The PI3K/Akt and mTOR/P70S6K Signaling Pathways in Human Uveal Melanoma Cells: Interaction with B-Raf/ERK

Abstract

Purpose.: Activated B-Raf alone cannot induce melanoma but must cooperate with other signaling pathways. The phosphatidylinositol 3-kinase (PI3K)/Akt and mammalian target of rapamycin (mTOR)/p70S6K pathways are critical for tumorigenesis. The authors investigated the role of these pathways in uveal melanoma cells.

Methods.: The effects of PI3K and mTOR activation and inhibition on the proliferation of human uveal melanoma cell lines expressing either activated ^WTB-Raf or ^V600EB-Raf were investigated. Interactions among PI3K, mTOR, and B-Raf/ERK were studied.

Results.: Inhibition of PI3K deactivated P70S6 kinase, reduced cell proliferation by 71% to 84%, and increased apoptosis by a factor of 5.0 to 8.4 without reducing ERK1/2 activation, indicating that ERK plays no role in mediating PI3K in these processes. In contrast, rapamycin-induced inhibition of mTOR did not significantly affect cell proliferation because it simultaneously stimulated PI3K/Akt activation and cyclin D1 expression. Regardless of B-Raf mutation status, cotreatment with the PI3K inhibitor effectively sensitized all melanoma cell lines to the B-Raf or ERK1/2 inhibition-induced reduction of cell proliferation. B-Raf/ERK and PI3K signaling, but not mTOR signaling, converged to control cyclin D1 expression. Moreover, p70S6K required the activation of ERK1/2. These data demonstrate that PI3K/Akt and mTOR/P70S6K interact with B-Raf/ERK.

Conclusions.: Activated PI3K/Akt attenuates the inhibitory effects of rapamycin on cell proliferation and thus serves as a negative feedback mechanism. This finding suggests that rapamycin is unlikely to inhibit uveal melanoma growth. In contrast, targeting PI3K while inhibiting B-Raf/ERK may be a promising approach to reduce the proliferation of uveal melanoma cells.

Uveal melanoma is the most common primary ocular neoplasm in adults in developed countries. Unlike cutaneous melanoma, which has been extensively studied, little is known about the molecular pathogenesis of this uveal cancer. Activating mutations in BRAF that lead to constitutive activation of ERK (extracellular signal-regulated kinase) are found in many primary tumors. ¹ Constitutive ERK1/2 activation has been shown in primary tumors and cell lines of uveal melanoma that express both ^WTRas and ^WTB-Raf. ^2–4 We previously showed that the B-Raf/MEK/ERK signaling pathway for cell proliferation and transformation plays the same role in human uveal melanoma cells lines that express ^WTB-Raf and in the rare human uveal melanoma cell lines that express the activating V600E mutation in B-Raf (^V600EB-Raf). ⁵ We also demonstrated that either siRNA-mediated deletion of B-Raf or BAY43–9006-mediated inhibition of B-Raf inhibits ERK1/2 activation, greatly reduces cyclin D1 expression, and almost abolishes cell proliferation and transformation in both ^WTB-Raf and ^V600EB-Raf uveal melanoma cell lines. ⁵ These findings highlight the role of the B-Raf/ERK1/2 signaling pathway in these cells. Activation of this signaling pathway is attributed to growth factor autocrine loops that participate in cell proliferation and transformation in ^WTB-Raf-expressing uveal melanoma cell lines. ^6,7 Even though B-Raf activation is an essential event in melanoma development, activated B-Raf alone is not sufficient to induce melanoma tumorigenesis; it must cooperate with other signaling pathways to induce the fully cancerous state. ⁸ One of the growth factor autocrine loops that induces ERK1/2 activation in uveal melanoma cell lines is the SCF-dependent activation of c-Kit. ⁷ Interestingly, SCF also activates phosphatidylinositol 3-kinase (PI3K) for tumorigenesis. Little is known about the role of the PI3K signaling pathway in uveal melanoma. Surprisingly, cell treatment with the PI3K inhibitor LY294002 almost totally inhibits cell proliferation in OCM-1, a ^V600EB-Raf uveal melanoma cell line that constitutively activates the B-Raf/ERK pathway and depends on it for proliferation. ⁹ These studies thus suggest that B-Raf activation is necessary but not sufficient for cell proliferation and that the PI3K signaling pathway may cooperate with B-Raf in uveal melanoma cells. 

PI3K triggers a network that positively regulates G1/S cell cycle progression and increases cyclin D1 expression. ¹⁰ The PI3K signaling pathway plays a major role in the tumorigenesis of cutaneous melanoma. ^11,12 The activation of one of its downstream kinases, mammalian target of rapamycin (mTOR), in most cutaneous melanomas confirms its key role. ¹³ Because mTOR integrates growth factor stimulation with protein synthesis and cell growth, its inhibition by rapamycin has been tested in several phase I and II trials in various cancers, including cutaneous melanoma. ^14,15 Moreover, substantial evidence indicates that the B-Raf/ERK axis and the multiple and separate downstream effectors of PI3K cooperate in the tumorigenesis of cutaneous melanoma. Combined targeting of ERK1/2 by UO126 and of PI3K by LY294002 greatly reduces the proliferation of cutaneous melanoma cell lines. ¹⁶ Combined treatment with BAY43–9006 and rapamycin, which respectively inhibit B-Raf and mTOR, synergistically inhibits the proliferation of cutaneous melanoma cells, ¹⁷ and clinical testing of these compounds for this melanoma is under way. 

