1990, The British Journal of Radiology, 63, 542-546
The effect of ultrasound on the cytoxicity of adriamycin By Pam Loverock, MPhil, Gail ter Haar, PhD, *M. G. Ormerod, PhD and *P. R. Imrie, MIBiol Physics Division and 'Section of Pathology Institute of Cancer Research, Sutton, Surrey (Received September 1989 and in revised form February 1990)
Abstract. The effect of continuous wave ultrasound exposures on the cytoxicity of adriamycin has been studied. It has been found that 2.6 MHz, 2.3 Wcm 2 (spatial average) ultrasound can enhance the cell killing potential of adriamycin both in suspensions of single V79 Chinese hamster fibroblast cells and in spheroids formed from these cells. The ratio of the slopes of the survival curves for single cell suspensions is 1.5. For spheroids, the growth delay is increased by 1.3 days by simultaneous ultrasound exposure. Flow cytometric studies of the intracellular concentration of adriamycin following ultrasound exposure reveals that this is increased when compared with that measured when the cells are only exposed to adriamycin. Evidence is presented to suggest that this is a non-thermal effect of ultrasound.
Ultrasound is one of the methods of choice for the induction of hyperthermia for cancer therapy. Hyperthermia is used not only as a single modality, on its own, but also in conjunction with radiotherapy or chemotherapy. For this reason, it seems sensible to study the effect of ultrasound exposures on the cytotoxic action of chemotherapy agents, both at physiological and at hyperthermic temperatures. There are a few reports in the literature concerning the effect of ultrasound on drug toxicity. Kremkau et al (1976) studied the survival on mice inoculated with L1210 Leukaemia cells that had been exposed to nitrogen mustard or to nitrogen mustard and ultrasound. They found that animals receiving cells that had been exposed to both nitrogen mustard and ultrasound had a significantly higher survival rate after 3 weeks than those that had received cells that had been exposed either to the drug alone or to ultrasound alone. The authors stated that the "increased cytotoxicity cannot be explained solely on the basis of a conventional thermal mechanism of action", although the maximum temperature rise measured in cell suspensions following ultrasound exposure was 5-6°C. Senepati et al (1982) looked at the effect of ultrasonically induced hyperthermia on the toxicity of cyclophosphamide, melphalan and procarbazine in human tumour xenografts (MX-1 and LX-1) in mice in vivo. They allowed the tumour temperature to rise to 43°C, and found evidence for a synergistic effect of ultrasound hyperthermia and melphalan on MX-1 tumours, and between ultrasound hyperthermia and procarbazine on LX-1 tumours. Akimoto (1985) studied the effect of the in vivo exposure of murine tumours to ultrasound and Mitomycin C. They found that the combined treatment significantly enhanced the amount of inhibition of tumour growth and increased the animals' survival rate. It was speculated that this was caused by an effect of ultrasound on membrane permeability. The tumour temperatures in this study were constrained to be below 35°C, as they were in the work reported by Yumita et al (1987). Yumita et al studied the effects of ultrasound on dauno542
mycin and adriamycin (ADM) toxicity in rats bearing Yoshida sarcoma. They found synergistic effects when ultrasound was combined with either drug. Interestingly, ultrasound exposure alone caused an inhibition in tumour growth, even though the tumour temperature was held below 35°C. We have embarked on a systematic investigation of the effect of ultrasound on a number of antitumour drugs in vitro. This report concerns the effect of ultrasound on ADM toxicity to cells in suspensions and to spheroid cultures. Adriamycin is an anthracycline antibiotic. It is known to intercalate with DNA. The cell surface is also thought to be a target for ADM action. (Karczmar & Tritton, 1979; Born & Eichholtz-Wirth, 1981). It is known that ADM cytotoxicity is enhanced by hyperthermia (Bleehen, 1984). Materials and methods Cell and spheroid culture
Chinese hamster lung fibroblasts V79-379A were grown in suspension using methods described elsewhere (Stratford & Adams, 1977). The cells were maintained in log phase at concentrations ranging between 105 and 106 per ml. These cells have a doubling time of 10-12 h, and therefore need diluting daily. When spheroids were required, 5 x 104 cells were inoculated into 10 ml of Eagles Minimum Essential Medium (MEM) plus 10% fetal calf serum (fcs) contained in 90 mm Petri dishes base-coated with 1% agar (noble, Difco). The dishes were kept in an air/5% CO2 incubator at 37°C for 2 days until small clumps of cells were seen. The medium was then replaced and the culture split into two or three Petri dishes. After 4-5 days the spheroids had reached approximately 200//m in diameter and were put into spinner flasks with fresh medium. The medium was replaced daily (West, 1983). Cell and spheroid assays
(i) Cells. After treatment, cells were washed by centrifugation and resuspension, counted, serially diluted, plated in MEM +15% fcs, and incubated at 37°C in an The British Journal of Radiology, July 1990
The effect of US on the cytoxicity of adriamycin
