. Experimental and theoretical study of the
antisymmetric magnetic behavior of copper inverse-9-metallacrown-3
compounds. Inorg. Chem. 47, 75457555.
Afrati, T., Pantazaki, A.A., Dendrinou-Samara, C., Raptopoulou, C., Terzis, A.,
Kessissoglou, D.P., 2010. Copper inverse-9-metallacrown-3 compounds
interacting with DNA. Dalton Trans. 39, 765775.
Al-Mossawi, J.M.A., 1983. The mutagenic effect of amaranth (FD and C Red No. 2) in
bacteria and yeast. Environ. Int. 9, 145148.
Arnold, D.W., Kennedy, G.L., Keplinger Jr., M.L., Calandra, J.C., 1976. Failure of FD & C
No. 2 to produce dominant lethal effects in the mouse. Food Cosmet. Toxicol. 14,
163165.
Auletta, E.A., Kuzava, M.J., Parmar, S.A., 1977. Lack of mutagenic activity of a series
of food dyes for Salmonella typhimurium. Mutat. Res. 5, 203206.
Bakopoulou, A., Mourelatos, D., Tsiftsoglou, A.S., Mioglou, E., Gares, P., 2008. Sisterchromatid exchange, chromosomal aberrations and delays in cell-cycle kinetics
in human lymphocytes induced by dental composite resin eluates. Mutat. Res.
649, 7990.
Borzelleca, J.F., Hallagan, J.B., 1988. A chronic toxicity/carcinogenicity studies of FD
& C Yellow No. 5 (tartrazine) in mice. Food Chem. Toxicol. 26, 189194.
Borzelleca, J.F., Hallagan, J.B., 1990. Multigeneration study of FD & C Red No. 3
(erythrosine) in SpragueDawley rats. Food Chem. Toxicol. 28, 813819.
Brown, J.P., Dietrich, P.S., 1983. Mutagenicity of selected sulfonated azo dyes in the
Salmonella/microsome assay: use of aerobic and anaerobic activation
procedures. Mutat. Res. 116, 305315.
Brown, P.J., Roehm, W.G., Brown, J.R., 1978. Mutagenicity testing of certied food
colors and related azo, xanthene and triphenylmethane dyes with the
Salmonella/microsome system. Mutat. Res. 56, 249271.
Burnett, C.M., Agersborg, H.P.K., Borzelleca, J.F., Eagle, E., Ebert, A.G., Pierce, E.C.,
Kirschman, J.C., Scala, R.A., 1974. Teratogenic studies with certied colors in rats
and rabbits. Toxicol. Appl. Pharmacol. 29, 75155.

