Irinotecan Toxicity to Human Blood Cells in vitro: Relationship between Various Biomarkers

Authors


Author for correspondence: Nevenka Kopjar, Mutagenesis Unit, Institute for Medical Research and Occupational Health, Ksaverska c.2, HR-10 000 Zagreb, Croatia (fax + 385-1-4673303, e-mail nkopjar@imi.hr).

Abstract

Abstract:  Toxic effects of the antineoplastic drug irinotecan on human blood cells at concentrations of 9.0 µg/ml and 4.6 µg/ml were evaluated in vitro. Using the alkaline and neutral comet assay significantly increased levels of primary DNA damage in lymphocytes were detected. The induction of apoptosis/necrosis, as determined by a fluorescent assay, was also notably increased. Cytogenetic outcomes of the treatment were assessed by the analysis of structural chromosome aberrations and fluorescence in situ hybridization. A significantly higher incidence of chromatid breaks and complex quadriradials was observed. Painted chromosomes 1, 2 and 4 were equally involved in translocations, but only the chromosome 1 was involved in the formation of quadriradials. Sister chromatid exchange analysis was performed in parallel with the analysis of lymphocyte proliferation kinetics. The higher concentration of irinotecan caused almost seven-time increase, while the lower one caused a five-time increase of the basal sister chromatid exchange frequency, accompanied with significant lowering of the lymphocyte proliferation index. Using the cytokinesis-block micronucleus assay, a dose-dependent increase in micronucleus frequency along with the formation of nuclear buds and nucleoplasmic bridges was noticed. Inhibitory effects of irinotecan on enzyme acetylcholinesterase (AChE) were studied in erythrocytes. An IC50 value of 5.0 × 10−7 was established. Irinotecan was found to be strong inhibitor of the acetylcholine hydrolysis and to cause a continuous decrease of catalytic activity of AChE. The results obtained on a single donor may contribute to the understanding of irinotecan toxicity, but further in vitro and in vivo studies are essential in order to clarify remaining issues, especially on possible inter-individual variability in genotoxic responses to the drug.

Irinotecan (7-ethyl-10-[4-(1-piperidino)-1-piperidino] carbonyloxycamptothecin; CPT-11) is clearly one of the most important new anticancer drugs developed in the last few decades. It is a member of the camptothecin drug family [1]. Irinotecan is a pro-drug that is biotransformed by tissue and serum carboxylesterases to an active metabolite, SN-38 (7-ethyl-10-hydroxycamptothecin) that has a 100–1000-time higher cytotoxic and antitumour activity [2]. It acts as an inhibitor of the nuclear enzyme topoisomerase I, which is involved in cellular DNA replication and transcription. During replication, topoisomerase I mediates the relaxation of super-coiled DNA, and its inhibition results in breakage of the DNA chain and likely induces apoptosis. Irinotecan and SN-38 both bind to the topoisomerase I-DNA complex and prevent religation of single-strand breaks [3]. Irinotecan has undergone extensive clinical investigation worldwide and demonstrated potent activity against many types of human cancer, but is particularly active in the treatment of gastrointestinal and pulmonary malignancies [1–4].

The principal dose-limiting toxicity of irinotecan is diarrhoea. It either causes acute diarrhoea related to a cholinergic surge from inhibition of acetyl cholinesterase, or a delayed diarrhoea syndrome, which is possibly related to the accumulation of the active metabolite of irinotecan in the bowel [5]. Other non-haematological toxicities include nausea, vomiting, anorexia, fatigue, abdominal pain, alopecia, asthenia and elevated alkaline phosphatase and/or hepatic transaminases. The most common haematological toxicity is dose-related neutropenia. Myelosuppression is generally not cumulative, and severe anaemia or thrombocytopenia is less common [1,6,7].

Cytogenetic consequences of chemotherapy with irinotecan to normal cells were not extensively studied. Because conventional antitumour drugs are indiscriminate, the adverse consequences of chemotherapy to non-tumour cells and tissues are almost always present. The aim of this study was to evaluate the toxicity profile of irinotecan on human non-target cells following in vitro treatment with two concentrations of the drug proportionate to its therapeutic doses. A multi-biomarker approach was used: lymphocyte viability and the induction of apoptosis/necrosis caused by the exposure to irinotecan were studied by simultaneous use of a fluorescent assay with ethidium bromide and acridine orange; the levels of primary DNA damage in lymphocyte genome and the dynamics of DNA repair were evaluated using the alkaline and neutral comet assay; the levels and nature of residual DNA damage were assessed by the analysis of structural chromosome aberrations and fluorescence in situ hybridisation (FISH). For the study of cytogenetic effects, the cytokinesis-block micronucleus (CBMN) assay and sister chromatid exchange (SCE) analysis were also employed, while the possible influences of treatment on the progression through the mitotic cycles were studied by analysing lymphocyte proliferative kinetics. To further address the question of whether irinotecan acts as specific blocker of enzyme acetylcholinesterase (AChE), we performed an in vitro relevant experiment on human erythrocyte AChE.

Material and Methods

Blood sampling.  Blood sample was obtained from a healthy female donor (age of 35 years, non-smoker) who gave informed consent for participation in the study. The donor had not been exposed to diagnostic or therapeutic irradiations as well as to known genotoxic chemicals for a year before blood sampling. Venous blood (40 ml) was collected under sterile conditions in heparinized vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ, USA) containing lithium heparin as anticoagulant.

