Role of mitogen-activated protein kinase kinase in regulation of the epidermal growth factor receptor by protein kinase C.

The epidermal growth factor receptor (EGFR) is regulated by at least two mechanisms involving protein kinase C (PKC), inhibition of EGF binding and inhibition of EGF-stimulated tyrosine kinase activity. In this study we investigated whether mitogen-activated protein kinase (MAPK) mediates the inhibitory effects of PKC on EGFR binding or kinase activity by pretreating NIH3T3 and Chinese hamster ovary cells expressing the EGFR with PD98059, an inhibitor of MAPK/extracellular signal-regulated kinase kinase (MEK). We also determined whether substitution of cysteine for threonine at residue 669, the site of MAPK phosphorylation of the EGFR, alters the inhibition of kinase activity by PKC. The results indicate that 1) PKC down-regulates EGFR tyrosine kinase activity by an MEK-dependent mechanism presumably involving MAPK; 2) the inhibition by PKC is not a direct result of phosphorylation of the EGFR by PKC or MAPK; 3) activation of MAPK is not sufficient to regulate EGFR kinase activity; and 4) PKC-mediated down-regulation of EGF binding and EGFR kinase activity occur by different mechanisms. These data are consistent with a model for regulation of the EGFR by other receptors whereby their activation of PKC, in conjunction with MAPK, results in the phosphorylation of a protein(s) that modulates EGFR kinase activity.

Cellular integration of signals transmitted simultaneously by multiple receptor systems involves modulation of one receptor by another and often leads to selective desensitization. Heterologous regulation of receptors can occur through several mechanisms, including modulation of receptor stability, expression, cellular localization, ligand binding, or activity. Understanding how receptor signaling cascades are regulated is fundamental to an understanding of how the process of signal transduction is controlled.
One of the major classes of growth factor receptors is the ligand-stimulated tyrosine kinases, exemplified by the epidermal growth factor receptor (EGFR) 1 (1). The EGFR is regulated by at least two mechanisms involving other receptors or agents such as phorbol esters that stimulate protein kinase C (PKC): inhibition of EGF binding and inhibition of EGF-stimulated tyrosine kinase activity. A key issue concerning the mechanism by which this occurs is whether the EGFR is regulated as a result of direct phosphorylation by PKC or whether the EGFR is regulated by phosphorylation of other interacting proteins.
One widely accepted hypothesis is that the receptor is regulated by direct PKC phosphorylation of the EGFR at residue Thr-654. The evidence is based largely on site-directed mutagenesis of Thr-654, which, when mutated to Ala-654, renders the EGFR resistant to regulation by PKC (2,3). However, an alternative explanation is that the threonine residue itself rather than the phosphothreonine derivative is required for stabilizing interactions between the EGF receptor and PKCactivated effectors of the receptor. Recent studies have shown that this juxtamembrane region of the EGFR does interact with a number of proteins important for signaling (4 -6).
Several lines of evidence suggest that direct phosphorylation of the EGFR at Thr-654 by PKC is not sufficient to modulate receptor kinase activity. First, the observation that other agents, such as the calcium-mobilizing tumor promoter thapsigargin, can similarly down-regulate EGFR autophosphorylation independent of PKC activation suggest that another mechanism exists (7). Additionally, the level of phosphorylation of Thr-654 by PKC is well below the near stoichiometric levels expected for such an effect in some responsive cell lines including CHO cells transfected with human EGFR (7,8). Finally, we have demonstrated that mutant EGF receptors containing glutamate residues substituted at either Thr-654 or Thr-669, to mimic constitutive phosphorylation, possess EGF binding affinities and EGF-inducible tyrosine kinase activities which are comparable to those of the wild-type receptor (8). Furthermore, the Glu-654 as well as the Glu-669 mutant receptors were still responsive to PKC-mediated down-regulation albeit with slightly diminished sensitivity to phorbol esters. These results indicate that introduction of a negative charge in the juxtamembrane region of the EGFR is not sufficient to modulate EGFR binding or receptor tyrosine kinase activity.
