Fcγ Receptor-mediated Mitogen-activated Protein Kinase Activation in Monocytes Is Independent of Ras*

Receptors for the Fc portion of immunoglobulin molecules (FcR) present on leukocyte cell membranes mediate a large number of cellular responses that are very important in host defense, including phagocytosis, cell cytotoxicity, production and secretion of inflammatory mediators, and modulation of the immune response. Cross-linking of FcR with immune complexes leads, first to activation of protein-tyrosine kinases. The molecular events that follow and that transduce signals from these receptors to the nucleus are still poorly defined. We have investigated the signal transduction pathway from Fc receptors that leads to gene activation and production of cytokines in monocytes. Cross-linking of FcR, on the THP-1 monocytic cell line, by immune complexes resulted in both activation of the transcription factor NF-κB and interleukin 1 production. These responses were completely blocked by tyrosine kinase inhibitors. In contrast, expression of dominant negative mutants of Ras and Raf-1, in these cells, did not have any effect on FcR-mediated nuclear factor activation, suggesting that the mitogen-activated protein kinase (MAPK) signaling pathway was not used by these receptors. However, MAPK activation was easily detected by in vitro kinase assays, after FcR cross-linking with immune complexes. Using the specific MAPK/extracellular signal-regulated kinase kinase (MAPK kinase) inhibitor PD98059, we found that MAPK activation is necessary for FcR-dependent activation of the nuclear factor NF-κB. These results strongly suggest that the signaling pathway from Fc receptors leading to expression of different genes important to leukocyte biology, initiates with tyrosine kinases and requires MAPK activation; but in contrast to other tyrosine kinase receptors, FcR-mediated MAPK activation does not involve Ras and Raf.

Antibodies (immunoglobulins) present two main functions in host defense: the binding to antigen via their antigen-combining sites and the mobilization of cellular defense mechanisms via their carboxyl-terminal Fc portion. Cross-linking of receptors for the Fc portion of immunoglobulin G molecules (Fc␥R) 1 on many cells of the immune system triggers various functions such as phagocytosis, antibody-dependent cell-mediated cytotoxicity, generation of the respiratory burst, and production of inflammatory mediators and cytokines (1)(2)(3).
Three classes of Fc␥R have been identified, Fc␥RI (CD64), Fc␥RII (CD32), and Fc␥RIII (CD16). They are coded for by different genes and differ in their relative avidity for IgG, molecular structure, and cellular distribution (4). Activation of Fc␥R as well as other immunoreceptors (such as TCR, BCR, and Fc⑀RI) results in common molecular events involving activation of Src family kinases followed by activation of Syk family kinases (5)(6)(7). The particular kinases involved depend on the particular immunoreceptor tyrosine-based activation motif (ITAM) present on the cytoplasmic portion of each receptor (8,9).
One of the major cellular responses initiated by Fc␥R crosslinking, specially in myelomonocytic and natural killer (NK) cells, is the activation of genes encoding cytokines important in inflammation, such as interleukin 1 (IL-1), IL-8, and tumor necrosis factor (TNF) (2,18,19). The signaling pathway from Fc␥R to the nucleus is not known, but it probably shares elements with the biochemical cascade used by other receptors known to activate gene transcription. In particular, receptors with intrinsic tyrosine kinase activity have been shown to induce transcription of genes via activation of the Ras signaling pathway (20), which turns on sequentially Ras, Raf-1, MEK, and mitogen-activated protein kinase (MAPK) (21,22). MAPK, also known as extracellular signal-regulated kinase (ERK) (23) phosphorylates and activates several transcription factors (24,25).
Due to the fact that recent reports indicate that MAPK is activated after Fc␥R cross-linking in various cell types (26 -32), it has been assumed that the classical Ras pathway is activated upon FcR signaling. However, no direct proof that Ras is used in Fc␥R signaling has been provided, except for a single report on NK cells (33).
Because activation of the transcription factor NF-B is required for IL-1 gene induction (34 -36), we decided to investigate directly if Fc␥R cross-linking on monocytic cells resulted in activation of this nuclear factor, and then we used this response as a final read-out to examine the involvement of the several elements of the Ras pathway in Fc␥R signaling, leading to gene activation and cytokine production.
