Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 May 13.
Published in final edited form as: Leukemia. 2013 Mar 7;27(10):2047–2055. doi: 10.1038/leu.2013.73

The eukaryotic translation initiation factor eIF4E is a direct transcriptional target of NF-κB and is aberrantly regulated in Acute Myeloid Leukemia

Fadi Hariri 1, Meztli Arguello 1,2, Laurent Volpon 1, Biljana Culjkovic-Kraljacic 1, Torsten Holm Nielsen 3, John Hiscott 4,5, Koren K Mann 3, Katherine LB Borden 1
PMCID: PMC4429918  NIHMSID: NIHMS688240  PMID: 23467026

Abstract

The eukaryotic translation initiation factor eIF4E is a potent oncogene elevated in many cancers including the M4 and M5 subtypes of acute myeloid leukemia (AML). While eIF4E RNA levels are elevated 3–10 fold in M4/M5 AML, the molecular underpinnings of this dysregulation were unknown. Here, we demonstrate that EIF4E is a direct transcriptional target of NF-κB that is dysregulated preferentially in M4 and M5 AML. In primary hematopoietic cells and in cell lines, eIF4E levels are induced by NF-κB activating stimuli. Pharmacological or genetic inhibition of NF-κB represses activation. The endogenous human EIF4E promoter recruits p65 and cRel to evolutionarily conserved κB sites in vitro and in vivo following NF-κB activation. Transcriptional activation is demonstrated by recruitment of p300 to the κB sites and phosphorylated Pol II to the transcriptional start site. In primary AML specimens, generally we observe that substantially more NF-κB complexes associate with eIF4E promoter elements in M4 and M5 AML specimens examined than in other subtypes or unstimulated normal primary hematopoietic cells. Consistently, genetic inhibition of NF-κB abrogates eIF4E RNA levels in this same population. These findings provide novel insights into the transcriptional control of eIF4E and a novel molecular basis for its dysregulation in at least a subset of M4/M5 AML specimens.

Keywords: AML, eIF4E, NF-κB

Introduction

The eukaryotic translation initiation factor 4E (EIF4E) is a potent oncogene which is inappropriately elevated in about 30% of human cancers including the M4 and M5 subtypes of Acute Myelogenous Leukemia (AML) and in blast crisis, but not chronic phase, CML (1). eIF4E overexpression leads to increased proliferation, evasion of apoptosis, oncogenic transformation, tumor invasion and metastases (1–5). eIF4E interacts with the methyl-7-guanosine cap moiety on the 5’ end of mRNAs (6) and via this activity plays a central role in cap dependent translation and in nucleo-cytoplasmic export of a subset of transcripts encoding proteins involved in cellular growth, survival and transformation such as Cyclin D1, VEGF, c-myc, Mcl1 and Pim1 (7, 8). Both the translation and export activities of eIF4E contribute to its transformation potential (9). Depletion of eIF4E in cancer cells using siRNA, anti-sense oligonucleotides or pharmacological inhibitors leads to cell cycle arrest and decreased tumorigenicity (10–13). Targeting of eIF4E with a competitive inhibitor of the m7G cap moiety, ribavirin, led to clinical responses in poor prognosis M4 and M5 AML patients, including remissions in a phase II trial (10).

Few studies have focused on how eIF4E RNA levels become elevated in malignant cells. In M4 and M5 AML specimens, eIF4E RNA and protein levels are elevated by ~ 3–10 fold relative to primary hematopoietic cells from healthy volunteers (10). Traditionally, transcription of EIF4E was thought to be controlled only by c-Myc (14). Other studies have implicated Sonic hedgehog signaling (15) and p53 (16) in the control of EIF4E transcription, but these too are ultimately considered to be mediated by c-Myc interaction with the EIF4E promoter. Interestingly, eIF4E expression is still stimulated in response to serum in Myc null fibroblasts indicating that there are other mechanisms to control EIF4E transcription (17). Consistent with this idea, a recent report suggests that EIF4E is also a C/EBP target (18). Our previous studies in primary M4 and M5 AML specimens suggested a tantalizing link between NF-κB and the transcription of EIF4E. Introduction of the NF-κB inhibitor IκB-super repressor (IκB-SR) into primary M4 or M5 AML specimens, which are characterized by constitutive NF-κB activity, resulted in a substantial reduction in eIF4E transcript and protein levels (19). However to date, whether the link between NF-κB and eIF4E was a direct one, had not been investigated.

The NF-κB family of transcription factors plays a central role in the regulation of growth, proliferation, inflammation and apoptosis (20). Importantly, NF-κB is constitutively active in primary AML specimens and this elevation contributes to the leukemogenic phenotype (21). NF-κB members include NF-κB1 (p50), NF-κB2 (p52), Rel A (p65), Rel B and cRel and they may form homo- or hetero-dimers. In resting cells, NF-κB subunits reside in the cytoplasm, kept inactive by the IκB family of inhibitors. Receptor stimulation leads to a signaling cascade that culminates in the activation of the IKK kinase complex, which phosphorylates the IκB molecules leading to their proteasomal degradation. Free NF-κB dimers translocate into the nucleus, where they bind cognate DNA sequences known as κB sites (consensus 5’- GGGRNYYYCC-3’) via their Rel homology domain (RHD) to regulate gene transcription (22, 23). Importantly, introduction of a dominant negative repressor of NF-κB, IκB-SR, which blocks its nuclear translocation leads to reduced growth of primary AML cells (19).

In this study, we establish that eIF4E is a direct transcriptional target of NF-κB in hematopoietic cell lines and primary normal hematopoietic cells. Further, studies in primary M4/M5 AML specimens indicate that at least for the specimens examined the eIF4E promoter elements are preferentially occupied relative to M1 and M2 AML subtypes with normal eIF4E levels or to healthy, unstimulated, hematopoietic cells. These studies suggest that NF-κB activation can differentially target subsets of genes in specific AML contexts. These findings provide a novel control mechanism for eIF4E expression and a novel basis for its dysregulation in AML, and likely in other malignancies characterized by activated NF-κB.