In order to identify an alternative mechanism for regulation of the EGF receptor, we focused on other kinases that are activated by PKC. The MAP kinase signaling pathway is activated by both phorbol esters and thapsigargin, two tumor promoters that regulate the EGFR kinase in a similar fashion. To test whether MAPK might be responsible for the inhibition of the EGFR kinase activity, we determined the effect of an in-hibitor of MEK (MAPK kinase), PD98059 (9, 10), on the regulation of the EGFR by PKC. The results indicate that the MAPK signaling pathway is a key step in EGFR regulation, but the effect does not appear to be mediated through direct phosphorylation of the receptor. We have previously shown that MAPK can activate an EGFR-associated phosphatase activity that inhibits the EGFR tyrosine phosphorylation (11). Taken together, these results are consistent with a model whereby PKC and MAPK activate a protein such as a tyrosine phosphatase that inhibits tyrosine phosphorylation by the EGFR.

Generation and Maintenance of Cell Lines Expressing EGF Receptor
Mutants-CHO cells, maintained in Ham's F-12 medium supplemented with 10% fetal bovine serum, were transfected by the Ca 3 (PO 4 ) 2 method (12) with a wild-type human EGF receptor construct (CHO 2WT) and a Cys-669 EGFR mutant (CHO C669) as described previously. The constructs contained a cytomegalovirus promoter linked to the receptor cDNA, and were co-transfected in a 10:1 ratio with a pWL-Neo (Stratagene, LaJolla, CA) expression vector, which confers resistance to G418. Cells were selected in 200 g/ml G418, and a fluorescenceactivated cell sorter was used to select clones or similarly expressing populations in some cases. For this purpose, the R1 EGF receptor antibody (␣-EGF-R) was used in conjunction with a fluorescein-conjugated sheep anti-mouse IgG (Sigma). NIH3T3 cells expressing human EGFR (N24), generously provided by Stuart Decker, were maintained in DMEM supplemented with 5% fetal bovine serum. The MEK inhibitor, PD98059, was described previously (13).
MAPK Activation-Cells were grown to confluence, serum-starved overnight and treated as described in the figure legends. Equivalent protein aliquots from whole cell protein extracts in radioimmune precipitation buffer were separated by 7.5% SDS-PAGE, immunoblotted as described below, and probed with an anti-ERK1 antibody (Santa Cruz).
Regulation of EGF Binding by PDBu-24-well plates of confluent serum-starved cells were untreated or preincubated with PD98059 for 10 min and then either left untreated or treated with the indicated concentration of PDBu for 10 min. Cells were subsequently placed on ice, washed with phosphate-buffered saline, and incubated for 2 h at 4°C with 0.1 nM 125 I-EGF (DuPont NEN) in binding buffer (DMEM with 50 mM HEPES, 0.1% bovine serum albumin, pH 7.4). Cells were thoroughly washed with phosphate-buffered saline to remove unbound ligand and then lysed in 1 N NaOH. Radioactivity was quantitated by counting the lysate in an LKB gamma counter.
Immunoblotting-Immunoblotting was performed by resolving cell extracts by 6 -7.5% SDS-PAGE under reducing conditions and then blotting the proteins onto nitrocellulose with a Fisher semi-dry transfer apparatus according to Fisher's protocol. The nitrocellulose was blocked with 5% bovine serum albumin in 50 mM Tris, pH 7.5, 150 mM NaCl, 0.2% Tween 20 (TBST), and the blots were then probed with desired antibodies in TBST. Immunoreactivity was localized by horseradish peroxidase-conjugated secondary antibodies (Sigma), visualized by Renaissance chemiluminescence (DuPont NEN), and the autoradiographic films were quantitated by Ambis densitometry.

Inhibition of MAPK Activity in N24 and CHO 2WT Cells by
PD98059 -In order to assess the efficacy of the MEK inhibitor, PD98059 was added to either murine NIH3T3 (N24) or Chinese hamster ovary (CHO 2WT) cells that had been stably transfected with human EGFR. Cells were preincubated with or without PD98059 for 10 min and then stimulated with EGF for 4 min. Samples were assayed for the ability of MEK to activate MAPK by two assays: 1) phosphorylation of myelin basic protein (MBP) by immunoprecipitated MAPK, or 2) MAPK gel shift in which activated forms of ERK1 and ERK2 are shifted to a slower mobility on SDS-PAGE. Fig. 1 shows dose-dependent inhibition of EGF-activated MAPK by PD98059. MAPK gel shift analysis demonstrated that this compound can inhibit the activation of ERK1 and ERK2 upon stimulation by EGF in both N24 and CHO 2WT cells (Fig. 1A), although the potency of PD98059 is greater in N24 cells. Whereas 50 -100 M PD98059 resulted in only 50 -75% inhibition of MAPK in CHO 2WT cells, quantitative analysis using the MBP phosphorylation assay showed that preincubation of N24 cells with 10 -50 M PD98059 blocked EGF-stimulated MAPK activity by Ͼ60 -100%. Inhibition of MEK by PD98059 was not limited to EGFstimulated cells. As shown in Fig. 1B, 50 M PD98059 also blocked activation of MAPK by 40 nM PDBu in N24 cells.
