• HGF;
  • melanoma;
  • PI-103;
  • immunocompetent animal model


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Melanoma is the most lethal human skin cancer. If metastatic, it becomes very aggressive and resistant to standard modalities of anticancer treatment. During the last 10 years, several therapeutic strategies have been tested including the use of single and combined small drugs. Experimental results indicate that RAS and PI3K pathways are important for the development and maintenance of melanoma. In this study, we assessed the in vitro and in vivo inhibition potential of PI-103, a PI3K (p110α)/mTOR inhibitor and sorafenib, a BRAF inhibitor, as single agents and in combination in primary melanoma cell lines. Although PI-103 and sorafenib inhibited melanoma in vitro cell proliferation and viability, the inhibition of RAS pathway appeared to be more effective. The combination of the two agents in in vitro showed a synergistic effect inhibiting RAS and PI3K pathways in a cell line dependent manner. However, no cooperative effect was observed in blocking in vivo tumor growth in immunocompetent mice. In contrary to the expected, the data indicate that PI-103 induced immunosuppression promoting in vivo tumor growth and inhibiting apoptosis. Furthermore, in vitro studies examining the effects of the PI3K/mTOR inhibitor in tumor derived cell lines indicated that PI-103 induced the anti-apoptotic BH3 family proteins Mcl1, Bcl2 and BclxL favoring, the in vitro survival of sorafenib treated melanoma cells. These data certainly makes an argument for investigating unexpected effects of rational drug combinations on immunocompetent animal models prior to conducting clinical studies.

Melanoma is a common skin cancer resulting in high morbidity and mortality. Recent studies have provided a much improved understanding of melanoma biology; however, this knowledge has yet to be translated into effective treatment strategies.

Several genetic alterations in signaling molecules have been described in human melanoma.1 The INK4a-ARF locus on chromosome 9p21, (encoding p16INK4a and p14ARF), is often deleted otherwise, inactivated frequently in human melanoma1 and, PTEN, the phosphatase that negatively regulates the PI3K-Akt pathway, is lost in 5–20% of late-stage melanomas.2, 3 Additionally, PI3K mutations occur in 3% of metastatic melanomas4 and Akt is over-expressed in 60% of the tumors.5

Ras-Erk1/2 is a key regulator pathway in melanoma cell proliferation.6 The most commonly mutated component of this pathway is BRAF (BRAFV600E, 50–70%).7 In addition to this, NRAS is mutated in 15% to 30% of melanomas and importantly oncogenic RAS can induce melanoma in p16INK4a-deficient mice.8

The ultraviolet (UV) radiation induced hepatocyte growth factor (HGF) transgenic malignant melanoma mouse model, recapitulates chronologically and histopathologically all the stages of the human disease.9, 10 HGF is a multifunctional cytokine able to elicit mitogenic, motogenic and morphogenic responses through the receptor tyrosine kinase c-Met.11 Notably, c-Met is highly expressed in diverse human and mouse tumors, including melanoma,11, 12 and has been correlated with metastatic progression.13, 14 Interestingly, HGF through its receptor tyrosine kinase constitutively activates Ras and PI3K pathways.15

Melanoma cells are relatively resistant to apoptosis.16 During the last decades immunotherapy and targeted therapy against the main pathways affected have been used without much success.17 Recent studies have revealed that suppression of BRAFV600E inhibits transformation in human melanoma cell lines.18 BAY-43-9006 (sorafenib) is the most common and widely used nonspecific BRAF inhibitor. Unfortunately, mono-therapy with sorafenib in advanced melanoma patients has not been effective.19, 20 However, the anti-tumor activity of sorafenib is enhanced when combined with other chemical agents.19 Some studies in melanoma mouse models provide evidence that the inhibition of both the PI3K pathway (by using, LY294002) and the Ras pathway (MEK inhibitor U0126) results in tumor regression.21 PI-103 is a PI3K/mTOR inhibitor that inhibits selectively the p110α subunit of PI3K22 and has been effective in a glioma xenograph model and leukemogenesis,23.24 Thus, our molecular knowledge of melanoma indicates that targeting BRAF and PI3K pathways simultaneously might be an effective therapeutic approach.

This study investigates the in vitro and in vivo effectiveness of PI-103 and sorafenib as single agents and in combination in melanoma treatment. Results indicate that sorafenib and PI-103 are able to inhibit Erk1/2 and Akt activation, respectively, in response to growth factor treatment (HGF). Interestingly, the combination of the drugs was synergistic blocking the activation of Ras-Erk1/2 and PI3K pathways in response to HGF in a cell line dependent manner. However, the combined treatment of orthotopic xenographs in immunocompetent FVB mice did not cooperate in blocking tumor growth. Surprisingly, the PI3K/mTOR kinase inhibitor (PI-103) induces immunosuppression promoting in vivo tumor growth and inhibiting apoptosis.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Cell lines

37-31E,25 37-31F3, 37-31F3K61 and 37-31F3K63 were maintained in DMEM with 10% FBS, penicillin/streptomycin supplemented with EGF (50 ng/ml) (Invitrogen) and Insulin (4 μg/ml) (Invitrogen). 37-31F3 is the 3rd generation of cells generated by consecutive passes of 37-31E cells through FVB mice with increased tumorigenic capabilities. 37-31F3K61 and 37-31F3K63 cells were obtained from tumors raised in xenograph control mice. All cell lines were grown at 37°C and 5% CO2 conditions.

Inhibitors, antibodies and western blot analysis

PI-103 (Genentech, and Calbiochem), sorafenib (Bayer), rapamycin (LC laboratories) and LY294002 (LC laboratories) were diluted in DMSO and used at the concentrations indicated. Cells were treated with the inhibitors for 2 hr in serum starvation and treated with HGF (40 ng/ml) for 10 min. Cells were lysed in RIPA buffer containing phosphatase and protease inhibitors (SIGMA). Liquid nitrogen frozen tumor samples were homogenized in RIPA buffer. Fifty micrograms of total protein lysates were separated by SDS-PAGE and transferred to a membrane. After blocking, membranes were blotted against different primary antibodies and developed using horseradish peroxidise-linked secondary antibodies and ECL (GE Healthcare). Bands were quantified using NIH 1.6 Image software. ERK2 and cyclin D1 were from Santa Cruz; phospho-Erk1/2 (Thr202/Tyr204), cleaved-caspase-3, p-S6 (Ser235/236), p-STAT3 (Tyr 705), STAT3, Akt and phospho-Akt (Ser473) were from Cell Signaling; Mcl1 was from DAKO; Bcl2 and BclX were from eBioscience; GAPDH was from Trevigen; and β-Actin was from Chemicon Inc. Ki67 was from Master Diagnostica.

Proliferation assays

Cells were seeded one day before treatment. Time point treatments were done in triplicates. Number of viable cells at different time points was analyzed by using Guava-Viacount reagent (Guava Technologies) in a cell counter (Viacount).

