Cisplatin accumulates in mitochondria, which are a major target for this drug in cancer cells. Thus alterations in mitochondrial function have been implicated in cancer cell resistance to chemotherapeutic agents. Moreover, cisplatin toxic side effects seem to be associated with mitochondrial injury in vivo and in vitro. In order to clarify the potential effect of cisplatin in mtDNA (mitochondrial DNA) maintenance and expression, we have analysed rat liver mtDNA and mtRNA (mitochondrial RNA) synthesis as well as their stability under the influence of in vivo treatment or in vitro exposure to cisplatin. We show that cisplatin causes a direct and significant impairment of mtDNA and mtRNA synthesis and decreases steady-state levels of mtRNAs in isolated mitochondria. Furthermore, in vivo treatment of the animals with cisplatin exerts a protective effect from the impairment of mtRNA metabolism caused by in vitro exposure to the drug, by means of increased mitochondrial GSH levels after in vivo cisplatin treatment.
- anticancer drug toxicity
- mitochondrial DNA (mtDNA) metabolism
- mitochondrial RNA (mtRNA)
Cisplatin is one of the most commonly used chemotherapeutic agents, with demonstrated activity against a diverse spectrum of malignancies including testicular cancer, ovarian carcinoma and head/neck tumours [1,2]. However, the therapeutic effectiveness of the drug is limited by the severity of its side effects and the potential progression of tumour cells to a cisplatin-resistant state .
In blood, where the chloride concentration is relatively high, cisplatin exists mainly in the dichloro neutral form. Inside the cell, the relatively low chloride concentration favours the replacement of one or both chloride moieties by water, resulting in a positively charged molecule that reacts with nucleophilic sites to cross-link DNA, RNA and protein . Cisplatin cytotoxicity correlates with DNA adduct formation . Some of the Pt (platinum)–DNA adducts can be removed by the NER (nucleotide excision repair) mechanism, but others cannot . Pt lesions on DNA impair template function, blocking DNA replication, induce a G2 cell cycle arrest, inhibit RNA transcription and finally promote cell death through apoptosis . The pathway from cisplatin-induced DNA damage to irreversible apoptotic commitment has so far not been fully defined .
Enhanced ability to repair DNA lesions through up-regulation of components of the NER pathway has been proposed to contribute to cisplatin resistance in a number of resistant tumours . Selection for cisplatin resistance in some models has been associated with development of mismatch repair deficiency . Failure to appropriately recognize cisplatin adducts in cells with defects in mismatch repair is hypothesized to lead to replication bypass of lesions, resulting in frequent incorporation of incorrect nucleotides and failure to trigger cell death.
Intracellular cisplatin inactivation by glutathione (GSH) has also been proposed as a mechanism of cisplatin resistance . GSH can either quench DNA–Pt mono-adducts before their conversion into cytotoxic DNA cross-links  or form a complex (or complexes) with cisplatin, thereby reducing the amount of intracellular drug available for interaction with DNA .
Although there is evidence that the mechanism of tumour killing involves the formation of DNA cross-links, the sensitivity of cells to Pt drugs does not always correlate with the formation of DNA adducts . Moreover, the acute side effects of cisplatin are not completely understood, indicating that other mechanisms may be involved in the toxicity of the drug.
Mitochondria are subcellular organelles involved in critical metabolic pathways including cellular energy metabolism. Besides the role of mitochondria in cellular ATP production, these organelles are known to be effectors of major apoptotic pathways . A direct inhibition of the respiratory chain rapidly depletes cellular ATP and promotes non-apoptotic (necrotic) cell death . Because of their critical role in cell survival, mitochondria are targets for cellular toxins and chemotherapeutic agents .
It has been shown that cisplatin accumulates in mitochondria . The electrochemical gradient resulting in a negative charge within mitochondria may play a role in the accumulation of positively charged cisplatin in this organelle . Indeed, mitochondria are thought to be a major target for cisplatin in cancer cells  and alterations in mitochondrial function have been implicated in cancer cell resistance to chemotherapeutic agents . Moreover, cisplatin toxic side effects seem to be associated with mitochondrial injury in vivo and in vitro [22,23]. Mitochondria contain their own DNA that is transcribed and translated to synthesize 13 proteins of the mitochondrial electron transport chain . It has been proposed that cisplatin may also induce mtDNA (mitochondrial DNA) damage [25,26]. However, it remains to be demonstrated if the damage of mtDNA is a primary or a secondary consequence of the cisplatin toxicity.