The incomplete suppression of EGF-stimulated MAPK in CHO 2WT cells by PD98059 could not be enhanced by longer preincubation times. Once added to the cells, the potency of PD98059 diminished gradually over time, but inhibition was still observed following overnight preincubation (Fig. 1C). Analysis of different preincubation times indicated that a 10min incubation with PD98059 was maximally effective for both cell lines ( Fig. 1D; data not shown). The different sensitivities of the two cell lines to PD98059 may be a reflection of differential uptake of the inhibitor or may be attributable to differences in the MEK/MAPK signaling pathways. Since PD98059 completely blocked EGF-stimulated MAPK activation in N24 cells, most of the following studies were done in this cell line.
Inhibition of MAPK Activity Does Not Alter EGFR Autophosphorylation-To determine whether PD98059 has any effect on EGFR kinase activity, receptor tyrosine phosphorylation in response to EGF was assessed in N24 cells that were either untreated or pretreated with PD98059. Addition of 50 M PD98059 to cells had no effect on unstimulated EGF receptor autophosphorylation (data not shown). Similarly, no effect was observed on EGF-stimulated autophosphorylation with doses of PD98059 up to 100 M when added to either N24 or CHO 2WT cells (Fig. 2). Finally, pretreatment with 75 M PD98059 did not change the time course of EGF-stimulated receptor autophosphorylation when EGF was added to NIH3T3 cells for up to 30 min (data not shown). These results confirm that PD98059 has no direct effect on EGFR tyrosine phosphorylation, consistent with the observed specificity of PD98059 for MEK when PD98059 was tested with a number of other serine/ threonine and tyrosine kinases including EGFR and PKC (10,13). Furthermore, the data indicate that EGF-stimulated MAPK does not act as a feedback regulator of EGFR autophosphorylation.
Inhibition of MEK and MAPK Activity Does Not Alter PKCmediated Down-regulation of EGF Binding Affinity-Although activators of PKC have been shown to inhibit high affinity EGF binding, the mechanism by which PKC mediates this effect is unclear (14). One possibility is through PKC activation of the MAP kinase cascade. To assess whether MAPK plays a role in modulating high affinity EGF binding, down-regulation of EGF binding by the phorbol ester PDBu was assessed in N24 cells that were untreated or preincubated with PD98059. As shown in Fig. 3, addition of 0 -100 M PD98059 had no effect on the ability of 50 nM PDBu to inhibit the binding of 100 pM 125 I-EGF to the EGFR. These data demonstrate that MEK and MAPK do not play a role in the regulation of EGF binding by PKC. Furthermore, these results show that PD98059 does not alter activation of PKC by PDBu.
MEK Activity Is Necessary for PKC-mediated Down-regulation of EGFR Tyrosine Phosphorylation-Previous studies have shown that both PKC activators and calcium mobilizers inhibit EGFR kinase activity (2, 7). Since these agents also activate MAPK (15), we determined whether MEK or MAPK play a role in the PKC-mediated inhibition of EGFR tyrosine phosphorylation. Dependent upon the particular cell line and the dose of PDBu, EGFR kinase activity, as measured by EGFR autophosphorylation, could be suppressed up to 90% by PDBu (Fig. 4). Near maximal down-regulation of EGFR tyrosine phosphorylation was achieved in N24 cells with 40 nM PDBu and in CHO 2WT cells with 8 nM PDBu. Preincubation of N24 cells with 100 M PD98059, which completely inhibited EGF-stimulated MAPK, was sufficient to prevent down-regulation of EGFR tyrosine phosphorylation by 40 nM PDBu (Fig. 4A). Similarly, preincubation of CHO 2WT cells with 100 M PD98059, which only partially inhibited MAPK, almost completely reversed the inhibitory effects of 8 nM PDBu (Fig. 4B). The ability of the PD98059 to block PDBu-mediated down-regulation of the EGFR was dose-dependent, with an IC 50 of ϳ10 M in N24 cells (Fig. 4C), and correlated with the relative extent of MAPK suppression (Fig. 4D). These results indicate that MEK, probably through activation of MAPK, is required for PKC-mediated down-regulation of EGFR autophosphorylation. Furthermore, these data demonstrate that direct phosphorylation of the EGFR by PKC is insufficient to suppress EGF-stimulated tyrosine phosphorylation.
