- Email: [email protected]
Cytotoxicity of commonly used nitroxide radical spin probes
Life Sciences, Vol. 40, pp. 495-498 Printed in the U.S.A.
Pergamon Journal~
CYTOTOXICITY OF COMMONLY USED NITROXIDE RADICAL SPIN PROBES Else G. Ankel,l, 2 Ching-San Lai,l, * Larry E. Hopwood 2 and Zorica Zivkovic 2 INational Biomedical ESR Center, Departments of Radiology and 2Radiation Oncology, Medical College of Wisconsin 8701 Watertown Plank Road, Milwaukee, Wisconsin 53226 (Received in final form November 3, 1986) Summary Since nitroxide radical spin probes are used frequently to test biophysical properties of cells, their use should be restricted to conditions that do not perturb normal cell growth and viability. Eight commonly used nitroxide radical spin probes have been tested for their effects on the survival of CHO cellst These include watersoluble spin probes Tempol, Tempamine, CTPO~ CTPC and 4-maleimidoTempo, and lipid soluble spin probes 5-Doxyl-~ 12-Doxyl-, and 16Doxylstearates. With the exception of 4-maleimido-Tempo, none of the water soluble spin labels inhibited cell survival at concentrations as high as I mM. At concentrations of 75 uM and higher, 4maleimido-Tempo inhibited cell survival in a dose dependent manner. At concentrations commonly used for spin labeling of cells (30-50 ~M) none of the lipid soluble spin probes tested was cytotoxic~ At I00 uM only 5-Doxylstearate inhibited cell survival, whereas 12-Doxylstearate and 16-Doxylstearate had no effect. The use of nitroxide radical spin probes for numerous biological studies has contributed significantly toward our understanding of the structure of proteins, membranes, and other biopolymers (i). In recent years, the use of nitroxide radical spin probes for studies of living organisms has greatly increased to include such parameters as cell volume (2), cytoplasmic viscosity (3)~ oxygen consumption (4), membrane fluidity (5) and membrane potentials (6) Since information concerning the cytotoxicity of these compounds is sporadic, we thought it of interest to investigate their effects on the cell survival using Chinese hamster ovary (CHO) cells, a widely used model system for mammalian cells in culture, under identical conditions employed for the biophysical measurements. The water soluble spin probes Tempol (4-hydroxy-Tempo), Tempamine (4-amino-Tempo), CTPC (2,2,5.5-tetramethyl-3-pyrroline-l-oxyl-3-carboxylic acid) and CTPO (3-carbamoyl-2,2~5 5-tetramethyl-3-pyrroline-l-yloxy) were obtained from Molecular Probes, Junction City, O R 5-Doxyl, 12-Doxyl and 16Doxyl stearates and 4-maleimido-Tempo were purchased from Aldrich Chemical Co., Milwaukee, WI. CHO cells were maintained routinely as monolayers subcultured three times weekly~ For the experiments cells were detached with trypsin and grown in suspension for 24 hours in a spinner flask in ~MEM ( ~minimum essential medium), 10% fetal calf serum and antibiotics ( ~ M E M IOFC) at 37oc in a humid 6% CO 2 atmosphere. *To whom reprint request should be addressed 0024-3205/87 $3.00 + .00 Copyright (c) 1987 Pergamon Journals Ltd.
