NSC23925 prevents the development of paclitaxel resistance by inhibiting the introduction of P-glycoprotein (Pgp) and enhancing apoptosis

Abstract

Strategies to prevent the emergence of drug resistance will increase the effectiveness of chemotherapy treatment and prolong survival of women with ovarian cancer. The aim of the current study is to determine the effects of NSC23925 on preventing the development of paclitaxel resistance in ovarian cancer both in cultured cells in vitro and in mouse xenograft models in vivo, and to further elucidate these underlying mechanisms. We first developed a paclitaxel-resistant ovarian cancer cell line, and demonstrated that NSC23925 could prevent the introduction of paclitaxel resistance by specifically inhibiting the overexpression of Pgp in vitro. The paclitaxel-resistant ovarian cancer cells were then established in a mouse model by continuous paclitaxel treatment in combination with or without NSC23925 administration in the mice. The majority of mice continuously treated with paclitaxel alone eventually developed paclitaxel resistance with overexpression of Pgp and anti-apoptotic proteins, whereas mice remained sensitivity to paclitaxel and displayed lower expression levels of Pgp and anti-apoptotic proteins after administered continuously with combination paclitaxel-NSC23925. Paclitaxel-NSC23925 treated mice experienced significantly longer overall survival time than paclitaxel-treated mice. Furthermore, the combination of paclitaxel and NSC23925 therapy did not induce obvious toxicity as measured by mice body weight changes, blood cell counts, and histology of internal organs. Collectively, our observations provide evidence that NSC23925 in combination with paclitaxel may prevent the onset of Pgp or anti-apoptotic-mediated paclitaxel resistance, and improve the long-term clinical outcome in patients with ovarian cancer.

Introduction

Ovarian cancer is the most lethal of all gynecologic malignancies, and ranks fifth in lethal tumors among women.¹ Paclitaxel was first approved by Food and Drug Administration (FDA) as the first-line chemotherapy agent for the treatment of ovarian cancer in 1992. There is an up to 80% response rate to paclitaxel in ovarian cancer patients with no prior treatment.² However, more than half of patients ultimately relapse and the response rate to paclitaxel dramatically decreases due to the development of drug resistance.^3-5 In recurrent ovarian cancer, conventionally dosed 3-weekly paclitaxel only leads to response rates of 16%–33%.⁵ The inevitable development of paclitaxel resistance remains the most significant challenge in the treatment of ovarian cancer.⁶

Overexpression of P-glycoprotein (Pgp), a member of the ATP-binding cassette (ABC) transporters superfamily, is a well-known molecular mechanism involved in the development of multidrug resistance (MDR), including paclitaxel resistance in cancer. Increased expression of Pgp actively pumps a structurally and functionally diverse set of chemotherapy agents out of the cancer cells, such as taxanes (paclitaxel, docetaxe, cabazitaxel), anthracyclines (doxorubicin, daunorubicin, epirubicin, idarubicin), and vinca alkaloids (vinblastine, vincristine, vindesine), which results in decreased intracellular accumulation of anticancer drugs. Paclitaxel is known to exert its anticancer activity through anti-mitotic and non-mitotic events, such as intracellular trafficking and motility, which eventually induce cell death by apoptosis or necrosis. Evasion of apoptosis regulated by apoptotic regulatory and mitosis checkpoint proteins has been identified to underlie paclitaxel resistance.^7-9 For example, survivin overexpression has been demonstrated to be associated with paclitaxel resistance in advanced ovarian cancer.¹⁰ Anti-apoptotic proteins Bcl-xL and MCL-1 are able to cooperate to protect ovarian cancer cells against chemotherapy drug-induced apoptosis.¹¹ Recent studies have suggested that a subset of cancer cells, called cancer stem cells, may also be an important source of tumor recurrence and drug resistance in ovarian cancer.^{12, 13}

Preventing or delaying the emergence of paclitaxel resistance may greatly aid to improve the clinical outcome of cancer patients. Inhibiting the overexpression of Pgp has become a promising strategy to prevent the development of MDR. Several compounds have been shown to prevent the overexpression of Pgp and therefore prevent the introduction of drug resistance in cancer. Studies have demonstrated that well-characterized Pgp inhibitors valspodar (PSC833), biricodar (VX-710), and tariquidar (XR9576) are able to prevent the development of vincristine resistance by inhibiting Pgp overexpression in pediatric rhabdomyosarcoma in vitro.¹⁴ The selective type 2 cyclooxygenase (COX-2) inhibitor NS-398 can prevent or reduce the introduction of doxorubicin resistance by suppressing Pgp levels and function in breast cancer.¹⁵ Dexrazoxane (ICRF-187), a catalytic DNA topoisomerase II inhibitor, also significantly prevents doxorubicin resistance in leukemia by delaying expression of functional Pgp.¹⁶ The anti-recombinogenic substance (E)-5-(2-bromovinyl)-2′-deoxyuridine (BVDU) held the potential to inhibit the development of doxorubicin resistance by preventing MDR1 amplification and Pgp overexpression in mouse leukemia cells.¹⁷ Similarly, Pluronic P85 (P85) block copolymer has been proven to suppress the induction of doxorubicin resistance by preventing the development of Pgp overexpression in leukemia and breast carcinoma.^{18, 19} Reports have also demonstrated that the selective retinoid X receptor agonist bexarotene (LGD1069, targretin) was able to prevent the development of paclitaxel resistance in advanced breast cancer and non-small cell lung cancer by modulating MDR1 expression through maintaining/increasing genomic integrity.^20-22 However, almost none of these agents have yielded a tolerable and effective therapy to prevent MDR in the clinical setting.

A number of clinical trials of the above agents have been conducted for numerous cancers, including ovarian cancer patients.^23-26 Unfortunately, clinical use of these agents has not generated substantial and satisfactory outcome benefits, which has largely precluded their widespread application in clinic. For example, valspodar has shown unexpected pharmacokinetic interaction with paclitaxel, doxorubicin, valspodar, etoposide, and mitoxantrone.^{23, 27, 28} Although no detrimental pharmacokinetic interactions between biricodar and paclitaxel have been detected, myelosuppression and nonhematologic toxicity are the major adverse effects.^{25, 29} Most importantly, the combination of vincristine, doxorubicin, and biricodar did not show significant benefits on antitumor activity or survival in small cell lung cancer patients.³⁰ A phase II study of tariquidar in stage III-IV breast carcinoma patients with chemotherapy resistance suggested tariquidar showed limited clinical activity to restore chemosensitivity to taxane or anthracycline chemotherapies.³¹ Therefore, more potent and selective MDR inhibitors and investigations on their mechanisms are needed.

NSC23925 (2-(4-methoxyphenyl)-4-quinolinyl) (2-piper-idinyl) methanol, http://pubchem.ncbi.nlm.nih.gov/) was identified previously as a novel, selective and effective Pgp inhibitor.³² NSC23925 has the ability to regulate Pgp ATPase activity and therefore increase intracellular accumulation of chemotherapeutic agents. NSC23925 in combination with doxorubicin significantly induces cell death and apoptosis compared with doxorubicin alone.³³ It has been shown that NSC23925 can restore chemosensitivity to anticancer drugs in a variety of multidrug resistance (MDR) cancer cells in vitro, such as MDR ovarian cancer, breast cancer, colon cancer, non-small lung cancer, and osteosarcoma. In addition, paclitaxel and NSC23925 combination therapy exhibited a more pronounced inhibitory effect on MDR ovarian cancer tumor growth than paclitaxel alone in vivo.³³ Recently, we demonstrated that the preclinical use of NSC23925 in combination with paclitaxel at the onset of chemotherapy produced a significant efficiency on preventing the emergence of MDR in osteosarcoma cell lines, U-2OS and Saos, in vitro.³⁴ However, the effects and underlying mechanisms of NSC23925 on impeding the induction of paclitaxel resistance in ovarian cancer, especially in vivo, need further investigation. The goal of our current study is to determine whether the combination of paclitaxel and NSC23925 can prevent the development of paclitaxel resistance in cultured cells in vitro and in xenograft mouse model in vivo, and to further elucidate the underlying molecular mechanisms.

Materials and Methods

Cell line, drugs, and antibodies

Human ovarian cancer cell line SKOV-3 was purchased from American Type Culture Collection (ATCC) (Manassas, VA). SKOV-3 cells were cultured in RPMI 1640 (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS), 100-U/mL penicillin, and 100 mg/mL streptomycin (Life Technologies) in 5% CO₂-95% air atmosphere at 37°C. Paclitaxel was provided by the pharmacy at the Massachusetts General Hospital. Previously, we demonstrated that 4 different isomers resulting from two chiral centers were exhibited in the structure of NSC23925. The most efficient one for the reversal paclitaxel resistance is isomer 11.³³ Therefore, isomer 11 of NSC23295 (abbreviated as NSC23925) was used in the present study and synthesized by ChemPartner Co. (Chengdu, China). The mouse monoclonal antibodies to human Pgp and MRP1 were obtained from Sigma-Aldrich (St. Louis, MO). The mouse monoclonal antibody to human BCRP was purchased from EMD Millipore (Bedford, MA). The apoptotic relative antibodies survivin, Bcl-xL, Cyclin D, and Cyclin E were purchased form Cell Signaling Technology (Beverly, MA). The mouse monoclonal antibodies to human Cytochrome C and MCL-1 were purchased from BD Biosciences (San Jose, CA). Mouse anti-CD44 and rabbit anti-Integrin β3 antibodies were purchased from Cell Signaling Technology. The respective secondary antibodies were purchased from LI-COR Bioscences (Lincoln, NE).

