6-(7-Nitro-2,1,3-benzoxadiazol-4-ylthio)hexanol, a specific glutathione S-transferase inhibitor, overcomes the multidrug resistance (MDR)-associated protein 1–mediated MDR in small cell lung cancer

Lung cancer is the leading cause of cancer-related deaths and is classified clinically into two major histologic types: small cell lung cancer (SCLC), which accounts for ∼25% of cases, and non-SCLC, which accounts for the remaining 75%. SCLC is a particularly aggressive form of lung cancer and has an extremely poor prognosis (1, 2). Current information suggests that, in most cells, defects in apoptosis signaling are the most critical determinants for tumor development and progression (3, 4). The antiapoptotic protein Bcl-2 is a critical survival factor for SCLC cells, and it has been shown that down-regulation of Bcl-2 induces apoptosis and enhances chemosensitivity in these tumor cells (5, 6). Moreover, after an initial response to chemotherapy, there is a relapse in most patients with SCLC, and their tumors become largely refractory to further treatments. P-glycoprotein (P-gp) and the multidrug resistance (MDR)-associated protein 1 (MRP1) are the most extensively studied ATP-binding cassette transporters and are often regarded as the prototypes for understanding the cell-based mechanisms of MDR (7–9). The clinical importance of P-gp and MRP1 stems from the fact that, in parallel with their overexpression, the tumor cells simultaneously exhibit acquired cross-resistance to diverse chemotherapeutic agents, resulting in the failure of chemotherapy for many cancers. To study the MDR phenotype of SCLC, Mirsk et al. (10) have developed a drug-resistant SCLC cell line by selection with Adriamycin. This cell line, designated H69AR, does not overexpress P-gp. Thus, it provides a model system for investigating MRP1-mediated MDR. We have recently reported that 6-(7-nitro-2,1,3-benzoxadiazol-4-ylthio)hexanol (NBDHEX) is not a substrate of the P-gp export pump and both P-gp–overexpressing tumor cells and their parental cell lines are efficiently committed to death by this molecule (11). NBDHEX is a nonpolar compound that behaves as a suicide inhibitor for glutathione S-transferases (GST). In fact, it is conjugated with reduced glutathione (GSH), so leading to a stable σ complex in the GST active site (12). The MRP1 export pump is able to transport hydrophobic anions, including drugs conjugated with GSH (GS-X pump), glucuronide, and sulfate. In addition, the MRP1 pump removes unmodified anticancer agents cotransported with GSH (13, 14). Therefore, the anionic adduct between NBDHEX and GSH could represent a substrate for the MRP1 transporter and it may be extruded from the cell. In the present work, we have analyzed the activity of NBDHEX against both H69 tumor cells and their Adriamycin-resistant counterpart H69AR, which overexpresses MRP1.

Human SCLC H69 and its Adriamycin-selected, multidrug-resistant H69AR cell variant were kindly provided by Dr. Susan P. Cole (Department of Pathology, Queen's University, Kingston, Ontario, Canada). Cells were grown in RPMI 1640 supplemented with 10% FCS and antibiotics. H69AR cells were maintained by alternate feedings with drug-free medium or medium containing 1.7 μmol/L Adriamycin. The stability of the MDR phenotype was periodically assessed by testing the MRP1 expression with the MRP1-specific monoclonal antibody MRPm6 (VinciBiochem; ref. 15). MDR cells were also typed for possible P-gp expression with the monoclonal antibody MM4.17, which recognizes an extracellular P-gp epitope (16).

Cell viability testing was done using the sulforhodamine B assay (17). In brief, H69 and H69AR cells were placed in 96-well microtiter plates at a density of 0.7 × 10⁶/mL and then incubated 48 h with NBDHEX at the desired final concentration (from 0.05 to 20 μmol/L). A positive control was also done by treating both H69 and H69AR cells with 10 μmol/L puromycin, a dose that causes 100% cell death after 48 h (data not shown). After 48 h of incubation, cell growth was evaluated by an in situ cell fixation procedure followed by sulforhodamine B staining of proteins. In order not to exceed the absorbance linearity range, a suboptimal wavelength (490 nm) was used to read the samples (17). The absorbance value of the positive control was subtracted from that of both NBDHEX-treated and untreated control samples. The dose-response profile obtained fulfills the LC₅₀ for NBDHEX (the concentration used to obtain 50% cellular mortality).

Treatments were done on cells cultured at 0.8 × 10⁶/mL. NBDHEX was synthesized as reported by Ricci et al. (12). Stock solutions of this compound were prepared in DMSO and diluted to the appropriate concentration in RPMI 1640 cell medium just before use.

