ERK and p38 inhibit the expression of 4E-BP1 repressor of translation through induction of Egr-1.

4E-BP1 plays a major role in translation by inhibiting cap-dependent translation initiation. Several reports have investigated the regulation of 4E-BP1 phosphorylation, which varies along with cell differentiation and upon various stimulations, but very little is known about the regulation of its expression. In a first part, we show that the expression of 4E-BP1 protein and transcript decreases in hematopoietic cell lines cultivated in the presence of phorbol 12-myristate 13-acetate (PMA). This decrease depends on the activation of the ERK/mitogen-activated protein kinases. 4E-BP1 expression also decreases when the p38/mitogen-activated protein kinase pathway is activated by granulocyte/macrophage colony-stimulating factor but to a lesser extent than with PMA. In a second part, we examine how 4e-bp1 promoter activity is regulated. PMA and granulocyte/macrophage colony-stimulating factor induce Egr-1 expression through ERK and p38 activation, respectively. Using a dominant negative mutant of Egr, ZnEgr, we show that this transcription factor is responsible for the inhibition of 4e-bp1 promoter activity. In a third part we show that histidine decarboxylase, whose activity and expression are inversely correlated with 4E-BP1 expression, is a potential target for the translational machinery. These data (i) are the first evidence of a new role of ERK and p38 on the translational machinery and (ii) demonstrate that 4E-BP1 is a new target for Egr-1.

Control of mRNA translation plays a pivotal role in regulating gene expression under a variety of conditions in mammalian cells (1). The predominant step in translational regulation is the initiation phase, which consists of the recruitment of the 40 S ribosomal subunit to the mRNA (2). This occurs through recognition of the 5Ј cap structure (m7GpppX, where X is any nucleotide) by the cap-binding protein complex eIF4F (eukaryotic initiation factor 4F) which, in higher eukaryotes, consists of three subunits: eIF4A, eIF4E, and eIF4G. The initiation process is largely regulated through changes in the phosphorylation state of eIFs and other components involved in this process (3,4). eIF4E activity is modulated by phosphorylation in response to mitogens, polypeptide hormones, tumor promot-ers, and growth factors in a mitogen-activated protein kinase (MAPK) 1 -MAPK signal-integrating kinase (MNK) pathway-dependent manner (5). In addition to the regulation of its phosphorylation, the activity of eIF4E is tightly controlled through reversible interaction with a family of inhibitory proteins termed 4E-BP (eIF4E-binding proteins). Of the three known proteins (4E-BP1, 4E-BP2, and 4E-BP3), 4E-BP1, also named PHAS-1, is the best characterized. 4E-BP1 specifically inhibits cap-dependent translation by competing with eIF4G for binding to the cap-binding factor eIF4E and consequently preventing the formation of the eIF4F complex (6). The affinity of the 4E-BPs to eIF4E depends on their phosphorylation state. Hypophosphorylated 4E-BPs interact with high affinity with eIF4E, whereas hyperphosphorylation of 4E-BPs, elicited by stimulation of cells with hormones, cytokines, or growth factors, results in an abrogation of eIF4E-binding activity. Activation of phosphatidylinositol 3-kinase or a downstream phosphatidylinositol 3-kinase effector, Akt/protein kinase B, and FRAP/mTOR (FKBP and rapamycin-associated protein), leads to 4E-BP1 hyperphosphorylation (7)(8)(9). Six phosphorylation sites have been identified in 4E-BP1: Thr 37 , Thr 46 , Ser 65 , Thr 70 , Ser 83 , and Ser 112 (numbering according to human 4E-BP1). FRAP/mTOR phosphorylates 4E-BP1 on Thr 37 , Thr 46 , and Thr 70 (9,12), activation of p38-MSK1 (mitogen and stress kinase 1) pathway by UV light leads to phosphorylation on Thr 37 and Ser 65 (10), and the activation of the ERK pathway induces phosphorylation on Ser 65 , Thr 37 , Thr 46 , and Thr 70 (11). A hierarchical phosphorylation of 4E-BP1 has been proposed: first on the Thr 37 and Thr 46 and then on Thr 70 and Ser 65 (12), showing that multiple phosphorylation events (most likely via different kinases) are required to release 4E-BP1 from eIF-4E. Recently it has been shown that 4E-BP1 cleavage by caspase is a new step in the regulation of translation in response to insulin (13).
