Sprouty2 inhibits the Ras/MAP kinase pathway by inhibiting the activation of Raf.

Several genetic studies in Drosophila have shown that the dSprouty (dSpry) protein inhibits the Ras/mitogen-activated protein (MAP) kinase pathway induced by various activated receptor tyrosine kinase receptors, most notably those of the epidermal growth factor receptor (EGFR) and fibroblast growth factor receptor (FGFR). Currently, the mode of action of dSpry is unknown, and the point of inhibition remains controversial. There are at least four mammalian Spry isoforms that have been shown to co-express preferentially with FGFRs as compared with EGFRs. In this study, we investigated the effects of the various mammalian Spry isoforms on the Ras/MAP kinase pathway in cells overexpressing constitutively active FGFR1. hSpry2 was significantly more potent than mSpry1 or mSpry4 in inhibiting the Ras/MAP kinase pathway. Additional experiments indicated that full-length hSpry2 was required for its full potency. hSpry2 had no inhibitory effect on either the JNK or the p38 pathway and displayed no inhibition of FRS2 phosphorylation, Akt activation, and Ras activation. Constitutively active mutants of Ras, Raf, and Mek were employed to locate the prospective point of inhibition of hSpry2 downstream of activated Ras. Results from this study indicated that hSpry2 exerted its inhibitory effect at the level of Raf, which was verified in a Raf activation assay in an FGF signaling context.

Several genetic studies in Drosophila have shown that the dSprouty (dSpry) protein inhibits the Ras/mitogenactivated protein (MAP) kinase pathway induced by various activated receptor tyrosine kinase receptors, most notably those of the epidermal growth factor receptor (EGFR) and fibroblast growth factor receptor (FGFR). Currently, the mode of action of dSpry is unknown, and the point of inhibition remains controversial. There are at least four mammalian Spry isoforms that have been shown to co-express preferentially with FGFRs as compared with EGFRs. In this study, we investigated the effects of the various mammalian Spry isoforms on the Ras/MAP kinase pathway in cells overexpressing constitutively active FGFR1. hSpry2 was significantly more potent than mSpry1 or mSpry4 in inhibiting the Ras/MAP kinase pathway. Additional experiments indicated that full-length hSpry2 was required for its full potency. hSpry2 had no inhibitory effect on either the JNK or the p38 pathway and displayed no inhibition of FRS2 phosphorylation, Akt activation, and Ras activation. Constitutively active mutants of Ras, Raf, and Mek were employed to locate the prospective point of inhibition of hSpry2 downstream of activated Ras. Results from this study indicated that hSpry2 exerted its inhibitory effect at the level of Raf, which was verified in a Raf activation assay in an FGF signaling context. Drosophila Sprouty (dSpry) 1 was discovered in genetic screens designed to detect novel genes involved in tracheal branching in the embryo (1). Fibroblast growth factor (FGF) is one of the agents involved in this phase of embryonic development, and dSpry was deemed to be an inhibitor of FGF signal-ing pathways. Correlative evidence suggested that the protein may be secreted and act as a competitive inhibitor at the receptor level (1). Another genetic screen, which analyzed genes associated with Drosophila eye development, further indicated that the inhibition by dSpry was not confined to FGF signaling but extended also to the epidermal factor receptor (EGFR) pathway. In contrast to the first study, dSpry was found to be an endogenous, membrane-located protein that inhibited the Ras/mitogen-activated protein (MAP) kinase pathway between receptor stimulation and the activation of Ras (2). The authors also provided preliminary evidence that dSpry binds directly to Drk, the Drosophila equivalent of mammalian Grb2, and to the GAP-1 protein. Another genetic-based study also found dSpry to be an endogenously acting protein that inhibited the Ras/MAP kinase pathway downstream of Ras activation around the level of Raf (3).
While Drosophila has only one Spry protein, mammals have at least four isoforms (1,3,4,5). This family of proteins is characterized by a highly conserved cysteine-rich C-terminal half and variable N-terminal sequences. Evidence from mammalian systems indicates that mSpry2 has an inhibitory effect on lung alveoli branching that parallels the effects on the tracheal system of insects (6).