Chemicals and reagents.  The test chemical, irinotecan was purchased from Aventis Pharma Ltd. (CAMPTO®, Dagenham, UK). It was used in the form of irinotecan hydrochloride trihydrate. Before use, it was dissolved in 5 ml of sterile concentrate solution for infusion to a final concentration of 20 mg/ml. The final concentrations of the antineoplastic drug in culture medium with human blood cells in our in vitro experiment were 9.0 µg/ml and 4.6 µg/ml. They were deduced form the therapeutic doses of irinotecan used in monotherapy (350 mg/m2), respectively in combination chemotherapy (180 mg/m2) [4] and calculated on the basis of average body weight and average surface area of an adult person. Other chemicals and reagents used, if not specified, were purchased form Sigma Chemical Co. (St. Louis, MO, USA).

Experimental procedure.  Two independent experiments were conducted on the same blood sample and the data were pooled. The evaluation of cell viability, apoptosis measurements and comet assay were performed on isolated lymphocytes. Six aliquots of isolated lymphocytes were placed in sterile Falcon tubes in final concentration of 2 × 106 cells/ml. Four aliquots were incubated with irinotecan for 2 hr at 37°C, while two were negative controls. After 2 hr of exposure, culture medium containing irinotecan was carefully removed and the cells were washed twice in fresh F-10 medium. Primary DNA damage and cell viability was evaluated immediately after the treatment (time-point 0′), while the effectiveness of DNA repair was checked at 30, 60, 90 and 120 min. after the treatment. The cell viability using a fluorescence assay was studied in parallel during the whole post-incubation period. The evaluation of A part of heparinized blood was used for the cytogenetic analyses. For each culture 0.8 ml of heparinized blood were added to 8.0 ml of standard Ham's F-10 culture medium supplemented with 20% foetal calf serum and antibiotics: penicillin 100 IU/ml (Pliva, Zagreb, Croatia) and streptomycin 100 µg/ml (Krka, Novo Mesto, Slovenia). For each end-point, an untreated control sample was also included.

Cell cultures were incubated in vitro at 37°C in humified atmosphere with 5.0% CO2 (Heraeus Heracell 240 incubator, Langenselbold, Germany). After 2 hr of in vitro treatment, the incubation medium was removed and replaced with fresh culture medium. Then, phytohemagglutinin (PHA; Apogent, Hudson, NH, USA) was added (0.2 ml per each culture). Subsequent steps of cultivation, cell harvesting and slides preparation were performed as recommended for the particular technique. One aliquot of heparinized blood was used for the determination of the AChE activity in erythrocytes and the study of the inhibition of AChE by irinotecan.

Lymphocyte isolation.  Anticoagulant-treated blood was mixed 1:1 (v/v) with balanced salts solution, layered on the Ficoll solution and centrifuged at 375 g for 40 min. at room temperature. The layer containing lymphocytes was carefully removed and cells were re-suspended in balanced salts solution. They were washed twice by centrifugation at 600 r.p.m. for 10 min. The final pellet was gently re-suspended in culture medium F-10. Viability of cells was checked by supravital staining with 0.1% trypan blue [8].

Assessment of cell viability, apoptosis and necrosis.  For studying cell death and morphological changes in the nuclei, we used dye exclusion method [9] in which viable (intact plasma membrane) and dead (damaged plasma membrane) cells can be visualized after staining with the fluorescent DNA-binding dyes. Ethidium bromide and acridine orange were added to the cell suspension in final concentrations of 100 µg/ml (1:1; v/v). Two parallel tests with aliquots of the same sample were performed and a total of 500 cells per sample were counted. Quantitative assessments were made by determination of the percentage of apoptotic and necrotic cells.

The comet assay.  The comet assay was carried out under alkaline conditions, as described by Singh et al. [10] and neutral conditions according to Wojewódzka et al. [11]. Two replicate slides per sample per method were prepared. Agarose gels were prepared on fully frosted slides coated with 1% and 0.6% normal melting point agarose. Lymphocyte samples (5 µl) were mixed with 0.5% low melting point agarose, placed on the slides and covered with a layer of 0.5% low melting point agarose. The slides were immersed for 1 hr in freshly prepared ice-cold lysis solution (2.5 M NaCl, 100 mM Na2EDTA, 10 mM Tris-HCl, 1% Na-sarcosinate, pH 10) with 1% Triton X-100 and 10% dimethyl sulfoxide (Kemika, Zagreb, Croatia). Alkaline denaturation and electrophoresis were carried out at 4°C under dim light in freshly prepared electrophoretic buffer (300 mM NaOH, 1 mM Na2EDTA, pH 13.0). After 20 min. of denaturation, the slides were randomly placed side by side in the horizontal gel-electrophoresis tank, facing the anode. Electrophoresis at 25 V lasted another 20 min. Neutral denaturation was carried out in the dark at 8°C and lasted for 1 hr in a buffer containing 300 mM sodium acetate and 100 mM Tris-HCl, pH 8.5. It was followed by electrophoresis at 14 V and 11–12 mA that also lasted for 1 hr. After electrophoresis, the slides were gently washed with a neutralisation buffer (0.4 M Tris-HCl, pH 7.5) three times at 5 min. intervals. Slides were stained with ethidium bromide (20 µg/ml) and stored at 4°C in humidified sealed containers until analysis. Each slide was examined using a 250× magnification fluorescence microscope (Zeiss, Oberkochen, Germany) equipped with an excitation filter of 515–560 nm and a barrier filter of 590 nm. A total of 100 comets per sample were scored (50 from each of two replicate slides). Comets were randomly captured at a constant depth of the gel, avoiding the edges of the gel, occasional dead cells and superimposed comets. Using a black and white camera, the microscope image was transferred to a computer-based image analysis system (Comet Assay II, Perceptive Instruments Ltd., Suffolk, UK). To avoid the variability, one well-trained scorer scored all comets. As a measure of DNA damage, tail moment was chosen.