We found that stimulation of the THP-1 monocytic cell line with insoluble immune complexes results in production of IL-1 and also in activation of the nuclear factor NF-B. Moreover, activation of this nuclear factor is mediated by MAPK but activation of this kinase does not seem to involve the classical Ras pathway defined for other receptor tyrosine kinases (20,37), since expression of dominant negative mutants of Ras and Raf did not have any effect on either MAPK activation or NF-B activation. In contrast, the MAPK kinase (MEK) specific inhibitor, PD98059, efficiently blocked activation of this nuclear factor to basal levels. These results indicate that MAPK is an important element in Fc␥R -mediated induction of cytokine genes (e.g. IL-1) in monocytes, and strongly suggest that activation of MEK and subsequently of MAPK occurs via a pathway that is independent of Ras and Raf.

EXPERIMENTAL PROCEDURES
Plasmids and Reagents-The following antibodies were used: Antipan ERK monoclonal antibody (catalog no. E171120, Transduction Laboratories, Lexington, KY), horseradish peroxidase-conjugated F(abЈ) 2 goat anti-mouse IgG (Cappel, Aurora, OH) The specific MEK (MAPK kinase) inhibitor PD98059 was from New England Biolabs, Inc. (Beverly, MA). The plasmids HIV-luc and E18pal-luc were a generous gift from Dr. John Westwick and Dr. David A. Brenner of the University of North Carolina, Chapel Hill, NC. HIV-luc contains NF-B-responsive elements within the human immunodeficiency virus long terminal repeat promoter placed upstream of the luciferase (luc) gene and directs the expression of luciferase in response to activation of the nuclear factor NF-B. E18pal-luc that activates luciferase transcription in response to the nuclear factor Ets. The plasmid encoding HA-MAPK was a gift from Mike Weber from the University of Virginia, Charlottesville, VA. Plasmids that direct the synthesis of normal or mutant forms of Ras and Raf were a gift from Dr. Channing Der from the University of North Carolina, Chapel Hill, NC. Ras constructs were all cloned in the retroviral vector pZIP. The Ras N17 (asparagine 17) mutant is a dominant negative form of this gene, while the Ras L61 (leucine 61) mutant is an active oncogenic form (38). The Raf 23-284 construct, which contains the amino-terminal domain of Raf-1 and acts as a dominant negative mutant (39), was cloned in the pCGN vector. To test the dominant negative mutants, cells were transfected with the Ras pathway-responsive reporter system GAL-Elk/5XGal-luc (40). The plasmid pEGFP-N1 (CLONTECH) containing the cDNA for the green fluorescent protein (GFP) was a gift of David García Díaz from the School of Medicine, University of Mexico, Mexico City. All other chemicals were from Sigma.
Insoluble Immune Complexes-Insoluble immune complexes (IIC) were prepared by mixing 300 l of rabbit anti-horse ferritin antibody (28 mg/ml) (Miles Laboratories Ltd., Slough, United Kingdom) and 30 l of horse ferritin type I (100 mg/ml) (Sigma) in Eppendorf tubes and incubating at 37°C for 60 min, followed by 12 h on ice. Insoluble immune complexes were separated by centrifugation at 20,000 ϫ g and were washed three times with sterile PBS. IIC were resuspended in 750 l of PBS and kept sterile at 4°C until use.
Preparation of F(abЈ) 2 Fragments-Anti-ferritin antibodies were subjected to pepsin digestion to prepare F(abЈ) 2 fragments. Briefly, 2.8 mg were dissolved in 0.1 M citrate buffer, pH 3.5, and pepsin (EC 3.4.23.1) (Sigma) was added at 25 g/ml. The mixture was incubated at 37°C for 4 h and then neutralized with 3 M Tris-HCl, pH ϭ 8.6. Undigested antibody was separated in a protein A-Sepharose column. Purity of F(abЈ) 2 fragments was confirmed by SDS-PAGE.
IL-1 Measurement-THP-1 cells (1 ϫ 10 6 ) were stimulated with 40 l insoluble immune complexes in 0.5 ml of RPMI 1640 complete medium for various times (0 -48 h) at 37°C. At the end of the incubation time, cells were centrifuged at 20,000 ϫ g and the supernatant collected and immediately frozen at Ϫ80°C. Interleukin 1 was measured in the supernatants with an ELISA kit (Amersham, Buckinghamshire, United Kingdom) according to the manufacturer's instructions. In some exper-iments, 30 M PD98059 or 10 M herbimycin A (Life Technologies, Inc.) was added 1 h before stimulation.