atmosphere of 95% air plus 5% CO2 for 7 days. Colonies containing more than 30 cells were scored as survivors and the surviving fraction was calculated as the number of colonies scored divided by the initial number of cells plated. The surviving fraction, or plating efficiency, of control cells maintained at 37°C was routinely greater than 0.84. Survival curves were plotted for which the logarithm of surviving fraction was expressed as a function of dose (either drug concentration, or exposure time). (ii) Spheroids. Following treatment, spheroids were washed by centrifugation and resuspension, and spheroids of the required size were selected and put into individual microwells (Linbro 17x15 mm) which had been coated with 1% Agar Noble. In any one experiment, 12 spheroids were selected for each experimental point. Two diameters at right angles were measured using a calibrated graticule on an inverted microscope, immediately after treatment, and daily thereafter. Spheroid volumes were calculated using the formula 4 volume = - n (radius)3 where the radius is taken as the average of the two measured radii. Growth curves were constructed, for which log
volume ~| initial volume I
was plotted as a function of time after treatment. Growth delay was taken as the difference in time taken to reach six times the initial volume for the different treatment conditions as compared to the control. Ultrasound exposures All ultrasound exposures were carried out using the experimental system described in detail elsewhere (ter Haar et al, 1980, 1988). A 2.6 MHz plane transducer (2.5 cm diameter) was mounted at the bottom of a thermostatically controlled water bath, at an angle of 45° to the horizontal, tilted upwards. An 8 ml aliquot of the sample suspension was inoculated into a cylindrical stainless steel chamber with acoustically transparent windows, (25 mm in diameter). The chamber was then mounted parallel to the transducer face, at a distance of 25 cm, the position of the last axial maximum. The ultrasonic beam hitting the water surface after transmission through the chamber is reflected through 90° away from the primary beam, thus avoiding standing waves. Chambers containing control cell suspensions were also placed in the tank, but away from the ultrasonic beam. Spatially averaged intensities were measured using a tethered float radiometer (Shotton, 1980). All intensities quoted here are spatial averages. The acoustic pressure profile across the beam was measured using a PVDF membrane hydrophone calibrated at the Vol. 63, No. 751
0-6
0-4
0-2
0
2 4 Distance
mm
Figure 1. Transverse beam plot at the position of the sample holder. The pressure profile was obtained using a PVDF membrane hydrophone. The profiles are shown both with the sample holder in position (O) and with it removed (#).
National Physical Laboratory (GEC Marconi). Sample pressure profiles taken 26 cm from the transducer are shown in Fig. 1. The profiles with and without the chamber in place are shown. It can be seen that the profiles are not changed appreciably by the presence of the sample holder. Adriamycin exposure Adriamycin was obtained from Montedison Pharmaceuticals Ltd. The contents of a 10 mg vial were dissolved in 5 ml of sterile water and kept frozen as stock solution. For each experiment, 0.5 ml of stock solution was diluted with PBS to give a 1000 /xM solution of ADM (molecular weight 543.8) which was kept in the dark. An appropriate volume was added to the cell suspension immediately before loading into the exposure chamber. Suspensions were exposed at a concentration of 2.105 cells/ml for 1 h. Immediately after exposure the suspensions were removed from the chamber, spun down, washed and resuspended in phosphate buffered saline and then assayed in the manner described above. Temperature studies In an attempt to determine the effect of small changes in temperature on ADM toxicity, some cell suspension experiments were carried out in the stainless steel chambers with the water bath held at temperatures in the range 37-40°C. Flow cytometry studies An Ortho Cytofluorograf 50 H equipped with a lexel 50 mW argon-ion laser tuned to 488 nm and an Ortho 2150 computer system was used. The parameters measured were forward and orthogonally scattered light and ADMflourescence(> 520 nm). The cells were gated on scattered light and fluorescence was displayed on a univariate histogram. The mean channel number from the histogram was recorded and used as a measure of intracellular ADM concentration. 543
P. Loverock, G. ter Haar, M. G. Ormerod and P. R. Imrie
Results
(i) Single cells in suspension. Figure 2 shows the survival curve for cells exposed to ADM concentrations in the range 1-8 /*M for 1 h at 37°C. The data shown is averaged over eight experiments and the error bars show the standard error of the mean. A linear regression analysis has been performed on the data. Figure 2 also shows the survival curve for cells exposed simultaneously to ultrasound (2.6 MHz, 2.3 Wcm~2 spatial average, continuous wave) and ADM (1-6 nM) for l h at 37°C. It can be seen that the ultrasound has rendered the ADM more toxic. The ratio of the slopes of the survival curves with and without ultrasound is 1.5. Ultrasound alone at this intensity at 37°C does not affect cell survival (Fig. 3). (ii) Spheroids. Figure 4 shows the growth curves for 200 fim diameter spheroids exposed to 8 pM ADM, with and without simultaneous ultrasound treatment, and for control spheroids maintained at 37°C without drug or ultrasound treatment. These are averaged over five experiments. Twelve spheroids were measured for each treatment condition, for each experiment.