The MAP kinase cascades constitute highly conserved signaling systems that have been deemed to play various roles in physiological responses such as in cell growth, differentiation, oncogenic transformation, immune responses, and apoptosis (7,8). The MAP kinase cascades are organized into core signaling modules consisting of three protein kinases: a MAP kinase kinase kinase (MKKK), a MAP kinase kinase (MKK), and a MAP kinase. Signals are transmitted through the module by sequential phosphorylation and activation of these component kinases (9). The MAP kinase family is composed of several sub-families including the ERKs, c-Jun N-terminal kinase (JNK), and p38 MAP kinase (10,11). In the mammalian system, the ERK cascade relays signals from receptor tyrosine kinases (RTKs) to the Ras family of small G-proteins, which then stimulates the sequential activation of Raf serine/threonine kinases (MKKK), MEK (MKK), and ERK1/2 (MAP kinase) (12,13).
Stimulation of cell proliferation and differentiation by FGF receptors acts via activation of the ERK pathway (14, 15). The small adaptor protein Grb2 has been shown to play an important role in linking RTK activation to the Ras/MAP kinase signaling pathway (16). Unlike most other RTKs, FGF receptors do not bind directly to Grb2 upon FGF stimulation. A novel lipid-anchored docking protein, FRS2, serves as a link between FGF receptor activation and the Ras/MAP kinase signaling pathway. FRS2 becomes tyrosine-phosphorylated upon FGF stimulation and associates with the Grb2⅐SOS complex to relay Phosphorylation of ERK2 by FGFR1 is inhibited by hSpry2 (A). 293T cells were transfected with wild type FLAG-ERK2, FGFR1, full-length HA-Sprys, and empty vector constructs. Anti-FLAG immunoprecipitates (IP) were resolved by Western blotting analysis to detect phospho-ERK2 and ERK2 using anti-phospho-ERK1/2 (phospho-p42/44) and anti-ERK2, respectively. Other associated proteins in the cell lysates were detected using anti-FGFR1 and anti-HA. IB, immunoblots. hSpry2 inhibits FGFR1-stimulated ERK2 kinase activity (B). 293T cells were transfected with wild type FLAG-ERK2, FGFR1, full-length HA-Sprys (S1, Sprouty1; S2, Sprouty2; S4, Sprouty4), and empty vector constructs (lane C). FLAG-ERK was immunoprecipitated, and kinase activity was determined using GST-Elk as described under "Materials and Methods." The ERK kinase activity was determined by densitometry quantitation of the 32 P GST-Elk band after autoradiography. The bar graph shown represents the mean value with standard error. Neither the N-nor the C-terminal halves of hSpry2 inhibit the phosphorylation of FGFR1-stimulated ERK2 (C). 293T cells were transfected with wild type FLAG-ERK2, FGFR1, HA-tagged Spry2, N-terminal or C-terminal hSpry2, and empty vector constructs, and the experiment was performed as in panel A. hSpry2 is an intracellular inhibitor of the Ras/MAP kinase pathway (D). 293T cells were transfected with wild type FLAG-ERK2, full-length or cytosolic domain of FGFR1, HA-Spry2, and empty vector constructs, and the experiment was performed activation signals down the Ras/MAP kinase pathway (17).
Apart from the genetic evidence, relatively little is known about the biochemical mode of action of the various Spry isoforms. One study demonstrated that a considerable part of the conserved C-terminal portion was involved in translocating the protein from cytosol to membrane upon RTK stimulation of cells (18). Assuming this highly conserved sequence is involved in directing the cellular localization of the protein, it is reasonable to hypothesize that the C-terminal half of the protein is involved in binding proteins that are strategically placed to execute physiological functions such as inhibition of the Ras/ MAP kinase pathway.
Currently, genetic evidence derived entirely from studies in Drosophila indicates that the physiological role of dSpry is as an inhibitor of the Ras/MAP kinase pathway. However, the three studies that employed genetic approaches in Drosophila are not in agreement as to the mode and site of the inhibition induced by dSpry (1)(2)(3). Mammalian Spry isoforms may have physiological activities that differ from those of dSpry and from each other. In this study, we asked several questions. In biochemical terms, do the various mammalian Spry isoforms inhibit the Ras/MAP kinase pathway? What features of the protein are required for the inhibition? Where does this inhibition take place?
