|Home | About | Journals | Submit | Contact Us | Français|
Eukaryotic translation initiation factor 3 (eIF3) is a multi-subunit complex that plays a critical role in translation initiation. Expression levels of eIF3 subunits are elevated or decreased in various cancers, suggesting a role for eIF3 in tumorigenesis. Recent studies have shown that the expression of the eIF3b subunit is elevated in bladder and prostate cancer, and eIF3b silencing inhibited glioblastoma growth and induced cellular apoptosis. In this study, we investigated the role of eIF3b in the survival of osteosarcoma cells.
To investigate the effect of eIF3b on cell viability and apoptosis in osteosarcoma cells, we first examined the silencing effect of eIF3b in U2OS cells. Cell viability and apoptosis were examined by the Cell Counting Kit-8 (CCK-8) assay and Western blot, respectively. We also performed gene profiling to identify genes affected by eIF3b silencing. Finally, the effect of eIF3b on cell viability and apoptosis was confirmed in multiple osteosarcoma cell lines.
eIF3b silencing decreased cell viability and induced apoptosis in U2OS cells, and by using gene profiling we discovered that eIF3b silencing also resulted in the upregulation of tumour necrosis factor receptor superfamily member 21 (TNFRSF21). We found that TNFRSF21 overexpression induced cell death in U2OS cells, and we confirmed that eIF3b silencing completely suppressed cell growth in multiple osteosarcoma cell lines. However, eIF3b silencing failed to suppress cell growth completely in normal fibroblast cells.
Our data led us to conclude that eIF3b may be required for osteosarcoma cell proliferation by regulating TNFRSF21 expression.
Cite this article: Y. J. Choi, Y. S. Lee, H. W. Lee, D. M. Shim, S. W. Seo. Silencing of translation initiation factor eIF3b promotes apoptosis in osteosarcoma cells. Bone Joint Res 2017;6:186–193. DOI: 10.1302/2046-3758.63.BJR-2016-0151.R2.
The control of protein translation is important for cancer development and progression. Global, as well as specific, mRNA translational regulation can promote tumour cell survival, angiogenesis, transformation, invasion and metastasis.1 Translation initiation is a rate-limiting step in protein synthesis from mRNA. A multitude of eukaryotic translation initiation factors (eIFs) are involved in the regulation of translation initiation.2 Several eIFs are overexpressed in cancer cells, and a considerable amount of research has been done to inhibit their function. For example, several human cancers exhibit eIF4E, eIF4A, and eIF5A overexpression, and these proteins have been considered targets of anticancer therapies.3-6 The function of eIF4E has been inhibited by using specific antisense oligonucleotides (4E-ASO)7 and ribavirin, a nucleoside inhibitor.8 Moreover, pateamine A,9 silvestrol,10 and hippuristanol11 have been reported to target eIF4A. In the case of eIF5A, N1-guanyl-1,7-diaminoheptane12 and ciclopirox olamine13 inhibited eIF5A by preventing its hypusination, a post-translational modification that is required for the full activity of eIF5A.
Another eIF called eIF3 is a multi-subunit complex, which plays a central role in translation initiation. The eIF3 complex is composed of at least 13 subunits,14,15 and the eIF3a,16-20 eIF3c21 and eIF3h22 subunits have been shown to be overexpressed in some carcinomas and in testicular seminoma. On the other hand, the eIF3f and eIF3e subunits have been reported to have decreased expression in melanoma and carcinoma cells.23,24 Indeed, eIF3 subunits can regulate the expression of proteins involved in cell cycle control. The eIF3a subunit negatively regulates the expression of p27, a cyclin dependent kinase (CDK) inhibitor.25 In addition, the expression of the proto-oncogene Myc correlated with the levels of eIF3h.22,26
The eIF3b subunit has been shown to be overexpressed in cancers of the prostate and bladder, and its overexpression has been linked to cancer prognosis.27 Although siRNA-mediated silencing of eIF3b inhibited the proliferation of colon cancer cells and glioblastoma,28,29 the potential role of eIF3b in the aggressive bone cancer osteosarcoma has not been elucidated. In this study, we investigated the effect of inhibiting eIF3b function on the survival of several osteosarcoma cell lines.
