Selective Toxicity of Rhodamine 123 in Carcinoma Cells in Vitro

Home , Rhodamine 123, Trypan blue

[CANCER RESEARCH 43, 716-720, February 1983] 0008-5472/83/0043-0000502.00 Selective Toxicity of Rhodamine 123 in Carcinoma Cells in Vitro

Theodore J. Lampidis, 1 Samuel D. Bernal, 2 lan C. Summerhayes, and Lan Bo Chen a

Sidney Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts 02115

ABSTRACT ing appears to be unique, since chemotherapeutic agents currently in clinical use have not been shown to be particularly The study of mitochondria in situ has recently been facilitated selective between tumorigenic and normal cells in vitro. through the use of rhodamine 123, a mitochondrial-specific fluorescent dye. It has been found to be nontoxic when applied MATERIALS AND METHODS for short periods to a variety of cell types and has thus become an invaluable tool for examining mitochondrial morphology and Cell Cultures. All cell types and cell lines were grown in Dulbecco's function in the intact living cell. In this report, however, we modified Eagle's medium supplemented with 10% calf serum (M. A. demonstrate that with continuous exposure, rhodamine 123 Bioproducts, Walkersville, Md.) at 37 ~ with 5% CO2. The following cell selectively kills carcinoma as compared to normal epithelial lines were obtained from the American Type Culture Collection: cells grown in vitro. At doses of rhodamine 123 which were CCL 105, CCL 15, CRL 1550, CRL 1420, CCL51, CCL77, CCL185, toxic to carcinoma cells, the conversion of mitochondrial-spe- CCL 34, PtK-1, CV-1, and BSC-I. The human bladder carcinoma line cific to cytoplasmic-nonspecific localization of the drug was (E J) was provided by Dr. L. M. Frank (Imperial Cancer Research Fund). observed prior to cell death. At 10 /~g/ml, >50% cell death EJ and mouse bladder epithelial cells were isolated and grown as described previously (10). MCF-7 was provided by Dr. M. Rich (Mich- occurred within 7 days in all nine of the carcinoma cell types igan Cancer Foundation). Human breast epithelial cells were gifts from and lines of different origin studied, while six of six normal Dr. N. S. Yang (Michigan Cancer Foundation) and Dr. M. Stampfer epithelial cell types and lines remained unaffected. Cotreating (Peralta Cancer Research Institute). carcinoma cells with 2-deoxyglucose and rhodamine 1 23 en- Cytotoxicity and Clonogenic Survival Assays. For the cytotoxicity hanced the inhibition of growth by rhodamine 123 alone in assays, cells were seeded at 5 x 10 ~ cells/60-mm plate and incubated clonogenic survival assays. The observation of the selective at 37 ~ in 5% CO2. Rhodamine 123 (Eastman Kodak, Rochester, N. Y.) toxicity of rhodamine 1 23 appears to be unique in view of the was applied the following day at the indicated doses, and cell numbers absence of selective toxicity reported in vitro for the various were monitored at various periods during continuous drug exposure by antitumor agents currently in clinical use. Preliminary results trypsinizing the appropriate cultures and counting the number of cells with rhodamine 1 23 in animal tumor systems indicate antitumor by hemocytometer. The exclusion of trypan blue (0.2%) was used as an indicator of the number of live cells. For the clonogenic assays, 300 activity for carcinomas. cells were seeded in 