Targeting p53-Null Neuroblastomas through RLIP76

Neuroblastomas are the most common solid tumors in infants and represent the third most common class of pediatric malignancies (1). Neuroblastomas arise from the neural crest tissue of the sympathetic ganglia. Neuroblastomas can metastasize to the liver where the metastatic tumors can be massive (2). The most common metastatic site is the skull where it metastasizes to sphenoid bone and retro bulbar tissue leading to common symptoms of proptosis and periorbital echymosis (3, 4). Surgical resection of localized neuroblastomas offers good clinical remission but advanced stages of neuroblastomas require aggressive multimodality treatment (1, 5, 6). The other treatment modalities currently being investigated in refractory and relapsed neuroblastomas include antiangiogenic drugs and fenretinide, a retinoid derivative (7, 8). Given the limited 3-year event-free survival of 15% in the incident population and the nature of symptoms in affected children, validation of novel drug targets becomes a vital focus of translational research to achieve complete therapeutic response in multidrug-resistant neuroblastomas (9–11).

The tumor suppressor p53 plays a significant role in multidrug resistance of neuroblastomas. Though the p53 mutations are relatively rare in primary neuroblastomas, advanced neuroblastomas acquire loss-of-function mutations in p53, which render them resistant to conventional chemotherapeutic regimens (12–14). It is an established fact that p53 mediates the induction of apoptosis in cells treated with chemotherapeutic drugs and loss-of-function mutations in p53 are associated with resistance to both anticancer drugs and radiation therapy (15, 16). The use of chemotherapeutic drugs such as cisplatin (CDDP), doxorubicin (DOX), and vincristine is highly limited in refractory and relapsed neuroblastomas due to frequent loss-of-function mutations in the tumor suppressor p53 (17, 18).

The human ral-binding protein 1(RALBP1 or RLIP76) is a 76 kDa multifunctional drug transport protein with substrate-stimulated ATPase activity and overexpressed in majority of cancers such as cancers of the lung, colon, kidney, melanoma, ovary, prostate, and breast (19, 20). Normally, RLIP76 functions as an antiapoptotic and stress-protective protein that prevents accumulation of oxidative-stress–induced toxic metabolites inside the cells. RLIP76 acts by catalyzing the ATP-dependent efflux of glutathione–electrophile conjugates (GS-E) of lipid peroxidation products (19). In the previous studies, we have shown that the inhibition of RLIP76 by antisense-induced selective apoptosis in several cell lines of lung, prostate, kidney, colon, and breast cancers (19, 20). The hallmark of inhibition of RLIP76 in our previous tumor models of various cancers has been selective toxicity toward targeted tumor cells sparing normal cells.

Until today, the investigations into the mechanisms of p53-induced apoptosis in neuroblastoma have been focused on the inhibition of cell cycle checkpoints on DNA damage and induction of the transcription of apoptotic proteins such as bax, bak, and PUMA on cytotoxic stimuli to cells (21). The nonnuclear and transcription-independent mechanisms of p53-mediated cell death include stimulation of caspase 3 activation by mitochondrial translocation of p53 on exposure to radiation (22). In this study, we investigated the interaction of p53 on the multidrug transporter RLIP76 followed by the analysis of the impact of RLIP76 inhibition in targeting multidrug-resistant neuroblastoma. We used 2 p53 wild-type (SMS-KCNR and LAN-5) and 2 p53-mutant (CHLA-90 and SKN-BE2) neuroblastoma cell lines for our studies. CHLA-90 is a neuroblastoma cell line derived at progressive stage of neuroblastoma following multiagent chemotherapy and SKN-BE2 is derived from recurrent neuroblastoma. Neuroblastomas established as cell lines following acquisition of drug resistance during multidrug therapy and maintained without exposure to drugs during regular cultures still retain the properties of specific drug resistance that was acquired during initial multidrug chemotherapy (23). Thus, the neuroblastoma cell lines with mutant p53 were relevant to specifically study the loss-of-p53–induced drug resistance in neuroblastoma.