Analysis of structural chromosome aberrations.  The chromosome aberration test was performed in agreement with International Programme on Chemical Safety guidelines [12]. One thousand metaphases per sample were scored for total numbers and types of aberrations, as well as the percentage of aberrant cells.

Fluorescence in situ hybridization.  Slides for metaphase FISH were prepared according to standard International Atomic Energy Agency [13] without final incineration. FISH was performed on 3-week-old metaphase spreads with whole chromosome-painting probes, following the instructions of the supplier (Cytocell Technologies Ltd., Cambridge, UK). Directly labelled Aquarius® chromosome-painting probes were used to paint chromosomes 1, 2 and 4. They were labelled with a red fluorophore (Texas Red spectrum), green fluorophore (FITC spectrum) and combination of both fluorophores, respectively. The chromosomes were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) prepared in an antifade solution (Cytocell Technologies Ltd.). Probed slides were scored using Olympus AX70 epifluorescence microscope (Olympus Optical, London, UK). A triple-band pass filter was used to permit simultaneous viewing of the probed chromosome pairs and other chromosomes counter stained by the DAPI. Furthermore, separate specific filters for Texas Red, FITC and DAPI were used. We used PAINT nomenclature system according to Tucker et al. [14].

Sister chromatid exchange assay and the analysis of lymphocyte proliferation kinetics.  Cell cultures were set up according to standard protocol [15]. 5-Bromodeoxyuridine (5 µg/ml) was added at the initiation of the cultures. To obtain harlequin chromosomes, slides were stained using a modified fluorescence plus Giemsa method as described by Perry and Wolff [16]. A total of 100 randomly selected second division metaphases (50 from each of two replicates) per each sample were analysed blindly. The number of SCE per metaphase and range of SCEs were determined.

In fluorescence plus Giemsa-stained preparations, cells dividing for the first (M1), second (M2) or third time (M3) in culture containing bromodeoxyuridine were determined by differential staining pattern of sister chromatids. Lymphocyte proliferation kinetics was studied on 200 differentially stained metaphases per each blood sample. The proliferation rate index (PRI) was calculated according to the formula: PRI = (M1 + 2M2 + 3M3)/total number of cells scored, as reported by Lamberti et al. [17].

Cytokinesis-block micronucleus assay.  Lymphocyte cultures were incubated according to standard protocol for CBMN assay [18]. Cytochalasin B in final concentration 6 µg/ml was added in culture at 44 hr. For MN identification the criteria of Fenech et al. [19] were used. The number of MN in 2000 binucleated (BN) cells was scored for each treatment and the number of BN cells with MN was also recorded. Two thousand cells were scored to determine the percentage of cells with one to four nuclei. A nuclear division index (NDI) was calculated according to the formula: NDI = (M1 + 2M2 + 3M3 + 4M4)/1000 [20].

Determination of the AChE activity.  In our experiments, the inhibitory power of irinotecan on AChE of human erythrocytes is determined and compared to carbamate physostigmine (eserine), which is tested as reference drug. Physostigmine salicylate salt was used at a dose of 0.1 mg/kg. Acetylthiocholine iodide (ATCh) and 5,5′dithiobis-2-nitrobenzoic acid (thiol reagent; DTNB) were of analytical grade.

Enzyme assay and preparation.  The activity of AChE was determined by means of a colorimetric assay based on the enzymatic conversion of ATCh to thiocholine, which reacts with DTNB to generate the chromogen compound 5-thio-2-nitrobenzoate [21]. The assay solution was maintained at a temperature of 37°C throughout the analytic procedure and consisted of a 0.1 mol/l phosphate buffer, pH 7.4 that contained 0.01 mol/l DTNB. The source of AChE was native human erythrocytes; the final dilution during enzyme assay was 400 times larger. The reaction was performed in the total volume of 3.0 ml. The increase in absorbance was read at 412 nm, at 15 sec. intervals; against a blank that contained the erythrocytes suspended in buffer and DTNB.

Inhibition of AChE by irinotecan and physostigmine.  (i) The concentration of the compound yielding a 50% enzyme inhibition (IC50) was determined by incubating erythrocytes with four or more different concentrations of irinotecan and physostigmine at 4°C and assaying for AChE activity after 15 min. Relative changes in the enzyme activity are presented as the percentage of activity of the respective of control. Only those values between 10% and 90% inhibition were used for calculation. (ii) For the purpose of biochemical experiments, we also determined catalytic activity of AChE previously inhibited with two different concentrations of irinotecan (9.0 and 4.6 µg/ml) or physostigmine (0.1 µ/ml). The reaction mixture was incubated for 10, 30, 60, 120, 150 and 180 min. at 37°C after addition of each of drugs and activity of AChE was measured. Each of samples was assayed in triplicate, and a mean value was then calculated. The effects of irinotecan and physostigmine were expressed as percentages of the AChE activity of the respective of control.

Statistical analyses.  Statistical analyses were carried out with a commercial programme Statistica 5.0 (Statsoft, Tulsa, OK, USA).