A differential regulation of the translational machinery during human myeloid differentiation has been reported. Induction of HL-60 and U-987 cell differentiation by PMA or interferon-␥ into monocytic/macrophages results in a dephosphorylation and consequent activation of 4E-BP1. In contrast induction of HL-60 into the granulocytic differentiation by Me 2 SO decreases 4E-BP1 expression level, whereas it increases 4E-BP2 expression level (44,45). Expression of 4E-BP2 is down-regulated during thymocyte maturation (46), but nothing is known about the mechanisms by which 4E-BP expression is modified. We derived from the pluripotent UT7 cell line a subpopulation, the UT7D1 cell line, which in the presence of the growth factor GM-CSF spontaneously expresses the mRNA coding for histidine decarboxylase (HDC). Stimulation of these cells during 24 h with PMA induces a basophil differentiation characterized by an induction of IL-4, IL-6, and IL-13 expression, the presentation of the basophil Bsp-1 antigen, and an increase in histamine production (47). Whereas the induction of IL-4, IL-6, and IL-13 expression is regulated at the transcriptional level, the HDC expression seems to be regulated at a post-transcriptional level (47,48). Thus we examined the activity of the translational initiation machinery in PMA-treated UT7D1 cells. Such a treatment induced a decrease in 4E-BP1 expression, but 4E-BP2 or eIF4E expressions did not change. We also have shown that 4E-BP1 expression is negatively regulated at the transcriptional level by the ERK and p38 via the induction of Egr-1 expression. Ultimately we have provided new insights into the regulation of HDC expression at the translational level.
Cloning of 4e-bp1 Promoter-A sequence of 1020 bp upstream of the ATG of 4e-bp1 gene was cloned in pCDRIVE vector (Qiagen) by PCR using the 5Ј-CGGGGGTACCCCGCCTCAAACCCCTGGGCTC-3Ј sense primer, the 5Ј-CCGCTCGAGCGGGTCTCCTGTGCGCTGCAC-3Ј antisense primer, and the BAC clone RPCI-11-701H6 (CHORI-BACPAC Resources) as matrix. This sequence was fused to the luciferase gene in the pGL2-Basic vector (Promega) using XhoI and KpnI restriction sites. Entire sequence was verified by sequencing the insert with pGL1 and pGL2 primers (Promega).
Transfection and Reporter Assays-HeLa cells were plated 24 h before transfection in 24-well plates so that they are 60 -70% confluent the day of transfection. Transfection was realized with LipofectAMINE Plus reagent (Invitrogen) according to the manufacturer's instructions. pIRES-EGFP was always cotransfected to estimate the level of transfection (fluorescence measure from lysates with the Victor2) and normalize the luciferase activity. This control level varied from 0.73 to 1.22 times. Luciferase reporter reagent (Promega) was used to measure the luciferase activity according to the manufacturer's instructions in a LB96V luminometer (Berthold Technologies).
Histamine and HDC Assays-Histamine concentrations in cell lysates were routinely determined by an automated continuous flow fluorometric technique (lower limit of sensitivity, 0.5 ng/ml), as previously described (49). Its specificity has been confirmed by radioimmunoassay (Immunotech, Marseille, France). HDC activity was measured by a radiochromatographic technique, as described before (49). Briefly, the cell pellets were resuspended in 50 mM ice-cold phosphate buffer and gently sonicated.