Three human osteosarcoma cell lines were used for the experiments and those were purchased from the American Type Culture Collection (ATCC, Manassas, Virginia). Human osteosarcoma cell lines (U2Os, MG63 and 143B) were authenticated using Short Tandem Repeat (STR) analysis. When compared with the reference profile by ATCC, cell lines with 80% match were considered to be related (Supplementary Table i). U2OS cells were grown in Modified Eagle’s medium (MEM). MG63 and 143B cells were grown in Dulbecco’s Modified Eagle’s medium (DMEM). As a control, the human foetal fibroblast cell, HDF-f (ScienCell Research Laboratories, Carlsbad, California) was used. Normal fibroblasts were grown in DMEM. All media were supplemented with 10% foetal bovine serum (FBS, HyClone Laboratories, Inc., Logan, Utah) and 1% antibiotic/antimycotic solution (Gibco Laboratories, Gaithersburg, Maryland). Cells were incubated in a humidified atmosphere containing 5% CO2 at 37°C.
The eIF3b siRNA sequence was sense: (5’-GGAGAGAAAUUCAAGCAAA(dTdT)-3’) and antisense: (5’-UUUGCUUGAAUUUCUCUCC(dTdT)-3’). The siRNA for eIF3b was designed based on the reported target site.28 Negative control siRNA with scrambled sequence (SN1003) and the eIF3b siRNA were purchased from Bioneer Corp. (Daejeon, Korea). The control and eIF3b siRNAs were transfected using Lipofectamine RNAiMAX Reagent (Life Technologies, Carlsbad, California) according to the manufacturer’s instructions.
The tumor necrosis factor receptor superfamily member 21 (TNFRSF21)-expressing plasmid, pcDNA3-TNFRSF21-Flag (HG10175-CF), was purchased from Sino Biological Inc. (Beijing, China). The sequence of the plasmid was confirmed by Sanger sequencing. Plasmids were transfected using Lipofectamine 2000 (Life Technologies) according to the manufacturer’s instructions.
Cell viability was monitored by Cell Counting Kit-8 (CCK-8) assay (Dojindo Laboratories, Kumamoto, Japan). For each assay, 1 × 103 cells were plated onto 96-well plates 24 hours after siRNA or expression plasmid transfection and were grown for the indicated days. CCK-8 solution (10 μl) was added to each well and incubated for one hour. The absorbance at 450 nm was measured using a SpectraMax Plus 384 instrument (Molecular Devices, München, Germany).
Cells were washed with ice-cold phosphate buffered saline (Biosesang, Gyeonggi, Korea) and the cell lysate was prepared in lysis buffer (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 10 mM EDTA, 1% NP-40). Protein concentrations were determined by Bradford protein assay reagent (Bio-Rad, Hercules, California). Equal amounts of protein were subjected to SDS-PAGE followed by transferral to a polyvinylidene difluoride (PVDF) membrane (Bio-Rad). The membrane was incubated with 5% skimmed milk in TBST (Biosesang, Gyeonggi, Korea) (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween-20) for one hour at room temperature. After being washed with TBST, the membrane was incubated overnight at 4°C with 5% bovine serum albumin in TBST containing the appropriate primary antibodies. After being washed with TBST, the membrane was then incubated with 5% skimmed milk in TBST containing the appropriate secondary antibodies conjugated to horseradish peroxidase (HRP) for one hour at room temperature. After being washed with TBST, the membrane was soaked in Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific Inc., Rockford, Illinois) and exposed to radiograph film. The following antibodies were used: Anti-β-actin (Sigma-Aldrich, St. Louis, Missouri), anti-eIF3b (Abcam, Cambridge, United Kingdom), anti-poly ADP ribose polymerase (PARP), anti-caspase-3, anti-rabbit IgG-HRP, anti-mouse IgG-HRP (Cell Signaling Technology Inc., Danvers, Massachusetts).