Materials and Methods

Primary cell isolation and treatments

Primary B cells were purchased from stem cell research. Primary AML samples from anonymous patients (M1, M2, M4 and M5) were obtained from the BCLQ (Banque de Cellules Leucémiques du Québec) with ethics committee approval from the University of Montreal (Comité d’éthique de la recherche en santé CEFRM#195). Leukaphereses from healthy donors were obtained at the Royal Victoria Hospital, Montreal, Quebec, Canada with ethics committee approval from the Jewish General Hospital and McGill University Research Ethics Committee (REC) board of the SMBD-Jewish General Hospital (protocol number# 06-103). Written informed consent was obtained from both healthy donors and AML patients in accordance with the Declaration of Helsinki. Characteristics of the AML primary specimens used in this study are presented in supplemental table 1. AML cells were thawed in RPMI 1640 (Invitrogen) supplemented with 10% heat inactivated FBS and 100 U of penicillin/streptomycin (Invitrogen). Nuclear lysates were prepared from 4 million cells respectively.

PBMCs were isolated with Ficoll-Paque PLUS (Invitrogen) according to the manufacturer’s instructions. Cells were re-suspended in RPMI 1640 (Invitrogen) supplemented with 15% heat inactivated FBS and 100 U of penicillin/streptomycin (Invitrogen). Cells were plated at a density of 30×106 cells in T75 flasks and treated with PMA (Sigma) at 20 ng/ml for the described time points. For NF-κB inhibition, cells were pretreated for 1 hour with 10 µM Bay 11-7082 (Sigma) prior to PMA stimulation and the inhibitor was kept in the media throughout the experiment.

Cell culture

BJAB (Burkitt’s cell lymphoma) and THP1 (M5 AML) cells were obtained from and cultured according to the American Type Culture Collection (ATCC). KM-H2 (Hodgkin’s) lymphoma cell line was a kind gift from Dr. Sigrun Smola. The BJAB cell line used was always cultured at low density, maintained at approximately 70% confluency and kept from reaching full confluency. BJAB cells (106 cells in 6 well plates) were treated with 20 ng/ml PMA (Sigma). NF-κB inhibition was carried out using the Bay 11-7082 (Sigma) by pre-treating the cells with 10 µM for 1 hour prior to stimulation with PMA.

Antibodies and Primers

The following antibodies were used: p50 (Rockland Immunochemicals, 100-4164), Rel A (p65) (Rockland Immunochemicals, 100-4165), c-Rel (Cell Signaling #4727S), p300 (Rockland Immunochemicals, 100-301-76), eIF4E (BD Transduction laboratory, 610270), S2/5 phospho-RNA polymerase II (Cell Signaling #4735S) and beta-actin (Sigma, A5441). Specific primers designed for gel shift assays, chromatin Immunoprecipitation (ChIP) and expression analysis are summarized in supplementary table 2.

Promoter Analysis and validation of NF-κB sites

The human eIF4E promoter sequence was obtained from the Transcriptional Regulatory Element Database (TRED, cold spring harbor, http://rulai.cshl.edu/cgi-bin/TRED/tred.cgi?process=home). Promoter analysis for transcription factor binding sites was performed using MatInspector (24) (http://www.genomatix.de/cgi-bin/matinspector_prof/). Validation of the putative NF-κB elements was carried out with Chromatin immunoprecipitation (ChIP) and Electrophoretic mobility shift assays (EMSA). These protocols are described in detail in the supporting text.

Expression Analysis

Total RNA was isolated from control and treated cells using Trizol (Invitrogen) according to the manufacturer’s instructions. RNA was additionally treated with RNase-free DNase I (Promega). cDNA was synthesized using Super script II (Invitrogen) and expression analysis was then performed with real time quantitative (Q) RT-PCR using the StepONE real-time PCR (Applied Biosystems). CT values were analyzed using the DDCT method (25) and normalized to β-Actin, Histone 2B and Ubiquitin. Total protein was extracted with RIPA buffer (Tris 50mM, NaCl 150 mM, SDS 0.1 %, Sodium Deoxycholate 0.5 % and NP40 1%) and analyzed by western blot using chemiluminescence (Thermoscientific). Western blot band densities were analyzed with ImageJ software.

Results

NF-κB activation stimulates eIF4E expression in hematopoietic cell lines

Our previous studies indicated that eIF4E is highly elevated (~ 3–10 fold) in M4/M5 AML specimens but generally not in other AML subtypes or in a variety of primary hematopoietic cells isolated from healthy volunteers including CD34+ cells, granulocytes and monocytes (10, 19, 26). Given our previous finding that IκB-SR expression in a primary M5 AML specimen repressed eIF4E RNA levels, we investigated the potential role of NF-κB activity in the direct transcriptional control of eIF4E. In M4/M5 AML cells, NF-κB activity is constitutive thus we used a hematopoietic cell line BJAB as a model system because of its low basal NF-κB activity, which enabled us to examine the effects of NF-κB activation (22, 23, 27). Cells treated with PMA, an NF-κB stimulator, exhibited a rapid increase in eIF4E mRNA, with levels doubling at 3 hours (Figure 1A), as determined by Q-PCR and a parallel increase in eIF4E protein levels up to 7 fold was also observed (Figure 1D). Known NF-κB targets MYC and CCND1 also displayed a similar increase (2–3 fold) upon PMA stimulation of these cells (Figure 1B and 1C). Given that PMA is a pleiotropic agent we assessed the effects of the pharmacological IKK complex inhibitor Bay 11-7082 (28), which blocks IκBα phosphorylation. The observed induction of eIF4E mRNA expression was completely abrogated when samples were treated with Bay 11-7082, consistent with the hypothesis that the increase in eIF4E mRNA observed is due to NF-κB activation (Figure 1A).