The caspase-dependent nature of the NBDHEX-induced apoptosis was confirmed by measuring caspase activity and apoptosis in cells treated with the general caspase inhibitor Z-VAD-fmk at a concentration of 40 μmol/L. The inhibitor was added to the cell culture 1 h before the addition of 3 μmol/L NBDHEX and maintained throughout the experimental time.

The protective effect of GSH on the NBDHEX-induced necrosis (H69) and apoptosis (H69AR) was determined after the addition to the cells of 1 mmol/L glutathione ethylester 1 h before the addition of 5 μmol/L NBDHEX.

Cells were incubated for 1 h in the presence of 1.7 μmol/L Adriamycin or 10 μmol/L NBDHEX in the presence or absence of 50 μmol/L MK571 (VinciBiochem), a specific inhibitor of MRP1 (18). Then, cells were incubated for 1 h in drug-free medium with or without MK571 to give the opportunity for efflux or efflux blocking to occur. Efflux was stopped by pelleting the cells and adding ice-cold medium. Fluorescence of Adriamycin or NBDHEX was analyzed by a FACSCalibur flow cytometer (BD Biosciences) equipped with an argon laser source.

Percentages of apoptosis and necrosis were determined on cells treated for 24 h with different concentrations of NBDHEX (from 0 to 5 μmol/L).

For apoptosis determination, aliquots of cells were stained for 10 min with Hoechst 33342 and examined by fluorescence microscopy. For necrosis determination, aliquots of cells were stained with trypan blue and examined by microscopy using a Neubauer hemacytometer. Alternatively, cells were stained with 50 μg/mL propidium iodide, permeabilized with NP40 (0.1%, w/v), and analyzed by a FACSCalibur instrument.

The intracellular levels of both reduced (GSH) and oxidized form (GSSG) of glutathione were determined at different times from the treatment with 3 μmol/L NBDHEX, as previously reported (19). Data are expressed as nmol of GSH or GSSG equivalents/mg of protein.

GST and caspase-3 activities were measured as described previously (20). Protein concentration was determined by the bicinchoninic acid protein assay reagent (Pierce).

Immunoblotting analysis was done on H69 and H69AR cell lines. At each time point analyzed, pellets of untreated control cells and cells treated with 3 μmol/L NBDHEX were lysed for Western blot analyses as described previously (20). Twenty micrograms of proteins were loaded on 12% polyacrylamide gel and transferred onto a nitrocellulose membrane (Bio-Rad). Monoclonal anti–phosphorylated c-Jun NH₂-terminal kinase (JNK) and anti–phosphorylated c-Jun (1:500; Santa Cruz Biotechnology); polyclonal anti-JNK and anti–c-Jun (1:1,000; Upstate Biotechnology), anti-p38^MAPK (1:1,000; Santa Cruz Biotechnology), anti–phosphorylated isoform of p38^MAPK (1:1,000; Cell Signaling Technology and New England Biolabs), anti–β-actin (1:5,000; Sigma), and anti–α-tubulin; and monoclonal anti–Bcl-2 (1:500; Dako) and anti–GSTP1-1 (1:2,000; Calbiochem) were used as primary antibodies. The protein complex formed on incubation with specific secondary antibodies (1:10,000; Bio-Rad) was identified using FluorChem Imaging System (Alpha Innotech) after incubation with ChemiGlow chemiluminescence substrate (Alpha Innotech).

Densitometric analyses of proteins were done using the software Quantity One (Bio-Rad), and protein levels were normalized to the density of related β-actin bands.

All of the experiments were repeated at least thrice. Results are presented as mean ± SD. Statistical evaluation was done, where appropriate, using the Student's t test or ANOVA followed by correction with Bonferroni's t test. The criterion for statistical significance used was P < 0.05.

Flow cytometry studies, conducted by specific antibodies, show that MRP1 is not present in the drug-sensitive H69 parental cell line. On the other hand, this export pump is highly expressed in the MDR cell variant H69AR, as assessed by the intracellular detection of the MRPm6 epitope by its cognate monoclonal antibody (Fig. 1A). In contrast, P-gp was not detected in both cell lines by the anti–P-gp monoclonal antibody MM4.17 (data not shown). Thus, the H69 and H69AR drug-sensitive/resistant cell pair provides a model system for investigating MRP1-mediated MDR. To clarify if NBDHEX efflux could be mediated by MRP1, we did experiments in the presence of MK571. This molecule is known to specifically inhibit the MRP1-mediated transport, so inducing intracellular accumulation of MRP1 substrates such as Adriamycin. As expected, MK571 inhibited Adriamycin efflux in H69AR cells (Fig. 1E), although it did not affect the intracellular concentration of Adriamycin in the sensitive variant (Fig. 1C). Conversely, MK571 did not affect the intracellular concentration of NBDHEX in both sensitive and resistant cell lines (Fig. 1B and D, respectively).

