Troglitazone Acts on Cellular pH and DNA Synthesis through a Peroxisome Proliferator-Activated Receptor γ-Independent Mechanism in Breast Cancer-Derived Cell Lines Free

The ability to maintain elevated intracellular pH (pH_i) despite growth in a progressively acidic extracellular milieu confers selective advantages to tumor cells for both proliferation and invasiveness (1, 2, 3). Recent studies have shown that an alkaline pH_i is a fundamental step in the acquisition of an induced tumorigenic phenotype in various cells (4, 5). Despite the fact that tumor cells grow in an acidic milieu, they present enhanced glycolysis that enables both endogenous acid production and their reliance on acid extrusion mechanisms (6, 7, 8). The “glycolytic phenotype” has been recently associated with acid-mediated tumor invasion in mathematical models in which the tumor cell plays a major role by altering the surrounding microenvironment with acid excretion (9, 10). The acidification of the extracellular environment leads to destruction of the normal tissue and triggers cell death, angiogenesis, extracellular matrix degradation, and inhibition of the immune function (10). Whereas extracellular acidosis favors tumor growth, cellular acidosis leads to p53-dependent induction of apoptosis in various tumor-derived cell lines (10, 11). On the other hand, the permissive role of elevated pH_i favors DNA synthesis and promotes tumor growth (4, 12, 13). Although the role of Na⁺/H⁺ exchanger (NHE) 1 in tumor growth has been well known for more than a decade, more recently, signaling pathways involved in the activation of NHE1 have been explored in cells with a malignant phenotype (5, 14, 15). In this perspective, the acid–base balance in tumors has been targeted for therapeutic purposes through interference with the function of NHE1 and related biochemical pathways (16, 17, 18).

Troglitazone (TRO) is a thiazoladinedione that exhibits antihyperglycemic and antiproliferative actions (19, 20, 21, 22). In growth factor-induced cell growth, the antiproliferative activity of TRO can be related to signaling via peroxisome proliferator-activated receptor (PPAR) γ pathway and down-regulation of cyclins and cyclin-related kinases as well as hypophosphorylation of the negative regulatory retinoblastoma protein (19, 23, 24, 25, 26). However, in serum-independent tumor cell growth, TRO reduces proliferation by both PPARγ-dependent and -independent pathways (15, 22, 27, 28, 29, 30, 31, 32, 33, 34). Recent studies of PPARγ-negative cancer cell lines have shown that TRO reduces proliferation in a dose-dependent manner by partial intracellular Ca²⁺ depletion and Ca²⁺-mediated inhibition of translation initiation, cyclin D and E expression, or expression of the hyperphosphorylated form of the retinoblastoma tumor suppressor gene product as well as cyclin E formation (31, 32, 33, 34). Therefore, depending on the concentration and physiologic conditions, the antitumorigenic action of TRO appears to involve both PPARγ and PPARγ-independent pathways.

Another apparent PPARγ-independent action that may play an anticancer role is the ability of TRO to induce cellular acidosis, which we have demonstrated previously in normal cells (35, 36, 37). Our previous findings are consistent with glitazones inducing a spontaneous cellular acidosis associated with a shift in glutamine amino nitrogen metabolism from a predominantly anabolic to a catabolic pathway in canine mesangial cells (35). Consequently, we have shown that TRO induces a spontaneous cellular acidosis resulting from a reduction in cellular acid extrusion in murine mesangial cells (36). In the current study, we explore the action of TRO on pH_i in MCF-7 and MDA-MB-231 breast cancer cell lines and confirm that acidosis does in fact develop in these cell lines. We also assess the ability of TRO-induced acidosis to inhibit DNA synthesis in a dose-dependent manner and partially independent from PPARγ activation. Our results present a novel action of TRO as an antiproliferative compound that broadens the spectrum of various PPARγ-independent effects described previously (35, 36, 37).

MCF-7 and MDA-MB-231 breast carcinoma-derived cell lines and murine NIH3T3 fibroblasts were purchased from American Type Culture Collection (Manassas, VA). Cells were grown to confluence in Dulbecco’s modified Eagle’s medium (DMEM) plus 10% fetal calf serum (FCS) containing 28 mmol/L sodium bicarbonate, 10 mmol/L sodium pyruvate, 5 mmol/L d-glucose, and 2 mmol/L l-glutamine at 37°C (pH 7.4). TRO (kindly supplied by Dr. Tagata; Sankyo, Tokyo Japan) was dissolved in dimethyl sulfoxide (DMSO), added to the media at 100 μmol/L, and diluted to desired concentration; controls received the DMSO vehicle.

