930 Intracellular Accumulation of Thallium as a Marker of Cisplatin Cytotoxicity in Nonsmall Cell Lung Carcinoma An Application of Inductively Coupled Plasma Mass Spectrometry Taro Hanada, M.D.1 Hiroshi Isobe, M.D.2 Takeshi Saito, Ph.D.3 Shigeaki Ogura, M.D.1 Hironori Takekawa, M.D.1 Kohichi Yamazaki, M.D.1 Yoshio Tokuchi, M.D.1 Yoshikazu Kawakami, M.D.1 1 First Department of Medicine, School of Medicine, Hokkaido University, Sapporo, Japan. 2 Department of Pulmonary Diseases and Clinical Research Institute, Sapporo National Hospital, Sapporo, Japan. 3 Hygene and Preventive Medicine, School of Medicine, Hokkaido University, Sapporo, Japan. Presented in part at the 37th Annual Meetings of the Japan Lung Cancer Society, Kobe, Japan, October 31 to November 1, 1996. The authors thank Dr. Kazuo Saito and the Hygiene and Preventive Medicine fellows for assisting in the ICP-MS and for their help in the design of this study. Address for reprints: Taro Hanada, M.D., Department of Internal Medicine, Tenshi Hospital, North 12, East 3, Higashi-ku, Sapporo, 065-8611, Japan. Received 27 October, 1997; revision received February 17, 1998; accepted February 17, 1998. © 1998 American Cancer Society BACKGROUND. Thallium-201 (201Tl) scintigraphy has been used to detect malignant pulmonary disease. The mechanism of Tl influx in tumor cells is believed to be similar to that of cisplatin (CDDP) mediated by sodium- and potassium-activated adenosine triphosphatase (Na-K ATPase), and the Na-K ATPase activity may determine the cellular CDDP accumulation and sensitivity to CDDP. The objective of this study was to determine the accumulation of CDDP and Tl in vitro by using inductively coupled plasma mass spectrometry (ICP-MS), a new analytic technique for detecting ultra trace elements, and to evaluate the correlations between cellular CDDP and Tl accumulation, between CDDP 50% inhibitory concentration (IC50) values and cellular CDDP accumulation, and between CDDP IC50 values and cellular Tl accumulation. METHODS. Eight nonsmall cell lung carcinoma (NSCLC) cell lines were used (five adenocarcinomas and three squamous cell carcinomas). The cell lines were exposed to CDDP or Tl for 1 hour, and the resulting cellular accumulation of platinum and Tl was determined by ICP-MS. CDDP IC50 values were determined by a soluble tetrazolium/formazan assay. RESULTS. The authors were able to measure cellular CDDP and Tl accumulation precisely, and heterogeneity in the cellular accumulation of CDDP and Tl existed among the NSCLC cell lines. A significant inverse correlation was observed between CDDP IC50 values and the cellular accumulation of both CDDP and Tl. CONCLUSIONS. ICP-MS is suitable for the determination of cellular CDDP and Tl accumulation in NSCLC cell lines. Cellular Tl accumulation determined by ICP-MS may reflect CDDP cytotoxicity rather than cellular CDDP accumulation. Cancer 1998;83:930 –5. © 1998 American Cancer Society. KEYWORDS. cisplatin accumulation, thallium accumulation, inductively coupled plasma mass spectrometry, Na-K-activated adenosine triphosphatase, thallium-201 scintigraphy. hallium-201 (201Tl) is used to diagnose myocardial infarction,1 myocardial ischemia,2 and thyroid tumor.3 Recently, 201Tl single photon emission computed tomography (SPECT) has been used for the detection of malignant pulmonary lesions4 and has been shown to be superior to gallium scintigraphy for lung carcinoma detection.5 The mechanism of 201Tl influx in tumor cells is believed to be active transport through the cell membrane mediated by the sodiumand potassium-activated adenosine triphosphatase (Na-K ATPase) system.6 – 8 In a similar manner, the Na-K ATPase system may also contribute to the cellular influx of cisplatin (CDDP).9 On the other T Tl Accumulation Determined by ICP-MS/Hanada et al. hand, Morikage et al.10 found that the expected CDDP accumulation was reduced consistently in a number of cell lines that had CDDP resistance. From these