Translation initiation and activity of eukaryotic initiation factor-alpha (eIF2α), the rate-limiting step of translation initiation, is often overactivated in malignant cells. Here, we investigated the regulation and role of eIF2α in acute promyelocytic (APL) and acute myeloid leukemia (AML) cells in response to all-trans retinoic acid (ATRA) and arsenic trioxide (ATO), the front-line therapies in APL. ATRA and ATO induce Ser-51 phosphorylation (inactivation) of eIF2α, through the induction of protein kinase C delta (PKCδ) and PKR, but not other eIF2α kinases, such as GCN2 and PERK in APL (NB4) and AML cells (HL60, U937, and THP-1). Inhibition of eIF2α reduced the expression of cellular proteins that are involved in apoptosis (DAP5/p97), cell cycle (p21Waf1/Cip1), differentiation (TG2) and induced those regulating proliferation (c-myc) and survival (p70S6K). PI3K/Akt/mTOR pathway is involved in regulation of eIF2α through PKCδ/PKR axis. PKCδ and p-eIF2α protein expression levels revealed a significant association between the reduced levels of PKCδ ( 𝑃 = 0 . 0 3 7 8 ) and peIF2 ( 𝑃 = 0 . 0 0 4 1 ) and relapses in AML patients ( 𝑛 = 4 7 ). In conclusion, our study provides the first evidence that PKCδ regulates/inhibits eIF2α through induction of PKR in AML cells and reveals a novel signaling mechanism regulating translation initiation.

1. Introduction

Differentiation block or arrest is one of the major characteristics of acute myeloid leukemia (AML) [1]. All-trans retinoic acid (ATRA), an active metabolite of vitamin A, is a potent inducer of cellular differentiation and growth arrest in various tumor cell lines and has been successfully used in the treatment of acute promyelocytic leukemia (APL) [15]. The success of ATRA in the treatment of APL introduced the concept of differentiation therapy in treating malignant diseases [1]. Arsenic trioxide (ATO), an FDA approved drug, induces both differentiation and apoptosis in APL and AML cells [5]. The molecular events that are involved in underlying mechanism of these drugs are not completely elucidated. Understanding the pathways regulating cell proliferation and differentiation may help designing new molecularly targeted therapies in AML.

Translation initiation is a highly regulated process of translation in response to cellular stress and mitogenic stimulation [611]. Increased translation and protein synthesis are associated with cell proliferation and malignant disease [6, 7]. Translational regulation plays a vital role in the expression of oncogenic, and growth-regulatory, differentiation, and apoptosis related proteins and is considered one of the important but understudied feature of malignant phenotype [610, 12, 13].

Increased activity of eukaryotic translation initiation factor-2α (eIF2α) is the rate-limiting step of translation initiation and phosphorylation of eIF2α at serine 51 converts eIF2 to a competitive inhibitor of eIF2B, resulting in the inhibition of translation [6, 1316]. Transfection of cells with eIF2α has been shown to cause malignant transformation of normal cells, suggesting that eIF2α plays a critical role in cellular pathways controlling cell proliferation [10, 11, 1727]. Phosphorylation of eIF2α on serine 51 (Ser51) by eIF2α kinases, such as PKR, GCN2, and PERK, leads to the increased affinity of eIF2α for eIF2B and converts the phosphorylated eIF2α into an inhibitor of the GDP-GTP exchange factor, thereby inhibiting eIF2α activity and translation initiation [14]. While reducing global translation, phosphorylation of eIF2α also induces preferential translation of specific mRNAs that assist in the regulation of genes involved in metabolism and apoptosis [25].

We and others reported that ATRA and ATO inhibit translation initiation through multiple posttranscriptional mechanisms, including downregulation of translation factors and upregulation of repressors of translation initiation, such as PDCD4 and DAP5/p97 in APL cells [28, 29]. However, the posttranscriptional mechanisms regulating in APL and AML cells remain largely unknown.

