RNA interference screening identifies lenalidomide sensitizers in multiple myeloma, including RSK2

Abstract

To identify molecular targets that modify sensitivity to lenalidomide, we measured proliferation in multiple myeloma (MM) cells transfected with 27 968 small interfering RNAs in the presence of increasing concentrations of drug and identified 63 genes that enhance activity of lenalidomide upon silencing. Ribosomal protein S6 kinase (RPS6KA3 or RSK2) was the most potent sensitizer. Other notable gene targets included 5 RAB family members, 3 potassium channel proteins, and 2 peroxisome family members. Single genes of interest included I-κ-B kinase-α (CHUK), and a phosphorylation dependent transcription factor (CREB1), which associate with RSK2 to regulate several signaling pathways. RSK2 knockdown induced cytotoxicity across a panel of MM cell lines and consistently increased sensitivity to lenalidomide. Accordingly, 3 small molecular inhibitors of RSK2 demonstrated synergy with lenalidomide cytotoxicity in MM cells even in the presence of stromal contact. Both RSK2 knockdown and small molecule inhibition downregulate interferon regulatory factor 4 and MYC, and provides an explanation for the synergy between lenalidomide and RSK2 inhibition. Interestingly, RSK2 inhibition also sensitized MM cells to bortezomib, melphalan, and dexamethasone, but did not downregulate Ikaros or influence lenalidomide-mediated downregulation of tumor necrosis factor-α or increase lenalidomide-induced IL-2 upregulation. In summary, inhibition of RSK2 may prove a broadly useful adjunct to MM therapy.

Figure 1

Top ranking lenalidomide sensitizers identified from druggable genome screens. The data collected from each siRNA were normalized and divided into untreated (siRNA only) and treated (siRNA+drug) classes, followed by analysis using the RIGER algorithm. When a stringent P value of <.01 was applied, 23 genes were selected as top lenalidomide modulators. (A) The 4 siRNAs targeting the top 23 sensitizer hits were ranked using a heat map plot based on their lethal activity changes after lenalidomide treatment. Relative: GENE-E converts values to heat map colors using the mean and maximum values for each row or the standard deviations from the row mean for each row (Source: https://www.broadinstitute.org/cancer/software/GENE-E/doc.html). “Untreated vs treated” are scores generated by the RIGER method for each siRNA. It is a score that is based on siRNA only (untreated) vs siRNA+lenalidomide (treated). (B) A list of top ranked genes with NES and P values are shown.

Figure 2

RSK2 knockdown inhibited myeloma growth and enhanced lenalidomide-mediated myeloma cytotoxicity. (A) JJN3 (left), MM1.S (middle), and H929 (right) cells were infected with control virus (NT) and RSK2 shRNAs (#52, #54, and #55) lentivirus, and immunoblotting was performed at 48 or 72 hours after virus infection to confirm RSK2 knockdown by shRNAs. (B) Cell viability was measured by MTT assay at day 6 after infection. The data from cells infected with RSK2 shRNAs were normalized to NT control. (C) Escalating doses of lenalidomide were added at 24 hours after virus infection and cell viability was measured at day 5 after treatment (the data from each treatment were normalized to vehicle-treated control). shRNA (#52) in H929 cells and shRNA (#55) in MM1.S cells alone were lethal to most of those cells and we were not able to evaluate drug response in them.

Figure 3

RSK2 inhibitor inhibited myeloma cell growth and synergized with lenalidomide to induce cytoxicity of myeloma cells. Four HMCLs (KMS11, XG1, MM1.S, and JJN3) were treated with various doses of lenalidomide (Len, μM) and (A) BI-D1870 (BI, μM) or (B) RMM46 (RM, μM) for 6 days, followed by MTT assay to measure cell viability. The data from each treatment were normalized to vehicle-treated control and then analyzed using CalcuSyn software to generate normalized isobolograms. An example of a normalized isobologram with the ranges indicative of synergy, additive effects, and antagonism has been provided above the experimental data. The normalized isobolograms below depict the combinations of lenalidomide (D1), BI-1870 or RMM46 (D2), and the effect of the drug observed at the corresponding dose (D × 1, D × 2). Any data points falling below the line have a combination index <1 and are synergistic. Data points on or above the line are considered additive or antagonistic, respectively. Each point on the normalized isobologram corresponds to a specific combination of lenalidomide and BI-D1870/RMM46.

Figure 4

RSK2 inhibition synergizes with lenalidomide to overcome BMSC-mediated drug resistance and it also enhances MM cytotoxicity mediated by other anti-myeloma drugs. (A) Myeloma cells (top panel, MM1.S cells; middle panel, XG1 cells) were incubated with indicated doses of drugs in the absence (left) or presence (right) of BMSCs. BMSCs were seeded at 5000/well in 96-well plates overnight before adding myeloma cells and indicated doses of drugs. MTT assays were read at day 5 after treatment. (B) JJN3 HMCLs were treated with indicated doses of BI-D1870, bortezomib (middle), dexamethasone (left), and melphalan (right) for 3 days, followed by MTT assay to measure cell viability.

Figure 5

Effects of RSK2 inhibition on lenalidomide-mediated immune-modulatory activities. (A) Jurkat T cells (top panel) were pretreated with vehicle (control) and indicated doses of lenalidomide (Len, 1 μM) and BI-D1870 (BI, 0.25 and 0.5 μM) for 1 hour, and then stimulated with PMA (1 ng/mL)/ionomycin (Ion, 1 μg/mL) for 40 hours. The cells were fixed, stained with anti-IL-2–APC and analyzed by flow cytometry. The expression of IL-2 after different combination of treatment (from left to right) are shown. (B) The treated Jurkat cells were also harvested for analysis of IKZF1 expression by western blot. (C) Human PBMCs were pretreated with lenalidomide (1 μM) or BI-D1870 (BI), or a combination of both drugs for 1 hour and then stimulated with LPS (1 ug/mL). The supernatants were harvested and tested for TNFα production by ELISA.

Figure 6

Knockdown or inhibition of RSK2 in myeloma cell lines induced apoptotic signaling and downregulation of IRF4, MYC, and MCL1. (A) Myeloma cell lines (MM1.S and H929) were infected with NT and RSK2 shRNAs lentivirus. After 48 hours, cells were harvested and used for immunoblotting assay to measure the expression of IRF4, MYC, MCL1, BIM, and PARP. (B) H929 cells were incubated with BI-D1870 (BI, 3 μM) at indicated time points, followed by immunoblotting analysis. (C) JJN3 cells were incubated with BI-D1870 at indicated doses for 6 hours, followed by immunoblotting analysis. (D) JJN3 cells were incubated with 2 μM of BI-D1870 for 3 hours, followed by qPCR analysis of MYC and IRF4 expression at transcription level.

Figure 7

The combination of lenalidomide and BI-D1870 demonstrated a significant enhancement of cytotoxicity and downregulation of IRF4 and MYC. H929 (A-B) and JJN3 (C-D) were treated with the indicated doses of lenalidomide and BI-D1870 (μM) for various times, and was followed by MTT assay (A and C, day 5) and immunoblotting assay (B and D, at various time points).