For these reasons, we studied the role of PI3K and mTOR on the cell proliferation mediated by the B-Raf/ERK signaling pathway in human uveal melanoma cell lines. We demonstrated that inhibition of PI3K, in contrast to the rapamycin-induced inhibition of mTOR, results in almost total inhibition of cell proliferation, regardless of the mutational status of the activated B-Raf. PI3K cooperates with the B-Raf/ERK signaling pathway for cell proliferation by controlling cyclin D1 expression. These findings suggest that targeting PI3K, alone or in combination with the B-Raf/ERK signaling pathway, may prove to be an effective strategy against uveal melanoma cell proliferation. 

Materials and Methods

Cell Cultures and Treatments

We grew four cell lines, 92.1 and Mel270 (both provided by Martine Jager, University of Leiden, the Netherlands), OCM-1 (provided by François Malecaze, Ophthalmology Department, CHU Toulouse, France), and TP31 (provided by Sylvain Guerin, Centre Hospitalo-Universitaire de Quebec, Canada) in RPMI 1640 medium supplemented with 5% FCS, 2.5 μg/mL Fungizone/amphotericin B, 50 μg/mL gentamicin, and 2 mM l-glutamine (Gibco/BRL, Invitrogen, Cergy Pontoise, France), as previously described. ⁶ Cells were cultured at 37°C in a humidified air/CO₂ (19:1) atmosphere. 

Cells were seeded in triplicate in 24-well plates at a density of 15,000 cells/well. The plates were incubated for 3 days and were treated with the MEK inhibitor UO126 (Calbiochem, San Diego, CA), the B-Raf inhibitor BAY43–9006 (Bayer, St. Gallen, Switzerland), the PI3K inhibitor LY294002 (Calbiochem), PI3K inhibitor 2 (inhibitor of the PI3K α and β isoforms; Cayman Chemical, Detroit, MI), or the mTOR inhibitor rapamycin (Sigma-Aldrich, Lyon, France). Inhibitors were added 1 hour before the induction of cell proliferation and at induction. This cell proliferation was induced 3 days after seeding by replacement of the culture medium with fresh medium and was assessed by counting the cells remaining in the culture dish after staining with trypan blue and by the 3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) colorimetric method. The number of trypan blue-positive cells never exceeded 3% of the total cell population. Cells were examined by phase-contrast microscopy before treatment and before MTT assay to measure cell death. The percentage of growth inhibition was calculated relative to control DMSO-treated cells. 

Cell Cycle Progression Analysis

We analyzed cell cycle progression by determining the cell DNA content with propidium iodide. Cells were washed in PBS and fixed in ice-cold 70% ethanol by incubation at 4°C for at least 2 hours. The cells were rehydrated in cold PBS, treated with 1 mg/mL RNase A (Boehringer Mannheim, Mannheim, Germany), and stained with 50 μg/mL propidium iodide by incubation at 4°C for at least 15 minutes. The stained cells were analyzed by flow cytometry (Epics ALTRA; Beckman Coulter, Villepinte, France). 

Western Blot Analysis

Cells were washed twice in PBS, lysed in ice-cold lysis buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.1% NP-40, 1% deoxycholate, 50 mM β-glycerophosphate, 0.2 mM sodium orthovanadate, 50 mM sodium fluoride, 1 μg/mL leupeptin, 5 μM pepstatin, 20 kIU/mL aprotinin, and 1 mM PMSF) and then were centrifuged for 10 minutes at 10,000g at 4°C. Protein concentrations were determined with the Bio-Rad (Hercules, CA) kit. Cell lysates were mixed with 3× Laemmli buffer and heated for 5 minutes at 95°C. They were then resolved by SDS-PAGE (10% or 15% polyacrylamide gels), electroblotted onto PVDF membrane (Immobilon; Millipore, Billerica, MA), and probed with polyclonal antibodies directed against ERK1/2 (dilution 1:1000), MEK1/2 (dilution 1:1000), Akt (dilution 1:1000), p70S6K (dilution 1:1000) (Cell Signaling Technology, Beverly, MA), cyclin D1 (dilution 1:1000; Thermo Fisher Scientific, Illkirch, France), GSK3α and β (dilution 1:1000; Euromedex, Souffelweyersheim, France), the p85α regulatory subunit of PI3K (dilution 1:500), and the p110β catalytic subunit of PI3K (dilution 1:1000) (both from Santa Cruz Biotechnology, Santa Cruz, CA). We used a polyclonal antibody directed against phospho-Akt (Ser473; dilution 1:1000), phospo-ERK1/2 (Thr202/Tyr204, dilution 1:1000), phospho-GSK3α/β (Ser21/Ser9; dilution 1:1000), and phospho-p70S6K (Thr421/Ser424, Thr389; dilution 1:1000) (Cell Signaling Technology) to analyze the activation of these kinases during melanoma cell proliferation. Membranes were probed with a goat antibody directed against actin (dilution 1:1000; Santa Cruz Biotechnology) to control for equal loading. The primary antibodies were tagged with specific secondary horseradish peroxidase-conjugated antibodies. Antibody complexes were detected by enhanced chemiluminescence (ECL; Amersham/GE Healthcare, Piscataway, NJ), and the membrane was placed against Kodak film (BioMax Light-1; Paris, France). Quantification was conducted with a Kodak image station (2000 MM) and software (1D3.6). 

