Apoptosis Induction and Micronucleus Formation after Exposure to the ...

ORIGINAL ARTICLE

Apoptosis Induction and Micronucleus Formation after Exposure to the Auger Electron Emitter Zinc-65 in a Human Cell Line Ralf Kriehuber and Myrtill Simko´ From the Institute of Cell Biology and Biosystem Technology, Department of Biology, University of Rostock, Germany Correspondence to: Dr R. Kriehuber, Institute of Cell Biology and Biosystem Technology, Department of Biology, University of Rostock, Universita¨tsplatz 2, DE-18055 Rostock, Germany. Tel: » 49 381 49 81 933. Fax: » 49 381 49 81 918. E-mail: [email protected]

Acta Oncologica Vol. 39, No. 6, pp. 699± 706, 2000 Induction of apoptosis and micronucleus formation has been studied in a transformed human squamous cell carcinoma cell line (SCL-II) after exposure to the Auger electron emitter Zinc-65 (65Zn) and after external low-LET radiation. Exposure to non-radioactive Zn and unirradiated cells served as controls. Studies on the cellular uptake of 65Zn2» have been carried out in vitro and conventional dosimetric models have been applied to derive the absorbed radiation dose. Auger electrons, generated during decay of 65Zn2» , are strong inducers of micronuclei as well as of apoptosis, in comparison to external low-LET irradiation. The relative biological effectiveness has been determined and was found to be in the range of 1.2± 2.7 for the two investigated biological endpoints, depending on which mathematical model for describing the dose-effect curves was used. A non-uniform distribution of intracellular Zn2» was observed, showing a strong signal in the perinuclear region. We conclude that separate radiation weighting factors for Auger electrons should be established depending on the nuclide and its ability to interact with the DNA (e.g. 65Zn by Zinc-® nger proteins). Recei×ed 29 September 1999 Accepted 15 March 2000

One of the most important tasks in radiation protection and dosimetry is to derive a broad database on the biological effectiveness of Auger electrons as well as establishing radiation weighting factors for these electrons (1). Auger electrons are produced by non-radiative (NR) Auger-, Coster-Kronig- and Super-Coster-Kronig processes resulting from inner shell ionizations caused by photoelectric effects, nuclear decay by orbital electron capture and internal conversion (2). Each NR transition ® lling an electron vacancy creates two new vacancies in higher atomic subshells. This vacancy multiplication, and the progressively decreasing transition energies result in a dense shower of low-energy electrons in the immediate vicinity of the Auger emitter (3). These low-energy electrons are characterized by a very short range in biological matter, thus leading to a high-energy deposition in a rather small volume. This is thought to cause severe biological damage, especially when the Auger emitter is located close to the DNA. However, no radiation weighting factor for Auger electrons has yet been established, and in current dosimetry models, Auger electrons are considered as highenergy electrons and are allocated a quality factor of one. © Taylor & Francis 2000. ISSN 0284-186X

There are several Auger electron emitters used in nuclear medicine, such as 51Cr, 55Fe, 67Ga, 75Se, 77Br, 80mBr, 99mTc, 110 In, 111In, 114mIn, 123I, 125I, 145Sm, 193mPt, 195mPt, and 201 Tl (4, 5), but many of them, including 65Zn, are also common in nuclear industry. However, most of the data are obtained for 125I, which can be integrated into the DNA as 125I-iododeoxyuridine (IUdR), a thymidine analogue (6, 7). The enormous radiotoxic capability of 125 IUdR could be shown in vivo and in vitro and is well in the range of the cytotoxicty of 5.3 MeV alpha-particles. But 125I-labelled DNA binding agents other than IUdR have also been shown to cause considerable damage to the DNA (8). However, the capability of the Auger emitter to traverse the cell membrane and, more essentially, to enter the nucleus and to come into the close vicinity of the DNA is supremely important. Zinc is not only an important trace element, it is also the stabilizing atom of a protein motif, which is known to interact with DNA. Many DNA-binding proteins, such as transcription factors, contain several repeats of such protein motifs, which are called Zinc ® ngers (9). Zinc ® ngers interact with the major grove of the DNA and the stabilizing Zinc atom is then as close as 5 nm to the DNA. Acta Oncologica

