| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Institute of Diagnostic and Interventional Radiology, University Hospital Jena, Postfach, D-07740 Jena, Germany (I.H., W.A.K.); and Department of Single Cell and Single Molecule Techniques, Institute of Molecular Biotechnology, Jena, Germany (A.R., K.O.G.). Received August 24, 2004; revision requested October 29; revision received December 2; accepted January 12, 2005. Address correspondence to I.H. (e-mail: ingrid.hilger{at}med.uni-jena.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Experimentation was approved by the regional animal care committee. Human adenocarcinoma cells (MX-1) and human fibroblasts (HTB-125) were exposed to defined temperatures of 45°–90°C for 4 minutes. Single- and double-strand DNA breaks (expressed as a percentage of the total DNA in tail) were identified by using the alkaline comet assay, and cell survival was determined by using the cloning assay and trypan blue exclusion. For in vivo experiments, MX-1 tumors were implanted into 14 mice. Magnetic heating at temperatures of 59°–96°C was subsequently performed by injecting magnetic material into the tumor (7 mg ± 3 [± standard deviation] magnetite per tumor) and applying an alternating magnetic field (8.8 kA/m; 400 kHz) for 4 minutes. The efficiency of the temperature-dependent induction of DNA damage in isolated tumor cells was quantified and compared with that in cultured cells.
RESULTS: Results of experiments with cell cultures revealed a strong correlation between DNA damage, cell survival, and temperature, as determined with the cloning assay and trypan blue exclusion. The threshold thermoablasive temperature for tumor cell elimination was found to be 55°–60°C. Moreover, a strong impairment in cell survival was found when damaged DNA accounted for more than 50% of the total DNA. The heating sensitivities of malignant and nonmalignant cells did not differ. After the magnetic heating of tumors in vivo to temperatures of up to 96°C (rectal temperatures between 27°C ± 2 and 29°C ± 2), isolated tumor cells showed a mean of 71.9% ± 24.5 of total DNA in the tail per cell compared with nontreated tumors, which showed 8.0% ± 3.1.
CONCLUSION: There appears to be a threshold temperature for the induction of irreversible DNA damage. This is reflected by the results of in vivo experiments in tumor-bearing mice after high temperatures were applied to tumors by using magnetic heating.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
For the hyperthermic treatment of tumors (ie, the heating of tumors to temperatures of up to approximately 43°C for 60 minutes), extensive investigations have been performed with respect to the effects of heating and temperature in combination with time; no systematic data, however, are available for thermoablative treatments (ie, the heating of tumors to temperatures higher than 47°C for several minutes). To our knowledge, only a few investigations have focused on the heating effects of thermoablative temperatures (eg, by measuring the activity of cellular dehydrogenases) (5).
It is known that the presence of irreversible DNA damage is closely related to cell death (6). Moreover, the induction of DNA damage plays an important role in oncology, particularly with respect to the currently discussed combination of hyperthermic treatment (ie, the heating of tumors to temperatures lower than approximately 47°C) and radiation therapy (7). For these two treatment methods, heat is used to reduce the ability of the cell to repair x-ray-induced DNA damage. In addition to the complementary effects of these techniques, the use of thermoablative treatment (ie, the heating of tumors to temperatures higher than 47°C) enables heat to be converted into a potent tool for the induction of DNA damage and cell death.
DNA damage and cellular repair processes can be detected by using the so-called comet assay, which has been described for the detection of both single- (8) and double-strand (9) breaks. The principle of the comet assay is that when the fragmented (ie, damaged) DNA of the cell nuclei are embedded in agarose and exposed to an applied electric field, the DNA fragments will migrate to create the shape of a comet while the undamaged DNA remains in the former cell nucleus volume. Information about the existence of DNA damage can be obtained by analyzing the intensity with which the DNA is distributed in the direction of the electric field. During the past years, this technique has been applied in several studies for the detection of environmental toxins (10), radiation (11), DNA synthesis and repair (12,13), apoptosis (14), and mutation in cancer patients and drug sensitivities at the level of single cells (14–16). Thus, the purpose of our study was to determine the effects of temperature on cell death for cells in culture and to compare these effects with the results of in vivo experiments in which heating is induced in mice with implanted human adenocarcinoma by using magnetic methods.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Temperature-dependent DNA Damage in Cells
To determine the relationship between temperature and DNA damage, two authors (A.R., I.H.) incubated cell suspensions (5 x 105 cells in 500 µL of cell culture medium, as specified earlier) for 4 minutes in a thermocycler (GeneAmp 9600; Perkin Elmer, Foster City, Calif) set at 45°C, 50°C, 55°C, 60°C, 70°C, 80°C, or 90°C. Temperature measurements were performed as previously described in the literature (17). After heat incubation, cells were immediately embedded in agarose gels and processed for the comet assay (described later) so that the amount of DNA damage without the induction of cellular repair processes could be measured. To determine the effects of repair processes, we performed the same experiment with the addition of a 1-hour repair period during which the cells are placed in an incubator at standard conditions (ie, at 37°C, with 5% CO2 and 95% humidity). Control cells were kept at 37°C for the duration of the heat treatment, and DNA damage was evaluated by using the comet assay.