4E-BP1 Expression Decreases in
Response to PMA-Phosphorylation of 4E-BP1 occurs in response to mitogens or growth factors by activation of the phosphatidylinositol 3-kinase and/or MAPK signal transduction pathways and is rapamycinsensitive (10,11,50). To assess to what extend PMA modifies 4E-BP1 phosphorylation and/or expression, we analyzed by Western blot the expression and phosphorylation of 4E-BP1 in UT7D1 cells treated with or without PMA for 24 h. It should be mentioned that in all experiments UT7D1 cells were placed in fresh culture medium and GM-CSF-starved for 120 min before the addition of the inhibitors. After 30 min, we added fresh GM-CSF alone or GM-CSF and PMA. 4E-BP1 migrates in 12% SDS-polyacrylamide gels as four bands (␣, ␤, ␥, and ␦; Fig. 1A). The ␣ band is the less phosphorylated form, and the ␦ band is the highly phosphorylated isoform. According to our personal observations and in line with the literature, the ␤ band represents the 4E-BP1 isoform phosphorylated on Thr 37 and Thr 46 ; the ␥ isoform is phosphorylated on Thr 37 , Thr 46 , and Thr 70 ; and the ␦ isoform is phosphorylated on these three threonines and on Ser 65 (11,12). When the cells were cultivated with GM-CSF, the four phosphorylated forms of 4E-BP1 were present. In contrast, when the cells were treated with PMA, only the ␣, ␤, and ␥ forms of 4E-BP1 were barely detected. These results not only reflect a change in the phosphorylation of 4E-BP1 but also a decrease in 4E-BP1 protein amount (Fig.  1A). In the presence of rapamycin, an inhibitor of FRAP/ mTOR, or LY294002, an inhibitor of phosphatidylinositol 3-kinase, we observed an accumulation of the ␣ less phosphorylated form of 4E-BP1 in the absence or in the presence of PMA (Fig. 1A). These results are in line with the data from literature that show that the Thr 37 , Thr 46 , and Thr 70 are phosphorylated by the FRAP/mTOR. p38 and ERK can phos-phorylate several sites of 4E-BP1, notably the Ser 65 (10,11). This regulation did not occur in our system because treatment of the cells with SB203580 or U0126 did not have any effect on the phosphorylation pattern of 4EBP1. 2 Nevertheless we could not rule out changes that could have occurred at shorter times of treatment. Northern blot analysis showed that neither rapamycin nor LY294002 affected the transcript level of 4E-BP1, whereas the PMA application resulted in a loss of 4E-BP1 transcript (Fig. 1B). We also looked at the effect of PMA on 4E-BP1 in three different cell lines: 11OC1, HEL, and K562. Twenty-four hours of PMA treatment led to an accumulation of the less phosphorylated isoforms of 4E-BP1 in 11OC1 and K562 cells, whereas in the HEL cell line only the most highly phosphorylated form of 4E-BP1 was The results are expressed as percentages from the quantification of the bands at time 0, which is considered to be 100%. 20 g of total RNA was analyzed by Northern blotting using 4e-bp1, 4e-bp2, or GAPDH DNA probes (loading control) as indicated (B and D). The means of quantification of two independent Northern blotting analysis Ϯ S.D. is represented in E. The results are expressed as percentages from the quantification of the bands from GM-CSF, 1 day, which is considered to be 100%. detectable (Fig. 1C). In all three cell lines PMA decreased 4E-BP1 expression (Fig. 1C).
4e-bp1 Transcript Level Decreases, Whereas the Expression of 4e-bp2 Is Not Sensitive to PMA-Time course analysis showed that 4E-BP1 expression transiently decreased after 4h of GM-CSF (47 Ϯ 17%; Fig. 2, A and C) and that the ␦ phosphorylated form of 4E-BP1 disappeared after 4 h of GM-CSF treatment. The ␣ less phosphorylated form was mainly detected after 24 h of GM-CSF treatment ( Fig. 2A). The expression of 4E-BP1 protein decreased after 4 h of PMA and was undetectable after 24 h (32.5 Ϯ 13% and 4 Ϯ 1%; Fig. 2, A and C). Northern blot analysis showed that the transcript level was strictly correlated with the protein expression level, that PMA negatively regulated 4E-BP1 expression at the transcript level (Fig. 2, B and C), and that this effect lasted 48 h (Fig. 2, D and E). Because 4E-BP2 expression was also known to vary during cell differentiation, we analyzed its expression by Northern blot. The transcript of 4e-bp2 was only present at 24 h of culture in the presence or absence of PMA (Fig. 2, D and E). Taken together, these results showed that the two repressors of translation have different ways of regulating.