700

Zinc is also known to be actively taken up by cells, to be co-factor of more than 300 enzymes and to be involved in other physiological important processes as well. However, Zinc is incorporated into Zinc ® nger motifs during protein synthesis. These proteins are guided to the cell nucleus and come into a close contact with the DNA, presumably with transcriptional active DNA. The special physiological role of Zinc and the fact that the Auger emitter 65Zn is a frequently used element in the nuclear industry makes 65Zn an appropriate and interesting candidate to study the effects of Auger electrons. Furthermore, little is known about the radiotoxicity of 65 Zn. Bingham et al. (10) discussed whether occupational exposure to 65Zn could explain the excess of prostate cancer reported amongst some nuclear workers, because it is known that the dorsolateral prostate contains high amounts of free zinc ions (11). Since studies on the cellular level revealed preferential uptake in the epithelial cells of the prostate, we used a human cell line of epithelial origin to investigate genotoxic effects employing the micronucleus assay and cytotoxic effects by studying the induction of apoptosis in order to determine the relative biological effectiveness (RBE) of Auger electrons emitted by 65Zn in comparison to external, low-LET radiation. MATERIALS AND METHODS Cell culture Human squamous cell carcinoma cell-line (SCL-II) cells were maintained in Eagle’s Minimum Essential Medium with Earle’s salt (MEM; Biochrom, Germany), supplemented with 20% fetal calf serum (FCS, Biochrom, Germany) and grown in plastic culture ¯ asks at 37°C in an atmosphere of 5% CO2. If necessary 1% penicillin:streptomycin (10.000 E: 10.000 mg:ml; Gibco, UK) was added. Viable cells were de® ned by their exclusion of Trypan blue (0.5 g Trypan blue in a 0.9% sodiumchloride solution) and cell numbers were microscopically determined. 65

Zn exposure, x-irradiation and preparation of samples

65

Acta Oncologica 39 (2000)

R. Kriehuber, M. SimkoÂ

Zn (Zinc-65 ZAS.2, Amersham, UK) was added to the cell culture medium of con¯ uent grown SCL-II cells to give a ® nal concentration of 65 kBq± 500 kBq:ml. After 24 h, the cells were extensively washed in phosphate buffered saline (PBS), trypsinized, transferred to a freezing medium (FCS, 10% DMSO) and stored in liquid nitrogen for accumulation of decays (5 d± 100 d). Unexposed cells were treated the same way and served as controls. External irradiation was performed with a LINEACunit (Neptun 16A), operated at 7.2 MeV at a dose rate of 1 Gy:min; 24 h prior irradiation cells were plated at a density of 1× 106 cells per ¯ ask (75 cm2). Controls were sham-irradiated. Cells were then washed in PBS, trypsinized, transferred to a freezing medium and stored in liquid nitrogen for the same length of time as 65Zn-treated cells.

For examination, the cells were thawed, washed in cell culture medium, and plated on cover slips, placed in six well plates, and cultured for 30 h. Apoptosis assay Thirty hours after replating, the cover slips were washed in pre-warmed (37°C) PBS and ® xed with ¼ 20°C methanol for at least 10 min. Air-dried cover slips were washed in PBS and stained for 1 min with the ¯ uorescent DNA dye bisbenzimide (Hoechst 33258; Sigma, Germany; 1 mg:ml). After thorough washing in distilled water, the cover slips were mounted for ¯ uorescence microscopy and evaluated for the presence of apoptotic and mitotic cells. Apoptotic cells were determined by at least two of the following morphological criteria: chromatin condensation, nuclear membrane blebbing and:or apoptotic bodies (12). The rate of apoptotic cells (APO) and the mitotic index (MI) were calculated by dividing the number of both the apoptotic cells and mitotic cells by the total number of scored cells. MI was used to ensure the proliferative capacity of the investigated cells (data not shown). At least 18000 cells were scored for each data point. Micronucleus assay Cells were grown on cover slips in six well plates. Cytochalasin-B (Cyt-B, Sigma, Germany) was made up to a stock solution (1 mg:ml) in dimethyl sulfoxide (DMSO, Aldrich, Germany) and added to the cell cultures 6 h after replating the cells, to give a ® nal concentration of 1.2 mg:ml. Twenty-four hours after adding Cyt-B, the medium was discarded, the cells were washed twice in prewarmed PBS and ® xed in methanol (¼ 20°C) for 15 min. Air-dried cover slips were stained in a bisbenzimide (1 mg:ml):ethidiumbromide (1 mg:ml) solution and embedded in DABCO antifading solution (1 part 0.2 M Tris-HCl, 0.02% NaN3 plus 9 parts 2.3% (w:v) 1.4 Diazabicyclo[2.2.2.]-octane in glycine, pH 8.0) or stained in a Giemsa staining solution and embedded in Entelan (Merck, Germany). For the identi® cation of micronuclei, the criteria of Countryman & Heddle (13) were applied: diameter of less than 1:3 of the main nucleus; staining the same as the main nucleus; location within 3± 4 nuclear diameters of the nucleus, but not touching the nucleus. Scored cells were divided into 7 classes: cells with no visible micronucleus (MN), cells containing 1 MN, 2 MN up to cells containing 6 MN. Micronucleus frequencies were calculated by dividing the total number of micronuclei by the number of scored cells. At least 3000 binucleated cells were scored for the presence of micronuclei per time point and dose point in each experiment. Cellular uptake of