In vivo studies

Five to six month old males of either FVB/N strain or nude BALB/c strain were injected subcutaneously with one million cells in PBS. When tumor reached between 50 and 100 mm3, mice were treated with the inhibitors. Treatments were done by IP injection daily with 10 mg/kg or 70 mg/kg of PI-103 and/or 50 mg/kg sorafenib. Control mice were treated with the same volume of DMSO. Tumor size and mice weight was monitored every 2 days. Tumor volume was calculated with the equation (d2 × D) × (π/6). When mice were sacrificed, tumors were dissected and processed. For immunosuppression experiments mice were treated with rapamycin (1 mg/kg) or LY294002 (25 mg/kg) by a daily IP injection for a total of 8 days.

Thymocytes and Splenocytes Quantification

Freshly isolated thymus and spleen were dispersed in RPMI 1640 medium (10% (vv) fetal-calf serum) with a slide glass and passed through a 100 μm mesh nylon screen to obtain thymocytes and splenocytes respectively. Then, the number of viable cells was analyzed by using Guava-Viacount reagent (Guava Technologies). Number of death cells was quantified by nuclear staining with 7-aminoactinomycin D (7AAD) (SIGMA) and flow cytometry analysis (FACScalibur, Becton Dickinson).

Immunohistochemistry, immunofluorescence and tunnel assay

Paraffin-embedded tumor samples were subjected to immunohistochemistry according the manufacturer's antibody protocol. Samples were developed either by using secondary antibodies linked to HRP and DBA as a substrate or by immunofluorescence. Tumor samples were used to perform a tunnel assay as described previously.26 Apoptosis and proliferating cells were quantified by calculating the average of positive cells in ten fields (10×).

mRNA samples and qRT-PCR

Fresh tumor tissues were disrupted using a rotor-stator homogenizer. mRNAs from tumors and cell lines were purified using RNeasy Kit (Quiagen). Amount and quality of RNA was assessed by spectrometrical measurements. Two hundred nanograms of RNA per sample were used to obtain cDNA using SuperScript™ III First-Strand Synthesis System for RT-PCR (Invitrogen). qRT-PCR was performed using validated Taqman Probes (Il-6, IL-10, VEGF, GAPDH) (Applied Biosystems). qRT-PCR was performed according to manufacturers recommendations in a SDS 7900HT System. GAPDH was used as an internal control. Results were calculated using ρρCt method.

Colony formation assays

Three hundred cells were seeded. Treatments were added next day. Plates were incubated at 37°C and 5% CO2 until differences between the treatment conditions were noticeable. Media was changed every 2 days. Plates were washed with PBS, fixed with 4% formaldehyde (SIGMA) in PBS for 10 min, and stained with crystal violet. Finally, representative pictures were taken. At least 2 experiments in triplicates per cell line were performed.


Comparisons protein expression and tumor size among cell lines or treatment groups were done by two-sided t-test (Microsoft Excel). Clonogenic assays were analyzed using Wilcoxon Signed-Rank Test (Vassar Stats).


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Determination of the effective concentration of sorafenib and PI-103 in 37-31E and 37-31E-F3 melanoma cell lines

In this study, we used two cell lines isolated from melanoma neoplasic lesions raised in the HGF transgenic-UV-irradiated mice. We decided to use these cell lines because they arise from tumors induced by a relevant environmental insult (UV) in a mouse model that recapitulate all the human melanoma stages.9 Furthermore, these cells can be used for in vivo tumor formation assays in a syngeneic immunocompetent context. Figure 1a shows that these cell lines have differential in vivo tumorigenic capabilities in FVB immunocompetent mice. 37-31E cells formed a tumor after an average of 30 ± 3 days post-injection whereas 37-31E-F3 took 7 ± 2 days to initiate the tumor (Fig. 1a). We determined the effective concentrations for sorafenib and PI-103 in response to HGF, by measuring the phosphorylation status of Erk1/2 (Thr202/Tyr204) and Akt (Ser473) as reliable indicators of the RAS and PI3K pathway activation, respectively. 37-31E HGF-transgenic tumor cells showed some constitutive activation of RAS and PI3K pathways under serum free conditions (Supporting Information Fig S1). To determine the effective concentration of the compounds for a particular pathway, we treated the cells with HGF to assure the full activation of these pathways avoiding the pleiotropic unknown signaling induced by serum treatment. In 37-31E cells, inhibition by 50% of HGF-induced phosphorylation of Erk1/2 was achieved at concentrations of 5 ± 2.1 μM of sorafenib (Fig. 1b), total inhibition was observed at concentrations above 50 μM (data not shown). PI-103 inhibited Akt phosphorylation in response to HGF by 50% at 17.5 ± 3.2 nM (Fig 1b). In 37-31E-F3 1 ± 2.1 μM concentration of sorafenib inhibited by 50% HGF-induced activation of RAS pathway (Fig. 1c) and 21 ± 1.2 nM of PI-103 reduced to 50% the HGF-induced Akt phosphorylation (Fig. 1c). These results indicate that sorafenib and PI-103 inhibitors have similar inhibitory capabilities in both cell lines blocking RAS and Akt pathways, respectively, upon HGF treatment.

thumbnail image

Figure 1. Determination of effective concentrations of sorafenib and PI-103 inhibitors in melanoma cells. (a) In vivo tumor growth of 37-31-E and 37-31E-F3 cells in FVB/N immunocompetent mice. Cells were subcutaneously injected. Tumor size was monitored every 3 days. (b) Western blots for p-Erk1/2 (Thr202/Tyr204) and p-Akt (Ser473) in 37-31E melanoma cells. Cells were starved and treated for 10 min with HGF (40ng/ml) and the indicated concentrations of sorafenib and PI-103. Graphs show the percentage of HGF-induced kinase phosphorylation upon the different treatments were derived following the quantification of western blots. Data presented corresponds to the average of 3 independent experiments. (c) Analogue experiments as in (b) were performed using 37-31E-F3 melanoma cells.

Download figure to PowerPoint

Sorafenib and PI-103 inhibit in vitro cell growth for proliferation of 37-31E and 37-31E-F3 melanoma cell lines

Cell proliferation and clonogenic capabilities following treatment with each agent were analyzed in vitro adding increasing concentration of the inhibitors (Fig. 2). The addition of HGF to 37-31E cells growing in 10% serum medium did not promote any significant proliferation over the control cells. Both cell lines responded to sorafenib in a dose dependent manner (Fig 2a). The addition of 5 μM sorafenib blocked 37-31E-cell proliferation by 50% and 10 μM concentration or above totally abolished cell division. 37-31E-F3 cells were more resistant to sorafenib treatment and only the addition of 30 μM sorafenib was able to totally block cell proliferation (Fig 2a). Both cell lines showed different sensitivities to proliferation inhibition by PI-103 in a dose dependent manner. 37-31E cell's growth for proliferation was inhibited by 50% at concentrations above 100 nM whereas 37-31E-F3 cell proliferation was inhibited by 50% at 40 ± 2 nM. Concentrations of 50 nM and 100 nM of PI-103 in 37-31E-F3-cell line were almost as effective as 500 nM in 37-31E-cells (Fig. 2b). These data correlated with the results obtained in clonogenic assays (Fig 2b).

thumbnail image

Figure 2. Effects of sorafenib and PI-103 on melanoma cell proliferation and viability. (a) Proliferation and clonogenic assays of 37-31E and 37-31E-F3 melanoma cells in complete medium plus HGF (40 ng/ml) in the presence of increasing concentration of sorafenib. Proliferation assays were performed by counting the number of viable cells at the time points indicated. For clonogenic assays 300 cells were seeded, clones were visualized by crystal violet staining 1 week after. (b) 37-31E and 37-31E-F3 melanoma cells were subjected to same assays as in (a) using increasing concentrations of the PI-103 inhibitor. [Color figure can be viewed in the online issue, which is available at]

Download figure to PowerPoint

Overall, these results indicate that both inhibitors targeting different pathways are effective in blocking in vitro cell proliferation of melanoma cells. However, cells showed a differential sensitivity to the agents in cell proliferation. Furthermore, the data also suggest that in order to block cell proliferation targeting RAS-Erk1/2 pathway seems to be more effective than targeting PI3K/mTOR pathway.

