Sequential dosing in chemosensitization: targeting the PI3K/Akt/mTOR pathway in neuroblastoma

Abstract

Breaking resistance to chemotherapy is a major goal of combination therapy in many tumors, including advanced neuroblastoma. We recently demonstrated that increased activity of the PI3K/Akt network is associated with poor prognosis, thus providing an ideal target for chemosensitization. Here we show that targeted therapy using the PI3K/mTOR inhibitor NVP-BEZ235 significantly enhances doxorubicin-induced apoptosis in neuroblastoma cells. Importantly, this increase in apoptosis was dependent on scheduling: while pretreatment with the inhibitor reduced doxorubicin-induced apoptosis, the sensitizing effect in co-treatment could further be increased by delayed addition of the inhibitor post chemotherapy. Desensitization for doxorubicin-induced apoptosis seemed to be mediated by a combination of cell cycle-arrest and autophagy induction, whereas sensitization was found to occur at the level of mitochondria within one hour of NVP-BEZ235 posttreatment, leading to a rapid loss of mitochondrial membrane potential with subsequent cytochrome c release and caspase-3 activation. Within the relevant time span we observed marked alterations in a ∼30 kDa protein associated with mitochondrial proteins and identified it as VDAC1/Porin protein, an integral part of the mitochondrial permeability transition pore complex. VDAC1 is negatively regulated by the PI3K/Akt pathway via GSK3β and inhibition of GSK3β, which is activated when Akt is blocked, ablated the sensitizing effect of NVP-BEZ235 posttreatment. Our findings show that cancer cells can be sensitized for chemotherapy induced cell death - at least in part - by NVP-BEZ235-mediated modulation of VDAC1. More generally, we show data that suggest that sequential dosing, in particular when multiple inhibitors of a single pathway are used in the optimal sequence, has important implications for the general design of combination therapies involving molecular targeted approaches towards the PI3K/Akt/mTOR signaling network.

Figure 1. The effects of the PI3K/mTOR inhibitor NVP-BEZ235 on SHEP NB cells.

A Cells were either left untreated, treated for 24 µM PI-103, a well-characterized pan-PI3K inhibitor used as positive control, or the indicated concentrations of NVP-BEZ235. Protein expression levels and phosphorylation status of Akt and S6 ribosomal protein served as surrogate read-outs for PI3K and mTOR activity, respectively, and were analyzed by Western blotting, β-actin served as loading control. B Cells were either left untreated, treated for 24 hrs with either 0.6 µM PI-103 as positive control, or 0.6 µM NVP-BEZ235 for the indicated lengths of time. Protein expression levels and phosphorylation status of Akt and S6 ribosomal protein were analyzed by Western blotting, β-actin served as loading control. C Cells were either left untreated, treated for indicated length of time with 0.6 µM NVP-BEZ235, or treated for 24 hrs with 0.2 µg/ml doxorubicin as positive control. Protein expression levels and cleavage of caspase-.3 protein were analyzed by Western blotting, β-actin served as loading control. D Cells were cultured either in the presence or absence of 0.6 µM of NVP-BEZ235 for 24, 48 and 72 hrs, followed by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei. The percentage of absolute DNA fragmentation is shown as readout of apoptosis. E 24, 48 and 72 hrs after treatment with 0.6 µM NVP-BEZ235 total cell numbers of treated and untreated cells were counted. F Cell cycle distribution (untreated control and samples treated with 0.6 µM of NVP-BEZ235) was determined after indicated times by FACS analysis of propidium iodide-stained nuclei. G Either untreated controls, or cells treated for 12 and 24 hrs with 0.6 µM NVP-BEZ235 were stained for Ki67 protein expression and evaluated by immunofluorescent microscopy. In A–C and F a representative result of two independent experiments is depicted, while in D and E mean+s.e.m. values of at least three independent experiments carried out in triplicate are shown. Shown in F is the mean of three independent experiments carried out in triplicate, in G the mean+SD of three independent experiments. Statistical analysis was carried out by two-sided Student's t-test; * P-value <0.01; ** P-value <0.001; # P-value <0.0001.

Figure 2. Altered timing affects the potency of NVP-BEZ235/doxorubicin combination therapy in SHEP NB cells.

Three different treatment combinations were tested on SHEP NB cells, giving NVP-BEZ235 12 hrs prior to doxorubicin (Pre), giving both substances concurrently (Co), or giving NVP-BEZ235 12 hrs after the chemotherapeutic (Post). Importantly, the maximal incubation time with doxorubicin was kept constant at 24 hrs (earlier time points also shown in C and D). A SHEP NB cells were treated with NVP-BEZ235 and indicated concentrations of doxorubicin for 24 hrs, according to the scheme outlined above. Apoptosis was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei, and percentage of specific DNA fragmentation is shown. B An alternative depiction of the data presented in A, highlighting the difference between the three NVP-BEZ235/doxorubicin combinations. For all following experiments 0.2 µg/ml doxorubicin was used. C Cells were either left untreated or treated as indicated and mitochondrial release of immunofluorescent-labeled cytochrome c was determined by FACS analysis. D Cells were either left untreated or treated as indicated. A Western blot analysis of caspase-3 processing served as surrogate read-out of caspase activation (appearance of the ∼12 kD cleavage fragment), β-actin was used as loading control. In A and B mean+s.e.m. values of three independent experiments carried out in triplicate, in C mean+s.d. of three independent experiments are shown, while in D a representative result of three independent experiments is depicted. Statistical analysis was carried out by two-sided Student's t-test; * P-value <0.01; ** P-value <0.001; # P-value <0.0001.

Figure 3. The superiority of posttreatment with NVP-BEZ235 is not restricted to one NB cell line and doxorubicin.