65

Zn and dose calculation

Cellular uptake was measured using liquid scintillation counting. Prior to the measurement, cell number was determined, cells were centrifuged and cell pellet size was


Apoptosis and micronucleus induction after exposure to Auger electrons

determined by volume and mass analysis. To assay for uniform uptake, exposed cells were plated at very low density on glass slides, cultured for 24 h, ® xed with ethanol and counted in a micro-imager (Biospace Mesures, France) for 24 h at 0.2 mm spatial resolution and 50 single cells were analysed. Accumulated decays were calculated for 24 h cell culture time (Ao ¾ 1.3 MBq) and for storage in liquid nitrogen (A1 ¾ 0.46± 0.72 kBq:mm3 cell pellet). Dose calculation during cell culture time is based on the assumption that the b » -component (energy× yield per decay: 4.78× 10 ¼ 3 MeV) and the Auger:conversion electron component (energy× yield: 2.09× 10 ¼ 3 MeV) are totally absorbed in the target volume (20 ml) and that 65Zn is uniformly distributed in the cells and cell culture medium. It is further assumed that the cells are uniformly distributed in the freezing medium during storage in liquid nitrogen and that the b » -component is totally absorbed in the target volume (1.5 ml freezing medium), whereas the Auger:conversion electron component is totally absorbed in the cell volume only, which is assumed to be the cell pellet size. The g-component is neglected, because of its low contribution ( B 1%) to the total absorbed dose. Intracellular zinc distribution SCL-II cells were grown on cover slips mounted in a Bionique chamber (35 mm, Bionique Laboratories, Saranac Lake, NY, USA), washed twice in pre-warmed PBS for 2 min and loaded for 30 min with the ¯ uorescent Zn2» probes Zinquin (25 mM; Alexis Corporation) or TSQ (50 mM; Te¯ ab). The cells were subsequently washed in PBS and analysed in MEM medium using a ¯ uorescence microscope equipped with a PlanApo 60:1.4 oil objective at 37°C. Data were analysed by comparing ¯ uorescence emissions in equal volumes of the cytoplasm and the cell nucleus using NIH-imaging software.

701

(RBEtestloe¾ Dref loe:Dtest loe). The RBE of 65Zn exposure was determined for the induction of apoptosis and MN induction in SCL-II cells and the LINEAC set was used as reference radiation. Linear, as well as linear-quadratic dose-response relationships were used for mathematical description of the data. For non-linear description of the data, RBEs were determined for two levels of damage: effect. Data were analysed versus unirradiated control data at a signi® cance level of pB 0.01, using a one-sided Student’s t-test for matched samples. RESULTS Cellular uptake and dose calculation of

65

Zn

SCL-II cells were shown to accumulate 65Zn during cell culture time. An 8.39 2.4-fold accumulation of 65Zn was found when compared to the concentration in the cell culture medium. A good correlation between cell pellet size and total activity could be demonstrated (data not shown). The measurement of single SCL-II cells after exposure to 1.3 MBq 65Zn for 24 h, using the micro-imager, indicated that the cells are uniformly labelled. The comparison between the calculated average of activity per cell, based on the scintillation data (6.2× 10 ¼4 9 1.9 × 10 ¼ 4 Bq), showed good correlation with the data of the micro-imager with an average of the measured activity per cell of 8.1 × 10 ¼4 9 3.2 × 10 ¼ 4 Bq. Comparing the dose contribution from the time in cell culture and the storage in liquid nitrogen as well as the in¯ uence of the different components (b » -component and Auger:conversion electron component) demonstrated that during storage in liquid nitrogen the Auger:conversion electron component contributes more than 98% to the total dose at longer storage times (\ 20 days) and about 95± 98% at shorter exposure times of 5 ± 20 days (Table 1). Induction of apoptosis