PI-103 and Sorafenib cooperate inhibiting Akt and Erk1/2 phosphorylation and in vitro cell growth for proliferation in response to HGF in a cell line dependent manner

Next, we investigated whether or not the combination of the two drugs had any additive or synergistic effect in blocking the activation of Erk1/2 and Akt pathways in response to HGF. We performed the experiments in a bidirectional way. First, we fixed the amount of one inhibitor at suboptimal concentration, adding increasing amounts of the second inhibitor and then, we repeated the experiment in a reverse way. In 37-31E cells, the combination of the two drugs did not show any additive or synergistic effect inhibiting Ras-Erk1/2 and PI3K pathways in response to HGF (Fig. 3a). However, in 37-31E-F3 cells, the addition of increasing concentrations of PI-103 over a fixed concentration of sorafenib (10 μM) resulted in the synergistic inhibition of the phosphorylation of both kinases and ribosomal protein S6 in response to the growth factor (Fig. 3b). The same results were observed when experiments were done with fixed concentration of PI-103 (20 nM) and the addition of increasing concentrations of sorafenib (Fig. 3b). Furthermore, combination of 5 μM of sorafenib and 20 nM of PI-103 were synergistic inhibiting the HGF-induced in vitro cell proliferation of 37-31E-F3 cells (Fig 3c).

thumbnail image

Figure 3. Inhibition of the HGF-induced Erk1/2 and Akt phosphorylation by combined treatment of sorafenib and PI-103. Western blots of p-Erk1/2 (Thr202/Tyr204), p-Akt (Ser473), Erk1/2 and Akt in melanoma cells. Cells were starved and treated as indicated. Cells were triggered with 40 ng/ml of HGF for 10 minutes. Representative experiments for (a) 37-31E cells and (b)37-31E-F3 cells are shown. Graphs show the average of the quantification from three independent experiments. Bars indicate the S.D. among the different experiments. (c) In vitro cell proliferation assay of 37-31E-F3. Cells were treated with the indicated concentrations of sorafenib and PI-103 for 72 hr. Number of viable cells was counted at the indicated time points. Bars indicate the S.D.

Download figure to PowerPoint

These data indicated that the drug combination was synergistically inhibiting the targeted pathways in response to HGF in a cell line dependent manner.

PI-103 promotes in vivo tumor growth increasing proliferation and blocking apoptosis

In humans, mono-therapy with sorafenib has not been very successful in melanoma treatment.19 The aforementioned results indicated that the combination of sorafenib and PI-103 abolished synergistically the HGF-induced phosphorylation of Erk1/2 and Akt in a cell line dependent manner. Thus, we investigated the effects of the combination of the two drugs in vivo by performing orthotopic xenographs in immunocompetent mice. 37-31E-F3 melanoma cells were injected subcutaneously in FVB/N wild type mice and animals were treated with a previously reported effective dose of PI-103 (10 mg/kg)23 or/and sorafenib (50 mg/kg). Treatments showed no apparent effect on general health of mice as indicated by their body weight (data not shown). Surprisingly, PI-103 treatment promoted a significant in vivo tumor growth compared with the DMSO treated mice. Sorafenib treatment reduced tumor growth and the treatment with the combination of the two agents did not show any further benefit on sorafenib induced-tumor reduction, moreover, tumors were slightly bigger than the ones treated with sorafenib alone (Fig 4a; Supporting Information S2).

thumbnail image

Figure 4. In vivo tumor growth experiments: (a) 37-31E-F3 melanoma cell line was subcutaneously injected into immunocompetent FVB/N mice. The animal groups were treated with DMSO (Control) (n = 5), PI-103 (10 mg/kg) (n = 5), Sorafenib (50 mg/kg) (n = 5) or PI-103 (10 mg/kg) plus Sorafenib (50 mg/kg) (n = 5). Tumor size was calculated as described in materials and methods. p-values were calculated performing a t-student test. (b) Immunohistochemistry with the indicated antibodies and a tunel assays were performed in paraffin-included sections of tumor samples. Specific nuclear costaining of tunel assay and dapi are shown in the amplified sections. Pictures show representative tumor sample sections. Graphs on the right show average of positive cells per ten fields (tunnel assay and Ki67). Bars indicate S.D. and significant differences are indicated as p-value. (c) PI-103-induced in vivo tumor growth is concentration and cell line independent. 37-31E-F3 melanoma cells were used to perform an in vivo tumor growth experiment in immunocompetent FVB/N mice (n = 5 per group) using either DMSO or 70 mg/kg of PI-103. Bars indicate S.D. Graph on the right, 37-31E cells were used for the in vivo tumor growth experiment. Mice were treated daily with either DMSO or PI-103 at 10mg/kg. (d) In vivo tumor growth of 37-31E cells in immunocompromised mice. Cells were subcutaneously injected into BALB/c nude mice. The mice groups (n = 5 per group) were treated with DMSO (Control), PI-103 (10 mg/kg). Tumor size was monitorized every 2 days. p-value was calculated performing a t-student test. [Color figure can be viewed in the online issue, which is available at]

Download figure to PowerPoint

We hypothesized that tumor cells from PI-103 treated mice were either proliferating more, having a lower rate of apoptosis or both. We used the tumor's tissue to investigate proliferation and apoptotic markers. Sorafenib-treated tumors showed very low signaling through the RAS-Erk1/2 pathway and the PI-103 PI3K/mTOR dual inhibitor was effective by partially inhibiting the Akt and S6 ribosomal protein phosphorylation (Fig 4b). Interestingly, tumors from PI-103-treated mice showed higher levels of cyclin D1 and more proliferating cells as indicated by the number of Ki67 positive cells (p < 0.01) (Fig 4b). Importantly, the data also showed that according to the tunel assay and the cleaved caspase-3, PI-103-treated tumors had the lowest apoptotic rate (Fig 4b). Sorafenib-treated tumors showed a very high rate of apoptosis by tunel assay that totally correlated with the cleaved caspase-3 levels (p < 0.01) (Fig 4b). Interestingly, tumors treated with the two agents showed less apoptotic cells than tumors treated with sorafenib alone (Fig 4b).