In order to normalize distribution and to equalize the variances, a logarithmic transformation of data was applied. The level of statistical significance was set at P < 0.05. The extent of DNA damage, as recorded by the alkaline and neutral comet assay, was analysed considering the mean (±S.E.M.), median and range of the comet tail moment. Comparisons between samples were done using the one-way anova and subsequently the Duncan's test for the calculations concerning pair-wise comparisons. The comparisons between values obtained for the cell viability, apoptosis and necrosis in irinotecan-treated and control samples were made by χ2-test. The same test was applied for the statistical evaluation of data considering the structural chromosome aberrations in lymphocytes, CBMN assay and lymphocyte proliferation kinetics. The statistical significance of data obtained by the SCE test was evaluated using the one-way anova and the Duncan's test for the calculations concerning pair-wise comparisons. Each sample was characterized for the extent of DNA damage by considering the mean (±S.D.), median and range of SCE per cell.

Results

Cell viability, apoptosis and necrosis.

After isolation, the lymphocyte viability in the sample prepared for the experiments was 98.8%. The percentages of viable and non-viable: apoptotic or necrotic lymphocytes after 2 hr in vitro exposure to irinotecan are reported in table 1. In vitro treatment with irinotecan caused a dose-dependent decrease in cell viability, accompanied by increases in the percentage of apoptotic and necrotic cells. Reduced cell viability compared to control was observed at each time-point following in vitro treatment with higher dose of irinotecan (P < 0.05, χ2-test) and only in sample analysed at time-point 0 after treatment with lower dose of irinotecan. Despite of reduced cell viability in sample treated with a higher dose of irinotecan, differences between this sample and a sample treated with a lower dose of drug were not statistically significant. The same was observed for the incidence of apoptotic cells in treated samples. During the 2 hr of experiment, the viability of lymphocytes in the control sample gradually decreased, along with an increase in the proportion of apoptotic cells. At time-point 120 min., the cell viability was significantly lower compared to time-points 0 and 30 min. (table 1).

Table 1. 
Results of the quantitative fluorescent assay for the simultaneous identification of apoptotic and necrotic cells in samples of isolated peripheral blood lymphocytes treated for 2 hr in vitro with irinotecan (IRI).
Time after treatmentViable cells (%)Non-viable cells (%)Necrotic
ΣApoptotic
Intact membraneDamaged membraneΣ Apo
  1. Evaluation was made by analysing 500 cells per sample per each experimental point. ↑significantly different as compared to control sample analysed at the same time; *significantly different as compared to samples 0′ and 30′ P < 0.05, χ2-test).

IRI 9.0 µg/ml
0′63.436.626.4 4.831.2 5.4
30′59.640.418.213.631.8 8.6
60′63.037.016.016.032.0 5.0
90′58.441.627.6 6.033.6 8.0
120′52.247.823.412.435.812.0
IRI 4.6 µg/ml
0′65.834.230.0 2.432.4 1.8
30′71.029.021.0 4.025.0 4.0
60′67.432.617.6 9.627.2 5.4
90′65.035.023.8 7.030.8 4.2
120′63.636.416.612.829.4 7.0
Control
0′85.814.212.212.2 2.0
30′82.417.615.2 0.816.0 1.6
60′77.822.217.6 3.421.0 1.2
90′74.625.421.0 2.423.4 2.0
120′67.2*32.824.2 4.829.0* 3.8

Alkaline comet assay.

Baseline DNA damage (mean tail moment) in freshly isolated lymphocytes prepared for the experiments was 0.12 ± 0.02 (median 0.05; min–max 0–1.87). The value of tail moment, observed at time-point 0, indicates that isolated lymphocytes during 2 hr of incubation in vitro effectively repaired minor DNA lesions inflicted by isolation procedure. The values of the comet tail moment in control cells steadily increased at later time-points, as displayed in fig. 1, but were still significantly lower than in irinotecan-treated samples. Both doses of irinotecan evoked significant increase in the tail moment. Unexpectedly, the increase was more pronounced at the lower dose applied, but the difference was not statistically significant. The values of tail moment in both irinotecan-treated samples were significantly increased as compared to the control sample, in most time-points following in vitro treatment (P < 0.05, analysis of variance; fig. 1). The highest level of DNA damage in both irinotecan-treated samples was observed at time-point 60 min. after treatment. Because the comet tail moments are positively correlated with the level of DNA breakage in the cell, it is likely that the amount of single strand breaks and alkali-labile sites in irinotecan-treated cells steadily increased in first 60 min. after treatment. During the later 30 min. of incubation, the levels of DNA damage in both samples were significantly lowered. According to our observations, the cells exposed to the lower dose of the drug were able to recover within the 120 min. incubation after treatment. However, in the sample treated with a higher dose at time-point 120 min., a new peak of DNA damage was recorded, and this damage level was significantly higher as compared to the samples 0, 30 and 90 min. (fig. 1). We assume that it was a transient increase of DNA damage related to DNA repair and oxidative processes that generate additional single strand breaks and alkali-labile sites detectable by the alkaline modification of the assay.

Figure 1.

Distribution of comet tail moments measured in peripheral blood lymphocytes throughout the post-incubation period after exposure to irinotecan (IRI) in concentrations of 9.0 and 4.6 µg/ml. In vitro treatment with irinotecan lasted for 2 hr. Alkaline comet assay was employed. Each sample was characterized for the extent of DNA damage by considering the mean ± S.E. (indicated with inline image) of comet tail moment. Statistical significance was evaluated on logarithmically transformed data, using the one-way anova and the Duncan's test for the calculations concerning pair-wise comparisons; the level of statistical significance was set as P < 0.05. Significantly increased compared to sample 0′– a; sample 30′– b; sample 60′– c; sample 90′– d; sample 120′– e; control sample – *; sample treated with lower dose of irinotecan – **.

Neutral comet assay.