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

Cameron, T.P., Hughes, T.J., Kirby, P.E., Fung, V.A., Dunkel, V.C., 1987. Mutagenic
activity of 27 dyes and related chemicals in the Salmonella/microsome and
mouse lymphoma TK+/ assays. Mutat. Res. 189, 223261.
Chaviara, A.T., Kioseoglou, E.E., Pantazaki, A.A., Tsipis, A.C., Karipidis, P.A., Kyriakidis,
D.A., Bolos, C.A., 2008. DNA interaction studies and evaluation of biological
activity of homo- and hetero-trihalide mononuclear Cu(II) Schiff base
complexes. Quantitative structure-activity relationships. J. Inorg. Biochem.
102, 17491764.
Chow, C.S., Barton, J.K., 1992. Transition metal complexes as probes of nucleic acids.
Methods Enzymol. 212, 219242.
Chung, K.T., 1983. The signicance of azo-reduction in the mutagenesis and
carcinogenesis of azo dyes. Mutat. Res. 114, 269281.
Chung, K.T., Fulk, E.G., Andrews, A.W., 1981. Mutagenicity testing of some
commonly used dyes. Appl. Environ. Microbiol. 42, 641648.
Collins, T.F., McLaughlin, J., 1972. Teratology studies on food colourings. Part I.
Embryotoxicity of amaranth (FD & C Red No. 2) in rats. Food Cosmet. Toxicol. 10,
619624.
Collins, T.F., Black, T.N., Brown, L.H., Bulhack, P., 1990. Study of the teratogenic
potential of FD & C Yellow No. 5 when given by gavage to rats. Food Chem.
Toxicol. 28, 821827.
Collins, T.F., Black, T.N., ODonnell, M.W., Bulhack, P., 1992. Study of the teratogenic
potential of FD & C Yellow No. 5 when given in drinking-water. Food Chem.
Toxicol. 30, 263268.
Collins, T.F., Black, T.N., ODonnell, M.W., Shackelford, M.E., Bulhack, P., 1993a.
Teratogenic potential of FD & C Red No. 3 when given in drinking water. Food
Chem. Toxicol. 3, 161167.
Collins, T.F., Black, T.N., Ruggles, D.I., 1993b. Teratogenic potential of FD & C Red No.
3 when given by gavage. Toxicol. Ind. Health 9, 605616.
Combes, R.D., 1986. On the mutagenicity of tartrazine (FD & C Yellow No 5). Arch.
Toxicol. 59, 6768.
Digkas, E., Kareli, D., Chrisa, S., Passadaki, T., Mantadakis, E., Hatzimichail, A.,
Vargemezis, V., Lialiaris, T., 2010. Attenuation of cytogenetic effects by
erythropoietin in human lymphocytes in vitro and P388 ascites tumor cells
in vivo treated with irinotecan (CPT-11). Food Chem. Toxicol. 48, 242249.
Dimitrakopoulou, A., Dendrinou-Samara, C., Pantazaki, A.A., Alexiou, M.,
Nordlander, E., Kessissoglou, D.P., 2008. Synthesis, structure and interactions
with DNA of novel tetranuclear, [Mn4(II/II/II/IV)] mixed valence complexes. J.
Inorg. Biochem. 102 (4), 618628.
Giri, A.K., Das, S.K., Talukder, G., Sharma, A., 1990. Sister chromatid exchange and
chromosome aberrations induced by curcumine and tartrazine on mammalian
cells in vivo. Cytobios 62, 111117.
Goto, K., Maeda, S., Kano, Y., Sugiyama, T., 1978. Factors involved in differential
Giemsa staining of sister chromatids. Chromosoma (Berl) 66, 351359.
Hagiwara, M., Watanabe, E., Barrett, C.J., Tsutsui, T., 2006. Assessment of
genotoxicity of 14 chemical agents used in dental practice. Ability to induce
chromosome aberrations in Syrian hamster embryo cells. Mutat. Res. 603, 111
120.
Haveland-Smith, R.B., Combes, R.D., 1980. Screening of food dyes for genotoxic
activity. Food Cosmet. Toxicol. 18, 215221.
Haveland-Smith, R.B., Combes, R.D., Bridges, B.A., 1981. Studies on the genotoxicity
of some uorescein dyes. Mutat. Res. 88, 115.
Heidelberg, C., Freeman, E.A., Pienta, J.R., Sivak, A., Bertram, S.J., Casto, C.B., Dunkel,
C.V., Francis, W.M., Kakunaga, T., Little, B.J., Schechtman, M.L., 1983. Cell
transformation by chemical agents a review and analysis of the literature. A
report of the U.S. Environmental Protection Agency Gene-Tox program. Mutat.
Res. 114, 283385.
Ioannidou, E., Lialiaris, T., Mourelatos, D., Dozi-Vassiliades, J., 1989. Synergistic
induction of cytogenetic damage by alkylating antineoplastics and 5azacytidine in human lymphocytes. Environ. Mol. Mutagen. 14, 612.
Ishidate, M., Sofuni Jr., T., Yoshikawa, K., Hayashi, M., Nohmi, T., Sawada, M.,
Matsuoka, A., 1984. Primary mutagenicity screening of food additives currently
used in Japan. Food Chem. Toxicol. 22, 623636.
Izbirak, A., Sumer, S., Diril, N., 1990. Mutagenicity testing of some azo dyes used as
food additives. Mikrobiyol. Bul. 24, 4856.
Johnson, D.R., Jasin, M., 2000. Sister chromatid gene conversion is a prominent
double-strand break repair pathway in mammalian cells. EMBO J. 19, 33983407.
Kada, T., Tutikawa, K., Sadaie, Y., 1972. In vitro and host-mediated rec-assay
procedures for screening chemical mutagens; and phloxine, a mutagenic red
dye detected. Mutat. Res. 16, 165174.
Karapidaki, I., Bakopoulou, A., Papageorgiou, A., Iakovidou, Z., Mioglou, E.,
Nikolaropoulos, S., Mourelatos, D., Lialiaris, T., 2009. Genotoxic, cytostatic,
antineoplastic and apoptotic effects of newly synthesized antitumour steroidal
esters. Mutat. Res. 675, 5159.
Karapidaki, I., Papageorgiou, A., Geromichalos, G., Fousteris, M., Papaconstadinou, I.,
Pairas, G., Koutsourea, A., Mourelatos, D., Nikolaropoulos, S., Lialiaris, T., 2010.
Cytogenetic and antineoplastic effects of modied steroidal alkylators. Genet.
Test. Mol. Biomark. 14 (1), 9397.
Kelly, T.M., Tossi, A.B., Mckonnell, D.J., OhUigin, C., 1985. A study of the interactions
of some polypyridylruthenium (II) complexes with DNA using uorescence
spectroscopy, topoisomerisation and thermal denaturation. Nucleic Acids Res.
13, 60176034.
Lakdawalla, A.A., Netrawali, M.S., 1988. Mutagenicity, comutagenicity, and
antimutagenicity of erythrosine (FD & C Red 3), a food dye, in the Ames/
Salmonella assay. Mutat. Res. 204, 131139.
Larsson, K.S., 1975. A teratologic study with the dyes amaranth and Ponceau 4R in
mice. Toxicology 4, 7582.