DNA Expression Plasmids-Full-length cDNA of hSpry2, mSprouty1, mSprouty4, N terminus, and C terminus of hSpry2 were cloned as described previously (18). The Sprouty cDNAs were subcloned into the pXJ40FLAG and pXJ40HA mammalian expression vectors (obtained from Dr. E. Manser, Glaxo Group, Institute of Molecular and Cell Biology, Singapore) utilizing BamHI/XhoI restriction sites.
Plasmids of FLAG-tagged JNK1, FLAG-p38, and FLAG-ERK2 were kindly provided by Dr. Y. Zhang (Institute of Molecular and Cell Biology). GST-Elk was provided by Dr. K. L. Guan (University of Michigan). Full-length FGFR1 (flg) and FRS2␣ were cloned into mammalian expression vectors as described previously (20). The cDNA of the cytosolic domain of FGFR1 (FGFR1(cyto)) was cloned as described previously (21), ␤-2 adrenergic receptor construct was from Prof. G. Milligan (University of Glasgow), and Ras V12 was kindly provided by Dr. E. Manser. Mouse Gab1 cDNA was kindly provided by Dr. W. Birchmeier (Max Delbrü ck Center for Molecular Medicine, Berlin, Germany). Myctagged Akt1 (wild type), activated Raf-1 (Y340D), and activated MEK2 (S222D/S226D double mutant) in pUSEamp vector were from Upstate Biotechnology. The integrity of all constructs was confirmed by DNA sequencing or restriction digestion analyses.
Expression of GST-Elk Fusion Protein-The GST-Elk fusion protein used in the in vitro kinase assay was prepared as described previously (22).
Immunoprecipitations and Western Blot Analyses-Cells were harvested at 40 h post-transfection and lysed in 1 ml of lysis buffer (20 mM HEPES (pH 7.4), 137 mM sodium chloride, 1.5 mM magnesium chloride, 1 mM EGTA, 10% (v/v) glycerol, 1% Triton X-100, a mixture of protease inhibitors (Roche Molecular Biochemicals), and 0.2 mM sodium orthovanadate). Protein concentrations of cell lysates were normalized using a BCA protein assay kit (Pierce) before incubation with various antibodies. FLAG-JNK1, FLAG-p38, and FLAG-ERK2 were immunoprecipitated using mouse monoclonal anti-FLAG M2-agarose-conjugated beads. Myc-tagged Akt was immunoprecipitated using anti-Myc beads. Cell extracts were incubated with 2.5 g of the appropriate agarose-conjugated antibody for 2 h at 4°C. Immunoprecipitates were collected by centrifugation and washed three times with lysis buffer. Eluted proteins were resolved on SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes. After blocking in phosphate-buffered saline containing 1% bovine serum albumin and 0.1% Tween 20 for 1 h at room temperature, the membranes were probed with 1 g/ml primary antibody followed by 0.5 g/ml horseradish peroxidase-conjugated secondary antibody. Immunoreactive protein bands were visualized by enhanced chemiluminescence reagent (Amersham Biosciences, Inc.).
Kinase Assay and Protein Activity-Kinase activity of ERK2 was measured from immunoprecipitated FLAG-ERK2 samples using GST-Elk-1 as substrate as described previously (23). Ras activity was assessed using the Ras activation assay kit from Upstate Biotechnology. Ras-GTP from various treated lysates was "pulled down" using the GST fusion protein corresponding to human Ras binding domain of Raf-1 bound to agarose. The presence of Ras-GTP was detected by Western blotting using anti-Ras antibody (Upstate Biotechnology). Raf-1 kinase assays were performed using the Raf-1 immunoprecipitation kinase cascade assay kit from Upstate Biotechnology with some modifications. Raf kinase activity was measured by a coupled assay using GST-MEK, GST-ERK, and GST-Elk or myelin basic protein as sequential substrates. The assay product was assessed using P81 phosphocellulose paper and scintillation counting. Alternatively, the radiolabeled product was analyzed by a combination of SDS-PAGE and autoradiography. The assay measures the phosphotransferase activity of an in vitro kinase cascade reaction initiated by an immunocomplex containing Raf-1 (Upstate Biotechnology protocol, Ref. 24).