In the present study, we have analysed the influence of the in vivo treatment and the in vitro addition of cisplatin on mtDNA and mtRNA (mitochondrial RNA) synthesis and stability. We show that cisplatin caused a direct and significant impairment of the synthesis of mtDNA and the synthesis and steady state of mtRNAs. Furthermore, we provide evidence indicating that GSH levels dramatically influence mtDNA metabolism. Thus the in vivo treatment of animals with cisplatin increases GSH levels, both in the cytoplasm and in mitochondrial compartments, with a parallel protection of mtDNA metabolism to cisplatin exposure. On the other hand, experimentally induced depletion of mitochondrial GSH impairs mtDNA expression.
These studies provide the basis for the understanding of the effects of cisplatin on mitochondrial activity and the mechanisms of cellular toxicity and resistance caused by this drug.
Cisplatin, glutathione (GSH), MPA (meta-phosphoric acid), NEM (N-ethylmaleimide), sodium borohydride (NaBH4), octan-1-ol, N-ethylmorpholine, mBBr (monobromobimane), acetonitrile (HPLC-grade), ammonium nitrate and ammonium formate were purchased from Sigma. Cisplatin was dissolved in 0.9% NaCl (1 mg/ml) within 1 h before use. Concentrated solutions of MPA (10%) and NEM (0.1 M) were prepared fresh immediately prior to use. The column and the guard column for HPLC were obtained from Waters.
Wistar rats weighing 200–250 g were fed laboratory chow and water ad libitum and used for the experiments. In cisplatin in vivo studies, groups of three rats were injected intraperitoneally with a simple dose of cisplatin: 10 or 20 mg/kg of body weight. The animals were killed by decapitation 5, 24 or 48 h after cisplatin administration. Controls were injected with saline alone, and one control animal was killed with each one of the experimental groups. All animal experimentation was performed in accordance with the requirements of the European Union and Spanish legislation.
Isolation of rat liver mitochondria
The liver was rapidly removed, chilled in homogenization medium and mitochondrial fractions were purified as described elsewhere . The final mitochondrial pellet was resuspended in the corresponding buffer for DNA or RNA synthesis or for GSH determination.
Sample preparation for GSH determination
The sample preparation for GSH analysis was performed using the method of Rebrink et al.  but with slight modifications. Total GSH determination was carried out on crude homogenates just after homogenization of the liver as mentioned above. Aliquots (200 μl) of the crude homogenates were mixed with 200 μl of ice-cold 10% (w/v) MPA for deproteinization. After 30 min of incubation on ice, samples were centrifuged at 14000 g for 20 min at 4 °C. Supernatants were used immediately for GSH determination or stored at −80 °C for up to 4 months.
For measurement of mitochondrial GSH content, 200 μl aliquots of fresh mitochondrial preparations containing 2 mg of protein/ml were washed once with a 10-fold excess of O2 consumption incubation medium (see below) and pelleted by centrifugation. The mitochondrial pellets were suspended in 200 μl of 5% MPA, incubated for 30 min on ice and centrifuged at 18000 g for 20 min at 4 °C. Supernatants were used immediately for GSH determination or stored at −80 °C until analysis.
Protein concentration in homogenates and mitochondrial fractions was determined by the Bradford method .
Derivatization of GSH and quantification by HPLC
The derivatization procedure was performed essentially as described by Pastore et al. . Briefly, 10 μl of 4 M NaBH4 (dissolved in 333 μl/ml DMSO and 66 mM NaOH), 5 μl of a solution containing 2 mM EDTA and 2 mM DTT (dithiothreitol), 5 μl of octan-1-ol, and 5 μl of 2 M HCl were mixed and spun down in a derivatization vial. A 10 μl portion of the homogenate or mitochondrial supernatant (in 5% MPA) was added, mixed and spun down. After 1 min of incubation, 25 μl of N-ethylmorpholine buffer (2 M N-ethylmorpholine, pH 8.0) and 10 μl of 25 mM bromobimane [in 1:1 (v/v) acetonitrile/water] were added to the derivatization vial. After 1 min of incubation, the reaction was stopped with 4.2 μl of acetic acid. All operations were carried out at room temperature (24 °C).
