Idelalisib Impacts Cell Growth through Inhibiting Translation-Regulatory Mechanisms in Mantle Cell Lymphoma

Abstract

Purpose: PI3K is a critical node in the B-cell receptor pathway, which is responsible for survival and proliferation of B-cell malignancies. Idelalisib, a PI3Kδ-isoform-specific inhibitor, has been approved to treat B-cell malignancies. Although biological activity of the drug has been evaluated, molecular mechanisms and signaling pathway disruption leading to the biological effects of idelalisib are not yet well defined. Prior laboratory reports have identified transcription and translation as the primary events for attenuation of PI3Kα isoform. We hypothesized that PI3Kδ-isoform inhibition by idelalisib should also affect gene transcription and protein translation.

Experimental design: Using three mantle cell lymphoma cell lines and primary cells from patients, biological consequences such as apoptosis/cell-cycle analysis, as well as RNA/protein synthesis were evaluated. Proteomics analyses (RPPA and immunoblot assays) defined molecular events downstream of PI3K/AKT cassette.

Results: Idelalisib treatment resulted in inhibition of protein synthesis, which correlated with reduction in cell size and cell growth. A moderate loss of viability without any change in cell-cycle profile was observed. Idelalisib treatment inhibited AKT activation, an immediate downstream PI3K effector, and also reduced phosphorylation levels of downstream AKT/mTOR pathway proteins such as PRAS40. In addition, idelalisib treatment impeded activation of the MAPK pathway, and MEK, ERK and p90RSK phosphorylation levels were reduced. Reduction in AKT, PDK1, and MEK phosphorylation correlated with protein synthesis inhibition.

Conclusions: Collectively, these results clarify the molecular mechanisms of actions and may provide biomarkers and targets for combination with idelalisib in B-cell malignancies. Clin Cancer Res; 23(1); 181-92. ©2016 AACR.

Figure 1. Effects of idelalisib treatment on decrease in cell numbers and cell sizes and cell death in MCL cells

Idelalisib-induced growth inhibition in three MCL cell lines JeKo-1, Mino and Granta 519. Cells were treated with DMSO or with 0.5 or 5μM idelalisib for 48h, and then the (A) cell number and (B) cell size were measured by Coulter counter and channelyzer. The values were normalized to DMSO control. Idelalisib-induced cell death in MCL cell lines and primary cells. (C) MCL cell lines (JeKo-1, Mino, and Granta 519) were treated with DMSO or with 0.5, or 5μM idelalisib for 48h, while (D) primary MCL cells (n=14) were treated with DMSO, or 0.05, 0.1, 0.5, 1, 3, or 5μM idelalisib for 24h, then stained with annexin-V/PI, and cell death levels were analyzed using a flow cytometer. Induction of autophagy after idelalisib treatment. (E) MCL cell lines were treated with DMSO or with 0.05, 0.1, 0.5, 1 or 5μM idelalisib for 24h, then stained with acridine orange, and levels of autophagy were measured by flow cytometer. All experiments in the MCL cell lines were performed in triplicate, and results are shown as means ± SEM. Student’s t-test was performed in Figure 1A, 1B, and 1C; *P<0.05 and **P<0.01, respectively. Linear mixed effects model and linear model were performed for quadratic treatment effects in Figure 1D (MCL primary cells experiments) and 1E, respectively. In Figure 1D, the idelalisib treatment resulted in a significant quadratic increase of cell death. From multiple comparison, 0.5, 1, and 5μM idelalisib treatment resulted in statistically significant increases in cell death compared with DMSO control, while induction of autophagy levels (Figure 1E) did not result in a statistically significance increase, except for 5μM treatment in Mino cells, as indicated by the asterisk.

Figure 2. Inhibition of global protein and RNA syntheses in MCL lines and primary cells by idelalisib treatment

Idelalisib-induced global protein or RNA synthesis inhibition in MCL lines. JeKo-1, Mino, and Granta 519 cells were treated with DMSO or with 0.5, or 5μM idelalisib for 48h. [³H]-leucine (2μCi/mL) or [³H]-uridine (1μCi/mL) was added to the cell media for 30–45min for co-culturing with cells before harvest, and then the incorporated [³H]-leucine or [³H]-uridine was determined using a scintillation counter to measure levels of (A) global protein or (C) global RNA synthesis. Disintegrations per minute (DPM)/cell values were calculated and normalized to DMSO control. Dose-dependent idelalisib-induced (B) global protein or (D) RNA synthesis inhibition in primary MCL cells. Patient MCL cells (n=9) were treated with DMSO or 0.05, 0.1, 0.3, 0.5, 1 or 5μM of idelalisib for 24h. Similar to the MCL cell lines, [³H]-labeled leucine (10μCi/mL) or [³H]-uridine (5μCi/mL) was added to the cell media for 30–45min before harvest, and then the incorporated [³H]-leucine or [³H]-uridine was measured for levels of global protein or RNA synthesis. Radioactivity values were also normalized to DMSO control. All cell line experiments were performed in triplicate, with results shown as means ± SEM. Student’s t-test was performed to analyze results shown; *P<0.05. From the linear mixed effects models, the idelalisib concentration and global protein synthesis level showed significant linear and quadratic relations in Figure 2B, 2D. All concentrations of idelalisib tested in Figure 2B resulted in statistically significant decreases of global protein synthesis levels compared with DMSO. However, in Figure 2D, only 0.1, 1, and 5μM idelalisib treatment resulted in statistically significant reductions of global RNA synthesis.

