DNA-damaging activity and mutagenicity of 16 newly synthesized thiazolo[5,4-a]acridine derivatives with high photo-inducible cytotoxicity

Article abstract of DOI:10.1016/j.mrgentox.2007.10.022

The discovery of the potent anticancer properties of natural alkaloids in the pyrido-thiazolo-acridine series has suggested that thiazolo-acridine derivatives could be of great interest. In a continuous attempt to develop DNA-binding molecules and DNA photo-cleavers, 16 new thiazolo[5,4-a]acridines were synthesized and studied for their photo-inducible DNA-intercalative, cytotoxic and mutagenic activities, by use of the DNA methyl-green bioassay, the Alamar Blue viability assay and the Salmonella mutagenicity test using strains TA97a and TA98 with and without metabolic activation and photo-activation. Without photo-activation, one compound showed a DNA-intercalative activity in the DNA major groove while three compounds displayed intercalating properties after photo-activation. In the dark, four molecules possessed cytotoxic activities against a THP1 acute monocytic leukemia cell line while 15 derivatives displayed photo-inducible cytotoxic activity against this cell line. All compounds were mutagenic in strain TA97a with metabolic activation (+S9mix) and 15 molecules were mutagenic in strain TA98 without activation (-S9mix). Study of the quantitative structure-activity relationships (QSAR) from the Salmonella mutagenicity data revealed that several descriptors could describe cytotoxic and mutagenic activities after photo-activation. From the results of the mutagenicity test, four compounds with elevated mutagenic activities were selected for additional experiments. Their capacities to induce single-strand breaks (SSB) and chromosome-damaging effects were monitored by the comet and the micronucleus assays in normal human keratinocytes. Comparison of the minimal genotoxic concentrations showed that two compounds possessed higher capacities to induce SSB after photo-activation. In the micronucleus assay, three molecules were able to induce high numbers of micronuclei following photo-activation. Overall, the results of this study confirm that acridines are predominantly genotoxic via a DNA-intercalating mechanism in the dark, while DNA-adducts were probably induced following photo-activation.

Full text of DOI:10.1016/j.mrgentox.2007.10.022

Available online at www.sciencedirect.com
Mutation Research 650 (2008) 104–114
DNA-damaging activity and mutagenicity of 16 newly synthesized
thiazolo[5,4-a]acridine derivatives with
high photo-inducible cytotoxicity
a
a
a
a
b
´
Carole Di Giorgio , Anna Nikoyan , Laetitia Decome , Celine Botta , Maxime Robin ,
c
a
b
a,∗
´
Jean-Pierre Reboul , Anne-Sophie Sabatier , Alain Matta , Michel De Meo
a Laboratoire de Bioge´notoxicologie et Mutagene`se Environnementale (EA 1784), Faculte´s de Me´decine et Pharmacie,
Universite´ de la Me´diterrane´e, 27 Bd Jean Moulin, 13385 Marseille Cedex 05, France
b UMR Symbio 6178, Universite´ Paul Ce´zanne Aix-Marseille III, Faculte´ des sciences, Case 552,
Avenue Escadrille Normandie Nie´men, 13397 Marseille Cedex 20, France
c Laboratoire de Chimie The´rapeutique, Faculte´ de Pharmacie, Universite´ de la Me´diterrane´e,
27 Bd Jean Moulin, 13385 Marseille Cedex 05, France
Received 31 January 2007; received in revised form 22 May 2007; accepted 7 October 2007
Available online 19 November 2007
Abstract
The discovery of the potent anticancer properties of natural alkaloids in the pyrido-thiazolo-acridine series has suggested that thiazolo-acridine
derivatives could be of great interest. In a continuous attempt to develop DNA-binding molecules and DNA photo-cleavers, 16 new thiazolo[5,4-
a]acridines were synthesized and studied for their photo-inducible DNA-intercalative, cytotoxic and mutagenic activities, by use of the DNA
methyl-green bioassay, the Alamar Blue® viability assay and the Salmonella mutagenicity test using strains TA97a and TA98 with and without
metabolic activation and photo-activation. Without photo-activation, one compound showed a DNA-intercalative activity in the DNA major groove
while three compounds displayed intercalating properties after photo-activation. In the dark, four molecules possessed cytotoxic activities against
a THP1 acute monocytic leukemia cell line while 15 derivatives displayed photo-inducible cytotoxic activity against this cell line. All compounds
were mutagenic in strain TA97a with metabolic activation (+S9mix) and 15 molecules were mutagenic in strain TA98 without activation (−S9mix).
Study of the quantitative structure–activity relationships (QSAR) from the Salmonella mutagenicity data revealed that several descriptors could
describe cytotoxic and mutagenic activities after photo-activation. From the results of the mutagenicity test, four compounds with elevated mutagenic
activities were selected for additional experiments. Their capacities to induce single-strand breaks (SSB) and chromosome-damaging effects were
monitored by the comet and the micronucleus assays in normal human keratinocytes. Comparison of the minimal genotoxic concentrations showed
that two compounds possessed higher capacities to induce SSB after photo-activation. In the micronucleus assay, three molecules were able to
induce high numbers of micronuclei following photo-activation. Overall, the results of this study confirm that acridines are predominantly genotoxic
via a DNA-intercalating mechanism in the dark, while DNA-adducts were probably induced following photo-activation.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Thiazolo-acridines; DNA-binding activity; Salmonella mutagenicity test; Comet assay; In vitro micronuclei; Cytotoxicity; QSAR; Photoactivation;
Simulated sunlight
1. Introduction
infectious activities [1,2]. In the mid 1960s, the discovery of
the activity of acridine in preventing the growth of tumor cells
led to the possibility of using such compounds in anticancer
therapy and stimulated the synthesis of new derivatives [3].
Recently, the synthesis of acridines with additional rings, e.g.,
acromycine [4,5], has focused the interest on tetracyclic com-
pounds. Furthermore, the discovery of natural alkaloids of the
pyrido-thiazolo-acridine series with potent anticancer properties
Since their first synthesis by Graebe and Caro in 1870,
acridine chromophores have revealed a broad spectrum of anti-
∗
Corresponding author. Tel.: +33 4 91 83 55 60; fax: +33 4 91 79 86 75.
E-mail address: michel.demeo@pharmacie.univ-mrs.fr (M. De Me´o).
1383-5718/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.mrgentox.2007.10.022

C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
105
[6,7] has suggested that thiazolo-acridine derivatives could be
of great interest.
rial for the synthesis of all other compounds. The reduction of the nitro-group
with iron powder in hydrochloric acid led to 2-chloro-amino-benzothiazole.
Then, the Ullmann condensation [30] with orthobromobenzoic acid gave the
corresponding N-6-aryl-anthranilic acid. The cyclation of N-6-aryl-anthranilic
acid was done according to the methodology of Friedel-Crafts [30] with POCl3
or H2SO4 to produce, respectively, compounds 3 and 11. The thiation of
compound 3 with P4S10 produced compound 7, and compound 5 after an
additional S-alkylation. Compound 10 was obtained by substitution of the
chloro-group with ammonium carbonate in phenol. In parallel, the thiation of
2,11-dichloro-thiazolo[5,4-a] acridine (11) with P2S5 provided compound 8,
while the substitution of chlorine in position 11 with phenol or alkylamines
gave, respectively, compound 13, and compounds 2, 6, 14 and 51.