4
5
6
7
JJM ADM Figure 2. Survival curves showing surviving fraction as a function of adriamycin concentration at 37°C, 60 min exposures, (#) and showing surviving fraction as a function of adriamycin concentration at 37°C for cells exposed simultaneously to adriamycin and continuous wave ultrasound (2.6 MHz; 2.3 Wcm~2 spatial average) for 60 min (O)544
120 Exposure time - mins Figure 3. Survival curve showing surviving fraction as a function of time for control cells (#) and for cells exposed to ultrasound alone at 37°C. (2.6 MHz; 2.2 Wcirr2 spatial average) «>)•
When averaged over five experiments (60 spheroids), it was found that the time to reach six times their volume at treatment for 200 ^m diameter spheroids exposed to 8 fiM ADM was 1.9 (±0.2) days longer than for the control group, whereas it took 3.2 (±0.3) days
1 2
3 Days
4 5 6 7 post-treatment
8
Figure 4. Growth curves for V79 spheroids (200 fim diameter). Curves are shown for control spheroids (#), for spheroids exposed to 8/xM adriamycin (A) and for spheroids exposed simultaneously to 8 //M adriamycin and continuous wave ultrasound (O). The British Journal of Radiology, July 1990
The effect of US on the cytoxicity of adriamycin
5 fim ADM, the ratio of the mean channel number of ADM fluorescence for cells treated with ADM plus ultrasound to that for ADM alone was 1.16±0.05 (average of measurements made on five separate occasions). For cells in spheroids treated with 8 pM ADM, the ratio was 1.24 ±0.07. If the cells were incubated at 37°C after treatment, these ratios were maintained despite the loss of adriamycin from the cells. Discussion
The investigations described here have been designed to address the question as to whether ultrasound exposures can modify the toxicity of adriamycin to cells. We have shown that the effect of adriamycin is greater if cells are exposed simultaneously to the drug and to ultrasound. The ultrasound intensity used is not sufficient to cause any change in cell survival on its own at 37°C. The effect is seen both in single cell suspensions and in spheroids. It is well known that the toxicity of ADM is enhanced at hyperthermic temperatures (Hahn et al, 1975). Previous studies have shown that the ultrasonically induced temperature rise in these cell suspensions under these exposure conditions is no more than 0.2°C (ter Haar et al, 1980). This has been re-confirmed for the work discussed here (data not shown). The temperature 38 39 studies performed here have demonstrated that the level Temperature of cell killing obtained from the combination of 5 pM ADM and ultrasound is considerably greater than that Figure 5. Survival curve showing surviving fractions for cells exposed to 5/iM adriamycin for 60min at different temperaobtained from 5JIM ADM when the cells are maintures. The hatched area shows the survival of cells exposed to tained at 40°C, a temperature elevation of 3°C above adriamycin and ultrasound for 60 min at 37°C. normal. Thus, it seems unlikely that the ultrasound effects seen are a result of bulk heating effects. Although the intraspheroid temperatures have not been measured, longer for spheroids subjected to the combination of it can be shown theoretically that spheroids of the sizes ultrasound and ADM. used here cannot maintain a temperature significantly Similar effects were seen with 400 jxm diameter sphe- above that of their suspending medium. roids. Averaged over three experiments, the growth The mechanism by which the ADM cytoxicity is delays were found to be 1.2 (±0.4) days for the ADM enhanced is still unknown. The flow cytometry results alone treatment, and 2.1 (±0.5) days for the combined indicate that there is an increase in intracellular ADM ADM and ultrasound exposure. concentration following ultrasound exposure. However, (iii) Temperature studies. Figure 5 shows the surviving the increases seen (~ 16%—25%) do not account in full fractions for cells exposed to 5 /xM ADM for 1 h while for the increase in toxicity. At 5 fiM, the level of cell being maintained at elevated temperatures (38°C, 39°C killing achieved by the combined exposure is the same as and 