Genotoxic, cytostatic, antineoplastic and apoptotic effects of newly synthesized antitumour steroidal esters

Mutation Research 675 (2009) 51–59

Contents lists available at ScienceDirect

Mutation Research/Genetic Toxicology and Environmental Mutagenesis journal homepage: www.elsevier.com/locate/gentox Community address: www.elsevier.com/locate/mutres

Genotoxic, cytostatic, antineoplastic and apoptotic effects of newly synthesized antitumour steroidal esters I. Karapidaki a,b , A. Bakopoulou a , A. Papageorgiou c , Z. Iakovidou a , E. Mioglou a , S. Nikolaropoulos d , D. Mourelatos a,∗ , T. Lialiaris b a

Laboratory of Biology and Genetics, Medical School, Aristotle University of Thessaloniki, Thessaloniki, 54124 Greece Laboratory of Genetics, Medical School, Democritus University of Thrace, Alexandroupolis, Greece Laboratory of Experimental Chemotherapy, Theagenion Anticancer Institute, Thessaloniki, Greece d Laboratory of Pharmaceutical Chemistry, Department of Pharmacy, University of Patras, Patra, Greece b c

a r t i c l e

i n f o

Article history: Received 30 September 2008 Received in revised form 22 January 2009 Accepted 9 February 2009 Available online 3 March 2009 Keywords: Modified steroidal esters of chlorambucil Sister-chromatid exchange (SCE) Proliferation rate index (PRI) Antileukemic activity Caspase-2 and caspase-3 activity

a b s t r a c t In this study, we have investigated the genotoxic, cytostatic, antineoplastic and apoptotic effects of three newly synthesized modified steroidal esters, having as alkylating agent p-N,N-bis(2-chloroethyl) aminophenyl butyrate (CHL) or p-N,N-bis(2-chloroethyl) aminophenyl acetate (PHE) esterified with the steroidal nucleus modified in the B- and D-ring. The genotoxic and cytotoxic effects of the compounds were investigated both in vitro, in lymphocyte cultures obtained from blood samples of healthy donors and in vivo, in ascites cells of P388 leukemia obtained from the peritoneal cavity of DBA/2 mice. Preparations were scored for sister-chromatid exchange (SCE) and proliferation-rate indices (PRI). The newly synthesized compounds were also studied for antineoplastic activity against lymphocytic P388 and lymphoid L1210 leukemias in mice, by calculating the mean of the median survival of the drug-treated animals (T) versus the untreated control (C) (T/C%). The activity of caspase-2 and caspase-3, indicators of apoptosis, was assessed biochemically in primary cultures of human lymphocytes. Our results show that the newly synthesized compounds caused severe genotoxic effects by significantly increasing the frequency of SCE and decreasing the PRI values in cultures of peripheral lymphocytes in vitro and in ascites cells of lymphocytic P388 leukemia in vivo. A significant correlation was also observed in both the in vitro and in vivo experiments: the higher the SCE frequency the lower the PRI value (r = −0.65, P < 0.001 and r = −0.99, P < 0.01, respectively). The measured antileukemic potency was statistically increased by all test compounds in both types of tumours, while the activity of caspase-2 and caspase-3 showed a statistically significant increase after two periods of exposure. The genotoxic (increase of SCE), cytostatic/cytotoxic (decrease of PRI) and antileukemic effects (increase of T/C%) in combination with the induction of apoptosis (activation of caspase-2 and caspase-3) caused by the newly synthesized compounds, lead us to propose them as agents with potentially antineoplastic properties. © 2009 Elsevier B.V. All rights reserved.

1. Introduction DNA-damaging agents that form cross-links, including nitrogen mustards, remain some of the most effective chemotherapeutics for the treatment of a variety of neoplasms [1]. Nitrogen mustards have played an important role in cancer chemotherapy for more than 50 years [2]. The most popular alkylating agents of nitrogen mustards with extended clinical application are chlorambucil and melphalan [3,4]. They express their antitumour activity by their interaction with the DNA molecule [5]. These agents may form adducts at all O- and N-atoms of nucleobases, as well as at O-atoms of phospho-

∗ Corresponding author. Tel.: +30 2310 999017; fax: +30 2310 999019. E-mail address: [email protected] (D. Mourelatos). 1383-5718/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.mrgentox.2009.02.010

diesters. The alkylation pattern depends on the agents used as well as the binding position in the DNA or RNA molecule, and seems to be different between single- and double-stranded nucleic acids [6]. Base alkylations have been found to be both genotoxic and cytotoxic. More specifically, O-alkylations are highly mutagenic and genotoxic, whereas N-alkylations are cytotoxic but relatively less mutagenic [6]. Only a small fraction of these alkylations forms effective cross-links due to their high inherent chemical activity, which results in their covalent binding to the nucleophilic sites of other biomolecules [7]. The increased toxicity and the lack of selectivity, which is due to their high inherent chemical reactivity, are also responsible for adverse effects observed by these agents, such as secondary neoplasms, anemia, cardiotoxicity, etc. [3,4]. Therefore, the main research goal in improving those agents has been to minimize the

52

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

above-mentioned undesirable effects [8]. One important means for improving drug efficacy while reducing toxicity is the chemical linkage of nitrogen mustards to carrier molecules (e.g. hormones) that have affinity for specific binding sites (hormone receptors) [9–11]. Steroidal hormones have been reported to influence the growth of many cancers and the presence of tumour-associated receptors for these hormones offers the opportunity for drug targeting through the use of drug-hormone conjugates [12]. Hormones can guide nitrogen mustards directly to the cell nucleus, where they naturally act as transcription factors via binding to their receptors [13]. In a previous study, we have shown that minor functional changes in the steroidal B-ring (e.g. 7-keto-group) of a steroidal ester of nitrogen mustard (androstan) had considerable effects on the molecule’s antileukemic, genotoxic and cytotoxic activity, leading us to assume that this modification is fundamental to the design of more effective molecules [14,15]. Therefore, in this study we extended those observations by synthesizing three new potentially chemotherapeutic compounds. These compounds contain p-N,N-bis(2-chloroethyl) aminophenyl butyrate (CHL) or p-N,N-bis(2-chloroethyl) aminophenyl acetate (PHE), as an alkylating agent and are conjugated with a steroidal ester modified in the steroidal B- and D-rings. These agents were comparatively studied on a molar basis regarding their ability to induce sister-chromatid exchange (SCE) and to disturb cell kinetics (proliferation rate index (PRI)) in normal human lymphocyte cultured in vitro and in ascites cells of lymphocytic P388 leukemia in mice in vivo. The SCE assay has a high predictive value as a clinical assay and has been widely used to predict both the sensitivity of human tumour cells to potential chemotherapeutics as well as the heterogeneity of drug sensitivity within individual tumours [16–20]. In addition, the newly synthesized compounds were tested for their antineoplastic potency in two well-established test systems of mice with lymphocytic P388 and lymphoid L1210 leukemias [21]. Finally, primary cultures of lymphocytes exposed to these agents were studied for possible activation of caspase-2 and caspase-3, in order to assess whether the genotoxic and cytostatic effects caused by these agents were correlated with the induction of programmed cell death (apoptosis). 2. Materials and methods 2.1. Newly synthesized compounds and concentrations tested All newly synthesized compounds (Fig. 1) carry a keto group in the B-ring of the steroidal nucleus. More specifically, compound 1 consists of CHL as alkylating agent in conjunction with a steroidal nucleus, with the insertion of a second keto function at position 17 of the D-ring. Compound 2 has PHE as alkylating agent in conjunction with a steroidal nucleus that carries a lactam group (–NHCO–) at the D ring (endocyclically). Finally, compound 3 has the same alkylating agent as compound 2, while the steroidal nucleus has the NHCOCH3 group (modified–NHCO– group) at position 17 of the D-ring (exocyclically). The synthetic procedure to obtain the steroidal ester derivatives is described in a previous paper of our group [22]. These compounds were tested in human primary lymphocytes cultured in vitro and in ascites cells of mice carrying lymphocytic P388 leukemia in vivo.