Mutating the MAPK Phosphorylation Site on the EGFR to Cysteine Does Not Alter Regulation of the EGFR Kinase by PKC-MAPK has been shown to directly phosphorylate the EGFR at residue Thr-669 (16). To determine whether direct phosphorylation of the EGFR by MAPK is the mechanism by which MEK modulates EGFR kinase activity, CHO cells expressing a mutant Cys-669 EGFR were utilized. These cells 6 -10) were treated with or without 10 nM EGF and varying concentrations of PD98059 as indicated. ERK1 and ERK2 activity were analyzed by immunoblotting and gel shift as described under "Experimental Procedures." B, inhibition of PDBu-stimulated MAPK activity by PD98059 in N24 cells. Equivalent numbers of cells were treated with or without 40 nM PDBu in the presence or absence of 50 M PD98059 as indicated. ERK1 and ERK2 activity were analyzed by immunoblotting and gel shift as described under "Experimental Procedures." C, dose response for PD98059 inhibition of EGF-stimulated MAPK in CHO 2WT cells. Equivalent numbers of serum-starved CHO 2WT cells were treated with differing amounts of PD98059 for 90 min prior to 4 min with 10 nM EGF. One lane was treated overnight with PD98059 prior to EGF treatment as indicated. ERK1 and ERK2 activity was analyzed by immunoblotting and gel shift as described under "Experimental Procedures." D, inhibition of MAPK activity in response to varying times of exposure of cells to PD98059. Equivalent numbers of serum-starved N24 cells were treated for different times with 75 M PD98059 prior to 4 min with 10 nM EGF. ERK1 and ERK2 activity was analyzed by immunoblotting and gel shift as described under "Experimental Procedures." The data in this figure represent more than five independent experiments.

FIG. 2. Inhibition of MAPK does not alter EGFR autophosphorylation.
Immunoblot analysis of EGFR tyrosine phosphorylation in N24 cells or CHO 2WT cells following cellular treatment with varying doses of PD98059. Equivalent numbers of serum-starved N24 cells (lanes 1-4) or CHO 2WT cells (lanes 5-8) were treated with varying amounts of PD98059 followed by 10 nM EGF for 4 min. EGFR autophosphorylation was detected by anti-phosphotyrosine immunoblotting as described under "Experimental Procedures." These data are representative of more than 10 independent experiments.

FIG. 3. Inhibition of MAPK does not alter PDBu down-regulation of high affinity EGF binding.
Serum-starved N24 cells were treated with 50 nM PDBu for 10 min with or without a 10-min preincubation with increasing concentrations of PD98059 prior to incubating at 4°C with 0.1 nM 125 I-EGF. The graph depicts percent EGF binding relative to cells that were not treated with PDBu. Error bars represent standard deviations from triplicate wells, and the data are representative of three independent experiments. express EGFR with the MAPK phosphorylation site substituted by cysteine, which has a structure similar to that of threonine but cannot be phosphorylated. Analysis of Cys-669 EGFR expression in a population of cells (CHO C669) that had been isolated by fluorescence-activated cell sorting showed comparable expression to that obtained for the wild-type EGFR in the CHO 2WT cells (Fig. 5A). Fig. 5B shows that the Cys-669 EGFR can be autophosphorylated at tyrosine residues in response to EGF treatment to a similar extent as the wild-type EGFR. Furthermore, the kinase activity of the Cys-669 EGFR is sensitive to PDBu down-regulation to the same extent as that of the wild-type receptor, with an IC 50 of 5 nM PDBu (Fig.  5C). These results suggest that the ability of MAPK to phosphorylate this site does not play a significant role in the regulation of EGFR tyrosine phosphorylation by PKC. DISCUSSION It has commonly been assumed that heterologous phosphorylation of the EGFR is both necessary and sufficient for downregulation of the receptor by PKC. In this study we demonstrate that PD98059, a MEK inhibitor, blocks PKC-mediated down-regulation of EGF-stimulated EGFR kinase activity in two different cell types. In contrast, inhibition of MAPK activation by this compound does not perturb PKC-mediated downregulation of high affinity EGF binding. These results suggest that the PKC-mediated pathway for down-regulating EGFR kinase activity is MEK-dependent, unlike its pathway for modulating high affinity binding. Furthermore, phosphorylation at the MAPK site is unnecessary, since a nonphosphorylatable Thr-669 mutant receptor (Cys-669) is still sensitive to this down-modulation. These results are consistent with a mechanism for PKC-mediated down-regulation of EGFR kinase activity whereby both PKC and MAPK regulate a modulator of EGFR tyrosine phosphorylation.