Doxorubicin was obtained from Adria Laboratories. Cisplatin was obtained from Bristol–Meyers Squibb. Source of polyclonal rabbit anti-human rec-RLIP76 antibodies was same as previously described (24). P53 antibodies were purchased from Santa Cruz Biotechnology. Maxfect transfection reagent was from MoleculA. PKCα was obtained from Molecular Probes. 4-Hydroxynonenal (4HNE) and malondialdehyde (MDA) measurements kit was purchased from Oxis International. Gene-specific primers, scrambled as well as RLIP76 antisense were purchased from Biosynthesis Inc. Radiolabeled ³H-GSHNE (specific activity, 3.1 × 10⁴ cpm/nmol) was synthesized as described by us previously (25). ¹⁴C-DOX (specific activity, 54 mCi/mmol) was purchased from Amersham Corporation.

CHLA-90 and SKN-BE2 (p53-mutant) cell lines have been established from tumors obtained from patients who had relapsed after intensive multiagent myeloablative chemoradiotherapy and bone-marrow transplantation. SMS-KCNR and LAN-5 (p53 wild-type) cell lines have been established from tumors obtained from neuroblastoma patients. Human neuroblastoma cell lines were kindly authenticated and provided in May 2010 by Dr. Patrick Reynolds, MD PhD, Children's Neuroblastoma Cancer Foundation/Children Oncology group, Texas Tech University, Department of Cell Biology & Biochemistry, School of Medicine, Lubbock, TX. The cells were tested for Mycoplasma once at the beginning immediately after receiving the cell lines and after 3 months. All the experiments were conducted within 5 months of securing the cell lines. Cells were grown in complete medium consisting of Iscove's modified Dulbecco's medium supplemented with 3 mmol/L l-glutamine, 5 μg/mL each of insulin and transferrin, 5 ng/mL of selenous acid (ITS culture supplement), and 20% FBS at 37°C in a humidified 5% CO₂ atmosphere. Cell lines were subcultured by detaching without trypsin from culture plates by using a modified Puck's Solution A plus EDTA (Puck's EDTA) which contain 140 mmol/L NaCl, 5 mmol/L KCl, 5.5 mmol/L glucose, 4 mmol/L NaHCO₃, 0.8 mmol/L EDTA, 13 μmol/L phenol red, and 9 mmol/L HEPES buffer (pH 7.3).

The protein–protein cross-linking and immunoprecipitation assay was carried out according to the method of Aumais and colleagues (26). Briefly, purified rec-RLIP76 and p53 (10 μg each) were cross-linked by incubation with 0.1 mmol/L N-succinimidyl 3-(2-pyridyldithio) propionate (SPDP; Sigma) in a total volume of 0.5 mL in 10 mmol/L sodium phosphate buffer, pH 7.4, for 30 minutes. Excess SPDP was removed by passing the solution through a Sephadex G-50 spin column pre-equilibrated with sodium phosphate buffer (pH 7.4). The samples were treated with 0.5 mmol/L N-ethylmaleimide for 10 minutes to block all free SH groups that could prematurely cleave cross-links. Protein complexes were immune-precipitated by incubating with either preimmune immunoglobulin G (IgG) or anti-RLIP76 IgG for 12 hours followed by with protein A-Sepharose (20 μL of 50% bead slurry) in radioimmunoprecipitation assay buffer [50 mmol/L Tris-HCl (pH 7.4), 4 mmol/L EDTA, 150 mmol/L NaCl, 1% Triton X-100, and 0.1% SDS] for 2 hours. Samples were sedimented by centrifugation at 10,000 × g, washed 3 times with radioimmunoprecipitation assay buffer, and then resuspended in 100 μL of SDS-PAGE sample buffer. To check the effect of PKCα in the interaction of RLIP76-p53, the immunoprecipitation was carried out in the presence of equimolar concentration of PKCα in the absence and presence of 1 mmol/L ATP. The reaction mixture was incubated for 30 minutes at 37°C and the cross-linking and immunoprecipitation was carried out as described earlier. Samples were analyzed by SDS-PAGE and Western blotting against anti-RLIP76, anti-PKCα, and anti-p53 IgG.