RESULTS
Based on the observation in mammalian embryo development that the expression of Spry isoforms coincide with the expression of FGF rather than EGF (25), we chose FGF-induced Ras/MAP kinase as our model system. Additionally we employed a cell culture system that best mimicked an inhibition of the pathway from the various in vivo systems reported: we reconstituted relatively long term stimulation of the FGF signaling pathway by overexpressing a wild type FGFR1 (flg) and observing the effects of overexpressed Spry isoforms at various assay points downstream with emphasis on the Ras/ MAP kinase pathway.
Full-length hSpry2 Is an Intracellular Inhibitor of the ERK Pathway but Not the JNK or p38 Pathways-FGFR1 was cotransfected into 293T cells with HA-tagged Spry isoforms and FLAG-tagged ERK2. Forty h after transfection, cells were lysed, and the lysates were precipitated with anti-FLAG. The immunoprecipitates and whole cell lysates were subjected to Western analysis using phospho-ERK1/2 (p42/44), ERK2, FGFR1, or HA antibodies to detect for phosphorylated ERK2 or the level of transfected proteins. The results in Fig. 1A show that hSpry2 significantly inhibits the phosphorylation of ERK2, whereas neither mSpry1 nor mSpry4 has any significant inhibitory effect.
as in panel A. The activated JNK pathway is not inhibited by Sprys (E). Cells were transfected with wild type FLAG-JNK1, FGFR1, HA-Sprys, and empty vector constructs. Anti-FLAG immunoprecipitates were resolved by Western blotting analysis to detect phospho-JNK and JNK using anti-phospho-JNK and anti-JNK, respectively. Other associated proteins in the cell lysates were detected using anti-FGFR1 and anti-HA. The activated p38 pathway is not inhibited by Sprys (F). Cells were transfected with wild type FLAG-p38, FGFR1, HA-Sprys, and empty vector constructs. Anti-FLAG immunoprecipitates were resolved by Western blotting analysis to detect phospho-p38 and p38 using anti-phospho-p38 and anti-p38, respectively. FGFR1 and Sprys were detected in the cell lysates with anti-FGFR1 and anti-HA. Similar experiments were next performed, except that the ERK precipitates were subjected to in vitro kinase analysis as outlined under "Materials and Methods." The data shown in Fig. 1B further demonstrate a profound inhibition induced by hSpry2, which is consistent with the results in Fig. 1A.
Previously we (18) have shown that the C-terminal half of the hSpry2 protein was responsible for the growth factor-stimulated translocation of the protein to membrane ruffles. We next asked whether this region was able to inhibit the Ras/ MAP kinase pathway when cells are stimulated with overexpressed FGFR1. Our results show that neither the C-terminal nor the N-terminal of hSpry2 was able to inhibit the Ras/MAP kinase pathway, which is in contrast to the inhibitory effects of full-length hSpry2. (Fig. 1C).
From an earlier study in Drosophila, it was believed that dSpry exerts its inhibitory effect by competing exogenously with FGF (1). Although subsequent evidence has demonstrated that dSpry2 operated on the inside of the plasma membrane (2), the issue has not been totally resolved. To investigate this issue in mammalian cells, we employed cytosolic FGFR1, which auto-activates to stimulate the Ras/MAP kinase pathway. Results from this experiment showed that FGFR1(cyto) stimulated phospho-ERK activity and that hSpry2 significantly inhibits this activation (Fig. 1D). Because there is no exogenous FGFR1 component, hSpry2 is exerting its inhibitory effect on the Ras/MAP kinase pathway endogenously.
We next determined whether any of the other two generic MAP kinase pathways were inhibited by any of the Spry isoforms. Similar experiments to those outlined above were carried out, except that FLAG-tagged JNK1 or FLAG-tagged p38 was substituted for the ERK2 construct. The assays were calibrated using UV light (for JNK1) or Traf2 (for p38) stimulation as positive controls. The results shown in Fig. 1, E and F, indicate that none of the three Spry isoforms exert any effect on the phosphorylation of JNK1 or p38. It is apparent from the data above that hSpry2 is the only isoform that inhibits a generic MAP kinase pathway, and this inhibition is confined to the ERK pathway.
hSpry2 Does Not Inhibit Either the Phosphorylation of FRS2 or the Associated Akt Pathway-ERK1/2 is somewhat distal from the receptor, and inhibition can potentially occur at various points along the pathway. To assess the site of inhibition of hSpry2, we decided to analyze key points along the canonical FGFR1 to ERK1/2 pathway.
