c-Jun Is a JNK-independent Coactivator of the PU.1 Transcription Factor*

The ETS domain transcription factor PU.1 is necessary for the development of monocytes and regulates, in particular, the expression of the monocyte-specific macrophage colony-stimulating factor (M-CSF) receptor, which is critical for monocytic cell survival, proliferation, and differentiation. The bZIP transcription factor c-Jun, which is part of the AP-1 transcription factor complex, is also important for monocytic differentiation, but the monocyte-specific M-CSF receptor promoter has no AP-1 consensus binding sites. We asked the question of whether c-Jun could promote the induction of the M-CSF receptor by collaborating with PU.1. We demonstrate that c-Jun enhances the ability of PU.1 to transactivate the M-CSF receptor promoter as well as a minimal thymidine kinase promoter containing only PU.1 DNA binding sites. c-Jun does not directly bind to the M-CSF receptor promoter but associates via its basic domain with the ETS domain of PU.1. Consistent with our observation that AP-1 binding does not contribute to c-Jun coactivation is the observation that the activation of PU.1 by c-Jun is blocked by overexpression of c-Fos. Phosphorylation of c-Jun by c-Jun NH2-terminal kinase on Ser-63 and -73 does not alter the ability of c-Jun to enhance PU.1 transactivation. Activated Ras enhances the transcriptional activity of PU.1 by up-regulating c-Jun expression without changing the phosphorylation pattern of PU.1. The activation of PU.1 by Ras is blocked by a mutant c-Jun protein lacking the basic domain. The expression of this mutant form of c-Jun also completely blocks 12-O-tetradecanoylphorbol-13-acetate-induced M-CSF receptor promoter activity during monocytic differentiation. We propose therefore that c-Jun acts as a c-Jun NH2-terminal kinase-independent coactivator of PU.1, resulting in M-CSF receptor expression and development of the monocytic lineage.

The ETS domain transcription factor PU.1 is preferentially expressed in myeloid and B cells (1,2) and plays a pivotal role in their development (3,4). Indeed, mice deficient in PU.1 display a complete block in development of monocytes, macrophages, and B cells (5,6). During hematopoietic development, PU.1 mRNA is expressed at low levels in murine embryonic stem cells and human CD34ϩ stem cells and is specifically up-regulated upon myeloid differentiation, and down-regulated upon erythroid differentiation (7,8). PU.1 regulates the expression of almost all characterized myeloid genes, including growth factor receptors, and in particular directs the monocytespecific expression of the macrophage colony-stimulating factor (M-CSF) 1 receptor (9,10). Thus, PU.1-deficient hematopoietic cells display minimal expression of granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor receptors and no detectable M-CSF receptors (11,12).
The M-CSF receptor is critical for monocytic cell survival, proliferation, and differentiation (3,13). M-CSF is known to augment monocyte survival and, therefore, to allow macrophage differentiation (14). The responsiveness of hematopoietic progenitor cells to M-CSF is regulated at the level of M-CSF receptor expression (15). Although the important role of the M-CSF receptor for the development of monocytes has been clearly demonstrated, little is known about the signaling molecules or protein-protein interactions that modulate the effect of PU.1 to regulate the M-CSF receptor promoter activity (3,16). c-Jun belongs to the bZIP group of DNA binding proteins and is a component of AP-1 transcription factor complexes (17). c-Jun forms homodimers or can heterodimerize with other Jun family members or with other bZIP proteins including members of the Fos and ATF/cAMP response element-binding protein (CREB) families (18,19). AP-1 has been shown to be involved in many cellular processes including proliferation, differentiation, apoptosis, and stress responses (18). In particular, there is evidence that c-Jun plays a role in monocytic differentiation. c-Jun mRNA is up-regulated upon monocyte differentiation of bipotential myeloid cell lines (20 -22), while stable transfection of c-Jun into myeloid cell lines results in partial differentiation (23,24).