To investigate whether the PI-103 effects were dose dependent, we performed another experiment using higher amounts of PI-103 (70 mg/kg). Figure 4c shows that PI-103 (70 mg/kg) also promoted a significant in vivo tumor growth (p < 0.05) indicating a dose-independent effect. We also investigated whether the PI-103-induced tumor growth was cell line dependent. 37-31E cells were used to repeat the in vivo experiment in response to PI-103 treatment. After 24 days treatment, mice treated with PI-103 (10 mg/kg) showed a significant higher tumor-growth rate than the control ones (p < 0.01) (Fig. 4c).

A number of studies have shown the effectiveness of PI-103 in tumor treatment in immunocompromised mouse models.24, 27 We investigated whether or not PI-103 was able to reduce the in vivo melanoma tumor growth in BALB/c nude mice. In this model, the tumor's latency of 37-31E cells was reduced from 1 month to 1 week and PI-103 (10 mg/kg) treatment significantly abolished (p < 0.01) tumor growth (Fig 4d).

These results indicate that according to the tumor latency, tumor cells might be differentially selected in the immunocompetent model. Although PI-103 is effective blocking melanoma tumor growth in an in vivo immunocompromised model, PI-103 induces melanoma tumor growth in an immunocompetent model. These data correlates with the higher number of proliferating cells and a low rate of apoptosis. Sorafenib treatment alone increased apoptosis reducing the tumor growth and, combination of the two agents did not represent any further benefit on sorafenib-treated mice. Furthermore, the addition of PI-103 reduced the sorafenib-induced tumor apoptosis.

PI-103 induces immunosuppression of FVB immunocompetent mice

Our data together with previous publications24, 27 indicated that the mouse background and/or the immune system could interfere with the drug effectiveness. It is known that mTOR inhibitors such as rapamycin, induces immunosuppression reducing the thymocytes and splenocytes cell populations.28 We investigated the possible effect of PI-103 modulating the mouse immune system analyzing whether the drug had any effect on the thymocytes and splenocytes population. FVB/N mice were treated daily either with DMSO or 10 mg/kg of PI-103 for 8 days. Then, the number of thymocytes and splenocytes were quantified and cell death was measured by Guava-Viacount reagent exclusion staining and by flow cytometry analysis and 7-Aminoactinomycin D (7-ADD) staining. PI-103 treated mice showed a significant (p < 0.05) reduction in the number of thymocytes (70%) and a slight decrease in the number of splenocytes (15%). Furthermore, control mice showed 8% of dead thymocytes while PI-103-treated mice showed a 26% dead (Fig 5a). However, no differences in dead cells were observed within the splenocytes population. We obtained similar results when we analyzed the thymus from animals used in the in vivo tumor growth experiment (Fig 5b). To investigate whether these results were extensive to other PI3K/mTOR inhibitors, we performed the same experiment using either 1 mg/kg rapamycin or 25 mg/kg of LY294002. As expected, the mTOR inhibitor rapamycin almost totally reduced the number of thymocytes (80%) after one-week treatment, producing also a significant decrease in the number of splenocytes (20%) (Fig 5c). Treatment with the PI3K/mTOR dual inhibitor LY294002 led to a slight decrease in the number of thymocytes (15–18%) (p < 0.1) that correlates with a significant increased percentage of dead cells (7%) (p < 0.05), no significant changes were observed in the splenocyte population (Fig 5c).

thumbnail image

Figure 5. PI-103 induces thymocytes depletion in an immunocompetent animal model. (a) FVB/N mice were treated daily for eight days either with DMSO (c) or PI-103 (10 mg/kg) (PI). Then, thymocytes and splenocytes were isolated and number of cells and viability was analyzed using Guava-Viacount system or by FACS analysis and 7AAD nuclear staining. Graphs on the left show number of thymocytes and splenocytes in control mice (n = 6) and PI-103 treated mice (n = 8). Graphs on the right show the percentage of cell death in both cell populations. p-value was calculated performing a t-student test. (b) Thymocytes and splenocytes from the mice used in the in vivo tumor growth experiment (Fig 4c, 37-31E-F3 cells) were isolated and analyzed as in (a). (c) FVB/N mice were treated daily for 8 days either with DMSO (n = 4) (C), LY294002 (25 mg/kg) (n = 4) (LY) or rapamycin (1 mg/kg) (n = 4) (R). Then, thymocytes and splenocytes from mice were isolated and analyzed as earlier.

Download figure to PowerPoint

Additionally, signal transducer and activator of transcription 3 (STAT3) represents a convergence point for numerous oncogenic signaling pathways. STAT3 activity promotes the production of immunosuppressive factors (i.e IL-6, IL10 and VEGF), restraining anti-tumor immune responses (for review29). In vitro treatment of 37-31 E-F3 cells for 48 hr with 50 nM PI-103 promoted an increase of p-STAT3 levels, however, treatment with 15 μM sorafenib induced the contrary response (Fig 6a). The increment of PI-103-induced p-STAT3 levels correlated with a slight increase of IL6 and VEGF mRNA levels, while IL10 mRNA was not detected (Fig 6a). Interestingly, PI-103-treated tumor samples from immunocompetent mice showed higher levels of p-STAT3 compared with the controls (Fig 6b). Sorafenib treated samples and tumor samples treated with both inhibitors did not show elevated levels of p-STAT3 (Fig 6b). To further investigate the immuno-modulating effects of PI-103, we checked the levels of STAT3-dependent transcriptional regulated imunosuppressor's genes (IL6, IL10 and VEGF) within the tumors. Figure 6c shows that PI-103-treated tumors samples from two different experiments (10 mg/kg and 70 mg/kg), performed in immunocompetent mice, had higher levels of immunosuppressors such as IL6, IL10 and VEGF than controls. Interestingly, PI-103-treated tumor samples from BALB/c nude mice showed lower mRNA levels of IL6, IL10 and VEGF than their controls (Fig 6d).

thumbnail image

Figure 6. PI-103 treatment induces STAT3 phosphorylation and transcriptional upregulation of immunosuppressors. (a) PI-103 increases p-STAT3, IL6 and VEGF levels. Fifty micrograms of total protein lysates from 37-31E-F3 cells treated for 48 hr in complete medium with either DMSO (C), 50 nM PI-103 (P) or 15 μM sorafenib (S) were resolved by SDS-PAGE. Membranes were blotted against p-STAT3 and STAT3 antibodies. Quantitative RT-PCR for IL6 and VEGF expression was performed in DMSO or PI-103-treated cells under the same conditions as above. (b) Immunohistochemistry of paraffin included tumor samples from an in vivo tumor growth experiment showing the levels of p-STAT3. Tumor samples were from mice treated as follow, control (DMSO treated), PI-103 (10 mg/kg), sorafenib (50 mg/kg) and PI-103+sorafenib (10 mg/kg and 50mg/kg respectively). Representative pictures are showed. Insets show a 40× magnification of the pictures. (c) PI-103 treated tumors from immunocompetent mice showed higher levels of IL6, IL10 and VEGF. mRNA of tumor samples obtained from experiments showed in Figures 4a and 4c (treatment of 10 mg/kg and 70 mg/kg of PI-103) were subjected to quantitative RT-PCR. Expression levels of IL6, IL10 and VEGF in the different tumor samples (n = 5 mice per group treatment) are shown. Bars indicate the S.D. (d) Same experiment as in (c) was performed using tumor samples raised in BALB/c nude mice (n = 5 mice per group treatment). Bars indicate the S.D. [Color figure can be viewed in the online issue, which is available at]