GSH was analysed with a Waters HPLC system, equipped with two 515 HPLC pumps. A 10 μl portion of derivatized sample was injected using a Waters 717 autosampler into a 150 mm×3.9 mm Nova-Pack C-18 column (particle size: 4 μm), equilibrated with 5% buffer A (acetonitrile)/95% buffer B (30 mM ammonium nitrate and 40 mM ammonium formate buffer, pH 3.6). Samples were eluted from the column with a 20 min gradient of buffer A (0–7.50 min, 5% buffer A/95% buffer B; 7.50–10.50 min, 100% buffer A; 10.50–16.00 min, 100% buffer A; 16.00–18.00 min, 5% buffer A/95% buffer B; 18.00–20.00 min, 5% buffer A/95% buffer B) at a flow rate of 1.5 ml/min. The column was run at 33 °C. Under these conditions, sample analysis was completed in 20 min and GSH was the only thiol eluting peak with a retention time of 7.90 min. Calibration standards for GSH were prepared in duplicate by diluting a 2 mM stock solution of GSH in 5% MPA. For the detection of derivatized GSH a Waters 474 scanning fluorescence detector operating at a λex of 390 nm and a λem of 478 nm was used. The Empower Login program for Windows 2000 (Waters) was used for controlling the HPLC system and analysing GSH results. Each sample was injected twice and the average of peak areas was used for calculations of the GSH concentrations.
Synthesis of DNA and RNA in isolated rat liver mitochondria
mtDNA and mtRNA synthesis assays were performed as described elsewhere . Briefly, reactions were performed in 0.5 ml at a mitochondrial concentration of 2 mg/ml in both cases. For DNA synthesis, 50 mM dATP, dGTP and dCTP and 20 μC [α-32P]dTTP (2000 μC/mmol) were added to the incubation test tube and incubation was carried out at 37 °C for 3 h on a rotating wheel .
For RNA synthesis, 20 μC [α-32P]UTP (800 μC/mmol) was added to the medium. The incubation was maintained at 37 °C for 60 min on a rotary shaker (12 rev./min) .
After DNA or RNA synthesis, mitochondria were centrifuged at 14000 g for 1 min and washed twice with 0.5 ml of 10% (v/v) glycerol, 0.15 mM MgCl2 and 10 mM Tris/HCl (pH 6.8).
Mitochondrial nucleic acid extraction
Mitochondrial lysis and nucleic acid extraction were carried out as described previously . Briefly, the mitochondrial pellet was resuspended in 500 μl of Pronase buffer (10 mM Tris/HCl, pH 7.4, 150 mM NaCl and 1 mM sodium EDTA), lysed with 2% (w/v) SDS in the presence of 100 μg of Pronase (autodigested to remove nucleases) and incubated for 15 min at 37 °C. Total mitochondrial nucleic acids were extracted twice with an equal volume of a phenol/chloroform/3-methylbutan-1-ol mixture (25:25:1, by volume) at room temperature. After ethanol precipitation and centrifugation, the pelleted nucleic acids were dissolved in 10 mM Tris/HCl (pH 7.4) and 1 mM EDTA.
mtDNA was analysed directly or after digestion with RNase-free DNase and the restriction enzyme BamHI by vertical 1% agarose slab gel electrophoresis . mtRNAs were analysed by methylmercury hydroxide–agarose gels as previously described .
After the run, the gels were first stained with ethidium bromide, photographed under UV light, dried and exposed for autoradiography either at −70 °C with a DuPont screen intensifier or at room temperature as needed.
The amount of total mtDNA or RNA was quantified by densitometry of the gel bands after electrophoresis, staining with ethidium bromide and photography of the gel with an LKB Ultroscan XL laser densitometer and Gel Scan XL software.
The amount of DNA or RNA synthesized by isolated mitochondria was quantified by densitometry of the autoradiograms with an LKB Ultroscan XL laser densitometer and Gel Scan XL software. Images were acquired with an Epson Perfection 1250 scanner and Adobe Photoshop CS 8.0.1 software.