findings, we speculated that a lung tumor that has a high degree of 201Tl uptake may also accumulate a high degree of CDDP and, therefore, may have a high CDDP sensitivity. In fact, in our previous study on small cell lung carcinoma (SCLC) patients who had undergone 201Tl SPECT, we found that 201Tl accumulation correlated to patient response to chemotherapy of platinum compounds.11 Based on this clinical study, we hypothesized that a significant correlation exists between the in vitro accumulations of CDDP and Tl. To evaluate this hypothesis, we investigated the cellular accumulation of CDDP and Tl in nonsmall cell lung carcinoma (NSCLC) cell lines by using inductively coupled plasma mass spectrometry (ICP-MS). ICP-MS is a powerful analytical technique for the detection of ultratrace elements in biological materials12 and has been applied to the analysis of the trace element composition of blood,13 multielement analysis of human tissue,14 and metabolic studies of minerals in humans.15 In a previous study, we were able to measure cellular CDDP accumulation by using ICP-MS.16 The objective of the present study was to analyze the cellular accumulation of CDDP and Tl by using ICP-MS and to determine the correlation between cellular accumulation of CDDP or Tl and the chemosensitivity to CDDP in NSCLC cell lines. MATERIALS AND METHODS 931 lung adenocarcinoma cell lines and EBC-1, LK-2 and PC-10 squamous lung carcinoma cell lines. Cell lines were obtained from the Japanese Cancer Research Resources Bank (Tokyo, Japan). All cell lines were cultivated in tissue culture flasks (Falcon; Beckton Dickinson Labware, Franklin Lakes, NJ) in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 100 units/mL penicillin, and 100 mg/mL streptomycin. Chemosensitivity Test Chemosensitivity to CDDP was determined by using the 2,3-bis(2-methoxy-4-nitro-5-sulphphenyl)-5-[(phenylamino)carbonyl]-2H-tetrazolium hydroxide, inner salt, sodium salt (XTT) assay, as described previously.18 Briefly, single-cell suspensions were harvested during the exponential growth phase and seeded into 96-well microplates. The cells were preincubated at 37°C overnight, after which, various concentrations of the required drugs were added. The cells were then incubated at 37°C for 96 hours. After incubation, 50 mg of XTT (Sigma Chemical Company) with 0.38 mg of phenazine methosulfate (PMS; Sigma Chemical Company) were added to each well and incubated at 37°C for 4 hours. The plates were agitated on a plate shaker for 5 minutes to solubilize the formazan crystals. Their absorbance was measured at 450 by nm using an enzyme-linked immunosorbent assay (ELISA) reader. The 50% inhibitory concentration (IC50) value was defined by determining the drug concentration that reduced absorbance by 50%. This value was derived graphically from the concentration-response curve. Instrumentation The ICP-MS apparatus used was an SPQ6500 (Seiko Instruments, Co., Shizuoka, Japan), with a nebulizer and a spray chamber made of borosilicate glass. Samples are introduced into a nebulizer, and positively charged ions are then produced by a high-temperature, inductively coupled plasma. The ions pass through a sampling cone interface into a high-vacuum, quadruple mass spectrometer, which has the ability to carry out multielement analysis.17 Drugs and Chemicals CDDP was purchased from Sigma Chemical Company (St. Louis, MO). Thallium chloride, standard solutions of platinum (Pt), and Tl were purchased from Wako Pure Chemical Company (Osaka, Japan). RPMI 1640 was purchased from Nissui Pharmaceutical Company (Tokyo, Japan). Cell Lines The human lung cancer cell lines used were A549, PC-3, RERF-LC-MS, RERF-LC-OK, and VMRC-LCD Drug Treatment The cells were incubated in 10-cm tissue culture plates (Falcon 3003; Beckton Dickinson Labware) at 5 3 106 cells in the 10-mL culture medium with 200 mL of 0.5 mg/mL CDDP (10 mg/mL) or 200 mL of 1.0 mg/mL TlCl (20 mg/mL). After 1 hour of incubation