Statistical Analysis

Two-tailed Student's t-tests (normal distributions with equal variances) and Mann-Whitney (nonparametric) tests were used for statistical analysis. Data were expressed as mean ± SD. 

Results

Key Role of PI3K in Controlling Cell Proliferation in ^WTB-Raf Uveal Melanoma Cells

Western blot analysis showed that the ^WTB-Raf uveal melanoma cell lines 92.1 and Mel270 expressed p85α and p110β (Fig. 1A). The two cell lines expressed similar levels of Akt and GSK3 (Fig. 1B). Constant phosphorylation/activation of Akt occurred over a 6-hour culture period in melanoma cells (Fig. 1B). Consistent with this activation, constant phosphorylation/inactivation of GSK3 also occurred in melanoma cells (Fig. 1B). Activation of Akt and inhibition of GSK3 depended on PI3K in ^WTB-Raf uveal melanoma cell lines because cell treatment with LY294002, an inhibitor of class I PI3K, greatly reduced both processes in 92.1 and Mel270 (Fig. 1B). 

Figure 1.

 

PI3K controls cell proliferation and apoptosis. (A) Western blot analysis of p85α subunit of PI3K, p110β subunits of PI3K, Akt, GSK3, and actin in ^WTB-Raf cells (92.1 and Mel270) after 48 hours of culture. (B) 92.1 cells were treated with 30 μM LY294002 or remained untreated and were lysed at various times after serum stimulation. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-phospho Akt, anti-phospho GSK3, and anti-actin antibodies. (C) 92.1 and Mel270 cells were treated with the indicated concentrations of LY294002 or remained untreated, and cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. SDs are not visible if they are smaller than the size of the cell-type symbol. (D) Cell death of 92.1 cells was analyzed by fluorescence-activated cell sorting after 48 hours of culture with 30 μM LY294002. Results in both sets of ^WTB-Raf cells (92.1 and Mel270) were similar (data not shown). Similar results were obtained in five (A, C), four (B), and three (D) independent experiments.

Then we used the MTT method to investigate the effects of PI3K inhibition on cell proliferation of these uveal melanoma cell lines. Cell treatment with the PI3K inhibitor LY294002 reduced proliferation in a concentration-dependent manner in ^WTB-Raf uveal melanoma cell lines (Fig. 1C). Maximum inhibition of cell proliferation (80%–84% inhibition), observed with 30 μM LY294002, did not differ significantly between the two ^WTB-Raf uveal melanoma cell lines (Fig. 1C). We quantified the efficacy of LY294002 by determining the concentration of the PI3K inhibitor needed to inhibit cell proliferation by 50% (IC₅₀). After 72 hours of culture, the IC₅₀ values were similar in both ^WTB-Raf melanoma cell lines (IC₅₀ of 16 μM for 92.1 and 18 μM for Mel270). Direct cell counts showed similar inhibition of cell proliferation by LY294002 (data not shown). Treatment of uveal melanoma cell lines with the PI3K inhibitor PI3Kα inhibitor 2, which inhibits PI3Kα and PI3Kβ more specifically than the PI3Kγ isoform, also greatly reduced cell proliferation in a concentration-dependent manner. Maximum inhibition of cell proliferation (89%–93% inhibition) was observed at a concentration of 10 μM, with no significant difference between the two ^WTB-Raf uveal melanoma cell lines (data not shown). The IC₅₀ values were also similar in both ^WTB-Raf melanoma cell lines after culturing for 72 hours (IC₅₀ of 1.2 μM for 92.1 and 2.0 μM for Mel270) (data not shown). Flow cytometry analysis showed that cells underwent apoptosis (sub-G1 peak) after 48 hours of LY294002 treatment, with cell death rates increasing by a factor of 5.0 in the ^WTB-Raf melanoma cell line Mel270 (Fig. 1D). 

Absence of a Role for mTOR/P70S6K Signaling in Cell Proliferation

The kinase mTOR is central in the PI3K class 1A signaling pathway, which is involved in the translational control of cell cycle progression regulators and thus of cell proliferation. ^18–20 Therefore, we started by analyzing the expression and activation levels of members of the mTOR signaling pathway. Western blot analysis showed that the two ^WTB-Raf uveal melanoma cell lines expressed different members of the Rheb/mTOR/p70S6K/S6 signaling pathway (Fig. 2A). Analysis of the kinetics of p70S6K phosphorylation/activation showed constant p70S6K phosphorylation over a 24-hour culture period in uveal melanoma cells (Fig. 2B) and suggested that p70S6K signaling is constitutively activated in these cells. 

Figure 2.