Although c-Jun and PU.1 are both pivotal for monocytic development, it is still unclear whether c-Jun is involved in the regulation of the M-CSF receptor, which is critical for monocyte survival, proliferation, and differentiation. It has been shown that during 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced monocytic differentiation of U937 cells, c-Jun and M-CSF receptor mRNA expression increases (25). However, the monocyte-specific M-CSF receptor promoter (9,26,27) contains * This work was supported by National Institutes of Health Grants CA41456 and P01 CA72009 (to D. G. T.) and Medical Research Council of Canada Grant MT-9734 (to T. H.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  Reporter Constructs and Expression Plasmids-The human monocyte-specific M-CSF receptor promoter ranging from bp Ϫ88 to ϩ71 with respect to the major monocytic transcription start site (9,26) was subcloned in the firefly luciferase vector pXP2 (28). pTK with PU.1 sites is a dimer of both PU.1 sites from the granulocyte colony-stimulating factor receptor promoter from bp ϩ28 to ϩ54 (29) subcloned into pTK81luc, a pXP2-based luciferase construct with a TATA box only as a minimal promoter (28). pTK with mutated PU.1 sites is a dimer of both mutated PU.1 sites from the granulocyte colony-stimulating factor receptor promoter from bp ϩ28 to ϩ54 (primers: 5Ј-TCG AGT GGT TTC ACA AAC TTT TGT TGA CGA GAG-3Ј and 5Ј-TCG ACT CTC GTC AAC AAA AGT TTG TGA AAC CAC-3Ј) subcloned into pTK81luc and was constructed as described for pTK with PU.1 sites (29). As an internal control plasmid for co-transfection assays, the pRL-null construct driving a Renilla luciferase gene (Promega, Madison, WI) was used (30).
Transient Transfections Using LipofectAMINE Plus or Electroporation and Reporter Assays for Firefly and Renilla Luciferase-CV-1 cells, F9 cells, or 293T cells were transfected using LipofectAMINE Plus (Life Technologies) as described by the manufacturer. U937 cells were transiently transfected by electroporation as described previously (9). Fire-fly luciferase activities from the constructs pM-CSFR, pXP2, and pTK with PU.1 sites and pTK with mutated PU.1 sites and Renilla luciferase activity from the internal control plasmid pRL-null were determined 24 h after the initiation of the transfection protocols using the Dual-Luciferase Reporter Assay System (Promega). Firefly luciferase activities were normalized to the Renilla luciferase values of pRL-null. Results are given as means Ϯ S.E. of at least six independent experiments. The following DNA concentrations of the reporter constructs and expression plasmids were used for LipofectAMINE Plus transfections: 0.3 g of the human monocyte-specific M-CSF receptor promoter in pXP2, pXP2, the TK promoter with PU.1 sites, and the TK promoter with mutated PU.1 sites; 0.05 g of the internal control plasmid pRL-null; 0.5 g of pEBG-PU.1; 0.2 g of the other expression plasmids for PU.1 and PU.1 mutants; 0.25 g of Ras(L61), Ras(N17), and MEKK1; 0.1 g of c-Jun, c-Jun mutants, and c-Fos; and the same concentrations of the empty expression vectors as controls, respectively. For electroporation, 10 g of the firefly luciferase reporter constructs, 5 g of expression plasmids, and 1 g of the internal control plasmid were used. pRL-null was chosen as internal control plasmid, because it was not transactivated by Ras (30) or by PU.1, c-Jun, c-Fos, or MEKK1 in CV-1, F9, 293T, or U937 cells (data not shown).
Electrophoretic Mobility Shift Assay-Electrophoretic mobility shift assays were performed as described previously (9,26,27,29). As a positive control for c-Jun binding, a double-stranded AP-1 probe from the collagenase promoter (5Ј-AAT TCG CTT GAT GAC TCA GCC GGA A-3Ј) was labeled with Klenow polymerase and [␣-32 P]dCTP (NEN Life Science Products) and incubated with 0.1 g/l of double-stranded poly(dI-dC) (Sigma) with 1 l of in vitro translated c-Jun or c-Fos. In some experiments, a 100-fold molar excess of the AP-1 probe was added as specific unlabeled competitor. Similarly, a double-stranded M-CSF receptor promoter oligonucleotide extending from position bp Ϫ88 to ϩ71 with respect to the major transcription start site (9,26) was Klenow-labeled with [␣-32 P]dCTP and incubated with 0.1 g/l doublestranded poly(dI-dC) with 1 l of in vitro translated PU.1, c-Jun, or c-Fos. In some experiments, a 100-fold molar excess of specific unlabeled competitor was added: an oligonucleotide with the PU.1 DNA binding site in the CD11b promoter (41), 3Ј-AGC CTA CTT CTC CTT TTC TGC CCT TCT TTG-5Ј, to compete for PU.1, and the AP-1 probe described above to compete for c-Jun.