Figure 1. Stimulation of BJAB cells with PMA leads to NF-κB dependent EIF4E transcriptional upregulation.

Figure 1

Open in a new tab

BJAB cells at approximately 50% confluency were stimulated with PMA (20 ng/mL) for the indicated time points, in the presence (black) or absence (grey) of BAY11-7082 (10 µM). A, B, C) Expression of eIF4E (EIF4E), cMyc (MYC) and cyclinD1 (CCND1) mRNA was analyzed by Q-PCR. Results shown are the average of 3 independent experiments each performed in triplicate (error bars are s.e.m, p values were calculated using the student t-test). D) Immunoblot of eIF4E and cMyc expression following PMA stimulation of BJAB cells for the indicated times; β-actin is provided as loading control. Band intensity was quantified with ImageJ.

The NF-κB subunits cRel and p65 directly alter EIF4E promoter activity

We analyzed the human EIF4E promoter (up to 863 bp upstream of the transcriptional start site) using bioinformatics (MatInspector, (24)) for the presence of κB sites. Previous analysis of the rat EIF4E promoter had suggested the presence of two putative Rel elements but these were never examined for activity (29). We identified four κB sites (denoted κB1 through 4 in the text) centered on positions −836, −808, −630 and −348 relative to the transcriptional start site (Figure 2A and supplemental figure 2A). These sites are evolutionary conserved in human, cow, monkey, mouse and rat (supplemental figure 2B). Note that the NF-κB pathway is absent in yeast and worms (20) although both express eIF4E suggesting an evolutionary emergence of an NF-κB mediated mechanism to induce eIF4E expression in higher organisms. Our analysis also revealed the presence of binding sites for several other transcription factors (PU.1, Pax.5, Octamer, NF-AT) (not shown). Although the significance of these novel binding sites needs further investigation, they are suggestive of a more intricate transcriptional control of eIF4E than the prevailing model of transcriptional regulation solely through Myc.

Figure 2. The EIF4E promoter contains four κB sites preferentially bound by cRel-p65 NF-κB complexes.

Figure 2

Open in a new tab

A) Schematic representation of the EIF4E promoter. Predicted κB sites (open circles) are indicated. B) Kinetics of NF-κB binding to the consensus κB following PMA stimulation of BJAB cells. C) EMSA of nuclear extracts (NE) from PMA-stimulated BJAB (90 min) using probes corresponding to κB3 and κB4 sites in the EIF4E promoter. Supershift analysis using antibodies against p65, p50, cRel and IgG control as well as competition with consensus cold probe (CCP) are indicated. Protein/DNA complexes are indicated by arrows, supershifted complexes by arrowheads. Free probe is also shown. D) EMSA of nuclear extracts from PMA-stimulated BJAB (90 min) using probes corresponding to κB3 and κB4 sites in the EIF4E promoter with cold probe competition using consensus cold probe (CCP) or mutant cold probe (MCP).

In order to determine whether activating signals induce binding of NF-κB complexes to the putative κB elements in the EIF4E promoter, we carried out electrophoretic mobility shift assays (EMSA) using nuclear extracts from BJAB cells in the presence and absence of PMA. These cells exhibited low basal levels of NF-κB activity, which could be dramatically induced by PMA treatment as determined by EMSA using the kappa light chain consensus motif (Figure 2B). At 90 minutes post-stimulation, all four of the predicted κB elements yielded inducible complexes when incubated with nuclear extracts (Figure 2C and supplemental figure 3). The complexes were specific, as they could be competed by excess cold probe corresponding to the consensus kappa light chain κB motif (30) (Figure 2C, lane 6), but not with a cold probe corresponding to mutant sites for κB3 and κB4 (Figure 2D). Supershift analysis with antibodies against p50, p65 and c-Rel revealed cRel/p65 heterodimers bound the promoter elements (Figure 2C, lanes 2 and 4). Although some p50 protein was supershifted (Figure 2C, lane 3), specific bands corresponding to p50-containing complexes could not be detected, suggesting that p50 is not a central component of the bound complexes. These experiments demonstrated that all four κB sites in the EIF4E promoter recruit cRel/p65 complexes.

NF-κB recruits p300 and Pol II to the EIF4E promoter in vivo

To establish that the NF-κB complexes detected by EMSA formed not only in vitro on eIF4E promoter fragments but also in cells, we carried out chromatin immunoprecipitation (ChIP) experiments in BJAB cells treated with PMA for one or two hours (Figure 3). Our results revealed maximum cRel binding at 1 hour post-treatment, with 3-fold enrichment at the κB3 site and a 2-fold enrichment at the κB4 site (Figure 3B). The p65 subunit was recruited to both sites following 1 hour of treatment with maximal 9-fold and 10-fold enrichment at the κB3 and κB4 sites, respectively (Figure 3A). Activated NF-κB (cRel and p65) dimers bound to the EIF4E promoter effectively recruited p300 histone acetyl transferase, a marker of transcriptionally active NF-κB complexes (31). Maximum p300 enrichment was again observed one hour post-treatment: 9-fold for κB3 and 7-fold for κB4 (Figure 3C). In contrast, no binding of NF-κB subunits was detected at the κB1 and κB2 sites by realtime PCR at 1 or 2 hours post-PMA treatment (data not shown). This suggests that the κB1 and κB2 sites are not active in vivo in this context, possibly due to the chromatin status at the eIF4E locus. Phosphorylated Pol II was significantly enriched at the EIF4E coding region with 2 fold at 1h and 5 fold at 2h (Figure 3D) confirming enhanced EIF4E gene transcription. Thus, shortly after NF-κB stimulation, binding of cRel/p65 dimers at the κB3 and κB4 but not κB1 and κB2 sites of the EIF4E promoter results in transcriptional upregulation of eIF4E expression via recruitment of p300.