 

Effects of the inhibition of the mTOR/p70S6K signaling pathway in cell proliferation and apoptosis. (A) Western blot analysis of Rheb, mTOR, p70S6K, S6, and actin in ^WTB-Raf cells (92.1 and Mel270) after 48 hours of culture. (B) 92.1 cells were treated with 30 μM LY294002 or remained untreated and were lysed at various times after serum stimulation. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-p70S6K and anti-phospho p70S6K antibodies. (C) 92.1 cells were treated with 50 nM rapamycin or remained untreated and were lysed at various times after serum stimulation. Equal amounts of protein extracts were analyzed by Western blot analysis with the anti-p70S6K and anti-actin antibodies. (D) 92.1 and Mel270 cells were treated with the indicated concentrations of rapamycin or remained untreated, and cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. SDs are not visible if they are smaller than the size of the cell-type symbol. (E) Cell cycle and (F) cell death of 92.1 cells were analyzed by fluorescence-activated cell sorting after 48 hours of culture with 250 nM rapamycin. Results in both sets of ^WTB-Raf cells (92.1 and Mel270) were similar (data not shown). Similar results were obtained in four (A–C), five (D), and three (E, F) independent experiments.

We hypothesized that the constitutive activation of PI3K might cause this p70S6K activation. To confirm this hypothesis, we analyzed the levels of phosphorylated/activated p70S6K expression in uveal melanoma cells treated with the PI3K inhibitor LY294002 at 30 μM. Inhibition of PI3K induced the rapid (within 10 minutes of treatment), prolonged (duration up to 24 hours), and complete inhibition of p70S6K activation in these cells (Fig. 2B). These data demonstrate that the mTOR/p70S6K signaling pathway lies downstream of PI3K in ^WTB-Raf uveal melanoma cell lines. 

Next, on speculation that mTOR/p70S6K mediates PI3K signaling for cell proliferation in ^WTB-Raf uveal melanoma cells, we used the MTT method to investigate the effects of rapamycin on this proliferation. Rapamycin, a lactone with antitumor properties resulting from its inhibition of mTOR, inhibits the proliferation of many types of tumoral cells. ¹⁸ We first confirmed that mTOR controlled p70S6K activation by showing that rapamycin (50 nM) completely inhibited the phosphorylation of p70S6K in melanoma cell lines within 15 minutes of treatment (Fig. 2C). Surprisingly, mTOR inhibition had a fairly small effect on the proliferation of the uveal melanoma cells: cell treatment with rapamycin (50 nM) at the concentration that reduced p70S6K phosphorylation levels by 100% reduced proliferation by only 9% in Mel270 and 21% in 92.1 (Fig. 2D). Even at the high concentration of 250 nM, rapamycin reduced uveal melanoma cell proliferation only moderately (10% and 25%, respectively; Fig. 2D). Additional treatment with rapamycin on day 3 of the culture did not modify cell proliferation (or its inhibition), as assessed on day 5 of the culture (data not shown). Direct cell counts showed similar inhibition of cell proliferation by rapamycin (data not shown). Flow cytometry analysis showed that mTOR inhibition by 250 nM rapamycin did not affect cell distribution throughout the cell cycle phases (Fig. 2E). Rapamycin did not induce significant apoptosis (sub-G1 peak; Fig. 2F). 

Attenuation of the Rapamycin-Induced Inhibition of Cell Proliferation by Activated PI3K

We next studied the mechanisms of activation of the mTOR/P70S6K pathway and of the paradoxic effects of rapamycin in uveal melanoma cells. We hypothesized that the relative insensitivity of uveal melanoma cells to rapamycin might be caused by resistance mechanisms that bypass the inhibition of proliferation caused by rapamycin-induced mTOR/p70S6K inhibition. There is evidence that activation of mTOR inhibits PI3K/Akt in some circumstances: rapamycin-induced inhibition of mTOR may enhance PI3K activation by an mTOR-dependent negative feedback mechanism for PI3K/Akt activation, at least in a few types of cells. ^21–23 Paradoxically, then, rapamycin, which inhibits mTOR/p70S6K-mediated cell proliferation signaling, concurrently increases Akt phosphorylation and thus increases cell survival and proliferation. We speculated that this might be the case for uveal melanoma cells. Analysis of Akt phosphorylation levels in rapamycin-treated cells showed that inhibition of mTOR greatly increased Akt phosphorylation without affecting Akt levels in uveal melanoma cell lines (Fig. 3A). This phenomenon was observed at a concentration as low as 10 nM rapamycin (Fig. 3A). Surprisingly, rapamycin treatment also increased levels of cyclin D1 expression in ^WTB-Raf melanoma cell lines (Fig. 3B). Given our previous finding that cyclin D1 mediates cell proliferation signaling in uveal melanoma cells, ⁵ these data suggest that the reduction of cell proliferation induced by the rapamycin-mediated mTOR inhibition may be counteracted by the activation of a proliferation signaling pathway mediated by cyclin D1 and a survival signaling pathway mediated by Akt. 

Figure 3.

