NCX-4040 (2-(acetyloxy)benzoic acid 4-(nitrooxymethyl) phenyl ester) was obtained from NicOx (Sophia Antipolis, France). GSH (L-glutamyl-L-cysteinylglycine), aspirin (ASA), dimethylsulfoxide (DMSO), buthionine sulfoximine (BSO), S-Nitroso-N-acetylpenicilamine (SNAP), MTT powder (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) were obtained from Sigma. 4-amino-5-methylamino-2', 7'-difluorofluorescein diacetate (DAF-FM DA was obtained from Molecular Probes. Cisplatin (cis-diamminedichloroplatinum), ammonium iron(II) sulfate hexahydrate, and acetonitrile were purchased from Aldrich. Cell culture medium (RPMI medium 1640), fetal bovine serum (FBS), antibiotics, sodium pyruvate, trypsin, and phosphate-buffered saline (PBS) were purchased from GIBCO/BRL. The imidazoline biradical probes R₁S-SR₁ (bis(2,2,5,5-tetramethyl-3-imidazoline-1-oxyl-4-il)disulfide) and R₂S-SR₂ (bis((2,2,3,5,5-pentamethyl-1-oxyl-imidazolidinyl-4)-methyl)-disulfide) were synthesized as previously reported 27. N-methyl-D-glucaminedithiocarbamate (MGD) was synthesized and purified in our laboratory 28. Polyvinylidene fluoride (PVDF) membrane and molecular weight markers were obtained from Bio-Rad (Hercules, CA). Antibodies for pEGFR (Tyr845, Tyr992), EGFR, pSTAT3 (Tyr705 and Ser727), poly-adenosine diphosphate ribose polymerase (PARP), cleaved caspases-3, Bcl-x_L, and Bax were purchased from Cell Signaling Technology (Beverly, MA). Antibodies for p53, Survivin, CycD1, and STAT3 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Enhanced chemiluminescence (ECL) reagents were obtained from Amersham Pharmacia Biotech (Buckinghamshire, UK).

Cisplatin-sensitive (A2780 WT) and cisplatin-resistant (A2780 cDDP) human ovarian cancer cell lines were used. The cisplatin-resistant cell line was originally developed from an in vivo tumor model by treating with cisplatin 29. Cells were grown in RPMI medium 1640 supplemented with 10% FBS, 2% sodium pyruvate, and 1% penicillin/streptomycin. Culture and drug-treatment of cells were carried out at 37°C in an atmosphere of 95% air/5% CO₂. Cells were routinely trypsinized (0.05% trypsin/EDTA) and counted using an automated counter (NucleoCounter, New Brunswick Scientific Co., Edison, NJ).

Cytotoxicity of the compounds was determined using the conversion of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) to formazan via mitochondrial oxidation. Cells were grown in 75-cm² flasks to 70% confluence, trypsinized, counted, and plated at a density of 20,000 cells per well in 96 well plates. Cells were incubated overnight and then treated with NCX-4040, ASA and/or cisplatin at variable concentrations for designated lengths of time. All experiments were repeated at least three times. NCX-4040 was dissolved in dimethyl sulfoxide (DMSO) and added into the culture. The final DMSO concentration never exceeded 1%, and this condition was used as the control in each experiment. Cisplatin was dissolved in DMSO (16 mg/ml) and stored at -20°C. The stock solution was diluted to the final concentration prior to each treatment.

The synergism of NCX-4040 and cisplatin was calculated as described 30. An expected value of cell survival, S_exp, was defined as the product of the survival observed for NCX-4040 and the survival observed for cisplatin (cDDP): S_exp = S_NCX-4040*S_cDDP. The actual survival observed for the combination treatment was defined as S_obs. The synergestic ratio, R, was calculated as: R = S_exp/S_obs. Synergy was defined for R > 1.

Generation of NO from NCX-4040 by cells incubated with the compound was measured by EPR spectroscopy using Fe(MGD)₂ spin trap. The spin trap reacts with NO and forms a stable complex Fe(MGD)₂-NO which can be detected using EPR. A2780 WT and A2780 cDDP cells were cultured in 75-cm² flasks. When cells reached 70% confluence, they were trypsinized and resuspended to a final concentration of 20 million cells per milliliter. The cell suspension was added with NCX-4040 (100 μM) and Fe(II)(MGD)₂, an NO spin-trap prepared in deoxygenated PBS by mixing 1 mM Fe²⁺ with 5 mM MGD. The NO measurements were performed using a Bruker ER300 X-band EPR spectrometer. The EPR spectra were collected at 20-mW microwave power using the following settings: modulation amplitude, 3 G; modulation frequency, 100 kHz; time constant, 8 ms; scan time, 21 s. SNAP was used as a positive control for NO detection.