40°C). The survival does not change significantly that achieved by 8/zM ADM in its own. This is an from that of the control (37°C) until a temperature of effective increase of 60%. There may be a number of 40°C is reached. Also marked on Fig. 5 is the surviving reasons that the full apparent increase in ADM concenfraction for 5/zM ADM + ultrasound at 37°C. tration is not measured by flow cytometry, a major one (iv) Flow cytometry studies. When separate samples of being that the efflux of ADM from cells is extremely cells were treated identically with adriamycin, the mean rapid following washing, and there is a time delay channel numbers of the ADM fluorescence lay within between the end of treatment and measurement on the ±3.5%. If cells were treated in suspension with ADM flow cytometer. The addition of Verapamil inhibits this and then held at 37°C, their fluorescence gradually efflux but does not prevent it. decreased owing to efflux of adriamycin from the cells. This effect was inhibited, but not completely eliminated, by Verapamil. Verapamil is a calcium influx blocker, Acknowledgments We should like to thank Professor C. R. Hill for his help and and has been shown to inhibit drug efflux (Helson, encouragement during the performance of this study, and Mr 1984). Thereafter, cells were held on ice immediately Ian Rivens for his help in calibration of the ultrasonic fields. after treatment with ADM. The work is funded by a programme grant from the MRC/ When single cell suspensions were incubated with CRC Joint Committee for the Institute of Cancer Research. Vol. 63, No. 751
545
P. Loverock, G. ter Haar, M. G. Ormerod and P. R. Imrie
References
KARCZMAR, G. S. & TRITTON, T. E., 1979. The interaction of
AKIMOTO, R., 1985. An experimental study on enhancement of the effect of anti-cancer drug by ultrasound. Journal of the Japanese Society of Cancer Therapy, 20, 562-570. BLEEHEN, N. M., 1984. Hyperthermia with drugs: current status. Strahlentherapie, 160, 721-724. BORN, R. & EICHHOLTZ-WIRTH, H., 1981. Effect of different
physiological conditions on the action of adriamycin on Chinese hamster cells in vitro. British Journal of Cancer, 44, 241-246. TER HAAR, G. R., STRATFORD, I. J. & HILL, C. R., 1980.
Ultrasonic irradiation of mammalian cells in vitro at hyperthermia temperatures. British Journal of Radiology, 53, 784-789. TER HAAR, G. R., WALLING, J., LOVEROCK, P. & TOWNSEND, S.,
1988. The effect of combined heat and ultrasound on multicellular tumour spheroids, International Journal of Radiation Biology, 53, 813-827. HAHN,
G.
M.,
BRAUN,
J.
&
HAR-KEDAR,
I.,
1975.
Thermochemotherapy: synergism between hyperthermia (42-43) and adriamycin (or bleomycin) in mammalian cell inactivation. Proceedings of the National Academy of Sciences of USA, 72, 937-940. HELSON, L., 1984. Calcium channel blocker enhancement of anticancer drug cytotoxicity—a review. Cancer Drug Delivery, 1, 353-361.
546
adriamycin with small unilamellar vesicle liposomes. A fluorescence study. Biochimica et Biophysica Ada, 557, 306-319. KREMKAU, F. W., KAUFMANN, J. S., WALKER, M., BURCH, P. &
SPURR, C. L., 1976. Ultrasonic enhancement of nitrogen mustard cytotoxicity in mouse leukaemia. Cancer, 37, 1463-1647. SENEPATI, N., HOUCHENS, D., OVEJERA, A., BEARD, R. & NINES,
R., 1982. Ultrasonic hyperthermia and drugs as therapy for human tumour xenografts. Cancer Treatment Reports, 66, 1635-1639. SHOTTON, K. C , 1980. A tethered float radiometer for measuring the output power from ultrasonic therapy equipment. Ultrasound in Medicine and Biology, 6, 131-133. STRATFORD, I. J. & ADAMS, G. E., 1977. Effect of hyperthermia
on the differential cytotoxicity of a hypoxic cell radiosensitizer Ro-07-0582, on mammalian cells in vitro. British Journal of Cancer, 35, 307-313. WEST, C. M. L., 1983. The effect of cytotoxic drugs and radiation on mammalian cells and multicellular spheroids in vitro. PhD Thesis (University of London). YUMITA, N., OKUMURA, A., NISHIGAKI, R., VNEMURA, K. &
UMEMURA, S., 1987. The combination treatment of ultrasound and antitumour drugs on Yoshida Sarcoma. Japanese Journal of Hyperthermic Oncology, 3, 175-182.