Baseline DNA damage (mean tail moment) in freshly isolated lymphocytes prepared for the experiments was 0.13 ± 0.02 (median 0.08; min–max 0–1.30). Contrary to the irinotecan-treated samples, primary DNA damage in control lymphocytes gradually increased within first 60 min. of the incubation after treatment. Later on, it steadily decreased to values comparable with those recorded in freshly isolated cells (fig. 2). Exposure to irinotecan evoked a dose-dependent increase in the tail moment of treated lymphocytes. However, only the value, recorded in the sample exposed to a higher dose of drug, was significantly increased compared to control sample (P < 0.05, analysis of variance; fig. 2). The lowest level of primary DNA damage after treatment with both doses of irinotecan was recorded at time-point 60 min. Later on, the levels of DNA damage increased again (fig. 2). Statistically significant differences are indicated in fig. 2. It should be stressed that the levels of DNA damage recorded at time-point 120 min. in both irinotecan-treated samples were relatively high and still increased compared to the control sample.

Figure 2.

Distribution of comet tail moments measured in peripheral blood lymphocytes throughout the post-incubation period after exposure to irinotecan (IRI) in concentrations of 9.0 and 4.6 µg/ml. In vitro treatment with irinotecan lasted for 2 hr. Neutral comet assay was employed. Each sample was characterized for the extent of DNA damage by considering the mean ± S.E. (indicated with inline image) of comet tail moment. Statistical significance was evaluated on logarithmically transformed data, using the one-way anova and the Duncan's test for the calculations concerning pair-wise comparisons; the level of statistical significance was set as P < 0.05. Significantly increased compared to sample 0′– a; sample 30′– b; sample 60′– c; sample 90′– d; sample 120′– e; control sample – *; sample treated with lower dose of irinotecan – **.

Analysis of structural chromosome aberrations.

The results obtained for the treatments with two doses of irinotecan and corresponding control are presented in table 2. Both doses of irinotecan produced increased frequencies of cells with chromosomal aberrations and abnormal metaphases compared to control (P < 0.05, χ2-test). The chromosome aberrations detected at the highest frequency were chromatid breaks and complex quadriradial chromosome exchange figures. Acentric chromosomes, chromosome breaks and triradial figures followed. The least frequent aberrations were dicentric chromosomes and rings (only one ring chromosome was found in sample treated with a higher dose of irinotecan).

Table 2. 
Structural chromosome aberrations in peripheral blood lymphocytes treated for 2 hr in vitro with irinotecan (IRI).
TreatmentChromosomal aberrations (CA)/1000 cellsCells with CA (%)
B1B2TQAcDcRTotal
  1. One thousand metaphases per sample per each experimental point were scored. B1– chromatid break; B2– chromosome break; T – triradial figure; Q – quadriradial figure; Ac – acentric fragment; Dc – dicentric; R – ring; ↑significantly different as compared to control sample (P < 0.05, χ2-test).

IRI 9.0 µg/ml34144241421937.6
IRI 4.6 µg/ml2812111131665.8
Control 1 1 20.2

Fluorescence in situ hybridization.

A dose-related increase in translocation yield was observed in lymphocytes treated with irinotecan (table 3). In the control lymphocytes, a single translocation was detected in 1183 genome equivalents, whereas in the cells treated with the higher concentration of the drug 15 in 890 genome equivalents were observed. Complex rearrangements involving more than two chromosomes and more than two chromosome breaks were observed only in the lymphocytes treated with a higher dose of irinotecan. The number of acentric fragments originating from one of the painted chromosomes also increased with the concentration (0 in the control, 1 for the lower dose and 9 for the higher dose of irinotecan). All three painted chromosomes were equally involved in translocations. However, among them only the chromosome 1 was involved in chromatid exchanges resulting in formation of quadriradials. In sample treated with the lower dose of irinotecan, 33.3% of quadriradials contained the chromosome 1, and in the sample treated with the higher dose 36.3%. Thus, there was no correlation between the frequency of chromosome 1 involvement in chromatid exchanges and the dose applied.

Table 3. 
Results of the fluorescence in situ hybridization (FISH) performed using three-colour painting probes for chromosomes 1, 2 and 4 on peripheral blood lymphocytes treated for 2 hr in vitro with irinotecan (IRI).
TreatmentΣ CellsΣ TΣ rTComplex RRΣ T involvingΣ deletions resulting in acentricΣ QΣ Q involving
Ch 1Ch 2Ch 4Ch 1Ch 2Ch 4Ch 1Ch 2Ch 4
  1. T – translocations; rT – reciprocal translocations; RR – rearrangements; Ch – chromosome; Q – quadriradials.

IRI 9.0 µg/ml 8901510466245114
IRI 4.6 µg/ml 808 51131 62
Control1183 11

Sister chromatid exchange and lymphocyte proliferation kinetics.

The frequencies of SCEs observed in second-division metaphases in irinotecan-treated and control samples are listed in table 4. In the control sample, the mean SCE frequency was 3.11 ± 1.21 SCEs/metaphase (median 3 SCEs/metaphase; range 0–6 SCEs/metaphase). The observed value was within the control range normally observed in our laboratory. The higher concentration of irinotecan tested caused an almost seven times increase of the basal SCE frequency: mean SCE frequency in treated lymphocytes was 20.70 ± 6.36 SCEs/metaphase (median 20 SCEs/metaphase; range 9–38 SCEs/metaphase) (table 4). The observed difference was highly significant (P < 0.01, anova). On the other hand, the lower concentration caused a five-time increase of the basal SCE frequency: mean SCE frequency in treated lymphocytes was 15.47 ± 7.35 SCEs/metaphase (median 13.5 SCEs/metaphase; range 6–37 SCEs/metaphase) (table 4).