Mitochondrial O2 consumption was measured polarographically at 37 °C in a closed 1.0 ml reaction chamber fitted with a Clark O2 electrode (Yellow Springs Instrument, Yellow Springs, OH, U.S.A.). Mitochondrial protein (1 mg) was added to 0.9 ml of incubation medium to initiate the study of mitochondrial respiration. The incubation medium was composed of 25 mM sucrose, 75 mM sorbitol, 100 mM KCl, 0.05 mM EDTA, 5 mM MgCl2, 10 mM H3PO4 and 10 mM Tris/HCl (pH 7.4), also containing 1 mg/ml fatty-acid-free BSA. Respiration started with the addition of Complex I substrates 10 mM glutamate and 2.5 mM malate. After 1–3 min, 0.2 mM ADP was added to initiate state 3 respiration. After ADP was fully converted into ATP, state 4 respiration was measured. KCN (0.2 μM) was added to inhibit Complex I activity. Mitochondrial respiration rates (state 3 and state 4) were expressed as nanoatom O2/min per mg of mitochondrial protein or nanoatom O2/min·per mtDNA [a.u. (arbitrary units)]. The RCR (respiratory control ratio) is indicative of mitochondrial coupling, and was expressed as the state 3/state 4 ratio.
The P/O ratio [moles of synthesized ATP per atom of oxygen transformed to water during OXPHOS (oxidative phosphorylation)] was also calculated as a measurement of mitochondrial coupling.
PCR amplification of mtDNA
Mitochondria and mtDNA were isolated as described above and incubated with and without cisplatin as described in the Figures. The oligonucleotide primers used in this study, as indicated below, were prepared by Gibco BRL. Various segments of mtDNA were amplified using the following oligonucleotides (H is the mtDNA heavy strand, L is the mtDNA light strand, OH is the heavy strand replication origin, and IL is the light strand transcription origin):
All amplification reactions were carried out in 0.2 ml PCR tubes in a final volume of 50 μl. Each reaction mixture consisted on 0.5 μg of mtDNA, 2.5 μM of each primer, 200 μM dNTP, 2 mM MgCl2, 2.5 units of Taq DNA polymerase (Epicenter Technologies, Madison, WI, U.S.A.), and the supplied reaction buffer. For the amplification of mtDNA, samples underwent an initial denaturation at 94 °C for 10 min, followed by 40 cycles of template denaturation at 94 °C for 1.30 min, primer–template annealing at 55 °C for 1.30 min and primer extension at 72 °C for 1.30 min. At the end of the 30 cycles, extension was allowed to continue for an additional 7 min. PCR amplification was performed on a programmable thermal cycler (Biometra Trio Thermocycler). The amplified products were separated by horizontal 1% agarose gel electrophoresis with 1% ethidium bromide. After the run, the gels were photographed under UV light. Amounts of amplified DNA were quantified by densitometry with an LKB Ultroscan XL laser densitometer and Gel Scan XL software.
Results are shown as the means plus the S.D. The statistical analysis was carried out by a Mann–Whitney test. P<0.05 was considered statistically significant.
Cisplatin impairs mitochondrial respiration
Mitochondria isolated from rat liver were incubated at 37 °C in the absence or in the presence of different cisplatin concentrations for 1, 2 or 3 h and subjected to oxygen consumption measurement. State 3 and state 4 respiration were measured and the RCR and the P/O ratio were calculated (Figure 1A). Cisplatin caused a dose- and time-dependent decrease in the RCR (Figure 1A, upper panel) and the P/O ratio was less sensitive to cisplatin, but it also showed a dose- and time-dependent increase (Figure 1A, lower panel) indicating an impairment of the OXPHOS system function.
Cisplatin inhibits mtDNA synthesis
Since cisplatin can form adducts with DNA, we investigated if mtDNA was functionally affected by the presence of the drug. To evaluate this possibility, the direct effect of cisplatin on mtDNA replication in isolated organelles was evaluated. Rat liver mitochondria were incubated for 3 h at 37 °C with [α-32P]TTP in the presence of cisplatin concentrations ranging from 0 to 0.7 mM. After incubation, mitochondrial nucleic acids were isolated and examined by electrophoresis. As shown in Figure 1(B), cisplatin significantly inhibited mtDNA replication in organello. The inhibition was dose-dependent with a range of cisplatin concentrations between 0 and 0.7 mM. At concentrations higher than 0.2 mM, cisplatin inhibited mtDNA synthesis within 1 h. The inhibition affected all the different mtDNA conformations . Cisplatin did not have any effect on radioisotope capture by the isolated mitochondria (Figure 1B, TTP uptake).