The British Journal of Radiology, July 1990
The effect of ultrasound on the cytoxicity of adriamycin By Pam Loverock, MPhil, Gail ter Haar, PhD, *M. G. Ormerod, PhD and *P. R. Imrie, MIBiol Physics Division and 'Section of Pathology Institute of Cancer Research, Sutton, Surrey (Received September 1989 and in revised form February 1990)
Abstract. The effect of continuous wave ultrasound exposures on the cytoxicity of adriamycin has been studied. It has been found that 2.6 MHz, 2.3 Wcm 2 (spatial average) ultrasound can enhance the cell killing potential of adriamycin both in suspensions of single V79 Chinese hamster fibroblast cells and in spheroids formed from these cells. The ratio of the slopes of the survival curves for single cell suspensions is 1.5. For spheroids, the growth delay is increased by 1.3 days by simultaneous ultrasound exposure. Flow cytometric studies of the intracellular concentration of adriamycin following ultrasound exposure reveals that this is increased when compared with that measured when the cells are only exposed to adriamycin. Evidence is presented to suggest that this is a non-thermal effect of ultrasound.
Ultrasound is one of the methods of choice for the induction of hyperthermia for cancer therapy. Hyperthermia is used not only as a single modality, on its own, but also in conjunction with radiotherapy or chemotherapy. For this reason, it seems sensible to study the effect of ultrasound exposures on the cytotoxic action of chemotherapy agents, both at physiological and at hyperthermic temperatures. There are a few reports in the literature concerning the effect of ultrasound on drug toxicity. Kremkau et al (1976) studied the survival on mice inoculated with L1210 Leukaemia cells that had been exposed to nitrogen mustard or to nitrogen mustard and ultrasound. They found that animals receiving cells that had been exposed to both nitrogen mustard and ultrasound had a significantly higher survival rate after 3 weeks than those that had received cells that had been exposed either to the drug alone or to ultrasound alone. The authors stated that the "increased cytotoxicity cannot be explained solely on the basis of a conventional thermal mechanism of action", although the maximum temperature rise measured in cell suspensions following ultrasound exposure was 5-6°C. Senepati et al (1982) looked at the effect of ultrasonically induced hyperthermia on the toxicity of cyclophosphamide, melphalan and procarbazine in human tumour xenografts (MX-1 and LX-1) in mice in vivo. They allowed the tumour temperature to rise to 43°C, and found evidence for a synergistic effect of ultrasound hyperthermia and melphalan on MX-1 tumours, and between ultrasound hyperthermia and procarbazine on LX-1 tumours. Akimoto (1985) studied the effect of the in vivo exposure of murine tumours to ultrasound and Mitomycin C. They found that the combined treatment significantly enhanced the amount of inhibition of tumour growth and increased the animals' survival rate. It was speculated that this was caused by an effect of ultrasound on membrane permeability. The tumour temperatures in this study were constrained to be below 35°C, as they were in the work reported by Yumita et al (1987). Yumita et al studied the effects of ultrasound on dauno542
mycin and adriamycin (ADM) toxicity in rats bearing Yoshida sarcoma. They found synergistic effects when ultrasound was combined with either drug. Interestingly, ultrasound exposure alone caused an inhibition in tumour growth, even though the tumour temperature was held below 35°C. We have embarked on a systematic investigation of the effect of ultrasound on a number of antitumour drugs in vitro. This report concerns the effect of ultrasound on ADM toxicity to cells in suspensions and to spheroid cultures. Adriamycin is an anthracycline antibiotic. It is known to intercalate with DNA. The cell surface is also thought to be a target for ADM action. (Karczmar & Tritton, 1979; Born & Eichholtz-Wirth, 1981). It is known that ADM cytotoxicity is enhanced by hyperthermia (Bleehen, 1984). Materials and methods Cell and spheroid culture
Chinese hamster lung fibroblasts V79-379A were grown in suspension using methods described elsewhere (Stratford & Adams, 1977). The cells were maintained in log phase at concentrations ranging between 105 and 106 per ml. These cells have a doubling time of 10-12 h, and therefore need diluting daily. When spheroids were required, 5 x 104 cells were inoculated into 10 ml of Eagles Minimum Essential Medium (MEM) plus 10% fetal calf serum (fcs) contained in 90 mm Petri dishes base-coated with 1% agar (noble, Difco). The dishes were kept in an air/5% CO2 incubator at 37°C for 2 days until small clumps of cells were seen. The medium was then replaced and the culture split into two or three Petri dishes. After 4-5 days the spheroids had reached approximately 200//m in diameter and were put into spinner flasks with fresh medium. The medium was replaced daily (West, 1983). Cell and spheroid assays
(i) Cells. After treatment, cells were washed by centrifugation and resuspension, counted, serially diluted, plated in MEM +15% fcs, and incubated at 37°C in an The British Journal of Radiology, July 1990
The effect of US on the cytoxicity of adriamycin