Download figure to PowerPoint

thumbnail image

Figure 7. PI-103 induces the anti-apoptotic BH3 family members in tumor derived cell lines. (a) Western blot showing the basal expression of the BH3-family members in 37-31E, 37-31E-F3, 37-31E-F3K61, 37-31E-F3K63 cell lines. (b) PI-103 but not sorafenib up-regulates the expression of the BH3-family members. Cell lines (37-31E-F3, 37-31E-F3K61, 37-31E-F3K63) were treated as indicated for 48 hr in complete medium. Then total protein lysates were checked for the status of the indicated proteins (Mcl1, Bcl2 and BclX). GAPDH was used as a loading control. Clonogenic assays with 37-31E-F3 (parental) and 37-31E-F3K63 (tumor derived) cell lines. After seeding, cells were treated with the indicated concentrations of PI-103 for 1 week. Number of viable clones was visualized by crystal violet staining. (c) Western-blot of total protein lysates from DMSO (control; samples 1-3) or PI-103 (samples 4–6) treated FVB/N mice tumor samples or DMSO (control; samples 7–9) or PI-103 (10 mg/kg) (samples 10–12) treated BALB/c nude mice tumor samples. Levels of BH3-family members are shown. Graphs show the quantification of the bands from the different groups. p-value was calculated performing a t-student test. (d) PI-103 increases viability of sorafenib treated cells. Cells were treated with the indicated concentrations of sorafenib and PI-103 for 1 week. Number of viable clones was visualized by crystal violet staining. The assays were performed three times as triplicates. Representative pictures are shown. Graph shows the quantification of all experiments. p-value was calculated performing a Wilcoxon-signed rank test (Vassar Stats). Western-blot showing the expression of Mcl1, Bcl2 and BclX and GAPDH (loading control) in 37-31E-F3 cells. Cells were treated for 48 hr with the combinations indicated.

Download figure to PowerPoint

Altogether, these data show evidences supporting that targeting PI3K/mTOR pathway with PI-103 in immunocompetent mice, induces immunosuppression reducing the number of thymocytes and regulating the expression of immunosuppressors such as IL6, IL10 and VEGF.

PI-103 regulates anti-apoptotic BH3-family members

In addition to the effects on the immune system, the in vivo experiments in FVB mice showed that PI-103-treated tumors had a lower apoptotic tumor rate. Importantly, it has been demonstrated that elevated levels of anti-apoptotic BH3-family members (Mcl1, Bcl2, and BclX) conferred resistance to the mTOR inhibitor rapamycin.30, 31 To investigate the possible PI-103 effects regulating tumor apoptosis we generated tumor-derived cell lines from the DMSO-treated FVB mice (37-31E-F3K61 and 37-31E-F3K63). We first examined the basal level of anti-apoptotic proteins in these cell lines compared with the parental cells (37-31E-F3). The two tumor-derived cell lines showed higher levels of Mcl-1 than the parental cells. 37-31E-F3 and 37-31E-F3K63 cells also showed higher levels of Bcl2 (Fig 7a). Then, we investigated whether the PI3K-mTOR inhibitor had any direct effect regulating the anti-apoptotic proteins in parental cells and tumor-derived cell lines. Cell lines were treated with increasing concentrations of either PI-103 or sorafenib for 48 hr. The addition of PI-103 promoted the up-regulation of Mcl1, Bcl2 and BclxL in the two tumor derived cell lines compared with the parental cells even at concentrations of 500 nM. Sorafenib treatment promoted an increase of the anti-apoptotic proteins at low concentrations. However at 15 μM the amount of the anti-apoptotic proteins was reduced in all cell lines (Fig 7b). Furthermore, clonogenic assays performed with 37-31E-F3 and 37-31E-F3K63 showed a correlation between the levels of the anti-apoptotic proteins and the number of clones present at 50 nM and 100 nM of PI-103 (Fig 7b). Interestingly, the tumor samples from PI-103-treated FVB/N mice and BALB/c mice showed different responses to the treatment in regard the levels of the anti-apoptotic proteins Mcl1, Bcl2 and BclXL. In FVB/N mice, PI-103 tumor-treatment induced an increase of the BH3-family members of anti-apoptotic proteins. However, PI-103-treated tumors from BALB/c mice showed a totally opposite response (Fig 7c).

Additionally, the in vivo experiments showed that PI-103 partially blocked the sorafenib-induced tumor apoptosis (Fig 4c). To check the possible role of PI-103 kinase inhibitor in blocking the sorafenib effect, we performed clonogenic assays combining both inhibitors in 37-31E-F3 cells. Importantly, the addition of increasing concentration of PI-103 over 10 μM sorafenib treated cells, resulted in a significant increase in the clone's number (p < 0.001) (Fig 7d) concomitant with the accumulation of Mcl1 and Bcl2 anti-apoptotic proteins.

These results show evidences indicating that PI-103, promoted the up-regulation of the anti-apoptotic BH3-family members in tumors and tumor-derived cells lines from immunocompetent mice, supporting the lower apoptosis rate observed in vivo in the PI-103 treated tumors.


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The incidence-rate for melanoma cases has dramatically increased during the last few decades. Having a wide age distribution, if melanoma reaches its more advance stages, it becomes very aggressive with no effective therapeutic treatment available. Several investigations have shown that RAS pathway and PI3K pathway seem to be important for melanoma development and maintenance.2, 8 In this study, we investigate the effectiveness in tumor growth inhibition of two small molecule inhibitors targeting BRAF (sorafenib) and PI3K (p110α)/mTOR (PI-103) as single agents and in combination. We show that both inhibitors as single agents were effective blocking in vitro proliferation. Interestingly, the drug combination appeared to be synergistic at the kinase pathway inhibition level in response to growth factor in a cell line dependent manner. However, in vivo tumor growth studies performed in immunocompetent mice revealed an unexpected effect of PI-103 inhibitor promoting immunosuppression and in vivo tumor growth, while the drug combination did not show any further benefit. Additionally, the analysis of tumor samples suggested that PI-103 was promoting in vivo tumor proliferation inhibiting tumor cell apoptosis. Furthermore, the biochemical analysis of derived tumor cell lines confirmed the PI-103-mediated up-regulation of the BH3-protein family members Mcl1, Bcl2 and BclX. Importantly, in vitro experiments showed that PI-103 was able to increase the viability of sorafenib-treated cells.