at 37°C, cells were harvested by scraping, transferred to plastic tubes, and washed, using centrifugation and resuspension, two times with cold phosphate-buffered saline. The cell pellets were resuspended in 1 mL of 0.9% sodium chloride and were taken up into Teflon resolution vessels (Flon Industry Company, Tokyo, Japan). The vessels were placed in a vacuum oven and heated at 120 °C for 2 hours. After cooling, the samples were diluted to 10 mL with Millipore MilliQ water (Millipore Corp., Bedford, MA). A CDDP concentration of 10 mg/mL, a TlCl concentration of 20 mg/mL, and an incubation time of 1 hour were employed, because the reproducibility of cellular CDDP and Tl accumulation were good under these experimental conditions using ICP-MS. 932 CANCER September 1, 1998 / Volume 83 / Number 5 TABLE 1 Resistance of Nonsmall Cell Lung Carcinoma Cell Lines to Cisplatin Cell lines Adenocarcinoma A549 PC-3 RERF-LC-MS RERF-LC-OK VMRC-LCD Squamous cell carcinoma EBC-1 LK-2 PC-10 a IC50 values (mM) 21.9 6 3.1a 8.3 6 1.5 2.8 6 1.4 7.8 6 0.8 20.6 6 9.3 17.0 6 1.2 17.1 6 4.0 4.2 6 0.6 Mean 6 S.D. (n $ 3). IC50: 50% inhibitory concentration. Inductively Coupled Plasma Mass Spectrometry The digested samples were nebulized into ICP-MS without further dilution. Quantification of all samples was performed by using external standard solutions made up in 1% nitric acid. The standard solutions used were 10, 50, and 100 mg/mL Pt or Tl, and the blank solution was 0.1% nitric acid. Protein Assay Protein was measured by using the bicinchoninic acid protein assay (Pierce Chemical Company, Rockford, IL). Statistics A calibration curve and a correlation coefficient between the CDDP IC50 values and cellular CDDP accumulation and between the CDDP IC50 values and cellular Tl accumulation were determined to test for the significance of the regression. Only P values ,0.05 were considered significant. RESULTS Cisplatin Sensitivity Chemosensitivity to CDDP in NSCLC cell lines was evaluated (Table 1). IC50 values among the NSCLC cell lines ranged from 2.8 mM to 21.9 mM by XTT assay. Calibration Curve The calibration curve was obtained by using four standard Pt and Tl solutions (Fig. 1). The correlation coefficient of the regression line was 0.999 among 0, 10, 50, and 100 mg/mL of Pt and Tl. Accumulation of CDDP and Tl in NSCLC Cell Lines By using ICP-MS, the cellular accumulation of CDDP and Tl could be determined after 1 hour of exposure to CDDP and Tl. Cellular accumulations of CDDP ranged from 5.4 3 1023 ng/mg protein to 2.6 3 1022 ng/mg protein, and Tl ranged from 1.7 3 1021 ng/mg protein to 8.8 3 1021 ng/mg protein (Fig. 2). The cellular accumulation of CDDP and Tl were heterogeneous among NSCLC cell lines. Correlations between Cellular CDDP and Tl Accumulation, between CDDP IC50 Values and Cellular CDDP Accumulation, and between CDDP IC50 Values and Cellular Tl Accumulation A significant inverse correlation between the CDDP IC50 values and its cellular accumulation was observed in the NSCLC cell lines (Fig. 3). A similar inverse correlation was obtained between the CDDP IC50 values and Tl cellular accumulation (Fig. 4). No significant correlation was observed between the cellular CDDP accumulation and Tl accumulation (data not shown). DISCUSSION Na-K ATPase has a number of features in addition to its contribution to the intracellular influx system of Tl. Kier19 reported that Na-K ATPase activities were higher in plasma membranes from metastatic cells than in those from primary tumor cells and proposed that this Na-K ATPase increase was related to the cell surface fluidity and metastatic ability of metastatic cells. On the other hand, Ohmori et al.9 reported that CDDP accumulation, Na-K ATPase activity, and Na-K ATPase mRNA expression were all lower in CDDPresistant NSCLC cell lines than in their parental cell lines. Accordingly, the authors speculated that Na-K ATPase plays a crucial role in the transport of CDDP and that decreased