Measurement of endogenous levels of 4HNE and MDA in a panel of neuroblastoma cells were carried out spectrophotometrically by using the LPO 586 measurement kit (Oxis International) according to manufacturer's instructions.

Cell density measurements were done by a hemacytometer to count dye-excluding cells resistant to staining with trypan blue. Approximately 2 × 10⁴ cells were plated into each well of a 96-well flat-bottomed microtiter plate 24 hours prior to addition of medium containing varying concentrations of antibodies or drugs. After 24 hours of incubation, 40 μL aliquots of drugs (DOX and CDDP) diluted in medium were added to get final concentration between 0.001 and 100 μmol/L to 8 replicate wells. After 96 hours of incubation, 20 μL (5mg/mL stock) MTT was added to each well and incubated for 2 hours at 37°C. The plates were centrifuged and medium was removed. Formazan dye trapped in cells was dissolved by addition of 100 μL dimethylsulfoxide with gentle shaking for 2 hours at room temperature, followed by measurement of absorbance at 570 nm. Depletion of RLIP76 expression in cells by RLIP76 antisense was measured as follows: cells were incubated for 6 hours with 0 to 10 μg/mL RLIP76 antisense in Maxfect transfection reagent (MoleculA) according to the manufacturer-provided protocol.

The proteoliposome preparation procedures used here have been validated and described in detail previously (24, 27). Briefly, proteoliposomes containing RLIP76 or p53 were prepared by addition of the purified recombinant protein to a sonicated emulsion of 1:4 cholesterol and phospholipids (soybean asolectin, 95% purity; Sigma Chemicals Co.) in liposome reconstitution buffer [10 mmol/L Tris-HCl (pH 7.4), 2 mmol/L MgCl₂, 1 mmol/L EGTA, 100 mmol/L KCl, 40 mmol/L sucrose, 2.8 mmol/L BME, 0.05 mmol/L BHT, and 0.025% polidocanol]. Liposome formation was initiated by addition of SM2 Biobeads (200 mg/mL). We have shown that vesiculation is complete within 4 hours and yields primarily unilammelar vesicles with median diameter of 0.25 μm and intravesicular/extravesicular volume ratio of 18 μL/mL (24). The control liposomes or the liposomes reconstituted with purified RLIP76 were used for transfecting cells in culture. Efficiency of delivery for RLIP76 proteoliposomes has been established previously (24).

A total of 2.5 × 10³ CHLA-90 and SMS-KCNR cells were treated with control liposome and RLIP76 proteoliposomes (50 μg/mL final conc.) for 24 hours prior to radiation at 100 to 500 cGy by using a Varian Clinac linear accelerator (2100 C; 6-MeV photon beams) at the Texas Cancer Center, Arlington, TX. Cells were inoculated into colony-forming assays immediately after radiation. After 7 days, colonies were stained with methylene blue and counted by using an image acquisition and analysis system (Innotech Alpha Imager HP). The results presented are the mean and SD from 3 separate experiments. Please note that CHLA-90 (6,255 colonies/mL); and SMS-KCNR (6,620 colonies/mL); plating efficiency of CHLA-90 and SMS-KCNR are 1 and 1.06, respectively. After RLIP76 proteoliposomes treatment; CHLA-90 (7,130 colonies/mL); and SMS-KCNR (7,035 colonies/mL); plating efficiency of CHLA-90 and SMS-KCNR after RLIP76 proteoliposomes treatment were 1.14 and 1.06, respectively.