Current evidence indicates that FRS2 is the major "distribution center" of FGFR-derived signals in mammalian cells. It has been shown to link to the Ras/MAP kinase pathway via Grb2, SHP-2, and SOS and to the Akt pathway via Grb2, Gab-1, phosphoinositide 3-kinase, and phosphoinositide-dependent kinases (17,26). We performed experiments to investigate whether any of the Spry isoforms affected FRS2␣, either via stimulation-induced tyrosine phosphorylation or by direct binding. FRS2 cDNA was co-transfected with FGFR1 and hSpry2 constructs. The cell lysates were later subjected to separation on SDS-PAGE and subsequent Western analysis employing PY20 antibody to detect tyrosine phosphorylation of FRS2␣. The data in Fig. 2A show that hSpry2 has no effect on the tyrosine phosphorylation of FRS2␣, which eliminates any effect of hSpry2 on the transactivation of FGFR1 as well as the ability of FGFR1 to phosphorylate FRS2␣. As these experiments involve relatively long term stimulation of the FGF pathway, it is possible that hSpry2 could either act as an in vitro competitor of the receptor or cause the induction of an exogenous inhibitor, as has been suggested elsewhere (27). The data shown preclude these possibilities as any impairment of receptor activity would present itself in the form of decreased tyrosine phosphorylation of its substrates, which is clearly not the case.
Using GST-Spry2 pull-down experiments it was also shown that FRS2␣ does not bind to the Spry proteins (data not shown). This further indicates that the target of hSpry2 inhibition on the FGF signaling pathway is downstream of the docker protein.
The activation of Akt via FRS2 and Gab1 represents a bifurcation from the Ras/MAP kinase pathway. Experiments were carried out using combinations of constructs for Myc-Akt, FGFR1, Gab1, and FLAG-tagged Spry isoforms to investigate whether hSpry2 elicited any effect on the Akt pathway. The cell lysates were subjected to precipitation with anti-Myc and assayed by Western analysis using antibodies against phospho-Akt. The results shown in Fig. 2B indicate that hSpry2 does not have any effect on the Akt pathway. Since hSpry2 did not have any effect at the level of receptor interaction with FRS2, our focus shifted to the next key point downstream; the possible effects on Ras activity.
hSpry2 Does Not Inhibit the Activity of Ras-The activation status of Ras was measured as described under "Materials and FIG. 2. hSpry2 does not inhibit either the phosphorylation of FRS2␣ or the associated Akt pathway. hSpry2 has no effect on the tyrosine phosphorylation of FRS2␣ (A). 293T cells were transfected with wild type FRS2␣, FGFR1, FLAG-Spry2, and empty vector constructs. Aliquots of the cell lysates were resolved by SDS-PAGE and tyrosinephosphorylated FRS2␣, and other related proteins were detected by immunoblotting (IB) with respective antibodies as indicated. Spry isoforms have no effect on the phosphoinositide 3-kinase pathway activated by FGFR1 (B). 293T cells were transfected with wild type Myc-Akt, FGFR1, Gab1, FLAG-tagged Sprys, and empty vector constructs. Anti-Myc immunoprecipitates were resolved by SDS-PAGE followed by Western blotting analysis to detect phospho-Akt and Akt using antiphospho-Akt and anti-Akt, respectively. Other associated proteins in the cell lysates were detected using anti-Gab1, anti-FGFR1, and anti-FLAG. IP, immunoprecipitates.