intracellular accumulation of CDDP is the primary factor of CDDP resistance. Based on these findings, we hypothesized and demonstrated previously that 201Tl accumulation within a tumor is closely related to that tumor’s lymph node metastasis20 or response to chemotherapy with CDDP. These results suggested that 201Tl SPECT might be useful as an indicator of metastasis in adenocarcinoma of the lung and as a predictor of chemotherapeutic responses to platinum compounds in SCLC patients. CDDP is an important anticancer drug for the treatment of many solid tumors, but intrinsic or acquired resistance to CDDP is a significant clinical problem, leading to treatment failure in NSCLC patients. If we could predict the chemosensitivity of the NSCLC patient to CDDP by using 201Tl scintigraphy, then we could avoid giving chemotherapy to a nonresponder. Because the patients with limited disease were treated by operation, and, for extensive disease, it was difficult to reach complete remission or partial remission in NSCLC patients, we were unable to ex- Tl Accumulation Determined by ICP-MS/Hanada et al. 933 FIGURE 1. Calibration curve for platinum (Pt; A) and thallium (Tl; B) determined by inductively coupled plasma mass spectrometry. The correlation coefficient of the regression line was 0.999 among 0, 10, 50, and 100 mg/mL of Pt and Tl. Because the standard deviations were less than the size of the symbol, error bars do not appear for dependent variables. conc., concentration. FIGURE 2. Cellular accumulation of cisplatin (CDDP; A) and thallium (Tl; B) after a 1 hour exposure to CDDP and Tl using inductively coupled plasma mass spectrometry. Heterogeneity in the cellular accumulation of CDDP and Tl was present among nonsmall cell lung carcinoma (NSCLC) cell lines. Each point represents the mean value from three or more independent experiments. Error bars indicate standard deviation. FIGURE 3. Correlation between accumulation of cisplatin (CDDP) and chemosensitivity to CDDP. A significant inverse correlation between the CDDP 50% inhibitory concentration (IC50) values and its cellular accumulation was observed in nonsmall cell lung carcinoma cell lines. Each point represents the mean value from three or more independent experiments. Error bars indicate standard deviation. amine the relation between 201Tl accumulation and responses to chemotherapy with platinum compounds in NSCLC patients. In this paper, we therefore investigated whether there was a significant correlation between CDDP IC50 and Tl accumulation in NSCLC cell lines. Although the most widely used techniques for the evaluation of cellular accumulation of CDDP and Tl in vitro are atomic absorption spectrometry (AAS), radiolabeled CDDP ([195mPt]-CDDP), and radiolabeled thallium chloride (201TlCl),8 –10,21,22 we employed ICP-MS to evaluated the cellular accumulation of CDDP and Tl in this study. The accuracy of ICP-MS was superior to AAS in the analysis of reference materials,14 and the sensitivity of ICP-MS was 100-fold greater than that of AAS.17 The accuracy and the sensitivity of radiolabeling methodology are not known. However, radiolabeled tracers are not widely available and involve safety hazards. Because ICP-MS offers a wide linear range (six orders of magnitude) and a low detection limit,17,23 the linearity of the calibration curve obtained by using 934 CANCER September 1, 1998 / Volume 83 / Number 5 FIGURE 4. Correlation between accumulation of thallium (Tl) and chemosensitivity to cisplatin (CDDP). A significant inverse correlation between the CDDP 50% inhibitory concentration (IC50) values and cellular accumulation of Tl was observed with NSCLC cell lines. Each point represents the mean value from three or more independent experiments. Error bars indicate standard deviation. four standard Pt and Tl solutions was excellent. Moreover, once the calibration curve was obtained, the cellular accumulation of CDDP and Tl in several NSCLC cell lines