Cells were pelleted and washed with balanced salt solution. Washed cells were lysed and sonicated for 30 seconds at 50 W and incubated for 4 hours at 4°C with occasional shaking. After incubation, the resultant preparation was centrifuged at 105,000 × g for 60 minutes at 4°C. The supernatant was collected and aliquots of crude fraction were applied to SDS-PAGE and Western blot analyses were carried out. Western blots were developed by enhanced chemiluminescence reagent (Pierce).

Colocalization of RLIP76 and p53 was carried out on neuroblastoma, LAN-5 (wild-type p53), and SKN-BE2 (mutant p53) fixed cells by method described previously with slight modifications (28). Briefly, cells were grown on glass coverslips and fixed with methanol and acetic acid (3:1). Nonspecific antibody interactions were minimized by pretreating the cells with 10% goat serum in TBS for 60 minutes at room temperature. The cells were subjected to immunocytochemistry by using anti-RLIP76 IgG (raised in rabbit) and anti-p53 IgG (raised in mice) as a primary antibody, and goat anti-rabbit rhodamine red-x conjugated (for RLIP76) or goat anti-mouse fluorescein isothiocyanate (FITC) conjugated (for p53) as secondary antibody. DAPI (4′,6-diamidino-2-phenylindole) was used as a nuclear counterstain. Slides were analyzed by confocal laser microscope (Leica TCS-SP5).

Depletion of RLIP76 expression by RLIP76 antisense (2–10 μg/mL) was done by using Maxfect Transfection Reagent according to manufacturer's protocol. Briefly, cells were incubated for 6 hours with 2 to 10 μg/mL RLIP76 antisense, washed with PBS, followed by 48 hours of incubation at 37°C in medium. For inhibition of RLIP76, anti-RLIP76 IgG at a concentration of 10 to 20 μg/mL was used.

For these experiments, fixed amount of purified rec-RLIP76 (250 ng) was reconstituted into proteoliposomes along with varying amounts (0–250 ng) of purified rec-p53, in the absence/presence of 12.5 ng of PKCα, 1 mmol/L ATP, and were incubated for 30 minutes at 37°C. Effect of PKCα-mediated phosphorylation on RLIP76-mediated ¹⁴C-DOX and ³H-GSHNE transports were measured as described (27, 29). In one control, p53 proteins were excluded whereas in another control equivalent amount of bovine serum albumin were reconstituted in proteoliposomes. Each determination was carried out in triplicate.

Hsd:athymic nude nu/nu mice were obtained from Harlan. All animal experiments were carried out in accordance with a protocol approved by the Institutional Animal Care and Use Committee. Twenty 11-week-old male mice were divided into 4 groups of 5 animals (treated with preimmune serum, scrambled antisense DNA, anti-RLIP76 IgG, and RLIP76 antisense). All 20 animals were injected with 2 × 10⁶ neuroblastoma cell (SMS-KCNR) suspensions in 100 μL of PBS, subcutaneously into 1 flank of each nu/nu nude mouse. Animals were examined daily for signs of tumor growth. Treatment was administered when the tumor surface area exceeded 40 mm² (∼28 days). Treatment consisted of 200 μg of anti-RLIP76 IgG or antisense in 100 μL PBS, intraperitoneally (i.p.). Control groups were treated with 200 μg/100 μL preimmune serum or scrambled antisense DNA. Tumors were measured in 2 dimensions by using calipers.

All experiments were conducted in triplicates. The experimental data was statistically analyzed by 2-tailed unpaired Student's t test or by 1-way ANOVA and are expressed as the mean ± SD. A value of P < 0.05 was considered statistically significant.