Transfections-THP-1 monocytic cells were transiently transfected with a DEAE-dextran method as described previously (41). Briefly, 1 ϫ 10 6 cells in 0.5 ml of serum-free RPMI 1640 medium were transfected with 5 g of plasmid DNA by incubating cells with 200 g/ml DEAEdextran (Pharmacia Biotech, Uppsala Sweden) for 60 min and after one wash, with 0.1 mM chloroquine for another hour at 37°C. Twenty-four hours after transfection, cells were resuspended in 4 ml of serum-free RPMI 1640 medium and stimulated with 40 l of insoluble immune complexes. Cells were collected after a 5-h incubation at 37°C and lysed with 65 l of lysis buffer (0.1 M Tris-HCl pH 7.8, 1% Triton X-100, 1 mM dithiothreitol, 2 mM EDTA). To evaluate transfection efficiencies in selected experiments, cells were transfected with the plasmid pGL3 control (Promega, Madison, WI), which constitutively expresses luciferase from the SV40 promoter. Cells were also transfected with the plasmid pEGFP-N1 (CLONTECH) containing the cDNA for the green fluorescent protein (GFP) under control of the cytomegalovirus promoter. Efficiency was estimated from the number of cells presenting green fluorescence at 24 h after transfection.
Western Blot-Total cell lysates or MAP kinase immunoprecipitates were resolved on 12% SDS-PAGE. Proteins were then electrotransferred onto polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA). Membranes were incubated in blocking buffer (1% bovine serum albumin, 5% nonfat dry milk (Carnation; Nestle Food Co., Glendde, CA) and 0.1% Tween 20 in PBS) overnight at room temperature. Membranes were subsequently probed with anti-pan ERK monoclonal antibody at 0.1 g/ml in blocking buffer, for 1 h at room temperature. Membranes were washed with PBS six times for 5 min each and incubated with horseradish peroxidase-conjugated F(abЈ) 2 goat antimouse IgG (Cappel, Aurora, OH), for 1 h at room temperature. After washing six more times with PBS, antibody-reactive proteins were detected with a chemiluminescence substrate (Pierce) according to the manufacturer's instructions.
Anti-phosphotyrosine Monoclonal Antibodies-A series of new monoclonal antibodies to anti-phosphotyrosine were obtained following standard techniques (42,43). Briefly, BALB/c mice were immunized with phosphotyrosine-coupled KLH in Freund's adjuvant. Splenocytes from these animals were fused to SP2/O myeloma cells and hybridomas selected by ELISA. Positive hybridomas secreted antibodies binding to phosphotyrosine-coupled ovalbumin, but not to tyrosine-coupled ovalbumin. Several hybridomas were cloned and characterized. Clone AFT8, an IgG1 producer, was selected for anti-phosphotyrosine Western blots. Monoclonal antibody AFT8 was purified from ascitis fluid by affinity chromatography in a protein A-Sepharose column.
Immune Complex Kinase Assay-MAP kinase was immunoprecipitated from THP-1 cell lysates (1.5 ϫ 10 7 cell equivalent) with 1 g of anti-pan ERK monoclonal antibody. The antibody was first incubated with 20 l of protein A-Sepharose (Pharmacia Biotech, Uppsala, Sweden) for 2 h at 4°C, and then mixed with the cell lysate for another 2 h at 4°C. Sepharose beads were then washed once with cold RIPA buffer and four more times with cold washing buffer (0.25 M Tris-HCl, pH 7.5, 0.1 M NaCl). Immunoprecipitates were resuspended in 40 l of kinase assay buffer (10 mM Tris-HCl, pH 7.5, 10 mM MgCl 2 , 1 mM dithiothreitol, 25 M ATP), containing 5 Ci of [␥-32 P]ATP (1.11 TBq/mmol; 2 mCi/ml) (NEN Life Science Products) and 10 g of myelin basic protein (MBP) (Sigma) and incubated at room temperature for 30 min. Reaction was stopped by adding 30 l of 3ϫ SDS sample buffer and boiling for 5 min. Samples were electrophoresed on 12% SDS-polyacrylamide gels. The phosphorylated substrate bands were analyzed by autoradiography. To evaluate the amount of protein immunoprecipitated, an aliquot of the sample was separated and Western blotted with anti-MAPK antibodies.
HA-MAPK Immune Complex Kinase Assay-In some experiments, HA epitope-tagged MAP kinase was transfected into THP-1 cells. After overnight culture to allow for cell recovery, the THP-1 cells were stimulated in various forms. HA-MAPK was immunoprecipitated from THP-1 cell lysates (1.5 ϫ 10 7 cell equivalent) with 14 g/ml of anti-HA monoclonal antibody 12CA5 (Boehringer Mannheim). The antibody was first incubated with cell lysates for 2 h and then 20 l of protein A-Sepharose (Pharmacia Biotech, Uppsala, Sweden) were added and the mixture incubated for another 4 h at 4°C. Sepharose beads were then washed once with cold RIPA buffer and four more times with cold washing buffer (0.25 M Tris-HCl, pH 7.5, 0.1 M NaCl). Immunoprecipitates were subjected to a kinase assay just as described above.

Fc␥ Receptor Stimulation Induces Interleukin 1 Production-
THP-1 monocytic cells were stimulated with insoluble immune complexes for various periods of time, and IL-1 produced and secreted in the supernatant was measured with a commercial ELISA kit. Fc␥R cross-linking by immune complexes induced a rapid and strong production of IL-1 by these cells (Fig. 1). IL-1 production reached a maximum (around 400 pg/ml) at about 24 h of stimulation. Previous treatment of cells with the selective tyrosine kinase inhibitor herbimycin A (44) abolished cytokine production, indicating that Fc␥R-mediated production of IL-1 requires protein-tyrosine kinase activity.
Fc␥ Receptor Stimulation Induces Tyrosine Phosphorylation-Because treatment of cells with the selective tyrosine kinase inhibitor herbimycin A (44) abolished cytokine production, we decided to look more directly at the effect of this drug on the FcR response. Stimulation of THP-1 cells with insoluble immune complexes caused rapid phosphorylation on tyrosine of several proteins. Prominent phosphotyrosine bands are increased at 1 min of stimulation at around 30, 35, 40, 44, and 70 kDa (Fig. 2). Treatment of cells with herbimycin A prior to stimulation completely abolished tyrosine phosphorylation of these proteins. Moreover, several other bands that were tyrosine-phosphorylated in the resting state also showed a significant reduction in presence of herbimycin A (Fig. 2). This result is in agreement with previous data indicating that Fc receptors recruit tyrosine kinases for their signaling (3,10). It also shows that herbimycin A is working well and it is blocking the tyrosine phosphorylation needed for Fc␥R-mediated IL-1 production.
Fc␥ Receptor Stimulation Induces Activation of the Nuclear Factor NF-B-Because the IL-1 gene, as well as other early immediate genes (such as those for IL-8 and TNF), requires activation of the nuclear factor NF-B for transcriptional acti-vation (34, 45), we evaluated NF-B activation in response to Fc␥R cross-linking in monocytes. THP-1 cells were transiently transfected with the NF-B reporter plasmid, HIV-luc, and luciferase activity was measured in cell lysates. Stimulation of transfected THP-1 cells by immune complexes resulted in a strong activation of the nuclear factor NF-B, as indicated by an increase (around 4-fold) in luciferase activity (Fig. 3A). Pretreatment of THP-1 cells with herbimycin A also completely blocked NF-B activation (Fig. 3A). Specificity of this response was tested by transfecting cells with the plasmid E18pal-luc that activates luciferase transcription in response to activation of the nuclear factor Ets. Stimulation of THP-1 cells with insoluble immune complexes did not induce luciferase activity from this plasmid (Fig. 3B). Treatment of the cells with 15 g/ml lipopolysaccharide induced luciferase activity from this plasmid (Fig. 3B), indicating that the response observed after immune complex stimulation is not due to general cell activation.
To confirm that only immune complexes were stimulating the cells via Fc receptors, transfected THP-1 cells were treated with various preparations of antibody and immune complexes (Fig. 4). None of the following stimuli caused activation of NF-B as indicated by an increase in luciferase activity: ferritin, the protein used to form the immune complexes, F(abЈ) 2 fragments of the anti-ferritin antibodies, and immune complexes formed with these F(abЈ) 2 fragments and ferritin (Fig.  4). The complete IgG molecule of anti-ferritin antibodies caused only a small activation, while the insoluble immune complexes gave the optimal response previously observed (Fig. 4). These data collectively indicate that Fc␥R aggregation is responsible for nuclear factor activation and production of IL-1, and that both events require protein-tyrosine kinase activity to take place.
Fc␥R-dependent NF-B Activation Is Independent of Ras-There is evidence that cross-linking of several immunoreceptors in leukocytes activates various elements of the Ras con- Cell lysates were prepared as described and proteins resolved by SDS-PAGE. Western blot for anti-phosphotyrosines was done with 5 g/ml monoclonal antibody AFT8. sensus signaling pathway (26 -32). To explore the possibility that Fc␥R signaling leading to nuclear factor activation in monocytic cells also involved elements of the Ras pathway, THP-1 cells were co-transfected with the NF-B-driven reporter plasmid and an expression plasmid directing the synthesis of a mutant form of Ras or Raf. Transfected THP-1 cells were then stimulated with insoluble immune complexes and NF-B activation evaluated by measuring luciferase activity in cell lysates. Expression of either wild-type Ras (not shown) or the dominant negative mutant Ras N17 did not affect Fc␥Rmediated NF-B activation (Fig. 5A). That the mutant Ras proteins were functional in these cells was confirmed by cotransfection of THP-1 cells with the corresponding Ras construct and the mitogen-responsive reporter system Gal-Elk/ 5XGal-luc, which detects Ras pathway activation (40,41). Cells were serum-starved for 48 h and then stimulated with 10% serum in the medium. Five hours after serum stimulation, a 3-fold increase in Ras activity was detected by measuring the luciferase activity (Fig. 5B). Co-expression of the dominant negative mutant Ras N17 blocked serum-induced activation of the reporter system to basal levels (Fig. 5B). Moreover, the presence of the activated oncogenic Ras L61 enhanced the Ras signaling initiated by serum (Fig. 5B). These results also con-firmed that, in these cells, the mutant forms of Ras were affecting Ras signaling activity as expected.
After Ras activation, the serine-threonine kinase Raf is the next element in the Ras signaling pathway (20,46). To explore if this kinase was involved in Fc␥R signaling in monocytes, THP-1 cells were also co-transfected with the NF-B reporter plasmid and the dominant negative mutant form of Raf-1, Raf 23-284. Expression of this altered protein did not prevent NF-B activation by insoluble immune complexes (Fig. 6A). Control experiments using serum stimulation of THP-1 cells transfected with the mitogen-responsive reporter system Gal-Elk/5XGal-luc indicated that Raf 23-284 can efficiently block serum-mediated activation of the Ras pathway (Fig. 6B). Thus, these results suggest that Ras and Raf are not directly involved in the Fc␥R-mediated signal transduction pathway in monocytic cells that leads to activation of NF-B. In order to exclude the possibility that these results were only valid on the THP-1 cell line used, the experiments were repeated in another monocytic cell line. U937 cells were transfected with the NF-Bdriven reporter plasmid and also with the dominant negative mutants of Ras and Raf (Fig. 7). Similarly to previous results, stimulation with insoluble immune complexes resulted in activation of the nuclear factor NF-B, as indicated by an increase in luciferase activity. The dominant negative forms Ras N17, and Raf 23-284 did not have any effect on NF-B activation (Fig. 7), supporting the previous data that FcR-mediated signaling for activation of genes does not involve the proteins Ras and Raf. Data in Fig. 7 also indicated that this results are valid for several monocytic cell lines.
Fc␥R-mediated NF-B Activation Requires MEK Activation-Although Ras and Raf did not seem to be involved in activation of the nuclear factor NF-B, many reports have indicated that MAPK is activated after cross-linking of Fc␥R. MEK is a cytoplasmic serine-threonine kinase that directly activates MAPK by phosphorylating its TEY domain on threonine and tyrosine residues (47). To determine if MEK was participating in Fc␥R signaling leading to nuclear factor activation, we used the selective MEK inhibitor PD98059, in THP-1 cells transiently transfected with the NF-B reporter plasmid. Treatment of cells with PD98059 for 60 min before stimulation with insoluble immune complexes resulted in complete inhibition of Fc␥R-mediated NF-B activation (Fig. 8). This result clearly showed the participation of MEK in the signal transduction pathway from Fc receptors leading to activation of this nuclear factor. Also, PD98059 affected the production of IL-1 by these cells after IIC stimulation. However, a concentration of PD98059 that completely blocked NF-B activation (Fig. 8) inhibited IL-1 production only about 40% (Fig.  9). Increasing concentrations of the MEK inhibitor did not further block IL-1 production.
Fc␥R Stimulation by Insoluble Immune Complexes Results in MAPK Activation-Our results, described above, indicated that Fc␥R signaling in monocytic cells did not involve Ras and Raf, but clearly activated MEK and the nuclear factor NF-B. To determine if MAPK was connecting MEK and NF-B, we decided to look directly at MAPK activation by immune complex kinase assays. Stimulation of THP-1 cells with IIC resulted in a clear and strong stimulation of MAPK activity (Fig. 10). The kinetics of Fc␥R-mediated activation of MAPK showed maximal kinase activity by 1 min of IIC stimulation. This activity had returned to basal levels around 3 min (Fig. 10). This result was in agreement with previous reports that Fc␥R cross-linking results in MAPK activation. Moreover, treatment of THP-1 cells with PD98059 for 1 h before IIC stimulation demonstrated that Fc␥R-mediated MAPK activation is completely blocked when MEK activation is inhibited (Fig. 11). These data suggested that, in Fc␥R signaling, MEK activation is an upstream event of MAP kinase activation, which then leads to nuclear factor NF-B activation.
Fc␥R-dependent MAPK Activation Is Independent of Ras-Data presented above indicated that MAPK is clearly activated by insoluble immune complexes and that the Ras and Raf dominant negative constructs did not inhibit NF-B activation. It was then important to determine directly if these dominant negative mutants did not blocked Fc␥R-dependent MAPK activation. It is not easy to see the effect of these mutants on MAPK directly because the efficiency of transfection of monocytic cells is rather low, approximately 5%, as estimated by transfections with a plasmid that expresses the GFP (data not shown). Therefore, the effect of the dominant negative constructs is only on those cells that were successfully transfected.
To test for the effect of these negative mutants on MAPK activity of only transfected cells, THP-1 cells were co-transfected with a MAPK that has the HA epitope tag and the corresponding Ras N17 or Raf 23-284 dominant negative mutants. After transfection cells were stimulated with insoluble immune complexes and cell lysates prepared. HA-MAPK was immunoprecipitated from these lysates with the HA-specific monoclonal antibody 12CA5, and its activity tested by in vitro kinase assays. The expression of Ras N17 did not inhibit the kinase activity stimulated by Fc␥R cross-linking with immune complexes (Fig. 12A). Control experiments using serum stimulation of transfected THP-1 cells indicated that Ras N17 could block MAPK activation induced by a different stimulus (Fig.  12C).
In a similar fashion we tested the effect of the Raf 23-284 dominant negative mutant on HA-MAPK activation after transfected THP-1 cells were treated with insoluble immune complexes. Expression of this mutant did not inhibit Fc␥R-dependent activation of MAPK (Fig. 13A). To confirm the efficacy of this negative constructs, transfected THP-1 cells were stimulated with serum and HA-MAPK activity measured in an in vitro kinase assay. Similarly to previous results, Raf 23-284 inhibited activation of MAPK induced by serum (Fig. 13C). These results all together support the idea that Fc␥R signaling does not use Ras or Raf to activate MEK and MAPK in the pathway that leads to nuclear factor NF-B activation. DISCUSSION Membrane receptors for the Fc portion of immunoglobulin G class antibodies (Fc␥R) are expressed on almost every type of hematopoietic cells. Cross-linking of these receptors by aggregated IgG, in the form of antigen-antibody complexes, triggers a very wide array of responses important for host defense and for modulation of the immune response (3). There is a great deal of interest in understanding the signaling mechanisms that lead to the various cell responses. One of the most important functions activated by immune complexes on myelomonocytic and NK cells is the production of inflammatory cytokines such as IL-1, IL-8, and TNF. This means that Fc␥R crosslinking induces transcription of the genes encoding these response (2,18). To have initiation of transcription of these genes, activation of diverse nuclear factors has to take place. Very little is known about the signal transduction pathway from Fc␥R to nuclear factors in the cell nucleus.
It has been observed that the 5Ј regulatory sequences of the cytokine genes induced by Fc␥R cross-linking (IL-1, IL-8, TNF), all contain sites for the nuclear factor NF-B (34 -36). We, therefore reasoned that NF-B activation would be an ideal way for monitoring the Fc␥R signaling pathway leading to gene induction. We decided to investigate directly if Fc␥R crosslinking on THP-1 monocytic cells also resulted in production of IL-1 and activation of the nuclear factor NF-B. Insoluble immune complexes indeed caused NF-B activation, as indicated by luciferase production from the NF-B-specific reporter plasmid (Fig. 3). This response is clearly mediated by Fc receptors because F(abЈ) 2 fragments of antibodies or antibody complexes made with them, were unable to stimulate the NF-Bspecific reporter plasmid (Fig. 4). So, Fc␥R stimulation in monocytic cells initiates a signal transduction pathway that activates nuclear factors and induces IL-1 gene expression.
Fc␥R, and also the antigen receptors on T lymphocytes and B lymphocytes, present a common feature that is important for signaling by all these immunoreceptors (5). They all contain a conserved motif, known as ITAM for immunoreceptor tyrosinebased activation motif (7,9), which contains phosphorylation sites important for signal transduction. Polyvalent ligands induce receptor cross-linking and activation of Src family (48,49) and Syk/ZAP-70 family related kinases (50 -52), which associate with the phosphorylated ITAM in the cytoplasmic tail of the receptor. After Fc␥R aggregation, these activated kinases catalyze the phosphorylation of cellular substrates on tyrosine residues (Fig. 2) (10). However, the nature of these substrates and other molecules involved in the signal transduction pathway is not clearly identified.
Activation of tyrosine kinases leading finally to nuclear factor activation resembles the signal transduction pathway defined for receptor tyrosine kinases that induces mitogenic signals in response to growth factors such as epidermal growth factor and platelet-derived growth factor. This signaling pathway is also known as the Ras pathway (20,37). Receptor tyrosine kinase activity induces transient formation of Ras-GTP and activation of Raf kinase at the membrane, followed by sequential activation of MEK and MAPK (22). MAPK is then responsible for activation of several transcription factors (25). Therefore, it has been thought that the classical Ras signal transduction pathway (37) may be involved in Fc␥R signaling leading to activation of nuclear factors and cytokine production.
Moreover, recent reports indicate that MAPK is activated after Fc␥R cross-linking in various cell types (26 -32), supporting the idea that the Ras pathway is used by Fc␥R to induce gene transcription. However, no direct evidence of Ras involvement has been provided in these reports. A recent publication indicates that an increase in Ras-GTP was observed after crosslinking of Fc␥RIIIA on NK cells (33), further supporting the idea for Fc␥R using the Ras signaling pathway. Because MAPK activation does not necessarily mean that the Ras pathway is being utilized, and because diverse signaling pathways may be used in different cell types, a direct evaluation for the involvement of the various elements of the Ras pathway in different cell types becomes important.
In this report, we investigated the participation of the Ras signaling pathway in the activation of cytokine genes upon Fc␥R cross-linking on monocytes, probing for the different elements involved in this signaling cascade.
The involvement of Ras signal pathway elements in the Fc␥R signal transduction pathway was investigated by measuring In order to confirm that this was a more general behavior of monocytes, the experiments were repeated in a different monocytic cell line. Data on U937 cells (Fig. 7) also showed that dominant negative mutant forms of Ras and Raf did not have any effect on activation of NF-B by immune complexes. Thus, in monocytic cells, Ras and Raf are not directly involved in the Fc␥R-mediated signal transduction path-way that leads to activation of NF-B. Although Ras and Raf did not seem to be involved in activation of the nuclear factor NF-B, many reports have indicated that MAPK is activated after cross-linking of Fc␥R, so downstream elements in the Ras pathway may still be used for Fc␥R signaling.
MEK is a cytoplasmic serine-threonine kinase that directly activates MAPK (47) and it is found downstream of Raf in the Ras signaling cascade activated by receptor tyrosine kinases (20). The selective MEK inhibitor, PD98059, completely blocked NF-B activation, indicating that MEK participated in Fc␥R-mediated signal transduction. Immune complex stimulation of THP-1 cells also resulted in activation of MAPK, as indicated by in vitro kinase assays using MBP as substrate for the kinase (Fig. 10). In addition, PD98059 was able to block MAPK activation back to basal levels (Fig. 11). This established a link between Fc␥R and the pathway MEK, MAPK, and NF-B. Moreover, evaluating directly MAPK activity only in transfected cells by using the HA-MAPK, it was found that the dominant negative mutants of Ras and Raf did not affect Fc␥Rdependent activation of this kinase (Figs. 12 and 13). These data further supported the idea that Fc␥R cross-linking activates a MAPK pathway without using the proteins Ras and Raf.
Treatment of THP-1 cells with a concentration of PD98059 that completely blocked NF-B activation resulted only in partial inhibition of IL-1 production. To initiate transcription of the IL-1 gene, more than one nuclear factor is required. The promoter region of this and many other genes contains multiple and different sites for nuclear factor binding (36,53,54). NF-B is one of the nuclear factors identified to bind at the 5Ј regulatory region of the IL-1 gene (34 -36). Thus, blockage of MEK and MAPK activity by the inhibitor PD98059 resulted in failure to activate NF-B, and therefore reduced Fc␥R-mediated IL-1 production in monocytes. Full transcriptional activation of the IL-1 gene needs NF-B and cooperation from other transcription factors (36,53,54). This cooperation effect has been demonstrated in the THP-1 cells for the IL-8 gene (41).
Stimulation of monocytic cells was done with immune complexes. These interact and stimulate all types of Fc receptors. However, it is known that the various types of Fc receptors activate different cellular responses (1,3,55), so it would be very interesting to know what type of receptor is responsible for the MAPK and nuclear factor activations that are connected to the induction of cytokine production by monocytes. We are now investigating this by stimulating cells with the monoclonal antibodies IV.3 and 3G8, which are specific for Fc␥RII and Fc␥RIII, respectively. Our preliminary results indicate that both receptors Fc␥RII and Fc␥RIII, expressed on monocytic cells, are capable of inducing NF-B activation, although they seem to do it at lower levels compared with immune complex stimulation.
How MEK is activated without Raf participation is not clear, but there is evidence for diverse ways to activate this kinase (56). MEK is phosphorylated and activated by an upstream kinase, which in the case of many serum growth factor receptors is the proto-oncogene Raf. Direct action of Raf over MEK is clearly established (57,58). Another MEK kinase, also called MEKK, has been described upstream of MEK in the signaling pathway from a different type of receptors (56). MEKK is the mammalian counterpart of the yeast protein kinases Byr2 (from Schizosaccharomyces pombe) and Ste11 (from Saccharomyces cerevisiae), which in turn activate the protein kinases Byr1 and Ste7, respectively. Byr1 and Ste7 have considerably sequence homology to MEK and function in the pheromoneinduced signaling pathway that leads to mating (56). Some pheromone receptors have a seven-membrane-spanning ser- pentine structure coupled to G proteins. Similarly, in mammalian cells, some serpentine receptors coupled to heterotrimeric G i2 proteins can stimulate DNA synthesis via MAPK activation (59). For example, G i2 -coupled acetylcholine muscarinic M2 receptors have been reported to activate MEK and MAPK independently of Raf. The kinase responsible for this effect is MEKK (59). Moreover, in mouse (NIH3T3) and rat (Rat1a) fibroblasts, MEK and MAPK are activated in response to epidermal growth factor (recognized by a receptor tyrosine kinase) and also in response to thrombin (recognized by a serpentine G protein-coupled receptor). Raf is activated by epidermal growth factor but not thrombin (60). It seems, then, that MEKK is a conserved kinase for the regulation of G protein-coupled signal pathways in yeast and vertebrates and Raf represents a divergence in vertebrates from the yeast pheromone-responsive protein kinase system (59,61). Whether G protein subunits activate MEKK directly or through an unknown intermediary molecule remains to be determined (61).
Fc␥R are not associated with G proteins. The kinase responsible for MEK activation after immune complexes stimulation remains unknown. Because G proteins may activate MEKK indirectly, it may be possible that this kinase is also used by Fc␥R to activate MEK. Another possibility is that Fc␥R may activate MEK via a different and yet undescribed kinase that has MEK kinase properties. It will be interesting to determine the involvement of MEKK in Fc␥R signaling to the nucleus to activate gene transcription.
Fc␥R signaling initiates with tyrosine phosphorylation (10). Herbimycin A, a selective inhibitor of tyrosine kinases, completely blocked both IL-1 production and activation of the NF-B-driven reporter plasmid in transient transfection assays, confirming that Fc␥R aggregation triggers activation of tyrosine kinases (see Fig. 2) as an early event in the signal transduction pathway from Fc receptors to gene activation and production of cytokines in monocytes. Syk kinase (72 kDa) has been implicated in Fc␥R signaling in several cell types. Syk belongs to the ZAP-70 kinase family. These enzymes are not myristoylated and therefore are exclusively cytoplasmic. Syk is present in all hematopoietic cells, whereas ZAP-70 is expressed in T cells and NK cells (62). In mast cells (RBL-2H3), MAPK activation has been clearly shown to by dependent of Syk, probably through the GTP/GDP exchange factor Vav (63). The link between Fc⑀R and MAPK may also be through Shc, which is phosphorylated by Syk and then binds to Grb2. This adaptor protein is known to associate with Sos to activate Ras upstream of MAPK (64), although in mast cells the Fc⑀R for IgE seems to connect Syk to MAPK via Ras. In monocytes, we did not find evidence for Ras or Raf involvement in MEK and MAPK activation after Fc␥R cross-linking. Syk is the most likely tyrosine kinase involved in Fc␥R signal transduction in THP-1 monocytic cells. In this report we did not look directly at Syk, but it will be interesting to confirm that Syk activation is required for MEK and MAPK activation in THP-1 cells. The mechanism that Syk may use to activate this downstream kinases bypassing Ras and Raf is unknown, but as discussed above, it may be through activation of MEKK. As of this report, there are no studies directly looking for a functional interaction between Syk and MEKK in any cell type.
Taken together, data presented in this work strongly suggest that the monocyte signaling pathway from Fc receptors leading to expression of different genes important to leukocyte biology (Fig. 14), initiates with tyrosine kinases and requires MAPK activation, but in contrast to other tyrosine kinase receptors, Fc␥R-mediated MAPK activation does not involve Ras and Raf.