Moreover, a recently isolated NBDHEX-resistant cell variant of the human T-lymphoblastic leukemia CCRF-CEM does not show detectable MRP1 or P-gp multidrug transporters (data not shown).⁷

The ability of NBDHEX to cause cell death in H69 and H69AR cells is affected by an increase of cell density. Thus, taking advantage of the wide range of linearity of the sulforhodamine B assay, the NBDHEX cytotoxic activity has been determined by using a cell density comparable with that used in the apoptosis/necrosis experiments. Under these experimental conditions, the dose-response profiles revealed a good cytotoxic activity both in sensitive H69 cell line (LC₅₀ of 2.3 ± 0.6 μmol/L) and in its Adriamycin-resistant counterpart H69AR (LC₅₀ of 4.5 ± 0.9 μmol/L; Fig. 1F).

To characterize the type of cell death induced by NBDHEX, we evaluated the amount of apoptotic/necrotic cells in both H69AR and H69 cell lines after incubation with the drug. Figure 2 reports the percentages of cells undergoing necrosis (Fig. 2A) or apoptosis (Fig. 2B and C) after incubation with NBDHEX. The occurrence of apoptosis in H69AR was revealed by the presence of cells with fragmented chromatin (data not shown). Time course analysis of cell death showed that apoptosis occurred within 3 h and continued to increase throughout the incubation period. Moreover, a clear dose-dependent effect was observed after 24 h of incubation with different amounts of NBDHEX. By contrast, the H69 cell line seemed quite insensitive to the treatment up to ∼1 μmol/L NBDHEX, whereas a further increase of drug concentration caused cell death by necrosis as determined by trypan blue staining. The necrotic H69 cells exhibited cytoplasmic swelling and diffuse nonfragmented chromatin, suggesting that the phenomenon observed is primary necrosis instead of apoptosis-derived secondary necrosis (data not shown). To corroborate this hypothesis, we did cytofluorimetric analysis of both H69 and H69AR cells on staining with propidium iodide. H69AR treated with different amounts of NBDHEX showed a dose-dependent increase of cell population with hypodiploid (sub-G₁) nuclei, in which DNA cleavage occurred as the ultimate step of apoptosis. Conversely, this trend was absent in the parental H69 cells, which showed no increase in sub-G₁ peak when treated with the same amounts of NBDHEX (Fig. 2D). Taken together, these results indicate that the observed necrosis in H69 cell line is a bona fide apoptosis-independent, primary necrosis.

Activation of the caspase family proteases is also a central biochemical event occurring only in apoptotic cells. Therefore, the specific proteolytic activity of caspase-3 was assayed at intervals after drug treatment in both H69 and H69AR cell lines. Figure 2E shows caspase-3 activity after 24 h of incubation with different amounts of NBDHEX. H69 cells showed only a slight increase in caspase-3 activity after 2 μmol/L NBDHEX treatment, which quickly dropped to background levels by increasing drug concentration. In contrast, caspase-3 activity was already observed in H69AR cells treated with 1 μmol/L NBDHEX and increased progressively with the increase of NBDHEX concentration. Caspase-3 inactivation by the general caspase inhibitor Z-VAD-fmk fully prevented apoptosis in H69AR cells (Fig. 2F), so confirming the caspase involvement in the apoptotic response of cells exposed to NBDHEX.

Mitogen-activated protein kinases (MAPK) are important components of signaling pathways, which transduce extracellular stimuli into intracellular responses. It has been reported that NBDHEX is able to induce the activation of JNK pathway in a variety of cancer cells (11, 20). Therefore, the effect of NBDHEX on the phosphorylation/activation of this MAPK was examined both in H69 cell line and in its Adriamycin-resistant counterpart. As shown in Fig. 3, treatment with NBDHEX increased the phosphorylation of JNK in H69AR cells in a time-dependent fashion; phosphorylated JNK activation occurred as early as 30 min and remained at high levels for at least 12 h. c-Jun, a direct downstream transcription factor activated by JNK, was also increased in the H69AR cells after 3 h of NBDHEX treatment, suggesting an involvement of the JNK/c-Jun pathway in the NBDHEX-induced apoptosis.