60-mm dishes and incubated at 37 ~ in 5% CO2. On the following day, drug was applied at the indicated doses, and INTRODUCTION CFU ~ were determined 2 weeks later by staining with methylene blue (0.2%). Rhodamine 1 23, a cationic fluorescent dye, has been used Fluorescence Localization. Cells were grown on 12-mm round glass previously to specifically localize mitochondria in living cells coverslips (Rochester Scientific, Rochester, N. Y.), treated with rho- (9). It has been a convenient and useful tool for mitochondrial damine 123 for various time periods, and rinsed and mounted on glass studies in situ, since their detectability is enhanced and the dye slides with silicone rubber chambers in rhodamine-free medium as is relatively nontoxic in most of the cell types studied (1-3, 5- described previously (9). Stained cells were examined by epifluores- 9, 1 1, 1 2). Recently, however, rhodamine 1 23 has been found cent illumination at 485 nm on a Zeiss Photomicroscope III equipped with a Zeiss Planapochromat objective lens (x40). to inhibit oxidative phosphorylation in isolated mitochondria at high doses" and has been reported to have a cytostatic effect in L1 21 0 cells (4). Since we had observed that carcinoma cells RESULTS accumulate and retain more rhodamine 123 than do normal Chart 1 illustrates the difference in the effect of rhodamine epithelial cells, fibroblasts, myoblasts, chondrocytes, and lym- 123 on the growth of transformed and normal epithelial cell phocytes (1 1), we were prompted to investigate the cytotoxic lines. Chart 1, A and C, illustrate the marked difference be- effects of this agent on carcinoma cells. In this report, we tween treated and untreated cultures in 2 tumor-derived epi- demonstrate that rhodamine 1 23 selectively inhibits the growth thelial cell lines, human pancreatic carcinoma line CRL 1420 and kills carcinoma-derived epithelial cells in vitro, while the and human breast carcinoma line MCF-7. At 2 days of treat- growth of normal epithelial cells remains unaffected. This find- ment, a marked cytotoxic effect is seen in cell line CRL 1420, 1 Supported by a Young Investigators Award (CA24771). To whom requests while MCF-7 cells are inhibited in their growth. With further for reprints should be addressed, at Sidney Farber Cancer Institute, Harvard incubation, 3 days for CRL 1420 and 7 days for MCF-7, the Medical School, 44 Binney Street, Boston, Mass. 02115. 2 Supported by a Postdoctoral Fellowship (CA06943) from the National Cancer cell number is reduced to less than 103 per plate in each of the Institute. rhodamine-treated cultures. In marked contrast to these re- 3 Recipient of an American Cancer Society Faculty Research Award. Sup- ported by grants from the National Cancer Institute, CA 22427, and the American sults, CV-1 and PtK-1, derived from normal African green Cancer Society, CD 92B. monkey kidney (epithelial) and normal marsupial kidney (epi- "T. J. Lampidis, C. Salet, G. Moreno, and L. B. Chen. Effects of the mito- thelial), respectively, show no cytotoxicity when exposed to the chondrial probe rhodamine 123 and related analogues on the function and viability of pulsating myocardial cells in culture, submitted for publication. Received June 21, 1982; accepted November 5, 1982. The abbreviation used is: CFU, colony-forming units.

716 CANCER RESEARCH VOL. 43

same dose of rhodamine 123 (Chart 1, B and D). With even hanced by cotreating with 2-deoxyglucose. longer exposure (2 weeks at 10 #g/ml) or at higher doses (50 To determine whether the selective toxicity of rhodamJne /zg/ml for 7 days), cell growth of CV-1 and PtK-1 remains 123 is a general phenomenon shared by other tumorigenic relatively unaffected. epithelial cells, a number of carcinoma lines were studied. In the course of these experiments, it was noted that a Table 2 is a composite of 15 different cell lines tested for significant number of cells had detached from the culture dish sensitivity to rhodamine 123 when exposed continuously to surface after drug treatment (3 days for CRL 1420 and 7 days various doses. It can be seen that all 9 tumorigenic carcinoma- for MCF-7). We followed the fate of these cells in the presence derived epithelial cell lines used in this study were susceptible of drug with increasing time and found that they converted to the cytotoxic effects of rhodamine 1 23 at 10 or 50/zg/ml. gradually from trypan blue negative to trypan blue positive. In The nontumorigenic epithelial cell lines derived from normal addition, we have tested the ability of these detached cells, tissues were unaffected by treatment at these doses and con- while still trypan-blue negative, to proliferate in drug-free me- tinued to grow normally as untreated cells. Thus, a dose of dium and found that they are no longer able to reattach to the --<10/~g/ml, at which 50% of the treated cells were killed, was culture dish surface nor divide in suspension. found for each of the carcinoma cell lines tested, whereas the The selective toxicity of rhodamine 123 for carcinoma cells dose at which 50% of the treated normal epithelial cell lines was also observed when clonogenic survival assays were per- were killed was >50/zg/ml. To confirm the epithelial origin of formed. Chart 2 illustrates the difference in CFU between the cell lines tested, immunofluorescence staining with keratin rhodamine 1 23-treated normal epithelial (CV-1) and carcinoma antibody was used. All cell lines desiqnated in this report as (1420) cell lines. At 5 /~g/ml, the carcinoma cells form no epithelial or carcinoma were keratin positive. colonies while the normal cells remain relatively unaffected In conjunction with these experiments, each cell line was (CFU are 83% of control). Thus, although there is a slight tested for its sensitivity to the potent, broad-spectrum antitumor reduction in normal colony formation, the same pattern of agent Adriamycin. Although some of the carcinoma cell lines carcinoma sensitivity versus normal cell resistance to rhoda- did appear to be more sensitive to Adriamycin than did the mine 1 23 treatment as seen in the growth experiments prevails. untransformed lines, the complete insensitivity of normal epi- Since rhodamine 123 inhibits oxidative phosphorylation in thelial cells to rhodamine 1 23 at high doses (50 #g/ml) and for isolated mitochondria, 4 we were interested in testing whether prolonged exposure times (7 days) suggests a stronger selec- an inhibitor of glycolysis, 2-deoxyglucose, could affect the selective toxicity of rhodamine 123 for carcinoma cells. In Table 1, the results of clonogenic survival assays in MCF-7 100 (carcinoma) and CV-1 (normal) epithelial cells are illustrated. At concentrations of 2 x 10 -4 M 2-deoxyglucose, no effects 80 on CFU were observed. At rhodamine 123 (0.5/zg/ml), 47% of the CFU were inhibited. When both drugs were applied simul- O 60 taneously, the CFU were reduced to zero. In contrast, normal epithelial cells, which are not affected by rhodamine 123 (0.5 /~g/ml), showed no inhibition of CFU when exposed to both 401 U. drugs. Thus, the selective toxicity of rhodamine 1 23 was en- 20

D _.4 0

~ooo 0.1 1.0 10 / Rh-123 (;ug/ml) Chart 2. Effects of rhodamine 123 on CFU of CV-1, normal monkey kidney o~o Untreated epithelial cells (0), and CRL 1420, human pancreatic carcinoma cells (m). Each ,~1761 \ e-.---e Treated cell line was plated at 300 cells/60-mm dish and 1 day after plating cells were L i i i , L J , ,o_~ , , i L i i L I treated continuously with the above concentrations of rhodamine 123 in Dul- 4 8 12 4 8 12 4 8 12 4 8 12 becco's modified Eagle's medium supplemented with 10% calf serum. The cells Days were incubated at 37 ~ in 5% CO2 and at the end of 2 weeks the number of Chart 1. Cell counts of carcinoma lines CRL 1420 (A) and MCF-7 (C) and colonies was counted. Plating efficiencies were 70% for CV-1 and 80% for CRL untransformed epithelial cell lines CV-1 (B) and PtK-1 (D) continuously exposed 1420. The results are expressed as percentage of control, and standard error to rhodamine 123 (10/~g/ml). Points, average count of triplicate cultures. for duplicate samples was --<5%.