Protein Interaction Assay-Protein interaction assays were performed as described previously (27,42). c-Jun and c-Fos were in vitro transcribed and translated using the TNT Reticulocyte Lysate System (Promega) and labeled with [ 35 S]methionine (NEN Life Science Products). 1 l of labeled in vitro translated c-Jun or c-Fos was mixed with 1 g of bacterially expressed GST-PU.1 or equivalent amounts of GST or glutathione-agarose beads (Sigma) for 1 h at 4°C in NETN buffer (20 mM Tris (pH 8.0), 200 mM NaCl, 1 mM EDTA, and 0.5% Nonidet P-40). GST-PU.1 was recovered using glutathione-agarose beads, washed seven times with NETN buffer, and separated by 10% SDS-polyacrylamide gel electrophoresis. Prior to autoradiography, the gel was stained with Coomassie Brilliant Blue (Bio-Rad) to verify that the protein concentrations of GST-PU.1 and GST were the same in all lanes.
In Vivo Labeling, Phosphoamino Acid Analysis, and Phosphopeptide Mapping-To detect changes in the phosphorylation pattern of PU.1 upon stimulation with activated Ras in vivo, 0.5 g of pEBG-PU.1 either with 0.25 g of activated Ras(L61) or with inactive Ras(N17) was transfected into 293T cells using LipofectAMINE Plus (Life Technologies). 3 h after transfection, cells were starved in serum-free Dulbecco's modified Eagle's medium. After 18 h, cells were placed into serumfree and phosphate-free Dulbecco's modified Eagle's medium (Life Technologies) for 30 min before they were metabolically labeled with [ 32 P]orthophosphate (2.5 mCi/ml). After 4 h, cells were lysed with radioimmunoprecipitation assay buffer containing 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 0.15 M NaCl, 5 mM EDTA, and 50 mM Tris (pH 8.0) and supplemented with aprotinin, phenylmethylsulfonyl fluoride, pepstatin A, leupeptin, antipain, and chymostatin as protease inhibitors (Sigma) and sodium pyrophosphate, sodium fluoride, and sodium vanadate as phosphatase inhibitors (Sigma). In parallel plates, 0.3 g of the M-CSF receptor promoter was co-transfected in 293T cells, and luciferase activities were determined to ensure that Ras enhances the transactivation function of PU.1 in the particular experiment used in in vivo labeling and subsequent phosphoamino acid analysis and phosphopeptide mapping.
Western Blot-24 h after the start of transfection, cells were lysed with radioimmunoprecipitation assay buffer. Equal amounts of total protein were separated on 10% SDS-polyacrylamide gels and transferred to Immobilon-P membrane (Millipore). Membranes were incubated with anti-c-Jun antibody (catalog no. SC-45; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), anti-PU.1 antibody (catalog no. SC-352; Santa Cruz Biotechnology), or anti-␤-tubulin antibody as an internal control (catalog no. 1111876; Boehringer Mannheim) for 60 min and then with Protein A-horseradish peroxidase conjugate (Amersham, Buckinghamshire, United Kingdom) for 45 min. For U937 cells, an anti-M-CSF receptor antibody (catalog no. SC-692, Santa Cruz Biotechnology) was used. Signals were detected with ECL Western blotting detection reagents (Amersham). In parallel plates, the M-CSF receptor promoter construct was co-transfected, and luciferase activities were determined to ensure that Ras enhances the transactivation function of PU.1 in the particular experiment used for Western blot analysis of c-Jun expression or PU.1 expression and that the transfection efficacy was the same (less than 10% difference between plates) in the particular experiment. Differences in protein expression were quantitated by ImageQuant software (Molecular Dynamics). We next asked the following questions: (a) whether the binding of PU.1 to DNA was necessary for its activation by c-Jun and (b) whether a PU.1 binding site alone was sufficient for the c-Jun-enhanced PU.1 activation. We observed enhanced PU.1 transactivation mediated by c-Jun using a reporter construct containing four PU.1 binding sites cloned upstream of a minimal TK promoter (Fig. 1C). In control experiments, no effect of c-Jun on PU.1 activity was observed when the PU.1 binding sites were mutated (Fig. 1C). These data indicate that PU.1 binding to DNA is necessary for its activation by c-Jun and that PU.1 binding sites are sufficient to mediate this effect. Since there are no AP-1 consensus sites