2943

Lialiaris, T., Mourelatos, D., Dozi-Vassiliades, J., 1987. Enhancement and attenuation
of cytogenetic damage by vitamin C in cultured human lymphocytes exposed to
thiotepa or L-ethionine. Cytogenet. Cell Genet. 44, 209214.
Lialiaris, T., Mourelatos, D., Dozi-Vassiliades, J., 1988. Enhancement of cytogenetic
damage by chlorpromazine in human lymphocytes treated with alkylating
antineoplastics and caffeine. Mutat. Res. 206, 361365.
Lialiaris, T., Mourelatos, D., Boutis, L., Papageorgiou, A., Christianopoulou, M.,
Papageorgiou, V., Dozi-Vassiliades, J., 1989. Comparative study on cytogenetic
effects by diplatinum complexes of the ligands of naphthazarine and squaric
acid in human lymphocytes. J. Pharmacol. Exp. Ther. 251, 368371.
Lialiaris, T., Mourelatos, D., Stergiadou, H.C., Constantinidou, H.A., 1990. Cytogenetic
study for possible mutagenic activity induced by ice-nucleation bacteria or their
metabolic products in human lymphocytes in vitro. Mutat. Res. 242, 163168.
Lialiaris, T., Pantazaki, A., Sivridis, E., Mourelatos, D., 1992. Chloropromazineinduced damage on nucleic acids: a combined cytogenetic and biochemical
study. Mutat. Res. 265, 155163.
Lialiaris, T.S., Pantazaki, A., Papachristou, F.E., Lyratzopoulos, E., Natsis, K., Kortsaris,
A.H., 2007. The mutagenic potential of vitamin C on human lymphocytes and
native nucleic acids. J. Biol. Res. Thessalon 8, 189197.
Lialiaris, T., Lyratzopoulos, E., Papachristou, F., Simopoulou, M., Mourelatos, C.,
Nikolettos, N., 2008. Supplementation of Melatonin protects human
lymphocytes in vitro from the genotoxic activity of Melphalan. Mutagenesis
23, 347354.
Lialiaris, T., Polyzou, A., Mpountoukas, P., Tsiggene, A., Kouskoukis, A., Pouliliou, S.,
Paraskakis, E., Tentes, I., Trypsianis, G., Chatzimichail, A., 2009a. Chromosome
instability on children with asthma. J. Asthma 46 (8), 841844.
Lialiaris, T.S., Kotsiou, E., Pouliliou, S., Kareli, D., Makrinou, H., Kouskoukis, A.,
Papachristou, F., Koukourakis, M., 2009b. Cytoprotective activity of amifostine
on cultured human lymphocytes exposed to irinotecan. Food Chem. Toxicol. 47
(10), 24452449.
Lialiaris, T.S., Papachristou, F., Mourelatos, C., Simopoulou, M., 2009c. Antineoplastic
and cytogenetic effects of chlorpromazine on human lymphocytes in vitro and
on Ehrlich ascites tumor cells in vivo. Anticancer Drugs 20 (8), 746751.
Lialiaris, T., Mavromatidou, P., Digkas, E., Passadaki, T., Mpountoukas, P.,
Panagoutsos, S., Vargemezis, V., 2010. Chromosome instability in patients
with chronic renal failure. Genet. Test. Mol. Biomark. 14 (1), 3741.
Lin, H.Y.G., Brusick, J.D., 1986. Mutagenicity studies on FD & C Red No. 3.
Mutagenesis 1, 253259.
Maskaleris, T., Lialiaris, T., Triantaphyllidis, C., 1998. Induction of cytogenetic
damage in human lymphocytes in vitro and of antineoplastic effects in Ehrlich
ascites tumor cells in vivo treated by methotrexate, hyperthermia and/or
caffeine. Mutat. Res. 422, 229236.