Table 4. 
Results of the sister chromatid exchange (SCE) assay and lymphocyte proliferation kinetics on peripheral blood lymphocytes treated for 2 hr in vitro with irinotecan (IRI).
TreatmentSCE/cellMitotic activity
Mean ± S.D.MedianRangeM1 (%)M2 (%)M3 (%)PRI
  1. Data on SCE are presented as mean values obtained by analysing of 100 second metaphases; lymphocyte proliferation kinetics was evaluated by analysing 200 cells per sample per each experimental point. M1, M2, M3 corresponded to the frequencies of cells in first, second and third in vitro division; PRI (proliferation rate index) was determined with the formula: 1M1 + 2M2 + 3M3/200; significantly increased as compared to control sample; significantly increased as compared to sample treated with lower concentration of irinotecan (P < 0.01; analysis of variance); *significantly different as compared to control sample P < 0.05, χ2-test).

IRI 9.0 µg/ml20.70 ± 6.36†,‡20.009–382862101.82*
IRI 4.6 µg/ml15.47 ± 7.3513.506–374352 51.62*
Control 3.11 ± 1.21 3.000–61470162.02

The majority of cells in the control sample were in second in vitro division (M2) at the time of analysis; moreover, the proportion of M3 cells was slightly higher than the proportion of M1 cells. The value of the PRI was 2.02 (table 4). In vitro administration of irinotecan significantly disturbed lymphocyte proliferation kinetics. An increase in the relative proportion of M1 indicates a delay in lymphocyte cell cycle, accompanied by decreases in the proportions of M2 and M3 cells and significant lowering of the PRI value in both irinotecan-treated samples (P < 0.05, χ2-test; table 4).

Cytokinesis-block micronucleus assay.

The induction of micronucleated lymphocytes by irinotecan is shown in table 5. Even at the lower dose, a significant increase in the MN frequency was observed. At this dose, irinotecan induced a frequency of 71 MN per 1000 BN lymphocytes (control 14 MN per 1000 BN lymphocytes). In contrast, exposure to a higher dose of irinotecan produced a frequency of 81 MN per 1000 BN cells. In the control sample, there were only 14 BN recorded and all of them contained only one MN (table 5). The majority of BN in the irinotecan-treated samples also contained only one MN. In both treated samples, a small proportion of BN with two MN was noticed, while BN with three MN was recorded only in a sample treated with a higher dose of the drug (table 5). Although a dose-dependent increase in MN frequency was observed, the differences were not statistically significant. Exposure to irinotecan also evoked formation of nuclear buds (NB) and nucleoplasmic bridges (NPB). In the treated samples, frequency of these types of damage was dose-dependent, but lower than the frequency of MN. In the control sample only NBs were recorded. CBMN assay takes into account all cells, both viable and non-viable, giving the better insight in actual events within the culture system. We observed that apoptosis plays an important role in the elimination of cells with DNA damage. While in the control sample only one apoptotic cell was detected, in treated samples a dose-dependent increase in frequency of apoptotic cells was observed (table 5). Both doses of irinotecan inhibited cell proliferation, compared to control. Moreover, exposure to the higher dose of irinotecan caused lower NDI compared to the value of the same index recorded in the sample treated with lower dose of irinotecan (table 5).

Table 5. 
Results of the cytokinesis-block micronucleus (CBMN) study of irinotecan toxicity on human peripheral lymphocytes treated for 2 hr in vitro with irinotecan (IRI).
TreatmentΣ MNed BNNo. of BN withΣMN Total No. ofLymphocyte proliferation
1MN2MN3MNNBNPBAPOM1M2M3M4NDI
  1. Two thousand binucleate cells (i.e. 4000 nuclei) were scored to determinate total number of micronuclei (MN) for each experimental point. MNed – micronucleated; BN – binuclear cells; NB – nuclear buds; NPB – nucleoplasmic bridges; NDI – nuclear division index; indicates the mean number of nuclei in all cells and is computed by the formula NDI = (M1+ 2M2 + 3M3 + 4M4)/N, with M1– M4 being the number of cells with 1–4 nuclei and N the number of screened cells. Two thousand cells per sample were screened for number of nuclei. significantly different as compared to control P < 0.05; χ2-test); significantly different as compared to sample treated with lower concentration of irinotecan.

IRI 9.0 µg/ml74685181375573501507108 641.958†,‡
IRI 4.6 µg/ml6865371252251461661104 892.068
Control141414 4 115215671531282.129

Measurement of AChE activity.

To further address the question of whether irinotecan acts as a specific AChE blocker, we performed an in vitro relevant experiment on human erythrocyte AChE. The results obtained in previous in vitro studies, suggest that this drug interacts directly with various molecular forms of AChE, leading to a non-competitive inhibition of this enzyme [22,23]. In our experiments, the inhibitory power (IC50) of irinotecan on AChE of human erythrocytes is determined and compared to carbamate physostigmine that is tested as reference drug. Namely, physostigmine has a strong inhibition potency to the same enzyme. Under identical conditions, changes in the enzymatic activity were detected when irinotecan or physostigmine were incubated with human erythrocyte AChE in the assay solution for 15 min. IC50 values were 5.0 × 10−7 for irinotecan and 2.0 × 10−8 mol/l for physostigmine (figs 3 and 4). The values indicate that irinotecan is an approximately 10 times less potent inhibitor than physostigmine. However, irinotecan was found to be strong inhibitor of the AChE hydrolysis of acetylcholine (ACh). In addition, a continuous decrease of catalytic activity of human erythrocyte AChE was obtained 10, 30, 60, 120, 150 and 180 min. after addition of a lower dose of irinotecan. A higher dose of irinotecan inhibited the AChE activity with potency similar to that estimated for physostigmine, and there was a slight reduction in the inhibition of AChE by both doses of irinotecan with increased incubation times (fig. 5).