A SH-SY5Y NB cells were treated as indicated by the scheme and apoptosis was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei. B Kelly NB cells were treated as indicated by the scheme and apoptosis was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei. C SHEP NB cells were treated as indicated by the scheme, substituting doxorubicin with either 1.0 µg/ml Cisplatin (Cis), 0.03 µg/ml Topotecan (Topo) or 5.0 µg/ml Etoposide (VP16). Apoptosis was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei. D D54 glioblastoma cells were treated as indicated by the scheme and apoptosis after doxorubicin (0.3 µg/ml doxorubicin) treatment was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei. In A to D mean+s.e.m. values of three independent experiments carried out in triplicate are shown. Statistical analysis was carried out by two-sided Student's t-test; * P-value <0.01; ** P-value <0.001; # P-value <0.0001.

Figure 4. Apoptosis sensitization occurs at the mitochondrial level.

A SHEP NB cells were either left untreated (control) or treated as indicated, followed by a Western blot analysis of the caspase-3 processing kinetic, with β-actin as loading control. B The loss of mitochondrial membrane potential (MMP) was analyzed after indicated treatment. Cells were incubated with TMRM dye prior to FACS analysis. C Cells were either left untreated or treated as indicated. The DNA damage was assayed by single cell gel electrophoresis (Comet) assay and expressed as Mean Olive Tail Moment. D Cells were treated for the indicated length of time with 0.6 µM NVP-BEZ235, 0.2 µg/ml doxorubicin, 20 nM Bafilomycin A1 (a inhibitor of the late stages of autophagy that blocks fusion between autophagosomes and lysosomes), or combinations thereof. The percentage of autophagic cells was then determined by counting cells with LC3 foci. In A representative results of two independent experiments are shown, in B mean+s.e.m. of three independent experiments carried out in triplicate are shown, while in C mean+s.d. of two independent experiments are depicted. In D mean+s.d. Of three independent experiments are depicted. Statistical analysis was carried out by two-sided Student's t-test; * P-value <0.01; ** P-value <0.001; # P-value <0.0001.

Figure 5. Sensitization for doxorubicin-induced apoptosis via posttreatment with NVP-BEZ235 is mediated via VDAC1.

A SHEP NB cells were treated for 12.5-BEZ235 for the last 0.5 hr. Either Bim, Bax or Bad was then immunoprecipitated and interaction partners that are phosphorylated on Serine or Threonine were visualized by Western blot analysis. A ∼30 kD protein, the presence of which appears to depend on NVP-BEZ235 addition, was identified as VDAC by VDAC1/Porin-specific antibody. IgG_H – heavy chain. B Cells were left untreated, treated for 12.5 hrs with Doxorubicin, or after 12 hrs for 0.5 hr with NVP-BEZ235, or a combination of both (first 12 hrs with Doxorubicin alone, followed by the addition of NVP-BEZ235 for 0.5 hr). VDAC was immunoprecipitated and its phosphorylation status was probed. IgG_L – light chain. C Cells were left untreated, treated for 12.5 hrs with doxorubicin, or after 12 hrs for 0.5 hr with NVP-BEZ235, or a combination of both (first 12 hrs with doxorubicin alone, followed by the addition of NVP-BEZ235 for 0.5 hr). Protein expression levels and phosphorylation status of GSK3β were analyzed by Western blotting, GAPDH served as loading control. D Cells were treated either for 12.5 hrs with doxorubicin, or a combination of doxorubicin and NVP-BEZ235, (first 12 hrs with doxorubicin alone, followed by the addition of NVP-BEZ235 for 0.5 hr). This was followed by immunoprecipitation of GSK3β and analysis of this protein's interaction with VDAC via immunoblotting. E Cells were again treated with a combination of doxorubicin and NVP-BEZ235, (first 12 hrs with doxorubicin alone, followed by the addition of NVP-BEZ235 for 0.5 hr), during the last hour in the absence or presence of the GSK3β-specific inhibitor SB415286. This was followed by immunoprecipitation of GSK3β and analysis of this protein's interaction with VDAC. F Apoptosis in cells treated for 24 hrs with doxorubicin, for 12.5 hrs with SB415286, for 12 hrs with NVP-BEZ235, or a combination of those substances was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei, and percentage of specific DNA fragmentation is shown. Shown in A to E are representative blots of at least two independent experiments, in F the mean+s.e.m. of three independent experiments performed in triplicate is depicted. Statistical analysis was carried out by two-sided Student's t-test; * P-value <0.01; ** P-value <0.001; # P-value <0.0001.

Figure 6. The effect of several inhibitors of PI3K/mTOR signaling on SHEP NB cell survival.

A SHEP NB cells were treated with doxorubicin for 24(control), or in the presence of the indicated pharmacological inhibitor, which was given 12 hrs prior to doxorubicin (Pre), or simultaneously with doxorubicin (Co), or 12 hrs after doxorubicin (Post). Apoptosis was determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei. B Comparison of different treatment strategies, either using the pharmacolgical inhibitors as single agents or in combination. The sensitization effect is depicted as X-fold increase in cell death (as determined by FACS analysis of the DNA fragmentation of propidium iodide-stained nuclei) over treatment with doxorubicin alone. C Cells were either left untreated or treated with either NVP-BEZ235 posttreatment, or the complex combination therapy shown to work best in B, in the presence or absence of doxorubicin. Cells were treated with Nu7026 for 24.5 hrs, with doxorubicin and/or rapamycin for 12.5 hrs, 0.5 hrs with NVP-BEZ235 and allowed to grow 10 days. In A and B mean+s.e.m. values of three independent experiments carried out in triplicate are shown, in C a representative result of two independent experiments is depicted. Statistical analysis was carried out by two-sided Student's t-test; * P-value <0.01; ** P-value <0.001; # P-value <0.0001.