ning of the culture. During the experiment all cultures were maintained in the dark to minimize photolysis of the BrdU. The cultures were incubated for 72 h at 37 ◦ C and 0.3 ␮g/mL colchicine (CAS 64-86-8, C9754, Sigma, St. Louis, USA) was added 2 h before the end of the culture period. The cells were then collected by centrifugation. Air-dried preparations were stained by the Fluorescence-plus Giemsa (FPG) procedure [23] and scored for cells undergoing metaphase of the first mitosis (with both chromatids staining dark), the second mitosis (with one chromatid of each chromosome staining dark) and the third and/or subsequent mitoses (a proportion of chromosomes with both chromatids staining light). From each culture, al least 30 properly spread metaphases of cells in second division were blindly scored for SCE. The cultures were set up in triplicate. 2.4. Assessment of the proliferation rate index (PRI) The PRI was calculated according to the formula PRI = (M1 + 2M2 + 3M3+ )/N, where “M1 , M2 , M3+ ” indicate the number of metaphases corresponding to first, second, and third or subsequent divisions, respectively, and “N” is the total number of metaphases scored (at least 100) for each culture, in in vitro experiments and for each treatment for the in vivo evaluation. 2.5. In vivo SCE assay Stock solutions of the compounds used in this study were prepared immediately before use. DBA/2 mice (both sexes, weighing 20–23 g, 6–9 weeks old) were implanted with 106 ascites cells of lymphocytic P388 leukemia derived from stock tumour-bearing DBA/2 mice [21]. The mice were kept under conditions of constant temperature and humidity in sterile cages with water and food ad libitum. The newly synthesized compounds were introduced at a concentration of 0.6 ␮mol/g body weight (0.6 ␮mol/g bw) on the fourth day after tumour implantation. One hour later animals were given intraperitoneal (ip) injections of 1 mL of BrdU (Sigma)–activated charcoal (CAS 7440-44-0, 102184, E. Merck, Darmstadt, F.R.G.) suspension at a concentration of 1 mg/g body weight. Animals were sacrificed by ether narcosis at 48 h, after a treatment with 0.1 mL of 1 mg/mL colchicine (CAS 64-86-8, C9754, Sigma, St. Louis, USA) at 46 h, i.e. two hours prior to harvesting ascites cells. The cells were washed once with a hypotonic solution of 0.075 M KCl, then treated with the same solution for 25 min, fixed with methanol/glacial acetic acid (Sigma) (3:1) and airdried. The cells were then processed according to the FPG procedure, as described for the in vitro SCE assay [18]. At least 30 properly spread second division metaphases of P388 leukemia ascites cells recovered from each treatment were blindly scored for SCE. Three mice were used for the control and for each drug treatment group. 2.6. In vivo experiments on acute toxicity and antitumour effects 2.6.1. Compounds for i.p. treatment Stock solutions were prepared immediately before use. Compounds were suspended in corn oil at the desired concentration (Table 3), following initial dissolution in 5% dimethyl sulphoxide (DMSO) (CAS 67-68-5, D2650, Sigma, St. Louis, USA). This concentration of DMSO by itself produced no observable toxic effects. 2.6.2. Tumours Lymphoid L1210 and lymphocytic P388 leukemia were maintained as ascites in DBA/2 mice by injection of 105 and 106 tumour cells, respectively, at seven day intervals, into the peritoneal cavity. The tumour was obtained from the experimental laboratory of the Theagenion anti-cancer hospital of Thessaloniki. 2.6.3. Animals Ascites fluid from stock tumour-bearing DBA/2 mice was implanted in (C57BL × DBA/2) BDF1 mice. BDF1 mice of both sexes, 6–9 weeks old and weighing 20–23 g, were used for the antitumour evaluation. Male and female animals were distributed equally, three males and three females for each group of the three different compounds tested and five males and five females for the control groups. They were kept under conditions of constant temperature and humidity in sterile cages with water and food ad libitum.

2.2. Cell culture Primary cultures of lymphocytes were set up by adding 11 drops of heparinized whole blood derived from three normal subjects, to 5 mL of cell culture medium (RPMI 1640, F1213, Biochrom, Berlin, Germany) supplemented with 20% FCS (S0113, Biochrom), 0.63% L-glutamine (K0282, Biochrom), 0.63% penicillin/streptomycin (A2210, Biochrom) and 2% phytohaemagglutinin (M5030, Biochrom) as mitogen. The individuals who donated their blood (25–35 years old) were healthy, not taking any medication, non-smokers and no alcohol consumers. All donors got acquainted with the study and they willingly signed permission for using their blood samples for scientific purposes. 2.3. In vitro SCE assay For determination of SCE, 5 ␮g/mL of 5-bromodeoxyuridine (BrdU) (CAS 59-143, B9285, Sigma, St. Louis, USA) and solutions of the newly synthesized compounds at three different concentrations (0.2, 0.4 and 0.6 ␮M) were added at the begin-

2.6.4. Estimation of acute toxicity The degree of toxicity of the compounds was determined following a single i.p. injection into BALB/C mice. There were groups of 10 mice per dose at three different dosages. The mice were observed for 30 days and the therapeutic dose of the compounds was determined after graphical estimation of the LD50 and LD10 (30-days curves). The highest dose used for a single treatment was equal to the LD10 value. 2.6.5. Antitumour experiments These were initiated on day 0 by i.p. implantation of 106 and 105 ascites cells of lymphocytic P388 and lymphoid L1210 leukemia, respectively, in (DBA/2 × C57BL) BDF1 mice, according to the protocol of the National Cancer Institute, USA [21]. The drugs were introduced at LD10 /2, intermittently on days 1, 4 and 7 after tumour implantation. The antitumour effect was assessed from the oncostatic parameter T/C% which is the percentage increase in median survival time of drug-treated animals (T) versus the corn oil-treated controls (C).