Cells were seeded on specially designed 30-mm chambers (Bioptechs, Biological Optical Technologies, Butler, PA). The chambers were placed uncapped inside a 60-mm covered tissue culture dish, incubated at 37°C in 5% CO₂ atmosphere, and allowed to gain confluence. The pH_i was determined with the pH-sensitive fluorescent dye (2,7)-biscarboxyethyl-5(6)-carboxyfluorescein [BCECF (Molecular Probes, Eugene, OR)] assay performed at 37°C by spectrofluorometry as described previously (35, 36). Changes in the emission ratio (490 nm/440 nm) were taken as an index of changes in pH_i. After each experiment, the high K⁺/nigericin technique was used to clamp the pH_i to media standards of known pH (confirmed on a Corning 240 pH meter at 37°C after withdrawing the sample from the chamber) obtaining a pH calibration of the 490 nm/440 nm signal ratio. To monitor the spontaneous pH_i response to TRO, the DMEM/10% FCS was replaced with fresh Krebs-Henseleit-HEPES (KHH) media in which 24 mmol/L HEPES buffer replaced bicarbonate, and the 490 nm/440 nm signal ratio was followed continuously for 12 minutes. Fresh KHH media minus TRO was then added to determine the reversibility of any TRO effect on pH_i. Cells were tested for pH_i in short-term (acute ΔpH_i) and long-term incubation with and without TRO (chronic ΔpH_i). To test for acid extrusion capability (acid load), the cells were incubated in KHH media (pH 7.4) and then acid loaded with a 4-minute exposure to KHH in which 20 mmol/L NaCl had been replaced with 20 mmol/L NH₄Cl. After returning the KHH media, the pH_i response was continuously monitored for 4 minutes. The recovery response was taken as the change in pH_i per time interval. Time control experiments for a repetitive acid load were also performed, establishing that the recovery rates were not different for the repetitive loadings.

Studies were performed on confluent cells previously grown in DMEM/10% FCS in which the media were replaced in the 12-well plates with KHH or KHH + TRO over a 12-minute period. Lactate concentration was determined enzymatically using lactate dehydrogenase (Sigma, St. Louis, MO) coupled to the reduction of NAD to NADH monitored as absorbance at 340 nm as described previously (38). The assay conditions included a hydrazine trap for pyruvate to force the reaction to completion. The production of lactate was expressed per milligram of proteins.

DNA synthesis was determined by following the incorporation of [methyl-³H]thymidine (70–86 Ci/mmol; Amersham, Piscataway, NJ). Briefly, the cells were incubated in 12-well plates in fresh DMEM/10% FCS plus 1 μCi/mL radiolabeled thymidine for 18 hours. The media were then withdrawn, and the cells washed three times with PBS followed by the addition of 5% trichloroacetic acid (TCA). After scraping TCA-treated cells, the cells were exposed to three freeze-thaw cycles and then pelleted by centrifugation. TCA supernatant was drawn off while the pellet was washed with 70% ethanol before the addition of 200 μL of 0.2 N NaOH and placed in a plate at 37°C for 24 hours. Aliquots of the digest were then added in duplicate to counting vials, and radioactivity was detected by liquid scintillation spectrometry. Protein concentration was determined by the dye binding assay (39). Results are expressed as cpm per milligram of protein.

For viability studies, cells were plated in duplicate (10⁵ cells per well) in 12-well dish in DMEM/10% FCS and allowed to attach overnight. The media were removed, and cells were washed with PBS to remove unattached cells. Media [DMEM/10% FCS (control) or DMEM/10% FCS + TRO] were added, and cells were harvested at different time points as specified. The viable cells were counted by trypan blue exclusion.