An additional synthetic pathway was developed to obtain acridine deriva-
tives bearing a 2-substituted thiazole-ring: 2-chloro-6-nitro-benzothiazole was
subjected to nucleophilic substitution by various alkylamines to give the corre-
sponding 2-substituted 6-nitro-benzothiazoles and the equivalent 2-substituted
6-amino-benzothiazoles were then obtained by reduction of the nitro-groups
according to the methodology of Katz, which finally yielded the thiazolo[5,4-
a]acridine analogues 1, 9, 67, 70 and 86 following Ullmann condensation with
orthobromobenzoic acid and cyclization according to the protocol developed by
Friedel-Crafts.
The mechanisms by which acridines exert their pharmaco-
logical properties have been shown to be closely related to their
capacity to reversibly bind with DNA [8], to interact with DNA
regulatory enzymes [9] and to disrupt DNA functions in the
cell. The concept of the intercalation of acridines between adja-
cent base pairs of the double-helical DNA, first proposed for
proflavine [8], has been extensively demonstrated with vari-
ous other derivatives [10]. Subsequently, acridine derivatives
have revealed genotoxic properties leading to mutagenicity [11],
reproductive toxicity and carcinogenicity [12]. However, while
simple acridines were ascertained as intercalating agents and
model frameshift mutagens in bacteriophages [13], bacteria
[14,15] and mammalian cells [16], several derivatives revealed
a large variety of mutagenic properties. Nitroacridines, capable
of reacting as alkylating agents depending on the position of
the nitro-group [17,18] and ICR compounds, in which the acri-
dine ring was used as a carrier for targeting alkylating agents to
DNA [19], produced both frameshift and base-pair substitutions
[20,21]. Acridines with additional fused aromatic rings such as
benzacridines were shown to interact covalently with DNA after
metabolic activation to mainly produce base-pair substitution
mutations [22,23]. This suggested that the DNA-binding prop-
erties of acridine-based compounds could substantially differ
according to their chemical structure, the intracellular envi-
ronment or the experimental conditions, and illustrated the
multi-factorial aspect of acridine-induced mutagenic events.
Among the various factors capable of modulating the cellu-
lar effects of acridine, the involvement of oxidative stress by
reactive oxygen species has been suggested since the 1980s
[24]. This was supported by the fact that acridine-induced DNA
damage was greatly enhanced with increasing oxidation poten-
tial of the compounds and demonstrated the importance of
electron-transfer reactions during acridine–DNA interactions
[24]. As a consequence, several members of the acridine series
exhibited photo-inducible properties, mainly photo-toxicity,
photo-bactericidal activity [2] and photo-enhanced genotoxicity
[25].
Chemical structures of acridine derivatives were characterized by 1D and
2D NMR analyses. TLC, microanalysis and HPLC confirmed that the purity of
the compounds was >99.5%. In all assays, acridine orange (CAS No. 65-61-
2, Sigma–Aldrich) was used as the reference compound. All compounds were
dissolved in spectra-grade dimethyl sulfoxide (DMSO) at their solubility limit
(5–10 mg/ml). The solutions were kept frozen at −80 ◦C until used.
2.2. Irradiation procedure
Irradiation was performed with a solar simulator Suntest CPS+ (Atlas Mate-
rial Testing Technology BV, Moussy le Neuf, France) equipped with a xenon-arc
lamp, a treated quartz filter to block infrared light and a glass filter to block
UVC light and to reduce UVB light (UVA/UVB: 0.5/0.001 mW/cm2, simu-
lated sunlight 800 lx). Preliminary dose-finding experiments performed with S.
typhimurium TA98 and various doses of acridine orange showed that the optimal
incident dose of light was 2 mJ/cm2 for a period of 1 min.
2.3. DNA–methyl green bioassay
The capacities of acridine derivatives to interact with the DNA major groove
were assessed by the DNA–methyl green bioassay [26], using the triphenyl
methane dye methyl green, a major groove binding agent which may reversibly
interact with DNA according to a reaction that can be followed spectropho-
tometrically. Twenty milligrams DNA–methyl green (Sigma) were suspended
in 100 ml of 0.05 M Tris–HCl buffer, pH 7.5, containing 7.5 mM of magnesium
sulfate and stirred at 37 ◦C during 24 h. A volume of 200 ␮l of the above solution
was put into each well of a multi-well plate and various concentrations of acri-
dine derivatives were added (from 0.1 to 100 ␮M). The initial absorbance of each
sample was measured at 630 nm. After an incubation period of 24 h in the dark at
room temperature, the final absorbance of the samples was measured again. The
readings were corrected for the initial absorbance and normalized as percentages
of the untreated DNA–methyl green absorbance value. Dose–response curves
were calculated by non-linear regression analysis (TableCurve 2.1), according to
a log-normal mathematical model. IC80-DNA corresponded to the concentration
of chemical compounds that induced a 20% decrease of absorbance.
In an attempt to discover new DNA photo-cleavers and to bet-
ter understand the genotoxic consequences of their interactions
with DNA, 16 thiazolo[5,4-a]acridines were synthesized and
assessed for their photo-inducible DNA-binding activity [26],
mutagenicity in Salmonella typhimurium [27], and cytoxicity
against human monocytes. The most mutagenic compounds of
this series were then tested for the ability to induce single-strands
breaks and heritable chromosomal damages, by use of the comet
assay [28] and the micronucleus assay in human keratinocytes
[29].
2.4. Salmonella mutagenicity test
2. Materials and methods
2.4.1. Salmonella tester strains and activation mixture
Salmonella tester strains TA97a, TA98, TA100 and TA102 were gifts from
Prof. B.N. Ames (Berkeley, CA, USA). The strains were stored at −80 ◦C
and regularly checked for genetic markers [33]. The liver homogenate used
for metabolic activation (S9) was prepared from male Sprague–Dawley rats
treated with Aroclor 1254 (500 mg/kg body weight). The protein concentration
was 26 mg/ml as determined by the technique of Lowry [31]. The S9 mix was a
mixture of 4% S9 and a solution of NADPH-generating factors (mix).
2.1. Chemicals
Two classical synthetic routes were used for the synthesis of acridine deriva-
tives [30] as shown in Fig. 1. 2-Chloro-6-nitro-benzothiazole was first prepared
by nitration of 2-chloro-benzothiazole (CAS No. 615-20-3, Sigma–Aldrich,
Saint Quentin Fallavier, France) with HNO3/H2SO4 and used as a starting mate-

106
C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
Fig. 1. Chemical structures and synthetic route of the 16 acridine derivatives.
2.4.2. Assay protocol
revertants/plate, with all compounds being tested in a final volume of 0.2 ml
during the contact period.