We first compared the neuroblastoma cells expressing wild-type p53 and mutant p53 for multidrug resistance on treatment with chemotherapeutic drugs, DOX and CDDP. The CHLA-90 and SKN-BE2 cells with mutant p53 were more resistant for all the tested drugs than the SMS-KCNR and LAN-5 cells with wild-type p53. The IC₅₀ values for DOX were 0.60 ± 0.1 μmol/L and 0.88 ± 0.1 μmol/L in CHLA-90 and SKN-BE2 p53-mutant cells, respectively, which were significantly higher by an order of magnitude of approximately 35-fold when compared with p53 wild-type neuroblastomas (IC₅₀: 0.02 ± 0.0 μmol/L). The IC₅₀ values for CDDP were 30 ± 4 μmol/L and 22 ± 2 μmol/L in CHLA-90 and SKN-BE2 p53-mutant cells, respectively, which were significantly higher by order of magnitude of 3- to 7-fold when compared with p53 wild-type neuroblastomas (IC₅₀: 4 ± 1 μmol/L and 10 ± 1 in SMS-KCNR and LAN-5 cells, respectively; Table 1). P53 is one of the key regulators of antioxidant network of the cells and the interplay between reactive oxygen species (ROS) and p53 is complex (30). Wild-type p53 in response to stress and DNA damage in normal cells activates pro-oxidant genes leading to increased ROS and apoptosis of damaged cells. Wild-type p53 in nonstressed cells stimulates the transcription of antioxidant genes leading to reduced ROS and thus protects the ROS-induced mutations in genes (30). Thus, the neuroblastoma cells with mutant p53 in regular cultures should possibly differ in the oxidative stress. The level of end products of lipid peroxidation of 4HNE and MDA were substantially higher in p53-mutant neuroblastoma cells compared with wild-type p53 expressing neuroblastoma cells (Fig. 1A; P < 0.001). But, interestingly, the p53-mutant neuroblastomas with higher and toxic oxidative-stress markers such as 4HNE and MDA were resistant to apoptosis induction by chemotherapy drugs relative to neuroblastomas with wild-type p53. Thus, in addition to lack of p53 function in multidrug-resistant neuroblastomas, it indicated a possibility for upregulation of cellular defense against toxic end products of ROS. Hence, we studied the expression of RLIP76, a multispecific transporter of GS-E of lipid peroxidation and chemotherapy drugs. We observed that the expression of RLIP76 was significantly higher in p53-mutant as compared with p53 wild-type neuroblastomas (Fig. 1B). These findings indicate that the p53-mutant neuroblastomas might be dependent on the function of RLIP76 to buffer the enhanced oxidative stress. RLIP76 regulates the cellular concentration of products of lipid peroxidation. 4HNE, a marker of oxidative stress, has concentration-dependent effects in cells. Moderate levels of 4HNE stimulate proliferation whereas very high levels of 4HNE are toxic to even transformed and relatively apoptosis-resistant cancer cells (31). We observed that p53-mutant neuroblastoma cells have higher levels of 4HNE in spite of higher expression of the GSHNE transporter RLIP76. This could be attributed to activation of cyp2E1 in p53-mutant neuroblastomas, which enhances the generation of 4HNE from ω-6 fatty acids (32, 33). Thus, p53-mutant neuroblastomas have higher 4HNE relative to p53 wild-type cells, but they may not be able to reach very toxic concentrations due to RLIP76 overexpression. Thus, the “moderately high levels of 4HNE” attained due to a fine calibration between cyp2E1-mediated enhanced generation of 4HNE and RLIP76-mediated efflux could have stimulatory effect on cellular proliferative pathways.

To further confirm the protective effects of RLIP76 in neuroblastomas, we supplemented 1 p53-mutant (CHLA-90) and 1 p53 wild-type (SMS-KCNR) neuroblastoma cells with RLIP76 by liposomal delivery and tested the radiation sensitivity in relation to respective cell lines transfected with control liposomes. RLIP76 delivery protected both the cell lines from toxic oxidative stress induced on radiation exposure (Fig. 1C).