could be analyzed at one time despite CDDP or Tl heterogeneity. In fact, it took less than 1 minute to analyze each sample in the present study. Because the NSCLC cell lines that were not exposed to CDDP or Tl scarcely contained Pt and Tl, we could determine their cellular CDDP and Tl accumulation precisely. Therefore, ICP-MS not only is suitable for the determination of cellular CDDP and Tl in tumor cells but also is applicable for determining the cellular accumulation of Tl in thyroid cells or myocardial cells in vitro. In a study using radiolabeled CDDP, Morikage et al.10 observed an inverse correlation between CDDP IC50 values and CDDP cellular accumulation in NSCLC cell lines and proposed that defective CDDP accumulation may play an important role in the natural resistance of NSCLC cells to CDDP. Our previous study using ICP-MS did not show a significant inverse correlation between the CDDP IC50 values and CDDP cellular accumulation.16 We speculated that this was due to the presence of amphotericin B in the culture medium, because it has been shown to increase CDDP cytotoxity by increasing cellular CDDP accumulation in several NSCLC cell lines.10 However, in the present study, we used a culture medium without amphotericin B, and, in good agreement with Morikage et al.,10 we found a significant inverse correlation between CDDP IC50 values and CDDP accumulation within NSCLC cell lines. Our present and past11 results suggest that intracellular accumulation of Tl could be a suitable marker of CDDP cytotoxity. On the other hand, no significant correlation was observed between the cellular accumulation of CDDP and Tl. Although the influx systems of both CDDP and Tl are mediated by Na-K ATPase, the intracellular metabolic system of CDDP is different from that of Tl. The mechanism of Tl metabolism has been thought to involve an affinity to sulfhydryl groups7,24,25 and a tendency to combine with riboflavin.25 In contrast, the metabolism of CDDP seems to be related to the following mechanisms: enhanced inactivation by cellular detoxification systems, such as glutathione (nonprotein-bound sulfhydryl groups) and metallothionein (protein-bound sulfhydryl groups), and decreased DNA damage and/or increased repair.26 These differences between the CDDP and Tl metabolic systems may contribute to the lack of correlation between CDDP and Tl accumulation. Although this is the first report demonstrating a significant inverse correlation between CDDP IC50 values and cellular Tl accumulation, the correlation was fairly weak. The weakness of the correlation was due to the fact that the CDDP IC50 values of the eight cell lines used were close to each other, so more experience is needed using a variety of cell lines that have heterogenous CDDP IC50 values. If a good correlation is obtained from a variety of NSCLC cell lines, then 201 Tl scintigraphy may be useful not only for the detection of malignant tumors but also for predicting the response of NSCLC patients to chemotherapy with platinum compounds and the detection of acquired resistance to platinum compounds in the process of chemotherapy. REFERENCES 1. 2. Strauss HW, Harrison K, Langan JK, Lebowitz E, Pitt B. Thallium-201 for myocardial imaging: relation of thallium201 to regional myocardial perfusion. Circulation 1975;51: 641–5. Strauss HW, Boucher CA. Myocardial perfusion studies: lessons from a decade of clinical use. Radiology 1986;160:577– 84. Tl Accumulation Determined by ICP-MS/Hanada et al. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Ochi H, Sawa H, Fukuda T, Inoue Y, Nakajima H. Thallium201-chloride thyroid scintigraphy to evaluate benign and/or malignant nodules. Cancer 1982;50:236 – 40. Tonami N, Shuke N, Yokoyama K, Seki H, Takayama T, Kinuya S, et al. Thallium-201 single photon emission computed tomography in the evaluation of suspected lung cancer. J Nucl Med 1989;30:997–1004. Matsuno S, Tanabe M, Kawasaki Y, Satoh K, Urrutia AE, Ohkawa M, et al. Effectiveness of planar image and single photon emission tomography of thallium-201 compared with gallium-67 in patients with primary lung cancer. Eur J Nucl Med 1991;19:86 –95. Britten JS, Blank M. Thallium activation of the (Na1-K1)activated ATPase of rabbit kidney. Biochim Biophys Acta 1968;159:160 – 6. Douglas KT, Bunni MA, Baindur SR. Thallium in biochemistry. Int J Biochem 1990;22:429 –38. Sehweil AM, Mckillop JH, Milroy R, Wilson R, Abdel-Dayem HM, Omar YT. Mechanism of 201Tl uptake in tumours. Eur J Nucl Med 1989;15:376 –9. Ohmori T, Nishio K, Ohta S, Kubota N, Adachi M, Komiya K, et al. Ouabain-resistant non-small cell lung-cancer cell line shows collateral sensitivity to cis-diamminedichloroplatinum(II) (CDDP). Int J Cancer 1994;57:111– 6. Morikage T, Ohmori T, Nishio K, Fujiwara Y, Takeda Y, Saijo N. Modulation of cisplatin sensitivity and accumulation by amphotericin B in cisplatin-resistant human lung cancer cell lines. Cancer Res 1993;53:3302–7. Tokuchi Y, Isobe H, Takekawa H, Hanada T, Ogura S, Itoh H, et al. Predicting chemotherapeutic response to small cell lung cancer of platinum compounds by thallium-201 SPECT. Br J Cancer 1998;77:1363– 8. Vanhoe H. A review of the capabilities of ICP-MS for trace element analysis in body fluids and tissues. J Trace Elem Electrolytes Health Dis 1993;7:131–9. Lutz TM, Nivel PMV, Schmidt B. Whole blood analysis by ICP-MS. In: Application of plasma source mass spectrometry. Holland H, Eaton AN, editors. Cambridge: Royal Society of Chemistry, 1991:96 –100. Lyon TDB, Fell GS. Accuracy of multi-element analysis of human tissue obtained at autopsy using inductively coupled mass spectrometry. J Anal At Spectrom 1991;6:559 – 64. Sun XF, Ting WTG, Ziesel SH, Janghorbani M. Accurate measurement of stable isotopes of lithium by induc- 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 935 tively coupled plasma mass spectrometry. Analyst 1987; 112:1223– 8. Hanada T, Isobe H, Saitoh T, Ogura S, Saitoh K, Kawakami Y. Inductively coupled plasma mass spectrometry for the determination of platinum accumulation in human non-small lung cancer cell lines. Int J Clin Oncol 1998;3:98 –101. Tothill P, Matheson LM, Mckay K, Smyth JF. Inductively coupled plasma mass spectrometry for the determination of platinum in animal tissues and a comparison with atomic absorption spectrometry. J Anal At Spectrom 1990;5:619 –22. Scudiero DA, Shoemaker RH, Paull KD, Monks A, Siobhan T, Nofziger TH, et al. Evaluation of a soluble tetrazoliium/ formazan assay for cell growth and drug sensitivity in culture using human and other tumor cell lines. Cancer Res 1988;48:4827–33. Kier AB. Plasma membrane properties of cultured local LM cell tumors and metastases from athymic (nude) mice. Cancer Lett 1990;50:19 –30. Takekawa H, Itoh K, Abe S, Ogura S, Isobe H, Sukou N, et al. Retension index of thallium-201 single phonton emission computerised tomography (SPECT) as an indicator of metastasis in adenocarcinoma of the lung. Br J Cancer 1994;70: 315– 8. Maublant JC, Zhang Z, Rapp M, Ollier M, Michelot J, Veyre A. In vitro uptake of Technetium-99m-teboroxime in carcinoma cell lines and normal cells: comparison with Technetium-99m-sestamibi and thallium-201. J Nucl Med 1993;34: 1949 –52. Ohmori T, Morikage T, Sugimoto Y, Fujiwara Y, Kasahara K, Ohta S, et al. The mechanism of the difference in cellular uptake of platinum derivatives in non-small cell lung cancer cell line (PC-14) and its cisplatin-resistant subline (PC-14/ CDDP). Jpn J Cancer Res 1993;84:83–92. Yoshinaga J. Inductiverly coupled plasma atomic emission spectrometry and ICP mass spectrometry [in Japanese]. Nippon Rinsho 1996;54:202– 6. Saddique A, Peterson C. Thallium poisoning: A review. Vet Hum Toxicol 1983;25:16 –22. Chandler HA, Scott M. A review of thallium toxicology. J R Nav Med Serv 1986;72:75–9. Andrews PA, Howell SB. Cellular pharmacology of cisplatin: perspectives on mechanisms of acquired resistance. Cancer Cells 1990;2:35– 43.