Cells were seeded on two-chamber permanox slides (Nalge-Nunc, Naperville, IL) and grown to approximately 70% confluence for immunofluorescence. Slides were fixed in 4% paraformaldehyde in PBS for 10 minutes, followed by 20 mmol/L glycine and 0.1% saponin in PBS for 20 minutes. Slides were incubated with either monoclonal antibody to PPARγ (Santa Cruz Biotechnology, Santa Cruz, CA) at 1:500 or monoclonal antibody to NHE1 (BD Biosciences, San Diego, CA) at 1:500 in BSP buffer (0.1% saponin and 2.5% bovine serum albumin in PBS) for 1 hour. After washing three times with PBS, AlexFluor 488 secondary antibody (Molecular Probes) was added at 1:500 for an additional hour in the dark. Slides were washed with PBS, and Hoechst 33342 (1 μg/mL; Molecular Probes) was added for 10 minutes for nuclear counterstain. Slides were mounted with Slowfade antifade reagent (Molecular Probes) and visualized.

For Western blot analysis, cells were seeded in a 6-well culture dish in duplicate at 2.5 × 10⁵ cells per well and allowed to attach and grow for 24 hours. Cells were lysed by the addition of 100 μL of ice-cold radioimmunoprecipitation assay buffer supplemented with Complete mini-protease inhibitor tablet (Roche Biochemicals, Indianapolis, IN). Protein concentration was determined by the BCA assay (Pierce, Rockford, IL). Sixty micrograms of total protein were loaded on a 10% SDS-PAGE gel. Proteins were transferred to a nitrocellulose filter and probed with primary antibody to PPARγ (Santa Cruz Biotechnology). The blots were stripped and reprobed for β-actin (Sigma) to serve as a loading control. Bands were developed using Immunostar-horseradish peroxidase chemiluminescent detection reagent (Bio-Rad, Hercules, CA).

Comparisons between multiple groups (nonrepeated and repeated measurements) were made using analysis of variance and a corrected Student’s t test, and comparisons between two groups were made using Student’s t test. A one-tailed t table was consulted when an a priori hypothesis was tested; otherwise a two-tailed table was used.

We have shown previously that TRO induces cellular acidosis in nontumorigenic cells expressing either NHE1 or NHE3 (35, 36, 37). Based on these observations, we asked whether TRO would decrease pH_i in breast carcinoma-derived cell lines MCF-7 and MDA-MB-231. As shown in Fig. 1,A and B, the steady-state pH_i decreased promptly after the replacement of KHH with KHH + TRO (25 μmol/L). The pH_i fell from the pretreatment average of 7.49 ± 0.10 to 6.77 ± 0.03 for MCF-7 cells (P < 0.001; n = 4), and from 7.38 ± 0.07 to 6.89 ± 0.09 for MDA-MB-231 cells (P < 0.05; n = 7) after 12 minutes, whereas pH_i for DMSO-treated time controls after 12 minutes was 7.72 ± 0.12 for MCF-7 cells (n = 3) and 7.27 ± 0.03 for MDA-MB-231 cells (n = 4), respectively (Fig. 1 C). Although the values of pH_i for DMSO-treated time controls were different compared with controls at time 0 for both cell lines, this difference was not statistically significant and was reproducible (data not shown). It is noteworthy that a significant (P < 0.01) decrease in the steady-state pH_i could be observed at 4 minutes after the addition of TRO in the MCF-7 cells. After 12 minutes, pH_i tended (P = 0.06) to be lower in the MCF-7 cells as compared with the MDA-MB-231 cells (6.66 ± 0.04 versus 6.89 ± 0.09, respectively).

TRO-induced cellular acidosis could result from an increased production of acid associated with accelerated glycolysis, particularly in these markedly acidogenic tumor cell lines (1, 2). To assess whether the spontaneous cellular acidosis observed in Fig. 1,C was associated with enhanced acid production, lactate appearance in the media was determined over the 12-minute time course. As shown in Fig. 1 D, the rate of lactate appearance was not increased by TRO in either cell line (4.44 ± 2.31 versus 5.89 ± 2.9 nmol/mg protein for MCF-7 cells and 15.72 ± 6.64 versus 15.59 ± 4.61 nmol/mg protein for MDA-MB-231 cells). These results point to the TRO-induced cellular acidosis resulting from decreased H⁺ extrusion in these tumorigenic cells.