Inhibition of mtDNA replication could be either due to impairment of the OXPHOS capacity induced by cisplatin or due to a direct modification of the mtDNA template, as has been postulated for nuclear DNA. In order to evaluate which one of the two mechanisms is determining the observed effect, we used a PCR-based assay  in which many DNA lesions would block the Taq DNA polymerase progression and result thereby in a specific decrease in amplification of the damaged DNA template. Mitochondria isolated from rat liver were incubated for 1 h at 37 °C in the presence of different concentrations of cisplatin (0–1.4 mM). After the incubation, nucleic acids were extracted and used as the template for PCR amplification. Also, 1.4 mM cisplatin was added to one of the DNA control samples just before the PCR amplification. PCR products were then analysed on agarose gels. As shown in Figure 1(C) (left panel), the amount of amplified mtDNA decreased in a dose-dependent manner after incubation with cisplatin, whereas it was not affected by the presence of cisplatin exclusively during the PCR amplification reaction (C1). Various mtDNA regions were affected by cisplatin to different extents, a fact that could be explained by the length of the amplification fragment (Figure 1C, right panel). The region amplified by H-Cys/L-Cys (1), the largest one, showed higher synthesis inhibition, whereas H-OH/L-OH (4), the shorter one, showed less amplification inhibition.
These results indicated that cisplatin produced a direct and non-specific structural alteration of mtDNA, which caused an inhibition of mtDNA amplification by PCR and, very likely, contributed to the inhibition of mtDNA synthesis in organello.
Cisplatin inhibits mtDNA transcription
The next question to be answered was whether cisplatin also interfered with mtDNA transcription. Again, mitochondria isolated from rat liver were incubated for 1 h at 37 °C with [α-32P]UTP in the presence of cisplatin concentrations ranging from 0 to 1.4 mM. After incubation, mitochondrial nucleic acids were isolated and examined by electrophoresis.
Analogous to the effect on replication, cisplatin caused a dramatic inhibition of mtDNA transcription in organello (Figure 2A). The inhibition was already detectable at concentrations as low as 0.17 mM and was complete at 0.7 mM. The densitometric analysis of the RNA electrophoretic patterns normalized by the mtDNA amount loaded on to each lane showed that the inhibition affected all mtRNA species: rRNA and mRNAs. Cisplatin did not have any effect on radioisotope capture by the isolated mitochondria (UTP uptake; Figure 2A, left panel). In addition, the effect of 0.17 mM cisplatin was time-dependent, showing increased inhibition from 1 to 3 h (Figure 2A, right panel).
Cisplatin decreases mtRNA stability
Incubation of the organelles in the presence of cisplatin did not affect the mtDNA steady state (Figure 2B, upper band), but the levels of the in vivo mtRNAs were decreased in a dose- and time-dependent manner (Figure 2B, lower bands) as determined from the stained gels (see the Experimental section). We observed a general decrease in the amount of the mtRNAs that was particularly evident for the 12S and 16S rRNAs. Thus we determined whether a decrease in mtRNA stability (Figure 2B) caused by cisplatin could explain the reduction in the amount of newly synthesized RNA in isolated mitochondria (Figure 2A). Comparison of the effect of cisplatin on mtRNA stability with the effect on RNA synthesis revealed that the decreased RNA levels could not be explained as resulting only from the reduction in the stability (results not shown).
The results described above demonstrated that cisplatin causes a direct damage of mitochondrial biogenesis by inhibiting mtDNA synthesis and transcription and by destabilizing the mtRNA, an effect that is already evident after a short time of exposure.
Effect of in vivo cisplatin treatment on mtRNA metabolism
In order to evaluate the in vivo relevance of the effects observed in vitro, we treated rats with cisplatin. The drug was injected intraperitoneally once at doses of either 10 or 20 mg per kg of body weight. For each dose, the animals were killed at 5, 24 or 48 h after the administration (see the Materials and methods section).
