Eukaryotic initiation factor 4B is a multi-functional RNA binding protein that regulates histone mRNAs

Abstract

RNA binding proteins drive proliferation and tumorigenesis by regulating the translation and stability of specific subsets of messenger RNAs (mRNAs). We have investigated the role of eukaryotic initiation factor 4B (eIF4B) in this process and identify 10-fold more RNA binding sites for eIF4B in tumour cells from patients with diffuse large B-cell lymphoma compared to control B cells and, using individual-nucleotide resolution UV cross-linking and immunoprecipitation, find that eIF4B binds the entire length of mRNA transcripts. eIF4B stimulates the helicase activity of eIF4A, thereby promoting the unwinding of RNA structure within the 5' untranslated regions of mRNAs. We have found that, in addition to its well-documented role in mRNA translation, eIF4B additionally interacts with proteins associated with RNA turnover, including UPF1 (up-frameshift protein 1), which plays a key role in histone mRNA degradation at the end of S phase. Consistent with these data, we locate an eIF4B binding site upstream of the stem-loop structure in histone mRNAs and show that decreased eIF4B expression alters histone mRNA turnover and delays cell cycle progression through S phase. Collectively, these data provide insight into how eIF4B promotes tumorigenesis.

Graphical Abstract

Figure 1.

Identification of novel eIF4B interacting proteins involved in nonsense-mediated decay (NMD). (A) Venn diagram of eIF4B interacting proteins identified by mass spectrometry in DoHH-2 and GM01953 cells. All co-immunoprecipitations were subjected to RNase digestion, allowing the isolation of direct eIF4B interacting proteins. Proteins were identified as eIF4B binding when showing a significant enrichment (P-value <0.05) relative to the IgG control immunoprecipitation. (B) GO analysis of the top 25 biological processes associated with eIF4B interacting proteins in DoHH-2 and GM01953 cells. Fold enrichment shows the enrichment of genes above the expected and significance was determined by Fisher’s exact test. (C) STRING interaction analysis of the eIF4B interacting proteins conserved between DoHH-2 and GM01953 cells. Proteins are coloured according to cellular function: yellow, ribosomal proteins; blue, initiation factors; red, NMD; green, helicases; pink, splicing factors. (D) Cell lysates from DoHH-2 and GM01953 cells were immunoprecipitated for eIF4B or IgG. After RNA digestion and washing to remove RNA-dependent interactions, directly interacting proteins were analysed by immunoblotting with the indicated antibodies. GAPDH was used as a negative control because it did not bind to eIF4B in the mass spectrometry experiments. Blots shown are representative of three independent experiments.

Figure 2.

Identification of eIF4B RNA binding sites by iCLIP in DLBCL. (A) Cell lysates from GM01953 and DoHH-2 were subjected to RIC after UVC (254 nM) cross-linking (Cl). RIC samples were analysed by immunoblotting for the indicated RBPs and loading was normalized to RNA. Blots shown are representative of three independent experiments. (B) Venn diagram of the number of significant (Wald test P-value <0.05) individual RNAs bound by eIF4B in DoHH-2, GM01953 or both cell lines. (C) Distribution of iCLIP peaks across RNA species bound by eIF4B in DoHH-2 and GM01953 cells. (D) GO analysis of mRNAs bound by eIF4B in DoHH-2 and GM01953 cells (BP, biological processes; CC, cellular components; MF, molecular functions). Fold enrichment shows the enrichment of genes above the expected and significance was determined by Fisher’s exact test.

Figure 3.

eIF4B binds across the entire length of mRNA transcripts. (A) A subset of mRNAs bound by eIF4B in DoHH-2 cells were classified within five distinct groups based on the location of eIF4B binding (5′ UTR = 164 genes; start codon = 199 genes; CDS = 56 genes; stop codon = 125 genes; 3′ UTR = 124 genes). Data values show mean normalized reads with standard deviation (n = 4). (B) GO analysis of the RNAs bound by eIF4B at the 5′ UTR, stop codon and 3′ UTR (BP, biological processes; CC, cellular components; MF, molecular functions). Fold enrichment shows the enrichment of genes above the expected and significance was determined by Fisher’s exact test. (C) eIF4B binding motifs were analysed within all peaks, 5′ UTR and 3′ UTR eIF4B binding groups shown in panel (A). Motifs are ranked by enrichment P-value calculated by HOMER (35). % of targets: number of target sequences with motif/total targets; % of background: number of background sequences with motif/total background; STD(Bg STD): standard deviation of position in target and background sequences. (D) The length and ΔG of 5′ and 3′ UTRs of mRNAs within each eIF4B binding cluster (shown in panel A) were analysed. mRNAs are grouped by the region they are bound by eIF4B (5′ UTR, start codon, CDS, stop codon, 3′ UTR).