We first confirmed the exchanger activity present in these two cell lines by immunostaining for NHE1, an important acid-extruding mechanism for maintaining the steady-state pH_i as well as for responding to a superimposed exogenous acid load (1, 2). NHE1 staining appeared highly expressed and uniformly distributed on the cell membrane of both cell lines (Fig. 2,A). To assess the effect of TRO on the ability of these cells to extrude acid, cells were exposed to an NH₄Cl acid pulse, and their recovery response was monitored. Representative responses for MCF-7 and MDA-MB-231 cells are presented in the left panels of Fig. 2,B and C. As shown in Fig. 2,B and C, left panels, addition of 20 mmol/L NH₄Cl to the media results in an initial alkalinization of the cells resulting from diffusion of NH₃ and intracellular formation of acid (NH₄⁺); removal of NH₄Cl from the media results in the rapid outward diffusion of NH₃ and a rapid cellular acidification. The lowest pH_i represents the “trough,” where the rates of acid loading and extrusion are equivalent, and the maximal stimulus for NHE activation. Thereafter, the activated acid extrusion exhibits a rapid response (RR) and a slower secondary response returning pH_i to the steady state. Addition of TRO to the recovery media virtually eliminated both the rapid and the low acid extrusion responses to the NH₄Cl load in the MCF-7 cells [ΔpH_i/Δt = 0.13 ± 0.01 to −0.012 ± 0.027 (RR; P < 0.001) and 0.082 ± 0.009 to 0.001 ± 0.009 (secondary response; P < 0.001)]. It is noteworthy that the trough obtained was no different between the first and second loadings (6.82 ± 0.09 and 6.74 ± 0.09, respectively), whereas the steady-state pH_i achieved was markedly reduced (P < 0.0001) in the presence of TRO (6.69 ± 0.06 versus 7.24 ± 0.08). The MDA-MB-231 acid extrusion response to the acid load is characterized by a more robust RR than that seen in the MCF-7 cells (ΔpH_i/Δt = 0.52 ± 0.30 versus 0.12 ± 0.01; P < 0.01), despite a similar trough pH_i consistent with the higher rate of acid production in MDA-MB-231 cells (Fig. 1 D). Nevertheless, TRO blunted the RR (0.52 ± 0.30 to 0.37 ± 0. 29; P < 0.01) and virtually eliminated the slow response (0.06 ± 0.02 to −0.002 ± 0.015; P < 0.001; n = 7). The steady-state pH_i achieved after the recovery from the NH₄Cl acid load was markedly reduced (P < 0.05) in the presence of TRO (7.73 ± 0.31 versus 7.10 ± 0.29).

To determine the effect of chronic TRO exposure on the ability of both cell lines to extrude acid (NHE activity), TRO or DMSO was added to the media for 18 hours, after which NHE activity was assayed in the absence of TRO. As shown in Fig. 3 A and B for representative experiments, 18-hour exposure to TRO (treatment) decreased assayable NHE activity (measured as the RR after the NH₄Cl pulse) as compared with the DMSO-treated control. For both the MCF-7 and MDA-MB-231 cells, NHE activity was decreased (P < 0.05) approximately 70% (ΔpH_i/Δt = 0.21 ± 0.03 to 0.07 ± 0.01 and 0.33 ± 0.08 to 0.10 ± 0.02, respectively). The chronic reduction in the capacity to extrude acid for both cell lines is consistent with a sustained cellular acidosis over the 18-hour time course during which DNA synthesis is measured.

Recent studies have shown that alkaline pH_i plays a major role in DNA synthesis, activation of tumor promoters, tumor transformation, and growth (5, 13). To determine whether TRO-induced acidosis inhibited DNA synthesis, we measured radiolabeled thymidine incorporation in both MCF-7 and MDA-MB-231 cells after 18 hours of incubation with TRO. As shown in Fig. 4 A, TRO inhibited [³H]thymidine incorporation in a dose-dependent manner in both the MCF-7 and MDA-MB-231 cell lines, although the 14% and 26% decrease at 10 μmol/L TRO did not achieve statistical significance. At 25 μmol/L TRO, both cell lines exhibited a marked reduction in DNA synthesis to 70 ± 3% (MCF-7; P < 0.05) and 43 ± 5% (MDA-MB-231; P < 0.01) of the control. At the highest concentration of TRO, 100 μmol/L, the synthesis of DNA was reduced >75% in both the cell lines (to 23 ± 2% and 22 ± 2% of control for MCF-7 and MDA-MB-231 cells, respectively). These results are consistent with the recent report that NHE activity is coupled via pH_i to DNA synthesis in tumor-promoting hepatocytes (13).

