Departments of Safety Assessment and Safety Assessment Statistics, GlaxoSmithKline, King of Prussia, PA 19406
To the Editor:
We welcome a chance to respond to the views expressed by Kostrubsky et al. regarding our article. We previously acknowledged that the relevance of our in vitro findings to the idiosyncratic hepatotoxicity of troglitazone (TRO) was uncertain. However, we wish to clarify that the uncertainties as regards findings in HepG2 cells to toxicity observed in patients does not arise from deficiencies in our experiments, but rather reflect the difficulty in extrapolating in vitro data to clinical experiences. We believe our experiments provide very convincing evidence that mitochondrial toxicity is an initiating event in TRO-induced toxicity in HepG2 cells. This conclusion was reached after reviewing data collected by ultrastructural, biochemical and confocal microscopic analysis from which there was remarkable agreement across these methodologies. This integrated approach to structural and functional assessment is not commonly seen in investigative toxicity studies, and we believe the concordance in observations strengthens our conclusions of a TRO-induced mitochondrial event. We provide the following answers to the concerns expressed by Kostrubsky et al.:
(1) The effects of TRO on both mitochondrial function and cell death were concentration- and time-dependent in HepG2 cells. In our experiments conducted in 12-well plates, 25 µM TRO did not induce cell death after 5-h incubations. However, further experiments are needed to determine if extended incubation at this concentration of TRO induces cell death. Our data suggests that loss of mitochondrial potential is an initiating event in TRO toxicity but is not in itself lethal to cells. Rather, induction of mitochondrial permeability transition is the key cellular event leading to cell death. Kostrubsky et al. seem to equate loss of mitochondrial potential with mitochondrial permeability transition. It is important to emphasize that mitochondrial depolarization and mitochondrial permeability transition are distinct events; depolarization is not always a consequence of mitochondrial permeability transition (Bernardi et al., 1999). In agreement with this concept, we found that the mitochondrial permeability transition inhibitor, cyclosporin A, prevented TRO-induced cell death but not the loss of mitochondrial potential. Mitochondrial membrane depolarization is known to promote opening of the permeability transition pore (Bernardi et al., 1999
), but the mechanism linking loss of mitochondrial potential with the triggering of the mitochondrial permeability transition is not known. Minamikawa et al. (1999)
found that cells treated with the mitochondrial uncoupler, carbonyl cyanide m-chlorophenylhydrazone (CCCP), could survive extended periods (6 h) with dissipated mitochondrial potentials without evidence of cell damage. In our studies, CCCP did not induce cell death in HepG2 cells over the 5-h incubation period despite decreasing cellular ATP and mitochondrial potentials. Higher concentration of TRO (
50 µM) may be more effective in promoting mitochondrial permeability transition and triggering cell death despite apparent similarities with lower concentrations (25 µM) in decreasing mitochondrial potential.
(2) Kostrubsky et al. suggest cyclosporin A inhibition of calcineurin-mediated cell death as an alternative mechanism. Calcineurin activity is calcium and calmodulin-dependent (Shatrov et al., 1997). We conducted several experiments in which HepG2 cells were preincubated with the calcium chelator BAPTA-AM (25 µM) for 30 min prior to addition of TRO. BAPTA-AM is membrane permeant and once taken up by cells is hydrolyzed by cellular esterases to produce the active calcium chelator, BAPTA. BAPTA-AM provided no protection against TRO-induced cell death. Since calcium chelation would be expected to inhibit calcineurin activity, this data argues against the involvement of calcineurin in TRO-induced cell death.
(3) MitoTracker Red® has been shown to be a mitochondria-specific dye (Gilmore and Wilson, 1999; Isenberg and Klaunig, 2000
). The size and morphology of structures stained with MitoTracker Red after exposure to 25 µM TRO for 1 h are consistent with those of mitochondria from treated cells observed by transmission electron microscopy (See Figure 7 in Tirmenstein et al., 2002
). Kostrubsky et al. state that calcein-AM should be excluded from normal mitochondria and should "redistribute into the mitochondria if there was severe mitochondria membrane damage". However, we find along with several other investigators (Bernardi et al., 1999
; Minamikawa et al. 1999
) that calcein-AM freely distributes into normal mitochondria. Petronilli et al. (1998)
developed a mitochondrial permeability assay based on the ability of intact mitochondria to exhibit calcein fluorescence while cytoplasmic calcein fluorescence is quenched by addition of cobalt chloride. Clearly this method relies on the ability of calcein-AM to distribute into normal mitochondria.