Calculation of RBE and statistical analysis The RBE of a radiation type is de® ned as the quotient of the dose of a reference radiation (Dref) to produce a certain level of effect (loe) and the dose of the radiation quality (Dtest) of interest to produce the same level of effect

Apoptosis was induced in SCL-II cells by both external irradiation and exposure to 65Zn (Fig. 1a). In general, apoptotic cells were rare and their frequency was constantly very low in the unirradiated control cells ( B 0.1%). Prolonged storage in liquid nitrogen had no effect on the

Table 1 Contribution of the different dose-components and exposure conditions to the total absorbed dose in SCL -II cells for different times of exposure to 65 Zn. E×en at short term exposure the Auger:ce component contributes up to 95% to the calculated total absorbed dose. At longer exposure times more than 98% of the total absorbed dose is contributed by the Auger:ce-component Exposure time (days)

5 20

Dose contribution during cell culture (Gy)

Dose contribution during storage in liquid nitrogen (Gy)

b»-component and Auger:ce-component

b»-component

Auger:ce-component

0.006 ($ 4%) 0.006 (B1%)

0.001 (B1%) 0.005 (B1%)

0.137 (\95%) 0.638 ( $98%)

Total dose (Gy)

0.145 0.65

702



Fig. 1. Cytotoxic effects in SCL-II cells after external x-irradiation ( ) and after exposure to 65Zn ( ), and different mathematical descriptions of dose-effect relationship as a function of the absorbed dose D. a) Induction of apoptosis (apoptotic cells %); each data point represents at least 18000 scored cells. b) Linear model of dose-effect relationship; data after x-irradiation are described best by apoptotic cells (%) ¾0.08 D » 0.1; R2 ¾0.36, whereas data after 65Zn exposure are described best by apoptotic cells (%)¾ 0.22 D » 0.1; R2 ¾0.35. c) Linear-quadratic model of dose-effect relationship; data after x-irradiation are described best by apoptotic cells (%)¾ ¼ 0.034 D2 »0.302 D »0.1, R2 ¾ 0.39 and data after 65Zn exposure are described best by apoptotic cells (%)¾ ¼ 0.0108 D2 » 0.353 D » 0.1, R2 ¾ 0.45.

rate of apoptotic cells in controls. External irradiation induced apoptosis in SCL-II cells at doses as low as 0.5 Gy. At a dose of 1 Gy up to 4 Gy a plateau of induction could be observed (0.4± 0.9%). A further increase could not be achieved at reasonable doses (8 Gy). Exposure to 65 Zn resulted in an induction of apoptosis, starting at doses as low as 0.6 Gy (0.5%). The rate of apoptotic cells increased much more prominently up to a dose of 15 Gy (3.5%). Although data showed enormous scattering at two dose points, a similar pronounced induction rate after external irradiation could not be reached. Data of apoptosis induction were ® tted using a linear model (Fig. 1b) and a linear-quadratic model (Fig. 1c) of dose-effect relationship. The linear-quadratic model ® tted the data better, although the data points turned over at relatively low doses (D¾ 4 Gy) after external irradiation. The linear ® t of the data with apoptotic cells (%)¾ 0.08 D » 0.1 for external radiation and with apoptotic cells (%)¾ 0.22 D » 0.1 for exposure to 65Zn led to an RBE of 2.7. The linear-quadratic ® t of the data with apoptotic cells (%)¾ ¼ 0.034 D2 » 0.302 D» 0.1 for external radiation