Methods." The main component of the assay was a recombinant protein derived from Raf that contains a domain capable of binding only to GTP-bound Ras (Raf-1 Ras binding domain). 293T cells were transfected with various combinations of constructs for FGFR1 and Spry isoforms. The results shown in Fig.  3 indicate that although transfected FGFR1 significantly enhanced the level of GTP-Ras, hSpry2 co-transfected into cells had no significant effect on these levels. Equal loading of whole cell lysates showed an inhibition of endogenous phospho-ERK by hSpry2, as demonstrated in previous experiments. Similar results were obtained in three such experiments. Accumulated data therefore indicated that hSpry2 did not inhibit the Ras/ MAP kinase pathway between the point of receptor activation and the activation of Ras. These data also rule out hSpry2's sequestration of Grb2 from SOS as a likely inhibitory mechanism. It was therefore necessary to assay the effect of hSpry2 on elements of the Ras/MAP kinase pathway downstream of Ras. It is apparent that there is some inhibition by mSpry1 and mSpry4, but it is very low compared with hSpry2.
hSpry2 Inhibits ERK Phosphorylation Downstream of Ras and Upstream of MEK and ERK-Experiments were carried out to locate the position of MAP kinase inhibition along the pathway from Ras to ERK. This pathway essentially involves three proteins: the G-protein Ras, which activates the protein kinase Raf; Raf which activates Mek; and Mek, which in turn activates ERK.
We took advantage of mutations in Ras, Raf, and Mek that cause an auto-activation of each of these respective proteins. Constructs were transfected into 293T cells along with FLAGtagged ERK2 and HA-tagged Spry isoforms. In each of these experiments, the cells were lysed, and the lysates were precipitated with anti-FLAG beads and analyzed by Western blotting using phospho-ERK1/2 antibody to ascertain the activation status of ERK (MAP kinase). The data from Fig. 4A show that hSpry2 can inhibit the phosphorylation (activation) of ERK2 downstream of active Ras (V12). A distinct inhibition of phosphorylated ERK2 was also observed in the mSpry1-transfected cells; however, the degree of inhibition by hSpry2 was more profound, and we subsequently focused more on the effects of this isoform. In addition, we demonstrated that the inhibition of activated ERK2 stimulated through the Ras mutant requires the full-length hSpry2 protein since neither the C-or N-terminal half was able to inhibit this activity (Fig. 4B). Contrary to the inhibition of activated Ras-generated signals, full-length hSpry2 did not cause any detectable inhibition of ERK2 phosphorylation when the constitutively active mutants of Raf (RafY340D) or Mek (MekS222D/S226D), respectively, were employed to activate the pathway (Fig. 4, C and D).
The above data indicate that hSpry2 is exerting its inhibitory effect at the level of activation of Raf. Raf has a complex mechanism of activation, and there currently appears to be some unidentified factors, especially kinases that may be involved. It would also appear from these data that hSpry2 should inhibit the MAP kinase pathway irrespective of what agonist stimulates it because the active Ras mutant used in this study is a generic stimulant that is independent of upstream pathways.
hSpry2 Inhibits ERK Activation Induced by ␤-2 Adrenergic Receptor Stimulation-A wide range of agonists activates the Ras/MAP kinase pathway. Sprouty has been shown to inhibit signaling pathways induced by receptor tyrosine kinases. Based on our results, we were interested in investigating the effects of hSpry when a non-RTK (for instance the isoprenaline activation of ␤2 adrenergic receptors (␤2ARs)), was used to stimulate the Ras/MAP kinase pathway. To address this, 293T cells were transfected with constructs for ␤2AR, HA-tagged hSpry2, and FLAG-tagged ERK2. Forty-eight h post-transfection, the cells were stimulated with isoprenaline (100 M) for various times, the cells were lysed, and the lysates were subjected to Western analysis using anti-phospho ERK1/2 to detect the level of ERK phosphorylation. The data shown in Fig. 5 demonstrate that ERK2 phosphorylation was elevated 5 min after isoprenaline addition and that this level was sustained for the next 25 min at least. hSpry2 substantially inhibited ERK2 phosphorylation to the degree that any level of phosphorylation in the data shown was only evident at 5 min. It has been reported that isoprenaline activation of ␤2AR can activate EGFR by "cross-talk" and that this activation can contribute to the subsequent activation of the Ras/MAP kinase pathway (28). The point illustrated by the above experiment is that it appears hSpry2 will inhibit the Ras/MAP kinase pathway downstream of activated Ras in a manner that is not dependent on the mode of pathway activation.