Nitric oxide production in A2780 cDDP cells was analyzed using confocal fluorescence microscopic imaging. DAF-FM DA (4-amino-5-methylamino-2',7'-difluorofluorescein diacetate), a green fluorescence probe specific for NO, was added (10 μM) to cells treated with NCX-4040 (100 μM; 0, 1, 2 and 4 h). The probe was diluted in serum and phenol red-free culture medium. Cells were incubated with DAF-FM DA for 30 min at 37°C and then for 15 min without probe to allow complete de-esterification of the intracellular diacetate. Fluorescence measurements were made using a Zeiss LSM 510 multiphoton confocal microscope (excitation, 488 nm; emission, 515 nm). Quantitative analysis of fluorescence intensity was performed using MetaMorph Software.

Female 6-week-old BALB/c nude mice were obtained from the National Cancer Institute. The animals were housed five per cage in a climate- and light-controlled room. Food and water were allowed ad libitum. All animals were used according to the Public Health Services Policy, the Federal Welfare Act, and ILACUC procedures and guidelines. The mice were anesthetized with ketamine (200 mg/kg b.w.) and xylazine (4 mg/kg b.w.) by intraperitoneal (i.p.) injection. The animals inhaled room room air (21% O₂) during the in vivo spectroscopy, as well as for imaging. The body temperature of the animal was maintained at 37 ± 1°C by an infrared lamp placed just above the animal during the measurements.

Cisplatin-sensitive (A2780 WT) and cisplatin-resistant (A2780 cDDP) human ovarian cancer cells (5 × 10⁶ cells in 60 μl of PBS) were injected subcutaneously (s.c.) into the upper portion of the right hind limb of BALB/c nude mice. Both types of cells grew, in vivo, as a solid tumor. The size of the tumor was measured three times per week using a Vernier caliper. The tumor volume was determined from the orthogonal dimensions (d₁, d₂, d₃) using the formula (d₁*d₂*d₃)* π/6.

On the 7^th day after inoculation, when the tumor size reached approximately 3–5 mm, the mice were subjected to the treatment regime. The mice received i.p. injections of (1) 5 mg/kg b.w. of NCX-4040 alone, daily; (2) 5 mg/kg b.w. of ASA alone, daily; (3) 8 mg/kg b.w. of cisplatin on the 11^th day alone; or (4) 5 mg/kg b.w. of NCX-4040 daily, followed by cisplatin treatment on the 11^th day (8 mg/kg b.w.). The control group was administered with the vehicle alone (5). At the end of experiment (25^th day) the mice were sacrificed. The tumor tissues were then resected and stored in liquid nitrogen up to the time of thiol level determination and immunoblotting analysis.

Tissue levels of thiols were determined by X-band EPR spectroscopy, using a thiol-specific imidazoline biradical probe 31. The probe reacts with reduced thiol and shows a characteristic EPR spectrum, which can be quantified by using known amounts of glutathione (GSH). One hundred mg of thawed tissue were homogenized, in 2 ml of PBS, using a tissue homogenizer (PCR Tissue Homogenizing Kit, Fisher Scientific) to obtain a uniform tissue suspension. The tissue homogenate was treated with the thiol-specific label R₁S-SR₁ (0.5 mM) at pH 7.0 for 7 min, and the EPR spectra were measured. The concentration of GSH was determined from a standard curve prepared with known concentrations of GSH under similar conditions.

Tumor-bearing mice were intratumorally (i.t.) injected with 15 μL of thiol-sensitive probe, R₂S-SR₂ 27. Then EPR measurements were taken immediately and every 1.5 min thereafter for a total of 30 min, during which the signal-decay was observed. The measurements were performed using a home-built L-band (1.2 GHz) EPR spectrometer with a bridged loop-gap resonator 32. In vivo EPR redox images were obtained on the 25^th day of tumor growth. Mice were intratumorally injected with 15 μL of R₂S-SR₂ 27. The probe upon reaction with thiols is converted to nitroxide. The total redox information can be obtained from the decay of the nitroxide EPR signal, which depends primary upon reduction factors in the tissue. Mapping of the EPR probe location and its decay was obtained by spatial EPR imaging (EPRI) methods that are well established in our laboratory 3233. The EPRI measurements were performed by using a home-built L-band (1.2 GHz) EPR spectrometer with a bridged loop-gap resonator 32.