In contrast, neither change in the intensity of the immunoreactive band of phosphorylated JNK nor c-Jun activation was detected in the NBDHEX-treated H69 cells. It is well known that the GST isoenzyme GSTP1-1 inhibits JNK activity by forming a GSTP1-1-JNK heterocomplex (21). Therefore, to rule out the possibility that a higher GSTP1-1 expression in H69 cells could be responsible for the lack of JNK activation (11), the amount of this enzyme was determined both in the H69 cell line and in its MDR counterpart. No differences in the GSTP1-1 levels were observed between these cell lines (Fig. 4A). Neither were any significant changes in the GST activity observed between NBDHEX-treated and untreated control cells up to 6 h of incubation (Fig. 4B). These findings indicate that NBDHEX triggers a different cell death mechanism in H69 cells.

To evaluate the effects induced by NBDHEX on proapoptotic members of MAPK family, we next analyzed the change in the phosphorylation/activation of p38^MAPK. Immunoblot analysis of H69 cell line showed a marked increase in phosphorylation of p38^MAPK, which occurred within 30 min of drug treatment, whereas NBDHEX induced only a negligible increase of p38^MAPK phosphorylation in the H69AR cell line (Fig. 5A). It has been suggested that p38^MAPK senses very low changes of the intracellular redox state (22). Thus, to show a possible change of this redox system after NBDHEX treatment, we measured the intracellular content of both reduced and disulfide form of glutathione, the most abundant low molecular weight antioxidant within the cell. Parental H69 cell line showed an average intracellular concentration of GSH of ∼60 nmol/mg protein, a value 3-fold higher than that of the Adriamycin-resistant counterpart (∼20 nmol/mg protein; Fig. 5B). Moreover, a time-dependent increase of GSSG content was detected in the NBDHEX-treated H69 cell line but not in the H69AR cells (Fig. 5C). This increase was not due to a reduced activity of the NADPH-dependent glutathione disulfide reductase that continuously reduces GSSG back into GSH. In fact, H69 cells show a GSH reductase activity of 0.090 unit/mg (μmol/min/mg a pH 7.6), which is comparable with the value found in the resistant H69AR cell line (0.080 unit/mg). It is known that the intracellular redox status depends on the relative amounts of the GSH and GSSG (22) and that the oxidation of a limited amount of GSH into GSSG can drastically change this ratio, so leading to activation of p38^MAPK and subsequent apoptosis. This was already reported for the K562 cell line treated with NBDHEX (20). However, the imbalance of the intracellular redox state and the activation of p38^MAPK found in H69 cells were not followed by the apoptotic cascade, and in this case, cells died by necrosis.

An increase in intracellular oxidation may determine the selection between apoptosis and necrosis. In particular, suppression of apoptosis and the induction of necrosis is often associated with the intracellular GSH content (23, 24). Therefore, we tested the possibility of switching the NBDHEX-induced necrosis in apoptosis by increasing the intracellular level of GSH. We incubated H69 cells with glutathione ethylester, a permeant compound that directly increases the intracellular levels of GSH (25). Figure 5E shows that treatment with glutathione ethylester significantly prevented NBDHEX-mediated necrosis. In fact, the percentage of necrotic cells, after 24 h of treatment with 5 μmol/L NBDHEX, decreased from 100% to 36% in the presence of 1 mmol/L glutathione ethylester, although no switch from necrosis to apoptosis was observed. On the other hand, an increase of the intracellular levels of GSH partially improved (∼25%) the survival of H69AR cells after 24 h of treatment with 5 μmol/L NBDHEX (Fig. 5D).

SCLCs, and particularly H69 cells, are known to express high levels of the antiapoptotic protein Bcl-2 (26). This was confirmed by Western blot analysis, which conversely showed a very low amount of Bcl-2 protein in the MPR1-overexpressing H69AR cell line (Fig. 6). It can be hypothesized that the strong down-regulation of this antiapoptotic protein in H69AR cells may favor the occurrence of apoptosis following NBDHEX treatment.