The assay was carried out according to [27] using a modified version of
the liquid incubation technique [31]. Salmonella strains were grown in Oxoid
Nutrient Broth No. 2 with ampicillin (25 ␮g/ml) for 12 h at 37 ◦C with gentle
shaking. After the incubation period, the following ingredients were placed in
12-mm × 75-mm polystyrene tubes: (1) various volumes of the tested molecules
(four tested doses per compound) not exceeding 10 ␮l (0.5%, v/v) to avoid
DMSOtoxicity, (2)0.1 mlofS9mixifneeded;(3)0.1 mloftheovernightculture.
Themixtureswereincubatedeitherfor60 mininthedarkorfor45 mininthedark
followed by irradiation during one minute. Then, 2-ml volumes of melted top
agar containing 0.045 mM histidine and biotin were added to the tubes and the
mixtures were poured onto Vogel-Bonner (VB) minimal plates. For each series
of experiments, the controls included: (i) 5 or 10 ␮l of DMSO without and with
S9 mix, respectively, to determine the numbers of spontaneous revertants/plate;
(ii) positive controls: 0.5 ␮g of acridine orange for quality control of S9 mix
in the dark and 0.05 ␮g of acridine orange for quality control of irradiation.
After a 48-h period of incubation, the revertants were counted with a laser
colony-counter equipped with a bacterial enumeration program (Spiral System
Instrument Inc., Bethesda, MD, USA). All experiments were conducted three
timesandthemutagenic activitywascalculated by non-linear regression analysis
using an arbitrary model on the ascending part of the dose–response curve, as
described previously [31]. The minimal mutagenic concentration was defined as
thelowestconcentration(␮M)inducingadoublingofthenumberofspontaneous
2.5. Cytotoxicity against human transformed monocytes
The cytotoxic activity of acridine derivatives was assessed on the THP1
human acute monocytic leukemia cell line (ATCC, Manassas VA, USA)
by colorimetric determination of the number of viable cells using the
oxidation–reduction indicator Alamar Blue® (Invitrogen, Cergy-Pontoise,
France) [32]. Late log-phase human monocytes were incubated in RPMI
1640 medium without phenol red supplemented with 2 mM l-glutamine, 10%
fetal calf serum and 10% Alamar Blue®. A range of acridine concentra-
tions (from 0.001 to 25 ␮M) was incorporated in duplicate cultures (final
DMSO concentration <1%). After a 48-h incubation period at 37 ◦C in a
5% CO2 atmosphere, enzymatic reduction of Alamar Blue® from blue to
pink in viable cells was measured by monitoring absorbance at 570 and
630 nm. Negative controls were treated with solvent (DMSO) and were included
to each set of experiments. Dose–response curves were calculated by non-
linear regression analysis (TableCurve 2.1 software, Systat Software Inc, USA)
from duplicated experiments according to a log-normal mathematical model.
The IC50-THP1 corresponded to the concentration of acridine compound able
to induce a 50% decrease of cell viability as compared with the control
culture.

C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
2.6. Induction of DNA-strand breaks (Comet assay) and 2.6.3. Micronucleus assay
107
A total of 50,000 human keratinocytes were plated in chamber slides and
incubated during 24 h at 37 ◦C in humidified atmosphere containing 5% CO2.
Solutions of acridine derivatives were thawed before use and added to the cell
cultures at concentrations ranging from 0.01 to 25 ␮M. After a 24-h incuba-
tion period, cells were rinsed and incubated in fresh medium. Cytochalasin B
(5 ␮g/ml) was added into each chamber slide to arrest cytokinesis and the cells
were incubated for an additional 24-h period. At 72 h, cells were submitted
to two successive washes with PBS and fixed with methanol/acetic acid (3/1,
v/v). Staining of air-dried slides was performed with 5% Giemsa stain in Milli-
Q water for 15 min. The binuclearity index (BN), considered as a measure of
cytotoxicity, was determined by scoring the number of binucleated keratinocytes
among 1000 Giemsa-stained cells with well-preserved cytoplasm. The micronu-
cleated cell rates were determined for acridine concentrations inducing less than
50% toxicity [29]: 2000 binucleated keratinocytes were examined and micronu-
clei were identified according to the morphological criteria previously defined
by Fenech [36]. To eliminate micronuclei resulting from apoptosis or necrosis,
cells containing two well-defined cytoplasms and nuclei were scored only [29].
Statistical analyses of differences between controls and treated samples were
performed by the contingency 2×2 chi-square test [37]. The minimal clasto-
genic concentration was defined as the lowest concentration (␮M) inducing a
statistically significant increase of micronucleated cell rates as compared with
the control culture.
micronuclei in normal human keratinocytes
Keratinocytes were used to study the photo-inducible DNA-damaging activ-
ity and clastogenic and/or aneugenic effects of the chemical compounds. These
constitutive cells of the skin provide an efficient barrier against sunlight, which
is essential for survival, and have developed resistance to the lethal effects of
UVB light [33]. Since the 1990s they have been used to assess the photo-
toxicity of molecules included in cosmetics, to detect the photo-protective
activity of sunscreens and to evaluate the genotoxic effects of radiation on
normal human cells [34]. Recently, keratinocytes have been proposed as an
alternative non-animal tool for ascertaining the safety of ingredients in cosmetics
[35].
2.6.1. Isolation and culture of normal human keratinocytes
Normal human keratinocytes were isolated from infant foreskin obtained
after circumcision according to the technique described by Decome et al. [28].
Tissues were cut into 5-mm × 5-mm pieces and incubated in 2.5 ml trypsin
(2.5%, v/v) in saline at 37 ◦C with constant shaking. After a 90-min incuba-
tion period, the trypsin was inactivated by addition of 50 ␮l fetal calf serum.
Dermis and epidermis were separated mechanically and epidermis fragments
were transferred into sterile PBS. After pipetting to break-up the cell pellets, the
cell suspension was centrifuged three times (1200 × g, 5 min). Cells were sus-
pended in K-SFM medium containing 25 ␮g/ml BPE and 0.1–0.2 ng/ml rEGF,
and seeded in flasks in complete K-SFM medium. Cultures were incubated at
37 ◦C in a humidified atmosphere containing 5% CO2, and the culture medium
was changed every 3 days. Pure keratinocyte cultures were obtained within 2
weeks.
2.7. Quantitative structure-activity relationship (QSAR) and
statistical analysis
Quantitativestructure–activityrelationship(QSAR)modelswereformulated
using Cerius2® software (Version 4.2, Accelrys Inc., Cambridge, UK). Multiple
linear regressions correlating thecytotoxicand themutagenicactivities tomolec-
ulardescriptorswereperformedwithrespecttotwoputativetautomericmolecule
series which could exist in aqueous solution, as shown in Fig. 2. A set of 32
constitutive, electronic, geometrical, topological and molecular descriptors was
analyzed by the Cerius2® Descriptor + software [38,39]. Electronic and steric
parameters were selected after a forward stepwise multiple-regression elimi-
nation procedure (Cerius2+). During the development of the best-fit equation
correlation model, molecular descriptors were selected when F-ratios were sta-
tisticallysignificant(P < 0.05). Thus, theselectedparametersincludedHydrogen
Acceptor index (HA), Hydrogen Donor index (HD), energy gap (GAP) related to
the difference between the energy of HOMO and LUMO, Hosoya index (log Z)
related to the number of sets of non-adjacent bonds, Wiener’s index (Wiener)
related to the sum of the distances between all pairs of vertices and the 3Xpv
descriptor involving ramification and steric effects. The statistical level of sig-
nificance of each regression model was evaluated by the correlation coefficient
r, the standard error s, the F-test value, the significance level of the model given
by the leave-one-out cross-validation coefficient Q2 and its standard deviation
of predictive residual sum of squares (SPRESS).