(4) We hypothesized that the parent compound, rather than a metabolite, was responsible for mitochondrial dysfunction. As such, metabolism would be expected to be protective, rather than required for toxicity. Hence, we purposely selected HepG2 cells to evaluate TRO-toxicity because they have low P450 activity. As such HepG2 cells represent a good in vitro model to evaluate the potential importance of P450 activation in TRO-induced toxicity.
(5) Many of the same observations we report in HepG2 cells have also been observed in primary rat and human hepatocytes, as well as in human peripheral blood mononuclear cells and reported by Haskins et al. (2001). Haskins et al. (2001)
concluded that TRO-induced disruption of mitochondrial activity was an initiating toxic event in human and rat hepatocytes. TRO at concentrations of 200 µM decreased ATP concentrations in rat hepatocytes (Haskins et al., 2001
). The IC50 values for TRO-induced decreases in mitochondrial potentials in rat hepatocytes and human lymphocytes were 270 and 83 µM TRO, respectively. Effects on mitochondrial potentials at concentrations of 100 µM TRO were also observed in human hepatocytes isolated from a diabetic donor. Cell culture studies conducted by Haskins et al. (2001)
contained serum while our studies used serum free media. It has previously been demonstrated that bovine serum albumin decreases the toxicity of TRO in cultured cells, presumably due to protein binding (Toyoda et al., 2001
). These results suggest that many of the mitochondrial effects we observed in HepG2 are also seen in normal (noncancerous) cells and as such may be relevant to human hepatotoxicities.
Consequently, data from other sources also supports the conclusion that TRO, as the parent molecule, disrupts mitochondrial function in hepatocytes, leading to mitochondrial permeability transition and ultimately hepatocellular death.
REFERENCES
Bernardi, P., Scorrano, L., Colonna, R., Petronilli, V., and Di Lisa, F. (1999). Mitochondria and cell death. Mechanistic aspects and methodological issues. Eur. J. Biochem. 264, 687701.
Gilmore, K., and Wilson, M. (1999). The use of chloromethyl-X-rosamine (MitoTracker Red) to measure loss of mitochondrial membrane potential in apoptotic cells is incompatible with cell fixation. Cytometry 36, 355358.[ISI][Medline]
Haskins, J. R., Rowse, P., Rahbari, R., and de la Iglesia, F. A. (2001). Thiazolidinedione toxicity to isolated hepatocytes revealed by coherent multiprobe fluorescence microscopy and correlated with multiparameter flow cytometry of peripheral leukocytes. Arch. Toxicol. 75, 425438.[ISI][Medline]
Isenberg, J. S., and Klaunig, J. E. (2000). Role of the mitochondrial permeability transition (MPT) in rotenone-induced apoptosis in liver cells. Toxicol. Sci. 53, 340351.
Minamikawa, T., Williams, D. A., Bowser, D. N. and Nagley, P. (1999). Mitochondrial permeability transition and swelling can occur reversibly without inducing cell death in intact human cells. Exp. Cell Res. 246, 2637.[ISI][Medline]
Petronilli, V., Miotto, G., Canton, M., Colonna, R., Bernardi, P., and Di Lisa, F. (1998). Imaging the mitochondrial permeability transition pore in intact cells. Biofactors 8, 263272.[ISI][Medline]
Shatrov, V. A., Lehmann, V., Chouaib, S. (1997). Sphingosine-1-phosphate mobilizes intracellular calcium and activates transcription factor NF-B in U937 cells. Biochem. Biophys. Res. Commun. 234, 121124.[ISI][Medline]
Tirmenstein, M. A., Hu, C. X., Gales, T. L., Maleeff, B. E., Narayanan, P. K., Kurali, E., Hart, T. K., Thomas, H. C., and Schwartz, L. W. (2002). Effects of Troglitazone on HepG2 Viability and Mitochondrial Function. Toxicol. Sci. 69, 131138.
Toyoda, Y., Tsuchida, A., Iwami, E. and Miwa, I. (2001). Toxic effect of troglitazone on cultured rat hepatocytes. Life Sci. 68, 18671876.[ISI][Medline]