Protein kinase C (PKC) is a family of serine/threonine protein kinases that are key regulatory enzymes in signal transduction [30]. The PKC family is divided in three groups, based on the differences in their sequence homology and cofactors required for their activation. The conventional PKCs (α, βI, βII, and γ) are activated by calcium and 1,2-diacyl-sn-glycerol (DAG), whereas the novel class of PKCs ( 𝛿 , 𝜀 , 𝜃 , 𝜂 ) are calcium independent but DAG dependent. The atypical PKCs ( 𝜆 , 𝜁 , and 𝜐 ) do not require calcium or DAG for their activation. Depending on the cell type, PKC 𝛿 can function as a tumor suppressor, a proapoptotic factor, and can regulate cell proliferation and cell survival functions [30]. The role of PKC 𝛿 in regulation of translational machinery is not well understood.

In the present study, we investigated the regulation of eIF2 𝛼 in APL and AML cells. We found that PKC 𝛿 regulates eIF2 𝛼 activity by phosphorylating it at Ser-51 through PKR, an eIF2 𝛼 kinase. We also found PI3K/Akt/mTOR pathway is involved in regulation of eIF2 𝛼 through PKC 𝛿 /PKR. Overall, our data provided the first evidence that PKCδ regulates phosphorylation/activity of eIF2α through PKR in APL and AML cells, revealing a novel role of PKCδ signaling and regulation of translation initiation.

2. Materials and Methods

2.1. Cell Lines and Culture Conditions

The human promyelocytic cell line NB4, (AML-M3 type by the FAB classification) harboring t(15;17), was purchased from German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). AML cell lines including myeloblastotic HL60 (M2-AML), myelomonocytic U937 (M4/M5-AML), and THP-1 (M4-AML) cells were purchased from American Type Culture Collection (Manassas, VA). Primary human hematopoietic progenitor cells (CD34+ bone marrow progenitor cells) were purchased from Cambrex Bioscience Inc. (Walkersville, CA). The cells were grown in RPMI 1640 medium (Life Technologies, Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum at 37°C under 5% CO2 in a humidified incubator. ATRA, arsenic trioxide (ATO), and rottlerin were purchased from Calbiochem, La Jolla, CA.

2.2. Evaluation of Cell Differentiation

Cell differentiation was identified by examining expression of granulocytic (CD11b and CD11c) and monocytic (CD14) differentiation markers, morphologic changes, electron microscopy images, and reformation of PML nuclear bodies. Cells were collected from 2 to 5 days after treatment and washed with phosphate-buffered saline solution (PBS). Cells (5 × 105) in 100 μL of PBS were incubated for 30 min with fluorescein isothiocyanate (FITC)—labeled or phycoerythrin (PE)—labeled anti-CD11b, anti-CD11c, or anti-CD14 antibodies (1 : 200; BD Biosciences, San Jose) and for 20 min with PE-labeled isotope control IgG2a (Becton Dickinson) on ice in the dark, as described by the manufacturer. The percentages of CD11b+, CD11c+, and CD14+ cells were determined by fluorescence-activated cell sorting (FACS) analysis at the Flow Cytometry and Cellular Imaging Facility, The University of Texas MD Anderson Cancer Center. Background staining was determined from the cells stained with the isotype control antibodies.

2.3. Cell Growth Assays

Cells were seeded at 1 × 105 cells/mL in RPMI medium in six-well tissue culture plates (Costar, Cambridge, MA). After dilution with saline from a 10 mM stock in DMSO, cells were treated with ATRA at a final concentration of 1 μM and incubated for 3 days. The maximum concentration of DMSO was kept at less than 0.001% (v/v). Cell viability was determined by two methods: trypan blue (Sigma) exclusion and the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) dye reduction assay [31]. Briefly, after cells were incubated with ATRA or ATO, 10 μL of MTT solution (10 mg/mL in PBS) was added to each well (in triplicate). The plates were then incubated for 4 h at 37°C, and the formazan crystals formed were dissolved by adding 100 μL of 0.04 N HCl in 2-propanol. Plates were read at 490 nm by a microplate reader (Molecular Devices, Sunnyvale, CA). In the trypan blue exclusion test, cells with intact membranes exclude the dye, whereas cells without intact membranes take up the coloring agent. For this assay, a cell suspension was prepared and mixed with an equal amount of a 0.4% trypan blue solution. Cell viability was assessed within 1-2 min by calculating the percentage of unstained (i.e., viable) cells. Untreated controls were used to determine the relative viability in both assays.