Liver mitochondria were isolated from animals treated in vivo with cisplatin and incubated in organello for 1 h with [α-32P]UTP. After incubation, mitochondrial nucleic acids were isolated and examined by electrophoresis. In vivo treatment with cisplatin caused various effects on mtRNA synthesis by isolated mitochondria depending on the dose and the time after the treatment (Figures 3A and 3B). Thus 10 mg/kg cisplatin increased the in organello RNA synthesis rate, whereas in vivo treatment with 20 mg/kg cisplatin slightly decreased it. These effects were already evident after 5 h of cisplatin administration, rising to a maximum after 24 h and returning to control levels after 48 h (Figure 3D, upper panel).
The effect of in vivo cisplatin treatment on mtRNA steady-state levels was next examined. An increase in the mitochondrial rRNA steady-state level was found for both doses assayed (Figure 3C). The change was moderate for the 20 mg/kg dose, but reached a 3-fold increase for the 10 mg/kg dose. This effect was already evident after 5 h of cisplatin administration, reaching a maximum after 24 h and returning to control levels after 48 h (Figure 3D, lower panel).
Effect of in vitro addition of cisplatin on mtRNA metabolism in mitochondria isolated from in vivo cisplatin-treated animals
The addition of cisplatin to the incubation medium was found to directly inhibit mtRNA synthesis in organello, but the in vivo treatment of the animals with the drug increased mitochondrial transcription. In an attempt to clarify the effect of cisplatin on mtDNA metabolism and on mitochondrial activity, organelles from animals treated in vivo with 10 mg/kg cisplatin for 24 h were incubated for 1 h with doses of cisplatin ranging from 0 to 1.3 mM. As expected, in organello RNA synthesis was inhibited by cisplatin both in mitochondria isolated from in vivo treated and non-treated animals, but surprisingly, this inhibition was less pronounced in the mitochondria isolated from in vivo cisplatin-treated animals (Figure 4A, upper panel). In vitro addition of cisplatin to the incubation medium decreased mtRNA steady-state levels both in mitochondria isolated from control and in mitochondria isolated from in vivo cisplatin-treated animals, but the decrease was far more dramatic in the former case (Figures 4A, lower panel, and 4B).
The above results indicate the existence of in vivo mechanisms that reduce the damage of cisplatin to mtRNA metabolism.
Effect of in vivo cisplatin treatment on GSH concentration
To test the effect of in vivo cisplatin treatment on GSH concentration, GSH metabolism was analysed. GSH concentration was measured in total liver homogenates and in mitochondrial preparations from controls and from animals treated in vivo with cisplatin with a 10 mg/kg dose 5, 24 or 48 h prior to killing. As shown in Table 1, after 24 h of in vivo cisplatin treatment, the GSH levels increased both in total homogenate and in the mitochondrial fraction. The increase was significant in both cases when a non-parametric test was applied (the Mann–Whitney test, P=0.0495 and P=0.0495 respectively). After 5 and 48 h of in vivo cisplatin treatment, the GSH levels were slightly higher in both the total homogenate and mitochondrial fraction, but this increase was not significant in any case.
Mitochondria isolated from control and animals treated in vivo with 10 mg/kg cisplatin for 24 h were next treated with 0.2 mM of NEM for 15 min at room temperature. NEM modifies thiol groups, rendering them incapable of binding cisplatin. The HPLC analyses showed that >75% of mitochondrial GSH in control samples and >70% of mitochondrial GSH in in vivo cisplatin-treated samples were modified by NEM (GSH concentration was 2.28±0.1 and 2.71±0.5 nmol/mg of mitochondrial protein for control and cisplatin in vivo treated mitochondria after NEM treatment respectively).
Next, the effect of mitochondrial GSH concentration on mtRNA synthesis in organello and mtRNA stability was analysed (Figure 5). As described above (Figure 4A), in organello RNA synthesis was inhibited by the in vitro addition of cisplatin both to mitochondria isolated from in vivo treated animals and mitochondria isolated from non-treated animals, the inhibition being higher in mitochondria isolated from in vivo cisplatin non-treated animals (Figures 5A and 5B). Very interestingly, when equivalent samples of mitochondria were used but pre-incubated with NEM (+NEM), in organello RNA synthesis was completely abolished even in the absence of cisplatin. Moreover, mitochondrial depletion of GSH impaired mtRNA stability and reversed the protective effect on mtRNA stability induced by in vivo treatment with cisplatin (Figures 5C and 5D).