ERK Is Activated by PMA and p38 Is Activated by GM-CSF in UT7D1 Cell Line-PMA and GM-CSF are known to activate ERK and p38 (27,(51)(52)(53), and the balance between these two pathways may be critical in determining cell function (54). To determine the activity of ERK and p38 in our system, we analyzed their phosphorylation by Western blot using phosphospecific antibodies. As previously mentioned, UT7D1 cells were placed in fresh culture medium and GM-CSF-starved for 120 min before addition of the inhibitors. After 30 min, we added fresh GM-CSF alone or GM-CSF and PMA. Activated ERK2/p42 was detected after 30 min of PMA treatment (Fig.  3A). This activation was maximal at 2 h of treatment and decreased thereafter to become undetectable after 48 h. It should be noted that although ERK1/p44 and ERK2/p42 were found at similar levels in UT7D1 cells, the signal was always weaker for phospho-ERK1/p44 than for phospho-ERK2/p42, and the former could not be detected in all experiments (Figs. 3A and 5A). The SB203580 inhibitor of p38/SAPK2␣ and ␤ did not affect ERK phosphorylation (Fig. 3A). The PD98059, an inhibitor of MEKs, almost completely prevented the effect of PMA on ERK phosphorylation (Fig. 3A). Phosphorylated p38 was detected after 2 h of GM-CSF treatment. We tested the activation of the p38 pathway in UT7D1 lysates by performing kinase assays using Hsp25 (56). p38 activation was maximal between 4 and 24 h and decreased thereafter (Fig. 3, B and C). The total amount of p38 did not vary during our experiments. PMA did not affect p38 phosphorylation, but the application of PD98059 increased its phosphorylation at 24 h and extended it at 48 h (Fig. 3, B and C). The SB203580 completely abolished p38 pathway activation (Fig. 3C) without affecting its phosphorylation (Fig. 3B).
Inhibition of ERK Abolished the Effect of PMA on 4E-BP1 Expression, and Inhibition of p38/SAPK2␣ and ␤ Increased 4E-BP1 Expression-To analyze the role of ERK or p38 in the regulation of 4E-BP1 expression, we treated UT7D1 (Fig. 4A) or HMC1 (Fig. 4B) cells with different specific inhibitors prior to the addition of GM-CSF or PMA. Northern blot analysis showed that the application of PMA for 24 h entirely abolished 4E-BP1 transcript expression (Fig. 4A). This effect was blocked by the U0126 inhibitor of MEK (71.5% Ϯ 19) or by the bisindolylmaleimide 1 inhibitor of protein kinase C (79% Ϯ 33), a well known kinase that transduces signal from PMA to ERK cascade (Fig. 4A). The SB203580 inhibitor of p38/SAPKs2 ␣ and ␤ had no effect on 4E-BP1 expression in the presence of PMA, but in the presence of GM-CSF alone, the pretreatment with SB203580 increased the transcript level of 4E-BP1 (165 Ϯ 33%; Fig. 4A). On another cell line, the HMC1 cell line that grows independently of GM-CSF, the SB203580 had no effect at all, whereas the bisindolylmaleimide 1 and the PD98059 still reversed the PMA effect on 4E-BP1 expression (110.5 Ϯ 18% and 59 Ϯ 18%, respectively; Fig. 4B). This showed that 4E-BP1 expression is inhibited by both PMA-activated ERK and GM-CSF-activated p38.
Hyperosmolarity-induced Activation of JNK/SAPK1 Does Not Affect 4E-BP1 Expression-We next wanted to determine the effect of different stimuli able to activate the MAPK pathways on 4e-bp1 expression. Using Western blot analysis with phosphospecific antibodies, we analyzed the phosphorylation and thereby the activation level of the three main MAPK pathways after 8 h stimuli. Whereas the NaCl barely activated the ERK, it activated the p38 and strongly activated the JNK/ SAPK1. The okadaic acid (OA) inhibitor of serine/threonine phosphatases indirectly activated the ERK and strongly acti- vated the p38 but not the JNK (Fig. 5A). As shown by Northern blot analysis, the 4E-BP1 mRNA expression was not affected by 10% fetal calf serum or NaCl but decreased when UT7D1 cells were treated with OA to the same level as when cells were treated with PMA (Fig. 5B).