2.6.2. Comet assay
The Comet assay was performed under alkaline conditions [28]. Duplicate
cultures of human keratinocytes were transferred to 35-mm Petri dishes and
treated with various acridine concentrations (0.35, 0.7, 0.14, and 1.75 ␮M). The
first set was directly incubated at 37 ◦C in a humidified atmosphere contain-
ing 5% CO2 for 2 h, while the second set was irradiated before incubation. At
the end of the incubation period, cells were trypsinized (0.05% trypsin, 0.02%
EDTA in PBS) for 3–5 min at 37 ◦C and centrifuged at 1,200 × g for 5 min. Cell
pellets were resuspended in 75 ␮l of low melting-point agarose (LMP, 0.5%
agarose in PBS), placed on pre-treated slides coated with 85 ␮l normal agarose
gel (0.8% in PBS) and finally covered with 75 ␮l LMP. Slides were immersed in
cold lysis solution (2.5 M NaCl, 100 mM Na2EDTA, 10 mM Tris-HCl, 1% N-
lauryl-sarcosinate, pH 10–12, 1% triton X-100 and 10% DMSO) during 90 min
at 4 ◦C. After cell lysis, the microscope slides were placed in an electrophoresis
tank and DNA was allowed to unwind in freshly prepared alkaline electrophore-
sis buffer (30 mM NaOH, 1 mM EDTA, pH 13) for 20 min at room temperature.
Then, electrophoresis was conducted for an additional 20 min period at 25 V and
300 mA. The slides were rinsed three times with cold neutralizing buffer (0.4 M
Tris–HCl, pH 7.5) for 5 min, dehydrated in pure methanol and kept in a dried-
atmosphere chamber until assessment. DNA damage was analyzed after staining
with ethidium bromide (2 ␮g/ml in ultrapure water) and examination of slides
was performed with a BH2-RFL fluorescence microscope (Olympus Optical
Co, Tokyo, Japan) equipped with a 20BG-W2 dichroic mirror (band-passfilter,
515–560 nm; long-pass filter, 590 nm) and a D plan-Apo 20× UV objective
(oil immersion). Images were analyzed with Fenestra Komet software (Kinet-
ics Imaging, Liverpool, UK, version 3.1) on 100 randomly selected cells from
duplicate slides. DNA damage was quantified as the increase of the Olive Tail
Moment (OTM). Normalized distribution frequencies of OTM were calculated
using 40 OTM classes between minimal and maximal values for each set of data.
Non-linear regression analyses were performed with a χ2-function model. The
calculated degree of freedom (n) of the function has been previously shown to be
a quantitative parameter of the level of DNA-damage in samples and was named
OTM χ2. The value of OTM χ2, expressed in arbitrary units, was used as the
only parameter for comparison purposes. The minimal DNA-damaging concen-
tration was defined as the lowest acridine concentration showing a statistically
significant DNA-damaging activity. The minimal DNA-damaging concentration
was calculated from regression models, assuming that an OTM χ2 of 2.4 0.11
was statistically significant from the control value (2.11 0.09).
3. Results
Complete data concerning the DNA-intercalating activity, the
mutagenicity and the cytotoxicity of acridine compounds are
reported in Table 1. Only a few acridine derivatives could com-
petitively displace the DNA–methyl green complex. Compound
6 displayed interactions with DNA in the dark (IC80 = 12.5 ␮M),
while compounds 3, 5 and 7 interacted with DNA following
irradiation (IC80 averaging 51.6, 69.2 and 8.4 ␮M, respectively).
Preliminary testing with the Salmonella mutagenicity test
showed that the mutagenic activity of two compounds (67 and
acridine orange, AO) required the presence of metabolic activa-
tion in the dark. Both were mutagenic in tester strains TA97a,
TA98, TA100 and TA102, with TA97a being the most sensi-
tive. Following photo-activation, AO and 67 were mutagenic
on all strains with and without S9mix. The presence of the

108
C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
Fig. 2. QSAR analysis with respect of molecule sets I and II.
metabolic fraction did not increase the mutagenicity and the
most sensitive strain was TA98 (data not shown). Thereafter,
the evaluation of the mutagenicity was conducted with these
two strains. Results reported in Table 1 showed that in the dark,
compound 6 had the strongest mutagenicity (mutagenic activ-
ity = 6401 rev/nmol), while after irradiation compounds 3, 5 and
7 displayed the highest photo-enhanced mutagenicity in TA98.
No significant correlation could be established between the
mutagenic activities observed in TA97a and in TA98 (P = 0.76),
suggesting two independent mechanisms of action.
Only four acridine derivatives produced a measurable cyto-
toxic activity against human monocytes in the dark: compounds
6 and 13 appeared the most active molecules with IC50s averag-
ing 0.71 and 0.029 ␮M, respectively, while compounds 11 and

C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
109
Table 1
Cytotoxicity, DNA-intercalation and mutagenicity in Salmonella typhimurium of acridine derivatives
Compounds
DNA–methyl green IC90 (␮M)
Mutagenic activity in S.
typhimurium (rev/nmol)
Cytotoxicity against human
transformed monocytes IC50 (␮M)
Without
With
TA 97a + S9 mix
TA98 with
Without
With
photo-activation
photo-activation
photo-activation
photo-activation
photo-activation
1
2
3
5
6
7
8
9
10
11
13
14
51
67
70
86
>100
>100
>100
>100
12.5
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
51.59
8.39
1.9
4.1
4.4
0.5
59.1
16,934.8
98,712.0
65.0
7,641.5
1.2
0.6
9.1
73.0
58.5
181.6
0.0
4.3
0.0
8.4
>25
>25
>25
>25
0.71
>25
>25
>25
>25
6.07
0.029
5.71
>25
>25
>25
>25
0.93
0.52
0.02
0.03
0.29
3.67
1.51
0.52
0.58
0.11
0.03
1.17
0.03
24.46
>25
4.1
>100
69.26
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
6,401.6
9.3
1.3
9.8
29.6
8.7
33.7
0.0
5.1
2.9
153.2
7.4
1.75
Acridine Orange
>100
>100
560.9
903.6
2.21
0.016
>100: IC90 could not determined and was assumed to be above the highest concentration tested (100 ␮M). >25: IC50 could not determined and was assumed to be
above the highest concentration tested (25 ␮M).
14 were far less cytotoxic (IC50 = 6.07 and 5.71 ␮M, respec-
tively). On the contrary, almost all acridines showed strong
cytotoxicity after irradiation, with compounds 3, 5 and 51 dis-
playing the strongest photo-inducible effects (IC50s ranging
from 0.02 to 0.04 ␮M).