Direct phosphorylation of the EGFR by either PKC or MAPK does not appear to be sufficient for regulation of EGFR kinase activity. Substitution of a negatively charged glutamate residue at either Thr-654 or Thr-669 does not block the ability of PKC to down-regulate the EGFR. Furthermore, EGF stimulation of the Glu-654 EGFR kinase is comparable to that of wild type, indicating that MAPK activation, even in conjunction with a negative charge at residue 654, is not sufficient to inhibit the receptor kinase (8). In addition, the present study and CHO 2WT cells (B) were incubated with (q) or without (E) 100 M PD98059, followed by 10 min with or without different concentrations of PDBu and then 4 min with 10 nM EGF. Tyrosine-phosphorylated EGFR was analyzed by immunoblotting with anti-phosphotyrosine antibody as described under "Experimental Procedures." Autoradiograms were quantitated by densitometric scanning from anti-phosphotyrosine Western blots. The graphs are plotted as percent tyrosine phosphorylation of EGFR relative to cells treated only with EGF. Error bars represent standard deviations of triplicate samples. C, dose response for effect of PD98059 on EGFR autophosphorylation. Confluent N24 cells were pretreated with varying concentrations of PD98059 and then incubated with 40 nM PDBu for 10 min and 10 nM EGF for 4 min. Tyrosine-phosphorylated EGFR was analyzed by immunoblotting with anti-phosphotyrosine antibody as described under "Experimental Procedures." The graph is plotted as percent tyrosine phosphorylation of EGFR relative to cells treated with only EGF. The experimental controls, EGF alone or with 40 nM PDBu, were in duplicate. Error bars represent deviation from the mean. The results of the PD98059 treatments are representative of more than five independent experiments. D, analysis of MAPK activity in cells pretreated with varying doses of PD98059. Confluent serum-starved CHO 2WT cells (lanes 1-8) and N24 cells (lanes 9 -16) were incubated with (lanes 5-8 and 13-16) or without (lanes 1-4 and 9 -12 shows that MEK activity is required for regulation by PKC, indicating that direct phosphorylation at Thr-654 by PKC is insufficient. Finally, the Cys-669 mutant, a nonphosphorylatable residue, is indistinguishable from the wild-type EGFR in response to PKC regulation, indicating that MAPK phosphorylation of Thr-669 is not required. This result supports previous studies from our laboratory showing that direct phosphorylation of the EGFR by MAPK in vitro has no effect on EGFR tyrosine phosphorylation (11).
Furthermore, several lines of evidence suggest that phosphorylation at Thr-654, the site of PKC phosphorylation, is not required. First, EGFR in several cell lines such as CHO 2WT (8) are not phosphorylated significantly by PKC. Second, similar regulation of the EGFR has been observed in response to calcium-mobilizing agents that do not activate PKC or phosphorylate Thr-654 (7). The studies suggesting that direct EGFR phosphorylation is required are based upon site-directed mutagenesis. Substitution at Thr-654 of either an alanine (2,3) or cysteine, 2 a closer threonine mimic, completely blocks the modulation of the EGFR kinase by PKC. However, while these studies indicate that this residue is important, they do not establish that phosphorylation is required. The interpretation of these studies is difficult due to the lack of structural information about the EGFR. Taken together, these studies suggest that this region is important in this modulation of the EGFR but that phosphorylation alone cannot account for kinase down-regulation.
To date, no study has demonstrated that a single phosphorylation site on the EGFR is sufficient for modulating EGFR signaling. Numerous studies have altered all known serine and threonine phosphorylation sites alone or in combinations without yielding a receptor that is defective in kinase activity (2,3,17,18). In vitro studies have demonstrated that receptors with mutated autophosphorylation sites have wild-type levels of kinase activity (19). A similar study demonstrated that a decrease in the extent of autophosphorylation of the purified receptor does not alter its kinase activity (20). The only mutants identified to date that have reduced or no kinase activity are the ATP binding site mutant at Lys-721 and the wa-2 mutant receptor which was identified in mice (21,22). Both mutations are in the kinase domain at sites which do not undergo modification in vivo. These studies would seem to indicate that receptor phosphorylation events do not modulate EGFR kinase activity, particularly since the stoichiometry of most of them is too low. These results are consistent with the involvement of another component that can alter EGFR activity such as a tyrosine phosphatase.