The HGF transgenic malignant melanoma mouse model recapitulates all the stages of the human disease.9, 10 This growth factor activates both RAS and PI3K pathways.15 By using this paracrine multifunctional cytokine naturally secreted in skin, we assure the activation of the pathways in a pertinent context avoiding serum pleiotropic effects. Although both drugs, sorafenib and PI-103, were effective blocking the kinase activation in response to HGF, proliferation and clonogenic assays showed that sorafenib was more effective than PI-103 blocking proliferation and inhibiting viability. These data is in agreement with the notions that, in melanoma, RAS-Erk1/2 signaling provides essential tumor maintenance functions32 and HGF-induced proliferation is largely dependent on RAS pathway.25 It was particularly interesting that the inhibitors cooperate at the kinase level inhibition in response to growth factors in a cell line-dependent manner. While 37-31E-F3 cells were sensitive to the double treatment at the kinase inhibition level the 37-31E cells did not show any cooperative response to the combination. Furthermore, the treatment of this cell line with 20 nM of the PI3K/mTOR inhibitor and 10 μM of sorafenib resulted in an increase of the p-Erk1/2 levels. These data highlights the complexity of the connections between mTORC1, PI3K and Erk1/2 pathways and is in agreement with recent investigations that show the existence of feedback loops that activate Erk1/2 after mTORC1 inhibition.33

The investigation of the drug's effectiveness in vivo in an immunocompetent model takes in account two important facts. First, tumors grown in such a context arose from cells that have to bypass the immune system in a more physiological way than cells that produce tumors in immunocompromised models. Second, any general side effect of the drugs on the immune system could be detected. As previously shown by others,34 in our model sorafenib was effective blocking in vivo tumor growth. Surprisingly, treatment of FVB/N mice with PI-103 alone clearly promoted in vivo tumor growth in a dose and a cell line independent manner and the drug's combination did not show any further benefit. Several publications have shown the potential antitumoral activity of PI-103 in glioblastoma.22, 27, 35, 36 However, these works also show different results in respect to the effective dose and the antitumoral activity depending on the animal tumor model used.22, 27, 35 Recently, it has been suggested that dual PI3K/mTOR inhibitors were advantageous to attenuate melanoma growth.37 However, our experiments using PI-103 inhibitor in FVB/N mice indicate opposite results. Besides the fact that a different tumor mouse model was used in this work, this investigation also showed that PI-103 was the less effective dual inhibitor tested, and importantly, did not promote tumor reductions in lymph node metastasis. Moreover, it decreased the tumor-necrosis index and did not suppress neovascularization.37 These pieces of data are in concordance with the PI-103-mediated apoptosis resistance observed in our model.

Targeting RAS and PI3K pathways in melanoma has been reported to be effective promoting melanoma regression in organotypic cultures, the DMBA-induced TPRas mouse melanoma model and the recently described Tyr::CreER;BrafCA/+xPTENlox/lox mouse melanoma model.21, 38–40 In our system, the inhibition of the same pathways with sorafenib and PI-103 did not show any further benefit. Furthermore, we observed that tumors treated with the drug combination showed less apoptosis than the ones treated with sorafenib alone. Interestingly, the experiments combining LY294002 and U0126 inhibitors in the DMBA-induced Ras-dependent melanoma model showed that 21.6% of the spontaneous tumors treated with the drug combination progressed. Unfortunately, this study does not provide any information about the treated tumor's size respect to the control's. Furthermore, drugs in this study were administered topically with a presumably minimal or none systemic effects on the animals. The study from Dankort et al. using a Tyr::CreER;BrafCA/+ x PTENlox/lox mouse melanoma model showed that targeting the RAF and mTORC1 pathways was effective blocking tumor growth. However, tumor development was totally dependent on the activation of both RAS and PI3K pathways. Mek1/2 inhibitors have been shown to be particularly effective in blocking tumor growth of BRAFV600E mutant melanoma cells,41 consequently it is not surprising the tumor's response to the combinated treatment. The cells used in our study are derived from the HGF-transgenic mouse and do not harbor mutations either in BRAF or PTEN, however, they seem to be partially sensitive to the multi-kinase inhibitor sorafenib. Altogether, these results suggest that combinational treatments targeting BRAF and PI3K/mTOR pathways might be particularly effective in RAS mutant melanoma tumors and/or BRAF mutated /PTEN null tumors.

The antitumoral effectiveness of PI-103 alone or in combination with γ-radiation has been mainly proved in immunocompromised mouse models.23, 42 Our experiments using BALB/c nude mice are consistent with these results. Besides the distinct genetic background of FVB/N and BALB/c nude mouse models, an important difference resides in the immune system. An intriguing possibility was that PI-103 could be compromising the immune system promoting tolerance to melanoma tumor growth. It is widely known that the mTOR inhibitor rapamycin induces immunosuppression.43 Notably, our data show that PI-103, promotes immunosuppression in immunocompetent mice by inducing thymus atrophy, reducing the thymocyte cell number and modulating the levels of p-STAT3 within the tumor. Constitutively activated STAT3 inhibits the expression of mediators necessary for immune activation against tumor cells, and promotes the production of immunosuppressive factors (for review29). In this matter, our results obtained from tumors raised in FVB/N mice are in agreement with these effects, suggesting that STAT3 could be participating in the regulation of the immunosuppressor factors (IL6, IL10 and VEGF) and providing a possible mechanism and connection between tumor cells and the host immune system that could partially explain the PI-103-induced tumor growth. The study from Dankort et al. does not report any effects of rapamycin in the immune system promoting tumor growth. To this extent, we cannot exclude that the effects observed by the dual PI3K/mTOR inhibitor PI-103 in our model could be due to the drug specificity, the mouse strain and/or the melanoma cells used in the study. Interestingly, the modulation of the immune system that we observed is in concordance with recent publications demonstrating that PI-103 blocks proliferation of T cells,24 enhances IgE production and reduces type-2 cytokine production in vivo.44 Importantly, these results highlight the relevance of the animal models in preclinical investigations.

Additionally, the data also show that PI-103 could have some direct effect on tumor cells regulating apoptosis. It has been well described that the anti-apoptotic proteins Mcl1 and Bcl2 are involved in rapamycin resistance and could be induced by rapamycin treatment.30, 31, 45, 46 In agreement with this, tumor-derived cell lines from the untreated mice not only showed elevated levels of the anti-apoptotic proteins Mcl1 and Bcl2 but also they were up-regulated after PI-103 treatment. Melanomas utilize multiple pathways to acquire the ability to attenuate signals that would normally lead to apoptosis.47 STAT3 is a downstream signaling molecule activated by HGF/SF-Met signaling, and is reported to contribute to cell transformation and resistance to apoptosis.47, 48 It is known that STAT3 negatively regulates TRAIL transcription protecting cells from apoptosis.49 Moreover, inhibition of PI3K was sufficient to increase transcriptional activities of Jun and STAT3 resulting in decreased Fas transcription leading to resistance to Fas-mediated cell death.50 Importantly, these data supports the diminished apoptotic index of the PI-103-treated tumors and the inhibition of sorafenib-mediated apoptosis in tumors treated with both agents. These results permit us to speculate that in addition RAS pathway, PI-103 treatment, probably through STAT3,29 increases the amount of these proteins contributing to the lower apoptotic index and tumor growth promotion.