Figure 3. NF-κB complexes are recruited to the EIF4E promoter and promote transactivation.

Figure 3

Open in a new tab

BJAB cells were stimulated with PMA for 0, 1 or 2 hours and subjected to chromatin immunoprecipitation using antibodies specific for p65 (A), cRel (B), p300 (C) and phosphorylated Pol II (Ser2/Ser5) (D). Data were normalized to IgG control and represented as fold enrichment with respect to untreated cells. Error bars represent standard deviations from triplicate measurements of a representative experiment.

NF-κB activation induces eIF4E transcription in primary human cells

To examine the biological relevance of our findings to the normal control of EIF4E transcription, we assessed eIF4E levels as a function of PMA stimulation in primary PBMCs isolated from two healthy individuals (Figure 4, supplemental figure 1). PMA induced eIF4E mRNA expression as early as two hours with maximal induction at 4 hours (4-fold upregulation relative to the untreated controls, Figure 4A). Increased eIF4E mRNA levels were observed up to 12 h post-treatment (3-fold). Established NF-κB targets c-Myc and Cyclin D1 mRNAs, had their RNA levels induced following PMA stimulation (supplemental figure 1A and 1B). Results were similar for eIF4E from another healthy volunteer (Figure 4B). Flow cytometry analysis of the same samples revealed a significant increase in the population of cells with elevated eIF4E protein levels (supplemental figure 1C and 1D). Upon treatment with the NF-κB inhibitor, Bay 11-7082, we observe that both eIF4E mRNA and protein expression were no longer induced by PMA consistent with these effects being mediated through NF-κB. Furthermore, eIF4E mRNA was monitored as a function of PMA stimulation in primary B lymphocytes as a direct primary companion to BJABs. eIF4E transcript doubled at two hours and reached a maximum of 4 fold at 6 hours in an NF-κB dependent manner (Figure 4C and 4D). Thus eIF4E is an NF-κB inducible gene in primary hematological cells.

Figure 4. PMA Stimulation of primary human PBMCs increases eIF4E expression in an NF-κB dependent manner.

Figure 4

Open in a new tab

(A) PBMCs (volunteer1) were stimulated with PMA (20 ng/mL) for the indicated time points in the presence or absence of the NF-κB inhibitor Bay 11-7082 (10 µM). Expression of eIF4E (EIF4E) was assessed at the mRNA level by realtime Q-PCR. (B) Same as (A) using PBMCs of a second healthy volunteer. Error bars indicate standard deviation. (C, D) Primary B lymphocytes were stimulated with PMA (20 ng/mL) for the indicated time points in the presence or absence of the NF-κB inhibitor Bay 11-7082 (10 µM). Expression of eIF4E and cMyc (MYC) were assessed at the mRNA level by realtime Q-PCR. Results shown are from cells obtained from one healthy donor in triplicate.

eIF4E transcription is elevated in cells with constitutively active NF-κB

Many cancers including AML are characterized by constitutively active NF-κB (21). Thus, we monitored eIF4E in KM-H2 cells which are hematopoietic cells characterized by constitutively active NF-κB, owing to a somatic mutation in the IκBα gene (32). In these cells, EMSA analysis revealed constitutive binding of NF-κB to the κB3 and κB4 sites of the EIF4E promoter (Figure 5A). Supershift analysis demonstrated the presence of cRel/cRel homodimers as well as cRel/p65 heterodimers; some p65/p50 complexes were also detected with κB4, although their unusually slow migration pattern was suggestive of the presence of an additional unidentified factor(s). To further demonstrate the direct role of NF-κB in this system, we transduced KM-H2 cells with a retroviral vector expressing IκB–SR to block NF-κB activity. Consistently, IκB-SR reduced eIF4E transcript and protein levels as well as those for the c-Myc control (Figure 5B and 5C). Thus, eIF4E transcription is elevated in the context of constitutively active NF-κB.

Figure 5. Constitutively active NF-κB regulates eIF4E expression in KM-H2 cells.

Figure 5

Open in a new tab

(A) EMSA analysis of KM-H2 nuclear extracts using probes corresponding to the κB3 and κB4 sites. Supershift analysis using antibodies against p65, p50, cRel and IgG control as well as competition with consensus cold probe (CCP) were done. Protein/DNA complexes are indicated by arrows and supershifted complexes by arrowheads. Free probe is also shown. (B) KM-H2 cells were transduced with IκB-SR or vector control and eIF4E and cMyc (positive control) RNA levels were assessed by Q-PCR. Error bars represent standard deviations. (C) Same samples as in (B) were analyzed by immunoblot.

Elevated NF-κB activity in M4 and M5 AML specimens underlies, at least in part, eIF4E dysregulation

To determine whether NF-κB transcriptional activity could underlie elevation of eIF4E in primary M4 and M5 AML, we examined NF-κB activity in the AML-M5 cell line, THP1 (33), characterized by elevated eIF4E (26). Consistent with previous findings (34), THP1 cells harbor constitutively active NF-κB. Nuclear lysates prepared from THP1 cells exhibited NF-κB specific binding for the four identified promoter elements (Figure 6A). ChIP analysis performed on these cells revealed Rel A enrichment on all NF-κB elements in the eIF4E promoter as seen in Figure 6B but not to a random region in exon 2 of eIF4E. Note that given the proximity of the κB1 and κB2 elements, the ChIP experiment cannot differentiate between these 2 sites and thus is referred to as κB1/κB2. Thus, NF-κB is constitutively found on the eIF4E promoter.

Figure 6. NF-κB recognition of the EIF4E promoter elements in AML cell lines.