in the human monocyte-specific M-CSF receptor promoter from bp Ϫ88 to ϩ71 with respect to the major monocytic transcription start site (9, 26) (Fig. 1, A and B) or in the TK promoter containing PU.1 sites (Fig. 1 C), our data suggested that c-Jun augmentation of PU.1 transactivation was not mediated by DNA binding of c-Jun. In order to formally exclude DNA binding by c-Jun, we performed an electrophoretic mobility shift assay using a bp Ϫ88 to ϩ71 M-CSF receptor promoter fragment or a bp Ϫ62 to Ϫ29 oligonucleotide containing the PU.1 binding site of the M-CSF receptor promoter (9). In vitro translated c-Jun or a mixture of c-Jun and c-Fos specifically bound to a doublestranded AP-1 oligonucleotide probe from the collagenase promoter ( Fig. 2A), while no specific binding was observed using the double-stranded bp Ϫ88 to ϩ71 M-CSF receptor promoter ( Fig. 2A) or bp Ϫ62 to Ϫ29 oligonucleotide containing the PU.1 binding site of the M-CSF receptor promoter (data not shown).

c-Jun Enhances the Ability of PU.1 to Transactivate the M-CSF Receptor Promoter and a Minimal TK Promoter Contain
In control experiments, in vitro translated PU.1 was shown to bind strongly and specifically to the same M-CSF receptor promoter fragments (Fig. 2A). These results indicate that binding of c-Jun to the M-CSF receptor promoter DNA is not required to mediate its activating effect on PU.1.
It has previously been reported that c-Jun can physically and functionally interact with ETS-1, which like PU.1 is an ETS family transcription factor (44). In these same studies, it was reported that c-Jun and PU.1 could physically interact, although the functional consequences of this interaction were not examined. We were able to confirm that c-Jun did indeed specifically interact with PU.1 in vitro (Fig. 2B). Because c-Jun can bind to PU.1 but does not bind to DNA itself ( Fig. 2A), we therefore conclude that c-Jun acts as a coactivator (45) of PU.1.
Since c-Jun can form a heterodimer with c-Fos in AP-1 transcription factor complexes (17)(18)(19), we asked the question of whether c-Fos expression could modulate the synergy between c-Jun and PU.1 in F9 cells. Co-transfection of c-Fos did not enhance the synergy between c-Jun and PU.1 but instead completely blocked it (Fig. 2C). c-Fos did not bind to PU.1 (Fig. 2B) and therefore might compete with PU.1 for the binding partner c-Jun. These results are consistent with a model in which c-Jun mediates its effects through direct interactions with PU.1 and not by independent DNA binding to an AP-1 site.

Phosphorylation of c-Jun by c-Jun NH 2 -terminal Kinase (JNK) on Ser-63 and -73 Does Not Alter the Ability of c-Jun to
Enhance PU.1 Transactivation-PU.1 (Fig. 3A) and c-Jun (Fig.  3B) are composed of a number of discrete domains. We used deletion mutants to determine which domains of PU.1 and c-Jun are critical for the activation of PU.1 by c-Jun in the context of the M-CSF receptor promoter. The transactivation domain of PU.1 (amino acids 1-118) was necessary for the activation by c-Jun (Fig. 3A). The transactivation domain of c-Jun (amino acids 1-87); the basic domain of c-Jun, which can physically interact with PU.1 (44) and mediates DNA binding to AP-1 sites; and the leucine zipper domain, which is responsible for homodimerization and heterodimerization with c-Fos, were all necessary for the activation of PU.1 by c-Jun (Fig. 3B). The transcriptional activity of c-Jun is increased following phosphorylation on Ser-63 and -73 by JNK (34,46). To determine whether phosphorylation of Ser-63 and -73 was important for the function of c-Jun as a transcriptional coactivator of PU.1, we used c-Jun constructs in which these sites were mutated to Ala residues. Surprisingly, we found no difference in the ability of wild type c-Jun and S63A/S73A c-Jun to enhance PU.1 transactivation of the M-CSF receptor promoter in F9 cells (Fig. 3B). Furthermore, co-transfection with an activated allele of Ras, which has been demonstrated to enhance the transcriptional activity of c-Jun via phosphorylation by JNK (34,46), did not enhance the coactivator function of wild type or S63A/S73A c-Jun (Fig. 3C). These data indicate that the coactivator function of c-Jun to enhance the transcriptional activity of PU.1 is independent of JNK phosphorylation.