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

53

Fig. 1. Newly synthesized compounds.

2.7. Assessment of caspase-2 and caspase-3 activity in primary lymphocyte cultures Normal human lymphocytes from blood samples of three healthy donors (different from those involved in the in vitro SCE assay) were used for the evaluation of caspase-2 and caspase-3 activity. Lymphocytes were isolated from whole blood by use of the Ficoll method (Biocoll Separating Solution, L6115, Biochrom, Berlin, Germany) [24]. Cells were cultured at a density of 1–3 × 105 cells/mL of medium (RPMI 1640, F1213, Biochrom, Berlin, Germany), supplemented with 20% FCS (S0113, Biochrom, Berlin, Germany), 0.63% l-glutamine (K0282, Biochrom, Berlin, Germany), 0.63% penicillin/streptomycin (A2210, Biochrom, Berlin, Germany) and 2% phytohaemagglutinin (M5030, Biochrom, Berlin, Germany) and kept at 37 ◦ C for 72 h. Only cells from cultures with viability >90%, as assessed by Trypan blue exclusion, (0.4%, CAS 72-57-1, 93595, Sigma–Aldrich, St. Louis, USA) were used for the experiments. The newly synthesized compounds were added at the beginning of the culture at

a concentration of 0.6 ␮M, which corresponded to the most effective concentration for SCE induction by these agents (Table 1). Mitomycin C (CAS 50-07-7, Kyowa Hakko Kogyo Co., Ltd., Tokyo, Japan), a well-known bifunctional alkylating agent, was used as a positive control at a concentration of 0.02 ␮M. 2.8. Colorimetric detection of caspase-3 and caspase-2 activation The activity of caspase-3, a key enzyme involved in programmed cell death, was measured in control and compound-treated human lymphocytes by use of a colorimetric assay (K106-100 kit, Bioline, Mountain View, USA). The induction of apoptosis caused by DNA damage was further confirmed by evaluating the activity of caspase2, also with a colorimetric assay (K117-100 kit, Biovision, Mountain View, USA). Lymphocyte cultures were exposed to the compounds for different time periods (12 or 24 hours), as described above. Cells were then washed with ice-cold PBS, lysed in buffer solution (K117-100 kit, Biovision, Mountain View, USA), and centrifuged

54

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

at 10,000 × g for 1 min. The supernatant was collected and subjected to Bradford protein assay (BioRad Laboratories GmbH, Muenchen, Germany) to determine the total protein content of each cell lysate. The caspase-3 and caspase-2 enzymatic reactions were carried out in 96-well plates. Briefly, estimation of caspase-3 activity of the lysates was based on the spectrophotometric detection at 405 nm of the chromophore pNA released after cleavage of the labeled substrate DEVD-pNA, in an ELISA plate reader (Stat Fax 2100 Automatic Micro Plate Reader). Purified human recombinant caspase-3 (RnD Systems 707-C3) at different concentrations was used to construct a standard curve for quantitative analysis of caspase-3 activity. Wells without cell lysate or without substrate served as additional controls. Measurement of caspase-2 activity of the lysates was based on the spectrophotometric detection at 405 nm of the chromophore pNA released after cleavage of the labeled substrate VDVAD-pNA [25,26]. Purified human recombinant caspase-2 (RnD Systems 702-C2) at different concentrations was used to construct a standard curve. 2.9. Statistical analysis For the statistical evaluation of the experimental data Student’s t-test was performed for the SCE frequencies, to determine whether any values deviated significantly from the controls (P < 0.05), whereas the 2 test was used for PRI comparisons to assess significant deviations compared with controls (P < 0.05). We also calculated the simple linear correlation between the values of the PRI index and the SCE frequencies. The formula for the Pearson product-moment correlation coefficient r was applied. Then a criterion for testing whether r differs significantly from zero was used, whose sampling distribution is Student’s t-test with n − 2 d.f. Statistical evaluation of the in vivo experimental data was made by the Wilcoxon test considering P values less than 0.05 as significant. For the caspase-3 and caspase-2 assays, statistical analysis of the data was performed with the t-test, considering P values less than 0.05 as significant.

3. Results 3.1. Enhancement of SCE frequency and reduction of PRI values in human lymphocyte cultures The cytogenetic effects of the test compounds in normal human lymphocyte cultures are shown in Table 1. The genotoxic and cytotoxic/cytostatic effects of the compounds are evaluated by calculating the SCE and PRI values, respectively. All newly synthesized compounds induced a statistically significant (P < 0.01) increase in SCE at all concentrations tested. Moreover, all the compounds were identified as being effective, on a molar basis, in causing cell-division delays. Compound 3 was found to have the most pronounced genotoxic activity by increasing the frequency of SCE up to 37.54 ± 2.06 SCEs/cell compared with a control value of 7.67 ± 0.48 SCE/cell (P < 0.01), while compounds 1 and 2 enhanced SCE to 21.30 ± 0.63 and 18.12 ± 0.54, respectively. Moreover, compounds 2 and 3 had also significant cytostatic effects by significantly reducing the PRI values (down to 2.19 for the 0.6 ␮M concentration, compared with control PRI = 2.67, Table 1 Induction of SCE and cell-division delays (PRI) by newly synthesized compounds in normal human lymphocyte cultures (in vitro).

Control Compound 1

Compound 2

Compound 3

Concentration (␮M)

SCE ± S.E.M.* /cell ± ± ± ± ± ± ±

PRI

0.2 0.4 0.6 0.2 0.4 0.6

7.67 18.3 20.03 21.30 19.88 24.11 18.12

0.48 0.72a 0.58a 0.63a 0.72a 1.02a 0.54a

2.67 2.23b 2.34b 2.28b 2.50 2.38b 2.19b

0.2 0.4 0.6

16.54 ± 0.68a 27.84 ± 2.12a 37.54 ± 2.06a

2.49 2.30b 2.19b

SCE have been correlated with corresponding PRI values (r = −0.65 and P < 0.01). * Results are shown as mean of SCE ± S.E.M. (standard error of mean). a Statistically significant increase in SCE compared to the control (P < 0.01, by ttest). b Statistically significant decrease in PRI compared to the control (P < 0.01, by 2 test).

Table 2 Induction of SCE and decrease of PRI by newly synthesized compounds in ascites cells of lymphocytic P388 leukemia (in vivo).

Control Compound 1 Compound 2 Compound 3

Concentration (␮mol/g b.w.)

