Interleukin-3 and Granulocyte-Macrophage Colony-stimulating Factor Mediate Rapid Phosphorylation and Activation of Cytosolic c-rap

Interleukin-3 (IL-3) and granulocyte-macrophage colony-stimulating factor induce the rapid phosphoryl- ation of the c-rafprotein in the growth factor-depend- ent FDC-Pl and DA-3 murine myeloid cell lines. Fur- thermore, immunoprecipitates of c-raf isolated from growth factor-stimulated cells demonstrate a marked increase in intrinsic protein kinase activity as meas- ured in vitro. of c-raf at both and tyrosine cytoplasmic c-raf with con- comitant c-raf activation provide a cogent sequential molecular model for linking external growth stimuli to intracellular signal transduction events.

Interleukin-3 (IL-3) and granulocyte-macrophage colony-stimulating factor induce the rapid phosphorylation of the c-rafprotein in the growth factor-dependent FDC-Pl and DA-3 murine myeloid cell lines. Furthermore, immunoprecipitates of c-raf isolated from growth factor-stimulated cells demonstrate a marked increase in intrinsic protein kinase activity as measured in vitro.
IL-3 and granulocyte-macrophage colony-stimulating factor induce phosphorylation of c-raf at both serine and tyrosine residues. Antiphosphotyrosine immunoprecipitates from IL-3-stimulated cells demonstrate the rapid and coordinate phosphorylation of both c-raf and a protein co-migrating with the 140-kDa putative IL-3 receptor component.
Collectively, the findings of rapid and coordinate ligand-induced phosphorylation of a potential IL-3 growth factor receptor component and cytoplasmic c-raf with concomitant c-raf activation provide a cogent sequential molecular model for linking external growth stimuli to intracellular signal transduction events.
The growth and development of normal bone marrow progenitors are orchestrated by soluble glycoproteins known as hematopoietic growth factors (1). Substantial evidence shows that these same growth factors may be important mediators of leukemic cell growth (2). Hematopoietic growth factors exert their effects on normal and leukemic progenitor cells by initially binding to specific, high affinity cell-surface receptors (3,4). However, the components and sequence of post-receptor hematopoietic growth signal transduction are largely unknown. Current data strongly suggest that protein phosphorylation is an important early regulatory mechanism of growth signaling. Thus, a variety of transmembrane growth factor receptors contain intrinsic tyrosine kinase activity and exhibit ligand-dependent tyrosine phosphorylation, including those for PDGF,' epidermal growth factor, colony-stimulating factor 1, insulin, and possibly IL-3 and GM-CSF (5)(6)(7)(8)(9)(10)(11)(12) confer growth factor independence are protein-tyrosine kinases. These include abl, fms, and WC (13)(14)(15). Furthermore, the transforming potential is abrogated by mutations in the protein-tyrosine kinase domain of these molecules (16-l@, underscoring the importance of this enzymatic function in cell growth. Moreover, cytosolic protein kinases are plausible links between receptor-ligand interactions and growth factorspecific cellular events (19)(20)(21). Recent evidence suggests that protein kinase C, a cytoplasmic serine/threonine kinase, may mediate some of the IL-3-induced cellular events (22). Additionally, the existence of transforming genes such as rczf and mos which encode cytoplamic serine/threonine protein kinases strengthens the evidence implicating this class of molecules in growth regulation (23)(24)(25).
c-raf is a 74-kDa cytosolic serine/threonine protein kinase present in normal hematopoietic and leukemic cells (26). The viral homolog, v-raf, is oncogenic and cooperates with myc to induce erythroleukemia and lymphoma in mice (27). The combination of v-raf and v-myc induces IL-3-independent growth in factor-dependent murine myeloid cells (28). Furthermore, phosphorylation and activation of the c-raf protein have been observed after PDGF addition to fibroblasts (29). Collectively, these data suggest that c-raf specifically may play a role in post-receptor signal transduction. Results described here demonstrate that the unique hematopoietic growth factors IL-3 and GM-CSF can mediate rapid postreceptor biochemical events which appear to involve and/or converge at the level of c-raf, suggesting that c-raf may play a pivotal role in hematopoietic growth signal transduction. IL-3 and GM-CSF Mediate c-raf Activation Antibodies-Immunoprecipitation of the c-raf protein was performed using a polyclonal rabbit antiserum (cuSP63) as described (35). IG, mouse monoclonal antiphosphotyrosine antibody covalently linked to Sepharose beads was used as described according to the manufacturer's instructions (Oncogene Sciences) (6).