The results presented in this study, together with previous studies, allow us to propose that cisplatin accumulates in the mitochondria, binds to mtDNA and most likely also to mtRNA, inducing their physical modification. This modification would be responsible for the inhibition of mtDNA and mtRNA synthesis and the decrease in mtRNA stability. Since mtDNA encodes polypeptides that are components of the mitochondrial respiratory chain complexes, a direct effect of cisplatin on mtDNA metabolism could be the cause of the long-term decline of the mitochondrial respiration activity and ATP synthesis.
We show here that cisplatin causes a decrease in the mitochondrial oxygen consumption in isolated mitochondria and an increase in mitochondrial uncoupling between oxygen consumption and ATP synthesis. This result is in good agreement with previous studies describing that cisplatin induces a decline in mitochondrial respiratory activity in vivo  and inhibition of the mitochondrial respiratory chain complexes in vitro [23,35]. However, this inhibition was observed at doses and/or exposure times higher than those needed to inhibit mtDNA synthesis and expression. This would agree with recent studies showing that cisplatin caused little or no immediate effect on mitochondrial function in various cancer cell lines, tumour cells and beef heart submitochondrial particles as measured by oxygen consumption , but caused a long-term inhibition of mitochondrial oxygen consumption, which could be an indirect effect of the drug on mitochondrial activity .
By contrast, mtDNA synthesis appeared to be greatly inhibited at low cisplatin concentrations. It is well known that cisplatin binds to mtDNA. In studies with human malignant melanoma cells, cisplatin binding to mtDNA was 50-fold greater than to chromosomal DNA . Cisplatin is able to induce intra-strand and inter-strand adduct formation on mtDNA , and mtDNA cross-linking has been described in animals treated in vivo  and in in vitro treated cancer cells . However, little is known about the consequences of these lesions for mitochondrial activity. DNA adducts can interfere with DNA synthesis. We have shown that cisplatin caused an impairment of mtDNA when used as a template for PCR amplification. It has previously been demonstrated that cisplatin–DNA adducts do block Taq-polymerase, and PCR assays have been used to analyse the effects of cisplatin on genomic DNA as well as on mtDNA . Therefore the formation of DNA adducts seems to be the more plausible explanation for the strong inhibition of mtDNA replication induced by cisplatin.
DNA adducts can interfere with DNA expression as well. We have shown that cisplatin directly inhibits mtDNA expression. Previous reports revealed that cisplatin–DNA adducts block elongation during in vitro RNA synthesis by prokaryotic and eukaryotic RNA polymerases [41,42]. Additionally, a 2–3-fold decrease in transcriptional activity was observed when Pt-modified reporter genes were transfected into human or hamster cells . Cisplatin was also shown to substantially reduce transcription from the mouse mammary tumour virus promoter stably incorporated into mouse cells .
The decrease in mtRNAs steady-state levels was an interesting and unexpected observation. To our knowledge, such an effect had never been described previously in cisplatin treatments. The decrease in mtRNAs newly synthesized by isolated mitochondria could have been either due to modifications in mtDNA transcriptional activity or due to alterations of the mtRNA stability. We must stress that cisplatin induced both inhibition of mtRNA synthesis and decreased mtRNA stability, both effects being independent. One explanation would be that cisplatin interacts directly with mtRNAs and causes their structural modification, inducing their earlier degradation. In any case, this hypothesis requires further investigation.
Unexpectedly, in vivo treatment with cisplatin (10 mg/kg) promoted an increase in mtRNA synthesis in organello. Similarly, the stability of mtRNAs was increased in mitochondria isolated from animals treated in vivo with the drug. These results were apparently contradictory to the results we found in mitochondria incubated in vitro in the presence of the drug. A scenario that conciliates both the in vitro and the in vivo results can be envisioned. After in vivo treatment of the animals with cisplatin, the drug inhibits mtDNA synthesis directly and expression and decreases mtRNA stability. As a consequence, mitochondrial respiration is impaired and the cellular ATP levels are decreased. The cell reacts, triggering compensatory mechanisms to maintain cellular ATP levels. These would be the induction of genes involved in: (i) mitochondrial biogenesis, (ii) repair of mtDNA damage, (iii) cellular protection mechanisms such as decreased cellular uptake, (iv) increased cellular efflux of cisplatin, or (v) inactivation of the drug by thiol groups. These in vivo putative compensatory/defence mechanisms could be maintained, at least in part, in the mitochondria after their isolation and would explain the resistance to a direct alteration of mtDNA metabolism by cisplatin in mitochondria isolated from in vivo treated animals.