2.4. Western Blot Analysis

NB4, HL60, THP-1, and U937 cells were harvested from exponentially growing cell cultures. After treatment, the cells were collected, centrifuged, and lysed with a lysis buffer. Total protein concentration of the resulting whole-cell lysates was determined using a DC protein assay kit (Bio-Rad, Hercules, CA). In the experiments in which the PKCδ inhibitor rottlerin (4 μM) were used, the cells were incubated with the inhibitor for 4 h and then with ATRA for the indicated time periods. Aliquots containing 30 μg of total protein from each sample were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and electrotransferred to the membranes as described previously [32]. The membranes were blocked with 5% dry milk in TBST (100 mM Tris-HCL [pH 8.0], 150 mM NaCl, and 0.05% Tween-20), probed with primary antibodies diluted in TBST containing 5% dry milk, and incubated at 4°C overnight. We used primary antibodies against eIF2, phosphorylated (p) eIF2e (Ser51), p-eIF4E (Ser209), p-4E-BP1 (Thr37/40), Akt (Ser473), P-PERP (The981), p-P70S6K (Thr421/Ser424) (all, Cell Signaling Technology, Danvers, MA), and p-PKR (Thr446) (Santa Cruz Biotechnology, Santa Cruz, CA); p 2 1 W a f 1 / C i p 1 , tissue transglutaminase 2 (TG2), and GCN2 also were purchased from Santa Cruz Biotechnology. eIF2α antibodies were diluted 1 : 1,000 in TBST. After being washed, the membranes were incubated with horseradish peroxidase-conjugated anti-rabbit secondary antibody (Amersham Life Science, Cleveland, OH). Mouse anti-β-actin and donkey anti-mouse secondary antibodies were purchased from Sigma to examine β-actin expression for equal loading. The bands were visualized by the enhanced chemiluminescence method (KPL, Gaithersburg, MD). Images were scanned and quantitated by a densitometer using the Alpha Imager application program (both from Alpha Innotech, San Leandro, CA). All experiments including treatments were repeated at least three times.

2.5. RNA Isolation and RT-PCR Analysis

Cells were seeded in six-well plates (1 × 106 cells/mL) and treated with ATRA at a final concentration of 1 μM. The cells were collected at various time points, and total cellular RNA was isolated with Trizol reagent (Life Technologies). cDNA was obtained from 5 μg of total RNA using a Superscript II RT kit (Life Technologies) as previously described [32]. Briefly, 5 μL of the total 20 μL of reverse-transcribed product was used for polymerase chain reaction (PCR) in 1× PCR buffer containing 1.5 mM MgCl2, 250 μM dNTPs, 0.5 units of Taq polymerase (Life Technologies), and 100 ng of primers for PKCδ (Santa Cruz Biotechnology) or β-actin (Sigma-Genosys, Houston, TX). The following programs were used for PCR amplification of PDCD4: 1 cycle at 94°C for 2 min; 35 cycles of denaturation at 94°C for 1 min; annealing at 55–65°C for 1 min; extension at 72°C for 1 min. A cycle of 72°C for 7 min was added to complete the reaction. The reaction products were analyzed on 2% agarose gels containing ethidium bromide, and cDNA synthesis was verified by detection of the β-actin transcript.

2.6. Knockdown of PKCδ and eIF2α by siRNA

Targeted downregulation of eIF2 𝛼 and PKC 𝛿 was achieved by using double-stranded small-interfering RNA (siRNA), which were purchased from Santa Cruz Biotechnology and Invitrogen Inc. The control and FITC-labeled siRNA were purchased from Qiagen (Valencia, CA). Transfection of siRNA was performed by using an optimized nucleofection protocol according to the manufacturer’s instructions (Amaxa Inc., Gaithersburg, MD). Exponentially growing NB4 cells were harvested and 2 × 106 cells were used for siRNA transfection experiments. Cells were also transfected with control (nonsilencing) siRNA [35]. Under these conditions, we consistently reached a transfection efficiency of 70% without significant reduction of viability in the cell lines. Untransfected cells and cells treated with transfection reagent or control siRNA were used as controls. Target protein expression was determined 24, 48, 72, and 96 h after transfection by Western blot analysis. Fresh medium containing 1 μM ATRA was added for the analysis at 72 h. After the treatment, the cells were harvested to assess differentiation markers by FACS analysis.