Development of paclitaxel resistant ovarian cancer cell line in vitro

To determine whether NSC23925 can prevent the emergence of paclitaxel resistance, paclitaxel resistant ovarian cancer cells were established as previously described.^{18, 19, 34} In brief, 1×10⁵ SKOV-3 cells were suspended in culture media containing paclitaxel alone, 1 μM NSC23925 alone, or paclitaxel in combination with 1 μM NSC23925. When the cells were cultured to 90% confluence, 1×10⁵ cells were reseeded in a new tissue culture flask, and the paclitaxel dose was increased stepwise. The initial concentration of paclitaxel was 0.0001 μM. At different selection points cells sublines were collected and stored at liquid nitrogen for further analysis.

MTT assay

To assess chemosensitivity of paclitaxel in different selected cell sublines, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay was conducted as previously described.³⁵

Western blot assay

Expression levels of ABC transporters, including Pgp, MRP1 and BCRP, were determined by Western blot as described previously ³⁶. The different selected cell sublines were harvested and lysed using RIPA lysis buffer (Upstate Biotechnology, Charlottesville, VA) added with complete protease inhibitor cocktail tablets (Roche Applied Science, IN, USA). The concentrations of proteins were assessed by the DC Protein Assay (Bio-Rad, Hercules, USA). Equal amounts of protein (20 μg) were resolved on a NuPAGE® 4–12% Bis-Tris Gel (Life Technologies) and transferred onto nitrocellulose membranes (Bio-Rad). After blocking with 5% nonfat milk in TBST for 2 h, the membranes were probed with primary antibodies (dilution: 1:1000) overnight at 4°C. Following, the membranes were washed with TBST and incubated with their respective secondary antibodies (dilution: 1:20000) for 1 h. Finally, the blot signals were scanned by Odyssey® CLx equipment (LI-COR Bioscences) and quantified using Odyssey software 3.0 (LI-COR Bioscences).

Establishment of SKOV-3 paclitaxel-resistant cells in vivo

The protocol for animal use in this project was approved by the Massachusetts General Hospital Subcommittee on Research Animal Care (SRAC) under the protocol number 2013N000121. The Crl:SHO-Prkdc^SCIDHr^hr nude female mice at approximately 3 to 4 weeks of age were purchased from Charles River Laboratories (Ann Arbor, MI). To evaluate the effects of NSC23925 on the induction of paclitaxel resistance in vivo, the paclitaxel resistant cells were established in human ovarian cancer xenograft models following previously described protocols with minor modifications ^{37, 38}. Briefly, on day 1, approximately 2 × 10⁶ parental sensitive SKOV-3 cells were injected subcutaneously with Matrigel (BD Biosciences) into the flanks of 3 - 4-week-old female nude mice. Administration was initiated 12 days after injection of tumor cells. The mice were randomized into 4 groups and treated intraperitoneally with either saline alone, NSC23925 alone (50mg/kg), paclitaxel (25 mg/kg) alone, or paclitaxel (25mg/kg) in combination with NSC23925 (50mg/kg) twice per week (Tuesday and Friday) for 3 weeks followed by a treatment-free interval of 2 weeks. The second round of treatment was then continued. The schedule of treatment was outlined as in Fig. 1.

Figure 1.

Schedule of paclitaxel treatment in ovarian cancer drug resistant xenograft mouse model. (a) The mice were treated with saline alone, 50 mg/kg NSC23925 alone, 25 mg/kg paclitaxel alone, or 25 mg/kg paclitaxel in combination with 50 mg/kg NSC23925 12 days after tumor implantation. The mice were treated with drug twice a week for 3 weeks followed by a treatment-free interval of 2 weeks. Administration was continued for another 3 weeks followed by a treatment-free interval of 1 week. (b) A cartoon explanation of treatment method in SKOV-3 cell line ovarian cancer drug resistant xenograft mouse model.

The size of tumors was recorded twice a week (Tuesday and Friday) beginning on day 13. Tumor volume (V_t, mm³) was measured with a digital caliper and calculated according to the formula (length × width²)/2. Relative tumor volume (V_r, mm³) and was estimated using the formula V_t/T₀ × 100_, where T₀ represents the tumor volume at the time of treatment commencement. Tumor volume percent (%) change (ΔT/T₀ %) for the treated mice was then calculated by the formula (T_s-T₀) ×100/T₀, where T_s was the volume on day sacrificed. Survival time was recorded from the date of tumor cells injection to the date of death or the last day of experiments. Animals were sacrificed once the tumor length exceeded 10 mm during the experiments. All remaining mice were sacrificed on day 72. Tumor tissue samples were excised from mice, weighed, and stored at -80°C for protein extraction.

Safety study

In the current study, safety evaluation following the treatment with the combination of paclitaxel and NSC23925 in tumor bearing mice was carried out by assessing the changes in body weight, white blood cell (WBC) and red blood cell (RBC) count, and the liver, kidney, spleen histology. The mice were weighed twice per week (Tuesday and Friday) and the body weights were calculated as percentage change to the weight at the start of treatment. Mice were sacrificed one week after the last dosing (day 72), and peripheral blood, liver, kidney, and spleen were collected. The blood was collected in tubes containing 0.1 M EDTA to avoid clotting. For WBC and RBC counts, 10 μL of blood was mixed with 190 μL of ACK lysing buffer (Life Technologies), or 2 μL blood was mixed in 1 mL saline, respectively. WBC and RBC cells were counted by a Z1 Coulter particle counter (Beckman Coulter). For histology of organs, paraffin sections of liver, kidney, and spleen from mice were deparaffinized, rehydrated, and stained by H&E according to manufacturer's protocol (ScyTek Laboratories, Inc. Logan, UT). Images were captured under a 40× objective using Nikon Eclipse Ti-U microscope (Nikon Instruments, Inc NY, CA) equipped with a SPOT RT^TM digital camera.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 5 software (GraphPad Software, Inc., La Jolla, CA). Data were expressed as the mean ± SEM. The unpaired two-sided Student's t-test and Mann-Whitney test were used accordingly to evaluate the differences between two groups. Percent survival was analyzed using Kaplan-Meier survival curves with Gehan-Breslow-Wilcoxon test for significance. P value < 0.05 was considered as statistically significant.

Results

Generation of paclitaxel-resistant ovarian cancer cells in vitro

In an attempt to investigate the effects of compound NSC23925 on the introduction of paclitaxel resistance in vitro, SKOV-3 cells were cultured with stepwise increasing dosages of paclitaxel in combination with or without 1 μM NSC23925. After over a half-year of drug exposure, cells cultured with paclitaxel alone (SKOV-3/paclitaxel) grew vigorously in culture medium with the presence of 0.3 μM paclitaxel. However, cells selected with paclitaxel in combination with NSC23925 (SKOV-3/paclitaxel-NSC23925) could not survive in the medium with more than 0.001 μM paclitaxel (Fig. 2a). Administration of paclitaxel alone or combination paclitaxel-NSC23925 to SKOV-3 cells was performed in duplicate and quadruplicate, respectively, and similar phenomena were observed. Meanwhile, SKOV-3 cells with long-term exposure of 1 μM NSC23925 (SKOV-3/NSC23925) showed stable growth in culture medium (data not shown). Selected cell sublines were harvested at different dose points for further experiments, marked by arrows in Fig. 2b.

Figure 2.

NSC23925 prevented the emergence of paclitaxel resistance in ovarian cancer by specifically preventing the overexpression of Pgp in cultured cells in vitro. (a) The time course to establish paclitaxel-resistant ovarian cancer cell sublines. Parental SKOV-3 cells were exposed to a stepwise increase in concentrations of paclitaxel in combination with or without NSC23925. Quadruplicate or duplicate experiments (labeled with a, b or a, b, c, d, respectively) were performed independently and simultaneously. Cell sublines selected at different dose points were marked by arrows and used in subsequent experiments. (b) IC₅₀ of paclitaxel in different selected cell sublines. SKOV-3/parental represent parental SKOV-3 cells, SKOV-3/NSC23925 represent SKOV-3 cells cultured with 1 μM NSC23925 alone, paclitaxel_0.3 represent SKOV-3 cells selected with 0.3 μM paclitaxel, and paclitaxel_0.001-NSC23925 represent SKOV-3 cells cultured with 0.001 μM paclitaxel in combination with 1 μM NSC23925. (c) Expression level of Pgp, MRP1 and BCRP in different selected cell sublines.

Subsequently, the IC₅₀ of paclitaxel was assessed to further identify paclitaxel resistance in selected cell sublines. As expected, the IC₅₀ of paclitaxel in SKOV-3/paclitaxel cells increased along with the stepwise increased drug exposure (time and dose), while no dramatic enhancement to the IC₅₀ for paclitaxel in SKOV-3/paclitaxel-NSC23925 cells was detected. Specifically, SKOV-3 cells treated with 0.3 μM paclitaxel alone (SKOV-3/paclitaxel_0.3) showed a significantly higher IC₅₀ for paclitaxel than its parental cells (SKOV-3/parental). An increase of 420.9-fold to the IC₅₀ for paclitaxel was shown in the SKOV-3/paclitaxel_0.3 cell subline in comparison with SKOV-3/parental. Compared with the IC₅₀ of paclitaxel in SKOV-3/parental cells, the IC₅₀ of paclitaxel increased by merely 0.8-fold in SKOV-3/paclitaxel_0.001-NSC23925 cells (Fig. 2b, Supporting Information Table S1). Additionally, no dramatic change in the IC₅₀ for paclitaxel was found in SKOV-3/NSC23925 cells. These results indicated that paclitaxel-NSC23925 combined treatment can prevent the development of paclitaxel resistance.

The expression level of Pgp was then determined to further clarify the underlying mechanisms in preventing the development of paclitaxel resistance by NSC23925. Pgp levels were increased in accordance with the increased concentration of paclitaxel tolerated by cells. Pgp was remarkably overexpressed in SKOV-3/paclitaxel_0.3 cells as compared with SKOV-3/parental, whereas no significant expression of Pgp was observed in SKOV-3/paclitaxel-NSC23925 cells (Fig. 2c). Furthermore, we evaluated other critical ABC transporters that account for drug resistance mechanisms. Although all of the SKOV-3 cell sublines expressed low levels of BCRP and undetectable protein expression of MRP1, there was no significant difference of BCRP and MRP1 levels among the cell sublines (Fig. 2c). Collectively, these observations indicated that NSC23925 specifically inhibited Pgp overexpression to prevent the emergence of paclitaxel resistance during paclitaxel treatment.

Establishment of paclitaxel-resistant ovarian cancer cells in vivo

To test the effects of NSC23925 on the emergence of paclitaxel resistance in vivo, tumor bearing mice were continuous treated with paclitaxel in combination with or without NSC23925. Individual tumor growth curves were presented in Fig. 3a. Both saline alone and NSC23925 alone treated tumors grew progressively (blue curves and red curves, respectively). Two types of response were observed in tumors treated with paclitaxel alone (green curves). Consistent with clinical observation, the majority of tumors became paclitaxel-refractory. 66.7% tumors treated with paclitaxel eventually showed increasing progressive tumor growth after no increase in tumor growth initially during continued paclitaxel treatment. 33.3% tumors showed no remarkable increase in tumor volume and revealed prolonged paclitaxel response. For comparison, all paclitaxel-NSC23925 treated tumors exhibited no obvious increase in tumor volume during the continued treatment (purple curves). Paclitaxel-NSC23925 treated mice showed marked and sustained response to paclitaxel without becoming paclitaxel-refractory. These results indicated that the usage of NSC23925 in paclitaxel chemotherapy produced significantly prolonged anticancer efficacy of paclitaxel.

Figure 3.

NSC23925 prevented the development of paclitaxel resistance in ovarian cancer mouse models in vivo. (a) Individual tumor growth dynamics in ovarian cancer mouse models. The mice bearing subcutaneous SKOV-3 xenografts were treated with saline alone, 50 mg/kg NSC23925 alone, 25 mg/kg paclitaxel alone, or 25 mg/kg paclitaxel in combination with 50 mg/kg NSC23925 (blue, red, green and purple curves, respectively). (b) Individual tumor images in four groups photographed on the day when the mice were sacrificed. The mice, from top to bottom, came from the group of mice treated with saline (S), NSC23925 (N), paclitaxel (P), and paclitaxel-NSC23925 combination (PN), respectively. n = 4-6 mice/group. (c) The mean relative tumor volume in each group. (d) Tumor volume percent change (ΔT/T₀ %) in each group. (e) Tumor weights on the day when the mice were sacrificed in four groups. S, N, P and PN represent the group of mice treated with saline, NSC23925, paclitaxel, and combination paclitaxel-NSC23925, respectively. The data were presented as mean ± SEM and analyzed using an unpaired two-sided Student's t-test or Mann-Whitney test accordingly. Statistical comparisons between paclitaxel treated mice and paclitaxel-NSC23925 combination treated mice are presented: * P < 0.05, ** P < 0.01, *** P < 0.001. (f) Survival of ovarian cancer bearing mice treated by different methods. Paclitaxel-treated mice experienced significant shorten overall survival time as compared with paclitaxel-NSC23925 treated mice (P < 0.05).

Individual tumor images were photographed on the day when the mice were sacrificed and shown in Fig. 3b. Relative tumor volumes and tumor weights in each mouse were also recorded, calculated, and analyzed. Our results suggest tumors treated with paclitaxel-NSC23925 presented with a significantly (P < 0.01) lower volume than paclitaxel-treated tumors, further indicating paclitaxel-NSC23925 treated mice remained responsive to paclitaxel (Fig. 3c). There was no significant alteration in tumor volume percent change in paclitaxel-NSC23925 treated mice. Tumor volume percent change in paclitaxel-treated mice was dramatically higher (P < 0.01) than that in paclitaxel-NSC23925 treated mice, owing to the development of paclitaxel resistance (Fig. 3d). Moreover, the tumor weight in paclitaxel-treated mice was also significantly higher compared with that in paclitaxel-NSC23925 treated mice (Fig. 3e, P < 0.01). Paclitaxel-treated tumors had a mean weight of 403.0 mg, while paclitaxel-NSC23925 treated tumors averaged 40.00 mg. The saline-treated and NSC23925-treated tumor weights at the time of sacrifice were 417.4 mg and 491.8 mg, respectively. Taken together, these findings demonstrated that NSC23925 prevented the development of paclitaxel resistance in vivo.

Because some paclitaxel-treated tumors developed paclitaxel resistance, the volume of tumors was too large to be sacrificed during the drug treatment. Paclitaxel-treated mice experienced significantly shorten overall survival times as compared with paclitaxel-NSC23925 treated mice (Fig. 3f, P < 0.05). The poor outcomes of paclitaxel-treated mice suggested that NSC23925 contributed to the notable survival benefit of paclitaxel-NSC23925 combination therapy in mice. These results may support the preclinical use of NSC23925 at the onset of chemotherapy to suppress the emergence of paclitaxel resistance as well as improve the overall survival rate.

Expression of Pgp in tumor tissues of ovarian cancer mice model

We next sought to identify the underlying molecular mechanisms of NSC23925 in preventing the development of paclitaxel resistance. The expression levels of Pgp were tested using Western blot in mice tumor tissues. As expected, a majority of paclitaxel treated mice showed a significant increase in Pgp expression level (Fig. 4a). Only a small number of tumors showed low levels of Pgp expression. In contrast, the expression level of Pgp was undetectable in all paclitaxel-NSC23925 treated mice. Interestingly, the mice treated with either saline or NSC23925 alone naturally expressed low levels of Pgp. Notably, saline or NSC23925 treated tumors expressed higher Pgp than paclitaxel-NSC2395 treated tumors. We also tested the expression level of other ABC transporter proteins, BCRP and MRP1 in the tumor tissues of mice (Fig. 4a). The densities of Western blot bands were quantified and the mean expression levels of Pgp and BCRP in different treated groups were shown (Fig. 4b). Paclitaxel-NSC23925 treated mice revealed a remarkably lower Pgp level as compared with paclitaxel treated mice (P < 0.01). The average Pgp expression level of paclitaxel-NSC23925 treated mice and paclitaxel treated mice was 0.0184 and 0.3678, respectively. No obvious differences in the expression level of BCRP and MRP1 were observed between paclitaxel-NSC23925 and paclitaxel treated mice (P > 0.05); this suggested that administration of NSC23925 to mice selectivity and specifically suppressed overexpression of Pgp and therefore prevented the emergence of paclitaxel resistance in continuous paclitaxel treatment. Moreover, we evaluated whether the expression level of Pgp was correlated with tumor volume and overall survival time. A positive and strong correlation was observed between Pgp expression level and tumor volume in paclitaxel-NSC23925 and paclitaxel treated mice (Fig. 4c, r = 0.8364, P < 0.01). The paclitaxel-NSC23925 treated mice with smaller tumor volumes had lower Pgp expression. The correlation coefficient of -0.7171 with P < 0.05 suggested that overall survival time was inversely and highly correlated with the expression of Pgp in paclitaxel-NSC23925 and paclitaxel treated mice (Fig. 4d). The combination of paclitaxel and NSC23925 held the potential to prevent the overexpression of Pgp, and therefore improve the long-term outcome. These results collectively indicated that NSC23925 held the ability to prevent the emergence of paclitaxel in mouse models by suppressing the overexpression of Pgp during the course of paclitaxel administration.

Figure 4.

NSC23925 specifically prevented the overexpression of Pgp in ovarian cancer mouse models during continuous paclitaxel treatment. (a) The expression level of Pgp, MRP1 and BCRP in all mice tumor tissues. S, N, P and PN represented the group of mice treated with saline, NSC23925, paclitaxel and paclitaxel-NSC23925 combination, respectively. (b) Densitometric quantification of the expression level of Pgp and BCRP. The data were presented as Mean ± SEM. Statistical comparisons between paclitaxel treated mice and paclitaxel-NSC23925 combination treated mice are presented: ** P < 0.01. (c) Correlation analysis of Pgp expression level and tumor volume in mice treated with paclitaxel alone and paclitaxel-NSC23925 combination (P = 0.0022, r = 0.8364). (d) A correlation between Pgp expression level and overall survival time (P = 0.0162, r = -0.7171).

Expression of apoptotic-related proteins in ovarian cancer mice model

To investigate whether the prolonged anticancer efficiency of paclitaxel observed in paclitaxel-NSC23925 combination therapy could involve the induction of increased apoptotic activity, the expression levels of several apoptotic-related proteins were evaluated in mouse tumor samples. Anti-apoptotic proteins survivin, Bcl-xL, and MCL-1 showed significantly lower expression levels in paclitaxel-NSC23925 treated mice than paclitaxel alone treated mice (Fig. 5a and 5b, P < 0.001, P < 0.05, P < 0.01, respectively). CD44 and Integrin β3, which are cancer stem cell markers, also displayed significantly decreased expression in paclitaxel-NSC23925 treated mice as compared with that in paclitaxel alone treated mice (Fig. 5a and 5b, P < 0.05 and P < 0.01, respectively). In addition, the expression level of Cyclin E decreased significantly in paclitaxel-NSC23925 treated mice (Fig. 5a and 5b, P < 0.001). There were no statistically significant differences in Cyclin D and Cytochrome C expressions between paclitaxel-NSC23925 and paclitaxel alone treated mice (Fig. 5a and 5c, P > 0.05). These findings suggest that treatment with combination paclitaxel-NSC23925 to mice could induce increased apoptosis, which may be contributed to the additional mechanism by which NSC23925 prevents the development of paclitaxel resistance in ovarian cancer.

Figure 5.

The paclitaxel-NSC23925 combination treatment promoted increased apoptosis in mouse models in vivo. (a) Expression levels of apoptotic regulatory proteins in all mice tumor tissues as determined by Western blot. S, N, P, and PN represent the group of mice treated with saline, NSC23925, paclitaxel, and paclitaxel-NSC23925 combination, respectively. (b) Densitometric quantification of the expression level of survvin, Bcl-xL, MCL-1, CD44, Integrin β3, and Cyclin E. The data were presented as mean ± SEM and analyzed using an unpaired two-sided Student's t-test or Mann-Whitney test accordingly. Statistical comparisons between paclitaxel treated mice and paclitaxel-NSC23925 combination treated mice are presented: * P < 0.05, ** P < 0.01, *** P < 0.001. (c) Densitometric quantification of the expression level of Cyclin D and Cytochrome C.

Toxicity study of NSC23925 in vivo

Although Pgp inhibitors, PSC833 and VX-710, were previously identified to impede drug resistance in vitro, the pharmacokinetic interactions with paclitaxel and/or induction of myelosuppression and nonhematologic toxicities restricts their clinical application.^23-26 Therefore, the toxicity of NSC23925 was determined after administration in vivo by evaluating body weight, blood cell count, and histology of organs in mouse models. The time course of body weight change was shown in Fig. 6a. There were no significant differences in the body weight among different groups during the course of drug treatment (P > 0.05). No significant differences were observed in RBC and WBC number between paclitaxel treated mice and paclitaxel-NSC23925 treated mice (Fig. 6b and 6c). In addition, no apparent pathological changes were detected in liver, kidney, and spleen (Fig. 6d). The representative photographs of major internal organs, including liver, kidney, spleen, and pancreas were displayed in Fig. 6e. Furthermore, mice undergoing NSC23925 or paclitaxel-NSC23925 combination displayed overall normal behavior and food intake during drug administration. On the basis of parameters evaluated, no remarkable toxicities were observed and the mice appeared to tolerate the treatments well when NSC23925 was administered either alone or in combination with paclitaxel over a long period of time.

Figure 6.

No significant toxicities were detected in mice treated long-term with NSC23925 alone or NSC23925 in combination with paclitaxel. (a) Body weight change of mice in each group. The data were presented as mean ± SEM. (b and c) RBC and WBC number in four groups. There were no significant differences observed between paclitaxel treated mice and paclitaxel-NSC23925 treated mice. (d) Basic organs from different groups were stained with H&E (original magnification, ×400). No apparent pathological changes were detected in liver, kidney, and spleen. (e) Representative photographs of liver, kidney, spleen, and pancreas from each group. S, N, P, and PN represent the group of mice treated with saline, NSC23925, paclitaxel, and combination paclitaxel-NSC23925, respectively.

Discussion

Our current study demonstrated that paclitaxel-NSC23925 combinations can prevent the development of paclitaxel resistance both in vitro and in vivo in ovarian cancer. The ability of NSC23925 to prevent the emergence of MDR was due to its function specifically inhibiting the overexpression of Pgp, as well as blocking the expression of anti-apoptotic proteins. The in vivo subcutaneous ovarian cancer xenograft model demonstrated that combination paclitaxel-NSC23925 produced a significantly greater antitumor efficacy than paclitaxel alone. Furthermore, the combination of paclitaxel and NSC23925 therapy was well tolerated at the dose and schedule tested in mice bearing ovarian cancer xenografts.

Paclitaxel is a substrate for Pgp-mediated efflux. Overexpression of Pgp is a well-characterized mechanism of paclitaxel resistance in tumor cells.^{8, 39} Beside the overexpression of Pgp, a number of other mechanisms of resistance are known to be triggered in cancer cells in response to exposure to chemotherapeutic drugs such as paclitaxel.^{40, 41} These include increased expression of anti-apoptotic proteins such as survivin, Bcl-xL, and MCL-1. More recently, cancer stem cell proteins, including CD44 and integrin β3 have also been found to be overexpressed in MDR cells.^{12, 42, 43} Overexpression of CD44 has been associated with tumor initiation, growth, cancer stem cells' specific behavior, development of drug resistance, and metastases.¹² In contrast with cells isolated from primary tumors, CD44 was highly overexpressed in metastatic ovarian cancer cells.⁴⁴ In the present study, we observed that ovarian cancer cells that developed MDR demonstrated an increased expression of Pgp and anti-apoptotic proteins, as well as cancer stem cell proteins, including CD44 and integrin β3; on the other hand, cells treated with the combination regimen of paclitaxel-NSC23925 remained chemosensitive and had no expression of Pgp or reduced expression of anti-apoptotic proteins include survivin, Bcl-xL, MCL-1, and cancer stem cell proteins CD44 and integrin β3. Co-overexpression of Pgp, anti-apoptotic proteins, and stem cell marker proteins has been observed in ovarian cancer MDR cells.^{12, 16, 39, 45} Tumor cells that co-express these proteins are due either to selection of more aggressive cells (MDR cells or cancer stem cells) or to an increase in the metastatic potential following chemotherapeutic treatment. It is likely that combination paclitaxel-NSC23925 treatment interfered with increased survival and tumor progression after drug exposure by maintaining and/or increasing genome stability. Combinations of NSC23925 with chemotherapy drug may be an efficacious treatment for advanced ovarian cancer by interfering with progression of tumor cells to a more malignant MDR condition, stem cell like states, and an invasive phenotype.

MDR inhibitors or chemosensitizing agents have usually been studied as “agents to reverse MDR” to downregulate the expression and inhibit the activity of Pgp, and improve the effectiveness of chemotherapy drugs. Previous studies have focused on MDR-reversing agents such as verapamil, cyclosporine A, valspodar (PSC833, Amdray), and biricodar (VX710, INCEL).^{1, 20, 21, 26, 31} Unfortunately, these compounds, initially developed for pharmacological uses other than MDR reversal, are relatively nonspecific and weak in potency. For most of these compounds, toxicities associated with their use at concentrations required to inhibit Pgp have precluded their clinical use. For example, the serum concentration of the verapamil that is required to produce in vitro reversal of MDR cannot be achieved in patients because of significant cardiotoxicities.^{33, 46} Valspodar and biricodar have shown significant side effects such as immunotoxicity and renal toxicity.^{6, 45, 47} The addition of biricodar to doxorubicin and vincristine therapy did not significantly increase antitumor activity or survival.^{25, 48} These agents usually fail in clinical trials when drug resistance is formed because the cancer cells have already expressed high levels of Pgp, accompanied with overexpression of anti-apoptotic proteins and downregulation of pro-apoptotic proteins. Therefore, it is important to develop more potent and less toxic MDR reversers and inhibitors for preventing or blocking the development of MDR preemptively. In our study, no significant toxicities were observed and the mice appeared to tolerate the treatment well when NSC23925 was administered either alone or in combination with paclitaxel over a long period of time, suggesting that NSC23925 may be safer in future clinical use than traditional MDR-reversing agents.

In conclusion, we demonstrated that the paclitaxel-NSC23925 combination can prevent acquired drug resistance in ovarian cancer cells in vitro. The benefit of the combination was also observed in an ovarian cancer mouse model. The mechanisms underlying the inhibiting effects of NSC23925 on MDR are correlated with Pgp regulation and the apoptosis pathway. These findings have important implications for patients with ovarian cancer, as well as other cancer types. For example, we have recently reported that the paclitaxel-NSC23925 combination was able to prevent and overcome MDR in osteosarcoma. Further research is needed for pharmacodynamic and pharmacokinetics study of the paclitaxel-NSC23925 combination. Results from those studies could have an impact on the use of NSC23925 as adjuvant in cancer treatment.

What's new?

Strategies to prevent the development of drug resistance will prolong the efficacy of chemotherapy and improve the long-term clinical outcome of ovarian cancer patients. The present study shows that combination paclitaxel-NSC23925 has the ability to prevent the emergence of paclitaxel resistance by specifically preventing the overexpression of Pgp and anti-apoptotic proteins in ovarian cancer. The combination paclitaxel-NSC23925 therapy was well tolerated at the dose and schedule tested in ovarian cancer mouse models. These findings suggest the clinical use of NSC23925 at the onset of chemotherapy may prevent the introduction of paclitaxel resistance and improve the clinical outcome of patients with ovarian cancer.

Abbreviations

FDA

Food and Drug Administration

Pgp

P-glycoprotein

ABC

ATP-binding cassette

MDR

multidrug resistance

COX-2

type 2 cyclooxygenase

ATCC

American Type Culture Collection

WBC

white blood cell

RBC

red blood cell