In support of increased mitochondrial biogenesis, it has been reported that in vivo cisplatin treatment caused an elevation of mtDNA levels in rat liver . The transcript level of mitochondrial 12S rRNA was increased by cisplatin treatment in two different cultured cell lines . Cisplatin may increase mitochondrial transcription by altering activities of nuclear-encoded transcription factors in response to mtDNA damage, oxidative stress or other metabolic signals. It has been shown in cell culture that incubation with cisplatin resulted in up-regulation of Tfam (mitochondrial transcription factor A) , which could explain the increase in mtRNA synthesis. However, the finding that mitochondria isolated from animals treated in vivo with the drug are more refractory, compared with control mitochondria, to the inhibition of transcription reinforces the idea that a protection mechanism is induced by cisplatin in vivo treatment.
In this respect, we analysed the levels of glutathione (GSH) as a potential protective pathway. GSH is critical in maintaining the protein thiol groups in a reduced state and in protection against oxidative stress through detoxification of reactive oxygen species . Glutathione can similarly detoxify many exogenous toxins, including cisplatin, through the formation of glutathione adducts . Pt ions entering the cell may preferentially bind to GSH and metallothionein, both present in millimolar concentrations in the cytoplasm. Formation of these complexes limits the amount of drug available for DNA binding and a positive correlation between GSH levels and resistance to cisplatin has been reported .
We found that in vivo treatment with cisplatin (10 mg/kg) induced an increase in the cytoplasmic and mitochondrial GSH concentrations as has been described in previous studies . Modification of mitochondrial GSH with NEM caused a dramatic impairment of mtRNA synthesis in organello. The influence of thiol groups on nuclear DNA platination by cisplatin has been demonstrated by an 8-fold increase in the amount of Pt–DNA adducts when cellular thiols were blocked by NEM . A similar effect could work for mitochondria incubated in vitro with NEM. Mitochondrial GSH depletion may leave mtDNA unprotected from cisplatin damage.
It should be noted that all the in vivo effects we have described here were observed when we used the 10 mg/kg dose. Treatment of the animals with a higher dose, 20 mg/kg, did not increase either mtRNA synthesis or stability. It is possible that raising the cisplatin dose overwhelmed the compensatory mechanisms and the cells were no longer able to overcome the toxicity. The fact that the animals treated with 40 mg/kg dose died before they were killed may show a strong alteration of cell viability.
The in vitro and in vivo cisplatin concentrations used in our studies are clinically relevant [35,37,51]. Cisplatin-induced nephrotoxicity and neurotoxicity are the most severe and dose-limiting side effects for the therapeutic use of cisplatin, but dose-dependent cisplatin-induced hepatotoxicity can alter the clinical situation in patients. It is known that cisplatin uptake by human liver is significant . Hepatotoxicity was reported in studies where similar doses of cisplatin to those used in our study were administrated to rats [52,53] and to humans .
In summary, the results presented in this paper are the first biochemical evidence of a direct effect of cisplatin on mtDNA metabolism. These results together with the evidence of the protective effect exerted by GSH are fundamental for a better understanding of the cisplatin mechanism of action on tumour cells and of its toxic side effects on non-tumour cells.
We thank Santiago Morales for his technical assistance. Our work was supported by the Spanish Ministry of Education (SAF2006-00428), the Health Institute Carlos III (REDEMETH-G03/054 and REDCIEN C03/06-Grupo RC-N34-3), the European Union [EUMITOCOMBAT (LSHM-CT-2004-503116)], DGA (Diputación General de Aragón) Research Group of Excellence grant (B55; to N. G., A. P.-M., P. F.-S. and J. A. E.), DGA Consolidated Research Group grant (B33; to M. J. L.-P. and J. M.) and ROSASNET (CSD2007–00020).
Abbreviations: MPA, meta-phosphoric acid; mtDNA, mitochondrial DNA; mtRNA, mitochondrial RNA; NEM, N-ethylmaleimide; NER, nucleotide excision repair; OXPHOS, oxidative phosphorylation; RCR, respiratory control ratio
- © The Authors Journal compilation © 2008 Biochemical Society