2.7. Reverse Phase Protein Array (RPPA)

The assay was performed at MD Anderson Cancer Center in collaboration with Dr. Steven Kornblau using AML samples under an approval protocol and consent. Paired samples from the same patients isolated at the time of diagnosis and relapse from the same patients were analyzed by RPPA assay as previously described [36, 37].

2.7.1. Statistical Analysis

The results are expressed as means ± standard deviations of three or more experiments. Statistical analysis was performed using the two-tailed Student’s t-test for paired data. P values less than 0.05 were considered significant. Comparison of the protein levels detected by reverse phase protein array (RPPA) in pairs (each pair are from the same patient) of “newly diagnosed” and “relapsed” AML Samples was analyzed by paired t-test to assess the difference.

3. Results

3.1. Evaluation of ATRA- and ATO-Induced Effects

We first determined the effects of ATTA and ATO on cell differentiation and apoptosis of NB4 cells. NB4 cells (M3-AML) are true APL cells that express t(15 : 17) PML-RARα. NB4 cells that were treated with ATRA (1 μM) at the indicated time points underwent granulocytic differentiation, as indicated by induction of CD11b and CD11c expression detected by FACS analysis (Figures 1(a), 1(b), and 1(c)). Induction of differentiation was also evidenced by PML nuclear body reformation (data not shown), May-Grünwald-Giemsa staining and transmission electron microscopy that ATRA-treated cells acquired granulocytic morphology: decreased nuclear/cytoplasm ratio, the appearance of cytoplasmic granules, and nuclear lobulation (Supplementary Figure 1). ATO induced growth inhibition by cell toxicity assay (Figure 1(d)) and apoptosis by activation of caspases 9 and 3 as well as cleavage of PARP was detected (Supplementary Figure 2) [5, 38].

3.2. ATRA and ATO Induce Ser-51 Phosphorylation of eIF2α in APL Cells

Previous studies suggest that growth inhibition and terminal differentiation are associated with inhibition of total protein synthesis and translational suppression by differentiation-inducing agents in APL, AML, and other cells [28, 33, 34, 39, 40]. We have previously reported the evidence that ATRA modulates the expression of proteins involved in suppression of translation initiation during granulocytic differentiation of APL cells [3234]. Phosphorylation of serine 51 on the alpha subunit of eukaryotic initiation factor eIF2 is a well-documented mechanism of inhibition of translation and global inhibition of protein synthesis under a variety of conditions in variety of cells including APL and AML cells [1416, 40, 41]. Therefore, we first investigated whether ATRA induces phosphorylation of eIF2 𝛼 for inhibition of translation and protein expression in APL cells. ATRA treatment induced marked phosphorylation (Ser 51) of eIF2 𝛼 in the cells at the differentiation-inducing concentrations of ATRA (0.1 or 1  𝜇 M) in NB4 cells (Figure 2(a)). ATRA-induced phosphorylation of eIF2α started at 24 h, reached its maximum at 48 h to 72 h of treatment, and correlated well with the level of differentiation observed in NB4 cells (Figure 1(a)). Total and unphosphorylated eIF2α expression were also elevated during differentiation. Densitometric analysis showed that p-eIF2α/eIF2α and p-eIF2α/β-actin ratios were increased about 7- to 10-fold (Figure 2(a), lower panel). Exposure of NB4 cells to ATO (0.4 and 1 μM) for 6–72 h also resulted in Ser-51 phosphorylation of eIF2α in NB4 cells as indicated by Western blot analysis (Figures 2(b) and 2(c)), suggesting that ATRA and ATO have similar effects in inducing phosphorylation of eIF2α.