Gene expression profiling was performed with GeneChip Human Gene 2.0 ST Arrays (Affymetrix, Santa Clara, California). Total RNA was extracted using RNeasy (Qiagen Inc., Valencia, California) and RNA purity and integrity were evaluated by the OD 260/280 ratio and analysed by Agilent 2100 Bioanalyzer (Agilent Technologies Inc., Palo Alto, California). The array process was executed by Macrogen Inc. (Seoul, Korea). cDNA was synthesised using the GeneChip WT (Whole Transcript) Amplification Kit as described by the manufacturer. The sense cDNA was then fragmented and biotin-labelled with TdT (terminal deoxynucleotidyl transferase) using the GeneChip WT Terminal Labeling Kit. Approximately 5.5 μg of labelled DNA target was hybridised to the Affymetrix GeneChip Human 2.0 ST Array at 45°C for 16 hours. Hybridised arrays were washed and stained on a GeneChip Fluidics Station 450 and scanned on a GCS3000 Scanner (Affymetrix). The experiment was performed in three replicates. The raw data were extracted automatically in Affymetrix data extraction protocol using the software provided by Affymetrix GeneChip Command Console Software (AGCC, Affymetrix). After importing CEL files, the data were summarised and normalised with the robust multi-average (RMA) method implemented in the Affymetrix Expression Console Software (Affymetrix). The comparative analysis was carried out using an local pooled error (LPE) test and fold change in which the null hypothesis was that no difference existed between the two groups. The false discovery rate (FDR) was controlled by adjusting the p-value using the Benjamini-Hochberg algorithm. All statistical tests and visualisation of differentially expressed genes were conducted using R statistical language v.3.0.2 (R foundation, Vienna, Austria).
Total RNA was extracted using RNeasy (Qiagen), and 500 ng of RNA was reverse-transcribed with SuperScript III (Life Technologies) according to the manufacturer’s instructions. PCR was performed with 1 μl of cDNA using AccuPower HotStart PCR PreMix (Bioneer Corp.). The PCR conditions were 95°C for five minutes, then 30 or 33 cycles (95°C for 15 seconds, 55°C for 15 seconds and 72°C for 30 seconds), and finally 72°C for seven minutes. Primer sequences used for RT-PCR are provided in Supplementary Tables ii and iii.
U2OS cells were transfected with control or eIF3b siRNA. Three days after transfection, cells were harvested for caspase-3 activity analysis using a Caspase-3 Colorimetric Activity Assay Kit (Chemicon International Inc., Temecula, California). Caspase-3 activity was measured using 30 μg of lysate by detecting the chromophore p-nitroaniline, which was cleaved from the labelled substrate N-acetyl-Asp-Glu-Val-Asp p-nitroanilide.
Data are expressed as mean and standard deviation (sd) from three independent experiments. Analyses were performed using the GraphPad Prism 5 (GraphPad Software Inc., La Jolla, California). The significance of the statistical comparisons was determined via two-tailed t-tests. A p-value < 0.05 were determined to be statistically significant and marked by asterisks (* p < 0.05; ** p < 0.01; *** p < 0.001; ns, not significant).
To determine whether eIF3b is important for osteosarcoma, U2OS cells were transfected with siRNA-targeting eIF3b to decrease its levels, and cell viability assays were performed. Indeed, the growth of the cells was significantly inhibited when eIF3b was knocked down by siRNA (Figs 1a and and1b).1b). We confirmed the depletion of eIF3b by siRNA using Western blot analysis (Fig. 1c). These results indicate that eIF3b is required for robust osteosarcoma cell growth.
To examine whether eIF3b silencing induces not only cell growth inhibition but also apoptosis, we analysed activation of apoptosis related proteins. Western blot analysis showed that eIF3b silencing increased PARP and caspase-3 cleavage (Fig. 2a), a readout of an active apoptotic cellular programme. Moreover, we also found that caspase-3 activity was increased in cells depleted of eIF3b (Fig. 2b).
We performed microarray analyses to screen for genes that are regulated by eIF3b. The microarray heat map showed different gene expression profiles between the eIF3b-silenced samples and the control non-silenced samples (i.e. si-Control and Mock) (Fig. 3a). We detected 542 genes that were upregulated, as well as 656 downregulated genes, when eIF3b was depleted from cells using siRNA. Furthermore, 101 genes displayed expression levels that were affected more than twofold by eIF3b silencing (p < 0.05). Eleven genes that showed the largest differences in expression levels were confirmed by semi-quantitative RT-PCR (Fig. 3b). Among these genes, we chose to examine TNFRSF21 as a potential regulator for the apoptosis induced by eIF3b silencing because TNFRSF21 is directly involved in cell death by inducing caspase-3 activity. We therefore performed cell viability assays to examine whether TNFRSF21 could induce cell death in U2OS cells and indeed, overexpression of TNFRSF21 did induce cell death in U2OS cells (Fig. 3c). However, when the apoptotic effect of TNFRSF21 overexpression was compared with that of eIF3b silencing, TNFRSF21 was not as strong an inducer of apoptosis compared with cells depleted of eIF3b. Therefore, although TNFRSF21 is a downstream effector of eIF3b silencing, additional effectors must participate in the apoptosis induced by eIF3b siRNA.