Here, we show that in melanoma cells, sorafenib and PI-103 were effective in blocking RAS and PI3K pathways in vitro, respectively. The combination of both drugs showed to be synergistic in inhibiting Erk1/2 and Akt phosphorylation and in vitro cell proliferation in response to HGF in a cell line dependent manner. In vivo experiments performed in immunocompetent mice indicated that sorafenib was partially effective blocking tumor growth. However, PI-103 inhibitor promoted an unexpected increase of in vivo tumor growth that correlated with an increased tumor proliferation and reduced tumor apoptosis, and unfortunately, the drug combination did not show any further benefit. Notably, the use of an immunocompetent animal model to test the drug's effectiveness uncover two different important issues to be considered in preclinical studies, tumor cell selection in an immunocompetent context and undesirable side effects on the immune system. PI-103 induced immunosuppression and promoted the up-regulation of anti-apoptotic BH3-family proteins in tumor derived cell lines that could account for some of the observed effects. Overall the data indicates that due to the melanoma intra-tumoral heterogeneicity some precautions should be taken when using these inhibitors for treatment. More importantly, these results certainly make an argument for investigating drug-mediated unexpected effects in more physiological animal models.


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors thank Dr. Josep Baselga for his useful input and Dr. Joan Seoane for his helpful discussions. They also thank Dr. Javier Cortes for providing them some useful reagents and Dr. Stephan Tenbaum for reading the manuscript.