3.3. ATRA- and ATO-Induced Phosphorylation of eIF2α Is Mediated PKR but Not PERK and GCN2

PKR is an eIF2α kinase that is known to phosphorylates eIF2α on the Ser51 residue [25, 4244]. To determine whether ATRA- and ATO-induced phosphorylation of eIF2α is mediated by PKR activity in APL cells we examined PKR expression. We found that ATRA and ATO induce marked PKR expression in NB4 cells (Figure 3(a)). We also observed activity of PKR as indicated by p-PKR (Thr448) was also induced (Figure 3(b)). We also observed similar effects in HL60 cells that undergo granulocytic differentiation by ATRA and ATO (Figure 3(c)).

GCN2 and PERK induce eIF2α phosphorylation in response to stress and unfolded protein response (UPR) [25]. We next examined whether other eIF2α kinases, including GCN2 and PERK, are involved eIF2α regulation during ATRA. Neither ATRA nor ATO treatment resulted in induction of GCN2 and PERK in NB4 cells (Figures 3(d) and 3(e)). We did not detect any change in PERK levels, suggesting no UPR type of response is involved in the process (Figure 3(e)).

To show a link between PKR and phosphorylation of eIF2α, we knocked down PKR by a specific siRNA and found that inhibition of PKR completely blocked ATRA-induced phosphorylation (Ser51) of eIF2α in NB4 cells (Figure 3(f)). To further eliminate possibility that PERK is not involved in ATRA-induced regulation of eIF2α, we knocked down of PERK by a specific siRNA and did not detect any change in phosphorylation status of eIF2α (Supplementary Figure 3). Overall, these findings suggested that PKR plays a major role in the phosphorylation of eIF2α.

3.4. ATRA and ATO Induce p-Ser51 eIF2α and PKR in AML Cells

We next investigated whether ATRA induces phosphorylation of eIF2α in AML cells during terminal differentiation. It is well established that ATRA induces monocytic differentiation of U937 (M4/M5-AML) [45] and THP-1 (M4-AML) [46, 47] cells. Monocytic differentiation of ATRA (1 μM) treatment of U937 and THP-1 cells was shown by CD14 expression by FACS analysis (Figure 4(a)). ATRA (1 μM) treatment of THP-1 and U937 cells led to an induction of phosphorylation (Ser51) of eIF2α, which is associated with marked PKR expression (Figures 4(b) and 4(c)). ATO (0.4 uM) also induced p-Ser51 eIF2α and PKR expression. These findings provided further evidence that phosphorylation of eIF2α is regulated by PKR by ATRA and ATO.

3.5. PKCδ Is Involved in Ser51-Phosphorylation of eIF2α in APL Cells

To identify the signaling mechanism that regulates/activates PKR in response to ATRA or ATO we examined PKCδ, a serine/threonine kinase that is induced during ATRA-induced differentiation in APL cells [30, 48]. We found that ATRA and ATO induce PKCδ expression, which is closely correlated with the increased phosphorylation (Ser 51) of eIF2α in NB4 cells by Western blot analysis (Figure 5(a)). Upregulation of PKCδ expression was also associated with phosphorylation of PKCδ on threonine 505 (Thr-505) in the activation loop (Figure 5(b)). RT-PCR analysis suggested that ATRA (1 μM) induces PKCδ expression at the transcriptional level (Figure 5(c)). ATRA also induced PKCδ expression also correlated with increased phosphorylation of eIF2α in U937 AML cells (Figure 5(d)), indicating that PKC activation is not APL cell line specific event.

To determine whether PKCδ plays a role in the regulation of eIF2α phosphorylation (Ser-51), we inhibited PKCδ using a PKCδ inhibitor rottlerin (at 4 μM specifically inhibits PKCδ) [49, 50]. Rottlerin treatment blocked basal levels and ATRA-induced phosphorylation of eIF2α in NB4 cells (Figure 5(e)). Densitometric analysis revealed that inhibition of PKCδ led to 5-fold reduction in p-eIF2α levels (lower panel). Overall, these findings suggest that PKCδ is involved in regulation of phosphorylation of eIF2α.