Recently, a number of phosphatases that can regulate the EGFR have been described. PTP1D (SH-PTP2, Syp, SH-PTP3, PTP2C), an SH2-containing tyrosine phosphatase, acts as a positive mediator of EGFR signaling (23). A related SH2-containing tyrosine phosphatase which is localized predominantly in hematopoietic cells, PTP1C (SHPTP1, HCP, SHP) (24,25), has recently been shown to associate with the EGFR in human epidermoid carcinoma (A431) cells and act as a negative regulator of the receptor upon activation by phosphatidic acid (26). We showed previously that purified MAPK can decrease the activity of EGFR isolated from A431 cells via stimulation of a tyrosine phosphatase that co-immunoprecipitated with the receptor (11). A similar model may explain the mechanism by which PKC regulates EGFR tyrosine phosphorylation in CHO and NIH3T3 cells in vivo; however, several differences must be noted. While MEK is required for in vivo, regulation by PKC, inhibition of MEK does not alter EGFR tyrosine phosphorylation in the absence of active PKC. This difference between the in vitro and in vivo studies may reflect the absence of key regulatory components in the in vitro system, or it may reflect differences in the phosphatases or other proteins expressed in the different cell types. PTP1C is just one member of a large tyrosine phosphatase family consisting of both cytoplasmic and transmembrane enzymes, and these enzymes have distinct tissue distributions (27). Thus, a tyrosine phosphatase with a similar ability to negatively regulate EGFR tyrosine phosphorylation that is responsive to MAPK and PKC might be the target of PKC in our system.
While the results in this study implicate the MEK signaling pathway in the regulation of EGFR tyrosine phosphorylation by PKC, they clearly show that regulation of EGF binding by 2 P. Morrison and M. R. Rosner, unpublished data. showing the expression of Cys-669 EGFR (C669) relative to that of the wild type (WT, CHO 2WT) EGFR in transfected CHO cells. Equivalent numbers of CHO stable transfectants were lysed and total cellular protein was analyzed for EGFR expression by immunoblotting as described under "Experimental Procedures." B, immunoblot analysis of EGF-stimulated tyrosine phosphorylation of Cys-669 (C669) mutant and wild-type (WT, CHO 2WT) EGFR from CHO cells. Equivalent numbers of confluent serum-starved cells were treated with or without 10 nM EGF for 4 min. Total cellular extracts were analyzed by immunoblotting as described under "Experimental Procedures." C, autophosphorylation of Cys-669 EGFR. CHO cells expressing Cys-669 mutant (Ç) or wild-type (E) EGFR were serum-starved and treated with varying concentrations of PDBu prior to stimulating with 10 nM EGF. EGFR tyrosine phosphorylation was assayed as described under "Experimental Procedures" and quantitated by densitometric scanning from the anti-phosphotyrosine immunoblot. The graph depicts percent autophosphorylation of EGFR relative to EGFR from cells treated with only EGF. The experimental controls (EGF alone) are in duplicate. Error bars represent deviation from the mean. The data in this figure are representative of three independent experiments. PKC occurs through a different mechanism. Numerous agents have been shown to perturb EGF binding, but no consensus has emerged as to the nature of this modulation. Some agents such as phorbol esters are dependent upon the activation of PKC, while others do so through PKC-independent mechanisms. For example, thapsigargin and A23187 both require extracellular Ca 2ϩ to down-regulate high affinity EGF binding (7). Palytoxin also acts through a PKC-independent mechanism but requires extracellular Na ϩ to down-regulate total EGF binding (28). These results indicate that several mechanisms must exist for EGFR regulation.
The data in this study support an alternative model by which PKC can down-regulate the EGFR kinase. Our results demonstrate that the mechanism requires a PKC-activated cascade involving MEK and presumably MAPK. Taken together with our previous data (8,11), these results suggest that PKC regulation of the EGFR is dependent upon MAPK phosphorylation of another component, such as a tyrosine phosphatase, that modulates tyrosine phosphorylation by the EGF receptor.