 

mTOR-mediated feedback to Akt activation and cyclin D1 expression: reversion by PI3K inhibition. (A) 92.1 cells were treated with 10 or 50 nM rapamycin or remained untreated over a 24-hour culture period, and then equal amounts of protein extracts were analyzed by Western blot analysis with anti-Akt, anti-phospho Akt and anti-cyclin D1 antibodies. (B) 92.1 cells were treated with 50 nM rapamycin (R), 30 μM LY294002 (LY), or a combination of both inhibitors (R+LY) over a 24-hour culture period, and then cells were lysed. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-phospho Akt, anti-cyclin D1, and anti-actin antibodies. (C) 92.1 cells were treated with 50 nM rapamycin (R), 30 μM LY294002 (LY), or a combination of both inhibitors (R+LY), and cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. *P ≤ 0.05. Results in both sets of ^WTB-Raf cells (92.1 and Mel270) were similar (data not shown). Similar results were obtained in four (A) and three (B, C) independent experiments.

If our hypothesis about the mTOR feedback mechanism on PI3K/Akt activation is correct, PI3K inhibition would reduce the rapamycin-induced phosphorylation of Akt. Therefore, we used Western blot analysis to examine the effects of cell pretreatment with LY294002 on the rapamycin-induced phosphorylation/activation of Akt. As expected, we found that PI3K inhibition greatly reduced rapamycin-induced Akt phosphorylation (Fig. 3B), thus confirming that rapamycin-induced Akt activation requires PI3K activation. Similarly, the failure of rapamycin to increase cyclin D1 expression when PI3K was inhibited (Fig. 4B) demonstrated that rapamycin-induced cyclin D1 expression also requires the activation of PI3K. Consequently, seeking further support for this model, we hypothesized that PI3K would counteract the proliferative/survival effects of mTOR inhibition and thus expected that blocking PI3K by LY294002 would enhance the inhibitory effect of rapamycin on cell proliferation. As expected, 10 μM LY294002 (a concentration of inhibitor that reduced cell proliferation by only 24%–36%; Figs. 1C, 3C), combined with 50 nM rapamycin (a concentration of inhibitor that reduced cell proliferation by 9%–21%; Figs. 2D, 3C) inhibited cell proliferation significantly (P ≤ 0.05) more (cell proliferation reduced by 69%–77%) than either rapamycin or LY29002 alone and more than the sum of the inhibitory effects caused by each agent alone (Fig. 3C). The combination thus produced synergistic effects. Direct cell counts showed similar synergistic effects on the inhibition of cell proliferation by a combination of rapamycin and LY294002 (data not shown). Taken together, these data indicate that mTOR exerts a feedback mechanism on the PI3K/Akt-mediated cell proliferation signaling pathway that can be reversed by the inhibition of PI3K in ^WTB-Raf uveal melanoma cells. 

Figure 4.

 

B-Raf/ERK inhibition sensitizes uveal melanoma cells to inhibition of PI3K and of mTOR. (A) 92.1 cells were treated with 30 μM LY294002 or remained untreated and were lysed at various times after serum stimulation. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-ERK1/2 and anti-phospho ERK1/2 antibodies. (B) 92.1 cells were cultured for 3 days and then treated with 10 μM LY294002 (LY) alone and in combination with 5 μM UO125 (UO), or with 1 μM BAY43–9006 (BAY). Cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. *P ≤ 0.05. (C) 92.1 cells were treated with 30 μM LY294002 or remained untreated and were lysed at various times after serum stimulation. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-cyclin D1 and anti-actin antibodies. (D) 92.1 cells were treated with 20 μM UO125 or remained untreated and were lysed at various times after serum stimulation. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-ERK1/2 and anti-phospho ERK1/2, anti-cyclin D1, anti-Akt, anti-phospho Akt, and anti-p70S6K, and anti-phospho p70S6K antibodies. (E) 92.1 cells were treated with 50 nM rapamycin (R), UO125 (UO) at a concentration of 5 μM (+), or 20 μM (++) alone or treated with a combination of UO125 (5 μM) plus rapamycin (50 nM). Then cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. *P ≤ 0.05. Results in both sets of ^WTB-Raf cells (92.1 and Mel270) were similar (data not shown). Similar results were obtained in three independent experiments.

Cooperation of PI3K with B-Raf/ERK to Regulate Cyclin D1-Dependent Cell Proliferation

Class I PI3Ks have been shown to activate ERK1/2. ^24–26 Because ERK1/2 controls cell proliferation in uveal melanoma cells, we examined whether LY294002-induced the inhibition of PI3K reduced cell proliferation by deactivating ERK1/2. Cell treatment with 30 μM LY294002, a concentration that reduced cell proliferation by 80% to 84%, did not affect the level of ERK1/2 phosphorylation/activation or of ERK1/2 expression (Fig. 4A). This shows that ERK1/2 does not mediate PI3K signaling. 