3.6. PKCδ Regulates Phosphorylation of eIF2α in HSC and AML Cells

We also investigated whether PKCδ regulates eIF2α activity in normal human hematopoietic stem cells (HSC) isolated from bone marrow of a healthy donor. Western blot analysis showed that inhibition of PKCδ by rottlerin (4 μM) markedly blocked ATRA-induced phosphorylation of eIF2α in HSCs, HL60 myeloblastic (AML-M2), and THP1 monocytic AML cells (AML-M5) (Figures 6(a)6(c)). Rottlerin treatment also caused a slight inhibition of total eIF2α. These findings indicated that PKCδ plays a role in regulation of eIF2α not only in APL and AML as well as normal HSCs.

3.7. PKCδ Regulates Phosphorylation of eIF2α through PKR

PKR, a component of the antiviral defense, is of the kinases known to phosphorylate eIF2α at Ser51 to inhibit its activity and translation initiation and protein synthesis [25]. Because our data previously indicated that inhibition of PKR (Figure 3(f)) or PKCδ (Figure 5(d)) prevented basal and ATRA-induced phosphorylation of eIF2α, we hypothesized that PKCδ, a Serine/Threonine kinase, may be responsible for activation of PKR through phosphorylation on Thr446. Therefore, we next examined whether PKCδ regulates PKR phosphorylation. Pretreatment of NB4 cells with PKCδ inhibitor rottlerin significantly inhibited ATRA-induced Thr446-phosphorylation of PKR (Figure 7(a)). The bar graph next to Figure 7(a) shows densitometric analysis of Figure 7(a). Figure 6(b) confirms that rottlerin treatment inhibited PKCδ in response to ATRA in NB4 cells. To provide further link between PKCδ and PKR we knockdown PKCδ by a specific siRNA and found that inhibition of PKCδ expression and reduced activity (phosphorylation) of PKR (Thr446) (Figure 7(c) and densitometry, right panel), indicating that PKCδ regulates PKR activity/phosphorylation in APL cells.

To evaluate the role of PKCδ signaling in cell differentiation of APL cells we specifically inhibited PKCδ and assessed ATRA-induced (48 h) differentiation in NB4 cells. Inhibition of PKCδ by rottlerin significantly inhibited ATRA-induced granulocytic differentiation in cells (Figure 8). While ATRA (1 μM for 48 h) treatment led to about 43.4% cell differentiation in NB4 cells, preincubation with rottlerin reduced ATRA-induced differentiation to 5.2% ( 𝑃 < 0 . 0 5 ). This finding suggested that PKCδ plays a key role in ATRA-induced granulocytic differentiation.

3.8. Inhibition of Mammalian Target of Rapamycin (mTOR) Induces p-(Ser51) eIF2α through PKCδ/PKR Axis

Inhibition of mammalian target of rapamycin (mTOR) signaling has been shown to potentiate the effects of ATRA and ATO to induce growth arrest and differentiation of APL (NB4) and AML cells in vitro and in vivo models [51, 52]. mTOR signaling is one of the major regulators of translation in cancer cells by altering 4EBP-1 and eIF4E activity [53]. We have previously shown that ATRA and ATO inhibit the activity PI3K/Akt/mTOR and p70S6 kinase [33, 34]. To investigate whether mTOR regulates eIF2α through PKR we inhibited the PI3K/Akt/mTOR pathway by a specific mTOR inhibitor rapamycin (20 nM). Treatment of NB4 cells with rapamycin for 24 h markedly induced p-PKR and total PKR expression that is closely associated with phosphorylation (Ser51) of eIF2α (Figure 9(a)). Inhibition of PI3K pathway by a specific inhibitor LY294002 (20 uM) [33] also resulted in increased p-(Ser51) eIF2α and PKCδ in NB4 cells (Figure 9(b)), indicating that the PI3K/Akt/mTOR pathway is involved in phosphorylation of eIF2α by PKCδ/PKR in APL cells.