Figure 6

Open in a new tab

(A) EMSA of nuclear extracts prepared from unstimulated THP1 cells (M5 AML) that were incubated with the κB elements in the eIF4E promoter. Supershift analysis (arrowheads) using antibodies against p65 and cRel as well as competition with consensus cold probe (CCP) are indicated. Free probe is also shown. (B) NF-κB complexes are recruited to the EIF4E promoter in THP1 and KG1a (M0 AML with high eIF4E) cell lines. Chromatin immunoprecipitation was carried out with Rel A antibody using chromatin from THP1 and KG1a cells. Recruitment to the κB elements as well as a non-specific control region in exon 2 of eIF4E was monitored by Q-PCR. Data is represented as fold enrichment over IgG. Error bars represent standard deviations from triplicate measurements.

Further, we examined NF-κB activity in KG1a cells, a varient AML M0 cell line derived (35). However, unlike the majority of primary M1 and M2 AML specimens that are characterized by normal eIF4E levels and distribution, KG1a cells represent the minority of these patients with highly elevated eIF4E that is mainly nuclear (supplemental Figure 4A and 4B). Similar to THP1, nuclear lysates from KG1a cells demonstrated NF-κB binding for all four identified promoter elements (supplemental figure 4C) and ChIP analysis indicated Rel A enrichment for all the elements (Figure 6B). Thus, elevation of eIF4E by NF-κB activity is not necessarily lineage restricted and could also explain how eIF4E levels become elevated in the 10–20% of M1 and M2 AML specimens characterized by elevated eIF4E.

To assess whether our findings could be translated to patient specimens, we extended our studies to primary AML specimens. We observed no substantial differences in constitutive NF-κB activity across AML subtypes as determined by EMSA experiments using the consensus kappa light chain element consistent with previous reports (21) (Figure 7A). To assess if the EIF4E promoter was specifically enriched for NF-κB proteins we carried out EMSA experiments with all four promoter elements in these AML specimens. We observed complexes highly enriched on the κB3 promoter elements in M4 and M5 specimens (in 2/2 M4 and 2/2 M5) relative to the M1 and M2 specimens (0/1 M1 and 0/2 M2 specimens), which were previously characterized by normal eIF4E levels (26) (Figure 7A and Supplemental table 1). Consistently, the κB4 element was preferentially occupied in M4 and M5 specimens relative to the M1 and M2 specimens (Figure 7A). Further, we observed supershifts using antibodies to Rel A and c-Rel consistent with these being NF-κB complexes with predominant Rel A species (Figure 7B). Additionally, M4/M5 nuclear lysates strongly bound the κB1 and κB2 elements (Figure 7C). This is consistent with our previous findings that eIF4E RNA and protein levels are preferentially elevated in M4 and M5 relative to M1 and M2 AML subtypes with approximately 44/44 M4/M5 AML specimens and 2/22 M1/M2 AML specimens relative to 5 healthy controls (1, 10, 19, 26). To assess the functional relevance of these associations we used ChIP analysis performed to monitor Rel A recruitment in the examined AML samples. Rel A was recruited to κB1/κB2 and κB3 but not κB4 in the M4/M5 samples but not the M2 (Figure 7D). Although all four elements were bound by nuclear lysates from PMA-induced BJABs as well as the NF-κB constitutive AML cell line (THP1) and primary specimens through EMSA assays, ChIP assays revealed Rel A recruitment to the κB1/κB2 elements in only the AML primary specimens and cell line but not in BJAB cells implying a lineage specific recognition for NF-κB proteins on the eIF4E promoter. Importantly, κB3 elements were also bound in the ChIP assays indicating these likely play important roles as well.

Figure 7. Selective NF-κB recognition of the EIF4E promoter elements in M4/M5 AML.

Figure 7

Open in a new tab

EMSA of nuclear extracts prepared from unstimulated PBMCs from a healthy volunteer (Norm) and different M1, M2, M4 and M5 AML primary specimen (corresponding to UPN samples 1–4 as presented in supplemental table1) incubated with consensus κB motif or with the κB3 and κB4 elements (A). (B) Supershift analysis (κB3/κB4) (arrowheads) using antibodies against p65 and cRel as well as competition with consensus cold probe (CCP) are indicated. The asterisk (*) indicates a break in the gel during drying. (C) EMSA of nuclear extracts prepared from M2, M4 and M5 AML primary specimen incubated with κB1/κB2 elements. Supershift analysis (arrowheads) using antibodies against p65 and cRel as well as competition with consensus cold probe (CCP) are indicated. Free probe is also shown. (D) Chromatin immunoprecipitation was carried out with Rel A antibody using chromatin from the different specimen (Norm, M2, M4 and M5) used in (A). Recruitment to the κB elements was monitored by Q-PCR. Data were normalized to the IgG control and depicted as fold enrichment with respect to the normal healthy specimen (Norm). Error bars represent standard deviations from triplicate measurements.

Thus, eIF4E promoter elements specifically recruit NF-κB complexes enabling increased EIF4E transcription relative to other AML subtypes. Consistently, previous studies showed that introduction of IκB-SR into primary M4 and M5 AML specimens resulted in reduced eIF4E mRNA and protein levels (19). It seems likely that NF-κB dysregulation in at least a subset of M4/M5 AML underlies, at least in part, aberrant elevation of eIF4E RNA levels in the evaluated samples. Clearly, more specimens will need to be examined to determine the generality of these findings.