To further confirm the effect of eIF3b silencing on the proliferation of osteosarcoma cells, we performed proliferation assays using the U2OS, MG63 and 143B cell lines. Strikingly, eIF3b silencing completely inhibited cell growth in each osteosarcoma cell line. Normal fibroblast cells (HDF-f) also exhibited growth inhibition due to eIF3b silencing (Fig. 4a). However, the growth of HDF-f was recovered at a later time point, by day 7 (Fig. 4b). We analysed PARP cleavage to evaluate the activity of apoptosis pathway induced by eIF3b silencing in those cell lines. PARP expression was significantly higher in osteosarcoma cell lines when compared with the HDF-f cell lines. cCleaved PARP was observed to be significantly higher among eIF3b siRNA-treated osteosarcoma cell lines, but not in HDF-f cells (Fig. 4c).
Several studies have shown that silencing of eIF3b inhibited the proliferation of different types of human cancers including bladder, prostate, and colon, as well as glioblastoma.27-29 However, whether eIF3b has a role in osteosarcoma proliferation has not previously been studied. In this study, we investigated the effect of altering eIF3b levels on osteosarcoma cell fate. eIF3b silencing inhibited U2OS cell proliferation while overexpression of eIF3b enhanced cell proliferation.
eIF3b is involved in protein synthesis, and eIF3b depletion can globally inhibit protein synthesis.27 Given that cell proliferation requires protein synthesis, the inhibited cell proliferation by eIF3b silencing might be a consequence of protein synthesis inhibition. However, in this study we observed activation of the caspase-3/PARP pathway by eIF3b silencing in U2OS cells. Moreover, other osteosarcoma cell lines, including MG63 and 143B, also showed complete growth inhibition and cell death by eIF3b silencing. In normal fibroblast cells (HDF-f), the growth inhibition was not complete and PARP activation was not detected. However, further study is needed toconfirm that eIF3b silencing is non-toxic for normal cells.
By using gene expression profiling, we found that the expression of TNFRSF21 was increased by eIF3b silencing. TNFRSF21 possesses a death domain and is known as Death Receptor 6 (DR6). It has been reported that TNFRSF21 overexpression induces apoptosis in a variety of cell types,30-33 and promotes apoptosis through activation of caspase-3.31,33 The apoptotic effect of TNFRSF21 has been shown in M17 neuroblastoma cells, human H4 neuroglioma cells, human cervical cancer HeLa cells and human embryonic kidney 293 (HEK293) cells.30-32 However, apoptosis was not induced by TNFRSF21 overexpression in human breast cancer MCF7 cells.32 We confirmed that cell death is induced by TNFRSF21 in U2OS cells, likely in part by the activation of caspase-3 in response to eIF3b silencing. However, TNFRSF21 is unlikely to be the main effector that induces cell death by eIF3b silencing. The rate of cell death induced by TNFRSF21 was less than that induced by eIF3b silencing. Thus, there may be other effectors that induce apoptosis when eIF3b expression is silenced. Gene profiling also revealed several genes that were up- or downregulated by eIF3b silencing. Even though we could not identify the relationship between these genes and cell death, the gene expression changes will likely give clues for further studies of eIF3b function in osteosarcoma.
The strength of our study is that we first showed the cross-talk between TNFRSF21 and eIF3b. Moreover, even though our results are somewhat limited due to in vitro approaches, targeting eIF3b in osteosarcoma may be beneficial because normal fibroblast cell death was not induced by eIF3b silencing. Further study using animal model will be required to confirm the eIF3b targeting effect.
Author Contribution: Y. J Choi: Acquisition of data, Analysis and interpretation of data, Drafting of manuscript.Y. S. Lee: Acquisition of data.
H. W. Lee: Acquisition of data.
D. M. Shim: Acquisition of data.
S. W. Seo: Study conception and design, Critical revision, Drafting of manuscript.
ICMJE Conflicts of Interest: None declared
A figure showing overexpression of elF3b enhanced survival into U20S cells and tables showing cell line authentication result, sequences and primers which have used for the studies and a list of genes used are available alongside this article at www.bjr.boneandjoint.org.uk
This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education and Science Technology (MEST) (NRF-2010-0024857), and by a grant from the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI12C1065).