atmosphere of 95% air plus 5% CO2 for 7 days. Colonies containing more than 30 cells were scored as survivors and the surviving fraction was calculated as the number of colonies scored divided by the initial number of cells plated. The surviving fraction, or plating efficiency, of control cells maintained at 37°C was routinely greater than 0.84. Survival curves were plotted for which the logarithm of surviving fraction was expressed as a function of dose (either drug concentration, or exposure time). (ii) Spheroids. Following treatment, spheroids were washed by centrifugation and resuspension, and spheroids of the required size were selected and put into individual microwells (Linbro 17x15 mm) which had been coated with 1% Agar Noble. In any one experiment, 12 spheroids were selected for each experimental point. Two diameters at right angles were measured using a calibrated graticule on an inverted microscope, immediately after treatment, and daily thereafter. Spheroid volumes were calculated using the formula 4 volume = - n (radius)3 where the radius is taken as the average of the two measured radii. Growth curves were constructed, for which log
volume ~| initial volume I
was plotted as a function of time after treatment. Growth delay was taken as the difference in time taken to reach six times the initial volume for the different treatment conditions as compared to the control. Ultrasound exposures All ultrasound exposures were carried out using the experimental system described in detail elsewhere (ter Haar et al, 1980, 1988). A 2.6 MHz plane transducer (2.5 cm diameter) was mounted at the bottom of a thermostatically controlled water bath, at an angle of 45° to the horizontal, tilted upwards. An 8 ml aliquot of the sample suspension was inoculated into a cylindrical stainless steel chamber with acoustically transparent windows, (25 mm in diameter). The chamber was then mounted parallel to the transducer face, at a distance of 25 cm, the position of the last axial maximum. The ultrasonic beam hitting the water surface after transmission through the chamber is reflected through 90° away from the primary beam, thus avoiding standing waves. Chambers containing control cell suspensions were also placed in the tank, but away from the ultrasonic beam. Spatially averaged intensities were measured using a tethered float radiometer (Shotton, 1980). All intensities quoted here are spatial averages. The acoustic pressure profile across the beam was measured using a PVDF membrane hydrophone calibrated at the Vol. 63, No. 751
0-6
0-4
0-2
0
2 4 Distance
mm
Figure 1. Transverse beam plot at the position of the sample holder. The pressure profile was obtained using a PVDF membrane hydrophone. The profiles are shown both with the sample holder in position (O) and with it removed (#).
National Physical Laboratory (GEC Marconi). Sample pressure profiles taken 26 cm from the transducer are shown in Fig. 1. The profiles with and without the chamber in place are shown. It can be seen that the profiles are not changed appreciably by the presence of the sample holder. Adriamycin exposure Adriamycin was obtained from Montedison Pharmaceuticals Ltd. The contents of a 10 mg vial were dissolved in 5 ml of sterile water and kept frozen as stock solution. For each experiment, 0.5 ml of stock solution was diluted with PBS to give a 1000 /xM solution of ADM (molecular weight 543.8) which was kept in the dark. An appropriate volume was added to the cell suspension immediately before loading into the exposure chamber. Suspensions were exposed at a concentration of 2.105 cells/ml for 1 h. Immediately after exposure the suspensions were removed from the chamber, spun down, washed and resuspended in phosphate buffered saline and then assayed in the manner described above. Temperature studies In an attempt to determine the effect of small changes in temperature on ADM toxicity, some cell suspension experiments were carried out in the stainless steel chambers with the water bath held at temperatures in the range 37-40°C. Flow cytometry studies An Ortho Cytofluorograf 50 H equipped with a lexel 50 mW argon-ion laser tuned to 488 nm and an Ortho 2150 computer system was used. The parameters measured were forward and orthogonally scattered light and ADMflourescence(> 520 nm). The cells were gated on scattered light and fluorescence was displayed on a univariate histogram. The mean channel number from the histogram was recorded and used as a measure of intracellular ADM concentration. 543
P. Loverock, G. ter Haar, M. G. Ormerod and P. R. Imrie
Results
(i) Single cells in suspension. Figure 2 shows the survival curve for cells exposed to ADM concentrations in the range 1-8 /*M for 1 h at 37°C. The data shown is averaged over eight experiments and the error bars show the standard error of the mean. A linear regression analysis has been performed on the data. Figure 2 also shows the survival curve for cells exposed simultaneously to ultrasound (2.6 MHz, 2.3 Wcm~2 spatial average, continuous wave) and ADM (1-6 nM) for l h at 37°C. It can be seen that the ultrasound has rendered the ADM more toxic. The ratio of the slopes of the survival curves with and without ultrasound is 1.5. Ultrasound alone at this intensity at 37°C does not affect cell survival (Fig. 3). (ii) Spheroids. Figure 4 shows the growth curves for 200 fim diameter spheroids exposed to 8 pM ADM, with and without simultaneous ultrasound treatment, and for control spheroids maintained at 37°C without drug or ultrasound treatment. These are averaged over five experiments. Twelve spheroids were measured for each treatment condition, for each experiment.