Results of the QSAR analysis are shown in Fig. 2. The
equation that best modeled the photo-inducible cytotoxicity
of acridine derivatives did not differ according to their tau-
tomeric form. This result established that the cytotoxicity mainly
depended on the capacity of the molecules to act as hydro-
gen donors (HD descriptor) and on the number of sets of
non-adjacent bonds (log Z). In contrast, two different equa-
tions modeled the photo-inducible mutagenicity of the acridines,
depending on their tautomeric forms. However, both equations
concerned the capacity of the molecules to act as hydrogen
donors (HD descriptor) or hydrogen acceptors (HA descrip-
tor), the molecular stability (GAP descriptor) and the molecular
topology (X and Wiener descriptors).
The effects of compounds 3, 5, 6 and 7 on human ker-
atinocytes are illustrated in Figs. 3 and 4 and summarized
in Table 2. Results observed with the comet assay showed
that both compounds induced single-strand breaks in the dark,
while only compounds 3 and 6 displayed photo-enhanced DNA-
damaging activity. Results observed in the micronucleus assay
indicated that irradiation with simulated sunlight produced a
non-significant increase of micronuclei in human keratinocytes,
since micronucleus levels in the control cultures were 16‰ in
the dark and 28‰ after photo-irradiation. Compounds 3, 5 and 6
inducedasignificantincreaseofmicronucleatedcellsinthedark,
whereas both compounds 3 and 5, but not 6, showed a significant
photo-enhancement of their clastogenic and/or aneugenic activ-
ity. Compounds 3 and 5 displayed the strongest photo-inducible
Table 2
Minimal genotoxic concentrations of the acridine derivatives 3, 5, 6 and 7 in the Salmonella mutagenicity test, the comet assay and the micronucleus assay
Compounds
Minimal mutagenic concentration
Minimal DNA-damaging
Minimal clastogenic concentration
Salmonella mutagenicity test (␮M)
concentration Comet assay (␮M)
micronuclei (␮M)
TA97a + S9 mix Without
photo-activation
TA98 − S9 mix With
Without
photo-activation
With
photo-activation
Without
photo-activation
With
photo-activation
photo-activation
3
5
6
7
337.27
404.88
0.26
1.24 × 10−2
2.13 × 10−3
3.23
2.75 × 10−2
1.06
0.22
0.70
0.02
0.44
0.36
0.09
0.81
3.84
2.06
0.07
NS
0.03
0.02
0.19
1.23
178.49
Acridine Orange
2.96
0.51
NS
0.30
0.05
0.009
Salmonella mutagenicity test: the minimal mutagenic concentrations were calculated from the regression models assuming that the doublings of the numbers of
spontaneous revertants were statistically significant responses for each tester strain (166 21 and 21 11 for TA97a + S9 mix and TA 98-S9 mix, respectively). Comet
assay: the minimal DNA-damaging concentrations were calculated from regression models assuming that an OTM χ2 of 2.4 0.11 was statistically significantly
different from the control value (2.11 0.09). Micronucleus assay: the minimal clastogenic concentrations were calculated from the regression models assuming
that 38 1‰ micronucleated cells was statistically significantly different from the control value (19 1).

110
C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
Fig. 3. Induction of single-strand breaks by the acridine derivatives 3, 5, 6 and 7 in keratinocytes using the comet assay without and with photo-activation, after a
2-h incubation period. The dotted line indicates an OTM χ2 = 2.4, significantly different from the control value (2.11 0.09).
clastogenic and/or aneugenic activities with MCD averaging
0.02–0.03 ␮M, which corresponds to a 100-fold increase of their
initial values.
This hypothesis was confirmed by data obtained with the
highly mutagenic N1-(2-chloro-thiazolo[5,4-a]acridin-11(6H)-
ylidene)-N,N-diethyl-1,3-propanediamine (compound 6) which
revealed a potent intercalative activity in the DNA major groove
[8]. Furthermore, the presence of the diethylamino-propylamine
side chain in position 11 of the acridine ring conferred to the
compound a chemical structure similar to that of the well-known
simple intercalator quinacrine [41], in which the aliphatic side
chain tended to bind non-covalently to complementary DNA
strands in the major groove [42]. In the dark, the necessity
of metabolic activation for a maximal mutagenic response in
the Salmonella mutagenicity test supposed that redox reac-
tions could potentiate the DNA-intercalation mechanism: the
metabolic fraction (S9 mix) could initiate electron-transfer
events and form reactive oxygen species capable of maintaining
the DNA of living cells in a continuous state of electron-transfer
reactions [24]. As a consequence, charge-transfer mechanisms
between DNA and small intercalative molecules could be
favored and could contribute to a stronger DNA-binding activity
[24].
4. Discussion
Acridines are typically considered as frameshift mutagens
through a reversible intercalation of the planar aromatic rings
within the macromolecules [19–21]. However, unlike simple
intercalators the reactive acridine derivatives have been shown
capable of forming covalent adducts with DNA [16,17].
Results observed in the present study demonstrate that
thiazolo[5,4-a]acridines are predominantly mutagenic in the
Salmonella tester strain TA97a after metabolic activation in
the dark. This strain includes the hisD6610 allele characterized
by an insertion of one cytosine in a run of six [21]. Such a
reverse mutation (−1 base frameshift), shown to be induced
by various mutagens including 2-aminoacridine, ICR191 and
dexon [40], indicates that thiazolo[5,4-a]acridines preferentially
interact with DNA through a simple intercalation mechanism.

C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
111
Fig. 4. Induction of micronuclei by the acridine derivatives 3, 5, 6 and 7 in keratinocytes without and with photo-activation.

112
C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
After photo-activation, acridine derivatives were directly
attachment of the thiazolo[5,4-a]acridines 3, 5 and 7 to DNA
could occur in the dark, according to a mode of action already
described for antitumor acridines such as ledacrine [52,53].
Results observed after irradiation also confirmed the large dif-
ferences between mammalian cells and bacteria. Compound
5, shown to exert a strong photo-enhanced mutagenicity in
Salmonella typhimurium, exhibited a decreased DNA-damaging
activity, while compound 6 was deprived of mutagenicity in
prokaryotic cells, but displayed strong photo-inducible DNA-
damaging properties.
mutagenic in strain TA98. Thisstrain carriesthe hisD3052allele,
a −1 frameshift mutation which affects the reading frame of a
repetitive (C-G)4 sequence. Reversion of hisD3052 back to the
wild type is possible through a one-base insertion [21]. Such
a mutation is induced by various frameshift mutagens such as
nitroaromatics, aromatic amines [43] and polycyclic aromatic
hydrocarbons [44], which bind covalently to DNA and gener-
ate bulky adducts. These results are consistent with previously
published data, which suggested that the photolytic enhance-
ment of acridine-induced mutagenesis in Salmonella appeared
to be due to repairable lesions following covalent attachment to
DNA [45,46]. Moreover, previous results indicated that irradia-
tion could generate the formation of acridinium salts expected to
be strong electron acceptors [47] and capable of forming hydro-
gen bonds. This hypothesis was supported by the QSAR analysis
since the equations that best described the photo-inducible muta-
genicity of acridines involved electronic descriptors such as HA
(Hydrogen Acceptor index) and HB (Hydrogen Donor index).
