Serum predominantly activates MAPK and Akt kinases in EGFR- and ErbB2-over-expressing cells, respectively

The ErbB receptor tyrosine kinase family consists of 4 members: epidermal growth factor receptor (EGFR), ErbB2 (also known as Her2/Neu), ErbB3 and ErbB4. They are characterized by their closely related structures: the presence of 2 extracellular cysteine-rich domains and a long intracellular C-terminal tail, on which most of the autophosphorylation sites are located (for review, Heldin, 1996). Members of the ErbB family, particularly EGFR and ErbB2, are over-expressed in various tumors and have been correlated with an adverse prognosis for cancer patients (Prigent and Lemoine, 1992). Both EGFR and ErbB2 can induce fibroblastic transformation and cellular proliferation (Di Fiore et al., 1987; Velu et al., 1987; Lehvaslaiho et al., 1989). These cellular processes may be induced by EGFR and ErbB2 using similar signal-transduction pathways. Upon activation, the tyrosine residues of EGFR and erbB2 become phosphorylated (for reviews, Heldin, 1996; Lemmon and Schlessinger, 1998). These phosphotyrosine residues serve as docking sites for adaptor proteins, such as growth factor receptor-bound protein-2 (Grb2) (for review, Pawson and Scott 1997), which can link EGFR and ErbB2 to the activation of Ras by stimulating the exchange of GDP for GTP on Ras. Ras is active when bound by GTP and inactive when bound by GDP (Downward et al., 1990). GTP-bound Ras in turn binds to Raf, resulting in activation of MAPK (for review, Marais and Marshall, 1996). Alternatively, GTP-bound Ras may bind to phosphoinositol-3 kinase (PI3K) (Rodriguez-Viciana et al., 1996), resulting in activation of Akt. EGFR and ErbB2 may therefore use similar Ras-dependent mechanisms to activate MAPK and Akt. Both EGFR and ErbB2 lack the binding motifs for PI3K, which are found in ErbB3 (Kim et al., 1994; Soltoff et al., 1994). EGFR and ErbB2 can induce the phosphorylation of ErbB3 and subsequently the binding of PI3K to ErbB3 (Kim et al., 1994; Soltoff et al., 1994). This binding has been correlated with the activation of PI3K and, presumably, Akt. Activated EGFR and ErbB2 may thus also use similar Ras-independent mechanisms to stimulate PI3K/Akt. Therefore, it is generally believed that EGFR and ErbB2 are very similar, if not identical, in how they activate MAPK and Akt. We previously used liposome-incorporated Grb2 anti-sense oligodeoxynucleotides (oligos) to down-regulate Grb2 protein expression (Tari et al., 1999). We found that such Grb2 down-regulation inhibited the growth of MDA-MB-468 and SK-Br-3 breast cancer cells, which over-express EGFR and ErbB2, respectively (Tari et al., 1999). We postulated that Grb2 down-regulation should, at least in part, be correlated with MAPK inactivation and found this to be the case for EGFR–over-expressing breast cancer cells but not ErbB2–over-expressing breast cancer cells (Tari et al., 1999). Our data therefore indicated that EGFR and ErbB2 do not use Grb2 protein in a similar fashion. This was surprising because EGFR and ErbB2 were presumed to use similar protein molecules to effect signal-transduction pathways. Serum and various growth factors are capable of activating MAPK and Akt. Some of these activation mechanisms could be mediated by EGFR and ErbB2. We postulated that EGFR and ErbB2 differ in how they regulate MAPK, and perhaps Akt, activities and that different effects on the activation of MAPK and Akt by serum and growth factors might be observed in EGFR– and ErbB2–over-expressing cells. We therefore examined whether serum and growth factors, such as EGF, may activate MAPK and Akt differently in EGFR– and ErbB2–over-expressing breast cancer cells. MDA-MB-468 and SK-Br-3 breast cancer cell lines (obtained from the ATCC, Rockville, MD) were cultured in DMEM/F12 medium supplemented with 10% heat-inactivated FBS at 37°C under 5% CO2 in a humidified incubator. Antibodies specific for MAPK and Akt were purchased from New England