Discussion

This study reveals novel insights into the control of EIF4E transcription in primary hematopoietic cells as well as its dysregulation in AML specimens. These are the first studies to show that EIF4E is a transcriptional target of NF-κB. Recently, C/EBP has also been shown to regulate EIF4E transcription (18) and thus NF-κB and C/EBP serve as examples that the transcriptional control of eIF4E, thought for nearly 16 years to be solely the purview of c-Myc, is more complicated. These findings suggest that there could be eventual clinical utility in controlling the transcription of EIF4E with the use of NF-κB inhibitors in addition to directly inhibiting eIF4E activity with ribavirin (10, 12, 26, 36, 37). Interestingly, many NF-κB target genes are in fact eIF4E mRNA export and/or translation targets (e.g. Myc and cyclinD1) suggesting these pathways cooperate to drive proliferative gene expression. Our results have shown that genetic and pharmacological inhibition of NF-κB result in downregulation of eIF4E targets suggesting that there is a nexus between transcriptional and post-transcriptional gene expression networks to modulate cell proliferation.

Beyond the control of eIF4E, our findings strongly suggest that NF-κB activity is likely heterogeneous amongst AML specimens with regard to other targeted promoters. In other words, the NF-κB dependent transcription of factors besides eIF4E may be differentially regulated between AML subtypes, potentially contributing to differences in leukemogenic potential. In this way, the kappa light chain occupancy may not be altered amongst subtypes as it is dependent on Rel components, but differences in the sequences of other promoters may lead to increased dependency of these promoters on non-Rel components for transcriptional activation. Furthermore, the selective in vivo recruitment of Rel A to the κB1/2 region in the M4/M5 AML specimens as well as the M5 AML THP1 cell line but not in the lymphocytic BJAB cells could be due to lineage differences. However, this is not strictly lineage restricted as ChIP studies indicate that in the high eIF4E cell line KG1a, enrichment on the κB1/2 binding elements is also observed. These differences could underlie the constitutive eIF4E upregulation in M4/M5 AML and suggest that non-Rel components play a role in this process. Finally, we did not have access to sufficient primary M3 AML (APL) specimens to examine NF-κB activity and thus cannot exclude that this is relevant to this leukemia subtype. However, primary APL specimens and NB4 cells are characterized by normal eIF4E levels and localization suggesting this may not be critical (19, 38).

Thus, specific non-Rel transcriptional co-factors of NF-κB may be specifically dysregulated in M4 and M5 AML, allowing preferential dysregulation of eIF4E, and potentially other promoters. These factors may be involved in the selective NF-κB recruitment to the eIF4E promoter. Examples of non-Rel proteins shown to selectively modulate DNA recognition and transactivation of NF-κB proteins in other contexts include: RPS3, CD40, BAFFR, Akirins, CHFR, PKAIP, AEG-1, ING-4 and others (Reviewed in (39)). Future studies should reveal the identity of non-Rel co-factors, which likely preferentially drive transcription of eIF4E in M4 and M5 AML. In addition to its role in the control of eIF4E expression as outlined here, NF-κB can also regulate eIF4E activity indirectly by modulating its subcellular localization and thus affecting eIF4E levels and activity (19).

In summary, we observe in normal primary hematopoietic cells and in cell lines with normal NF-κB activity, that EIF4E transcription is stimulated by NF-κB activation. We demonstrate that EIF4E is a direct transcriptional target of NF-κB. We observe that in the M4 and M5 AML specimens examined, the eIF4E promoter is highly occupied by NF-κB complexes. These studies elucidate a novel mechanism of transcriptional control for EIF4E and thus potentially a new point at which to target it.

Key Points.

  1. eIF4E is a direct inducible transcriptional target of NF-κB. Its upregulation is abrogated with genetic or pharmacological NF-κB inhibition

  2. NF-κB in M4/M5 AML is strongly and constitutively associated with the eIF4E promoter contributing to its transcriptional upregulation.

Acknowledgements

These studies were supported by grants from the NIH (RO1 98571) and Leukemia and Lymphoma Society Translational Research Program to KLBB, the Ride to Conquer Cancer and the Canadian Foundation for Innovation to KKM. FH is supported by PhD studentships from the Fonds de la recherche en santé du Québec (FRSQ) and the Cole Foundation. MA is a fellow of the Leukemia and Lymphoma Society. KKM is a Chercheur Boursier of the FRSQ. KLBB holds a Canada Research Chair. The Institute for Research in Immunology and Cancer receives infrastructure support from the CIHR and FRSQ. Clinical specimens were collected and analyzed by the BCLQ of the Reseau de recherche sur le cancer of the FRSQ, which is affiliated with the Canadian Tumour Repository Network.

Abbreviations Footnote

NF-κB

Nuclear factor κ in activated B Lymphocytes

eIF4E

eukaryotic translation initiation factor 4E

PMA

phorbol 12-myristate-13-acetate

PBMCs

peripheral blood mononuclear cells

AML

Acute Myeloid Leukemia

Footnotes

Authorship Contributions: FH, MA and KLBB designed research. FH, MA, LV and THN performed research. FH, MA, BCK and KLBB analyzed data. KKM and JH contributed valuable reagents. FH and KLBB wrote the manuscript.

Conflict of interest disclosure: The Authors declare no competing financial interest.