Figure 4.

eIF4B binds histone mRNAs. (A) Schematic representation of replication-dependent histone 3′ UTRs and tool RNAs generated for fluorescence-based RNA binding assays. (B) Distribution of eIF4B binding sites relative to the stem–loop on replication-dependent histone mRNAs from DoHH-2 iCLIP analysis. Proportion is an arbitrary value of eIF4B binding across all histone mRNAs. (C) eIF4B binding motifs were analysed within histone mRNAs in DoHH-2 cells. Motifs are ranked by enrichment P-value calculated by HOMER (35). % of targets: number of target sequences with motif/total targets; % of background: number of background sequences with motif/total background; STD(Bg STD): standard deviation of position in target and background sequences. (D) Fluorescence-based RNA binding assay showing eIF4B binding to tool RNAs based on histone 1 mRNA. Dissociation constants of RNA binding affinity of eIF4B alone and in the presence of UPF1 and SLBP. Error bars represent mean ± standard deviation (n = 3 independent experiments); *P ≤ 0.05 and ***P ≤ 0.001 by ANOVA. His1-FL: full-length sequence; His1-SL: containing histone stem–loop only; His1-ssRNA: full-length sequence minus the stem–loop; 20 nt CAA: negative control RNA; 30 nt AG: positive control RNA. Binding to 20 nt CAA and 30 nt AG was only carried out in the presence of eIF4B alone. (E) Comparison of eIF4B iCLIP binding sites from DoHH-2 (red) and GM01953 (peach) with UPF1 eCLIP binding sites from K562 (dark blue) and HepG2 (light blue) on histone mRNAs. All mRNAs (grey) were aligned at the stop codon and the stem–loop location is indicated in black.

Figure 5.

eIF4B stabilizes histone mRNAs and accurate S-phase progression. (A) (i) HeLa cells were transfected with a control non-targeting siRNA or an siRNA specific for eIF4B (30 nM) for 72 h. Cells were lysed and knockdown efficiency was determined by immunoblotting with the indicated antibodies. Blots are representative of three independent experiments. (ii) Quantification of eIF4B protein levels from panel (i) normalized to β-tubulin levels. Error bars represent mean ± standard deviation (n = 3 independent experiments); ***P ≤ 0.001 by unpaired Student’s t-test. (B) HeLa cells were transfected with an siRNA specific for eIF4B (30 nM) or a non-targeting control for 72 h and treated with flavopiridol (1 μM) to inhibit transcription for the indicated time. Total RNA was isolated and synthesized cDNA was used to quantify the indicated histone mRNA levels using qPCR. Estimation of mRNA half-life (t_1/2) was determined using a one-phase decay model (dotted line). Actin mRNA was used as a negative control and eIF4B depletion did not reduce actin mRNA half-life within the time frames analysed. Mean values for each time point are shown (solid line) and error bars represent standard deviation (n = 3 independent experiments). (C) HeLa cells were transfected with an siRNA specific for eIF4B (30 nM) or a non-targeting control and subjected to a double thymidine block to arrest cells at the G1/S checkpoint (G1/S). Cells were released from the block for either 4 or 8 h and immunoblotted with the indicated antibodies to determine the expression of cyclin proteins. Async, asynchronous cells. Blots shown are representative of three independent experiments. (D) Cells were treated as in panel (C) and collected and fixed for cell cycle analysis. DNA was stained using FxCycle Violet dye and quantified by flow cytometry. Histograms of stained DNA are representative of three independent experiments. (E) HeLa cells were transfected with an siRNA specific for eIF4B (30 nM) or a non-targeting control and subjected to a double thymidine block to arrest cells at the G1/S checkpoint (G1/S). Cells were released from the block for 24 h and fixed for cell cycle analysis. DNA was stained using FxCycle Violet dye and quantified by flow cytometry. Histogram of stained DNA is representative of three independent experiments. (F) Quantification of cell cycle stage from panel (E) using Watson’s (pragmatic) model. Error bars represent mean ± standard deviation (n = 3 independent experiments); *P ≤ 0.05 and **P ≤ 0.01 by unpaired Student’s t-test (ns, not significant).

Figure 6.

Proposed model of eIF4B regulation of histone mRNAs and cell cycle progression The loss of eIF4B leads to reduction in histone mRNA half-life and disrupted S-phase progression and cell proliferation. We propose that eIF4B stabilizes histone mRNAs by binding directly to histone mRNAs upstream of the stem–loop and to UPF1 protein, promoting rapid and accurate S-phase progression. Created with BioRender.com.