Moreover, it is likely that forms of the set II could have different
behaviors than those of set I, especially when hydrogen bonding
was involved. For the tautomeric forms, the electronic behavior
of the substituted-groups in the thiazole-ring could be modified
possibly to favor the formation of reactive species capable to
generate DNA adducts. Three derivatives (3, 5 and 7) revealed
high photo-enhanced mutagenic activities in TA98. Compounds
5 and 7, bearing a butylsulfanyl-group and a sulfur in position
2 of the thiazole-ring, respectively, were highly susceptible of
forming DNA adducts after S-protonation. Compound 3, bear-
ing a chloro-group in position 2 of the thiazole-ring could react
in the same way as acridine mustards by spontaneous loss of the
chloro-group after aromatic nucleophilic substitution and for-
mation of positively charged intermediates [15]. Unexpectedly,
these three compounds also exhibited a positive result in the
DNA methyl–green complex assay, indicating that a concomi-
tant non-covalent DNA-intercalation could occur. It confirmed
that an additional intercalation of acridines into DNA could be
due to oxidative mechanisms and suggested that the boundary
between reversible intercalation and DNA-adduct formationwas
not sharp for these compounds [15].
In mammalian cells, acridine derivatives have been shown
to induce single-strand breaks [48], sister-chromatid exchange
[49] and chromosome rearrangements [50]. Results observed in
the comet assay are in line with these data. Compounds 3, 5
and 7, poorly active in Salmonella typhimurium TA97a, exhib-
ited a significant DNA-damaging activity in the dark. On the
other hand, the DNA-intercalating compound 6, demonstrat-
ing a high mutagenicity in Salmonella typhimurium TA97a,
did not show a potent DNA-damaging activity in human ker-
atinocytes. This result is consistent with the genotoxic profile
of quinacrine, which exhibited a decreased mutagenic activ-
ity in mammalian cells compared with that in bacteria [42].
This could be explained by the capacity of mammalian cells to
transform quinacrine, and probably its analogues, into the less
active mono-desethyl quinacrine intermediate [51]. As a con-
sequence, the production of reactive metabolites by enzymatic
oxidation could be hypothesized, suggesting that a covalent
The consequences of DNA-damage induced by thiazolo[5,4-
a]acridines on chromosomes have been evaluated by the scoring
of micronucleated keratinocytes after a 48-h incubation period.
Micronuclei can be defined as acentric chromosome fragments
or whole chromosomes clearly visible in the cytoplasm of
interphase cells. They originate from chromosome rearrange-
ments or non-disjunctions induced by clastogenic or aneugenic
agents, respectively [54]. Results observed in the present study
confirmed that irradiation with simulated sunlight induced
micronuclei in human keratinocytes and demonstrated that some
thiazolo[5,4-a]acridines induced photo-inducible micronuclei.
Compound 6, characterized by a slight DNA-damaging activ-
ity, appeared as the most potent clastogenic and/or aneugenic
derivative in the dark. In contrast, after irradiation the activity
of compound 6 was lower while that of compounds 3, 5 and 7
greatly increased. These results suggest that micronuclei could
originate through various mechanisms. The first one comprises
direct DNA-interaction that leads to the formation of double-
strand breaks inaccurately repaired [54,55]. This mechanism
could reasonably be envisaged for most of the thiazolo[5,4-
a]acridines derivatives. In this case, the clastogenic potential
of the compounds depended on both the type and the num-
ber of lesions in the DNA and the capacity of the cells to
correctlyrepairDNA-damage. Afterirradiation, additionalchro-
mosome rearrangements could appear due to the excitation of
endogenous photosensitizers, leading to oxidative DNA mod-
ifications [24,56] and to the formation of reactive acridine
intermediates. The second mechanism assumes the possible
interaction of acridines with enzymes involved in the regulation
of conformational changes in DNA [57]. Among these, DNA
topoisomerases I and II have been shown to play an important
role in the relaxation of DNA supercoils generated by cellular
processes such as transcription, recombination and replication,
and the topoisomerases appeared essential for the condensa-
tion of chromosomes and their segregation during mitosis [57].
The stabilization of covalent DNA-topoisomerase complexes
and the inhibition of their activity by exogenous agents have
been identified as different sources of double-strand breaks in
mammalian DNA [48]. Compound 6, which demonstrated a
slight DNA-damaging activity and a strong clastogenic effect,
may induce chromosome rearrangements by inhibiting topoiso-
merase activity according to the same mechanism as proposed
for its anticancer analogue quinacrine [57]. This mechanism of
action, coupled with the possible interaction of acridines with
various other cellular pathways, could be at the origin of the
strong antiproliferative activity observed in the micronucleus
assay.

C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
113
As a consequence of their interaction with DNA, acridine
compounds have demonstrated strong anticancer properties [1].
Structure–activity studies on diverse classes of acridines have
shown that intercalation into DNA leading to the formation of
ternary acridine/DNA/topoisomerase enzyme complexes, also
named cleavable complexes, was a requirement for a maximal
anticancer property [57]. Unfortunately, this pathway generated
genotoxic events and increased the risk for secondary malig-
nancies following anticancer chemotherapy. Results observed
in the present study revealed that only compounds 6 and 13 dis-
played a significant cytotoxic activity in human monocytes in the
dark. Concerning compound 6, results obtained with Salmonella
typhimurium and human keratinocytes suggested that genotoxic
and cytotoxic activities were generated by the same mechanism
as proposed for the anticancer drug quinacrine, namely inter-
calation between DNA strands and formation of topoisomerase
II cleavable complexes [57]. However, results observed with
compound 13 clearly indicate that the cytotoxicity of acridines
could involve mechanisms other than DNA-interactions. After
irradiation by simulated sunlight, almost all the thiazolo[5,4-
a]acridines displayed strong cytotoxic activity. Since it has been
demonstrated that the formation of DNA adducts could more
efficiently inhibit DNA replication and transcription than inter-
calation [45,46], one could suppose that the covalent attachment
of reactive acridine intermediates to DNA was at the origin
of cell apoptosis. Nevertheless, the absence of a correlation
between cytotoxic and mutagenic activities implied that other
targets could be involved in the interaction between acridines
and mammalian cells, depending on their lipophilicity, their
degree of ionization and their capacity to migrate to different
organelles. This was illustrated by the strong antiproliferative
activity observed in CHO cells during the assessment of binu-
cleated cell counts and by the presence of high levels of apoptotic
cells.
[6] N.S. Burres, S. Sazesh, G.P. Gunawardana, J.J. Clement, Antitumor activity
and nucleic acid binding properties of dercitin, a new acridine alkaloid
isolated from a marine Dercitus species sponge, Cancer Res. 49 (1989)
5267–5274.
[7] I.B. Taraporewala, J.W. Cessac, T.C. Chanh, A.V. Delgado, R.F. Schi-
nazi, HIV-1 neutralization and tumor cell proliferation inhibition in vitro
by simplified analogues of pyrido[4,3,2-mn]thiazolo[5,4-b]acridine marine
alkaloids, J. Med. Chem. 35 (1992) 2744–2752.
[8] A. Rajendran, A. Nair, BU Unprecedented dual binding behaviour of acri-
dine group of dye: a combined experimental and theoretical investigation
for the development of anticancer chemotherapeutic agents, Biochim. Bio-
phys. Acta 1760 (12) (2006) 1794–1801.