Metabolic Labeling and Immunoprecipitation-Exponentially growing cells were washed three times and resuspended at 1 x lo6 cells/ ml in RPM1 1640 medium containing either 1 or 10% FBS. Incubations were performed from 2 to 14 h at the designated temperatures (see "Results") to deprive cells of IL-3. After growth factor deprivation, cells were washed and resuspended at 1 x 10' cells/ml in phosphate-free RPM1 1640 medium and equilibrated with ortho[32P] phosphoric acid at 100 rCi/ml for 60 min. Radiolabeled cells were pulsed with growth factor for the times indicated (see "Results") and then immediately plunged into 10 volumes of ice-cold TBS. Cell pellets were washed twice in TBS at 4 "C and then lysed in TBS containing 1% Triton X-100 (Sigma), 20 rg/ml leupeptin, 1.9 pg/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride, 100 pM sodium metavanadate, and 20 mM sodium fluoride for 60 min at 4 "C. Extracts were centrifuged for 15 min in a Microfuge (Beckman Instruments), and the clarified lysates were incubated for 30 min with 100 ~1 of staphylococcal protein A-Sepharose beads (Calbiochem) per ml. The protein A-Sepharose beads were removed by centrifugation in a Microfuge, and the precleared lysate was subjected to quantitative protein determination using a bicinchoninic acid method according to the manufacturer's instructions (Pierce Chemical Co.). Lysates were normalized for protein content prior to immunoprecipitation using cuSP63 (anti-c-raf) or antiphosDhotvrosine antibodies. For &P163 immunoprecipitation, 1 ml of lysate &as incubated with 10 ~1 of (uSP63 antiserum and 100 pl of protein A-Sepharose beads for 14 h at 4 "C. c-raf antigen excess control immunoprecipitation experiments employed 10 ~1 of c&P63 antiserum preincubated with 5 rg of SP63 peptide for 30 min at 37 "C. This antigen/antibody mixture was then used for immunoprecipitation as described above. Immunoprecipitates were washed five times in ice-cold TBS containing 1% Triton X-100, 100 jtM sodium metavanadate, and 20 mM sodium fluoride. The washed immune complex-containing beads were suspended in 150 ~1 of SDS sample buffer, boiled for 5 min, and microcentrifuged. The resulting supernatants were resolved by SDS-PAGE and analyzed by autoradiography using Kodak X-Omat film. Immunoadsorption and elution of phosphotyrosine-containing proteins were performed according to-the manufacturer's instr&$ons. Eluates were boiled in SDS samnle buffer. resolved bv SDS-PAGE. and analvzed by immunoblotting and autoradiography as described ibove.
" In Vitro Kinase Reactions-Cell lysates were prepared and immunoprecipitated with (uSP63 with or without excess SP63 antigen as described above. Washed immunoprecipitates containing c-raf were resuspended in 100 ~1 of kinase buffer consisting of 100 rg of chicken erythrocyte histone H5, 5 mM MgC$, and 2 mM dithiothreitol in TBS. Kinase reactions were initiated by the addition of 50 pM ATP and 1 &i of [y-3ZP]ATP. Reaction mixtures were agitated at room temperature for 5 min and then terminated by boiling in SDS sample buffer. Samples were microcentrifuged, and the supernatants were resolved by SDS-PAGE and analyzed by autoradiography as described above. Optical densitometry was performed on the phosphorylated histone bands observed on the resulting autoradiographs. Phosphoamino Acid Analysis-The 32P-labeled c-raf protein bands identified on polyacrylamide gels were excised, and phosphoamino acid analysis was performed as described (36).

IL-3 Induces Rapid Phosphorylution
and Actiuation of c-raf Protein-Growth factor-deprived FDC-Pl cells exhibit a rapid increase in c-raf protein phosphorylation after IL-3 addition (Fig. 1, upper). An increase in the apparent molecular mass of the c-raf protein from 74 to 78 kDa accompanies phosphorylation, a feature consistent with c-raf phosphorylation also observed by others (29). The c-raf peptide SP63, to which the aSP63 antibody was raised, completely blocks the immunoprecipitation of this protein, confirming antibody specificity for c-raf. The immunoprecipitation of two other phosphoproteins of M, -105 and -88 is also blocked by the peptide. These proteins are not detected by (uSP63 immunoblotting (data not shown), suggesting that their coimmunoprecipitation with c-raf is not due to a shared epitope, but rather may result from avid association with the c-raf protein.