Cycloheximide Treatment Abolished PMA Effect on 4E-BP1 Expression-The MAPKs are known to regulate gene expression directly by phosphorylation of transcription factors or indirectly by controlling immediate early genes, which are themselves transcription factors. To distinguish which of these two kinds of transcription factors, ERK and p38, inhibit 4e-bp1 transcription, we blocked the protein synthesis with cycloheximide (CHX). When cells were cultivated in GM-CSF, CHX had no effect on the 4E-BP1 transcript level, neither at 4 h nor at 8 h (Fig. 6). In contrast, the PMA-induced inhibition of 4E-BP1 mRNA expression was completely reversed by the presence of CHX, demonstrating the necessity of protein neosynthesis for 4e-bp1 repres-sion. We addressed the question of 4e-bp1 mRNA stability by performing a time course analysis of 4e-bp1 mRNA expression in the presence of actinomycin D (10 g/ml from 0 to 270 min). The decrease of its expression was the same in the presence or the absence of PMA, showing that GM-CSF or PMA did not stabilize or destabilize 4e-bp1 mRNA. GM-CSF-starved (2 h) UT7D1 cells were cultivated in the presence or absence of PMA for different times as indicated. Where shown, the cells were preincubated (30 min) with cycloheximide (10 g/ml). 20 g of total RNA was analyzed by Northern blotting using 4e-bp1, egr-1, or GAPDH cDNA probes (loading control) as indicated.

Egr-1 Expression Is Induced by PMA or GM-CSF in an ERKor p38-dependent Manner, Respectively-Because
Egr-1 expression is known to be under the control of MAPK activities under different conditions (55,56), we analyzed its expression in UT7D1 cells. Northern blot analysis of egr-1 expression showed an induction of its transcript by PMA (Fig. 6). CHX pretreatment increased this effect (Fig. 6), suggesting that in UT7D1 cells Egr-1 exerts a negative feedback on its expression as was demonstrated (57). Western and Northern blot analysis showed that GM-CSF induced the expressions of Egr-1 protein and transcript, which were maximal at 1 h and decreased thereafter to reach the control level at 24 h (Fig. 7, A and B). The addition of PMA prolonged the Egr-1 induction until 24 h. Pretreatment of the UT7DI cells with U0126 or SB203580 abolished egr-1 induction by PMA or GM-CSF, respectively (Fig. 7C). This demonstrated that PMA-activated ERK or GM-CSF activated p38 induce Egr-1 expression.
4e-bp1 Promoter Activity-To examine whether Egr-1 is involved in ERK and/or p38 inhibition of 4e-bp1 transcription, we have cloned 1020 bp upstream of the ATG of 4e-bp1 human gene sequence as described under "Experimental Procedures." This sequence contains some potential Egr response element, Elk1, Sp1, AP4, or NFB regulatory elements (Fig. 8A). We have fused them to the luciferase gene to measure whether this potential promoter responds to ERK and p38 activities and whether this activity depends on Egr-1. HeLa cells were transfected, and luciferase assays were performed as described under "Experimental Procedures." 4e-bp1 promoter activity was almost entirely abolished when the cells were stimulated with PMA, and the U0126 inhibitor reversed this effect, whereas the SB203580 did not (Fig. 8B). These inhibitors applied together did not have more effect than the U0126 alone. The OA dimin-ished 4e-bp1 promoter activity too, and this effect was partially reversed by U0126 and significantly by SB203580. The two MAPK inhibitors had an additional effect (Fig. 8B). Coexpression of the ZnEgr dominant negative mutant of Egr did not have any effect on 4e-bp1 promoter activity in our control condition. In the presence of PMA or OA it not only reversed the PMA-or OA-induced inhibition of 4e-bp1 promoter but also increased it approximately 2-fold (Fig. 8B). Coexpression of the ZnEgr did not change the pGL2B nor the pGL2C activities, showing that the effect observed is specific to the 4e-bp1 promoter.
U0126 and SB203580 Inhibit Histamine Production and HDC Activity without Modifying hdc mRNA Expression-Stim-

FIG. 7. Egr-1 expression is induced by GM-CSF-activated p38
or PMA-activated ERK. GM-CSF-starved (2 h) UT7D1 cells were cultivated in the presence or absence of PMA for different times as indicated. Samples of cell lysates were subjected to SDS-PAGE followed by Western blotting using anti-egr-1 or anti-actin antibodies (loading control) as indicated (A). 20 g of total RNA was analyzed by Northern blotting using egr-1 or GAPDH cDNA probes (loading control) as indicated (B). In C cells are GM-CSF starved for 2 h and pretreated with U0126 (10 M) or SB203580 (10 M) before 1 h of treatment with GM-CSF alone or GM-CSF and PMA. 20 g of total RNA was analyzed by Northern blotting as described in B.