SCE ± S.E.M.* /cell

0.6 0.6 0.6

9.05 12.85 17.85 18.93

± ± ± ±

PRI

0.35 0.49a 1.24a 1.01a

2.55 2.05b 1.54b 1.40b

SCE have been correlated with corresponding PRI values (r = −0.99 and P < 0.01). * Results are shown as mean of SCE ± S.E.M. a Statistically significant increase in SCE compared to the control (P < 0.01, by ttest). b Statistically significant decrease in PRI compared to the control (P < 0.01, by 2 test).

P < 0.01). Moreover, the decrease of the PRI values was found to be concentration-dependent for both compounds 2 and 3, whereas compound 1 caused almost the same PRI decrease for all concentrations tested. Additionally, the three test compounds showed a different pattern in their genotoxic potential. Compound 1 caused a statistically significant increase in SCE starting from the lowest concentration of 0.2 ␮M, but this increase reached a plateau at the higher concentrations of 0.4 and 0.6 ␮M. Compound 2 increased SCE at 0.2 and 0.4 ␮M in a concentration-dependent manner but its effects were reduced at the concentration of 0.6 ␮M. Finally, compound 3 displayed a clearly concentration-dependent pattern of genotoxicity, as SCE frequencies increased at increasing concentrations of this compound. Overall, a correlation was observed between the magnitude of the SCE response and the reduction in PRI values for all compounds tested (r = −0.65 and P < 0.01). 3.2. In vivo induction of SCE and reduction of PRI in ascites cells of lymphocytic P388 leukemia-bearing mice The cytogenetic effects of the test compounds in ascites cells of lymphocytic P388 leukemia-bearing mice are shown in Table 2. The same parameters, i.e., SCE and PRI values were also evaluated. We have injected 0.6 ␮mol/g bw of the compounds, as this concentration was found to be the most effective in increasing SCE (compounds 1 and 3) and in reducing PRI values (mainly for compounds 2 and 3) in the in vitro experiments. All newly synthesized compounds induced a statistically significant (P < 0.01) increase in SCE and a decrease in PRI values (P < 0.01). Compound 3 had the most pronounced genotoxic activity, increasing the frequency of SCE up to 18.93 ± 1.01 and also the most severe cytotostatic activity, reducing the PRI value down to 1.40 compared with 9.05 ± 0.35 and 2.55, respectively, in the controls. This finding is in agreement with the results obtained in the in vitro evaluation. The PRI value was also remarkably decreased (PRI = 1.54) in ascites cells exposed to compound 2. A significant correlation was observed between the magnitude of the increase in SCE and the reduction of the PRI (r = −0.99 and P < 0.01). The order of decreasing cytogenetic potency of the newly synthesized compounds in human Table 3 Toxicity of the compounds.

Compound 1 Compound 2 Compound 3

LD50 a (mg/kg)

LD10 (mg/kg)

112 52 48

30 14 12

a LD50 values were estimated graphically, where the percentage of deaths due to the toxicity of each dose is shown in the ordinate, while the administered doses are indicated on the abscissa on semilogarithmic paper. For chemotherapy testing the highest dose used for a single treatment was LD10 . Therefore the drugs in the survival experiments were compared at equitoxic doses.

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

55

Table 4 The effect of new compounds on median survival time administered on days 1, 4 and 7 after tumour implantation on leukemia P388-bearing mice (BDF1).

Control Compound 1 Compound 2 Compound 3 a b c

Treatment schedule

Treatment dose (mg/kg)

MST ± S.E.a

1,4,7 1,4,7 1,4,7

Corn oil 15 7 6

10.7 18.0 19.8 18.8

± ± ± ±

0.61 0.68 1.81 0.79

T/C%b

Cures

100 168c 185c 178c

0/6 0/6 0/6 0/6

Median survival time ± standard error of the mean. T/C% is the present increase median survival time of treated animals over the control. Significant increase (P < 0.01, by the Wilcoxon test) over the control.

Table 5 The effect of new compounds on median survival time administered on days 1, 4 and 7 after tumour implantation on leukemia L1210-bearing mice (BDF1).

Control Compound 1 Compound 2 Compound 3 a b c

Treatment schedule

Treatment dose (mg/kg)

MST ± S.E.a

1, 4, 7 1, 4, 7 1, 4, 7

Corn oil 15 7 6

9.30 12.5 14.1 14.7

± ± ± ±

0.52 0.84 0.98 1.21

T/C%b 100 134c 152c 158c

Cures 0/6 0/6 0/6 0/6

Median survival time ± standard error of the mean. T/C% is the present increase median survival time of treated animals over the control. Significant increase (P < 0.01, by the Wilcoxon test) over the control.

lymphocytes (in vitro) was: compound 3 > compound 2 > compound 1, while in ascites cells of lymphocytic P388 leukemia (in vivo) the order was as follows: compound 3 = compound 2 > compound 1. The increase in SCE frequencies was slightly less pronounced in ascites cells in vivo compared with that in vitro, whereas the decrease in PRI was much more pronounced in vivo than in the in vitro experiments, for the same compounds tested (compare Tables 1 and 2). 3.3. The newly synthesized compounds increase the survival time of mice bearing lymphocytic P388 and lymphoid L1210 leukemia The effective compounds were tested for antineoplastic activity in vivo against the transplantable, lymphocytic P388 (Table 4) and lymphoid L1210 leukemias (Table 5). In the antileukemic experiments, the therapeutic dose was calculated by evaluating the LD10 acute toxicity (D). Leukemia P388bearing mice (BDF1) were treated with the compounds at doses of (D/2) × 3 (Table 3) on days 1, 4 and 7 after tumour transplantation. Compounds 1, 2 and 3 caused a statistically significant increase (P < 0.01) of the survival time (Table 4) compared with the control, with no statistically significant difference between the three compounds. Compound 2 had the most pronounced antineoplastic activity, increasing the survival time of the mice up to 185%. Compounds 1 and 3 had less but still considerable antineoplastic activity, significantly increasing the survival time of the mice to 168% and 178%, respectively. Lymphoid L1210 leukemia-bearing mice (BDF1) received a dose of (D/2) × 3 of each test compound on days 1, 4 and 7 after tumour transplantation (Table 5). Compounds 1, 2 and 3 caused a statistically significant increase of the survival time (Table 5). Compounds 3 and 2 induced the most pronounced increase, upto 158% and 152%, respectively, while compound 1 increased the survival time of the mice only to 134%. Our findings are in agreement with the cytogenetic in vitro and in vivo results, which show that compounds 3 and 2 have the highest activity.