The significance of these potentially associated phosphoproteins in signal transduction, if any, remains to be determined. c-raf immunoprecipitates from FDC-Pl cells treated with IL-3 for 15 min demonstrate markedly enhanced histone phosphorylation in the in uitro kinase reaction compared with immunoprecipitates from untreated control cells (Fig. 1,  lower). These findings suggest that IL-3 can mediate the rapid phosphorylation and activation of intracellular c-raf. FDC-Pl cells were withdrawn from IL-3 for 14 h. Upper, cells were metabolically labeled with ortho[32P]phosphoric acid and treated with 50 units/ml IL-3 for 15 min. Immunoprecipitates prepared using orSP63 with or without excess SP63 antigen were resolved by 10% SDS-PAGE and analyzed by autoradiography as described under "Experimental Procedures." CTL, control. Arrowhead indicates the position of the c-ruf protein. Lower, cells were treated with 50 units/ml IL-3 for 15 min. Immunoprecipitates prepared as described above were added to an in uitro kinase reaction mixture containing chicken erythrocyte histone H5 and [T~~-P]ATP as described under "Exnerimental Procedures." Reaction mixtures were resolved by 15% SDg-PAGE, and phosphorylation was assessed by autoradiography. resents a physiologically significant event, time-and doseof c-ruf in DA-3 Cells-To determine whether the stimulatory response studies were performed. Results demonstrate that effect on c-raf was unique and limited to IL-3, GM-CSF was IL-3 induces rapid and time-dependent c-raf phosphorylation also studied. Using DA-3 cells, where growth can be supported in growth factor-deprived FDC-Pl cells (Fig. 2, upper). c-raf by either IL-3 or GM-CSF (Fig. 3), we find that both growth immunoblotting indicates that the increase in c-raf phos-factors can induce rapid, time-dependent phosphorylation of phorylation is not the result of increased c-rafprotein content c-raf (Fig. 4, upper). Furthermore, either of these growth after IL-3 addition (Fig. 2, center). c-raf phosphorylation factors can stimulate c-ruf protein kinase activity as demonoccurs at biologically significant concentrations of IL-3, and strated by increased in uitro histone phosphorylation (Fig. 4, the extent of phosphorylation correlates with mitogenicity as lower). Histone phosphorylation by cusp63 immunoprecipimeasured by ["Hlthymidine incorporation (Figs. 2 (lower) and tates prepared in the presence of excess SP63 antigen is nearly 3); that is, FDC-Pl cells demonstrate increased c-raf phosundetectable, confirming that this in vitro kinase activity is phorylation after treatment with 0.1 unit/ml IL-3, and phosspecifically associated with c-raf. phorylation is approximately half-maximal after treatment IL-3-and GM-CSF-induced c-raf Protein Phosphorylation with 1.0 unit/ml IL-3.
Occurs at Both Serine and Tyrosine Residues-Both IL-3 and

IL-3 and GM-CSF Induce Phosphorylation and Activation
GM-CSF can mediate the rapid tyrosine phosphorylation of proteins in hematopoietic cells (6,12). Additionally, IL-3 has been shown to induce the serine phosphorylation of several intracellular substrates in IL-3-dependent cells (22). Characterization of the amino acid sites on c-raf which are targeted for phosphorylation by IL-3 and GM-CSF could provide fundamental information regarding the mechanism(s) of c-raf activation. Therefore, phosphoamino acid analysis was performed on 32P-labeled c-raf protein from DA-3 cells treated with either IL-3 or GM-CSF. Results demonstrate that the growth factors induce a marked increase in both phosphoserine and phosphotyrosine content compared with c-raf protein from untreated cells (Fig. 5). These findings suggest a similar mechanism for IL-3-and GM-CSF-induced c-raf phosphorylation.

IL-3 Induces Rapid and Coordinate Phosphorylution
of c-raf and a 140-kDa Protein-Studies were performed to directly compare the kinetics of IL-3-mediated c-raf tyrosine phosphorylation with the known rapid tyrosine phosphorylation of a 140-kDa IL-3 receptor component (5). Antiphosphotyrosine immunoprecipitation demonstrates that IL-3 rapidly induces phosphorylation of several proteins in DA-3 cells (Fig.  6, upper). Three phosphoproteins predominate, which include a 74-and 55-kDa band as well as a protein which co-migrates with a 140-kDa putative IL-3 receptor component (5). Phosphorylation of these three proteins is coordinate and rapid, apparent 1 min after IL-3 addition and peaking at 5 min. Whereas some of the protein phosphorylation seen in Fig. 6A (upper) likely occurs on serine or threonine residues, the specificity of the antibody for phosphotyrosyl residues indicates that tyrosine phosphorylation accounts, at least in part, for the results observed. Anti-c-raf immunoblotting of antiphosphotyrosine immunoprecipitates confirms the identity of the 74-kDa phosphoprotein as c-raf (Fig. 6, center). Whereas IL-3-induced tyrosine phosphorylation of c-raf in DA-3 cells peaks at 5 min and rapidly diminishes, total c-raf phosphorylation, reflecting phosphorylation of both serine and tyrosine residues, continues to increase up to 60 min (Fig. 6, lower).
Thus, it appears that increased tyrosine phosphorylation of c-raf is rapid and transient, whereas increased serine phosphorylation may be sustained. 45 and treated with 10 units/ml IL-3 for the times indicated (upper and center) or incubated for 15 min with IL-3 at the various concentrations indicated (lower). &P63 immunoprecipitates were resolved by 10% SDS-PAGE and analyzed by autoradiography (upper and lower). c-raf phosphorylation, estimated using optical densitometry, is depicted in the histograms. After SDS-PAGE, the unlabeled immunoprecipitates (center) were electrophoretically transfered to a nitrocellulose membrane and probed with cuSP63 antiserum. c-raf was detected using lz51-protein A as described under "Experimental Procedures."

Murine
Myeloid Cells Expressing v-abl Are IL-3 Growthindependent and Demonstrate Constitutive c-raf Protein Phosphorylution-Since several oncogenes which encode proteintyrosine kinases including abl, fms, and src can abrogate cellular dependence on IL-3 for growth (13-15), we reasoned that such a regulatory perturbation might occur at a proximal intermediate step in the growth signal transduction pathway, i.e. c-raf phosphorylation. Such an effect might thereby circumvent the otherwise obligate early growth factor-mediated signal(s) such as ligand-induced tyrosine kinase activation. To directly test this hypothesis, we examined the effects of a Upper, cells were metabolically labeled with ortho[32P]phosphoric acid and then incubated with 10 units/ml IL-3 or GM-CSF for the times indicated. The 0 lane represents the untreated control. Immunoprecipitates prepared using c&P63 with or without excess SP63 antigen as described under "Experimental Procedures" were resolved by 10% SDS-PAGE, and c-raf phosphorylation was assessed by autoradiography. Lower, cells were treated with 10 units/ml IL-3 or GM-CSF for 15 min. Immunoprecipitates prepared using aSP63 with or without excess SP63 antigen were added to in vitro kinase reaction mixtures containing chicken erythrocyte histone H5 and [T-~*P]ATP as described under "Experimental Procedures." Reaction mixtures were resolved by 15% SDS-PAGE and analyzed by autoradiography. Histone phosphorylation was estimated using optical densitometry.
temperature-sensitive mutant of v-abl on IL-3 requirements for cell growth and on c-raf phosphorylation.
This mutant vabl protein (DP) contains an amino acid insertion in its tyrosine kinase domain which renders it exquisitely temperature-sensitive with respect to its tyrosine kinase activity and transforming efficiency (32). The FDDP-2 cell line was derived by introducing a retroviral construct containing this DA-3 cells were growth factor-deprived for 3 h, metabolically labeled with ortho[32P] phosphoric acid, and stimulated with 10 units/ml IL-3 or GM-CSF for 15 min. (uSP63 immunoprecipitates were prepared and resolved by 10% SDS-PAGE. 32P-Labeled c-raf protein was identified on the gels by autoradiography, excised, and subjected to phosphoamino acid analysis as described under "Experimental Procedures." temperature-sensitive v-abl into FDC-Pl cells (31). At the nonpermissive temperature (39 "C), temperature-sensitive v-abl has low tyrosine kinase activity, and FDDP-2 cells are dependent on IL-3 for growth. Under these conditions, IL-3 addition to cells induces rapid c-raf phosphorylation (Fig. 7). However, at the permissive temperature (32 "C), temperaturesensitive v-abl tyrosine kinase activity is increased II-fold (32), and the cells become independent of IL-3 for growth (31). In this instance, c-raf is found to be constitutively phosphorylated, and the addition of IL-3 has no further effect (Fig. 7). No temperature-associated changes in specific IL-3 surface binding or c-raf content are seen in FDDP-2 cells (data not shown). Phosphoamino acid analysis of 32P-labeled c-ruf protein isolated from FDDP-2 cells incubated at the permissive temperature reveals high levels of both phosphotyrosine and phosphoserine, consistent with the increased protein-tyrosine kinase activity of temperature-sensitive vabl at 32 "C (data not shown). DISCUSSION We have observed that IL-3 induces c-ruf protein phosphorylation as well as a significant increase in associated DA-3 cells were growth factor-deprived for 3 h, metabolically labeled with ortho['"P]phosphoric acid, and incubated with 10 units/ml IL-3 for the times indicated. The 0 lanes represent the untreated controls. Cell equivalents (5 x 10') were used per lane. Upper, immunoprecipitates prepared using IG, antiphosphotyrosine antibody were resolved by 10% SDS-PAGE and electrophoretically transferred to a nitrocellulose membrane as described under "Experimental Procedures." Phosphorylation was assessed by autoradiography. The exposure of the 5-min lane has been lessened to allow discrimination of individual bands. The expected positions of a 140-kDa putative IL-3 receptor and the c-rafprotein are indicated by arrowheads.
Center, the nitrocellulose membrane prepared as described (upper) was probed with c&P63 antiserum, and secondary immunodetection was performed using a calorimetric method as described under "Experimental Procedures." Lower, cuSP63 immunoprecipitates were resolved by 7.5% SDS-PAGE and analyzed by autoradiography. c-raf phosphorylation was estimated using optical densitometry. FDDP-2 cells were IL-3-deprived in RPM1 1640 medium supplemented with 10% FBS for 14 h at 32 or 39 "C. Cells were then metabolically labeled with ortho["'P]phosphoric acid and treated with 10 units/ml IL-3 for 15 min at the temperatures indicated. olSP63 immunoprecipitates were resolved by 10% SDS-PAGE and analyzed by autoradiography as described under "Experimental Procedures." intrinsic protein kinase activity. Furthermore, IL-3-induced c-rczf protein phosphorylation and activation occur very rapidly and at IL-3 concentrations relevant to its physiological effects (growth and mitogenesis). These findings form the basis for a model of early molecular events in cell growth whereby a specific ligand/receptor interaction at the plasma membrane provokes a biochemical reaction, i.e. c-raf phosphorylation, resulting in a potential phosphorylation cascade at the cytoplasmic level. This rapid activation step can also be mediated by GM-CSF, suggesting a general role for c-raf in hematopoietic growth signal transduction. c-raf phosphorylation and activation have been observed by others (29) following the addition of PDGF and the phorbol ester mitogen phorbol 12-myristate 13-acetate to murine fibroblasts. Our results extend these findings. Collectively, these data suggest that the c-ruf protein may play a pivotal role in growth signal transduction, perhaps serving as a cytoplasmic focal point in the transmission of a variety of mitogenic stimuli. Many growth factor receptors are known to be ligandactivated tyrosine kinases including those for PDGF, epider-ma1 growth factor, insulin, colony-stimulating factor 1, and possibly IL-3 and GM-CSF (5)(6)(7)(8)(9)(10)(11)(12). This enzymatic activity appears to be necessary for cell growth as mutations which interrupt tyrosine kinase activity abrogate growth factormediated cellular responses (11,(37)(38)(39). Significantly, IL-3and GM-CSF-induced c-raf phosphorylation occurs at both serine and tyrosine residues. IL-3-induced c-ruf protein tyrosine phosphorylation is very rapid, occurring by 1 min, and is coordinated precisely with the phosphorylation of a protein co-migrating with the 140-kDa putative IL-3 receptor ( Fig. 6 and Refs. 5 and 7). This finding raises the possibility that IL-3-induced c-raf tyrosine phosphorylation may be mediated directly by an IL-3 receptor-associated tyrosine kinase. In support of such an hypothesis, others (40) have found that the c-rufprotein is directly associated with the PDGF receptor in whole fibroblasts and that the PDGF receptor can bind the c-raf protein and mediate its tyrosine phosphorylation in vitro. Although we (5) and others (41) have previously reported the putative IL-3 receptor to be a phosphotyrosine-containing protein, no consensus sequence for protein-tyrosine kinase activity is present in a recently cloned, low affinity IL-3binding protein (42). These data could suggest that an associated protein-tyrosine kinase activity, distinct from the IL-3-binding component but activated by ligand binding may be responsible for the apparent ligand-induced tyrosine phosphorylation of the receptor and other proteins including craf. These intriguing possibilities remain to be determined.