We then investigated whether B-Raf/ERK signaling cooperates with PI3K to control cell proliferation by treating cells with UO126 (5 μM) or with LY294002 (10 μM) alone, at low concentrations corresponding approximately to their respective IC₅₀, or with a combination of both inhibitors. The combination was twice as effective as either inhibitor alone at inhibiting serum-stimulated melanoma cell proliferation (Fig. 4B). This significant decrease (P ≤ 0.05) in the concentration required for maximal inhibition of cell proliferation suggests that PI3K inhibition sensitized uveal melanoma cell lines to ERK1/2 inhibition. The B-Raf inhibitor BAY43–9006, which efficiently inhibits ERK1/2-mediated proliferation of uveal melanoma cells, ⁵ is still being tested in clinical trials for various cancers, including melanoma. Therefore, we tested whether LY294002-mediated inhibition of PI3K sensitized uveal melanoma cell lines to BAY43–9006. As expected, similar additive effects were observed for low concentrations of LY294002 (10 μM) and BAY43–9006 (1 μM) in ^WTB-Raf uveal melanoma cell lines (Fig. 4B). These data show that the PI3K and B-Raf/ERK signaling pathways are parallel pathways that cooperate in the control of uveal melanoma cell proliferation. 

We previously showed that ERK1/2 controls the expression of cyclin D1, the inhibition of which greatly reduces the proliferation of uveal melanoma cells. ⁵ Therefore, we investigated whether cyclin D1 is a common effector of both PI3K and ERK1/2 in uveal melanoma cells. Treatment with 30 μM LY294002 induced rapid and prolonged inhibition of cyclin D1 expression in ^WTB-Raf uveal melanoma cells (Fig. 4C). This shows that PI3K, as well as Raf/ERK1/2, controls cyclin D1 regulation of uveal melanoma cell proliferation. 

We then tested the possibility that ERK1/2 regulates the Akt or the mTOR/p70S6K signaling pathways in uveal melanoma cells. As expected, treatment with UO126 (20 μM) rapidly and completely inhibited ERK1/2 activation without affecting its expression (Fig. 4D). We confirmed that the inhibition of ERK1/2 induced a major rapid and prolonged reduction of cyclin D1 expression (Fig. 4D) without deactivating Akt (Fig. 4D). This suggests that any interaction of ERK1/2 with the PI3K signaling pathway lies downstream of Akt. We were particularly interested in the possibility that p70S6K activity might be regulated by ERK1/2 because cross-talk between ERK1/2 and mTOR/p70S6K has been shown in some circumstances. ^27–30 This was indeed the case for uveal melanoma cells: ERK1/2 inhibition rapidly and strongly deactivated p70S6K, even though total p70S6K expression was not affected (Fig. 4D). 

Next we investigated whether ERK1/2 inhibition might sensitize cells to the very modest inhibitory effect that rapamycin has on their proliferation. Thus, cells were treated with a concentration of UO126 corresponding to its IC₅₀ (5 μM), rapamycin at the concentration of its maximal inhibitory effect (50 μM), or a combination of both. The inhibitory effects on cell proliferation were compared with those of UO126 at the concentration that induces its maximal inhibitory effects (20 μM). A combination of the two inhibitors was significantly (P ≤ 0.05) more effective than either inhibitor alone in inhibiting serum-stimulated melanoma cell proliferation (Fig. 4E). The inhibitory effects of 5 μM UO126 in the presence of rapamycin were similar to those obtained with 20 μM UO126, suggesting that mTOR inhibition sensitized uveal melanoma cell lines to UO126-mediated inhibition of ERK1/2. 

Role of the V600E Mutation in Uveal Melanoma Cell Sensitivity to PI3K Inhibition or mTOR-Mediated Feedback Mechanism on PI3K/Akt Signaling for Cell Proliferation

A recent study found a higher prevalence of activated mutant ^V600EB-Raf in uveal melanoma than previously reported. ³¹ The strong ^V600EB-Raf-dependent cell proliferation of uveal melanoma cell lines compared with normal uveal melanocytes ² suggests that melanoma cells with this mutation are less dependent on other signaling pathways involved in cell proliferation than the normal uveal melanocytes and uveal melanoma cells that express ^WTB-Raf. Therefore, we compared the effects of PI3K and mTOR inhibition on cell proliferation in the ^V600EB-Raf in the OCM-1 and TP31 uveal melanoma cell lines. Constitutive activation of the PI3K/Akt and mTOR/p70S6K signaling pathways occurred in ^V600EB-Raf melanoma cells (data not shown). Cell treatment with LY294002 reduced proliferation by similar percentages in the ^V600B-Raf and ^WTB-Raf uveal melanoma cell lines (Fig. 5A), and, after 72 hours of culture, the IC₅₀ values were similar in both ^WTB-Raf and ^V600EB-Raf melanoma cell lines (IC₅₀ of 16–18 μM for ^WTB-Raf cells and IC₅₀ of 18–19 μM for ^V600EB-Raf cells). Inhibition of cell proliferation by rapamycin was equally low in both (Fig. 5B). 

Figure 5.

 

Role of the V600E mutation in B-Raf on the sensitivity of uveal melanoma cells to PI3K and mTOR inhibition. ^WTB-Raf cells (92.1 and Mel270) and ^V600EB-Raf cells (OCM-1 and TP31) were cultured for 3 days. Then cells were treated with the indicated concentrations of LY294002 (A) or rapamycin (B) or remained untreated, and cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. SDs are not visible if they are smaller than the size of the cell-type symbol. (C) OCM-1 cells were treated with 50 nM rapamycin (R), 30 μM LY294002 (LY), or a combination of both inhibitors (R+LY), and cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. (D) OCM-1 cells were treated with 50 nM rapamycin (R), 30 μM LY294002 (LY), or a combination of both inhibitors (R+LY) over a 24-hour culture period, and then cells were lysed. Equal amounts of protein extracts were analyzed by Western blot analysis with anti-phospho Akt, anti-cyclin D1, and anti-actin antibodies. (E) Cells were treated with UO125 (UO, 5 μM) or LY294002 (LY, 30 μM) alone or with a combination of UO125 (UO, 5 μM) plus LY294002 (LY, 30 μM). Cell proliferation was measured with the MTT colorimetric assay on day 3 of treatment. Results in both sets of ^WTB-Raf cells (92.1 and Mel270) were similar, as were the results for both sets of ^V600EB-Raf cells (OCM-1 and TP31) (data not shown). *P ≤ 0.05. Similar results were obtained in five (A, B) and three (C–E) independent experiments.

Moreover, mTOR exerts a feedback mechanism on the PI3K/Akt-mediated cell proliferation signaling pathway that can be reversed by the inhibition of PI3K in ^V600EB-Raf uveal melanoma cells: 10 μM LY294002 combined with 50 nM rapamycin inhibited cell proliferation significantly (P ≤ 0.05) more than either rapamycin or LY29002 alone and more than the sum of the inhibitory effects caused by each alone (Fig. 5C). PI3K inhibition greatly reduced rapamycin-induced Akt phosphorylation (Fig. 5D), thus showing that rapamycin-induced Akt activation requires PI3K activation. Similarly, the failure of rapamycin to increase cyclin D1 expression when PI3K was inhibited (Fig. 5D) demonstrated that rapamycin-induced cyclin D1 expression also required the activation of PI3K. Finally, PI3K inhibition sensitized ^V600EB-Raf uveal melanoma cell lines to UO126- and BAY43–9006-mediated inhibition of the B-Raf/ERK signaling because additive effects were observed for low concentrations of LY294002 (10 μM) and UO125 (5μM) or BAY43–9006 (1 μM) in the ^V600EB-Raf uveal melanoma cell lines (Fig. 5E). Together these data showed that the inhibition of PI3K reduced cell proliferation substantially and similarly in ^WTB-Raf and ^V600EB-Raf uveal melanoma cell lines and that the PI3K and B-Raf/ERK signaling pathways are parallel pathways that converge at the level of cyclin D1 and cooperate in the control of uveal melanoma cell proliferation, regardless of the B-Raf mutational status. A proposed scheme for the cross-talk among the activated PI3K, mTOR/p70S6K, and B-Raf/ERK signaling pathways in uveal melanoma cells is presented in Figure 6. 

Figure 6.

 

Schematic diagram of the interaction between the ERK and the PI3K/mTOR pathways. The B-Raf/ERK and the PI3K/mTOR signaling pathways interact for the regulation of cyclin D1 expression and Akt phosphorylation and for the control of cell proliferation in uveal melanoma cells. Arrows indicate activation but do not necessarily represent direct interactions, whereas bars represent inhibition. Cyclin D1 expression is coregulated by the PI3K and ERK pathways (large white arrows). The mTOR-mediated feedback inhibition of Akt is responsible for the small inhibitory effect that rapamycin has on cell proliferation (thin black bar), whereas ERK1/2 positively controls the activation of p70S6K (thick black arrow). The tuberous sclerosis complex (TSC) gene products lie between the PI3K/Akt and Rheb/mTOR signaling pathways. Alteration of TSC is a model for the mTOR feedback mechanism in some cancer cells. To identify TSC1/2 as the connection between PI3K/Akt and mTOR signaling in uveal melanoma cells, the expression and activation levels of TSC must be characterized in these cells.

Discussion

No oncogene or tumor suppressor has yet been convincingly linked to uveal melanoma. ERK1/2 is activated in this disease, ³² and it is well documented that B-Raf/ERK1/2 signaling is responsible for the anchorage-dependent and -independent proliferation of uveal melanoma cells that express ^WTRas/^WTB-Raf. ^5–7 Activation of the PI3K signaling pathway has not been studied, however, and it is not known what role PI3K plays in uveal melanoma, though activated Akt has been observed in some of these tumors. ³³  

Our study is the first to show the constant phosphorylation/activation of Akt in uveal melanoma cell lines expressing both ^WTRas/^WTB-Raf (92.1 and Mel270) and ^WTRas/^V600EB-Raf (OCM-1 and TP31). The expression by all these cell lines of the activated B-Raf/ERK signaling pathway shows that both signaling pathways are simultaneously activated in uveal melanoma. Although loss of PTEN expression from submicroscopic deletion has been observed in a few uveal melanomas, ³⁴ this gene does not appear to be absent, structurally altered, or somatically mutated in OCM-1, 92.1, or Mel270. ³⁵ Therefore, the constitutive activation of PI3K/Akt does not appear to result from changes in this gene. Moreover, the failure to detect any activating RAS mutations in uveal melanoma ^36,37 and the relative insensitivity of uveal melanoma cell proliferation to Ras inhibition ² together suggest that Ras plays no role in the activation of PI3K signaling. In contrast, IGF-1, SCF, and FGF2 are known to be involved in the proliferation or survival of uveal melanoma cell lines and in the growth of uveal melanoma. ^6,38–41 All these growth factors stimulate activation of either or both of the PI3K and ERK1/2 signaling pathways, depending on the cell type and context. Moreover, PI3K involvement in the HGF-induced migration of the ^V600EB-Raf-expressing uveal melanoma cell line, SP6.5, ⁴² confirms that uveal melanoma cells can coexpress activated PI3K and ERK1/2 signaling pathways. Because autocrine growth factor activation loops are involved in the acquisition by ^WTB-Raf uveal melanoma cells of the capacity for autonomous proliferation, constitutive activation of the corresponding receptor tyrosine kinase of these growth factors may be one part of the mechanism for activating PI3K in these cells. 