4
5
6
7
JJM ADM Figure 2. Survival curves showing surviving fraction as a function of adriamycin concentration at 37°C, 60 min exposures, (#) and showing surviving fraction as a function of adriamycin concentration at 37°C for cells exposed simultaneously to adriamycin and continuous wave ultrasound (2.6 MHz; 2.3 Wcm~2 spatial average) for 60 min (O)544
120 Exposure time - mins Figure 3. Survival curve showing surviving fraction as a function of time for control cells (#) and for cells exposed to ultrasound alone at 37°C. (2.6 MHz; 2.2 Wcirr2 spatial average) «>)•
When averaged over five experiments (60 spheroids), it was found that the time to reach six times their volume at treatment for 200 ^m diameter spheroids exposed to 8 fiM ADM was 1.9 (±0.2) days longer than for the control group, whereas it took 3.2 (±0.3) days
1 2
3 Days
4 5 6 7 post-treatment
8
Figure 4. Growth curves for V79 spheroids (200 fim diameter). Curves are shown for control spheroids (#), for spheroids exposed to 8/xM adriamycin (A) and for spheroids exposed simultaneously to 8 //M adriamycin and continuous wave ultrasound (O). The British Journal of Radiology, July 1990
The effect of US on the cytoxicity of adriamycin
5 fim ADM, the ratio of the mean channel number of ADM fluorescence for cells treated with ADM plus ultrasound to that for ADM alone was 1.16±0.05 (average of measurements made on five separate occasions). For cells in spheroids treated with 8 pM ADM, the ratio was 1.24 ±0.07. If the cells were incubated at 37°C after treatment, these ratios were maintained despite the loss of adriamycin from the cells. Discussion
The investigations described here have been designed to address the question as to whether ultrasound exposures can modify the toxicity of adriamycin to cells. We have shown that the effect of adriamycin is greater if cells are exposed simultaneously to the drug and to ultrasound. The ultrasound intensity used is not sufficient to cause any change in cell survival on its own at 37°C. The effect is seen both in single cell suspensions and in spheroids. It is well known that the toxicity of ADM is enhanced at hyperthermic temperatures (Hahn et al, 1975). Previous studies have shown that the ultrasonically induced temperature rise in these cell suspensions under these exposure conditions is no more than 0.2°C (ter Haar et al, 1980). This has been re-confirmed for the work discussed here (data not shown). The temperature 38 39 studies performed here have demonstrated that the level Temperature of cell killing obtained from the combination of 5 pM ADM and ultrasound is considerably greater than that Figure 5. Survival curve showing surviving fractions for cells exposed to 5/iM adriamycin for 60min at different temperaobtained from 5JIM ADM when the cells are maintures. The hatched area shows the survival of cells exposed to tained at 40°C, a temperature elevation of 3°C above adriamycin and ultrasound for 60 min at 37°C. normal. Thus, it seems unlikely that the ultrasound effects seen are a result of bulk heating effects. Although the intraspheroid temperatures have not been measured, longer for spheroids subjected to the combination of it can be shown theoretically that spheroids of the sizes ultrasound and ADM. used here cannot maintain a temperature significantly Similar effects were seen with 400 jxm diameter sphe- above that of their suspending medium. roids. Averaged over three experiments, the growth The mechanism by which the ADM cytoxicity is delays were found to be 1.2 (±0.4) days for the ADM enhanced is still unknown. The flow cytometry results alone treatment, and 2.1 (±0.5) days for the combined indicate that there is an increase in intracellular ADM ADM and ultrasound exposure. concentration following ultrasound exposure. However, (iii) Temperature studies. Figure 5 shows the surviving the increases seen (~ 16%—25%) do not account in full fractions for cells exposed to 5 /xM ADM for 1 h while for the increase in toxicity. At 5 fiM, the level of cell being maintained at elevated temperatures (38°C, 39°C killing achieved by the combined exposure is the same as and 40°C). The survival does