Activated Ras Enhances the Transcriptional Activity of PU.1 by Up-regulating c-Jun Expression-Although
Ras signal transduction has been demonstrated to play an important role in myeloid differentiation (47-50), it has not been shown whether Ras increases the activity of the monocyte-specific M-CSF receptor promoter. However, Ras is known to induce the expression of c-Jun (51). Therefore, we asked the question of whether Ras could augment the transcriptional activity of PU.1 and if this was mediated by enhancing c-Jun expression. In fact, activated Ras(L61) enhanced the ability of PU.1 to transactivate the M-CSF receptor promoter in CV-1 cells to a similar degree as c-Jun (Fig. 4A). Furthermore, a dominant negative c-Jun mutant lacking the basic domain, which is required for physical interaction with PU.1 in vitro (44), blocked the activation of PU.1 by Ras (Fig. 4A). In the same experiment, Ras did not change the protein expression of transfected PU.1 (Fig. 4B) but up-regulated endogenous c-Jun protein expression 4-fold (Fig. 4B). Since the transfection efficiency in the CV-1 cells was 40% (Fig. 4B), we estimate that Ras up-regulated c-Jun expression 10-fold in transfected cells.
The c-Jun promoter contains important TPA response elements that preferentially bind heterodimers of c-Jun and ATF-2, both of which are activated upon phosphorylation by JNK (17-19). We therefore tested the effect of another JNK activator, the MAP kinase kinase kinase MEKK1 (40), on the ability of PU.1 to transactivate the M-CSF receptor promoter. In fact, MEKK1 enhanced the PU.1 transactivation function to a similar level as Ras (Fig. 4A).
Ras is known to modulate the activity of the ETS domain transcription factors ETS-1 and ETS-2 by phosphorylation (52). However, co-expression of Ras did not alter the phosphorylation pattern of PU.1 in vivo (Fig. 4C), and furthermore, Ras enhanced the transcriptional activity of known phosphorylation site mutants of PU.1 (Ser-41, -45, and -148) (53, 54) similar to wild type PU.1 (data not shown). In conclusion, these data suggest a model in which Ras enhances the transcriptional activity of PU.1 by increasing the expression of its coactivator c-Jun.

The Expression of the Mutant Form of c-Jun That Lacks the Basic Domain Completely Blocks TPA-induced M-CSF Receptor
Promoter Activity during Monocytic Differentiation-We next asked the question of whether the coactivator function of c-Jun could play a biological role during monocytic differentiation in vivo. We first showed that Ras enhanced the ability of PU.1 to transactivate the M-CSF receptor promoter in myeloid U937 cells (Fig. 5A). U937 cells can be differentiated to monocytic cells upon treatment with TPA, and during this process c-Jun and M-CSF receptor mRNA expression increases (25). TPA increased reporter gene expression from our M-CSF receptor promoter construct (Fig. 5A) and also increased the expression of endogenous M-CSF receptor (Fig. 5B). This effect on the human monocytic M-CSF receptor promoter was blocked by co-expression with a dominant negative c-Jun mutant lacking the basic domain (Fig. 5A). These results indicate that c-Jun function is required for the increase in M-CSF receptor promoter activity observed in myeloid cells differentiated toward the monocytic lineage with TPA.

DISCUSSION
The transcription factor c-Jun (20 -24) and the regulation of the M-CSF receptor by the transcription factor PU.1 (11,12,55) both play important roles in the development of the monocytic lineage. Therefore, we asked the question of whether c-Jun could promote the induction of the M-CSF receptor by PU.1. Here we demonstrate that c-Jun enhances the ability of PU.1 to transactivate the human monocyte-specific M-CSF receptor promoter (Fig. 1, A and B).