References

  • 1.Borden KL, Culjkovic-Kraljacic B. Ribavirin as an anti-cancer therapy: acute myeloid leukemia and beyond? Leuk Lymphoma. 2010 Oct;51(10):1805–1815. doi: 10.3109/10428194.2010.496506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lazaris-Karatzas A, Montine KS, Sonenberg N. Malignant transformation by a eukaryotic initiation factor subunit that binds to mRNA 5' cap. Nature. 1990 Jun 7;345(6275):544–547. doi: 10.1038/345544a0. [DOI] [PubMed] [Google Scholar]
  • 3.Polunovsky VA, Rosenwald IB, Tan AT, White J, Chiang L, Sonenberg N, et al. Translational control of programmed cell death: eukaryotic translation initiation factor 4E blocks apoptosis in growth-factor-restricted fibroblasts with physiologically expressed or deregulated Myc. Mol Cell Biol. 1996 Nov;16(11):6573–6581. doi: 10.1128/mcb.16.11.6573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Graff JR, Boghaert ER, De Benedetti A, Tudor DL, Zimmer CC, Chan SK, et al. Reduction of translation initiation factor 4E decreases the malignancy of ras-transformed cloned rat embryo fibroblasts. Int J Cancer. 1995 Jan 17;60(2):255–263. doi: 10.1002/ijc.2910600221. [DOI] [PubMed] [Google Scholar]
  • 5.De Benedetti A, Rhoads RE. Overexpression of eukaryotic protein synthesis initiation factor 4E in HeLa cells results in aberrant growth and morphology. Proc Natl Acad Sci U S A. 1990 Nov;87(21):8212–8216. doi: 10.1073/pnas.87.21.8212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Filipowicz W, Furuichi Y, Sierra JM, Muthukrishnan S, Shatkin AJ, Ochoa S. A protein binding the methylated 5'-terminal sequence, m7GpppN, of eukaryotic messenger RNA. Proc Natl Acad Sci U S A. 1976 May;73(5):1559–1563. doi: 10.1073/pnas.73.5.1559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Culjkovic B, Topisirovic I, Borden KL. Controlling gene expression through RNA regulons: the role of the eukaryotic translation initiation factor eIF4E. Cell Cycle. 2007 Jan 1;6(1):65–69. doi: 10.4161/cc.6.1.3688. [DOI] [PubMed] [Google Scholar]
  • 8.Gingras AC, Raught B, Sonenberg N. eIF4 initiation factors: effectors of mRNA recruitment to ribosomes and regulators of translation. Annu Rev Biochem. 1999;68:913–963. doi: 10.1146/annurev.biochem.68.1.913. [DOI] [PubMed] [Google Scholar]
  • 9.Culjkovic B, Topisirovic I, Skrabanek L, Ruiz-Gutierrez M, Borden KL. eIF4E is a central node of an RNA regulon that governs cellular proliferation. J Cell Biol. 2006 Nov 6;175(3):415–426. doi: 10.1083/jcb.200607020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Assouline S, Culjkovic B, Cocolakis E, Rousseau C, Beslu N, Amri A, et al. Molecular targeting of the oncogene eIF4E in acute myeloid leukemia (AML): a proof-of-principle clinical trial with ribavirin. Blood. 2009 Jul 9;114(2):257–260. doi: 10.1182/blood-2009-02-205153. [DOI] [PubMed] [Google Scholar]
  • 11.Oridate N, Kim HJ, Xu X, Lotan R. Growth inhibition of head and neck squamous carcinoma cells by small interfering RNAs targeting eIF4E or cyclin D1 alone or combined with cisplatin. Cancer Biol Ther. 2005 Mar;4(3):318–323. doi: 10.4161/cbt.4.3.1504. [DOI] [PubMed] [Google Scholar]
  • 12.Kentsis A, Topisirovic I, Culjkovic B, Shao L, Borden KL. Ribavirin suppresses eIF4E-mediated oncogenic transformation by physical mimicry of the 7-methyl guanosine mRNA cap. Proc Natl Acad Sci U S A. 2004 Dec 28;101(52):18105–18110. doi: 10.1073/pnas.0406927102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Graff JR, Konicek BW, Vincent TM, Lynch RL, Monteith D, Weir SN, et al. Therapeutic suppression of translation initiation factor eIF4E expression reduces tumor growth without toxicity. J Clin Invest. 2007 Sep;117(9):2638–2648. doi: 10.1172/JCI32044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Jones RM, Branda J, Johnston KA, Polymenis M, Gadd M, Rustgi A, et al. An essential E box in the promoter of the gene encoding the mRNA cap-binding protein (eukaryotic initiation factor 4E) is a target for activation by c-myc. Mol Cell Biol. 1996 Sep;16(9):4754–4764. doi: 10.1128/mcb.16.9.4754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Mainwaring LA, Kenney AM. Divergent functions for eIF4E and S6 kinase by sonic hedgehog mitogenic signaling in the developing cerebellum. Oncogene. 2011 Apr 14;30(15):1784–1797. doi: 10.1038/onc.2010.564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Zhu N, Gu L, Findley HW, Zhou M. Transcriptional repression of the eukaryotic initiation factor 4E gene by wild type p53. Biochem Biophys Res Commun. 2005 Oct 7;335(4):1272–1279. doi: 10.1016/j.bbrc.2005.08.026. [DOI] [PubMed] [Google Scholar]
  • 17.Bush A, Mateyak M, Dugan K, Obaya A, Adachi S, Sedivy J, et al. c-myc null cells misregulate cad and gadd45 but not other proposed c-Myc targets. Genes Dev. 1998 Dec 15;12(24):3797–3802. doi: 10.1101/gad.12.24.3797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Khanna-Gupta A, Abayasekara N, Levine M, Sun H, Virgilio M, Nia N, et al. Up-regulation of Translation Eukaryotic Initiation Factor 4E in Nucleophosmin 1 Haploinsufficient Cells Results in Changes in CCAAT Enhancer-binding Protein alpha Activity: IMPLICATIONS IN MYELODYSPLASTIC SYNDROME AND ACUTE MYELOID LEUKEMIA. J Biol Chem. 2012 Sep 21;287(39):32728–32737. doi: 10.1074/jbc.M112.373274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Topisirovic I, Guzman ML, McConnell MJ, Licht JD, Culjkovic B, Neering SJ, et al. Aberrant eukaryotic translation initiation factor 4E-dependent mRNA transport impedes hematopoietic differentiation and contributes to leukemogenesis. Mol Cell Biol. 2003 Dec;23(24):8992–9002. doi: 10.1128/MCB.23.24.8992-9002.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Gilmore TD. Introduction to NF-kappaB: players, pathways, perspectives. Oncogene. 2006 Oct 30;25(51):6680–6684. doi: 10.1038/sj.onc.1209954. [DOI] [PubMed] [Google Scholar]