Figure 3. Impact of idelalisib treatment on the PI3K immediate downstream targets in MCL cells

JeKo-1, Mino, and Granta 519 cells or cells from four different MCL patients were serum-starved for 1h and then treated with DMSO, or with 0.5, 1, or 3μM for 1h; the cells were then co-cultured with IgM (10ng/μL) for 15min before harvesting. Cell extracts were prepared, and 30μg (cell lines) or 50μg (primary cells) protein was loaded for immunoblot analyses. The effects of idelalisib on Akt (Thr308) and total Akt, PDK1 (Ser241) and total PDK1, GS3K-3β (Ser9) and total GSK-3β protein expression levels were detected in (A) JeKo-1, Mino and Granta 519 cells and (B) Four MCL primary samples (MCL 12, MCL 17, MCL 20, MCL 21) were tested and representative immunoblots for specific proteins are shown. For each immunoblot panel, GAPDH loading control was measured. (C) Immunoblots from both MCL cell lines and primary cells were quantified and phospho- to total protein ratios were calculated. Values are presented as means ± SEM of independent experiments. All three cell line experiments were performed in triplicate; primary cell experiments were performed in four samples. Multiple statistical comparison of each concentration of idelalisib versus the DMSO control was performed on the results shown, and AKT (Thr308) and GSK-3β (Ser9) phosphorylation levels showed statistically significant decreases compared with vehicle control at at concentrations above 0.5 μM and at 0.5, 1 and 3μM idelalisib for phospho-PDK1. Note: the same protein loading controls appear in different figures; please refer to Supplemental Figure 4 for more details.

Figure 3. Impact of idelalisib treatment on the PI3K immediate downstream targets in MCL cells

JeKo-1, Mino, and Granta 519 cells or cells from four different MCL patients were serum-starved for 1h and then treated with DMSO, or with 0.5, 1, or 3μM for 1h; the cells were then co-cultured with IgM (10ng/μL) for 15min before harvesting. Cell extracts were prepared, and 30μg (cell lines) or 50μg (primary cells) protein was loaded for immunoblot analyses. The effects of idelalisib on Akt (Thr308) and total Akt, PDK1 (Ser241) and total PDK1, GS3K-3β (Ser9) and total GSK-3β protein expression levels were detected in (A) JeKo-1, Mino and Granta 519 cells and (B) Four MCL primary samples (MCL 12, MCL 17, MCL 20, MCL 21) were tested and representative immunoblots for specific proteins are shown. For each immunoblot panel, GAPDH loading control was measured. (C) Immunoblots from both MCL cell lines and primary cells were quantified and phospho- to total protein ratios were calculated. Values are presented as means ± SEM of independent experiments. All three cell line experiments were performed in triplicate; primary cell experiments were performed in four samples. Multiple statistical comparison of each concentration of idelalisib versus the DMSO control was performed on the results shown, and AKT (Thr308) and GSK-3β (Ser9) phosphorylation levels showed statistically significant decreases compared with vehicle control at at concentrations above 0.5 μM and at 0.5, 1 and 3μM idelalisib for phospho-PDK1. Note: the same protein loading controls appear in different figures; please refer to Supplemental Figure 4 for more details.

Figure 4. Effect of idelalisib treatment on the AKT/mTOR-mediated translation regulation pathway in primary MCL cells

Both MCL cell lines and primary cells were serum-starved for 1hr and then treated with DMSO or with various concentrations of idelalisib for 1hr followed by 15min of IgM stimulation before harvest. Immunoblots were performed to measure PRAS40 (Ser246), S6 (Ser235/236) and total S6, 4E-BP1 (Ser65) and total 4E-BP1 and mTOR (Ser2448) and total mTOR protein expression levels in (A) MCL cell lines (only Mino results are shown here) and (B) MCL primary cells (one to two sets of representative immunoblots from MCL primary cells are shown here). Accordingly, loading controls GAPDH or Vinculin levels were also measured. (C) Immunoblots from both MCL cell lines and primary cells were quantified and phospho- to total protein ratios were calculated. Values are presented as means ± SEM of independent experiments. All three cell line experiments were performed in triplicate; primary cell experiments were performed in four samples. PRAS40 (Ser246; all concentrations), S6 (Ser235/236; 0.5, 1 and 3μM idelalisib) and 4E-BP1 (Ser65; 0.1, 1 and 3μM idelalisib) levels showed statistically significant decreases compared with vehicle control. Note: the same protein loading controls appear in different figures, please refer to Supplemental Figure 4 for more details.