The accumulated data indicate that hSpry2 is the predominant inhibitor of the Ras/MAP kinase and that its inhibitory effect is exerted at the level of Raf activation. In this case, hSpry2 should inhibit the activation of Raf.
hSpry2 Inhibits the Activation of Raf-The effect of the various Spry isoforms on the status of Raf stimulation downstream of active FGFR1 was analyzed. FGFR1 was transiently expressed in 293T cells, and Raf kinase activity was measured by a coupled assay using GST-MEK, GST-ERK, and GST-Elk or myelin basic protein as sequential substrates. The assay measures the phosphotransferase activity in an in vitro kinase cascade reaction initiated in the immunocomplex by active Raf-1. The results show that FGFR1 significantly stimulated the Raf kinase activity in comparison with the vector control. When the cells were co-transfected with the various Spry isoforms, it can be seen that hSpry2 profoundly reduced the level of Raf kinase activity, whereas mSpry1 and mSpry4 caused lesser inhibition (Fig. 6A). A parallel inhibition of phosphorylated ERK2 was observed with hSpry2 overexpression, using the same cell lysates and Western blotting detection (Fig. 6B). In essence, hSpry2, in comparison with mSpry1 and mSpry4, causes a more profound inhibition of the Ras/MAP kinase pathway, and this down-regulation occurs at the level of Raf kinase activation. DISCUSSION The major question asked at the onset of this study was: in biochemical terms, can we detect a point on the Ras/MAP FIG. 3. Sprys do not inhibit FGFR1 activated Ras. 293T cells were transfected with FGFR1 FLAG-tagged Sprys (S1, S2, and S4) or empty vector constructs as indicated (C). Cell lysates were incubated with GST-Raf-1 Ras binding domain (RBD) agarose-conjugated beads. Bound proteins were denatured by boiling in reducing sample buffer and then resolved by SDS-PAGE followed by Western blotting analysis to detect Ras-GTP using a monoclonal anti-Ras. Aliquots of the same cell lysates were resolved by SDS-PAGE, and phospho-ERK1/2 and other related proteins were detected by immunoblotting (IB) as indicated. (S1, Sprouty1; S2, Sprouty2; S4, Sprouty4; PD, pull-down.) kinase pathway where any of the Spry isoforms inhibit? For reasons outlined previously, we have mainly used a model protocol based around constitutive FGFR activation. The results clearly indicate that hSpry2 inhibits the Ras/MAP kinase pathway at the level of Raf activation. The other Spry isoforms tested, mSpry1 and mSpry4, displayed some inhibitory activity, but the effects are small compared with that seen with hSpry2. Other groups have reported an inhibition of the Ras/ MAP kinase pathway by mSpry1 or mSpry4 in a different context (29,30). It is possible that other activation systems may provide a different spectrum of inhibition from that shown above. However, ours is the first in-depth and comparative study addressing Spry inhibition on the FGFR signaling pathway.
The 293T cells were transfected with the constitutively active Ras (RasV12) mutant, wild type FLAG-ERK2, HA-Sprys, or empty vector constructs as indicated (A). Anti-FLAG immunoprecipitates (IP) were resolved by SDS-PAGE followed by Western blotting analysis to detect phospho-ERK2 and ERK2 levels using anti-phospho-ERK1/2 and anti-ERK2, respectively. Ras and Spry proteins were detected using anti-Ras and anti-HA in the total cell lysates, respectively. IB, immunoblots. Inhibition of phosphorylated ERK through active Ras requires the full-length hSpry2 (B). 293T cells were transfected with wild type FLAG-ERK2, constitutively active Ras (RasV12), HA-tagged full-length Spry2, the N-or C-terminal half of hSpry2, and empty vector constructs. Anti-FLAG immunoprecipitates were resolved by Western blotting analysis to detect phospho-ERK2 and ERK2 using anti-phospho-ERK1/2 (phospho-p42/44) and anti-ERK2, respectively. Other associated proteins in the cell lysates were detected using anti-Ras and anti-HA, respectively. Sprys do not inhibit ERK1/2 activation by the Raf (Y340D) mutant (C). 293T cells were transfected with wild type FLAG-ERK2, constitutively active mutants of Raf (RafY340D), HA-Sprys, or empty vector constructs as indicated. Anti-FLAG immunoprecipitates were resolved by SDS-PAGE followed by Western blotting analysis to detect phospho-ERK2 and ERK2 using anti-phospho-ERK1/2 and anti-ERK2, respectively. Raf and Spry proteins were detected in the cell lysates using anti-Raf and anti-HA antibodies, respectively. hSpry2 acts on the Ras/MAP kinase pathway upstream of MEK (D). 293T cells were transfected with wild type FLAG-ERK2, constitutively active mutants of Mek (MekS222D/S226D), HA-Sprys, or empty vector constructs as indicated. Anti-FLAG immunoprecipitates were resolved by SDS-PAGE followed by Western blotting analysis to detect phospho-ERK2 and ERK2 proteins using anti-phospho-ERK1/2 and anti-ERK2, respectively. MEK and Spry proteins were detected in the cell lysates using anti-HA.