  • 21.Guzman ML, Neering SJ, Upchurch D, Grimes B, Howard DS, Rizzieri DA, et al. Nuclear factor-kappaB is constitutively activated in primitive human acute myelogenous leukemia cells. Blood. 2001 Oct 15;98(8):2301–2307. doi: 10.1182/blood.v98.8.2301. [DOI] [PubMed] [Google Scholar]
  • 22.Karin M. NF-kappaB as a critical link between inflammation and cancer. Cold Spring Harb Perspect Biol. 2009 Nov;1(5):a000141. doi: 10.1101/cshperspect.a000141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Gerondakis S, Siebenlist U. Roles of the NF-kappaB pathway in lymphocyte development and function. Cold Spring Harb Perspect Biol. 2010 May;2(5):a000182. doi: 10.1101/cshperspect.a000182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Quandt K, Frech K, Karas H, Wingender E, Werner T. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 1995 Dec 11;23(23):4878–4884. doi: 10.1093/nar/23.23.4878. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001 May 1;29(9):e45. doi: 10.1093/nar/29.9.e45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kraljacic BC, Arguello M, Amri A, Cormack G, Borden K. Inhibition of eIF4E with ribavirin cooperates with common chemotherapies in primary acute myeloid leukemia specimens. Leukemia. 2011 Jul;25(7):1197–1200. doi: 10.1038/leu.2011.57. [DOI] [PubMed] [Google Scholar]
  • 27.Chin M, Herscovitch M, Zhang N, Waxman DJ, Gilmore TD. Overexpression of an activated REL mutant enhances the transformed state of the human B-lymphoma BJAB cell line and alters its gene expression profile. Oncogene. 2009 May 21;28(20):2100–2111. doi: 10.1038/onc.2009.74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mori N, Yamada Y, Ikeda S, Yamasaki Y, Tsukasaki K, Tanaka Y, et al. Bay 11-7082 inhibits transcription factor NF-kappaB and induces apoptosis of HTLV-I-infected T-cell lines and primary adult T-cell leukemia cells. Blood. 2002 Sep 1;100(5):1828–1834. doi: 10.1182/blood-2002-01-0151. [DOI] [PubMed] [Google Scholar]
  • 29.Makhlouf AA, Namboodiri AM, McDermott PJ. Transcriptional regulation of the rat eIF4E gene in cardiac muscle cells: the role of specific elements in the promoter region. Gene. 2001 Apr 4;267(1):1–12. doi: 10.1016/s0378-1119(01)00399-7. [DOI] [PubMed] [Google Scholar]
  • 30.Sen R, Baltimore D. Inducibility of kappa immunoglobulin enhancer-binding protein Nf-kappa B by a posttranslational mechanism. Cell. 1986 Dec 26;47(6):921–928. doi: 10.1016/0092-8674(86)90807-x. [DOI] [PubMed] [Google Scholar]
  • 31.Chen LF, Mu Y, Greene WC. Acetylation of RelA at discrete sites regulates distinct nuclear functions of NF-kappaB. EMBO J. 2002 Dec 2;21(23):6539–6548. doi: 10.1093/emboj/cdf660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Gruss HJ, Brach MA, Drexler HG, Bonifer R, Mertelsmann RH, Herrmann F. Expression of cytokine genes, cytokine receptor genes, and transcription factors in cultured Hodgkin and Reed-Sternberg cells. Cancer Res. 1992 Jun 15;52(12):3353–3360. [PubMed] [Google Scholar]
  • 33.Odero MD, Zeleznik-Le NJ, Chinwalla V, Rowley JD. Cytogenetic and molecular analysis of the acute monocytic leukemia cell line THP-1 with an MLL-AF9 translocation. Genes Chromosomes Cancer. 2000 Dec;29(4):333–338. [PubMed] [Google Scholar]
  • 34.Mufson RA, Myers C, Turpin JA, Meltzer M. Phorbol ester reduces constitutive nuclear NF kappa B and inhibits HIV-1 production in mature human monocytic cells. J Leukoc Biol. 1992 Dec;52(6):637–644. doi: 10.1002/jlb.52.6.637. [DOI] [PubMed] [Google Scholar]
  • 35.Landry B, Aliabadi HM, Samuel A, Gul-Uludag H, Jiang X, Kutsch O, et al. Effective non-viral delivery of siRNA to acute myeloid leukemia cells with lipid-substituted polyethylenimines. PLoS One. 2012;7(8):e44197. doi: 10.1371/journal.pone.0044197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Pettersson F, Yau C, Dobocan MC, Culjkovic-Kraljacic B, Retrouvay H, Puckett R, et al. Ribavirin Treatment Effects on Breast Cancers Overexpressing eIF4E, a Biomarker with Prognostic Specificity for Luminal B-Type Breast Cancer. Clin Cancer Res. 2011 May 1;17(9):2874–2884. doi: 10.1158/1078-0432.CCR-10-2334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kentsis A, Volpon L, Topisirovic I, Soll CE, Culjkovic B, Shao L, et al. Further evidence that ribavirin interacts with eIF4E. RNA. 2005 Dec;11(12):1762–1766. doi: 10.1261/rna.2238705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Cohen N, Sharma M, Kentsis A, Perez JM, Strudwick S, Borden KL. PML RING suppresses oncogenic transformation by reducing the affinity of eIF4E for mRNA. EMBO J. 2001 Aug 15;20(16):4547–4559. doi: 10.1093/emboj/20.16.4547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Wan F, Lenardo MJ. The nuclear signaling of NF-kappaB: current knowledge, new insights, and future perspectives. Cell Res. 2010 Jan;20(1):24–33. doi: 10.1038/cr.2009.137. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES