Role of Erk1/2 Signaling in the Regulation of Neutrophil Versus Monocyte Development in Response to G-CSF and M-CSF*

Background: G-CSF and M-CSF are cytokines that support the development of neutrophils and monocytes, respectively. Results: The duration of Erk1/2 activation by G-CSF and M-CSF affects the lineage commitment of myeloid precursors. Conclusion: G-CSF and M-CSF instruct neutrophil versus monocyte development through differential activation of Erk1/2. Significance: This reveals a key mechanism by which G-CSF and M-CSF control lineage specification. Lineage specification in the hematopoietic system depends on the expression of lineage specific transcription factors. However, the role of hematopoietic cytokines in this process has been controversial and little is known about the intracellular signaling mechanisms by which cytokines instruct lineage choice. G-CSF and M-CSF are two lineage-specific cytokines that play a dominant role in granulopoiesis and monopoiesis, respectively. We show here that a G-CSFR mutant in which tyrosine 729 had been mutated to phenylalanine (Y729F) promoted monocyte rather than neutrophil development in myeloid precursors, which was associated with prolonged activation of Erk1/2 and augmented activation of downstream targets c-Fos and Egr1. Inhibition of Erk1/2 activation or knockdown of c-Fos or Egr1 largely rescued neutrophil development in cells expressing G-CSFR Y729F. We also show that M-CSF, but not G-CSF, stimulated strong and sustained activation of Erk1/2 in mouse lineage marker negative (Lin−) bone marrow cells. Significantly, inhibition of Erk1/2 signaling in these cells favored neutrophil over monocyte development in response to M-CSF. Thus, prolonged Erk1/2 activation resulted in monocyte development following G-CSF induction whereas inhibition of Erk1/2 signaling promoted neutrophil development at the expense of monocyte formation in response to M-CSF. These results reveal an important mechanism by which G-CSF and M-CSF instruct neutrophil versus monocyte lineage choice, i.e. differential activation of Erk1/2 pathway.

Neutrophils and monocytes/macrophages, key components of host innate immune defense against infections, are derived from hematopoietic stem cells (HSCs) 2 through a process called myelopoiesis, in which HSCs, through progressive commitment, give rise to common myeloid progenitors (CMPs) that in turn develop into granulocyte-monocyte progenitors (GMPs) (1,2). GMPs are bipotent and can terminally differentiate into either granulocytes or monocytes in response to cell intrinsic and external signals. Myelopoiesis is a dynamic process that is tightly controlled by two interacting mechanisms, i.e. a transcription factor network that regulates the expression of lineage-specific genes and a group of hematopoietic cytokines that stimulate intracellular signaling by binding to cell surface cytokine receptors (3)(4)(5). These two mechanisms act in collaboration to regulate the commitment, differentiation, proliferation, and survival of HSCs and myeloid precursors. Disruption of the regulatory mechanisms is often associated with myeloid leukemia.
The lineage specification of HSCs and precursors depends on the expression and activities of lineage specific transcription factors. Monocyte and neutrophil lineage specifications require the transcription factors C/EBP␣ and PU.1 that are components of a myeloid transcriptional regulatory circuit, which includes Egr1, Egr2, Nab2, and Gfi1, among others (5,6). A high C/EBP␣/PU.1 ratio supports neutrophil development whereas increased expression of PU.1 favors monocyte over granulocyte lineage decision (7). C/EBP␣ instructs neutrophil cell fate in part through activating Gfi1 that promotes neutrophil development and suppresses the alternative monocyte development (8 -10). PU.1 acts in a graded manner to direct distinct cell fates with a high expression promoting monocyte development and a low expression required for B lymphocyte development (11). PU.1 activates IRF8, Klf4, Egr2, and Nab2 that direct monocyte development at the expense of neutrophil cell fate (12)(13)(14). In addition, transcription factors c-Fos and c-Jun have been shown to positively regulate monocyte development (5,15,16).
G-CSF and M-CSF are two lineage-specific hematopoietic cytokines that play a dominant role in granulopoiesis and monopoiesis, respectively. Hematopoietic cytokines have been shown to stimulate cell proliferation and survival; however, their role in lineage specification remains controversial (17)(18)(19). According to the stochastic model, cell fate choice is stochastic and cytokines simply provide nonspecific permissive signals for the survival and proliferation of already committed cells. The instructive model, on the other hand, proposes that cytokines actively instruct uncommitted cells to differentiate into distinct types of mature blood cells. While both models are backed by experimental data, two recent reports lend strong support to the instructive model, at least for G-CSF and M-CSF. Using the bio-imaging approaches that permit continuous long-term observation at the single-cell level, it was shown that G-CSF and M-CSF can instruct myeloid lineage choice in HSCs and GMPs (20,21). However, the intracellular signaling mechanisms by which G-CSF and M-CSF instruct granulocyte versus monocyte lineage commitment are unknown.
In this report, we show that substitution of Tyr-729 of G-CSFR with phenylalanine (F) resulted in monocyte development in response to G-CSF, which was associated with prolonged activation of Erk1/2 and augmented activation of c-Fos and Egr1. Treatment of cells with Mek1/2 inhibitors or knockdown of c-Fos or Egr1 essentially rescued neutrophil development. Notably, the Mek1/2 inhibitors also promoted neutrophil development at the expense of monocyte formation induced by M-CSF. Our data reveal an important signaling mechanism by which G-CSF and M-CSF direct neutrophil versus monocyte lineage specification.
Flow Cytometry-Cells were washed in PBS with 2% horse serum and blocked with Fc block (eBioscience) for 15 min. Cells were then incubated with isotype control anti-mouse IgG antibody conjugated with phycoerythrin (PE), anti-mouse IgG antibody conjugated with fluorescein isothiocyanate (FITC) or PEconjugated anti-F4/80 antibody for 30 min prior to washing in PBS with 2% horse serum. All antibodies were purchased from eBioscience. Samples were analyzed by flow cytometry using a FACSCalibur and the CellQuest software system (BD Biosciences).
Western Blot Analysis-Cells were lysed in SDS lysis buffer and proteins were separated by SDS-PAGE prior to transfer onto polyvinylidenedifluoride (PVDF) membranes. The membranes were incubated with the appropriate antibodies and signals were detected by enhanced chemiluminescence. The antibodies against phospho-Stat5, phospho-Stat3, phospho-Erk1/2, c-Fos, phospho-c-Fos (Ser32), Egr-1, and ␤-actin were purchased from Cell Signaling.
Luciferase Reporter Assay-Cells were transfected with the reporter constructs TRE 3 -tk-Luc (a gift from Dr. Lirim Shemshedini, The University of Toledo) or pEBS2 4 -Luc (a gift from Dr. Gerald Thiel, University of the Saarland Medical Center).
Sixteen hours after transfection, cells were washed and stimulated with G-CSF (10 ng/ml) for 8 h. Luciferase activities were measured using the Molecular Devices Lmaxluminometer (Sunnyvale, CA).
RNA Interference-Lentiviral constructs containing murine c-Fos and Egr-1 shRNAs were purchased from Thermo Scientific. Lentiviral vector encoding a murine ERK2 shRNA (TRCN54729) was purchased from Dharmacon. To target murine ERK1, oligonucleotides were designed to generate a mature antisense AATGTAAACATCTCTCATGGC and cloned into pLKO.1-Hygro (Addgene plasmid 24150). 293T cells were transfected with the lentiviral constructs along with packaging plasmids psPAX2 and pMD2G using the calcium phosphate coprecipitation procedure. Supernatants containing viral particles were harvested at 48 and 72 h post-transfection, concentrated, and used to infect cells in the presence of 8 g/ml polybrene (Santa Cruz Biotechnology). Cells were selected in 2 g/ml puromycin for 48 h or hygromycin (1 mg/ml) for 4 days prior to evaluation of gene knockdown by Western blot analysis.
Real-time Reverse Transcription Polymerase Chain Reaction (qRT-PCR)-Total RNA was extracted using TRIzol reagent (Invitrogen), and cDNA was synthesized using the GoScript TM Reverse Transcription System and Oligo(dT)15 primer (Promega, Madison, WI). qRT-PCR was performed using the SsoFast TM EvaGreen Supermix kit (Bio-Rad), and the relative levels of mRNAs for the different myeloid differentiation markers were normalized to GAPDH mRNA expression.
Apoptosis Assay-Apoptosis was examined using the Annexin V-PE apoptosis detection kit (BD Biosciences). Briefly, 0.3 ϫ 10 6 cells were collected and incubated with Annexin V-PE and 7 amino-actinomycin (7-AAD). Cells were analyzed by two-color flow cytometry.
Statistics-GraphPad Prism software (GraphPad Software, La Jolla, CA) was used for all statistical analysis. Data are shown as mean Ϯ S.D. in all figures. A p value of Ͻ0.05 was considered significant for all analyses and shown as *. ** denotes p Ͻ 0.01, *** denotes p Ͻ 0.001, and **** denotes p Ͻ 0.0001.

Results
Tyrosine 729 of G-CSFR Is Essential for Instructing Neutrophil Lineage Choice-The human G-CSFR contains four tyrosine (Y) residues in the cytoplasmic domain, i.e. Y704, Y729, Y744, and Y764 (Fig. 1A), that have been implicated in G-CSF-stimulated proliferation, survival and differentiation (23,(25)(26)(27)(28). Notably, Y729, Y744, and Y764 are located in the C-terminal region of G-CSFR required for differentiation signaling (22,29). To address the roles of the C-terminal tyrosine residues in G-CSF response, we evaluated the G-CSF response of 32D cells transfected with the G-CSFR mA mutant (32D/mA) in which the three C-terminal tyrosine (Y) residues were mutated to phenylalanine (F) (23,30). As reported previously (22,31), 32D cells expressing the wild    type (WT) G-CSFR (32D/GR) underwent terminal neutrophilic differentiation after culture in G-CSF for 6 to 9 days. Interestingly, upon culture in G-CSF, 32D/mA cells showed increased cell sizes, adherence to the culture dishes and morphological features characteristic of monocytes (Fig. 1, B and C). Consistent with this, the surface expression of macrophage marker F4/80 and the mRNA levels of M-CSF and Mmp-12 were significantly higher in 32D/mA cells than in 32D/GR cells (Fig. 1, D and E). Thus, mutations of the C-terminal tyrosine residues of G-CSFR resulted in monocyte rather than neutrophil development in response to G-CSF, suggesting that the C-terminal tyrosine residues function to promote neutrophil cell fate and suppress the alternative monocyte development.
To identify the involved tyrosine residue, we evaluated the G-CSF responses of 32D cells transfected with the G-CSFR mutants containing individual Y-to-F substitutions in the C terminus (Fig. 1A). Y744F and Y764F mutations had no significant effect on neutrophil development; however, 32D cells express-ing G-CSFR Y729F (32D/Y729F) displayed features associated with monocyte development (Fig. 2). To confirm the observations in a different cell line, we expressed the WT G-CSFR and Y729F mutant in murine multipotent FDCP-mix A4 cells. Parental FDCP-mix A4 cells expressed no detectable levels of endogenous G-CSFR and M-CSFR on cell surface, and the cells died in G-CSF or M-CSF within 24 h (data not shown). FDCPmix A4 cells transfected with the WT G-CSFR (FDCP/GR) developed into mature neutrophils after culture in G-CSF for 6 days (Fig. 3). In contrast, FDCP/Y729F rapidly developed into monocytes within 2-3 days following G-CSF induction and died subsequently. Thus, G-CSFR Y729 is required for neutrophil development in response to G-CSF.
Prolonged Activation of Erk1/2 Is Required for Monocyte Development Mediated by G-CSFR Y729F-We previously showed that Y729 of G-CSFR controls the duration of G-CSFR signaling (23). The kinetics of G-CSF-stimulated activation of different downstream pathways was examined in more detail in 32D and FDCP-mix A4 cells. As shown in Fig. 4, A and B, G-CSFR Y729F mediated prolonged activation of Stat5, Akt, and Erk1/2 in both cell lines. Activation of these pathways was also enhanced in FDCP-mix A4 cells. It has been shown that M-CSF stimulates more potent activation of Erk1/2 as compared with G-CSF and that sustained activation of Erk1/2 is required for M-CSF-induced monocytic differentiation (32,33). We therefore examined the activation status of Erk1/2 in cells continuously cultured in G-CSF for up to 48 h. Erk1/2 phosphorylation was barely detectable at 24 h in cells expressing the WT G-CSFR, but readily detected in cells expressing G-CSFR Y729F, and even at 48 h in 32D/Y729F cells (Fig. 4, C  and D). Thus, prolonged activation of Erk1/2 led to their persistent activation when cells were cultured continuously in G-CSF.
We addressed whether inhibition of Erk1/2 activation rescued neutrophil development. As shown in Fig. 5, treatment of 32D/Y729F and FDCP/Y729F cells with the Mek1/2 inhibitor U0126 or PD0325901 resulted in the typical neutrophil development with reduced cell sizes, loss of adherent phenotype and diminished expression of M-CSF and Mmp-12 although the expression of F4/80 was not significantly altered. Comparable results were obtained with another Mek1/2 inhibitor PD98059 (data not shown). To further address the role of Erk1/2 pathway in monocyte development mediated by G-CSFR Y729F, we knocked down their expression using shRNAs specifically targeting murine Erk1 and Erk2 (32). As shown in Fig. 6, knockdown of Erk1 and Erk2 largely rescued neutrophil development in response to G-CSF in both 32D/Y729F and FDCP/Y729 cells. Thus, prolonged activation of Erk1/2 pathway is essential for monocyte development directed by G-CSFR Y729F. (20,21). However, how M-CSF instructs monocyte cell fate remains unresolved. We examined Erk1/2 activation by G-CSF and M-CSF in mouse Lin Ϫ BM cells. Treatment with M-CSF, but not G-CSF, resulted in strong and sustained activation of Erk1/2 for at least six hours (Fig. 7A). In contrast, G-CSF, but not M-CSF, stimulated Stat3 phosphorylation. When cultured in M-CSF for 3 days, more than 70% of Lin Ϫ BM cells developed into monocytes/macrophages (Fig. 7, B and C). Interestingly, addition of U0126 or PD0325901 to the cultures increased the numbers of mature neutrophils, but markedly suppressed macrophage development. The decrease in macrophage population was unlikely due to increased apoptosis as U0126 and PD0325901 had only a modest effect on cell survival (data not shown). We further performed colony formation assays to assess the effect of the Mek1/2 inhibitors on the development of myeloid precursors. As shown in Fig. 7D, U0126 and PD0325901 caused an approximate 2-fold increase in the number of CFU-G, but significantly decreased the number of CFU-M. In support of increased neutrophil development at the expense of monocyte formation, the two Mek1/2 inhibitors downregulated the expression of M-CSF and Mmp-12, but upregulated the mRNA levels of the neutrophil differentiation markers, including the primary granule proteins myeloperoxidase (MPO) and neutrophil elastase (NE), secondary granule protein lactoferrin (LF) and tertiary granule protein gelatinase B in BM cells cultured in M-CSF (Fig. 7E). Thus, upon suppres- G-CSFR Y729F Mediates Enhanced Activation of c-Fos and Egr1-c-Fos and Egr1 are immediate early genes (IEGs) that are activated by the Erk1/2 pathway (34,35). When persistently activated, Erk1/2 also directly phosphorylates c-Fos that is rapidly induced following Erk1/2 activation and thus c-Fos phosphorylation serves as a sensor for Erk1/2 signal duration (36,37). Erk1/2-mediated phosphorylation stabilizes c-Fos and primes additional phosphorylation by exposing an Erk1/2 docking site, called DEF domain. Interestingly, Egr1 also has a DEF domain and several potential Erk1/2 phosphorylation sites (37). We investigated whether prolonged activation of Erk1/2 mediated by G-CSFR Y729F led to enhanced phosphorylation and activation of c-Fos and Egr1. In 32D and FDCP-mix A4 cells expressing the WT or Y729F form of G-CSFR, G-CSF stimulation resulted in rapid induction of c-Fos and Egr1, accompanied by their electrophoretic mobility shifts that became increasingly more significant up to 8 h after G-CSF stimulation (Fig. 8A). Both U0126 and PD0325901 blocked c-Fos and Egr1 induction by G-CSF (data not shown). Notably, the induction of c-Fos and Egr1 was more sustained, and their mobility shifts greater in cells expressing G-CSFR Y729F than in cells expressing WT G-CSFR. Consistent with this, phosphorylation of c-Fos on serine 32, which occurred one hour after c-Fos induction, was greater and more sustained in cells expressing G-CSFR Y729F.

Suppression of Erk1/2 Pathway Favors Neutrophil Development at the Expense of Monocyte Cell Fate in Response to M-CSF-Previous studies have provided supportive evidence for an instructive role of M-CSF in monocyte development
Egr1 has been shown to promote monocyte development at the expense of neutrophil cell fate (14,16,38). A potential role of c-Fos in monocyte development has also been proposed (5,15,39). We next examined whether sustained induction and phosphorylation of c-Fos and Egr1 enhanced their transcrip-  OCTOBER 2, 2015 • VOLUME 290 • NUMBER 40 tional activities. c-Fos heterodimerizes with Jun family transcription factors to form AP-1 proteins that activate transcription of target genes via the tetradecanoyl phorbol acetate response element (TRE). In both 32D and FDCP-mix A4 cells, G-CSFR Y729F mediated augmented activation of the luciferase reporter construct containing three repeats of TRE (40)  ( Fig. 8B). Activation of the reporter construct with four repeats of conserved Egr1 binding site (41) was also augmented in cells expressing G-CSFR Y729F (Fig. 8C). As expected, activation of the two reporter constructs by G-CSF was inhibited by U0126 and PD0325901 (data not shown). Thus, prolonged activation of Erk1/2 was associated with augmented activation of AP1 and Egr1.

G-CSF in Neutrophil Lineage Specification
Neutrophil Development Is Partially Restored upon Knockdown of c-Fos or Egr1 in Cells Expressing G-CSFR Y729F-To address whether enhanced activation of c-Fos contributed to monocyte development, we transduced 32D/Y729F cells with the lentiviral constructs containing two different shRNAs against c-Fos. shRNA 79 markedly and shRNA 80 moderately inhibited c-Fos induction by G-CSF (Fig. 9A). When cultured in G-CSF for 6 days, the c-Fos knocked down 32D/Y729F cells developed into morphologically mature neutrophils with significantly less adherence to culture dishes and reduced expression of M-CSF and Mmp-12 (Fig. 9, B-D). Notably, as in cells treated with U0126 and PD0325901, the expression of F4/80 was not altered by c-Fos knockdown (Fig. 9E). To address the role of enhanced Egr1 activation in monocyte development, we expressed 3 different Egr1 shRNAs in 32D/Y729F cells. As shown in Fig. 10A, shRNAs 25 and 26 inhibited Egr1 induction by G-CSF whereas shRNA 24 showed no effect. Notably, shRNAs 25 and 26, but not shRNA 24 supported neutrophil development at the expense of monocyte development (Fig. 10, B-D). Similar to c-Fos knockdown, F4/80 expression was not affected following Egr1 knockdown (Fig. 10E).
We also knocked down c-Fos and Egr1 in FDCP/Y729F cells. As shown in Fig. 11, knockdown of either c-Fos or Egr1 resulted in a shift in cell morphology toward neutrophils following culture in G-CSF for 2 days, which was associated with downregulation of M-CSF and Mmp-12, and up-regulation of NE and MPO. No significant changes in F4/80 expression were observed (data not shown). Thus, knockdown of c-Fos or Egr1 partially rescued neutrophil development in both 32D/Y729F and FDCP/Y729F cells. The data also indicated that c-Fos and Egr1 were not involved in the regulation of F4/80 expression.

Discussion
It has long been debated whether hematopoietic cytokines direct lineage specification and, if they do, little is known about the underlying signaling mechanisms. Two recent studies provide strong evidence for an instructive role of G-CSF and M-CSF in the regulation of lineage commitment toward neutrophils and monocytes, respectively, in GMPs and HSCs (20,21). However, how G-CSF and M-CSF regulate the intracellular signaling pathways to resolve neutrophil versus monocyte cell fate decision is still unknown.
In this report, we have shown that G-CSFR Y729F promotes monocyte rather than neutrophil development. Interestingly, G-CSFR Y729F has previously been shown to induce macrophage-like morphology in murine myeloid L-GM cells (25) and support significantly increased formation of macrophage colonies but reduced number of granulocyte colonies in mouse primary BM cells (27). Our data further indicate that monocyte development directed by G-CSFR Y729F is associated with prolonged activation of Erk1/2 and inhibition of Erk1/2 signaling largely rescues neutrophil development. Importantly, M-CSF, but not G-CSF, induces sustained and strong activation of Erk1/2 in mouse Lin Ϫ BM cells, and inhibition of Erk1/2 pathway favors neutrophil over monocyte/macrophage development in Lin Ϫ BM cells cultured in M-CSF. Thus, prolonged Erk1/2 activation results in monocyte development following G-CSF induction whereas inhibition of Erk1/2 signaling promotes neutrophil development at the expense of monocyte cell fate in response to M-CSF. It appears that the signals for terminal differentiation transduced by G-CSFR and M-CSFR might be similar, but the decision whether to develop along the neutrophil or monocytes/macrophage lineage largely depends on the duration and probably also the magnitude of Erk1/2 activation. These results point to an important mechanism by which G-CSF and M-CSF instruct neutrophil versus monocyte lineage choice, i.e. differential activation of Erk1/2 pathway. In support of this, it has been shown that inhibition of Erk1/2 signaling with U0126 or through overexpression of the Erk1/2-specific nuclear phosphatase DUSP5 led to neutrophil instead of monocyte development in response to M-CSF in a mouse E2A-Pbx1immortalized pro-T cell line transfected with M-CSFR (42,43).
The Erk1/2 signaling pathway has been shown to play a pivotal role in cell differentiation in different cellular systems. For instance, sustained ERK1/2 activation by nerve growth factor (NGF) induces neuronal differentiation in PC12 cells (44,45). Thrombopoietin-induced megakaryocytic differentiation is dependent on prolonged activation of the Ras-Erk1/2 pathway (46). Persistent Erk1/2 signaling has also been shown to be required for M-CSF-induced monocytic differentiation in myeloid FDCP1 cells (33) and that the Mek1/2 inhibitors U0126 and PD98059 inhibit the production of monocytes/macrophages from primary BM cells in vitro (32,47). The data presented here indicate that Erk1/2 pathway regulates neutrophil versus monocyte lineage choice, but is not required for the process of terminal differentiation. It is of note that treatment of Lin Ϫ BM cells cultured in M-CSF with U0126 and PD0325901 not only decreases the production of monocytes and macrophages, but also increases the total number of neutrophils, indicating that inhibition of Erk1/2 signaling redirects the Lin Ϫ BM cells to develop along the alternative neutrophil lineage in response to M-CSF. However, monocyte/macrophage development is not completely blocked by the Mek1/2 inhibitors in Lin Ϫ BM cells. It is possible that Erk1/2 pathway promotes, but is not essential for monocyte development. An alternative explanation is that some Lin Ϫ cells are already committed to the monocyte lineage and their terminal differentiation is not dependent on Erk1/2 signaling.
Our data also indicate that the Erk1/2 pathway promotes monocyte cell fate through c-Fos and Egr1. Prolonged Erk1/2 activation by G-CSFR Y729F is associated with more sustained induction, greater mobility shifts and augmented activation of c-Fos and Egr1, consistent with the previous study demonstrating that the persistently activated Erk1/2 directly phosphorylate and stabilize c-Fos and likely other early response gene products including Egr1 (36,37). Phosphorylation of c-Fos by Erk1/2 also primes it for additional phosphorylation (37). Indeed, G-CSF-induced c-Fos phosphorylation on Ser 32 was more sustained in cells expressing G-CSFR Y729F than in cells expressing WT G-CSFR. Egr1 has been shown to support monopoiesis and suppresses the alternative neutrophil cell fate (14,38). c-Fos has also been suggested to be involved in monopoiesis (5,15,39). Significantly, similar to the effects of Erk1/2 knockdown and the Mek1/2 inhibitors, knockdown of c-Fos or Egr1 largely restores neutrophil cell fate in response to G-CSF. Thus, c-Fos and Egr1 represent the key transcription factors that are differentially activated by G-CSF and M-CSF in an Erk1/2-dependent manner to resolve neutrophil versus monocyte cell fate. Additionally, our data reveal for the first time a critical role of c-Fos in monopoiesis.
In addition to activating c-Fos and Egr1, Erk1/2 have been shown to phosphorylate serine 21 of C/EBP␣ and thereby inhibit its activity (48). However, we have observed no significant difference in C/EBP␣ activity upon G-CSF stimulation of 32D cells expressing the WT or Y729F form of G-CSFR in lucif-erase reporter assays (data not shown). In contrast, we have consistently observed that G-CSFR Y729F, but not the WT G-CSFR activates a luciferase reporter construct containing three repeats of the conserved PU.1 binding site, which is not blocked by the Mek1/2 inhibitors. 3 In this aspect, it is interesting to note that the expression of F4/80 is not suppressed following inhibition of Erk1/2 signaling, or knockdown of c-Fos or Egr1. Notably, PU.1 activation has been associated with up-regulation of F4/80 expression (7, 49 -51). It is possible that G-CSFR Y729F may also activate other signaling pathways to promote monocyte development. Further studies are needed to examine the signaling pathway downstream of G-CSFR Y729F leading to PU.1 activation and determine whether PU.1 directly activates F4/80 expression.