Table 1 Enhancement of rhodamine 123 inhibition of clonogenic survival by 2-deoxyglucose CFU Cell line Drug Dose (% of control) MCF-7 (carcinoma) 2-Deoxyglucose 3 x 10-4M 102 +_ 6a Rhodamine 123 0.5/~g/ml 53 _+ 3 2-Deoxyglucose + rhodamine 123 3 x 10 -4 M + 0.5/~g/ml 0

CV-1 (normal) 2-Deoxyglucose 3 x 10 -4 M 101 4- 6 Rhodamine 123 0.5/~g/ml 100 +_ 3 2-Deoxyglucose + rhodamine 123 3 x 10 -4 M + 0.5 p.g/ml 100 +_ 5 a Mean _+ S.E.

FEBRUARY 1983 71 7

Table 2 Selective killing by rhodamine 123 in carcinoma versus normal epithelial cells Days of treatment required to produce Conversionof -->50% cell death rhodamine 123 rhodamine123 staining from mitochondrial- specific to cy- Type nonspecific 10/~g/ml a 50/~g/ml a toplasmic Carcinoma CRL 1420 (human pancreatic) 3 2 Yes EJ (human bladder) 3 2 Yes CCL 77 (mouse-Ehrlich-Lettre ascites) 3 2 Yes CCL 105 (human adrenal cortex) 3 3 Yes CCL 51 (mouse mammary) 5 5 Yes MCF-7 (human breast) 7 3 Yes CCL 15 (Syrian hamster kidney) 7 4 Yes CCL 185 (human lung) 7 6 Yes CRL 1550 (human cervix) b 6 Yes

Normal epithelial CV-1 (monkey kidney) No effect (7)c (7) No BSC-1 (monkey kidney) No effect (7) (7) No CCL 34 (canine kidney) No effect (7) (7) No PtK-1 (marsupial kidney) No effect (7) (7) No Tertiary (human breast epithelial) No effect (7) (7) No Primary (mouse bladder epithelial) No effect (7) (7) NTd a Applied continuously. b Growth inhibited. c Numbers in parentheses, number of days of treatment at which no cytotoxic effects could be observed. Cell viability was monitored by trypan blue exclusion. d NT, Not tested.

tivity for rhodamine 123 than for Adriamycin in the cell lines mine 1 23 are essentially unable to replicate and subsequently tested in this study. undergo cell death. Previously, it had been reported that, with short exposure times (10 to 30 min), rhodamine 1 23 localized and accumulated DISCUSSION specifically in mitochondria of live cells (1). However, with prolonged exposure (4 to 24 hr), we have found that rhodamine Since rhodamine 123 localizes at the mitochondria in the 123 localization in carcinoma cells is no longer restricted to intact cell and inhibits oxidative phosphorylation in isolated mitochondria. Fig. 1 illustrates dye localization for MCF-7 ex- mitochondria in vitro, 4 it is conceivable that its cytotoxicity may posed to rhodamine 123 (10/~g/ml) for 10 min (a) and 4 hr be attributable, at least in part, to this inhibitory action in (b), respectively. The change in dye localization from mito- mitochondria. The enhanced inhibition of CFU we report with chondrial specific (10 min) to cytoplasmic nonspecific (4 hr) is 2-deoxyglucose in combination with rhodamine 123 as com- restricted to the carcinoma cell lines tested in this study (Table pared to rhodamine alone is consistent with this interpretation. 2). Fig. I c illustrates the maintenance of mitochondrial-specific Since 2-deoxyglucose inhibits glycolysis, which serves as an- dye localization in a nontumorigenic cell line (CV-1) exposed other source of ATP, the cell may become compromised more to rhodamine 123 (10 #g/ml) for 7 days. An apparent correla- quickly by no longer being able to produce enough ATP by tion between the susceptibility to the cytocidal effect of rho- either pathway to maintain its metabolic integrity. damine 123 and the change from mitochondrial-specific to However, the molecular basis for the selective toxic effects cytoplasmic-nonspecific localization after prolonged exposure of rhodamine 1 23 in carcinoma as compared to normal epithe- is noted in Table 2. lial cells remains to be investigated. One possibility is that In Fig. 2, the various stages of dye localization in MCF-7 carcinoma cells contain rhodamine 123-binding sites which exposed to rhodamine 123 for 24 hr are illustrated. The cell on allow the dye to accumulate more readily than in untransformed the far left maintains the mitochondria-specific fluorescence epithelial cells which contain fewer such sites. Another possi- indicative of a viable, unaffected cell, while the cells in the bility relates to the electrical potential of the cell. It has been middle illustrate what we believe to be damaged and dying shown previously that only cationic permeant of compounds cells. Two kinds of cells that contain nonspecific cytoplasmic are specifically accumulated by mitochondria in situ (8). It was staining are apparent, those that remain flattened and elon- suggested that the high electric potential across the mitochon- gated (cell A) and those that round up (cell B). Cells first seen drial membrane (negative inside) accounts for the attraction to contain nonspecific cytoplasmic staining (2 to 4 hr of contin- and accumulation of these positively charged molecules. Thus, uous exposure) remain stretched out. With continuous expo- it is conceivable that carcinoma cells in comparison to normal sure, they round up, gradually become trypan blue positive, epithelial cells may have a higher mitochondrial or plasma and finally lyse. In the course of this dying process, cells lose transmembrane potential or both. Consequently, more posi- all of their rhodamine 123-specific fluorescence prior to be- tively charged dye would be accumulated and retained in coming trypan blue positive (cell C). Bernal et al. (1) have carcinoma than in normal epithelial cells (1 1 ). recently found that all cells that are not stainable with rhoda- This latter possibility may also be related to the striking