As RLIP76 was increased in p53-mutant neuroblastoma cell lines, we next investigated the functional status of RLIP76 in the presence or absence of p53. In this regard, we first analyzed whether p53 interacts with RLIP76 in the presence of equimolar concentrations of RLIP76 and PKCα, a known activator of RLIP76. Our previous studies have indicated that PKCα phosphorylates RLIP76 on S¹¹⁸, T²⁹⁷, S³⁵³, and S⁵⁰⁹ (29). Hence, we wanted to study whether the PKCα-mediated phosphorylations have any impact on the binding assay. Immunoprecipitation studies revealed that RLIP76 binds with p53 both in the presence and absence of PKCα and that any phosphorylation by PKCα is not essential for the binding of p53 to RLIP76. But, the binding of p53 to RLIP76 was enhanced in the presence of PKCα and ATP (Fig. 2A, lane 5). Next, we analyzed the binding of p53 to RLIP76 in SKN-BE2 cells expressing mutant p53 and LAN-5 cells expressing wild-type p53. Immunocytochemistry for the binding of RLIP76 and p53 revealed enhanced p53 binding to RLIP76 in LAN-5 neuroblastoma cells relative to p53-mutant SKN-BE2 cells as revealed by an yellow overlay fluorescence when stained with fluorescent antibodies for RLIP76 (red, rhodamine) and p53 (green, FITC; Fig. 2B). Thus, the confirmation of enhanced binding of wild-type 53 over mutant p53 to RLIP76 further lead to investigation of the effect of p53 binding on RLIP76 transport function.

The effect of p53 on the transport function of RLIP76 was analyzed in reconstituted liposomes treated with increasing concentrations of p53. P53 caused a concentration-dependent decrease in the transport activity of RLIP76. Both the transport of glutathione conjugates of 4HNE (GSHNE) and DOX was decreased in the presence of p53 (Fig. 3A and B). In the immunoprecipitation studies, we noted an enhanced binding of p53 to RLIP76 in the presence of PKCα (Fig. 2A). Hence, we further analyzed the impact of PKCα on p53-mediated inhibition of RLIP76 drug transport. Cotreatment with PKCα significantly enhanced the p53-induced inhibition of RLIP76-mediated drug transport in liposomes. In accordance with previously published studies, PKCα in the presence of 1 mmol/L ATP activated the drug transport by RLIP76. P53 caused significant decline in both the basal and PKCα-stimulated transport function of RLIP76 (Fig. 3C). Also, p53 caused maximum inhibition of RLIP76 transport in the absence of PKCα, but not in the presence of PKCα. Thus, though PKCα stimulates the binding of p53 to RLIP76; functionally, it partially reverses the p53-induced inhibition of RLIP76 transport. Thus, PKCα and p53 have opposite effects on the function of RLIP76 with either of them being able to partially reverse the effect of other. This is significant because previous studies by others have established that malignant neuroblastoma cells express a high level of PKCα along with nMyc and treatment with retinoic acid, which induces maturation of neuroblastoma cells to normal neuronal phenotype causes a decline of PKCα levels (34). Thus, the loss-of-p53 tilts the balance toward enhanced multidrug resistance that is mediated by RLIP76 both due to overexpression of RLIP76 and uninhibited stimulation of drug transport by RLIP76 by activators such as PKCα in advanced neuroblastomas.

We further investigated the effect of inhibition of RLIP76 by using antibodies and antisense. Depletion of RLIP76 by antisense increased the cell death in neuroblastoma cells (Fig. 4A). RLIP76 inhibition substantially enhanced the sensitivity to CDDP in both p53-mutant neuroblastoma cell lines and those expressing wild-type p53 (Fig. 4B; P < 0.01). The marginally enhanced susceptibility of p53 wild-type SMS-KCNR and LAN-5 neuroblastoma cells to RLIP76 antisense–induced cell death and potentiation of CDDP toxicity may be due to the activation of alternate p53-dependent apoptotic pathways consequent to RLIP76 depletion-induced oxidative stress and enhanced intracellular drug concentrations in a p53 wild-type background. But, the substantial cell death in p53-mutant neuroblastoma cell lines reflects the ability of RLIP76 depletion alone to induce cell death, which, along with the RLIP76 inhibition–induced increase in drug sensitivity in p53-null background, is of specific relevance in the control of aggressive and drug-resistant p53-null neuroblastomas.