These data explain how c-Jun, PU.1, and the M-CSF receptor might collaborate as major factors in the process of monocytic differentiation. Moreover, c-Jun and PU.1 are already known to cooperate in the induction of other monocytic genes containing PU.1 and AP-1 DNA binding sites, such as the macrosialin gene (56) and macrophage scavenger receptor gene (57). The c-Jun enhancement of the ability of PU.1 to transactivate the M-CSF receptor promoter is novel, because the human monocyte-specific M-CSF receptor promoter contains no AP-1 consensus binding sites. In contrast, the macrosialin promoter and macrophage scavenger receptor promoter contain PU.1 and AP-1 binding sites, which are critical for their monocyte-specific expression (56,57). In fact, our data suggest a novel mechanism by which c-Jun can induce gene expression. c-Jun does not bind to the M-CSF receptor promoter ( Fig. 2A) and, moreover, enhances the ability of PU.1 to transactivate a minimal promoter driven by PU.1 sites alone (Fig. 1C). Furthermore, c-Jun physically binds to PU.1 (Fig. 2B). Since c-Jun binds to PU.1 and functionally activates PU.1 without binding to the M-CSF receptor promoter DNA, we conclude that c-Jun acts as a coactivator (45) of PU.1. This is the first report demonstrating a coactivator function for c-Jun.
Usually, c-Jun forms heterodimers with c-Fos in AP-1 transcription factor complexes (17)(18)(19). However, c-Fos does not cooperate with c-Jun in its coactivator function. In contrast, c-Fos completely blocks the coactivation of PU.1 by c-Jun (Fig.  2C). Since c-Fos does not physically bind to PU.1 (Fig. 2B) and not by independent DNA binding to an AP-1 site (Fig. 5C).
Usually, the transcriptional activity of c-Jun is increased following phosphorylation on Ser-63 and -73 by JNK (34,46). Furthermore, the general coactivator CBP/p300 stimulates c-Jun-dependent transcription, and the c-Jun residues Ser-63 and -73 are required for CBP/p300 stimulation in vivo and CBP/p300 binding in vitro (58). Surprisingly, we found no difference in the ability of wild type c-Jun and the S63A/S73A c-Jun mutant to enhance PU.1 transactivation of the M-CSF receptor promoter (Fig. 3B). Furthermore, co-transfection with an activated allele of Ras, which has been demonstrated to enhance the transcriptional activity of c-Jun via phosphorylation by JNK (34,46), did not enhance the coactivator function of wild type or S63A/S73A c-Jun (Fig. 3C). Our data indicate that the coactivator function of c-Jun to enhance the transcriptional activity of PU.1 is independent of JNK phosphorylation.
In accordance with its pivotal role in B cell development (5,6), PU.1 binds to the B cell-specific immunoglobulin kappa (Ig) 3Ј enhancer and can control transcriptional activity (59). The immunoglobulin 3Ј enhancer is activated by PU.1, c-Jun, PIP, and c-Fos and contains respective DNA binding sites for these factors (60). In this context, mutants of PU.1 that lack the transcriptional activation domain are as efficient at stimulating enhancer activity as the wild-type PU.1 protein (60). In contrast, the transactivation domain of PU.1 (amino acids 1-118) is necessary for the basal transactivation of the M-CSF receptor promoter by PU.1 (Fig. 3A), and c-Jun cannot exert its coactivator function on PU.1 mutants that lack the transactivation domain (Fig. 3A). These data suggest that the function of different PU.1 domains might vary depending on the cell type (monocyte versus B cell) and on the cooperating proteins (c-Jun as part of the AP-1 transcription factor complex versus c-Jun as a JNK independent coactivator).
How is the coactivator function of c-Jun regulated? Since the coactivator function of c-Jun cannot be regulated via phosphorylation by JNK, the regulation of c-Jun expression might be crucial for the capacity of c-Jun to coactivate PU.1 and induce monocytic differentiation. The Ras signal transduction pathway, for example, induces the expression of c-Jun (51). The Ras family of proteins are GTP-dependent molecular switches that are essential for cell growth and differentiation (61,62). Ras exerts its effect on cell growth mainly via ETS (63) and AP-1 (64) transcription factors. For example, cells with a null mutation in the c-jun gene lack many characteristics of Ras transformation (64), and dominant negative mutants of ETS-1, ETS-2, or PU.1 with just the DNA binding domain inhibit Ras activation of transcription and revert Ras-transformed cells (63). In particular, Ras has been demonstrated to play an important role in myeloid differentiation. Macrophage differentiation and M-CSF-dependent survival are altered in transgenic mice that express dominant suppressors of Ras signaling (47), while a number of hematopoietic cell lines undergo spontaneous monocytic differentiation in response to expression of activated Ras (48,49). In addition, M-CSF, granulocyte-macrophage colony-stimulating factor, or interleukin-3-induced monocytopoiesis of CD34ϩ cells is inhibited by N-Ras antisense oligonucleotides (50).