respectively (1,25). Although other mammalian Sprys have been shown to be co-expressed with various FGFs, it is not known what role they play in embryo development or signal transduction in the mature animal.
The Drosophila system, via its accessible genetics, has been invaluable for discovering novel proteins and locating them in various signaling pathways. There has been some controversy with respect to the site of action of dSpry. Originally it was thought that dSpry was a secreted protein that competed with growth factors for binding to RTKs (1). A second study indi-cated that dSpry was an endogenous cellular protein that was bound to the internal face of the plasma membrane via the conserved C-terminal, cysteine-rich sequence (2). Using FGFR1(cyto) to stimulate the Ras/MAP kinase pathway, we confirmed that hSpry2 exerts its inhibitory effect endogenously. Genetic analysis indicated that dSpry inhibited the Ras/MAP kinase pathway somewhere between the active RTK and the activation of Ras (2). An alternate point of action was presented following another Drosophila-based genetic study. In this case, it was postulated that dSpry inhibited the Ras/MAP kinase pathway at the level of Raf activation (3). Our biochemical study, albeit in a different system, would support Raf activation as the intersection point.
Raf has a complex and variable mode of activation that depends on a number of factors with phosphorylation being a crucial event. Raf associates with various inhibitory proteins and translocates to the plasma membrane before interacting with GTP-Ras. Current concepts suggest there are immediate upstream kinases of Raf that await discovery. We currently do not know the exact mechanism whereby hSpry2 inhibits Raf activation, and because our study mostly involves transient transfection, and with the associated time lag before analysis, we cannot exclude the possibility that hSpry2 induces the expression of a specific Raf inhibitor.
It is noteworthy from our studies that hSpry2 binds c-Cbl (31) and also inhibits the Ras/MAP kinase pathway. mSpry1 and mSpry4 display mediocre effects in the majority of these studies. These data indicate that the various isoforms of Spry are almost certain to have different functions, and the full spectrum of these awaits elucidation.
FIG. 5. hSpry2 inhibits ERK activation induced by ␤2 adrenergic receptor stimulation. 293T cells transfected with HA-hSpry2 and ␤2 adrenergic receptor were serum-depleted overnight and subsequently stimulated with isoprenaline (100 M) for various times. Cells were lysed and immunoprecipitated with FLAG antibody. The anti-FLAG immunoprecipitates (IP) were resolved by Western blotting analysis to detect phospho-ERK2 and ERK2 using anti-phospho-ERK1/2 (phospho-p42/44) and anti-ERK2, respectively. Other associated proteins in the cell lysates were detected using anti-HA. IB, immunoblots.
FIG. 6. Effect of Sprys on activated Raf-1 by FGFR1. 293T cells were transfected with empty vector (V), FGFR1 (C), or FGFR1 and FLAG-Sprys (S1, S2, and S4) constructs as indicated. Raf-1 was immunoprecipitated from lysates, and Raf activity was assessed by an in vitro coupled kinase assay in which recombinant MEK, ERK, and Elk-1 were used. Raf-1 activity was determined quantitatively by scintillation counting of the P81 phosphocellulose squares spotted with the assay product. Background counts were subtracted from each sample, and the resultant net counts were expressed as a percentage of the stimulated control in which no Sprouty was expressed. Aliquots of the same cell lysates were resolved by SDS-PAGE, and phospho-ERK1/2 and other related proteins were detected by immunoblotting (IB) as indicated. The data shown are representative of three independent assays. The bar graph shown represents the mean value with standard error. (S1, Sprouty1; S2, Sprouty2; S4, Sprouty4.)