The novel finding that RLIP76 depletion alone using antisense was effective in targeting neuroblastomas lead to further investigations. Both the p53-mutant CHLA-90 and SKN-BE2 cell lines did not form tumors in mice xenografts. Hence, we tested the impact of RLIP76 inhibition on mice xenografts of SMS-KCNR cells. A total of 2 × 10⁶ neuroblastoma cells were injected in 100 μL PBS into subcutaneous flank of each nu/nu mice and tumors were allowed to develop. Approximately at day 30, when the tumors reached approximately 40 mm² the mice were treated with 200 μg of RLIP76 antibodies or antisense in 100 μL of PBS, and tumor growth was observed everyday by measuring the tumor cross-sectional area by using calipers. The RLIP76 inhibition alone, either by using antibody or antisense caused significant tumor regression (Fig. 5A). The average weight of tumors of treated mice was substantially less than that of the controls (tumor weight at day 44: control, 1.88 g; RLIP76 antibody or antisense treated, 0.39 g; Fig. 5B and C; P < 0.001). There was no loss in the weight of treated mice when compared with controls, which ruled out any dose-limiting overt off-target cytotoxicity consequent to RLIP76 inhibition. This confirmed the ability of RLIP76 inhibition to successfully induce neuroblastoma tumor regression. Though, p53-mutant neuroblastomas did not develop tumors in mice xenograft studies, the significant efficacy of RLIP76 inhibition or depletion in p53-normal neuroblastoma was a striking finding for the ability to target RLIP76 alone in neuroblastomas. Taken together, our mice xenograft studies in p53-normal neuroblastomas along with the preceding drug transport studies where p53 inhibits RLIP76 and cell survival studies where RLIP76 is effective to a greater extent in p53-mutant neuroblastoma cell lines, the current research findings provide strong mechanistic and biological rationale for targeting RLIP76 in both p53-normal and p53-mutant neuroblastomas to achieve effective therapeutic response.

Neuroblastomas are the most aggressive form of neuroblastic tumors with ganglioneuroblastoma and ganglioneuroma being the other succeeding aggressive tumors originating from the neural crest cells. Neuroblastomas have a median age of onset at 23 months and cause approximately 15% of the total cancer deaths in pediatric population. The neuroblastomas diagnosed before 18 months have good prognosis and many cases show spontaneous regression even without any clinical interventions. But, the survival of neuroblastomas incident after 18 months of age is poor at 42% (35). The neuroblastomas that occur in adolescents represent immense challenge to current interventions and are highly fatal (36). Approximately 60% of neuroblastomas present with metastases at different organs such as liver, bone-marrow, and cortical bones at the time of presentation, which limits localized surgical approaches and necessitates aggressive and effective chemotherapeutic interventions to achieve complete cure (37). Thus, targeting drug-resistant and relapsed tumors represents a contemporary focus of management in neuroblastomas.

Amplification of nMyc is an established cause in primary neuroblastomas, but nMyc is positively associated with aggressive course of disease in children ages more than 1 year at the time of diagnosis but not in infants (38–40). Treatment of neuroblastoma cells with cyclophosphamide induces apoptotic response by activating p53-dependent transcription of apoptotic genes such as PUMA and activation of proapoptotic proteins bax, bim, caspase 3, and caspase 9. P53-mediated apoptotic pathway is frequently altered in multidrug-resistant and relapsed neuroblastomas due to several causative factors (41). The nMyc amplification leads to attenuation of p53 function by stimulating the overexpression of p53 inhibitor Mdm2 (42).