The tumor tissues were homogenized in a lysis buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 0.3 mM phenylmethylsulfonyl fluoride (PMSF), 0.2 mM sodium orthovanadate, 0.5% NP40, 1 μg/ml aprotinin, and 1 μg/ml leupetin. The homogenate was clarified by centrifugation. The protein concentration in the lysates was determined using a Pierce detergent-compatible protein assay kit (Rockford, IL). For immunoblotting, 50 to 100 μg of tissue protein per sample was denatured in 2× SDS-PAGE sample buffer (0.5 mM Tris HCl, 60% glycerol, 10% SDS, and β-mercapto-ethanol, bromophenol blue) and subjected to SDS-PAGE on a 10% or 12% tris-glycine gel. The separated proteins were transferred onto a PVDF membrane. The membrane was blocked with 5% nonfat milk powder (w/v) in Tris-buffered saline Tween-20 (TBST; 10 mM Tris, 100 mM NaCl, 0.1% Tween 20) for 1 h at room temperature or overnight at 4°C. The membranes were incubated overnight with the respective primary antibodies. The bound antibodies were detected with horseradish peroxidase (HRP)-labeled sheep anti-mouse IgG or HRP-labeled donkey anti-rabbit IgG (Amersham Pharmacia Biotech) using an ECL detection system. The ECL image was digitized using a scanner (ScanJet 7400c, Hewlett-Packard, Palo Alto, CA) and quantified using densitometry software (Scion Image v0.4.0.2, Scion, Frederick, MD).

Data were expressed as mean ± SD (in vitro study) or mean ± SEM (in vivo study). Comparisons among groups were performed by using a Student's t test. The significance level was set at p < 0.05.

NCX-4040 is a positional isomer of NCX-4016 (Figure 1A), which we have previously reported to be cytotoxic to human ovarian cancer cell lines 18. In order to evaluate the anticancer efficacy of NCX-4040 and to compare with that of NCX-4016, we performed a dose-response study using cisplatin-resistant cell line (A2780 cDDP). As seen in Figure 1B, a dose-dependent decrease of cell viability was observed in both cases. However, NCX-4040 showed a substantial decrease in cell viability when compared to NCX-4016. The cytotoxic effects of NCX-4040, either alone or in combination with cisplatin, on A2780 WT and A2780 cDDP cell lines were determined. Cells were treated with NCX-4040 (25 μM) for 6 h followed by cisplatin (IC₆₇ dose for A2780 WT cells) for 1 h with a 24-h follow up time. Cytotoxicity was assessed using an MTT viability assay. The results (Figure 2) showed that NCX-4040 alone significantly decreased the viability of both A2780 WT (34.9 ± 8.7%) and A2780 cDDP (41.7 ± 7.6%) cell lines. Cisplatin (cDDP) alone significantly decreased the viability of A2780 WT (31.5 ± 3.4%), but not A2780 cDDP (80.6 ± 11.8%) cell lines. On the other hand, combination of cisplatin and NCX-4040 was more effective than cisplatin or NCX-4040 alone in decreasing the viability of both A2780 WT (9.4 ± 6.0%) and A2780 cDDP (26.4 ± 7.6%) cell lines. A weak synergism was observed (R = 1.27, see Methods for calculation) between NCX-4040 and cisplatin in the killing of A2780 cDDP cells. The results clearly demonstrated that NCX-4040 was not only cytotoxic, but also capable of sensitizing cisplatin-resistant ovarian cancer cells to cisplatin.

To determine whether NCX-4040 can generate NO in cells, we performed EPR spectroscopic measurements on A2780 WT and A2780 cDDP cells incubated with 100 μM of NCX-4040. Fe(MGD)₂ was used as a spin-trap for real-time detection of NO. Fe(MGD)₂ reacts with NO and forms a stable paramagnetic adduct, Fe(MGD)₂-NO, which can be detected by EPR spectroscopy 34. NCX-4040 or cells alone in media did not generate any signal (Figure 3A). However, a prominent triplet signal with a hyperfine coupling constant of 12.5 G, characteristic of Fe(MGD)₂-NO, was observed upon incubation of NCX-4040 with the cells. The authenticity of the spectra was verified using S-nitroso-N-acetylpenicillamine (SNAP, 100 μM), a known NO-releasing compound. Both cell lines demonstrated a time-dependent increase in the intensity of Fe(MGD)₂-NO signal (Figure 3B) starting at 40–50 min of co-incubation time. While the rate of NO generation was steady in the case of A2780 WT cells, a rapid increase followed by slow attenuation was observed in the case of A2780 cDDP cells. In order to determine whether or not the increased rate of NO generation in A2780 cDDP cells was due to thiols, we pre-incubated A2780 cDDP cells with 1 mM buthionine sulfoximine (BSO), an inhibitor of glutathione synthesis, for 24 h and then treated with NCX-4040. As seen in Figure 3B, the NO generation was significantly decreased upon BSO treatment, suggesting that the NO generation was mediated by cellular thiols. In contrast, NO generation by SNAP was instant and lasted only for a few minutes (data not shown). We further imaged the intracellular NO generation using DAF-FM DA (Figure 3C). These results shown in Figure 3D clearly demonstrated a sustained generation of NO by NCX-4040 in the cancer cells. Thus, the NO generation by NCX-4040 in cancer cells was low, but persisted for longer periods.