Additional evidence implicating c-raf tyrosine phosphorylation in hematopoietic cell growth regulation may be gleaned from an investigation of oncogene products which abrogate the growth requirement of FDC-Pl cells for IL-3. Oncogenes within this class such as v-fms, v-abl, and v-src encode proteintyrosine kinases (13)(14)(15). Mutations which interrupt the tyrosine kinase activity also abrogate the physiological effects of these proteins, underscoring the importance of this enzymatic activity in growth regulation (16)(17)(18). The FDDP-2 cell line provides a unique, "reversible" system whereby a direct comparison of the subcellular events in IL-3-dependent growth can be made with those occurring during functional v-abl expression and IL-3-independent growth. At the nonpermissive temperature (39 "C), the temperature-sensitive vabl protein has low tyrosine kinase activity, and the cells are dependent on IL-3 for growth (31,32). Under these conditions, IL-3 induces c-raf phosphorylation. However, at 32 "C, where temperature-sensitive v-abl protein-tyrosine kinase activity is high and cell growth is IL-3-independent, a constitutive phosphorylation of c-raf is observed (Fig. 7). In this case, IL-3 has little to no effect on c-raf phosphorylation.
This finding supports a potential role for c-ruf in cell growth regulation IL-3 and GM-CSF Mediate c-raf Activation and suggests a mechanism by which this protooncogene may mediate the transforming effect(s) of v-abl.
Our results indicate that c-ruf protein phosphorylation apparently consists of at least two distinct phoshorylation events presumably mediated by different protein kinases which result in the modification of both tyrosine and serine residues. Whereas IL-3-mediated c-raf tyrosine phosphorylation peaks 5 min after growth factor addition and declines to near undetectable levels by 1 h, total c-raf phosphorylation continues to increase for up to 1 h. The mechanism of this apparent protracted serine phosphorylation is not known. One possibility involves c-raf autophosphorylation. However, several lines of evidence suggest that this is unlikely to be the only explanation for the substantial serine phosphorylation. First, although the in uitro histone phosphorylation observed was specifically c-raf-associated (Fig. 4, lower, 4th bar), at no time did we observe in vitro phosphorylation of a 74-kDa protein in the anti-c-raf immunoprecipitates derived from growth factor-stimulated cells. Second, even in the absence of histone in the kinase reaction, which might be a competitive substrate for activated c-raf, no c-raf autophosphorylation could be detected (data not shown). Third, it has been recently reported (43) that c-raf immunoprecipitates derived from fibroblasts demonstrate little autokinase activity. An alternative explanation for c-raf serine phosphorylation involves the action of another serine protein kinase which is distinct from c-raf. Two observations support protein kinase C as a potential candidate. First, protein kinase C has been strongly implicated in the mechanisms of action for several growth factors including IL-3 (22,44) and PDGF (45), suggesting at least a circumstantial association between this serine/threonine kinase and the c-raf protein. Second, two mitogenic activators of protein kinase C can induce c-raf phosphorylation and activation. Bryostatin 1, a macrocyclic lactone derived from the marine organism Bug&u neritinu, is a potent activator of protein kinase C (46) and is mitogenic in FDC-Pl cells (22). In addition, this compound induces rapid phosphorylation and activation of the c-ruf protein in FDC-Pl cells.' Recently, others (29) have observed that another protein kinase C activator, phorbol 12-myristate 13-acetate, also induces c-raf protein phosphorylation and activation. As such, these data are intriguing, but no direct evidence linking protein kinase C to c-raf serine phosphorylation and activation has been demonstrated. Although recently published studies (40) suggest that in vitro PDGF receptor-mediated c-raf tyrosine phosphorylation triggers c-raf kinase activation, no physiological role for c-raf serine phosphorylation is yet known. The aforementioned effects of protein kinase C activators indicate that c-raf phosphorylation at serine may also serve to activate its intrinsic kinase function.
In summary, these results, demonstrating IL-3-and GM-CSF-mediated phosphorylation and activation of cytosolic craf, suggest a potential rapid regulatory mechanism by which hematopoietic growth signals, generated at the plasma membrane, may be transmitted to the interior of the cell. Whether c-raf activation is necessary and/or sufficient to transduce these mitogenic stimuli remains to be determined.