FIG. 8. Egr-1 is responsible for PMA-or OA-induced inhibition of 4e-bp1 promoter activity. Egr response element (shaded), Elk1 (underlined), Sp1(boxed), AP4 (underlined), or NFB (underlined) potential regulatory elements were found in the first 1020 bp of the sequence of the 5Ј upstream region of the human 4e-bp1 gene (A). This promoter fused to the luciferase gene (bp1-p) was transfected (as described under "Experimental Procedures") in HeLa cells and after 24 h stimulated with OA or PMA for additional 12 h. Where indicated, the cells were preincubated with U0126 (U; 10 M) or SB203580 (SB; 10 M) inhibitors. The dominant negative mutant ZnEgr was cotransfected as indicated, and the basal luciferase expression was estimated by transfection of the pGL2B vector. Luciferase activity was measured as described under "Experimental Procedures" and normalized by measuring the fluorescence of EGFP expressed by the pIRES-EGFP vector, which was always cotransfected. The data are the means Ϯ S.D. of three independent experiments performed in duplicate (B). The columns labeled with asterisks show significant differences (p Ͻ 0.01) compared with the same treatment without the different inhibitors or without the ZnEgr.
ulation of UT7D1 cells for 24 h with PMA induces a basophil differentiation characterized by an induction of IL-4, IL-6, and IL-13 expression, the presentation of the basophil Bsp-1 antigen, and an increase in histamine production (47). Whereas the induction of IL-4, IL-6, and IL-13 expression is regulated at the transcriptional level, the HDC expression seems to be regulated at a post-transcriptional level (47,48). To determine whether the post-transcriptional regulation of HDC could be translational, we examined histamine production and HDC activity and expression in UT7D1 cells cultivated with GM-CSF or GM-CSF and PMA in the presence or absence of U0126 and SB203580. As we have previously shown (47), PMA increased the intracellular histamine concentration about 10 times at 24 and 48 h (Fig. 9A). This production resulted from de novo synthesis and was the result of an increased HDC activity (47). As expected the HDC activity was increased ϳ6 and 16 times at 24 and 48 h, respectively (Fig. 9A). Whereas U0126 had no effect, the presence of SB203580 significantly decreased both the histamine production and the HDC activity when cells were cultivated in GM-CSF (Fig. 9A). When UT7D1 cells were stimulated with PMA, the histamine production and the HDC activity were significantly inhibited in the presence of U0126 (Fig. 9A). This effect seemed to be amplified when U0126 and SB203580 inhibitors were both present in the culture medium (Fig. 9A). Northern blot analysis showed that the hdc mRNAs expression did not change in the presence of PMA nor in the presence of U0126 or SB203580 inhibitors (Fig. 9B). Taken together these data suggest that the post-transcriptional regulation of HDC could be done at the translational level, depending on 4E-BP1. DISCUSSION In UT7D1 cells, PMA quickly and transiently activates the ERK through a protein kinase C-dependent pathway but does not affect p38, which is activated by the GM-CSF. Activation of one of these two pathways represses 4E-BP1 transcription but at different levels. In contrast the JNK pathway, activated by hyperosmolarity, does not affect 4E-BP1 expression. This work is the first evidence for a role of the ERK and p38 on translational machinery member expression. As determined by our CHX experiments, inhibition of 4E-BP1 expression by the MAPK depends on protein neosynthesis. ERK and p38 have common targets that could play a role in this regulation: the immediate early genes egr-1, c-jun, and c-fos, for example. Egr-1 can be the activator of transcription as well as the inhibitor of transcription and, in UT7D1 cells, we have shown that Egr-1 expression is strictly inversely correlated with 4E-BP1 expression. On the 1020-bp sequence of 4e-bp1 promoter some putative Egr response element, AP4, SRF, SP1, Elk-1, or NFB-binding sites can be identified. Performing reporter assays in HeLa cells with the ZnEgr dominant negative mutant, we have shown that 4e-bp1 is a new gene, the expression of which is inhibited by Egr transcription factors.