SCLC is characterized by a highly aggressive clinical course and a strong tendency to develop distant metastases. This makes chemotherapy more effective than surgical or radiation therapy. Consequently, we have investigated the efficacy of the nitrobenzoxadiazole-derivative NBDHEX on the human SCLC H69 cell line and on its MDR variant H69AR. Surprisingly, we found that NBDHEX was cytotoxic in both cell lines with LC₅₀s in the micromolar range. Interestingly, this drug triggers two different types of cell death: a caspase-dependent apoptotic pathway in the Adriamycin-resistant H69AR and a necrotic phenotype in the parental H69 cell line. To explain this result, some of the internal cell factors that counteract apoptosis in lung cancer have been analyzed. We found higher GSH levels and Bcl-2 overexpression in H69 cells compared with levels found in the Adriamycin-resistant H69AR cells characterized by an almost undetectable Bcl-2 content. Such evidence matches several studies showing a positive correlation between elevated Bcl-2 and GSH levels and suppression of apoptosis (27, 28). The important role that these antiapoptotic factors play in lung cancer progression is confirmed by the action mechanism of Adriamycin, which triggers apoptosis in H69 cells by down-regulation of Bcl-2 expression and depletion of GSH (29). However, under our experimental conditions, neither GSH nor Bcl-2 content decreased in H69 cells after treatment with NBDHEX (Figs. 5 and 6, respectively). Moreover, an increase of intracellular GSH concentration did not switch necrosis to apoptosis, suggesting that the amount of intracellular GSH does not control by itself the mode of cell death in H69 cell line. It has also been reported that Bcl-2 influences intracellular signaling cascades, thereby preventing apoptosis. In particular, Bcl-2 blocks the activation of the JNK signaling pathway by various apoptotic stresses (i.e., staurosporine UV light and anisomycin) by suppressing the stimulation of the upstream protein kinase MAPK kinase kinase (30, 31). These findings may explain the lack of JNK/c-Jun activation and the absence of apoptosis in H69 cells despite their being provided with an effective apoptotic machinery. The resulting necrosis observed after NBDHEX treatment may constitute a default cell death pathway induced when the activation of specific signal transduction cascades leading to cell death is inhibited (32). In spite of the adverse effects of necrosis, it could be therapeutically advantageous to trigger cell death by necrosis in lung cancer cells selected to resist apoptosis and characterized by extremely poor prognosis.

As far as MDR H69AR cells are concerned, we have found that the MRP1 transporter overexpressed in this cell line does not counteract the cytotoxicity of NBDHEX, which is not a substrate for this pump. NBDHEX induces a caspase-dependent apoptosis and this process may be promoted by the very low amount of Bcl-2 found in H69AR cells. The decrease of this antiapoptotic protein seems linked to the MDR phenotype. In fact, previous data show the concomitant down-regulation of Bcl-2 gene in response to a significant overexpression of the MRP1 gene in H69 cells (33, 34). Moreover, it has been shown that the amount of intracellular GSH levels is a critical factor for MRP1-expressing cells. In particular, MRP1 expression causes a reduction in cellular GSH levels and this makes H69AR cells hypersensitive to drugs, such as verapamil and flavonoids, which increase the MRP1-mediated GSH transport. Accordingly, addition of GSH to the medium efficiently prevents apoptosis induced by these drugs in MRP1-expressing cells (35, 36). Under our experimental conditions, we observed that an increase in intracellular GSH levels partially improved the survival of H69AR cells treated with NBDHEX. Moreover, NBDHEX does not seem to affect the correct redox equilibrium of this cell line. In fact, no decrease in GSH levels and no accumulation of oxidized glutathione are observed during NBDHEX treatment (Fig. 5). We cannot rule out the hypothesis that GSH could be extruded from the cells under our experimental conditions; however, this probably occurs in a long time frame so as to allow the cell to induce GSH neosynthesis. Therefore, the role played by GSH in the apoptosis of NBDHEX-treated H69AR cells is not completely understood at this time and will be the goal of future investigations. Whatever the case, a down-regulation of factors crucial for cell survival makes H69AR cells more sensitive to the cytotoxic action of drugs that are not MRP1 substrates. Actually, MRP1 transports not only a wide variety of neutral hydrophobic compounds but also numerous glutathione, glucuronate, and sulfate conjugates (13), and only a few anticancer drugs are poor substrates (37) or not substrates of MRP1 at all (38). Thus, many attempts have been made to identify agents able to reverse the MDR mediated by MRP1 (39). However, many of these molecules in clinical trials revealed intrinsic toxicity at doses effective in reversing MDR (40). Finally, to enhance the efficacy of chemotherapy in MDR lung cancer, complex strategies based on the simultaneous administration of an anticancer drug and of antisense oligonucleotides targeting antiapoptotic defense have been proposed (41). Taken together, all this evidence emphasizes the potential anticancer role of NBDHEX, a compound that is effective against drug-resistant tumors overexpressing both P-gp (11) and MRP1 and that also shows low toxicity in vivo (20).