To determine the effect of TRO on cell growth, cells were counted over a 48-hour period with the results presented in Fig. 4,B and C for MCF-7 and MDA-MB-231 cells, respectively. TRO decreased the cell number in a dose-dependent fashion over the time, in line with the decrease in DNA synthesis at 18 hours shown above. At the highest TRO concentration, 100 μmol/L, the number of MCF-7 and MDA-MB-231 cells was 65% (P < 0.05) and 50% (P < 0.05) of their controls after 18 hours, respectively. These results show that cell numbers are reduced in parallel over a prolonged time with the TRO-induced inhibition of DNA synthesis shown in Fig. 4 A.

To determine whether the inhibitory effect of TRO-induced acidosis on [³H]thymidine incorporation was PPARγ mediated, we initially assessed the expression and cellular localization of PPARγ in MDA-MB-231 cells. As shown in Fig. 5,A and B, MDA-MD-231 cells show staining that is predominantly localized to the nucleus. To demonstrate that TRO reduces DNA synthesis independent of this signaling system, a selective and irreversible PPARγ antagonist, GW9662 (5 μmol/L), was added with TRO (25 μmol/L) or without TRO (DMSO vehicle), and the response was compared with that of TRO (25 μmol/L) alone. As shown in Fig. 5 C, TRO-induced inhibition of DNA synthesis (from 100 ± 7% in controls to 62 ± 4% with TRO; P < 0.05; n = 3) was not blocked by GW9662 (51 ± 8% with TRO + GW9662, P < 0.02 versus control). These results are consistent with the inhibitory effect of TRO on DNA synthesis being independent of the nuclear PPARγ signaling system.

To demonstrate that the above responses can be elicited in a cell line that does not express PPARγ, we performed parallel studies in PPARγ^−/− NIH3T3 cells (ref. 19; cells were confirmed to be negative in our laboratory; data not shown). The nontumorigenic NIH3T3 cell line exhibited a significantly (P < 0.05) lower pH_i than either tumorigenic cell line (7.18 ± 0.05 versus 7.54 ± 0.08 and 7.38 ± 0.07) as noted previously for nontumorigenic and tumorigenic cell lines (15, 18). Exposing PPARγ^−/− NIH3T3 cells to TRO (25 μmol/L) results in a rapid (Fig. 6,A) and profound cellular acidosis, with pH_i dropping significantly (P < 0.05) lower than either tumorigenic cell line (6.52 ± 0.02 versus 6.89 ± 0.09 and 6.66 ± 0.04 for MDA-MB-231 and MCF-7 cell lines, respectively) at 12 minutes (Fig. 6,B). The effect of TRO DNA synthesis as evaluated by tritiated thymidine incorporation is shown in Fig. 6 C. Even at the lowest concentration (10 μmol/L), TRO reduces DNA synthesis by >90%, despite the absence of a PPARγ signaling pathway in the NIH3T3 cell line (P < 0.001). These results clearly confirm that the effects of TRO-induced cellular acidosis on DNA synthesis are PPARγ independent.

TRO is a well-established antiproliferative and apoptotic agent with potential antitumorigenic activity (31, 32). Previous studies have shown that TRO has the possibility to influence cell proliferation through PPARγ-dependent and/or PPARγ-independent signaling pathways (23, 24, 27, 28, 29, 30, 31, 32, 33, 34, 40, 41). Our study is predicated on the observations in normal cells that TRO induces a spontaneous cellular acidosis, that cellular proliferation is pH dependent, and, that an alkaline pH_i is permissive for tumorigenesis (3, 4, 12, 15, 35, 36, 37, 41, 42, 43, 44, 45). The fact that cellular alkalinization plays this permissive role in transformation and proliferation has lead to therapeutic strategies based on inducing cellular acidosis through inhibiting NHE activity and hence reducing tumor growth and inducing apoptosis (17, 18, 46).

Precisely how a decrease in cytosolic pH acts to block transformation and inhibit the proliferative response is unclear, but there are several possibilities. One possibility is via pH-sensitive metabolic pathways whose contribution is required both to initiate and support cellular proliferation; in this regard, glycolysis is well known to be enhanced in cancer cells and to be acidogenic and thus requires an up-regulated acid-extruding system (10, 12, 15, 18). Another possibility is that acidosis induces down-regulation of retinoblastoma protein hyperphosphorylation resulting in cell cycle arrest. This event has been described to occur in both human colonic adenoma-derived epithelial cell lines and normal tubule-derived cells (11, 47). Interestingly, TRO down-regulates retinoblastoma gene expression and its conversion to the hyperphosphorylated form in both PPARγ^+/+ and PPARγ^−/− cells (29, 30, 31, 34).