Diagram 1. Fold increase of caspase-3 in primary lymphocyte cultures derived from donor 1. Compounds and time-points tested compared with the relevant control.

sible activation of caspase-3, a key enzyme involved in apoptosis, as well as caspase-2, which is the most apical caspase in the DNA damage-induced apoptotic cascade [27]. Lymphocytes treated with each one of the three experimental compounds displayed a statistically significant activation of both caspase-2 and caspase-3, compared with untreated control cultures. However, there were major differences in the patterns of activation of these two apoptotic enzymes, as well as significant variations among primary lymphocyte cultures derived from different donors, among the three compounds and also between the two time intervals tested (12 and 24 h), (Table 6). More specifically, caspase-3 displayed a 1.5–4-fold increase compared with the control depending on the donor and the compound tested (Diagrams 1–3). In most cases, the activation seemed to be more pronounced at the 24 h time point than at the 12 h time point. Moreover, in most cases the fold increase of caspase-3 seemed to be more pronounced after treatment with compounds 2 and 3 than for compound 1. In contrast, caspase-2 displayed a completely dif-

3.4. The newly synthesized compounds activate caspase-2 and caspase-3 in cultured human lymphocytes In order to elucidate whether the strong decrease in PRI values caused by the test compounds was mainly due to cell-cycle delays as a result of DNA damage, or whether the induction of programmed cell death (apoptosis) was also involved, we investigated the pos-

Diagram 2. Fold increase of caspase-3 in primary lymphocyte cultures derived from donor 2. Compounds and time-points tested compared with the relevant control.

56

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

Table 6 Caspase-3 and caspase-2 activation by newly synthesized compounds and mitomycin C in normal human lymphocyte cultures. Results are shown as mean (±SD) (n = 3). Sample 1st donor Control Mitomycin C Compound 1 Compound 2 Compound 3 Control Mitomycin C Compound 1 Compound 2 Compound 3 2nd donor Control Mitomycin C Compound 1 Compound 2 Compound 3 Control Mitomycin C Compound 1 Compound 2 Compound 3 3rd donor Control Mitomycin C Compound 1 Compound 2 Compound 3 Control Mitomycin C Compound 1 Compound 2 Compound 3 * **

Time-point (h)

Ng of caspase-3 mg of total cytoplasmic protein

Time-point (h)

Ng of caspase-2 mg of total cytoplasmic protein

12 12 12 12 12 24 24 24 24 24

1.89 2.41 5.14 1.95 6.89 4.90 10.97 10.51 10.82 10.30

± ± ± ± ± ± ± ± ± ±

0.25 0.46 0.12* 0.20 0.19* 0.41 0.05* 0.12* 0.20* 0.19*

12 12 12 12 12 24 24 24 24 24

0.24 2.45 3.20 1.85 3.63 1.08 1.93 6.50 6.28 6.30

± ± ± ± ± ± ± ± ± ±

0.09 0.10* 0.04* 0.11* 0.11* 0.09 0.10 0.04* 0.11* 0.11*

12 12 12 12 12 24 24 24 24 24

1.29 2.44 1.90 2.40 2.60 0.44 1.26 0.84 1.17 0.97

± ± ± ± ± ± ± ± ± ±

0.09 0.08** 0.10** 0.12** 0.06** 0.03 0.02* 0.01* 0.02* 0.13*

12 12 12 12 12 24 24 24 24 24

0.59 0.64 0.78 0.73 0.80 0.31 1.18 0.47 0.71 0.79

± ± ± ± ± ± ± ± ± ±

0.09 0.05 0.05** 0.03** 0.05** 0.02 0.09* 0.06** 0.09* 0.07*

12 12 12 12 12 24 24 24 24 24

3.67 5.72 7.00 6.77 5.30 2.38 6.89 8.73 9.83 8.98

± ± ± ± ± ± ± ± ± ±

0.06 0.14* 0.07* 0.35* 0.07* 0.05 0.14* 0.15* 0.19* 0.07*

12 12 12 12 12 24 24 24 24 24

0.017 0.53 0.37 0.57 0.13 0.89 2.52 3.11 3.44 3.11

± ± ± ± ± ± ± ± ± ±

0.00 0.02* 0.02* 0.04* 0.01* 0.05 0.05* 0.11* 0.11* 0.02*

Statistically significant difference (P < 0.01, by t-test) compared with the control (untreated cells). Statistically significant difference (P < 0.02, by t-test) compared with the control (untreated cells).

Diagram 3. Fold increase of caspase-3 in primary lymphocyte cultures derived from donor 3. Compounds and time-points tested compared with the relevant control.

ferent pattern of activation. There was indeed a very pronounced early activation of this enzyme after a 12 h treatment, displaying a 7–15-fold increase for donor 1 (Diagram 4) and 7–33-fold increase for donor 3 (Diagram 6), but this effect diminished after the 24 h treatment. Surprisingly, donor 2 displayed a less pronounced activation of this enzyme only after 24 h of exposure (Diagram 5). The

Diagram 5. Fold increase of caspase-2 in primary lymphocyte cultures derived from donor 2. Compounds and time-points tested compared with the relevant control.

activation of caspase-2 was also much more variable among treatments with the different compounds than activation of caspase-3. Variations encountered among lymphocyte cultures derived from different donors can be mainly attributed to different DNA-repair capacities between individuals, which can significantly affect the cellular response to genotoxic stimuli. 4. Discussion

Diagram 4. Fold increase of caspase-2 in primary lymphocyte cultures derived from donor 1. Compounds and time-points tested compared with the relevant control.

It has been established by now that different modifications in the steroidal nucleus of several chemotherapeutic DNA-damaging molecules are able to significantly affect the genotoxic activity and consequently the effectiveness of these compounds [14,15,28]. In this study, we have investigated the genotoxic, cytostatic/cytotoxic, antineoplastic and apoptotic effects of three different newly synthesized compounds that contain CHL or PHE as alkylating agents and are also esterified with a steroidal skeleton modified in the Band D-rings.