Matula, T.I., Downie, R.H., 1984. Genetic toxicity of erythrosine in yeast. Mutat. Res.
138, 153156.
Minioti, S.K., Sakellariou, F.C., Thomaidis, S.N., 2007. Determination of 13 synthetic
food colorants in water-soluble foods by reversed-phase high-performance
liquid chromatography coupled with diode-array detector. Anal. Chim. Acta
583, 103110.
Miyachi, T., Tsutsui, T., 2005. Ability of 13 chemical agents used in dental practice to
induce sister-chromatid exchanges in Syrian hamster embryo cells. Odontology
93, 2429.
Mourelatos, D., Dozi-Vassiliades, J., Kotsis, A., Gourtsas, K., 1988. Enhancement of
cytogenetic damage and of antineoplastic effect by caffeine in Ehrlich ascites
tumor cells treated with cyclophosphamide in vivo. Cancer Res. 48, 11291131.
Mpountoukas, P., Vantarakis, A., Sivridis, E., Lialiaris, T., 2008. Cytogenetic study in
cultured human lymphocytes treated with three commonly used preservatives.
Food Chem. Toxicol. 46, 23902393.
Munzner, R., 1979. Mutagenicity testing of the urine of rats treated with amaranth.
Food Cosmet. Toxicol. 17, 563.
Pan, X., Ushio, H., Ohshima, T., 2005. Effects of molecular congurations of food
colorants on their efcacies as photosensitizers in lipid oxidation. Food Chem.
92, 3744.
Pantazaki, A.A., Lialiaris, Th.S., 1999. A combined biochemical and cytogenetic study
of thioridazine-induced damage to nucleic acids. Mutagenesis 14, 243248.
Papachristou, F., Lialiaris, T., Touloupidis, S., Kalaitzis, C., Simopoulos, C., Sokitis, N.,
2006. Evidence of increased chromosomal instability in infertile males after
exposure to mitomycin C and caffeine. Asian J. Androl. 8, 199204.
Papachristou, F., Simopoulou, M., Touloupidis, S., Tsalikidis, C., Sokitis, N., Lialiaris,
T., 2008. DNA damage and chromosomal aberrations in various types of male
factor infertility. Fertil. Steril. 90, 17741781.
Parry, J.M., 1977. The use of yeast cultures for the detection of environmental
mutagens using a uctuation test. Mutat. Res. 46, 165175.
Patterson, R.M., Butler, J.S., 1982. Tartrazine-induced chromosomal aberrations in
mammalian cells. Food Chem. Toxicol. 20, 461465.
Piersma, H.A., Attenon, P., Bechter, R., Govers, J.A.P.M., Krafft, N., Schmid, P.B., Stadler,
J., Verhoef, A., Verseil, C., 1995. Interlaboratory evaluation of embryotoxicity in
the postimplantation rat embryo culture. Reprod. Toxicol. 9, 275280.
Polyanichko, A.M., Andrushchenko, V.V., Chikhirzhina, E.V., Vorobev, V.I., Wieser,
H., 2004. The effect of manganese(II) on DNA structure: electronic and
vibrational circular dichroism studies. Nucleic Acids Res. 32, 989996.
Price, J.P., Suk, A.W., Freeman, E.A., Lane, T.W., Peters, L.R., Vernon, M.L., Huebner,
J.R., 1978. In vitro and in vivo indications of the carcinogenicity and toxicity of
food dyes. Int. J. Cancer 21, 361367.
Prival, J.M., Davis, M.V., Peiperl, D.M., Bell, J.S., 1988. Evaluation of azo food dyes for
mutagenicity and inhibition of mutagenicity by methods using Salmonella
typhimurium. Mutat. Res. 206, 247259.