718 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1983 American Association for Cancer Research. Rhodamine 123 Kills Carcinoma Cells in vitro difference between carcinoma and untransformed epithelial a median survival time of 260% for Ehrlich ascites tumor and cell lines in the ultimate localization of rhodamine 123 after 140% for MB49 as compared to untreated controls. prolonged treatment. The phenomenon of nonspecific cyto- REFERENCES plasmic staining, which we describe in Table 2, at doses toxic 1. Bernal, S. D., Shapiro, H. M., and Chen, L. B. Monitoring the effect of anti- to carcinoma cells may be due, in part, to the inhibitory action cancer drugs on L1210 cells by a mitochondrial probe, rhodamine 123. Int. J. Cancer, 30:219-224, 1982. of rhodamine 123 on mitochondria. Johnson et al. (8) and 2. Chen, L. B., Summerhayes, I. C., Johnson, L. V., Walsh, M. L., Bernal, S. D. others later (5) have demonstrated that cells exposed for short and Lampidis, T. J. Probing mitochondria in living cells with rhodamine 123_ periods to rhodamine 123 and subsequently treated with Cold Spring Harbor Symp. Quant. Biol., 46. 141-155, 1982. 3. Darzynkiewicz, A., Staiano-Coico, L., and Melamed, M. R. Increased mito- agents that should dissipate mitochondrial transmembrane po- chondrial uptake of rhodamine 123 during lymphocyte stimulation. Proc. tential such as carbonyl cyanide p-trifluoromethoxyphenylhy- Natl. Acad. Sci. U. S. A., 78: 2383-2387, 1981. drazone, dinitrophenol, and valinomycin, immediately lose their 4. Darzynkiewicz, Z., Traganos, F., Staiano-Coico, L., Kapuscinski, J., and Melamed, M. R. Interactions of rhodamine 123 with living cells studied by mitochondrial staining and cytoplasmic fluorescence in- flow cytometry. Cancer Res., 42: 799-806, 1982. creases. Carcinoma cells treated for prolonged periods with 5. Goldstein, S., and Korczack, L. B. Status of mitochondria in living human rhodamine 123 alone mimic the above phenomenon. In turn, fibroblasts during growth and senescense. Use of the laser dye rhodamine 123. J. Cell Biol., 91: 392-398, 1981. the function of the plasma membrane, nucleus, or other cyto- 6. James, W. T., and Bohman, R. J. Proliferation of mitochondria during the plasmic organelles in addition to mitochondria could be af- cell cycle of the human cell line (HL-60). J. Cell Biol., 88. 256-260, 1981. 7. Johnson, L. V., Summerhayes, I. C., and Chen, L. B. Decreased uptake and fected by the accumulation of the dye. Finally, any one or retention of rhodamine 123 by mitochondria in feline sarcoma virus-trans- combination of these considerations may contribute to cell formed mink cells. Cell, 28:7-14, 1982. death. 8. Johnson, L. V., Walsh, M. L., Bockus, B. J., and Chen, L. B. Monitoring of relative mitochondrial potential in living cells by fluorescence microscopy. J. Generally, the cytotoxicity of new antitumor agents is evalu- Cell Biol., 88: 526-535, 1981. ated by clonogenic survival or growth inhibition assays. Rarely 9. Johnson, L. V., Walsh, M. L., and Chert, L. B. Localization of mitochondria have studies attempted to distinguish carcinoma from normal in living cells with rhodamine 123. Proc. Natl. Acad. Sci. U. S. A., 77: 990- 994, 1980. cells with respect to drug susceptibility. Since rhodamine 1 23 10. Summerhayes, I. C., and Franks, L. M. The effect of donor age on neoplastic has shown a strong selectivity in killing carcinoma cells in vitro, transformation of adult mouse bladder epithelium in vitro. J. Natl. Cancer Inst., 62: 1017-1023, 1979. the efficacy of using it as a chemotherapeutic agent in vivo is 11. Summerhayes, I. C., Lampidis, T. J., Bernal, S. D., Nadakavukaren, J. J., now being tested in animals. In preliminary experiments, mice Nadakavukaren, K. K., Shepherd, E. L., and Chen, L. B. Unusual retention bearing Ehrlich ascites tumor and mouse bladder carcinoma of rhodamine 123 by mitochondria in muscle and carcinoma cells. Proc. Natl. Acad. Sci. U. S. A., 79: 5292-5296, 1982. (MB49), respectively, were given i.p. injections of rhodamine 12. Ziegler, M. L., and Davidson, R. L. Elimination of mitochondrial elements 123 (15 mg/kg) on Days 1, 3, and 5. This regimen produced and improved viability in hybrid cells. Somatic Cell Genet., 7: 73-88, 1981.