Loss-of-function mutations in p53 lead to overexpression and uninhibited action of RLIP76 which contributes to chemoresistance. The p53-mutant neuroblastomas have enhanced levels of oxidative stress as revealed by increased concentration of 4HNE and MDA. Previous studies from our laboratory have shown that increased oxidative stress stimulates the expression of RLIP76 in K562 cells (43). Thus, p53-mutant neuroblastomas could be activating the transcription of RLIP76 as a stress-responsive protein to defend against toxic effects of enhanced oxidative stress. It is also possible that p53 could have a direct transcriptional role in regulating the expression of RLIP76. In this regard, the exact mechanisms that contribute to overexpression of RLIP76 in p53-mutant neuroblastomas need to be investigated further.

In addition to enhancing the chemosensitivity of coadministered drugs such as CDDP, RLIP76 alone represents a highly significant therapeutic target in multidrug-resistant neuroblastomas. Inhibition of RLIP76 alone induced significant cell death, which was more pronounced in p53-mutant neuroblastomas. Overexpression of RLIP76 in p53-mutant cancers itself could be a major pathogenetic factor for aggressive and metastatic neuroblastomas. In addition to its fundamental role in efflux of GS-E of lipid peroxidation products and chemotherapy drugs, RLIP76 regulates a spectrum of signaling events in cancer cells that are of importance in proliferation and metastases (Fig. 6). CDK1 binds to RLIP76 and translocates it to nucleus where it mediates the separation of chromosomal spindles during mitosis (44). PKCα-mediated proliferation and drug resistance require RLIP76 (29). RLIP76, through its RhoGAP domain, directly binds to and mediates adhesion-dependent activation of the GTPase R-Ras, leading to enhanced cell motility that was mediated by downstream Rac activation (45). RLIP76 is also negatively regulated by previously known inhibitors such as HSF-1, POB-1, and CDC2 (46, 47). P53, as evident from our current studies, is a novel inhibitor and arguably the most significant inhibitor in the context of multidrug resistance associated with loss-of-p53 in multiple malignancies. Most importantly, RLIP76 inhibition in our previous tumor models of mice did not reveal any deficits in the CNS functions and the experimental mice were active and normal without any loss of weight (48–50).

In summary, RLIP76 is a potential target for therapeutic interventions in refractory and relapsed neuroblastomas. In addition to well-defined nuclear and mitochondrial mechanisms of initiating p53-mediated apoptosis, for the first time, our results have established a novel “cell membrane” mechanism, in which active p53 inhibits the multidrug transporter RLIP76 to initiate a cascade of cell-death events that are mediated by an increase in intracellular concentration of ROS to toxic levels along with enhancing the intracellular concentration of administered chemotherapy drugs. The observed absence of off-target cytotoxicity as evident from our previous studies in RLIP76-targeted mice further corroborates the relevance of RLIP76-targeted chemotherapy for pediatric neuroblastomas, as such intervention would not compromise the normal growth and development of treated children. Thus, RLIP76 represents a novel and effective therapeutic target to treat multidrug-resistant neuroblastomas.

No potential conflicts of interest were disclosed.

We thank Dr. Xiangle Sun, core facility at the University of North Texas Health Science Center, Fort Worth, TX, for helping with flow cytometry and laser capture microdissection (supported by NIH Grant ISIORR018999-01A1). We also thank Dr. Sumihiro Suzuki, Department of Biostatistics, School of Public Health, University of North Texas Health Science Center, Fort Worth, TX, for his assistance in the statistical analyses of the data.

This work was supported in part by USPHS grant CA 77495 (S. Awasthi and S.S. Singhal), the Cancer Research Foundation of North Texas (S.S. Singhal and S. Yadav) and Institute for Cancer Research and the Joe & Jessie Crump Fund for Medical Education (S.S. Singhal).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.