Although Ras signaling plays an important role in monocytic differentiation, Ras has not previously been shown to increase activity of the monocyte-specific M-CSF receptor promoter. Our data suggest that Ras activates the M-CSF receptor promoter via c-Jun and PU.1, because in c-Jun-deficient F9 cells there is no effect of Ras on PU.1 (Fig. 3C), whereas Ras enhances the ability of PU.1 to transactivate the M-CSF receptor promoter in c-Jun-containing CV-1 cells (Fig. 4A). Moreover, a c-Jun mutant lacking the basic domain blocks the Ras enhancement of PU.1 transactivation of the M-CSF receptor promoter (Fig. 4A). Ras signaling has been reported to induce the expression of c-Jun (51), and here we demonstrate that Ras induces c-Jun expression in CV-1 cells during activation of PU.1 (Fig. 4B). Thus, these data suggest that Ras enhances the transcriptional activity of PU.1 by up-regulating the expression of its coactivator c-Jun. In accordance with this model, another inducer of c-Jun expression, MEKK1 (40), enhanced the PU.1 transactivation function to a similar level as Ras or c-Jun (Fig. 4A).
Ras modulates the activity of ETS domain transcription factors such as ETS-1 or ETS-2 by phosphorylation (52). More-over, the ETS domain factor PU.1, in particular, can be phosphorylated by casein kinase (53,65) or JNK in vitro (66), and the activity of PU.1 is known to be regulated by phosphorylation (53,54,65). Phosphorylation of PU.1 at Ser-148 is necessary for interaction with NF-EM5 (53) or stimulation by lipopolysaccharide (65), and a PU.1 mutant at Ser-41 and -45 decreases the M-CSF-or granulocyte-macrophage colony-stimulating factor-dependent proliferation of bone marrow macrophages (54). However, co-expression of activated Ras does not alter the phosphoamino acid (data not shown) or phosphopeptide pattern (Fig. 4C) of PU.1 in vivo. Furthermore, Ras enhances the transcriptional activity of known phosphorylation site mutants of PU.1 (Ser-41, -45, and -148) (53, 54) similar to wild type PU.1 (data not shown). Our data are in accordance with a report that the ability of PU.1 to rescue macrophage development in PU.1Ϫ/Ϫ ES cells is not impaired by these same phosphorylation mutants (67). In conclusion, these results suggest a model in which Ras enhances the transcriptional activity of PU.1 by increasing the expression of its coactivator c-Jun without modifying the activity of PU.1 by phosphorylation.
Finally, we determined whether Ras and c-Jun could regulate the M-CSF receptor promoter during differentiation of myeloid cells. In fact, Ras enhanced the ability of PU.1 to transactivate the M-CSF receptor promoter in myeloid U937 cells (Fig. 5A). It has been shown that U937 cells can be differentiated with TPA to monocytic cells, and during this process c-Jun and M-CSF receptor mRNA expression increases (25). TPA increased reporter gene expression from our M-CSF receptor promoter construct (Fig. 5A) and also increased the expression of endogenous M-CSF receptor (Fig. 5B). This effect on the M-CSF receptor promoter is blocked by co-expression with a dominant negative c-Jun mutant lacking the basic domain, which interacts with PU.1 (Fig. 5A). These results indicate that c-Jun function is required for the increase in M-CSF receptor promoter activity observed in myeloid cells differentiated toward the monocytic lineage with TPA.
In summary, our data indicate that c-Jun is a JNK-independent coactivator of PU.1 (Fig. 5C). These results suggest a model in which growth factors or other signals (3,68) activate the Ras pathway, which in turn increases c-Jun expression in monocytic progenitors. Increased c-Jun expression activates PU.1, resulting in increased M-CSF receptor up-regulation and survival, proliferation, and differentiation of the monocytic lineage.