[9] W.A. Denny, B.C. Baguley, Dual topoisomerase I/II inhibitors in cancer
therapy, Curr. Top Med. Chem. 3 (2003) 339–353.
[10] L.R. Ferguson, W.A. Denny, Genotoxicity of non-covalent interactions:
DNA intercalators, Mutat. Res. (2007) 6 (Epub ahead of print).
[11] L.R. Ferguson, W.A. Denny, The genetic toxicology of acridines, Mutat.
Res. 258 (1991) 123–160.
[12] A.M. Cancel, T. Smith, U. Rehkemper, J.E. Dillberger, D. Sokal, M.
McClain, A one-year neonatal mouse carcinogenesis study of quinacrine
dihydrochloride, Int. J. Toxicol. 25 (2006) 109–118.
[13] M.D. Brown, L.S. Ripley, D.H. Hall, A proflavin-induced frameshift
hotspot in the thymidylate synthase gene of bacteriophage T4, Mutation
Res. 286 (1993) 189–197.
[14] R. Narayanan, P. Tiwari, D. Inoa, B.T. Ashok, Comparative analysis of
mutagenic potency of 1-nitro-acridine derivatives, Life Sci. 77 (2005)
2312–2323.
[15] G.R. Hoffmann, M.A. Calciano, B.M. Lawless, K.M. Mahoney, Frameshift
mutations induced by three classes of acridines in the lacZ reversion assay
in Escherichia coli: potency of responses and relationship to slipped mis-
pairing models, Environ. Mol. Mutagen. 42 (2003) 111–121.
[16] L.R. Ferguson, P.M. Turner, W.A. Denny, The mutagenic spectrum of
acridine-linked aniline nitrogen mustards in AS52 cells: implications of
DNA targeting with high selectivity for adenine or guanine bases, Mutat.
Res. 469 (2000) 115–126.
[17] B.R. Brown, W.J. Firth, L.W. Yielding, Acridine structure correlated with
mutagenic activity in Salmonella, Mutat. Res. 72 (1980) 373–388.
[18] H. Tomosaka, S. Omata, E. Hasegawa, K. Anzai, The effects of sub-
stituents introduced into 9-aminoacridine on frameshift mutagenicity and
DNA binding affinity, Biosci. Biotechnol. Biochem. 61 (1997) 1121–1125.
[19] L.R. Ferguson, D.G. MacPhee, Frameshift mutagenesis by 9-
aminoacridine, ICR191, AMSA and related experimental antitumour
acridines in recA+ and recA1 strains of Salmonella typhimurium, Mutat.
Res. 116 (1983) 289–296.
Overall, the results of this study confirmed that acridines
were predominantly genotoxic by a DNA-intercalating mech-
anism in the dark while DNA-adducts were probably induced
following photo-activation. These photo-inducible properties
demonstrated the necessity to consider the mechanisms of pho-
toactivation when the pharmacological activities of acridines are
evaluated.
[20] L.R. Ferguson, W.A. Denny, Frameshift mutagenesis by acridines and other
reversibly-binding DNA ligands, Mutagenesis 5 (1990) 529–540.
[21] D.M. De Marini, M.L. Shelton, A. Abu-Shakra, A. Szakmary, J.G. Levine,
Spectra of spontaneous frameshift mutations at the hisD3052 allele of
Salmonella typhimurium in four repair backgrounds, Genetics 149 (1998)
17–36.
[22] W.D. Chen, J.R. Eshleman, M.R. Aminoshariae, A.H. Ma, N. Veloso, S.D.
Markowitz, W.D. Sedwick, M.L. Veigl, Cytotoxicity and mutagenicity
of frameshift-inducing agent ICR191 in mismatch repair-deficient colon
cancer cells, J. Natl. Cancer Inst. 92 (2000) 440–441.
[23] M. Tanaka, K. Csuri, J. Molnar, N. Motohashi, Diverse mutagenicity of
methylbenz(c)acridines in the direct Ames’ Salmonella mutagenicity test,
Anticancer Res. 16 (1996) 2837–2841.
References
[1] W.A. Denny, Acridine derivatives as chemotherapeutic agents, Curr. Med.
Chem. 9 (2002) 1655–1665.
[2] M. Wainwright, Acridine-a neglected antibacterial chromophore, J.
Antimicrob. Chemother. 47 (2001) 1–13.
[24] B. Baguley, J. Wakelin, J.D. Jacintho, P. Kovacic, Mechanisms of action of
DNA intercalating acridine-based drugs: how important are contributions
from electron transfer and oxidative stress? Curr. Med. Chem. 10 (2003)
2643–2649.
[25] P. Yang, Q. Yang, X. Qian, L. Tong, X. Li, Isoquino[4,5-bc]acridines:
design, synthesis and evaluation of DNA binding, anti-tumor and DNA
photo-damaging ability, J. Photochem. Photobiol. B 84 (2006) 221–
226.
[3] P. Belmont, J. Bosson, T. Godet, M. Tiano, Acridine and acridone deriva-
tives, anticancer properties and synthetic methods: where are we now?
Anticancer Agents Med. Chem. 7 (2007) 139–169.
[4] J.B. Bongui, A. Elomri, D. Cahard, F. Tillequin, B. Pfeiffer, A. Pierre, E.
Seguin, Synthesis and cytotoxic activity of acronycine analogues in the
benzo[c]pyrano[3,2-h]acridin-7-one and naphtho[1,2-b][1,7] and [1,10]-
phenanthrolin-7(14H)-one series, Chem. Pharm. Bull. (Tokyo) 53 (2005)
1540–1546.
[26] N.S. Burres, A. Frigo, R.R. Rasmussen, J.B. McAlpine, A colorimetric
microassay for the detection of agents that interact with DNA, J. Nat. Prod.
55 (1992) 1582–1587.
[5] E. Seguin, F. Tillequin, Structure activity relationships and mechanism of
action of antitumor benzo[b]acronycine antitumor agents, Ann. Pharm. Fr.
63 (2005) 44–52.

114
C. Di Giorgio et al. / Mutation Research 650 (2008) 104–114
[27] D.M. Maron, B.N. Ames, RevisedmethodsfortheSalmonellamutagenicity
test, Mutat. Res. 113 (1983) 173–215.
[28] L. Decome, M.P. De Meo, A. Geffard, O. Doucet, G. Dumenil, A. Botta,
Evaluation of photolyase repair activity in human keratinocytes after a
single dose ultraviolet B irradiation using the comet assay, J. Photochem.
Photobiol. B 79 (2005) 101–108.
[29] M. Kirsch-Volders, T. Sofuni, M. Aardema, S. Albertini, D. Eastmond, M.
Fenech, M. Ishidate Jr., S. Kirchner, E. Lorge, T. Morita, H. Norppa, J. Sur-
ralles, A. Vanhauwaert, A. Wakata, Report from the in vitro micronucleus
assay working group, Mutat. Res. 540 (2003) 153–163, Erratum in: Mut.
Res. 564 (2004) 97–100.