and with apoptotic cells (%)¾ ¼ 0.010 D2 » 0.351 D » 0.1 for exposure to 65Zn resulted in an RBE of 1.2 at the chosen level of effect of 0.25% apoptotic cells and an RBE of 1.4 at the chosen level of effect of 0.5% of apoptotic cells. Micronucleus induction MN was induced in SCL-II cells after external irradiation and also after exposure to 65Zn (Fig. 2a, b). Unirradiated controls showed a MN frequency of 149 6 MN per 1000 binucleated cells and storage in liquid nitrogen up to 100 days had no effect on the MN frequency in control cells. The induction of MN after external irradiation was described best by a linear dose-effect relationship. A signi® cant (p B 0.01) increase of MN frequency in binucleated SCL-II cells could be observed at a dose as low as 0.25 Gy (419 16 MN:1000 cells). Exposure to 65Zn resulted in a much stronger induction of MN when compared to external irradiation. A signi® cant increase of MN, when compared with own control (pB 0.01), could be stated at doses as low as 0.2 Gy



(639 13 MN:1 000 cells). A large scattering of the data was observed owing to cells containing multiple micronuclei (more than 4 MN). The peak of MN induction could be observed at a dose of 1.6 Gy, followed by a decline in MN frequency. In the dose range above 2 Gy only low frequencies of binucleated cells (B 5%) could be obtained, few mitotic cells could be observed and the morphology of cells was altered (strong vacuolization in the perinuclear region), indicating heavy cytotoxic damage. However the few observed binucleated cells were scored for micronuclei but neglected for the mathematical description of the dose response after 65Zn exposure. Therefore the mathematical description of the MN dose response is ® tted only to data in the dose range of 0 ± 2 Gy. Data of MN induction after external irradiation were ® tted using a linear model (Fig. 2c) and a linear-quadratic model (Fig. 2d) of dose-effect relationship. The linearquadratic model described the 65Zn data best (MN:1 000 cells¾ 145.5 D2 » 95.12 D » 14), followed by the linear model (MN:1000 cells¾ 251.13 D» 14). However, the

703

data are not very well described by any models used, due to the large scattering of the data. The MN data after external radiation were very good described by either a linear (MN:1 000 cells¾ 123.41 D» 14) or a linearquadratic model (MN:1000 cells¾ ¼ 0.6426 D2 » 127.81 D » 14) of dose-effect relationship. The linear ® t of the data after external irradiation and after exposure to 65 Zn led to an RBE of 2. The linear-quadratic ® t of both data sets resulted in an RBE of 1.7 at the chosen level of effect at 20% MN induction, and an RBE of 2.3 at a chosen level of effect at 40% MN induction in binucleated cells (Table 2). Intracellular Zn 2 » localization Most of the intracellular Zn2» in SCL-II cells was found to be located in the perinuclear region (Fig. 3). The cell nucleus showed only a weak ¯ uorescence signal when compared to the overall signal intensity in the cytoplasm. The analysis of the signal intensity in equal volumes of the cytoplasm and cell nucleus, revealed a three times higher

Fig. 2. Genotoxic effects in SCL-II cells after external x-irradiation ( ) and after exposure to 65Zn ( ), and different mathematical descriptions of dose-effect relationship as a function of the absorbed dose D. a) Micronucleus formation (micronuclei:1 000 cells) in binucleated SCL-II cells, for each data point at least 9 000 cells were scored. b) Micronucleus data of the low dose range (shown for better graphical illustration). c) Linear model of dose-effect relationship; data after x-irradiation are described best by MN:1 000 cells ¾123.41 D » 14, R2 ¾0.98, data after 65Zn exposure are described best by MN:1 000 cells ¾251.13 D » 14, R2 ¾0.63. For mathematical description of MN induction, only data in the dose range of 0 ± 2 Gy were considered. d) Linear-quadratic model of dose-effect relationship; data after x-irradiation can be described best by MN:1000 cells¾ ¼0.6426 D2 »127.81 D » 14, R2 ¾0.98, and data after 65Zn exposure are described best by MN:1 000 cells¾ 145.5 D2 »95.12 D »14, R2 ¾0.74.

704



Table 2 Relati×e biological effecti×eness (RBE) for Auger electrons, generated during decay of 65Zn, for induction of apoptosis (APO) and micronucleus (MN) formation in SCL -II cells, when compared to an external low-LET reference radiation (x-rays). For linear-quadratic modelling of dose-effect relationship, the RBEs were calculated for two le×els of effect. Dose response for MN formation after 65Zn exposure was ® tted only to data in the dose range 0± 2 Gy Biological endpoint

Mathematical model

APO APO

Linear Linear-quadratic

APO

Linear-quadratic

MN MN MN

Linear Linear-quadratic Linear-quadratic

Level of effect

0.25% Apoptotic cells 0.5% Apoptotic cells 20% micronuclei 40% micronuclei

RBE

2.7 1.2 1.4 2 1.7 2.3

signal intensity in the cytoplasm (759 8.4). It can be deduced from the distribution and localization data of Zn2» that most of the intracellular 65Zn is located within the cytoplasm. Owing to the fact that more than 95% of

Fig. 3. SCL-II cells stained with Zinquin. The cell nuclei (arrows) show a weak ¯ uorescence signal.

the calculated absorbed dose is contributed by short-ranging Auger electrons (average range of a few nanometers), it can be assumed, that the delivered energy to the cell nucleus is largely overestimated by the performed dose calculations.

DISCUSSION The accumulation of intracellular Zn2 » in the perinuclear region of SCL-II cells is in good agreement with the observed distribution of Zn2» in murine NIH 3T3 ® broblasts described by Back et al. (14). The observed low concentration of Zn2» in the cell nucleus corresponds also to the low amounts of Zn2 » detected in the nuclei of pancreatic islet cells, lymphocytes, neutrophils and ® broblasts, as has been reported by Zalewski et al. (15). It seems that the overall intracellular distribution pattern of Zn2» does not vary greatly in different cell types, re¯ ecting the important physiological role of Zn2» , especially in DNArelated processes. However, more information about the ultrastructural localization has to be gained. Cytoplasmatic metallothionein (MT) seems to be the most important intracellular Zn2» pool, and therefore the cellular redox state as well as the concentration of other biological chelating agents might well determine the direction of Zn2» transfer and ultimately affect the intracellular Zn2» distribution (16). Most of the intracellular zinc is stored in the cytoplasm and might explain the rather high amount of free Zn2 » located in the cytoplasm. The accumulation of 65Zn2» in SCL-II cells is comparable to data gained in primary cultured rat hepatocytes, where a 5-fold increase after 24 h of 65Zn exposure was observed (17). The Auger electron emitter 65Zn caused substantial genotoxic and cytotoxic effects in human SCL-II cells in vitro, detected through micronucleus formation and induction of apoptosis. However, the Auger electron emitter, which can be incorporated into the DNA, such as 125I as 125 Iododesoxyuridine, shows somewhat higher relative biological effectiveness (RBE) for different biological endpoints (18± 21). But it has to be kept in mind that 65Zn is primarily associated with the DNA via Zinc-® nger motifs and it can be argued that only a small portion of the available 65Zn is actually in close contact with the DNA. Furthermore, the determined RBEs are based on a dose calculation for the whole cell, assuming a uniform intracellular 65Zn distribution and therefore make no allowances for intracellular Zn2» distribution pattern. Since most intracellular Zn2» is associated with the cytoplasm, this approach overestimates the absorbed energy in the cell nucleus, which is known to be the critical target for radiation-induced genotoxic effects (22). However, single Zinc-® nger proteins contain several repeats of the Zinc® nger motif and this could possibly lead to multiply damaged sites, especially when decays are accumulated for several weeks and no protein turnover occurs under the



used experimental design. Multiply damaged sites are thought to be the critical damage caused after high LET radiation and frequently become transformed into chromatin breaks (23). Recent experiments with 125I also suggest that more than one DNA double-strand break per decay might be realistic (24). Either of these suggestions would adequately explain the strong induction of micronuclei at rather low doses, with only few decays in the close vicinity of the DNA, including the frequently observed cells containing multiple micronuclei. The observed decrease of MN at doses \ 2 Gy re¯ ects the observed strong cytotoxicity of 65Zn in the colony-forming assay (data not shown). The few observed binucleated cells contained only few micronuclei, representing the portion of undamaged cells, indicating at least the strong mitotic inhibition of SCL-II cells exposed to 65Zn. Zn2» seems to be of importance also during apoptosis. Zn2» is mobilized during the early state of apoptotic cell death and is even reported to suppress apoptosis (25, 26). This might explain the observed suppressive effect of Zn2» on the apoptosis frequency in control cells, well below the frequency observed in unlabelled SCL-II cells (data not shown). However, apoptosis is strongly induced in SCL-II cells following exposure to 65Zn, when compared with uniform external x-irradiation. In fact, the observed overall amount of apoptosis after exposure to 65Zn could not be reached by external radiation, which raises the question of a special mechanism of apoptosis induction in SCL-II cells by 65Zn. Different mechanisms of inducing apoptotic cell death, not requiring DNA damage, are known (27). The preferred localization of 65Zn in the perinuclear region, and subsequently, the high-energy deposition in this subcellular structure, might induce apoptosis by damaging non-DNA structures. Further research is required to identify the nature of this mechanism. The observed scattering of the data may be due to the fact, that only a few decays in the near vicinity of the DNA cause strong biological effects, such as multiple micronucleus induction. Small ¯ uctuations of the intracellular distribution of Zn2 » , as a result of the physiological status of the cells, might therefore cause dramatic changes in the investigated biological effect. It could be shown that 65Zn2» is actively taken up by the studied cell line and develops a non-uniform intracellular distribution pattern. Despite that only a fraction of the 65 Zn2» enters the cell nucleus, prominent biological effects could be observed, when compared to external low-LET radiation. The Auger and conversion electrons contribute more than 95% of the calculated dose, indicating that the observed severe damage at rather low doses is caused by the emitted short-ranging, low-energy Auger electrons. RBEs have been derived for the investigated endpoints and range between 1.2 and 2.7, if a uniform intracellular Zn2» distribution is assumed for the dose calculations, and are in good agreement with reported RBEs for other Auger

705

electron emitters capable of entering the cell nucleus. The increased RBE in relation to low-LET radiation, the active uptake of 65Zn2» , including the capability to enter the nucleus, as well as the already achieved basic data on intracellular distribution of Zn2 » and on cell morphology of the investigated cell line represent an interesting data set to be further used for re® ned dose calculations and microdosimetric modelling. Our results indicate that in vitro studies on micronucleus formation and induction of apoptosis in a well-de® ned cell line are appropriate tools to assess quantitatively radiation-induced biological damage (28), even in the low dose range, and are suitable to study the biological effectiveness of Auger electrons. More information on the intracellular distribution of the investigated Auger electron emitting nuclides is desirable in order to understand the induced complex biological response and to implement the data in an applicable and reliable dosimetry system for Auger emitters. ACKNOWLEDGEMENTS We thank Professor H.-A. Schulze (Head of the Experimentelles Forschungszentrum der Medizinischen FakultaÈt, UniversitaÈt Rostock) for enabling the scintillation measurements, Professor Filthut (Biospace Measure) for enabling the micro-imager measurements and the staff of the LINEAC radiotherapy unit (Radiologische Klinik und Poliklinik, UniversitaÈt Rostock) for cooperation during the external irradiation experiments. This work was supported by the 4th framework programme of the Commission of the European Community (FI4P-CT95-0011).

REFERENCES 1. Persson L. The Auger electron effect in radiation dosimetry. Health Phys 1994; 67: 471± 6. 2. Sastry KSR, Howell RW, Rao DV, et al. Dosimetry of Auger emitters: physical and phenomenological approaches. In: Baverstock KF, Charlton DE, eds. DNA damage by Auger emitters. London: Taylor & Francis, 1988: 27± 38. 3. Howell RW, Narra VR, Sastry KSR, Rao DV. On the equivalent dose for Auger electron emitter. Radiat Res 1993; 134: 71± 8. 4. Baverstock KF, Charlton DE, eds. DNA damage by Auger emitters. Report of a workshop on 17th July 1987. London: Taylor & Francis, 1988. 5. Hofer KG. Symposium reportÐ biophysical aspects of Auger processes. Int J Radiat Biol 1992; 61: 289± 92. 6. Narra VR, Howell RW, Thanki KH, Rao DV. Radiotoxicity of 125I-iododeoxyuridine in pre-implantation mouse embryos. Int J Radiat Biol 1991; 60: 525± 32. 7. Rao DV, Narra VR, Howell RW, Govelitz GF, Sastry KSR. In-vitro radiotoxicity of DNA-incorporated 125I compared with that of densely ionising alpha-particles. Lancet II 1989; 650± 53. 8. Ludwikow G, Ludwikow F, Johnson KJ. Kinetics of micronucleus induction by 125I-labelled thyroid hormone in hormone-responsive cells. Int J Radiat Biol 1992; 61: 639± 53. 9. Klug A, Rhodes D. Zinc ® ngers: a novel protein motif for nucleic acid recognition. Cold Spring Harb Symp Quant Biol 1987; 52: 473± 82.

706



10. Bingham D, Harrison JD, Phipps AW. Biokinetics and dosimetry of chromium, cobalt, hydrogen, iron and zinc radionuclides in male reproductive tissues of the rat. Int J Radiat Biol 1997; 72: 235± 48. 11. Sorensen MB, Stoltenberg M, Juhl S, Danscher G, Ernst E. Ultrastructural localization of zinc ions in the rat prostate: an autometallographic study. Prostate 1997; 31: 125± 30. 12. Dini L, Coppola S, Ruzittu MT, Ghibelli L. Multiple pathways for apoptotic nuclear fragmentation. Exp Cell Res 1996; 15, 223: 340± 7. 13. Countryman PI, Heddle JA. The production of micronuclei from chromosome aberrations in irradiated cultures of human lymphocytes. Mutat Res 1976; 41: 321± 32. 14. Back CJ, Sistonen L, Enkvist MOK, HeikkilaÈ JE, Akerman KEO. Ca2» and Zn2» dependence of DNA synthesis in untransformed and in Ha-rasval-12-expressing NIH 3T3 cells. Exp Cell Res 1993; 208: 303± 10. 15. Zalewski PD, Millard SH, Forbes IJ, et al. Video image analysis of labile zinc in viable pancreatic islet cells using a speci® c ¯ uorescent probe for zinc. J Histochem Cytochem 1994; 42: 877± 84. 16. Jacob C, Maret W, Vallee BL. Control of zinc transfer between thionein, metallothionein, and zinc proteins. Proc Natl Acad Sci USA 1998; 95: 3489± 94. 17. Steinebach OM, Wolterbeek HT. Effects of zinc on rat hepatoma HTC cells and primary cultured rat hepatocytes. Toxicol Appl Pharmacol 1993; 118: 245± 54. 18. Hofer KG, Harris CR, Smith JM. Radiotoxicity of intracellular 67Ga, 125I and 3H. Nuclear versus cytoplasmic radiation effects in murine L1210 leukemia. Int J Radiat Biol 1975; 28: 225± 41.

.

19. Kassis AI, Fayad F, Kinsey BM, Sastry KSR, Taube RA, Adelstein SJ. Radiotoxicity of 125I in mammalian cells. Radiat Res 1987; 111: 305± 18. 20. Makrigiorgos GM, Adelstein SJ, Kassis AI. Auger electron emitters: insight gained from in vitro experiments. Radiat Environ Biophys 1990; 29: 75± 91. 21. Hofer KG, Bao SP. Low-LET and high-LET radiation action of 125I decays in DNA: effect of cysteamine on micronucleus formation and cell killing. Radiat Res 1995; 141: 183± 92. 22. Warters RL, Hofer KG, Harris CR. Radionuclide toxicity in cultured mammalian cells: elucidation of the primary site of radiation damage. Curr Top Radiat Res Q 1977; 12: 398± 407. 23. Rydberg B. Clusters of DNA damage induced by ionizing radiation: formation of short DNA fragments. II. Experimental detection. Radiat Res 1996; 145: 200± 9. 24. Walicka MA, Adelstein SJ, Kassis AI. Indirect mechanisms contribute to biological effects produced by decay of DNAincorporated iodine-125 in mammalian cells in vitro: doublestrand breaks. Radiat Res 1998; 149: 134± 41. 25. Zalewski PD, Forbes IJ, Seamark RF, et al. Flux of intracellular labile zinc during apoptosis (gene-directed cell death) revealed by a speci® c chemical probe, Zinquin. Chem Biol 1994; 1: 153± 61. 26. Mathieu J, Ferlat S, Ferrand D, et al. DNA fragmentation induced in lymphocytes by gamma irradiation or dexamethasone: inhibition by diethyldithiocarbamate (DTC), potentiated by zinc. Biochem Mol Biol Int 1995; 36: 733± 44. 27. Raff M. Cell suicide for beginners. Nature 1998; 396: 119± 22. 28. Kriehuber R, SimkoÂ M, Schiffmann D, Trott K-R. Delayed cytotoxic and genotoxic effects in a human cell line following x-irradiation. Int J Radiat Biol 1999; 75: 1021± 7.