3.9. eIF2α Phosphorylation Leads to Differential Expression of Important Cellular Proteins

Studies have shown that ATRA regulates important regulatory proteins such as cyclin-dependent kinase inhibitor p 2 1 W a f 1 / C i p 1 [54], death-associated protein 5 (DAP5/p97) [55, 56], c-myc [23, 57], p-P70S6K [58], and TG2 [46, 47, 59]. To gain some insight about the ATRA-mediated downstream changes in response to eIF2α regulation we knocked down eIF2α by siRNA and examined if expression of these proteins is altered. Knockdown of eIF2α inhibited ATRA-induced upregulation of p 2 1 W a f 1 / C i p 1 , DAP5/p97, and TG2 (Figure 10(a)). We also observed inhibition of ATRA-induced downregulation of c-myc and p-P70S6K, a downstream mediator of PI3K/Akt/mTOR pathway (Figure 10(a)). These findings suggest that eIF2α in fact is involved in expression of some of the proteins involved in cell cycle arrest, proliferation, survival, differentiation, and apoptosis.

3.9.1. PKCδ and p-eIF2α Protein Expressions Are Associated with Relapses in AML Patient

PKCδ and p-eIF2α protein expression was assessed by RPPA assay in 47 paired samples from newly diagnosed and relapsed AML patients to determine if the protein level changes when the disease status changes. The comparison of PKCδ (Figures 11(a) and 11(c)) and p-eIF2α protein (Figures 11(b) and 11(d)) expression and distributions of the protein levels between pairs (newly diagnosed and relapsed) were analyzed. Data suggest that there is a significant relationship between the reduced levels of PKCδ ( 𝑃 = 0 . 0 3 7 8 ) and p-eIF2α ( 𝑃 = 0 . 0 0 4 1 ) and relapses in AMP patients, suggesting that higher levels of PKCδ and p-eIF2α is associated with better response.

4. Discussion

Overactivity of translation initiation factors, such as eIF2α, eIF4E, and eIF4G, results in malignant transformation, indicating that regulation of the activity of these factors is critical in controlling survival pathways and cell proliferation [613, 22]. Previous reports including ours suggest that growth inhibition and terminal cell differentiation are associated with suppression of translation initiation [28, 3234, 39, 60]. However, molecular mechanisms responsible for regulation of eIF2α during these events have not been elucidated.

The present study provides the first evidence that PKCδ regulates activity of eIF2α through induction of PKR, an eIF2α kinase, but not PERK and GCN2, leading to inhibition of translation initiation during terminal differentiation of APL and AML cells (Figure 12). Moreover, our study suggested that PI3K/Akt pathway and its downstream mediator mTOR, a major hub of translational control, is involved in regulation of eIF2α through PKCδ/PKR axis. Overall our findings revealed a novel mechanistic insight on actions of ATRA and ATO in regulation of eIF2α, the rate limiting step of translation initiation in APL and AML cells.

PKCδ, a serine/threonine kinase, can function as a tumor suppressor and a proapoptotic factor and can regulate cell proliferation and cell survival functions [30]. Induction of PKCδ has been previously shown in APL cells [48]. PKCδ is critical to ATRA-induced terminal differentiation because inhibition of PKCδ by rottlerin resulted in almost complete blockage of ATRA-induced differentiation (Figure 8). Our data also indicated that induction of PKCδ signaling is critical for regulation of eIF2α, linking for the first time PKCδ with the regulation of translation initiation, which is often over activated in AML cells. Most importantly, our data suggested that PKCδ regulates PKR and eIF2α and thereby revealing a novel function of PKCδ.

PKR, GCN2, and PERK are eIF2α kinases that are known to phosphorylate and inhibit the activity of eIF2α [9, 25]. However, we found that ATO and ATRA treatments induce PKR but not GCN2 and PERK. The activation of unfolded protein response (UPR) is manifested by phosphorylation of protein kinase RNA-like endoplasmic reticulum kinase (PERK) and eIF2α. In our study we did not detect any change in the activity (or phosphorylation) of PERK on phosphorylation of eIF2α, suggesting that ATRA and ATO do not induce UPR response. PKR has been shown to be induced by interferon in myeloid leukemia cells [44]. Interferon is known to inhibit translation and protein synthesis during viral infections, limiting production of viral particles. It is possible that ATRA-induced interferon secretion from the cells may lead to upregulation and activation of PKR in APL by autocrine mechanisms. This hypothesis remains to be tested by future studies.