496
Cytotoxicity of Spin Probes
Vol. 40, No. 5, 1987
The procedure for determining the cytotoxicity of spin probes to CHO cells was as follows: The spin probe solution in phosphate buffered saline (PBS), pH 7.4 was added to CHO cells (5 x 106 cells/ml) for i0 minutes with gentle stirring at room temperature. (For lipid-solution spin probes, the spin probe solution was prepared by drying the ethanolic spin probe solution in a small beaker under N 2 gas and redispersing it in PBS buffer). Subsequently the ceils were washed with PBS buffer and the amounts of free spin probe in supernatant and spin probe associated with cells were determined by the electron spin resonance (ESR) method (7). Briefly, for free spin components the concentration was estimated by comparing the peak-to-peak signal amplitude with that of the same probe of known concentration. For partially immobilized components such as in the case of fatty-acid spin probe labeled cells, the spin concentration was determined by double-integration of the spectra. For cell survival determination after ESR measurements, cells ranging in number from 200 to 104 were placed in 60-mm plastic petri dishes in 5 ml of aMEM-10FC and incubated at 37°C in a humid 6% C02 atmosphere for 6 to 9 days. Colonies greater than 50 cells were scored and cell survival was calculated by conventional methods (7). Plating efficiencies for control cells treated in the identical manner ranged from 76 to 92%. The water-soluble spin probes Tempol and CTPO have been used as oxygen probes for studying cell respiration (4,8), CTPC for measuring cell volume (9), and Tempamlne for measuring intracellular microviscosity (i0). As summarized in Table 1 these spin probes at concentrations as high as 1 mM have no measurable effects on the cell viability. ESR studies show that the majority of the probes remain in supernatant (Table i) suggesting that the low intracellular concentrations are due to their poor retention in the cell rather than their fast reduction. Mehlhorn and Packer reported previously Table 1
Effects of Spin Probes on the Survival of CHO Cells +
Spin Probe Water soluble Tempol Tempamine CTPO CTPC 4 Maleimido-Tempo
Lipid-soluble 5-Doxylstearate
12-Doxylstearate 16-Doxylstearate
Concentration added (IJM)
I000 i000 I000 i000 i000 i00 75 50
I00 75 50 25 i00 I00
Con~centrations ~ ' ~ $ @M) in cells in supernatant
3 2 8 34 140 32 27 23
78 ND# 45 BID 40 43
920 893 990 930 850 49 26 I0
Cell Survival % of Control
92 87 114 95 0 4 60 103
0 21 89 92 90 85
+Values are averages of two experiments for both concentration and survival measurements. *The concentration was determined by the ESR method as described in the text. The concentration in cells plus in supernatant are not equal to the concentration added, suggesting probe reduction in cells, particularly for 4-Maleimido-Tempo. #ND-not determined.
Vol. 40, No. 5, 1987
Cytotoxicity of Spin Probes
497
that these water-soluble spin probes equilibrate rapidly across biological membranes (ii). On the other hand. 4-maleimido-Tempo which has been employed for measuring intracellular microviscosity was found to be toxic to the cells at concentrations higher than 75 DM. Similar results were obtained with 3-maleimido-Tempo (data not shown). Daveloose et al. (12) have shown that 4-maleimido-Tempo reacts with cytoplasmic glutathione when added to intact erytnrocytes Glutathione is present in very high concentrations (millimolars) in cells (13). Small alteration of its content may not have an effect on cell viability. In fact addition of N-ethyl maleimide (i00 ~M) to CHO cells at 37°C for one hour was found to have no effect on cell survival (data not shown). Maleimido-Tempo has been commonly used for labeling membrane proteins for ESR investigation of membrane structure. At 50 ~ M concentration~ the typical concentration used in membrane studies the probe was previously found to be non-toxic to erythrocytes (14). At i mM concentration~ however, the effect of maleimido-Tempo on the membrane structure of erythrocytes was noted (15), It is not clear whether the observed cytotoxicity of maleimido-Tempo at higher concentrations reported in this study is due to alteration of membrane structure of the cells. Incorporation of fatty-acid spin labels into the membrane of cultured mammalian cells has been used for monitoring membrane fluidity of the intact cells (i~4..5). These spin probes incorporated into CHO cells gave rise to ESR Spectra typical of probes in biological membranes (data not shown). At concentrations commonly used for spin labeling of cells (30-50 ~M) none of the fatty acid spin probes has any effect on the cell survival. However at 75 ~M concentration, 5-Doxylstearate was shown to be highly toxic to the cell. Since 5-Doxylstearate is the most commonly used fatty-acid spin probe for measurement of cellular membrane fluidity proper controls of its concentration during spin labeling of cells probably are of prime importance for obtaining meaningful data In conclusion~ the results presented in this communication suggest that with the exception of 4-maleimido-TEMPO and 5-Doxylstearate, many commonly used nitroxide radical spin probes at concentrations as high as i00 ~ M have no effect on the cell survival. Recently, nitroxide radical spin probes have been suggested as potential paramagnetic contrast agents for NMR imaging (16-19). The information provided in this in v i t r _ o o study suggest that these nitroxide compounds may be suitable for in vivo NMR imaging and should be tested further for in__vivo toxicity at tL1e appropriate concentrations. Ackn@wled~ements This work was supported in part by NIH Grant GM-35719, RR-01008 and CA-37930 We thank Donna Volk for her excellent technical assistance. References i. 2 3 4 5 6
L J. BERLINER, ed , Spin Labeling~ Theory and Application Vol. I and 2 Academic Press~ New York (1976, 1978), B SCHOBERT and D. MARCH, Biochim. Biophys. Acta 720 87-95 (1982). A~ M. MASTRO and A. D. KEITH~ The transformed cell, ed. I. L. Cameron and T B. Pool, p. 327~ Academic Press~ New York (1981). C -S. LAI, L. E. HOPWOOD, J. S. HYDE and S. LUKIEWICZ. P r o c Natl. Acad. Sci. USA 79 1166-1170 (1982). C -S. LAI~--E, G ANKEL and L. E. HOPWOOD Exp Cell Res. 150 77-83 (1984) D. S CAFISO W. L HUBBELL and A. QUINTANILHA, Hethods in Enzymol. 8_~8 682-696 (1982).
498
7 8. 9 10. 11. 12. 13 14. 15. 16. 17. 18 19.
Cytotoxicity
of Spin Probes
Vol. 40, No. 5, 1987
C.-S. LAI. L. E, HOPWOOD and H M. SWARTZ~ Biochim. Biophys. Acta 6 0 2 117-126 (1980) P. D. MORSE II and H. M. SWARTZ, Magn. Reson. Med. 2 114-127 (1985). HAMMERSTEDT, R. H,, AMANN R P. RUCINSKY T.~ MORSE, P. E II LEPOCK, J., SNIPES W. and KEITH A. D Biol. Reproduction 14 381-397 (1976) P. D. MORSE II Biochem. Biophys. Res. Commun. 77 1486-1491 (1977). R J MEHLHORN and L. PACKER, Ann. N Y. Acad. Sci. 414 180-189 (1983). D. DAVELOOSE G FABRE, F BERLEUR G TESTYLIER and F. LETERRIER Biochim. Biophys. Acta 763 41-49 (1983). M. E ANDERSON, F POWRIE, R. N. PURl and A. M E I S T E R Arch. Biochem. Biophys. 239 538-548 (1985). D. A. BUTTERFIELD. Biol. Mag. Reson, 4 1-78 (1982). B. T. FARMER, II, T M. HARMON and D. A BUTTERFIELD Biochim. Biophys Acta 821 420-430 (1985). C -S. LAI S J STAIR, H. MIZIORKO and J. S. HYDE J Mag. Res 57 447-452 (1984). L. K. GRIFFETH, G. M. ROSEN, E J. RAUCEMAN and B. P. D R A Y E R In~est. Radiol. 19 553-562 (1984). L. J. BERLINER and H. FUJII, Science 227 517-519 (1985). R. L. EHMAN, G. E WESBY, K L. MOON,--~. D WILLIAMS M. T. M c N ~ A R A W. R COUTET T. N. TOZER and R. C. BRASCH, Magn. Reson. Imaging 3 89-97 (1985).
Pergamon Journal~
CYTOTOXICITY OF COMMONLY USED NITROXIDE RADICAL SPIN PROBES Else G. Ankel,l, 2 Ching-San Lai,l, * Larry E. Hopwood 2 and Zorica Zivkovic 2 INational Biomedical ESR Center, Departments of Radiology and 2Radiation Oncology, Medical College of Wisconsin 8701 Watertown Plank Road, Milwaukee, Wisconsin 53226 (Received in final form November 3, 1986) Summary Since nitroxide radical spin probes are used frequently to test biophysical properties of cells, their use should be restricted to conditions that do not perturb normal cell growth and viability. Eight commonly used nitroxide radical spin probes have been tested for their effects on the survival of CHO cellst These include watersoluble spin probes Tempol, Tempamine, CTPO~ CTPC and 4-maleimidoTempo, and lipid soluble spin probes 5-Doxyl-~ 12-Doxyl-, and 16Doxylstearates. With the exception of 4-maleimido-Tempo, none of the water soluble spin labels inhibited cell survival at concentrations as high as I mM. At concentrations of 75 uM and higher, 4maleimido-Tempo inhibited cell survival in a dose dependent manner. At concentrations commonly used for spin labeling of cells (30-50 ~M) none of the lipid soluble spin probes tested was cytotoxic~ At I00 uM only 5-Doxylstearate inhibited cell survival, whereas 12-Doxylstearate and 16-Doxylstearate had no effect. The use of nitroxide radical spin probes for numerous biological studies has contributed significantly toward our understanding of the structure of proteins, membranes, and other biopolymers (i). In recent years, the use of nitroxide radical spin probes for studies of living organisms has greatly increased to include such parameters as cell volume (2), cytoplasmic viscosity (3)~ oxygen consumption (4), membrane fluidity (5) and membrane potentials (6) Since information concerning the cytotoxicity of these compounds is sporadic, we thought it of interest to investigate their effects on the cell survival using Chinese hamster ovary (CHO) cells, a widely used model system for mammalian cells in culture, under identical conditions employed for the biophysical measurements. The water soluble spin probes Tempol (4-hydroxy-Tempo), Tempamine (4-amino-Tempo), CTPC (2,2,5.5-tetramethyl-3-pyrroline-l-oxyl-3-carboxylic acid) and CTPO (3-carbamoyl-2,2~5 5-tetramethyl-3-pyrroline-l-yloxy) were obtained from Molecular Probes, Junction City, O R 5-Doxyl, 12-Doxyl and 16Doxyl stearates and 4-maleimido-Tempo were purchased from Aldrich Chemical Co., Milwaukee, WI. CHO cells were maintained routinely as monolayers subcultured three times weekly~ For the experiments cells were detached with trypsin and grown in suspension for 24 hours in a spinner flask in ~MEM ( ~minimum essential medium), 10% fetal calf serum and antibiotics ( ~ M E M IOFC) at 37oc in a humid 6% CO 2 atmosphere. *To whom reprint request should be addressed 0024-3205/87 $3.00 + .00 Copyright (c) 1987 Pergamon Journals Ltd.
496
Cytotoxicity of Spin Probes
Vol. 40, No. 5, 1987
The procedure for determining the cytotoxicity of spin probes to CHO cells was as follows: The spin probe solution in phosphate buffered saline (PBS), pH 7.4 was added to CHO cells (5 x 106 cells/ml) for i0 minutes with gentle stirring at room temperature. (For lipid-solution spin probes, the spin probe solution was prepared by drying the ethanolic spin probe solution in a small beaker under N 2 gas and redispersing it in PBS buffer). Subsequently the ceils were washed with PBS buffer and the amounts of free spin probe in supernatant and spin probe associated with cells were determined by the electron spin resonance (ESR) method (7). Briefly, for free spin components the concentration was estimated by comparing the peak-to-peak signal amplitude with that of the same probe of known concentration. For partially immobilized components such as in the case of fatty-acid spin probe labeled cells, the spin concentration was determined by double-integration of the spectra. For cell survival determination after ESR measurements, cells ranging in number from 200 to 104 were placed in 60-mm plastic petri dishes in 5 ml of aMEM-10FC and incubated at 37°C in a humid 6% C02 atmosphere for 6 to 9 days. Colonies greater than 50 cells were scored and cell survival was calculated by conventional methods (7). Plating efficiencies for control cells treated in the identical manner ranged from 76 to 92%. The water-soluble spin probes Tempol and CTPO have been used as oxygen probes for studying cell respiration (4,8), CTPC for measuring cell volume (9), and Tempamlne for measuring intracellular microviscosity (i0). As summarized in Table 1 these spin probes at concentrations as high as 1 mM have no measurable effects on the cell viability. ESR studies show that the majority of the probes remain in supernatant (Table i) suggesting that the low intracellular concentrations are due to their poor retention in the cell rather than their fast reduction. Mehlhorn and Packer reported previously Table 1
Effects of Spin Probes on the Survival of CHO Cells +
Spin Probe Water soluble Tempol Tempamine CTPO CTPC 4 Maleimido-Tempo
Lipid-soluble 5-Doxylstearate
12-Doxylstearate 16-Doxylstearate
Concentration added (IJM)
I000 i000 I000 i000 i000 i00 75 50
I00 75 50 25 i00 I00
Con~centrations ~ ' ~ $ @M) in cells in supernatant
3 2 8 34 140 32 27 23
78 ND# 45 BID 40 43
920 893 990 930 850 49 26 I0
Cell Survival % of Control
92 87 114 95 0 4 60 103
0 21 89 92 90 85
+Values are averages of two experiments for both concentration and survival measurements. *The concentration was determined by the ESR method as described in the text. The concentration in cells plus in supernatant are not equal to the concentration added, suggesting probe reduction in cells, particularly for 4-Maleimido-Tempo. #ND-not determined.