Figure 5. Effect of idelalisib treatment on the MAPK/MEK-mediated translation regulation pathway in MCL lines and primary cells

JeKo-1, Mino and Granta 519 and primary MCL cells from four different patients were serum-starved for 1h and then treated with DMSO, or with 0.5, 1 or 3μM idelalisib for 1h, and then the cells were co-cultured with IgM (10ng/μL) for 15min before harvesting. Cell extracts were prepared, and 30μg (cell lines) or 50μg (primary cells) of protein was loaded for immunoblot analyses. Immunoblots were performed to measure the phosphorylation and total protein expression levels of Raf (Ser259) and total Raf, MEK1/2 (Ser217) and total MEK1/2, p90RSK (Thr359/Ser363) and total p90RSK, and ERK1/2 (Thr202/Tyr204) and total ERK1/2 in (A) MCL cell lines and (B) MCL primary cells. Representative immunoblots from two sets of MCL primary samples were shown. (C) Quantitation of immunoblots to depict changes in proteins in three cell lines and four patient samples. Immunoblots were quantified, and phospho- to- total protein ratios were calculated. Values are presented as means ± SEM of independent experiments. Cell line experiments were performed in triplicate; primary cell experiments were performed in two to four samples. Multiple statistical comparison of each concentration of idelalisib versus the DMSO control was performed on the results shown. Among all the phospho-protein levels tested, MEK1/2 (Ser217/221; 0.5, 1, 3, 5μM idelalisib), ERK1/2 (Thr202/Y204; all concentrations of idelalisib) and p90RSK (Thr359/Ser363; 0.1, 0.5, 1, 3, 5μM idelalisib) showed statistically significant decrease compared with DMSO control. Note: the same protein loading controls appear in different figures; please refer to Supplemental Figure 4 for more details.

Figure 5. Effect of idelalisib treatment on the MAPK/MEK-mediated translation regulation pathway in MCL lines and primary cells

JeKo-1, Mino and Granta 519 and primary MCL cells from four different patients were serum-starved for 1h and then treated with DMSO, or with 0.5, 1 or 3μM idelalisib for 1h, and then the cells were co-cultured with IgM (10ng/μL) for 15min before harvesting. Cell extracts were prepared, and 30μg (cell lines) or 50μg (primary cells) of protein was loaded for immunoblot analyses. Immunoblots were performed to measure the phosphorylation and total protein expression levels of Raf (Ser259) and total Raf, MEK1/2 (Ser217) and total MEK1/2, p90RSK (Thr359/Ser363) and total p90RSK, and ERK1/2 (Thr202/Tyr204) and total ERK1/2 in (A) MCL cell lines and (B) MCL primary cells. Representative immunoblots from two sets of MCL primary samples were shown. (C) Quantitation of immunoblots to depict changes in proteins in three cell lines and four patient samples. Immunoblots were quantified, and phospho- to- total protein ratios were calculated. Values are presented as means ± SEM of independent experiments. Cell line experiments were performed in triplicate; primary cell experiments were performed in two to four samples. Multiple statistical comparison of each concentration of idelalisib versus the DMSO control was performed on the results shown. Among all the phospho-protein levels tested, MEK1/2 (Ser217/221; 0.5, 1, 3, 5μM idelalisib), ERK1/2 (Thr202/Y204; all concentrations of idelalisib) and p90RSK (Thr359/Ser363; 0.1, 0.5, 1, 3, 5μM idelalisib) showed statistically significant decrease compared with DMSO control. Note: the same protein loading controls appear in different figures; please refer to Supplemental Figure 4 for more details.

Figure 6. Confirmation and validation of idelalisib’s actions on the PI3K, AKT/mTOR and MAPK pathway proteins in MCL primary cells

(A–C) MCL primary samples (n=7 labeled as S1–S7) were treated with DMSO or with 1μM of idelalisib for 24h. Then the cells were harvested, and protein lysates were analyzed using the RPPA assay. Results were normalized to loading controls and then further normalized to DMSO control, and fold changes of each protein target in (A) upstream PI3K, (B) MAPK, and (C) AKT/mTOR pathways were plotted in groups in heat map format. Red color indicates an increase whereas blue color indicates a decrease, while white color indicates no change; the color intensities of red and blue indicates the degree of increase or decrease, respectively. The corresponding fold changes to these colors are shown. (D) RPPA results were validated using immunoblots of one representative patient sample, MCL #11, which corresponds to S1 in the heat map. The immunoblots were quantified and the values were indicated below each western blot band. The numbers listed are ratios of phospho- to- total protein expression or total protein/GAPDH.

Figure 7. A proposed model of BCR-mediated activation and idelalisib-induced inhibition of PI3Kδ pathway in mantle cell lymphoma

BCR-mediated PI3K-driven molecular pathway plot was generated, in which arrows indicate phosphorylation and dotted arrows indicate interaction through the PH domain. Protein name color blocks with dotted edges indicated mechanism un-confirmed in this study. The red crosses mark inhibitory effect of idelalisib on these kinases.