FEBRUARY 1983 719

Fig. 2. MCF-7 cells grown on glass discs, treated with rhodamine 123 for 24 hr, rinsed for 10 rain, and mounted on rubber chambers. (a) cells observed under fluorescent optic; (b) identical field under phase contrast. Note the cytoplasmi- cally stained flattened cell (A) and rounded cell (B) (both alive) and the rounded, unstained cell (C) (dead) which represent further stages in the course of carcinoma cell killing by rhodamine 123. Bar, 20/~m.

Fig. 1. Human breast carcinoma cells, MCF-7, were exposed to rhodamine 123 (10/~g/ml) for 10 min (a) and 4 hr (b), respectively, rinsed, mounted on rubber chambers as described previously (1) in drug-free medium, and examined under fluorescence microscopy. Note the conversion from mitochondrial-specific to cytoplasmic-nonspecific staining. CV-1 (untransformed epithelial) cells were exposed to rhodamine 123 for 7 days (c) and were rinsed, mounted, and examined as above. Note the maintenance of mitochondrial-specific staining. Bars: 15 ,u,m (a); 12 #m (b); 10 #m (c).

720 CANCER RESEARCH VOL. 43

Theodore J. Lampidis, Samuel D. Bernal, Ian C. Summerhayes, et al.

Cancer Res 1983;43:716-720.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/2/716

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/43/2/716. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.