2944

P. Mpountoukas et al. / Food and Chemical Toxicology 48 (2010) 29342944

Renner, H.W., 1984. Tartrazine a reinvestigation by in vivo cytogenetic methods.

Food Chem. Toxicol. 22, 327.
Revankar, S.M., Lele, S.S., 2007. Synthetic dye decolorization by white rot fungus,
Ganoderma sp. WR-1. Bioresour. Technol 98, 775780.
Rogers, C.G., Boyes, B.G., Matula, T.I., Heroux-Metcalf, C., Clayson, D.B., 1988. A case
report: A multiple end-point approach to evaluation of cytotoxicity and
genotoxicity of erythrosine (FD & C Red No. 3) in a V79 hepatocyte-mediated
mutation assay. Mutat. Res. 205, 415423.
Roychoudhury, A., Giri, A.K., 1989. Effects of certain food dyes on chromosome of
Allium cepa. Mutat. Res. 223, 313319.
Sankaranarayanan, N., Murthy, M.S., 1979. Testing of some permitted food colours
for the induction of gene conversion in diploid yeast. Mutat. Res. 67, 309314.
Sasaki, Y.F., Kawaguchi, S., Kamaya, A., Ohshita, M., Kabasawa, K., Iwama, K.,
Taniguchi, K., Tsuda, S., 2002. The comet assay with 8 mouse organs: results
with 39 currently used food additives. Mutat. Res. 519, 103119.

Sonoda, E., Sasaki, M.S., Morrison, C., Yamaguchi-Iwai, Y., Takata, M., Takeda, S.,
1999. Sister Chromatid Exchanges are mediated by homologous recombination
in vertebrate cells. Mol. Cell. Biol. 19, 51665169.
Stanimirovic, Z., Stevanovic, J., Jovanovic, S., Andjelkovic, M., 2005. Evaluation of
genotoxic effects of Apitol (cymiazole hydrochloride) in vitro by measurement
of sister chromatid exchange. Mutat. Res. 588, 152157.
Tanaka, T., 2001. Reproductive and neurobehavioural toxicity study of erythrosine
administered to mice in the diet. Food Chem. Toxicol. 39, 447454.
Tripathy, N.K., Nabi, Md.J., Sahu, G.P., Kumar Anand, A., 1995. Genotoxicity testing of
two red dyes in the somatic and germ line cells of Drosophila. Food Chem.
Toxicol. 33, 923927.
Zijno, A., Marcon, F., Leopardi, P., Salvatore, G., Carere, A., Crebelli, R., 1994. An
assessment of the in vivo clastogenicity of erythrosine. Food Chem. Toxicol. 32,
159163.