As far as we know, the only repressor effect of Egr-1 was described on egr-1 gene itself to realize a negative feedback (57), and how Egr-1 is inhibitor rather than activator of transcription is not fully understood. It was first shown that ZnEgr inhibits Egr-1-dependent transcription by binding to the Egr response element of a promoter, but a recent work has demonstrated that ZnEgr disrupts the formation of a Egr-1/c-Jun complex (58). In our system c-Jun expression is induced after 4 h of PMA treatment of UT7D1. 2 We cannot exclude the possibility that a Egr-1/c-Jun complex is formed on 4e-bp1 promoter to inhibit its activity, but at that time Egr-1/c-Jun has been shown to activate the transcription of the MAO-B gene (59). A detailed study of 4e-bp1 promoter regulation must be done to identify the mechanism by which retinoic acid or Me 2 SO represses 4E-BP1 expression. Nevertheless Egr-1 has a critical role in a variety of processes that include proliferation, apoptosis, and cell differentiation (and thereby oncogenesis) neuronal plasticity and ischemia. The new link that our work makes between Egr-1 and 4E-BP1 allows us to reconsider the role that 4E-BP1 could play in this process by being regulated through Egr-1. The induction of Egr-1 expression by PMA has been implicated in the megakaryocytic differentiation process of K562 cell line (60). According to our data, we can suppose that 4E-BP1 could be responsible for one or more of the characteristic changes that appear, like variations of cell morphology, adhesive properties, endomitosis, and expression of markers associated with megakaryocytes.
This work is the first evidence for a role of the ERK and p38 on translational machinery member expression, but a lot of works have been done demonstrating the role of the MAPK in the regulation of phosphorylation of eukaryotic initiation factors. MAPK regulate the eIF4E phosphorylation state through . Intracellular histamine concentration and HDC activity were measured as described under "Experimental Procedures." The data are the means Ϯ S.D. of two to five independent experiments. The columns labeled with asterisks show significant differences (p Ͻ 0.01) compared with the same treatment without inhibitor (A). 20 g of total RNA was analyzed by Northern blotting using hdc or GAPDH cDNA probes (loading control) as indicated (B). MNK1 and MNK2, which can integrate signals emanating from both types of MAPK pathway in response to mitogens, polypeptide hormones, tumor promoters, and growth factors (61)(62)(63)(64). This phosphorylation was first correlated with an increase rate in protein synthesis, and on the other hand, dephosphorylation coincides with a reduction of protein synthesis at metaphase, upon heat shock, and during adenovirus infection. However, a correlation between eIF4E phosphorylation and the overall translation rate is not observed in every situation (5), and the effects of phosphorylation on eIF4E are not completely understood. Whereas Minich et al. (66) have described an increased affinity of phosphorylated eIF4E for the cap, Scheper et al. (65) have shown that phosphorylation of eIF4E on Ser 209 diminishes its ability to bind capped mRNA. A recent work of Knauf et al. (68) demonstrates that the phosphorylation of eIF4E by MNK1 and MNK2 inhibits the cap-dependent translation. Two recent studies have shown that ERK and p38, through p90/RSK1, can influence the translational machinery at another level. The eukaryotic elongation factor 2 kinase phosphorylates and inactivates eEF2. Insulin induces dephosphorylation of eEF2 and inactivation of eukaryotic elongation factor 2 kinase, and these effects are blocked by rapamycin. In contrast, regulation of eEF2 by stimuli that activate ERK or p38 is insensitive to rapamycin but blocked by inhibitors of MEKs or p38, respectively, consistent with the involvement of p90/RSK (69,70). Taken together, these results put on to the fore the translational level of MAPK regulation of protein expression. In UT7D1 cell line, neither the 4E-BP2 nor the eIF4E expressions vary along with PMA stimulation. 2 Nevertheless, we have measured an increase in protein neosynthesis when cells are cultivated with PMA, and the concentration of protein/cell is two times higher in PMA-stimulated cells than in GM-CSF-cultivated cells. 2 Of course PMA induces a great increase in transcription level, and we cannot assign this effect to a 4E-BP1 decrease of expression. Nevertheless we have demonstrated that concerning the HDC, its PMA-induced increased activity does not depend on transcriptional regulation but seems to be translational. We have now to determine other transcripts, the translation of which depends on 4E-BP1. A recent work from Grolleau et al. (67) shows an interesting application of microarrays and proteomics that could be used to identify such transcripts.