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Chin L, Garraway LA, Fisher DE. Malignant melanoma: genetics and therapeutics in the genomic era. Genes Dev 2006; 20: 214982.
  • 2
    Wu H, Goel V, Haluska FG. PTEN signaling pathways in melanoma. Oncogene 2003; 22: 311322.
  • 3
    Li L, Ross AH. Why is PTEN an important tumor suppressor? J Cell Biochem 2007; 102: 136874.
  • 4
    Omholt K, Krockel D, Ringborg U, Hansson J. Mutations of PIK3CA are rare in cutaneous melanoma. Melanoma Res 2006; 16: 197200.
  • 5
    Stahl JM, Sharma A, Cheung M, Zimmerman M, Cheng JQ, Bosenberg MW, Kester M, Sandirasegarane L, Robertson GP. Deregulated Akt3 activity promotes development of malignant melanoma. Cancer Res 2004; 64: 700210.
  • 6
    Cohen C, Zavala-Pompa A, Sequeira JH, Shoji M, Sexton DG, Cotsonis G, Cerimele F, Govindarajan B, Macaron N, Arbiser JL. Mitogen-actived protein kinase activation is an early event in melanoma progression. Clin Cancer Res 2002; 8: 372833.
  • 7
    Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, Garnett MJ, Bottomley W, Davis N, Dicks E, et al. Mutations of the BRAF gene in human cancer. Nature 2002; 417: 94954.
  • 8
    Chin L, Tam A, Pomerantz J, Wong M, Holash J, Bardeesy N, Shen Q, O'Hagan R, Pantginis J, Zhou H, Horner JW,II, Cordon-Cardo C, et al. Essential role for oncogenic Ras in tumour maintenance. Nature. 1999; 400: 46872.
  • 9
    Noonan FP, Recio JA, Takayama H, Duray P, Anver MR, Rush WL, De Fabo EC, Merlino G. Neonatal sunburn and melanoma in mice. Nature 2001; 413: 2712.
  • 10
    Recio JA, Noonan FP, Takayama H, Anver MR, Duray P, Rush WL, Lindner G, De Fabo EC, DePinho RA, Merlino G. Ink4a/arf deficiency promotes ultraviolet radiation-induced melanomagenesis. Cancer Res 2002; 62: 672430.
  • 11
    Lesko E, Majka M. The biological role of HGF-MET axis in tumor growth and development of metastasis. Front Biosci 2008; 13: 127180.
  • 12
    Benvenuti S, Comoglio PM. The MET receptor tyrosine kinase in invasion and metastasis. J Cell Physiol 2007; 213: 31625.
  • 13
    Cruz J, Reis-Filho JS, Silva P, Lopes JM. Expression of c-met tyrosine kinase receptor is biologically and prognostically relevant for primary cutaneous malignant melanomas. Oncology 2003; 65: 7282.
  • 14
    Natali PG, Nicotra MR, Di Renzo MF, Prat M, Bigotti A, Cavaliere R, Comoglio PM. Expression of the c-Met/HGF receptor in human melanocytic neoplasms: demonstration of the relationship to malignant melanoma tumour progression. Br J Cancer 1993; 68: 74650.
  • 15
    Gao CF, Vande Woude GF. HGF/SF-Met signaling in tumor progression. Cell. Res. 2005; 15: 4951.
  • 16
    Soengas MS, Lowe SW. Apoptosis and melanoma chemoresistance. Oncogene 2003; 22: 313851.
  • 17
    Becker JC, Kirkwood JM, Agarwala SS, Dummer R, Schrama D, Hauschild A. Molecularly targeted therapy for melanoma: current reality and future options. Cancer 2006; 107: 231727.
  • 18
    Hingorani SR, Jacobetz MA, Robertson GP, Herlyn M, Tuveson DA. Suppression of BRAF(V599E) in human melanoma abrogates transformation. Cancer Res 2003; 63: 5198202.
  • 19
    Eisen T, Ahmad T, Flaherty KT, Gore M, Kaye S, Marais R, Gibbens I, Hackett S, James M, Schuchter LM, Nathanson KL, Xia C, et al. Sorafenib in advanced melanoma: a Phase II randomised discontinuation trial analysis. Br J Cancer 2006; 95: 5816.
  • 20
    Hauschild A, Agarwala SS, Trefzer U, Hogg D, Robert C, Hersey P, Eggermont A, Grabbe S, Gonzalez R, Gille J, Peschel C, Schadendorf D, et al. Results of a phase III, randomized, placebo-controlled study of sorafenib in combination with carboplatin and paclitaxel as second-line treatment in patients with unresectable stage III or stage IV melanoma. J Clin Oncol 2009 Jul 20; 27: 3496502
  • 21
    Bedogni B, Welford SM, Kwan AC, Ranger-Moore J, Saboda K, Powell MB. Inhibition of phosphatidylinositol-3-kinase and mitogen-activated protein kinase kinase 1/2 prevents melanoma development and promotes melanoma regression in the transgenic TPRas mouse model. Mol Cancer Ther 2006; 5: 30717.
  • 22
    Raynaud FI, Eccles S, Clarke PA, Hayes A, Nutley B, Alix S, Henley A, Di-Stefano F, Ahmad Z, Guillard S, Bjerke LM, Kelland L, et al. Pharmacologic characterization of a potent inhibitor of class I phosphatidylinositide 3-kinases. Cancer Res 2007; 67: 584050.
  • 23
    Fan QW, Cheng CK, Nicolaides TP, Hackett CS, Knight ZA, Shokat KM, Weiss WA. A dual phosphoinositide-3-kinase {alpha}/mTOR inhibitor cooperates with blockade of epidermal growth factor receptor in PTEN-mutant glioma. Cancer Res 2007; 67: 79605.
  • 24
    Kharas MG, Janes MR, Scarfone VM, Lilly MB, Knight ZA, Shokat KM, Fruman DA. Ablation of PI3K blocks BCR-ABL leukemogenesis in mice, and a dual PI3K/mTOR inhibitor prevents expansion of human BCR-ABL+ leukemia cells. J Clin Invest 2008; 118: 303850.
  • 25
    Recio JA, Merlino G. Hepatocyte growth factor/scatter factor activates proliferation in melanoma cells through p38 MAPK. ATF-2 and cyclin. D1. Oncogene 2002; 21: 10008.
  • 26
    Kingsley-Kallesen M, Mukhopadhyay SS, Wyszomierski SL, Schanler S, Schutz G, Rosen JM. The mineralocorticoid receptor may compensate for the loss of the glucocorticoid receptor at specific stages of mammary gland development. Mol Endocrinol 2002; 16: 200818.
  • 27
    Fan QW, Knight ZA, Goldenberg DD, Yu W, Mostov KE, Stokoe D, Shokat KM, Weiss WA. A dual PI3 kinase/mTOR inhibitor reveals emergent efficacy in glioma. Cancer Cell 2006; 9: 3419.
  • 28
    Tian L, Lu L, Yuan Z, Lamb JR, Tam PK. Acceleration of apoptosis in CD4+CD8+ thymocytes by rapamycin accompanied by increased CD4+CD25+ T cells in the periphery. Transplantation 2004; 77: 1839.
  • 29
    Yu H, Kortylewski M, Pardoll D. Crosstalk between cancer and immune cells: role of STAT3 in the tumour microenvironment. Nat Rev Immunol 2007; 7: 4151.
  • 30
    Wendel HG, Lowe SW. Reversing drug resistance in vivo. Cell Cycle 2004; 3: 8479.
  • 31
    Slavik JM, Lim DG, Burakoff SJ, Hafler DA. Rapamycin-resistant proliferation of CD8+ T cells correlates with p27kip1 down-regulation and bcl-xL induction, and is prevented by an inhibitor of phosphoinositide 3-kinase activity. J Biol Chem 2004; 279: 91019.
  • 32
    Gray-Schopfer VC, da Rocha Dias S, Marais R. The role of B-RAF in melanoma. Cancer Metastasis Rev 2005; 24: 16583.
  • 33
    Carracedo A, Ma L, Teruya-Feldstein J, Rojo F, Salmena L, Alimonti A, Egia A, Sasaki AT, Thomas G, Kozma SC, Papa A, Nardella C, et al. Inhibition of mTORC1 leads to MAPK pathway activation through a PI3K-dependent feedback loop in human cancer. J Clin Invest 2008; 118: 306574.
  • 34
    Sharma A, Trivedi NR, Zimmerman MA, Tuveson DA, Smith CD, Robertson GP. Mutant V599EB-Raf regulates growth and vascular development of malignant melanoma tumors. Cancer Res 2005; 65: 241221.
  • 35
    Chen JS, Zhou LJ, Entin-Meer M, Yang X, Donker M, Knight ZA, Weiss W, Shokat KM, Haas-Kogan D, Stokoe D. Characterization of structurally distinct, isoform-selective phosphoinositide 3′-kinase inhibitors in combination with radiation in the treatment of glioblastoma. Mol Cancer Ther 2008; 7: 84150.
  • 36
    Stallone G, Infante B, Grandaliano G, Schena FP, Gesualdo L. Kaposi's sarcoma and mTOR: a crossroad between viral infection neoangiogenesis and immunosuppression. Transpl Int 2008; 21: 82532.
  • 37
    Marone R, Erhart D, Mertz AC, Bohnacker T, Schnell C, Cmiljanovic V, Stauffer F, Garcia-Echeverria C, Giese B, Maira SM, Wymann MP. Targeting melanoma with dual phosphoinositide 3-kinase/mammalian target of rapamycin inhibitors. Mol Cancer Res 2009; 7: 60113.
  • 38
    Lasithiotakis KG, Sinnberg TW, Schittek B, Flaherty KT, Kulms D, Maczey E, Garbe C, Meier FE. Combined inhibition of MAPK and mTOR signaling inhibits growth, induces cell death, and abrogates invasive growth of melanoma cells. J Invest Dermatol 2008; 128: 201323.
  • 39
    Smalley KS, Haass NK, Brafford PA, Lioni M, Flaherty KT, Herlyn M. Multiple signaling pathways must be targeted to overcome drug resistance in cell lines derived from melanoma metastases. Mol Cancer Ther 2006; 5: 113644.
  • 40
    Dankort D, Curley DP, Cartlidge RA, Nelson B, Karnezis AN, Damsky WE,Jr, You MJ, DePinho RA, McMahon M, Bosenberg M. Braf(V600E) cooperates with Pten loss to induce metastatic melanoma. Nat Genet 2009; 41: 54452.
  • 41
    Solit DB, Garraway LA, Pratilas CA, Sawai A, Getz G, Basso A, Ye Q, Lobo JM, She Y, Osman I, Golub TR, Sebolt-Leopold J, et al. BRAF mutation predicts sensitivity to MEK inhibition. Nature 2006; 439: 35862.
  • 42
    Chaisuparat R, Hu J, Jham BC, Knight ZA, Shokat KM, Montaner S. Dual inhibition of PI3Kalpha and mTOR as an alternative treatment for Kaposi's sarcoma. Cancer Res 2008; 68: 83618.
  • 43
    Heitman J, Movva NR, Hall MN. Targets for cell cycle arrest by the immunosuppressant rapamycin in yeast. Science 1991; 253: 9059.
  • 44
    Zhang TT, Okkenhaug K, Nashed BF, Puri KD, Knight ZA, Shokat KM, Vanhaesebroeck B, Marshall AJ. Genetic or pharmaceutical blockade of p110delta phosphoinositide 3-kinase enhances IgE production. J Allergy Clin Immunol 2008; 122: 81119 e2.
  • 45
    Wei G, Twomey D, Lamb J, Schlis K, Agarwal J, Stam RW, Opferman JT, Sallan SE, den Boer ML, Pieters R, Golub TR, Armstrong SA. Gene expression-based chemical genomics identifies rapamycin as a modulator of MCL1 and glucocorticoid resistance. Cancer Cell 2006; 10: 33142.
  • 46
    Calastretti A, Rancati F, Ceriani MC, Asnaghi L, Canti G, Nicolin A. Rapamycin increases the cellular concentration of the BCL-2 protein and exerts an anti-apoptotic effect. Eur J Cancer 2001; 37: 21218.
  • 47
    Ivanov VN, Bhoumik A, Ronai Z. Death receptors and melanoma resistance to apoptosis. Oncogene 2003; 22: 315261.
  • 48
    Zhang YW, Wang LM, Jove R, Vande Woude GF. Requirement of Stat3 signaling for HGF/SF-Met mediated tumorigenesis. Oncogene 2002; 21: 21726.
  • 49
    Niu G, Shain KH, Huang M, Ravi R, Bedi A, Dalton WS, Jove R, Yu H. Overexpression of a dominant-negative signal transducer and activator of transcription 3 variant in tumor cells leads to production of soluble factors that induce apoptosis and cell cycle arrest. Cancer Res 2001; 61: 327680.
  • 50
    Ivanov VN, Krasilnikov M, Ronai Z. Regulation of Fas expression by STAT3 and c-Jun is mediated by phosphatidylinositol 3-kinase-AKT signaling. J Biol Chem 2002; 277: 493244.

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article

IJC_24926_sm_SuppFig1.tif19801KSupporting Figure 1
IJC_24926_sm_SuppFig2.tif19802KSupporting Figure 2

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.