Because we focused specifically on the heretofore little recognized role of TRO-induced cellular acidosis, our design was to study the effect of TRO on pH_i and how the induced acidosis acts on [³H]thymidine incorporation in alkalinized malignant MCF-7 and MDA-MB-231 cells, which express high levels of PPARγ. Therefore, our study did not investigate the solicited signaling pathways responsible for the cellular acidosis and associated reduced DNA synthesis. In fact, we show that TRO exerts a rapid effect on pH_i within 1 minute of incubation in both cell lines (Fig. 1 A and B). The rapidity of action implies an immediate effect that inevitably bypasses the slower signaling pathways solicited by activation of PPARγ, as shown previously in breast cancer-derived cell lines (27, 28). The decrement in steady-state pH_i with TRO did not differ between malignant (MCF-7, and MDA-MD-231) and nonmalignant cells (NIH3T3; ref. 37), but because the resting pH_i was higher in the malignant cells, the extent of cellular acidosis was less severe in the tumorigenic cells. We also show that TRO-induced cellular acidosis was attributable to inhibition of the active extrusion of acid via NHE, previously as shown for normal cells (35, 37, 38). TRO-induced cellular acidosis in the cell lines tested in this study is shared by another compound of the same family of thiazoladinedione, such as rosiglitazone, although the magnitude of acidosis induced by the latter is less severe than that induced by TRO (data not shown).

Although the mechanism of the effect of TRO to slow acid extrusion by this system has not been elucidated, it does not seem to be at the exchanger’s Na⁺-binding site because removal of the compound from the media did not restore exchanger activity, as occurs with inhibition by amiloride derivatives (43). Furthermore, the action of TRO on either spontaneous pH_i or NHE1 activity is additive rather than competitive with the action of amiloride analog 5-(N,N-dimethyl)amiloride on the exchanger.⁴ It is noteworthy that previous studies have shown that TRO induces an activation of signaling pathways that could potentially exert an effect on cellular acidosis (22, 34, 40). If TRO acts through an internal regulatory mechanism that maintains steady-state pH_i, rather than directly competing at the NHE Na⁺-binding site, then this compound or compounds modeled after TRO would have the therapeutic effectiveness of being able to act on tumors expressing both NHE1 and NHE3 isoforms, something that may provide a basis for looking at the subsequent acidosis-induced effects on proliferation in terms of the spontaneous acidosis. In fact, for example, a greater inhibition of DNA synthesis might be expected in MCF-7 cells exposed to TRO because, besides the developing cellular acidosis, these cells also express PPARγ. In a previous study of MCF-7 cell growth, TRO slowed growth in a dose-dependent manner, with a nearly 50% reduction after 48 hours with 40 μmol/L TRO; after 96 h, approximately 70% of the cells present at 48 hours had undergone cell death (30). Thus, these findings indicated that TRO not only suppressed cell growth but also apparently activated cell death pathways. The growth-suppressing effect occurring over the first 24 to 48 hours was associated with decreased expression of cyclin D1, retinoblastoma protein dephosphorylation, and G₁ cycle arrest in previous studies (30).

In our study, 25 μmol/L TRO reduced the pH_i, and inhibited DNA synthesis as reflected by reduced [³H]thymidine incorporation after 18 hours (Fig. 4,A). It is noteworthy that the nontumorigenic cell line NIH3T3 exhibited a lower pH_i with TRO than did the tumorigenic cells MCF-7 and MDA-MD-231, although the activity of NHE-mediated acid extrusion was not different. This relatively higher pH_i presumably explains why TRO was less effective in inhibiting DNA synthesis in the tumorigenic cell line than in normal cells (Fig. 5 B and C). Alternatively, cancer cells may defend themselves from excessive lowering of pH_i by different mechanisms than normal cells in an attempt to protect their proliferation and growth. In this case, a prolonged acidification as observed with TRO may prime the cell for activation of either PPARγ-dependent or -independent mechanisms of cell death occurring at a later time after the initial cellular acidosis. Further work is under way in this perspective.