Figure 3.

Inhibition of acetylcholinesterase (AChE) by irinotecan.

Figure 4.

Inhibition of acetylcholinesterase (AChE) by physostigmine.

Figure 5.

Progressive inhibition of acetylcholinesterase (AChE) by irinotecan and physostigmine.

Discussion

The present study reports for the first time the results of simultaneous application of alkaline and neutral comet assay on irinotecan-treated human lymphocytes. Previous study with related drugs, camptothecin and topotecan demonstrates the power of the alkaline comet assay to detect DNA damage in treated Chinese hamster ovary cells, as well as its repair after the drug removal [24]. Positive results obtained in our research indicate that in lymphocyte DNA after treatment with irinotecan a lot of strand breaks were induced. Dynamics of damage infliction as observed both in alkaline and neutral modifications of the comet assay during the post-incubation period clearly reflects the ‘poisoning’ of the topoisomerase I [25]. It is known that topoisomerase I binds to single-strand DNA breaks. The reversible Topo I-irinotecan-DNA cleavable complex is not lethal to the cells by itself. However, by its collision with the advancing replication forks, the formation of a double-strand DNA break occurs, leading to irreversible arrest of the replication fork and cell death [26]. The double-strand DNA breaks caused by the treatment are of special interest because they are more cytotoxic lesions and considered as major source of stabile chromosome aberrations and rearrangements [27]. Furthermore, their repair is much slower and more complicated as compared with single-strand DNA breaks [28]. It is known that the repair process itself also generates additional breaks (observed in this study as well), while a part of strand DNA breaks could be induced by indirect action of reactive oxygen species generated during the treatment. This assumption is also sustained by the results of an earlier study on camptothecin-treated Jurkat cells [29]. In the present study, a positive correlation between the results of the comet assay and the results obtained by quantitative determination of apoptotic and necrotic cells in the same samples was observed. The mortality of irinotecan-treated lymphocytes was primarily caused by apoptosis. Irinotecan was established earlier as an effective inducer of apoptosis [30]. It is assumed that the formation of DNA-protein complex stabilized by DNA topoisomerase I inhibitor ultimately signals the onset of apoptosis [31]. Moreover, at higher concentrations of irinotecan non-S-phase cells, besides apoptosis, could also be killed due to transcriptionally mediated DNA damage [32].

Conventional analysis of structural chromosome aberrations revealed a lot of chromatid breaks and complex quadriradial chromosome exchange figures in irinotecan-treated cells. These results are also in agreement with the reports of other researchers who investigated the effects of camptothecin and related drugs [33]. There are no published data on involvement of specific chromosomes in irinotecan-induced rearrangements yet. In the present study, three-colour painting probes for chromosomes 1, 2 and 4 were used. These chromosome pairs represent 22.34% of the DNA content of the human genome in a female and 22.70% in a male, as derived from Morton [34]. Thus, the labelling of chromosomes 1, 2 and 4 in cells from a female donor is expected to detect as paint/non-paint interchanges about 34.7% of the chromosomal interchanges occurring in the complete genome [35]. According to literature data [14], ‘one-way’ exchanges constitute typically 20–30% of the total exchange patterns. In our study, it appeared to be 33% at the highest dose of irinotecan applied. However, other authors using telomere probes have shown that very often, when the complete exchanges occurred, a terminal segment was so small that it does not register as a distinct visible signal [36,37]. This also must have been the case in our study because the number of ‘one-way’ translocations did no change with the concentration. For both doses of irinotecan tested, five ‘non-reciprocal’ translocations were detected. There were no significant differences in the frequency of involvement of specific chromosomes in translocations. However, at the higher irinotecan concentration, chromosome 4 was under-represented in the chromosome re-arrangements. Ganguly et al. [38] observed the same in X-irradiated human lymphocytes, whereas Braselmann et al. [39] showed over-involvement of chromosome 4 in reciprocal translocations. However, all published studies on involvement of specific chromosomes in their re-arrangements were done on irradiated cells. In the present study, a chemical agent, irinotecan, was tested. Due to its chemical structure, it may be possible that the drug exhibits certain specificity towards specific DNA regions on different chromosomes, which gives different results on translocations distribution from those obtained in irradiated lymphocytes. Recently, Anderson et al. [40] reported that more breaks than expected were observed on chromosome 2 indicating a deviation from randomness distribution. This finding is partially in agreement with our results. We detected an over-involvement of chromosomes 2 and 4 in acentric formation for the higher dose of irinotecan. However, when the results for involvement of the chromosomes 1, 2 and 4 in acentric formation and translocation are taken together, they appear to be in correlation with other studies [41,42]. They indicated that involvement of different chromosomes in re-arrangements was not proportional with their DNA content. Satoh et al. [43] using mitomycin C treated WTK-1 cell line deduced that chromosome 1 possesses significantly higher number of exchange sites than chromosomes 2 or 4. Thus, it is more frequently involved in quadriradials formation. In our study, only chromosome 1 was detected as the part of quadriradials. At tested concentrations of irinotecan, the frequency of quadriradial formation was too low to permit detection of quadriradials involving other two painted chromosomes with a single exchange site. Testing of higher concentrations of irinotecan for chromosomal aberration induction was not possible due to its cytotoxicity. Furthermore, Satoh et al. [43] deduced that asymmetrical quadriradials derive unstable chromosome-type aberrations (e.g. dicentrics and acentric fragments) and symmetrical quadriradials derive stable aberrations (e.g. reciprocal translocations). Because we detected involvement of chromosomes 2 and 4 in acentric fragments and reciprocal translocations, it could be presumed that some of those rearrangements were derived from quadriradials. Indirectly, it might suggest that also those two chromosomes are involved in chromatid exchange and quadriradials formation.