The only previous indication that PI3K might be important in the growth of uveal melanoma came from Casagrande et al., ⁹ who showed that inhibition of PI3K by LY294002 results in almost complete inhibition of cell proliferation in the OCM-1 uveal melanoma cell line. Interestingly, OCM-1 expresses the mutant ^V600EB-Raf, which leads to the constitutive activation of the B-Raf/ERK pathway and cell proliferation. ^2,5,6 This suggests that the PI3K signaling pathway may cooperate with the B-Raf/ERK pathway in uveal melanoma cells. We confirmed the role of PI3K in these cells by demonstrating that cell proliferation is substantially reduced because of the inhibition of PI3K in the ^WTB-Raf uveal melanoma cell lines 92.1 and Mel270 and in the ^V600EB-Raf uveal melanoma cell lines OCM-1 and TP31. All four cell lines express an activated B-Raf/ERK signaling pathway that also controls cell proliferation. ^2,5,6 These data strongly indicate that the PI3K and B-Raf/ERK signaling pathways are necessary for the control of cell proliferation in uveal melanoma cells. We thus speculated that the two pathways might cooperate to control cell proliferation in these cells. Cross-talk between PI3K and B-Raf/ERK was confirmed by showing additive effects of the inhibition of PI3K and either ERK1/2 or B-Raf on cell proliferation. We previously demonstrated that the B-Raf/ERK signaling pathway controls cell proliferation by regulating cyclin D1 expression. ⁵ Our present study shows that inhibition of PI3K substantially decreases cyclin D1 expression and thus demonstrates that both the PI3K and the B-Raf/ERK pathways converge to cyclin D1 to control cell proliferation in uveal melanoma. This finding, the first link between PI3K-mediated regulation of cyclin D1 expression and cell proliferation in uveal melanoma, may have implications for the inhibition of uveal melanoma cell proliferation. The overall mortality rate from uveal melanoma remains high because of the frequent development of metastases. Targeting the PI3K signaling pathway alone or with other therapeutic agents, such as BAY43–9006, an inhibitor of activated B-Raf, may be one strategy for future clinical trials in uveal melanoma. In cutaneous melanoma, both the B-Raf/ERK and the PI3K signaling pathways are constitutively activated through multiple mechanisms, including autocrine and paracrine growth factor loops, and thus exert several key functions in melanoma development and progression. Combined targeting of ERK1/2 and PI3K with pharmacologic inhibitors reduces the proliferation of cutaneous melanoma cell lines substantially, ¹⁶ a finding indicating that common intracellular signaling pathways control cell proliferation in uveal melanoma and cutaneous melanoma. 

A major downstream target of PI3K is mTOR; it regulates p70S6K, which is activated in uveal melanoma cells. Inhibition of mTOR potently inhibits tumor cells with activated PI3K/Akt signaling resulting from PTEN loss, Akt overexpression, or growth factor activation. ^43–45 We thus predicted that inhibiting mTOR and inhibiting PI3K would affect cell proliferation, as previously shown for other types of tumor cells. ^46,47 Unexpectedly, however, rapamycin-mediated inhibition of mTOR failed to inhibit the proliferation of uveal melanoma cells. Even more surprisingly, cells treated with rapamycin increased Akt activation and cyclin D1 expression. The rapamycin-induced Akt activation in uveal melanoma cells is likely to reduce its antitumor effects by activating pathways that attenuate its effects on proliferation and apoptosis. This phenomenon has been observed only once, in NSCLC cells. ²¹ The increase in Akt activation levels induced by cell treatment with rapamycin also suggests that Akt activation in our untreated uveal melanoma cells was less than maximal. This may explain why pharmacologic inhibition of Akt did not significantly affect uveal melanoma cell proliferation, ^6,7 even though activated Akt has been detected in uveal melanoma. ³³ Stimulation by the B-Raf/ERK-mediated expression of cyclin D1, which compensated for the inhibitory mTOR/p70S6K feedback to Akt, may also explain the insensitivity of uveal melanoma cells to pharmacologic inhibition of Akt. Although we demonstrated that rapamycin can induce Akt activation in uveal melanoma cells, further studies on the molecular mechanisms of the inhibitory mTOR/p70S6K feedback to Akt are needed to confirm that its potential antitumor activity may be attenuated in uveal melanoma. 

Footnotes

 Disclosure: N. Babchia, None; A. Calipel, None; F. Mouriaux, None; A.-M. Faussat, None; F. Mascarelli, None