not change significantly that achieved by 8/zM ADM in its own. This is an from that of the control (37°C) until a temperature of effective increase of 60%. There may be a number of 40°C is reached. Also marked on Fig. 5 is the surviving reasons that the full apparent increase in ADM concenfraction for 5/zM ADM + ultrasound at 37°C. tration is not measured by flow cytometry, a major one (iv) Flow cytometry studies. When separate samples of being that the efflux of ADM from cells is extremely cells were treated identically with adriamycin, the mean rapid following washing, and there is a time delay channel numbers of the ADM fluorescence lay within between the end of treatment and measurement on the ±3.5%. If cells were treated in suspension with ADM flow cytometer. The addition of Verapamil inhibits this and then held at 37°C, their fluorescence gradually efflux but does not prevent it. decreased owing to efflux of adriamycin from the cells. This effect was inhibited, but not completely eliminated, by Verapamil. Verapamil is a calcium influx blocker, Acknowledgments We should like to thank Professor C. R. Hill for his help and and has been shown to inhibit drug efflux (Helson, encouragement during the performance of this study, and Mr 1984). Thereafter, cells were held on ice immediately Ian Rivens for his help in calibration of the ultrasonic fields. after treatment with ADM. The work is funded by a programme grant from the MRC/ When single cell suspensions were incubated with CRC Joint Committee for the Institute of Cancer Research. Vol. 63, No. 751
545
P. Loverock, G. ter Haar, M. G. Ormerod and P. R. Imrie
References
KARCZMAR, G. S. & TRITTON, T. E., 1979. The interaction of
AKIMOTO, R., 1985. An experimental study on enhancement of the effect of anti-cancer drug by ultrasound. Journal of the Japanese Society of Cancer Therapy, 20, 562-570. BLEEHEN, N. M., 1984. Hyperthermia with drugs: current status. Strahlentherapie, 160, 721-724. BORN, R. & EICHHOLTZ-WIRTH, H., 1981. Effect of different
physiological conditions on the action of adriamycin on Chinese hamster cells in vitro. British Journal of Cancer, 44, 241-246. TER HAAR, G. R., STRATFORD, I. J. & HILL, C. R., 1980.
Ultrasonic irradiation of mammalian cells in vitro at hyperthermia temperatures. British Journal of Radiology, 53, 784-789. TER HAAR, G. R., WALLING, J., LOVEROCK, P. & TOWNSEND, S.,
1988. The effect of combined heat and ultrasound on multicellular tumour spheroids, International Journal of Radiation Biology, 53, 813-827. HAHN,
G.
M.,
BRAUN,
J.
&
HAR-KEDAR,
I.,
1975.
Thermochemotherapy: synergism between hyperthermia (42-43) and adriamycin (or bleomycin) in mammalian cell inactivation. Proceedings of the National Academy of Sciences of USA, 72, 937-940. HELSON, L., 1984. Calcium channel blocker enhancement of anticancer drug cytotoxicity—a review. Cancer Drug Delivery, 1, 353-361.
546
adriamycin with small unilamellar vesicle liposomes. A fluorescence study. Biochimica et Biophysica Ada, 557, 306-319. KREMKAU, F. W., KAUFMANN, J. S., WALKER, M., BURCH, P. &
SPURR, C. L., 1976. Ultrasonic enhancement of nitrogen mustard cytotoxicity in mouse leukaemia. Cancer, 37, 1463-1647. SENEPATI, N., HOUCHENS, D., OVEJERA, A., BEARD, R. & NINES,
R., 1982. Ultrasonic hyperthermia and drugs as therapy for human tumour xenografts. Cancer Treatment Reports, 66, 1635-1639. SHOTTON, K. C , 1980. A tethered float radiometer for measuring the output power from ultrasonic therapy equipment. Ultrasound in Medicine and Biology, 6, 131-133. STRATFORD, I. J. & ADAMS, G. E., 1977. Effect of hyperthermia
on the differential cytotoxicity of a hypoxic cell radiosensitizer Ro-07-0582, on mammalian cells in vitro. British Journal of Cancer, 35, 307-313. WEST, C. M. L., 1983. The effect of cytotoxic drugs and radiation on mammalian cells and multicellular spheroids in vitro. PhD Thesis (University of London). YUMITA, N., OKUMURA, A., NISHIGAKI, R., VNEMURA, K. &
UMEMURA, S., 1987. The combination treatment of ultrasound and antitumour drugs on Yoshida Sarcoma. Japanese Journal of Hyperthermic Oncology, 3, 175-182.
The British Journal of Radiology, July 1990