[30] M. Robin, S. Mialhe, V. Pique, R. Faure, J.-P. Galy, Synthesis
of two novel classes of tetracycles bearing tetrahydro ring system
from benzothiazole: 7,8,9,10-tetrahydro[1,3]thiazolo[5,4-a]acridine and
1,2,3,4-tetrahydro-12H-[1,3]benzothiazole [2,3-b] quinazolin-12-one, J.
Heterocycl. Chem. 39 (2002) 295–298.
[42] J.J. Clarke, D.C. Sokal, A.M. Cancel, D.B. Campen, R. Gudi, V.O. Wagner,
R.H. San, J.D. Jacobson-Kram, Re-eveluation of the mutagenic poten-
tial of quinacrine dihydrochloride dehydrate, Mutat. Res. 494 (2001) 41–
53.
[43] R.J. Turesky, Heterocyclic aromatic amine metabolism, DNA adduct for-
mation, mutagenesis, and carcinogenesis, Drug Metab. Rev. 34 (2002)
625–650.
[44] W. Xue, D. Warshawsky, Metabolic activation of polycyclic and hetero-
cyclic aromatic hydrocarbons and DNA damage: a review, Toxicol. Appl.
Pharmacol. 206 (2005) 73–93.
[45] W.J. Firth, S.G. Rock, B.R. Brown, L.W. Yielding, Azido analogs of
acridine: photoaffinity probes for frameshift mutagenesis in Salmonella
typhimurium, Mutat. Res. 81 (1981) 295–309.
[46] R.D. Snyder, M. Arnone, Putative identification of functional interactions
between DNA intercalating agents and topoisomerase II using the V79 in
vitro micronucleus assay, Mutat. Res. 503 (2002) 21–35.
[31] M.P. De Meo, M. Laget, C. Di Giorgio, H. Guiraud, A. Botta, M. Casteg-
naro, G. Dumenil, Optimization of the Salmonella/mammalian microsome
assay for urine mutagenesis by experimental designs, Mutat. Res. 340
(1996) 51–65.
[32] M. Davoren, E. Herzog, A. Casey, B. Cottineau, G. Chambers, H.J. Byrne,
F.M. Lyng, In vitro toxicity evaluation of single walled carbon nanotubes
on human A549 lung cells, Toxicol. In Vitro 21 (2007) 438–448.
[33] M. D’Errico, M. Teson, A. Calcagnile, L. Proietti De Santis, O. Nikaido,
E. Botta, G. Zambruno, M. Stefanini, E. Dogliotti, Apoptosis and efficient
repair of DNA damage protect human keratinocytes against UVB, Cell
Death Differ. 10 (2003) 754–756.
[34] R.P. Phillipson, S.E. Tobi, J.A. Morris, T.J. McMillan, UV-A induces per-
sistent genomic instability in human keratinocytes through an oxidative
stress mechanism, Free Radic. Biol. Med. 32 (2002) 474–480.
[35] R.D. Curren, G.C. Mun, D.P. Gibson, M.J. Aardema, Development of a
method for assessing micronucleus induction in a 3D human skin model
(EpiDerm), Mutat. Res. 607 (2006) 192–204.
[47] C. Bohne, K. Faulhaber, B. Giese, A. Hafner, A. Hofmann, H. Ihmels, A.K.
Kohler, S. Pera, F. Schneider, M.A. Sheepwash, Studies on the mechanism
of the photo-induced DNA damage in the presence of acridizinium salts-
involvement of singlet oxygen and an unusual source for hydroxyl radicals,
J. Am. Chem. Soc. 127 (2005) 76–85.
[48] G. Boos, H. Stopper, Genotoxicity of several clinically used topoisomerase
II inhibitors, Toxicol. Lett. 116 (2000) 7–16.
[49] G. Ribas, N. Xamena, A. Creus, R. Marcos, Sister-chromatid exchanges
(SCE) induction by inhibitors of DNA topoisomerases in cultured human
lymphocytes, Mutat. Res. 368 (1996) 205–211.
[50] E.M. Bomhard, High-dose clastogenic activity of aniline in the rat bone
marrow and its relationship to the carcinogenicity in the spleen of rats,
Arch. Toxicol. 77 (2003) 291–297.
[51] Y. Huang, H. Okochi, B.C. May, G. Legname, S.B. Prusiner, L.Z. Benet,
B.J. Guglielmo, E.T. Lin, Quinacrine is mainly metabolized to mono-
desethyl quinacrine by CYP3A4/5 and its brain accumulation is limited
by P-glycoprotein, Drug Metab. Dispos. 34 (2006) 1136–1144.
[52] K. Gorlewska, Z. Mazerska, P. Sowinski, J. Konopa, Products of metabolic
activation of the antitumor drug ledakrin (nitracrine) in vitro, Chem. Res.
Toxicol. 14 (2001) 1–10.
[53] Z. Mazerska, Similarity between enzymatic and electrochemical oxidation
of 2-hydroxyacridinone, the reference compound of antitumor mida-
zoacridinones, Acta Biochim. Pol. 50 (2003) 515–525.
[54] P. Pfeiffer, W. Goedecke, G. Obe, Mechanisms of DNA double-strand break
repair and their potential to induce chromosomal aberrations, Mutagenesis
15 (2000) 289–302.
[36] M. Fenech, The in vitro micronucleus technique, Mutat. Res. 455 (2000)
81–95.
[37] D. Anderson, D.P. Lovell, R. Albanese, G.E. Amphlett, G. Clare, R. Fer-
gusson, M. Richold, D.G. Papworth, J.R.K. Savage, Statistical analysis of
in vivo cytogenetic assays, in: D.J. Kirkland (Ed.), Statistical Evaluation
of Mutagenicity Test Data, 1989, Cambridge University press, Cambridge,
1989, pp. 184–232.
[38] D. Rogers, A.J. Hopfinger, Application of genetic function approximation
to quantitative structure activity relationships and quantitative structure
property relationships, J. Chem. Inf. Comp. Sci. 34 (1994) 854–866.
[39] R. Todeschini, V. Consonni (Eds.), Handbook of Molecular Descriptors,
Wiley-VCH, Weinheim, Germany, 2000, p. 667.
[40] D.E. Levin, B.N. Ames, Classifying mutagens as to their specificity in
causing the six possible transitions and transversions: a simple analysis
using the Salmonella mutagenicity assay, Environ. Mutagen. 8 (1986) 9–28.
[41] M. Aslanoglu, G. Ayne, Voltametric studies of the interaction of quinacrine
with DNA, Anal. Bioanal. Chem. 380 (2004) 658–663.
[55] M. Van Den Bosch, P. Lohman, A. Pastink, DNA double-strand break repair
by homologous recombination, Biol. Chem. 383 (2002) 873–892.
[56] J. Blasiak, E. Gloc, J. Drzewoski, K. Wozniak, M. Zadrozny, T. Skorski,
T. Pertynski, Free radical scavengers can differentially modulate the geno-
toxicity of amsacrine in normal and cancer cells, Mutat. Res. 535 (2003)
25–34.
[57] Z. Topcu, DNA topoisomerases as target for anticancer drugs, J. Clin.
Pharm. Ther. 26 (2001) 405–416.