Cell Viability Test
To estimate the effects of temperature-dependent DNA damage on cell survival, we investigated the viability of treated cells by using trypan blue staining to identify those cells with intact or damaged cell membranes; viability was also investigated by using the cloning assay to determine the proportion of proliferative cells (A.R.) (18). After a 4-minute heating period, the cell suspensions (5 x 105 cells) were split into three aliquots. The first aliquot was directly stained with 0.1% (vol/vol) trypan blue solution in 0.9% (vol/vol) NaCl. The second aliquot was incubated in fresh medium for 1 hour to allow for DNA repair and cell recovery after heating; the cell suspension was then stained for 5 minutes in trypan blue solution. After staining, 100–200 cells were analyzed by using a microscope, and the cells with corrupted membranes (ie, those that had taken up the dye) were identified. The number of cells with corrupted membranes was counted as fraction of the total cell number. All analyses were performed in triplicate by three different researchers (I.H., A.R.). The third aliquot was seeded in 5 mL of fresh medium and was cultivated for 10 days, as described earlier, to show the clonogenic survival (cloning assay). The medium was subsequently aspirated, and the cells were washed with 5 mL of phosphate-buffered saline (PBS; Sigma-Aldrich) and fixed with 50% (vol/vol) methanol at –20°C for 5 minutes. Staining was performed with 30% (wt/vol) methyl blue (Sigma-Aldrich), and colonies of more than 10 cells were counted by using an inverse microscope (Axiovert; Zeiss, Jena, Germany).
To detect apoptosis and necrosis, two authors (A.R.) incubated 105 MX-1 cells in the thermocycler to the previously mentioned temperatures (ie, 45°C, 50°C, 55°C, 60°C, 70°C, 80°C, and 90°C). After a postincubation period of 60 minutes, cells were centrifuged at 300g for 5 minutes and resuspended in 50 µL of the annexin V staining solution according to the recommendations of the manufacturer (In Situ Cell Death Kit, annexin V-Fluos; Roche, Mannheim, Germany). Twenty minutes later, cells were analyzed by using a fluorescence microscope (Axioskope; Zeiss) that was equipped with filter sets for fluorescein isothiocyanate and rhodamine at x250 magnification. Two independent analyses were performed with 100 cells per data point by two authors (A.R.). Cells were scored as apoptotic (annexin V signal alone) or necrotic (propidium iodide signal alone or propidium iodide and annexin V signal combined). Total cell counts were performed with phase contrast.
Animals and Tumor Implantation
Fourteen female immunodeficient mice that weighed 24 g ± 2.3 (± standard deviation) were group housed in a solid-floor cage (Ehret, Berlin, Germany). Sawdust (Altromin Tierlabor Service, Lage, Germany) was used as bedding. Room temperature was controlled at 21°C ± 2. A 24-hour light-dark cycle that consisted of 10 hours of light and 14 hours of dark was maintained. Animals received a diet of commercially available pellets (Altrombin Tierlabor Service) and water ad libitum. Experimental tumors were grown subcutaneously after 0.35 mL of Dulbecco's minimal essential medium that contained 2 x 106 MX-1 cells was injected into the lateral portion of the abdomen (I.H.). All experiments were approved by the regional animal care committee.
Magnetic Heating of Tumors in Mice
Experiments were started approximately 6 weeks after tumor implantation (I.H.). Tumor volumes were calculated to be approximately 356 x 168 mm3, as described by Steel (19). Prior to the experiments, the animals received anesthetic agents (0.5 mg of medetomidinhydrochloride [Dormitor; Pfitzer, Karlsruhe, Germany], 5 mg of midazolamhydrochloride [Dormicum; Hoffmann-La Roche, Grenzach-Wyhlen, Germany], and 0.05 mg of fentanyldihydrochloride [Fentanyl; Janssen-Cilag, Neuss, Germany], all per kilogram body weight). On the basis of tumor volume, 59 ± 23 µL (approximately 15% of the tumor volume) of a magnetic fluid sample (Ferrofluidics, Nürtingen, Germany) was injected into the tumor (n = 11). The amount of magnetite per 350 mm3 of tumor tissue was calculated to be 7 mg ± 3. The intratumoral distribution of iron oxide was determined at radiography. Four animals were used as controls (ie, no injection of magnetic material). All mice were exposed to an alternating magnetic field (frequency, 400 kHz; amplitude, 8.8 kA/m), which was induced by a circular coil (diameter, 9 cm) for 242 seconds. Rectal temperatures and temperatures at the tumor periphery (Fig 1) were assessed by using thermocouples, as described previously in the literature (4). For data interpretation, temperatures at the tumor periphery were averaged.
|
Comet Assay
The comet assay was performed by one author (A.R.) according to the alkaline version first described by Singh et al (20) and later by Tice (21). In short, frosted slides (Labcraft, London, England) were coated with 0.5% (wt/vol) type II regular agarose (Sigma) in phosphate-buffered saline and were air dried. A second agarose layer (400 µL of 1% [wt/vol] type II regular agarose in phosphate-buffered saline) was then added to the slides and flattened with a coverslip. After the agarose had cooled, the coverslips were removed, and the microgels were stored overnight in a humidified chamber at 4°C.
The cell suspension (5 x 105 cells per milliliter) that was prepared from the tumor sample was then diluted 1:4 with 1% (wt/vol) type VII low-melting-point agarose (Sigma) in phosphate-buffered saline, which was incubated to a temperature of 40°C. A 100-µL volume of this suspension was added to the slides to create a third agarose layer on top of the previously prepared microgels. This new layer was then flattened with a coverslip and cooled until the agarose became solid.
The gels were subsequently incubated for 1 hour at 4°C in a lysis solution (2.4 mol/L of NaCl, 100 mmol/L of Na2EDTA, 10 mmol/L of Tris [pH 10.0], 1% [vol/vol] sodium N-lauryl-sarcosinate, 10% [vol/vol] dimethylsulfoxide, and 1% [wt/vol] Triton X-100, all manufactured by Sigma). The slides were then drained and incubated at 4°C for another 60 minutes in an electrophoresis buffer (1 mmol/L of Na2EDTA and 300 mmol/L of NaOH [pH 13.1]). The gels were then transferred to a precooled electrophoresis tank filled with fresh electrophoresis buffer. Electrophoresis was performed at 1 V/cm for 35 minutes at 4°C, with continual stirring to avoid buffer degradation.
After electrophoresis, the slides were neutralized in 100 mmol/L of Tris (pH 7.5) for 5 minutes at room temperature and were stored in absolute ethanol at 4°C. Immediately before image analysis, the cells were rehydrated twice for 5 minutes in water at room temperature. DNA was stained with 30 µL of 1:500 diluted SYBR Green (Molecular Probes, Leiden, the Netherlands), which contained 50% (vol/vol) antifade (Qbiogene, Heidelberg, Germany) in a buffer solution (10 mmol/L of Tris HCl and 1 mmol/L of EDTA [pH 7.4]). Image analysis was performed with a fluorescence microscope (Axioskope; Zeiss) that was equipped with a 50-W HBO 50 mercury vapor lamp and fluorescein isothiocyanate filter setting (number 10; Zeiss). Live images were recorded with an intensified charge-coupled device camera (Variocam; PCO Computer Optics, Kelheim, Germany) and were transferred to a personal computer running image analysis software (Komet 4.0; Kineticimaging, Liverpool, United Kingdom). Images were analyzed in relation to the temperature-dependent morphologic features. The percentage of DNA in the tail (ie, the percentage of damaged DNA) was chosen for quantifying the extent of DNA damage. These values were calculated from the integrated fluorescence intensity distribution of the microscopic images. Two independent experimental treatments were performed. For each temperature and experiment, two slides were prepared. A total of 150 cells per sample point were analyzed on at least three independent slides. This analysis was performed by using a computer-aided software program (Komet 4.0; Kineticimaging) and was conducted independently and randomly by three researchers (A.R., I.H.).
Data Analysis
The data were combined, and the percentage of DNA in the tail was calculated by using a spreadsheet software Excel macro (Kineticimaging). From all measurements of each individual slide, the median value was calculated. From three replicated slides, the mean of the median was calculated (22). For the experiments with cells in culture (both cell lines), the extent of DNA damage was evaluated in relation to the applied temperature with and without repair incubation. Cell survival was assessed in relation to DNA damage, and the heating effects (ie, DNA damage and cell survival) were directly compared with respect to temperature dependence. For the in vivo experiments, DNA damage was analyzed with respect to its dependence on the average temperatures at the tumor periphery. The specimens were evaluated by two authors independently (A.R., I.H.).
Statistical calculations were performed by using a commercially available software program (Sigma Stat 3.1; Jandel Scientific, Corte Madera, Calif) (A.R.). The Kruskal-Wallis one-way analysis of variance on ranks, which was used to evaluate the results of the comet assay, was applied to the combined measurements of at least two independent repeated treatments. For the differences in median values, P < .001 was considered to indicate a statistically significant difference. The Dunn test was used to compare individual measurements with control measurements, with a significance level of P < .05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
Microscopically, typical cometlike features were observed after cells were heated for 4 minutes (Fig 4). Cells that were kept at 37°C were round, indicating the absence of DNA breaks. As temperatures increased, the migration of DNA out of the nuclear volume that formed the comet tail also increased owing to a greater amount of fragmented DNA. The effects of DNA repair were visibly demonstrated as a reduction in the amount of DNA fragments in the cells that underwent repair incubation, as compared with cells that were exposed to the same temperature without repair incubation (Fig 4).
|
Typical comets were also seen in cells isolated from tumors after magnetic thermoablation (Fig 5). Isolated tumor cells without the application of magnetite (ie, no temperature increase) were round, and virtually no DNA damage could be detected in these cells. Nevertheless, tumor cells treated with magnetic thermoablation at a maximum temperature of 75°C demonstrated comets with oval tails.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Our findings relate to the interactions of heating effects, and data for both temperature and cell survival imply that the threshold temperature for cell death by using short treatment times (4 minutes) is 55°–60°C. Particularly, heat treatments of 55°–65°C led to a greater fluorescence intensity (greater than 50% of total fluorescence intensity) in the tail of the comet after a 1-hour repair period than was observed in cells processed directly after heat treatment. On the basis of the shape of the comet (14,23), we determined that the observed effects were related to the induction of apoptosis. These effects take place if the amount of DNA damage exceeds a repairable limit.
Our findings supplement—and refine—the temperature range reported by Heisterkamp et al (5). According to Heisterkamp et al, cell death, as determined by the activity of cellular dehydrogenases immediately after heating, could be achieved by heating cells to 50°–60°C for a period of 3 minutes. In comparison with the findings of the present study, this is a relatively broad temperature range that could be attributed to the fact that cellular dehydrogenases are involved in a wide range of different cellular metabolic pathways, including apoptosis. Moreover, we did not observe any differential heat sensitivity between malignant (MX-1) and nonmalignant (HTB-125) cells within the investigated temperature range. This finding agrees with the fact that the hypersensitivity of malignant cells reported in relation to hyperthermic tumor therapy (ie, treatments of up to approximately 45°C) is not caused by the temperature effects themselves but by the physiologic factors of the tumor region (24).
The curves that were related to the dependence of DNA damage on temperature indicate that inactivation of the DNA repair enzymes occurs at temperatures around 55°C. At temperatures lower than 55°C, no inactivation of the DNA repair enzymes seems to take place (ie, damages are being repaired), whereas at higher temperatures, this system is inactivated. According to Xu and Qian (25), this inactivation could be the result of protein coagulation.
Our results show that cell viability is dramatically decreased when more than 50% of the total fluorescence intensity is located in the comet's tail. This is in good agreement with earlier reported values of the so-called olive tail moment, which is another parameter for data analysis that is related to the comet assay (16). Obviously, in this case, if the amount of DNA damage exceeds a reparable threshold, then the cells are prone to die.
One major finding is that heat-induced DNA damage is correlated with cell survival and clonogenicity. This suggests that heat-induced irreparable DNA damage manifests by means of cell cycle arrest and by activation of the apoptosis cascade, which occurs after the activation of stress-associated signaling pathways. These findings justify further investigations to clarify if the comet assay could be used for the assessment of heat-induced damage during tumor treatments in patients. Compared with the standard cell survival test, the comet assay can be easily applied to cells isolated from biopsy material to determine the efficacy of a therapeutic treatment.
The results obtained from in vivo experiments show that the corresponding heating effects could be assessed by performing a comet assay after tumor cell isolation. During magnetic thermoablation, selective tumor heating was generated by means of hysteresis or relaxational losses to the magnetic material (26). The temperature data are based on subcutaneously grown tumors. One side of the tumor was surrounded by air, which provided greater insulation than water, which is the major component of biologic tissue. Therefore, it is conceivable that the effects of temperature washout are higher in tumors surrounded completely by vascularized tissue (eg, tumors located in solid organs) than in tumors located subcutaneously.
The amount of DNA damage recorded after magnetic thermoablation of tumors in mice is in agreement with the results obtained from the heating of cells in a thermocycler. This is exemplified by the sigmoid fit function obtained from the curve of cells heated in vitro. Compared with in vitro measurements of the cloning assay and trypan blue exclusion test, in vivo results demonstrated that effective tumor elimination was achieved. The different heating techniques (magnetic heating of isolated tumor cells and water bath, as obtained from cell culture experiments) should not have had different effects on the induction of DNA damage and cell survival.
The relative fluorescence intensities of the treated animal group were significantly higher than those of the nontreated controls. The fact that DNA damage was found in the controls represents the presence of dead cells in the tumor itself, particularly in the central necrotic core of the tumor. The fact that two tumors showed a comparatively lower amount of DNA damage is attributed to the inhomogeneous temperature distribution within the tumor region as a consequence of the inadequate placement of magnetic material within the tumor.
When looking at the fluorescence intensity in the tail, we observed that the data obtained from the in vivo studies in mice seemed to reflect the data obtained from the in vitro experiments in cultured cells. Most of the cells that were isolated from the tumors in mice showed more than 50% of total of DNA in the tail when exposed to temperatures higher than 60°C. A comparable situation was found when cultured cells were heated with a thermocycler. On the basis of the results of cell survival for in vitro experiments, we determined that isolated tumor cells should no longer be viable or able to divide in the long term. Therefore, the observed temperature threshold of 55°–60°C should also be applicable to the in vivo situation, even when applying other thermoablation methods for the treatment of tumors.
To the best of our knowledge, this is the first time that DNA damage has been associated with thermoablation temperatures both in vitro and in vivo. Most other studies focused on the relationships associated with radiation treatments (27–30).
Possible limitations of the present study are related to interindividual variability of the intratumoral magnetic nanoparticle distribution, which led to local variations in the generated temperatures.
Taken together, our results on cells in culture show that there appears to be a dependence between heat-induced DNA damage, cell survival, and temperature and that a narrow temperature threshold leading to irreparable DNA damage should be considered for adequate tumor elimination. These relationships were also reflected in the in vivo experiments during which thermoablative temperatures were applied to tumors in mice.
Practical application: Results from the present study indicate the temperature thresholds that are to be considered for an efficient thermoablative elimination of tumor cells. Moreover, such information may be important for a more sophisticated monitoring of therapeutic outcome.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, I.H., W.A.K.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, all authors; experimental studies, I.H., A.R.; statistical analysis, A.R., K.O.G.; and manuscript editing, I.H., A.R., K.O.G.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
) and after 1-hour recovery period (
). Error bars indicate standard errors. Correlation between DNA damage and target temperature without repair incubation was calculated as r2 = 0.98. Dramatic increase in dead cells at temperatures higher than 55°C highlights critical temperature for successful tumor elimination. * = measurement values in MX-1 cells that were significantly different from those in control cells (P < .05, Dunn test).
). All three techniques showed comparable results and sensitivities.