PI3K/Akt/mTOR signaling pathway is overactivated in APL and AML cells and plays an important role in proliferation, drug resistance, inhibition of apoptosis in cancer cells [50, 61]. mTOR signaling is a critical inducer of translational activity by phosphorylating 4EBP-1 and releasing eIF4E from 4EBP1 and increasing activity translation initiation complex [53]. We found that ATRA and ATO inhibits phosphorylation of 4E-BP1 in NB4 APL cells (unpublished observation). Dephosphorylated 4E-BP1 inhibits translation initiation by binding to eIF4E, which normally binds to the cap-structure of mRNA to form an initiation complex [62, 63]. Reduced phosphorylation of 4E-BP1 by ATRA facilitates its binding to eIF4E, leading to inhibition of eIF4E. We have also shown that ATRA and ATO inhibit the activity PI3K/Akt/mTOR and p70S6 kinase in APL cells [33, 34]. The current study is in agreement with our previous findings that ATRA inhibits translation initiation by multiple mechanisms, including inhibition of initiation factors and induction of PDCD4 and DAP5 (inhibitors of translation initiation), inhibition of p-4E-BP1 and EF4E (Figure 11) [3234]. DAP5 and PDCD4, a novel tumor suppressor protein, were recently identified as inhibitors of translation initiation. PDCD4 binds to eIF4A and specifically inhibits its helicase activity [64]. DAP5 functions as a repressor of translation initiation by forming translationally inactive complexes with eIF4A and eIF3 [56]. Overall, data suggest that translation initiation and protein synthesis are suppressed by several mechanisms by ATRA and ATO in APL cells.

Several specific mRNAs have been found to be selectively regulated in response to inhibition of eIF2α. These selectively upregulated proteins include ATF-2, ATF-3, ATF-4, GADD34, and CHOP/GADD153 [25, 28, 65], suggesting that even during inhibition of translation some of the mRNAs are being translated. Our study suggest that eIF2α is involved in expression of p 2 1 W a f 1 / C i p 1 , DAP5, and TG2 in response to ATRA during differentiation, supporting the hypothesis that subset of mRNAs encoding critical proteins are differentially regulated by eIF2α. On the other hand, increased expression of eIF2α in response to growth induction by c-myc [23] and transformed cells [24] suggest that eIF2α plays a critical role in regulation of cell proliferation and is strictly regulated because of its oncogenic potential.

In conclusion, a better understanding of translation initiation and posttranscriptional mechanisms may help identify novel targets for targeted therapies. Antitumor agents such as rapamycin, a specific mTOR inhibitor, or related compounds that inhibit translation by inhibiting phosphorylation of 4E-BP1 and P70S6K have been shown to induce remissions even in AML patients with relapsed disease, suggesting that the targeted inhibition of mRNA translational pathways might offer therapeutic benefits for patients with certain malignancies.

Authors’ Contribution

B. Ozpolat and U. Akar contributed equally to this work.


The authors would like to thank Pierette Lo for editing the paper. This study was supported in part by grants from the Ladies Leukemia League (B. Ozpolat) and the National Cancer Institute (U54 RFA CA096300) (G. Lopez-Berestein).

Supplementary Materials

Supplementary Figure 1: Morphology and cellular structures of NB4 cells. Cells were analyzed by May-Grünwald-Giemsa staining (upper panels) and transmission electron microscopy (lower panels), respectively, before and after 72 h of ATRA (1 μM) treatment.

Supplementary Figure 2: ATO at high concentrations induces apoptosis in APL cells. NB4 cells were treated with ATO (2 μM) for 24 h and collected for Western blot analysis. ATO induced apoptosis in NB4 cells as indicated by cleavage (activation) of caspases 9, 3 and PARP cleavage.

Supplementary Figure 3: Knockdown of PERK by siRNA has no effect on phosphorylation of eIF2α in NB4 cells. Cells were transfected with PERK or control siRNA for 48 h. 24 h after ATRA treatment cells were collected, lysed and analyzed by Western blot.

  1. Supplementary Figures