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

This investigation was based on the analysis of SCE and PRI both in lymphocyte cultures in vitro and in ascites cells of lymphocytic P388 leukemia-bearing-mice in vivo, on T/C% calculations in P388 and L1210 leukemia-bearing mice, as well as on biochemical measurements of the activation of key enzymes, such as caspase-2 and caspase-3, involved in apoptosis caused by DNA-damaging agents. SCE have been frequently used as highly sensitive indicators of DNA damage and/or subsequent repair [29]. Non-repaired DNA damage expressed as SCE in normal cells caused by certain chemicals may indicate the inability to repair the damage-induced by the same chemicals in cancer cells since both cell types have similar DNArepair mechanisms [7]. Moreover, there are findings indicating that effectiveness in SCE induction by potential antitumour agents in cancer cells in vitro and in vivo is positively correlated with the in vivo tumour response to these agents [29]. According to our results all three compounds proved to be cytogenetically active: they significantly increased the frequencies of SCE while reducing PRI values in both normal human lymphocyte cultures in vitro and in ascites tumour cells of lymphocytic P388 leukemia-bearing mice in vivo (Tables 1 and 2). The order of decreasing cytogenetic activity was “compound 3 > compound 2 > compound 1” for the in vitro assays and “compound 3 = compound 2 > compound 1” for the in vivo assays, respectively. The cytogenetic effects (increase in SCE frequency) were more pronounced in vitro (Table 1), whereas the cytostatic activity, assessed by the PRI values, was more pronounced in vivo (Table 1). All derivatives have a keto group in the steroidal B-ring, whereas they differ in the D-ring and in the alkylator congener (Fig. 1). Compound 1 was not as effective in increasing the SCE frequency and in reducing the PRI, compared with compound 3. Moreover, compound 3 presented a dose-dependent statistically significant increase in SCE and decrease in PRI and was more active than compound 2, which has the same alkylating agent and differs at the modification of the steroidal D-ring (Fig. 1, Table 2). Especially, compound 3 has an exocyclic insertion of an acetamide group that could be responsible for the better cytogenetic and cytostatic activities. In the case of compound 2, the SCE frequency starts falling after reaching a plateau, whereas the PRI value continues to decrease. The latter can be explained by the fact that, as increasing doses of the compounds disturb the cell-cycle (Table 1), those cells with a higher number of SCE are eliminated before being able to reach the second division metaphase [7]. Comparing the acute toxicity values of the compounds, it is obvious that an esterified steroidal nucleus with PHE as alkylating agent (compounds 2 and 3) results in an increased toxicity, in contrast to esterification with CHL (compound 1). Moreover, as shown in Tables 4 and 5, compounds 2 and 3 caused a more pronounced increase in T/C% values than compound 1. Therefore, we can conclude from the results of the in vivo studies (cytogenetic and antineoplastic) that nitrogen mustards having PHE esterified with a steroidal skeleton modified by a–NHCO–group (endocyclically or exocyclically) have much stronger cytostatic and antineoplastic effects in comparison with those having CHL esterified to the steroidal skeleton. These results could be attributed to stereochemical hindrance (the ester bond in CHL esters is less efficiently cleaved by esterases) in the compounds, thus resulting in a decrease of the proportion of free alkylating agent in the cell and subsequently to the reduction of the toxic effects derived from CHL itself [28]. Additionally, the presence of the D-lactam group improves the activity of the final derivatives [30]. The presence of –NHCO– moiety in the steroidal skeleton of nitrogen mustard esters appears to alter the biological behavior of the molecule, possibly because of the multiple interactions of this group with similar groups present in proteins and nucleic acids [31]. The biological action of active esterified steroidal derivatives of chlorambucil, such as compounds 2 and 3 (Fig. 1) bearing the

57

–NHCO– moiety, may be structurally specific and therefore more persistent, as long as this group is inserted to the proper position in the steroidal skeleton. Nitrogen mustard-induced alkylation of DNA results predominately in the development of purine-drug complexes [32]. The nitrogen mustards, including chlorambucil and melphalan, may also form intrastrand and/or interstrand cross-links [33]. The presence of a crosslink in the template strand may inhibit DNA replication and cause SCE induction. The latter is a mechanism by which the cell is trying to circumvent this damage and continue DNA replication [34,35]. Moreover, in response to genotoxic injury from various agents, including modified steroidal esters of alkylators [36], poly (ADP-ribose) polymerase (PARP) catalyses the synthesis of poly (ADP-ribose) from its substrate beta NAD+ and as a consequence the DNA-repair programme is activated. PARP is an enzyme that detects DNA strand-breaks during DNA excision-repair to maintain genomic stability [36,37,38]. Inhibition or loss of PARP activity by various agents leads to a recombinogenic phenotype characterized by an increased frequency of SCE [36,38]. On the other hand, inhibition of DNA replication by cross-links may initiate processes leading to programmed cell death when the damage cannot be repaired [39,40,41]. This could be confirmed in the present study for all three newly synthesized compounds by testing the activation of caspase-3 and caspase-2 (Table 6). Caspase3 is a main executor enzyme of apoptosis, which cleaves a specific set of cellular substrates and enzymes, including PARP [37], resulting in the morphological and biochemical changes associated with the apoptotic phenotype [42]. Caspase-2 is unique among caspases in that it has both the features of an initiator and an effector caspase. Caspase-2 appears to be necessary for the onset of apoptosis triggered by genotoxic agents [27,43]. Pro-caspase 2 is the only one present constitutively in the nucleus [44,45]. Upon activation by DNA damage, caspase-2 mediates the recruitment of the mitochondrial pathway of apoptosis [27]. It appears that caspase2 activity is required for translocation of the death protein Bax to the mitochondria, as well as for the release of the mitochondrial proteins cytochrome c and Smac/DIABLO into the cytosol [46,47]. Moreover, Susin et al. [48] suggest that caspase-2 zymogens are essentially localized in mitochondria and that the disruption of the outer mitochondrial membrane occurring early during apoptosis may be critical for their sub-cellular redistribution and activation. Alternatively, caspase-2 could cleave the pro-apoptotic protein Bid, followed by its translocation to the mitochondria to induce cytochrome c release [47]. Caspase-2 can serve as an apical caspase in the proteolytic cascade triggered by DNA damage [27]. Therefore, the pronounced activation of caspase-2 by the compounds tested (Diagrams 4 and 6) could suggest that the alkylating steroidal esters act in the nucleus, causing DNA damage and finally leading the cell to apoptosis. Moreover, the activation of caspase2 induced by these agents was observed as early as 12 h after the start of the treatment, while in most of the cases it was reduced at the 24 h time point. The latter can be attributed to its early activation as initiator caspase in the apoptotic cascade [44,45,46,47]. The

Diagram 6. Fold increase of caspase-2 in primary lymphocyte cultures derived from donor 3. Compounds and time-points tested compared with the relevant control.

58

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59

activation of caspase-3, the main effector caspase, was more pronounced after the 24 h treatment (Diagrams 1–3) [46,47]. We did not study more prolonged exposure times, as the viability of the cultures treated with the compounds for more than 24 h was very low (less than 30%) because of the increased toxicity. In our study, there were some variations in the potency for cytogenetic damage and the apoptotic effects observed in lymphocyte cultures derived from different donors. These variations could be attributed to biochemical pathways genetically controlled on an individual basis, which are responsible for DNA repair enzymes, cell proliferation, DNA transcription and apoptosis [49,50]. This variation in DNA-repair capacity of each individual is also expressed by differences in the expression of SCE in different donors. Bender et al. reported that the mean unadjusted frequency of SCE was 8.29 ± 0.08/cell in a large population sample, whereas there is a great variance (5–14 SCE/cell) of the SCE frequencies reported by various laboratories for healthy populations [51]. In conclusion, the current study shows that the newly synthesized compounds caused both in vitro and in vivo a significant increase in the SCE frequencies. This increase was followed by (a) cytostatic activity (decrease of PRI), (b) antileukemic activity and (c) induction of apoptosis, as expressed by the activation of caspase-2 and caspase-3. These findings lead us to propose these compounds agents with potential antineoplastic properties. In addition, the present work seems to further substantiate the assumption that the SCE assay has an application in the clinical prediction of tumour sensitivity to potential chemotherapeutics. Conflict of interest None. References [1] L.H. Hurley, DNA and its associated processes as targets for cancer therapy, Nat. Rev. Cancer 2 (3) (2002) 188–200. [2] K.W. Kohn, Beyond DNA cross-linking: history and prospects of DNA-targeted cancer treatment—fifteenth Bruce F. Cain Memorial Award Lecture, Cancer Res. 56 (24) (1996) 5533–5546. [3] K.E. Dusenbery, E.E. Bellairs, R.A. Potish, L.B. Twiggs, M.P. Boente, Twenty-five year outcome of sequential abdominal radiotherapy and melphalan: implications for future management of epithelial carcinoma of the ovary, Gynecol. Oncol. 96 (2) (2005) 307–313. [4] J. Shamash, G. Dancey, C. Barlow, P. Wilson, W. Ansell, R.T. Oliver, Chlorambucil and iomustine (CL56) in absolute hormone refractory prostate cancer: re-induction of endocrine sensitivity an unexpected finding, Br. J. Cancer 92 (1) (2005) 36–40. [5] P.D. Lawley, Alkylation of DNA and its aftermath, Bioassays 17 (6) (1995) 561–568. ˜ [6] F. Drabløs, E. Feyzi, P.A. Aas, C.B. Vaagbø, B. Kavli, M.S. Bratlie, J. Pena-Diaz, M. Otterlei, G. Slupphauq, H.E. Krokan, Alkylation damage in DNA and RNA—repair mechanisms and medical significance, DNA Repair (Amst.) 3 (11) (2004) 1389–1407. [7] V. Karayianni, E. Mioglou, Z. Iakovidou, D. Mourelatos, M. Fousteris, A. Koutsourea, E. Arsenou, S. Nikolaropoulos, A new approach for evaluating in vivo anti-leukemic activity using the SCE assay. An application on three newly synthesized anti-tumour steroidal esters, Mutat. Res. 535 (1) (2003) 79–86. [8] D. Mourelatos, Z. Iakovidou, E. Mioglou, I. Karapidaki, A. Koutsourea, E. Arsenou, M. Fousteris, S. Nikolaropoulos, SCEs and PRIs as indicators of structurebiological activity relationship of newly synthesized steroidal esters, Rev. Clin. Pharmakol. Pharmacokinet. 18 (2004) 52–54. [9] F.I. Carroll, A. Philip, J.T. Blackwell, D.J. Taylor, M.E. Wall, Antitumor and antileukemic effects of some steroids and other biologically interesting compounds containing an alkylating agent, J. Med. Chem. 15 (11) (1972) 1158–1161. [10] K.M. Kasiotis, P. Magiatis, H. Pratsinis, A. Skaltsounis, V. Abadji, A. Charalambous, P. Moutsatsou, S.A. Haroutounian, Synthesis and biological evaluation of novel daunorubicin–estrogen conjugates, Steroids 66 (10) (2001) 785–791. [11] A. Kouloumenta, G. Stephanou, N.A. Demopoulos, S.S. Nikolaropoulos, Genetic effects caused by potent antileukemic steroidal esters of chlorambucil’s active metabolite, Anticancer Drugs 16 (1) (2005) 67–75. [12] H. Wasan, J. Waxman, in: L.L. Pusztai, C.E. Lewis, E. Yap (Eds.), Cell Proliferation in Cancer: Regulatory Mechanisms of Neoplastic Cell Growth, Oxford University Press, Oxford, 1996, pp. 260–281. [13] A.P. Lupulescu, Hormones, vitamins and growth factors in cancer treatment and prevention. A critical appraisal, Cancer 78 (11) (1996) 2264–2280.

[14] E.S. Arsenou, M.A. Fousteris, A.I. Koutsourea, A. Papageorgiou, V. Karayianni, E. Mioglou, Z. Iakovidou, D. Mourelatos, S.S. Nikolaropoulos, The allylic 7-ketone at the steroidal skeleton is crucial for the antileukemic potency of chlorambucil’s active metabolite steroidal esters, AntiCancer Drugs 15 (10) (2004) 983–990. [15] M.A. Fousteris, A. Papageorgiou, E.S. Arsenou, A.I. Koutsourea, E. Karaberis, D. Mourelatos, D.O. Onyango, S.S. Nikolaropoulos, Antileukemic and cytogenetic activity by triple administration of three modified steroidal derivatives of nitrogen mustards, Chemotherapy 53 (2) (2007) 118–126. [16] D.F. Deen, L.E. Kendall, L.J. Marton, P.J. Tofilon, Prediction of human tumor cell chemosensitivity using the sister chromatid exchange assay, Cancer Res. 46 (4 Pt 1) (1986) 1599–1602. [17] E. Mylonaki-Charalambous, D. Mourelatos, A. Kotsis, A comparative study of the cytogenetic and antineoplastic effects induced by carboplatin in combination with niacin in human lymphocytes in vitro and in Ehrlich ascites tumor cells in vivo, Chemotherapy 44 (1998) 121–128. [18] D. Mourelatos, J. Dozi-Vassiliades, A. Kotsis, C. Goutsas, Enhancement of cytogenetic damage and of antineoplastic effect by caffeine in Ehrlich ascites tumor cells treated with cyclophosphamide in vivo, Cancer Res. 48 (5) (1988) 1129–1131. [19] P.J. Tofilon, I. Basic, L. Milas, Prediction of in vivo tumor response to chemotherapeutic agents by the in vitro sister chromatid exchange assay, Cancer Res. 45 (5) (1985) 2025–2030. [20] C. Camoutsis, D. Catsoulacos, V. Karayianni, A. Papageorgiou, D. Mourelatos, E. Mioglou, Z. Kritsi, S. Nikolaropoulos, Comparative study of sister chromatid exchange induction and anti-tumor effects by homo-aza-steroidal esters of {p-[bis(2-chloroethyl)amino]phenyl}butyric acid, Cancer Detect. Prev. 25 (6) (2001) 558–564. [21] A. Goldin, Z. Sofina, A. Syrkin, Experimental evaluation of antitumor drugs in the U.S.A. and U.S.S.R. and clinical correlations, Natl. Cancer Inst. Monogr. 55 (1980) 25–26. [22] A.I. Koutsourea, M.A. Fousteris, E.S. Arsenou, A. Papageorgiou, G.N. Pairas, S.S. Nikolaropoulos, Synthesis, in vivo antileukemic evaluation and comparative study of novel 5alpha-7-keto steroidal esters of chlorambucil and its active metabolite, In Vivo 22 (3) (2008) 345–352. [23] K. Goto, S. Maeda, Y. Kano, T. Sugiyama, Factors involved in differential giemsastaining of sister chromatids, Chromosoma 66 (4) (1978) 351–359. [24] P.B. Noble, J.H. Cutts, K.K. Carroll, Ficoll flotation for the separation of blood leukocyte types, Blood 31 (1) (1968) 66–73. [25] R.V. Talanian, C. Quinlan, S. Trautz, M.C. Hackett, J.A. Mankovich, D. Banach, T. Ghayur, K.D. Brady, W.W. Wong, Substrate specificities of caspase family proteases, J. Biol. Chem. 272 (15) (1997) 9677–9682. [26] M. Carcia-Calvo, E.P. Peterson, B. Leiting, R. Ruel, D.W. Nicholson, N.A. Thornberry, Inhibition of human caspases by peptide-based and macromolecular inhibitors, J. Biol. Chem. 273 (49) (1998) 32608–32613. [27] B. Zhivotovsky, S. Orrenius, Caspase-2 function in response to DNA damage, Biochem. Biophys. Res. Commun. 331 (3) (2005) 859–867. [28] A. Papageorgiou, A.I. Koutsourea, E.S. Arsenou, M.A. Fousteris, D. Mourelatos, S.S. Nikolaropoulos, Structure-anti-leukemic activity relationship study of Band D-ring modified and non-modified steroidal esters of chlorambucil’s active metabolite, Anticancer Drugs 16 (10) (2005) 1075–1082. [29] D. Mourelatos, Chromosomes study as predictor of chemoresponse of tumours, Cancer J. 9 (3) (1996) 136–141. [30] P. Catsoulacos, D. Politis, G.L. Wampler, A new steroidal alkylating agent with improved activity in advanced murine leukemias, Cancer Chemother. Pharmacol. 3 (1) (1979) 67–70. [31] J.A. Hartley, M.D. Berardini, R.L. Souhami, An agarose gel method for the determination of DNA interstrand crosslinking applicable to the measurement of the rate of total and ‘second arm’ crosslink reactions, Anal. Biochem. 193 (1) (1991) 131–134. [32] B.B. Bank, D. Kanganis, L.F. Liebes, R. Silber, Chlorambucil pharmacokinetics and DNA binding in chronic lymphocytic leukemia lymphocytes, Cancer Res. 49 (3) (1989) 554–559. [33] P.D. Lawley, P. Brookes, Interstrand cross-linking of DNA by bifunctional alkylating agents, J. Mol. Biol. 25 (1) (1967) 143–160. [34] D.A. Shafer, Replication bypass model of sister chromatid exchanges and implications for Bloom’s syndrome and Fanconi’s anemia, Hum. Genet. 39 (2) (1977) 177–190. [35] Y. Kano, Y. Fujiwara, Higher inductions of twin and single sister chromatid exchanges by cross-linking agents in Fanconi’s anemia cells, Hum. Genet. 60 (3) (1982) 233–238. [36] C. Petrou, D. Mourelatos, E. Mioglou, J. Dozi-Vassiliades, P. Catsoulacos, Effects of alkylating antineoplastics alone or in combination with 3-aminobenzamide on genotoxicity, antitumor activity and NAD levels in human lymphocytes in vitro and on Ehrlich ascites tumor cells in vivo, Teratog. Carcinog. Mutagen. 10 (4) (1990) 321–331. [37] P. Decker, S. Muller, Modulating poly (ADP-ribose) polymerase activity: potential for the prevention and therapy of pathogenic situations involving DNA damage and oxidative stress, Curr. Pharm. Biotechnol. 3 (3) (2002) 275–283. [38] S. Kyle, H.D. Thomas, J. Mitchell, N.J. Curtin, Exploiting the Achilles heel of cancer: the therapeutic potential of poly(ADP-ribose) polymerase inhibitors in BRCA2-defective cancer, Br. J. Radiol. 81 (1) (2008) 6–11. [39] J.J. Roberts, T.P. Brent, A.R. Crathorn, Evidence for the inactivation and repair of the mammalian DNA template after alkylation by mustard gas and half mustard gas, Eur. J. Cancer 7 (6) (1971) 515–524. [40] R.B. Painter, Effect of caffeine on DNA synthesis in irradiated and unirradiated mammalian cells, J. Mol. Biol. 143 (3) (1980) 289–301.

I. Karapidaki et al. / Mutation Research 675 (2009) 51–59 [41] K.W. Kohn, Biological aspects of DNA damage by crosslink agents, in: S. Neidle, M.J. Waring (Eds.), Molecular Aspects of Anti-Cancer Drug Action, 1983, pp. 315–361. [42] G.S. Salvesen, V.M. Dixit, Caspases: intracellular signalling by proteolysis, Cell 91 (4) (1997) 443–446. [43] T.J. Fan, L.H. Han, R.S. Cong, J. Liang, Caspase family proteases and apoptosis, Acta Biochim. Biophys. Sin. 37 (11) (2005) 719–727. [44] M. Mancini, C.E. Machamer, S. Roy, D.W. Nicholson, N.A. Thornberry, L.A. Casciola- Rosen, A. Rosen, Caspase-2 is localized at the Golgi complex and cleaves golgin-160 during apoptosis, J. Cell Biol. 149 (3) (2000) 603–612. [45] B. Zhivotovsky, A. Samali, A. Gahm, S. Orrenius, Caspases: their intracellular localization and translocation during apoptosis, Cell Death Differ. 6 (7) (1999) 644–651. [46] P. Lassus, X. Opitz-Araya, Y. Lazebnik, Requirement for caspase-2 in stressinduced apoptosis before mitochondrial permeabilization, Science 297 (5585) (2002) 1352–1354.

59

[47] Y. Guo, S.M. Srinivasula, A. Druilhe, T. Fernandes-Alnemri, E.S. Alnemri, Caspase2 induces apoptosis by releasing proapoptotic proteins from mitochondria, J. Biol. Chem. 277 (16) (2002) 13430–13437. [48] S.A. Susin, H.K. Lorenzo, N. Zamzami, I. Marzo, C. Brenner, N. Larochette, M.C. Prévost, P.M. Alzari, G. Kroemer, Mitochondrial release of caspase-2 and -9 during the apoptotic process, J. Exp. Med. 189 (2) (1999) 381–393. [49] J.K. Wiencke, M.R. Wrensch, R. Miike, N.L. Petrakis, Individual susceptibility to induced chromosome damage and its implications for detecting genotoxic exposures in human populations, Cancer Res. 51 (19) (1991) 5266– 5269. [50] L. Hagmar, U. Strömberg, H. Tinnerberg, Z. Mikoczy, The usefulness of cytogenetic biomarkers as intermediate endpoints in carcinogenesis, Int. J. Hyg. Environ. Health 204 (1) (2001) 43–47. [51] M.K. Conner, N. Wald, Chromosomal methods in population studies, Environ. Health Perspect. 42 (1981) 107–113.