High rates of SCE caused by both doses of irinotecan are also indirect evidence of a large quantity of single- and double-strand DNA breaks produced by treatment. These results are in agreement with the reports of other authors [44–46]. Besides an increased SCE frequency, irinotecan caused a delay in lymphocyte proliferation in vitro. Because topoisomerase I is involved in the processes of chromosome segregation [47] as well as in cell transition from G0 to G1 phase and in DNA replication [46], disturbed functions of this enzyme lead to the retardation of the cell cycle following treatment. Previous studies with the irinotecan metabolite SN-38 on human glioblastoma cells indicate that cell cycle delays were induced by a decrease in the percentage of cells in the G0/G1 phase and an increase in the percentage of cells in S and G2/M phases [48].

Formation of MN after exposure to irinotecan can mirror clastogenic action of the drug tested. On the other hand, NPBs between nuclei in BN cells provide a measure of chromosome rearrangement [49]. The proportions of both MN and NPBs as detected in CBMN assay correlated positively with the types of chromosome aberrations recorded in the same blood samples. A high proportion of NBs observed after treatment with both doses of irinotecan indicate the possibility of gene amplification [49]. To clearly explain the origin of NBs induced after exposure to irinotecan, further studies are needed. The results obtained by CBMN assay provided comparable results with a fluorescence dye exclusion test and confirmed that apoptosis plays an important role in the elimination of cells with DNA damage.

The results of our study indicate that irinotecan as the parent drug also induces remarkable toxic effects in human peripheral blood lymphocytes treated in vitro, both on the subcellular and the cellular level. The same was observed by Pavilliard et al. [50] in their in vitro study with two human colorectal tumour cell lines. Therefore, irinotecan is not totally devoid of activity, which was concluded in some earlier studies. The actual levels of DNA damage produced both in cancer and non-target cells in human evidently are much higher due to the presence of SN-38, a metabolite with an augmented potency and cytotoxicity [26]. While the present study was performed in vitro, and the blood cells were exposed to irinotecan in a serum-free medium, there was no possibility for extensive conversion of the parent drug to SN-38, although some studies indicate that human plasma esterases, especially butyrylcholinesterase, present in serum also possess irinotecan-activating activity [51,52].

In our study, the activity of erythrocyte AChE after in vitro exposure to irinotecan was studied simultaneously with other end-points. The best-characterized function of AChE is the hydrolysis of ACh at cholinergic synapses [53,54]. Inhibition of its enzyme leads to accumulation of ACh resulting in over-stimulation of the whole cholinergic system. Muscular and nerve AChE are only present in the synaptic cleft and cannot be measured directly. Because erythrocyte AChE has a similar structure as the synaptic enzyme, it appears to be a suitable parameter to reflect the various reactions at the synaptic site. Therefore, its measurement is of important for therapy management, especially during the course of the intoxication with different chemicals or drugs that inhibits the activity of the enzyme. Clinical studies indicate the possibility of acute cholinergic side effects after intravenous administration of irinotecan [55–57], which, on the basis of in vitro experimental findings, have been ascribed to a direct blockade of AChE [22,23]. In this study, an attempt was made to establish whether impairment in the activity of AChE occurs in whole-blood samples previously treated with irinotecan. All our results indicate high affinity of irinotecan for AChE in vitro. The enzymatic activity decreased up to 50% when the enzyme was exposed to irinotecan at a concentration of 5.0 × 10−7 mol/l. These findings were consistent with the results obtained in previous in vitro studies, suggesting that this drug acts as potent inhibitor of AChE [22,23]. Furthermore, it is noteworthy that physostigmine, a well-known AChE blocker [58], inhibited the activity of AChE with a potency similar to that estimated for the same drug in previous studies [59,60]. Moreover, there is evidence that human erythrocyte AChE activity inhibited by irinotecan in doses comparable to those recommended in mono- or combination therapy correlates with the activity of the same enzyme after physostigmine administration. Thus, irinotecan may provide an interesting lead compound for the investigation of the inhibition of AChE. Judging from experimental in vitro data, we assume that measurement of AChE activity in vivo could be an appropriate method with possible implementation in control and management of acute cholinergic syndrome. However, it has to be further investigated. In addition, the current evidence is insufficient to indicate whether atropine sulfate is beneficial in the management of this syndrome, that is, whether muscarinic symptoms such as nausea, vomiting, diarrhoea and bowel movements can be stopped by administration of atropine sulfate. Future in vivo research will help to determine whether administration of atropine or functionally related compounds has a potential in medical treatment in patients or not.

Despite of the limitations, the results of the present study may contribute to the understanding of irinotecan toxicity and constitute a helpful step to recognize the side effects of the drug, even though the consequences of in vivo drug treatment may not be completely like those occurring in a living organism, they may indicate the existence of certain mutations in non-target cells with cancer predictive value. Using a battery of end-points in surrogate cells, we have obtained an insight into the levels of the cyto- and genotoxicity, as well as the inhibitory potency of irinotecan against AChE. However, further in vitro and in vivo studies are essential in order to clarify remaining issues, especially to elucidate possible inter-individual variability in genotoxic responses to the drug.

Ancillary