The anti-tumor efficacy of NCX-4040 was studied using A2780 cDDP tumor xenografts in mice. Tumor-bearing animals, on day 7 post-inoculation, were treated with NCX-4040 and/or cisplatin, as described in the Methods section. The tumor volumes were measured every 2 days for 19 days post-inoculation. As shown in Figure 4, NCX-4040 treatment alone did not show any significant reduction in tumor volume (85.8 ± 9.8% versus Control on day 19; Figure 4B). Cisplatin (cDDP) treatment alone significantly reduced the tumor growth volume (74.0 ± 4.4% versus Control). However, animals treated with a combination of cisplatin and NCX-4040 showed a significant reduction in tumor volume visible from second day after cisplatin treatment and reached 56.4 ± 7.8% versus Control on day 19. Aspirin (ASA), a metabolic product of NCX-4040, did not show any significant effect on tumor growth volume (Figure 4B). The results clearly demonstrated that NCX-4040, in combination with cisplatin, was more effective than cisplatin alone in inhibiting the growth of cisplatin-resistant ovarian cancer xenografts in mice.

Since cisplatin-resistance in ovarian cancer is associated with excessive thiol levels, we next wanted to check whether NCX-4040 treatment could modify cellular thiol levels in the solid tumor, in vivo. We performed in vivo redox imaging of tumor xenografts in mice treated with NCX-4040. As we have reported previously, the in vivo redox mapping using EPR imaging enables noninvasive and real-time visualization of tumor redox status, which is a measure of total reducing equivalents, including reduced thiols, in the tissue 35. We used a thiol-specific di-nitroxyl redox probe R₂S-SR₂, which on reaction with SH groups breaks down to a mono-nitroxyl with a characteristic triplet EPR signal. The rate of conversion of di- to mono-nitroxyl is a measure of thiol concentration in the tumor. However, in solid tumors, the conversion is very rapid due to excess concentration of thiols compared to the probe. However, the subsequent slower reduction of the EPR-active mono-nitroxyl to EPR-silent hydroxylamine by the cellular reductants including thiols can be monitored and used as a measure of tissue "redox" state. Figure 5A shows typical time-course "redox" images obtained after intra-tumoral injection of R₂S-SR₂ (100 mM, 15 μl) in the cisplatin-sensitive (A2780 WT) and cisplatin-resistant (A2780 cDDP) ovarian xenograft tumors treated with NCX-4040. The time-course images showed that the decay of the redox signal was faster in the cisplatin-resistant tumor as compared to that of the cisplatin-sensitive tumor, presumably due to higher cellular reductants in the former cell type. Further, it was observed that NCX-4040 treatment resulted in a substantial slow-down of the redox decay, suggesting that the treatment might have depleted the reductants in the cisplatin-resistant tumor. Quantitative redox data from multiple experiments (Figure 5B) showed that the total redox level was significantly elevated in A2780 cDDP tumor, 182.1 ± 47.4% of A2780 WT tumors' redox. NCX-4040 treatment significantly depressed the total redox level, to 33.3 ± 12.8% (5.5-fold decrease), in A2780 cDDP tumor. We next wanted to measure the modulation of thiol-only levels in the tumors. We performed GSH assay from excised tumor tissues (in vitro) using EPR spectroscopy with another thiol-specific probe R₁SSR₁, as described in Methods. As seen in Figure 5C, the GSH level was 179.5 ± 42.3% in A2780 cDDP tumor and it was depleted to 38.5 ± 14.1% (a 4.7-fold decrease) after treatment with NCX-4040. The results clearly showed that NCX-4040 depleted thiol level in the cisplatin-resistant tumor.

We recently observed that NCX-4016, an isomeric form of NCX-4040, inhibited tumor growth by modulating EGFR/PI3K and STAT3 signaling pathways 19. Therefore, we analyzed the EGFR and STAT3 proteins in the A2780 cDDP tumor xenografts from mice treated with NCX-4040 and/or cisplatin. The immunoblots showed that the expressions of pEGFR (Tyr845 and Tyr992) and pSTAT3 (Tyr705, Ser727) were clearly downregulated in the tumors treated with NCX-4040 and cisplatin, when compared to NCX-4040 or aspirin alone groups (Figure 6A). The combination treatment also downregulated the cell-survival proteins including survivin, cyclin D1, Bcl-x_L, and upregulated the p53 protein. The expression of the proapoptotic protein Bax and apoptotic markers, including cleaved caspase-3 and PARP, were upregulated in the co-treatment group (Figure 6B–D). The results suggested that administration of NCX-4040, in combination with cisplatin, inhibited tumor growth by inducing apoptosis through downregulation of EGFR and STAT3 signaling in the cisplatin-resistant ovarian cancer xenograft in mice.