Biolabs (Beverly, MA), as were antibodies phospho-specific for MAPK (Thr202/Tyr204) and for Akt (Ser473). We (Tari et al., 1999) and others (Davies et al., 1998; Li et al., 1998) have shown that these phospho-specific antibodies can be reliably used to study the activation status/levels of MAPK and Akt. Breast cancer cells were plated at 1 × 105 cells/well in a 6-well plate in DMEM/F12 containing 10% FBS. Three days later, the medium was changed to DMEM/F12 containing 0.5% FBS. Various concentrations of EGF (Sigma, St. Louis, MO) were then added, and cells were harvested 20 hr later. Protein lysates were prepared and normalized before loading onto a 12% SDS-PAGE gel. Western blot analyses were then performed as previously described (Tari et al., 1999). Both MAPK and Akt were activated in EGFR–over-expressing MDA-MB-468 breast cancer cells (Fig. 1) and in ErbB2–over-expressing SK-Br-3 breast cancer cells (Fig. 2) in the presence of 10% FBS. In MDA-MB-468 cells, overnight serum starvation predominantly decreased MAPK activity and, to a much lower extent, Akt activity (Fig. 1). However, in SK-Br-3 cells, serum starvation decreased Akt activity but not MAPK activity (Fig. 2). Similar data were obtained with A431 and BT-474 cells, which over-express EGFR and ErbB2, respectively (data not shown). Thus, serum starvation appeared to have very different effects on MAPK and Akt activation in the EGFR– and the ErbB2–over-expressing breast cancer cells. In EGFR–over-expressing cells, activation of MAPK, but not Akt, was dependent on the serum concentration, while in ErbB2–over-expressing cells, activation of Akt, but not MAPK, depended on the serum concentration. MAPK and Akt activation in MDA-MB-468 cells incubated with 10% or 0.5% FBS and various concentrations (0, 25, 50, 75, 100 ng/ml) of EGF for about 20 hr. Protein lysates were then prepared, and Western blot analysis was performed. MAPK and Akt activation in SK-Br-3 cells incubated with 10% or 0.5% FBS and various concentrations (0, 50, 100, 200, 300 ng/ml) of EGF for about 20 hr. Protein lysates were then prepared, and Western blot analysis was performed. When the serum concentration was low, increasing concentrations of EGF could be used to activate MAPK in MDA-MB-468 cells (Fig. 1). Interestingly, EGF did not stimulate Akt activity in MDA-MB-468 cells (Fig. 1), despite the fact that EGF had been found to induce PI3K activity in these cells (Kim et al., 1994; Soltoff et al., 1994). EGFR expression in SK-Br-3 cells, though not as high as that in MDA-MB-468 cells, has previously been detected (Wang et al., 1999). Therefore, we also incubated SK-Br-3 cells with increasing concentrations of EGF. Under these conditions, EGF did not further increase the activity of MAPK or Akt in SK-Br-3 cells (Fig. 2). These results indicate that MAPK could be activated by EGF in MDA-MB-468 cells but not in SK-Br-3 cells. In other words, EGFR, but not ErbB2, could use EGF to activate MAPK. Also, the regulation of Akt activity was independent of EGF in both MDA-MB-468 and SK-Br-3 cells. Therefore, both EGFR and ErbB2 appear to use factors other than EGF to stimulate Akt activity. Our data suggest that EGFR and ErbB2 have different effects on the activation of MAPK and Akt by serum as well as by EGF. The activation of EGFR by serum and/or EGF leads mainly to MAPK activation, while that of ErbB2 by serum leads mainly to Akt activation. These different effects on MAPK and Akt activation may be due to the fundamental differences between EGFR and ErbB2. EGFR is activated by ligand binding, whereas ErbB2 is activated by over-expression. Activated EGFR mainly undergoes homodimerization, whereas activated ErbB2 mainly undergoes heterodimerization (for review, Lemmon and Schlessinger, 1998). Thus, EGFR and ErbB2, though closely related, are still biologically distinct. Yours sincerely, This research was supported by a grant from The University of Texas M.D. Anderson Cancer Center Breast Cancer Research Program (to AMT).