Vol. 40, No. 5, 1987
Cytotoxicity of Spin Probes
497
that these water-soluble spin probes equilibrate rapidly across biological membranes (ii). On the other hand. 4-maleimido-Tempo which has been employed for measuring intracellular microviscosity was found to be toxic to the cells at concentrations higher than 75 DM. Similar results were obtained with 3-maleimido-Tempo (data not shown). Daveloose et al. (12) have shown that 4-maleimido-Tempo reacts with cytoplasmic glutathione when added to intact erytnrocytes Glutathione is present in very high concentrations (millimolars) in cells (13). Small alteration of its content may not have an effect on cell viability. In fact addition of N-ethyl maleimide (i00 ~M) to CHO cells at 37°C for one hour was found to have no effect on cell survival (data not shown). Maleimido-Tempo has been commonly used for labeling membrane proteins for ESR investigation of membrane structure. At 50 ~ M concentration~ the typical concentration used in membrane studies the probe was previously found to be non-toxic to erythrocytes (14). At i mM concentration~ however, the effect of maleimido-Tempo on the membrane structure of erythrocytes was noted (15), It is not clear whether the observed cytotoxicity of maleimido-Tempo at higher concentrations reported in this study is due to alteration of membrane structure of the cells. Incorporation of fatty-acid spin labels into the membrane of cultured mammalian cells has been used for monitoring membrane fluidity of the intact cells (i~4..5). These spin probes incorporated into CHO cells gave rise to ESR Spectra typical of probes in biological membranes (data not shown). At concentrations commonly used for spin labeling of cells (30-50 ~M) none of the fatty acid spin probes has any effect on the cell survival. However at 75 ~M concentration, 5-Doxylstearate was shown to be highly toxic to the cell. Since 5-Doxylstearate is the most commonly used fatty-acid spin probe for measurement of cellular membrane fluidity proper controls of its concentration during spin labeling of cells probably are of prime importance for obtaining meaningful data In conclusion~ the results presented in this communication suggest that with the exception of 4-maleimido-TEMPO and 5-Doxylstearate, many commonly used nitroxide radical spin probes at concentrations as high as i00 ~ M have no effect on the cell survival. Recently, nitroxide radical spin probes have been suggested as potential paramagnetic contrast agents for NMR imaging (16-19). The information provided in this in v i t r _ o o study suggest that these nitroxide compounds may be suitable for in vivo NMR imaging and should be tested further for in__vivo toxicity at tL1e appropriate concentrations. Ackn@wled~ements This work was supported in part by NIH Grant GM-35719, RR-01008 and CA-37930 We thank Donna Volk for her excellent technical assistance. References i. 2 3 4 5 6
L J. BERLINER, ed , Spin Labeling~ Theory and Application Vol. I and 2 Academic Press~ New York (1976, 1978), B SCHOBERT and D. MARCH, Biochim. Biophys. Acta 720 87-95 (1982). A~ M. MASTRO and A. D. KEITH~ The transformed cell, ed. I. L. Cameron and T B. Pool, p. 327~ Academic Press~ New York (1981). C -S. LAI, L. E. HOPWOOD, J. S. HYDE and S. LUKIEWICZ. P r o c Natl. Acad. Sci. USA 79 1166-1170 (1982). C -S. LAI~--E, G ANKEL and L. E. HOPWOOD Exp Cell Res. 150 77-83 (1984) D. S CAFISO W. L HUBBELL and A. QUINTANILHA, Hethods in Enzymol. 8_~8 682-696 (1982).
498
7 8. 9 10. 11. 12. 13 14. 15. 16. 17. 18 19.
Cytotoxicity
of Spin Probes
Vol. 40, No. 5, 1987
C.-S. LAI. L. E, HOPWOOD and H M. SWARTZ~ Biochim. Biophys. Acta 6 0 2 117-126 (1980) P. D. MORSE II and H. M. SWARTZ, Magn. Reson. Med. 2 114-127 (1985). HAMMERSTEDT, R. H,, AMANN R P. RUCINSKY T.~ MORSE, P. E II LEPOCK, J., SNIPES W. and KEITH A. D Biol. Reproduction 14 381-397 (1976) P. D. MORSE II Biochem. Biophys. Res. Commun. 77 1486-1491 (1977). R J MEHLHORN and L. PACKER, Ann. N Y. Acad. Sci. 414 180-189 (1983). D. DAVELOOSE G FABRE, F BERLEUR G TESTYLIER and F. LETERRIER Biochim. Biophys. Acta 763 41-49 (1983). M. E ANDERSON, F POWRIE, R. N. PURl and A. M E I S T E R Arch. Biochem. Biophys. 239 538-548 (1985). D. A. BUTTERFIELD. Biol. Mag. Reson, 4 1-78 (1982). B. T. FARMER, II, T M. HARMON and D. A BUTTERFIELD Biochim. Biophys Acta 821 420-430 (1985). C -S. LAI S J STAIR, H. MIZIORKO and J. S. HYDE J Mag. Res 57 447-452 (1984). L. K. GRIFFETH, G. M. ROSEN, E J. RAUCEMAN and B. P. D R A Y E R In~est. Radiol. 19 553-562 (1984). L. J. BERLINER and H. FUJII, Science 227 517-519 (1985). R. L. EHMAN, G. E WESBY, K L. MOON,--~. D WILLIAMS M. T. M c N ~ A R A W. R COUTET T. N. TOZER and R. C. BRASCH, Magn. Reson. Imaging 3 89-97 (1985).