Introduction

Top Abstract Introduction Materials & Methods Results Discussion Summary References

The EGF and the PDGF receptors are two highly distinct members of an expanding catalog of receptor tyrosine kinases implicated in the regulation of cellular growth control, morphogenesis and differentiation. The binding of either EGF or PDGF to their cognate receptors rapidly leads to tyrosine phosphorylation events including tyrosine autophosphorylation of the receptor, phosphorylation of proximal downstream targets, protein association and activation of enzymatic activities (for a review, see Ullrich and Schlessinger 1990).

Over the last decade, the signal transduction events resulting from activation of the EGF and PDGF receptors have been well studied. Moreover, it has been shown that the tyrosine phosphorylation events, including receptor autophosphorylation, are required for receptor activation leading to mitosis in cultured fibroblasts. The activation of receptor tyrosine kinases causes the phosphorylation of substrates including PI3K, phospholipase C and GTPase-activating protein (for a review, see van der Geer et al., 1994). A common feature shared by these molecules include a src homology domain 2, or SH2 domain, that has been shown to be involved in protein-protein interactions mediated by tyrosine phosphorylation (for reviews, see Pawson 1995; Cohen et al., 1995). It is widely accepted that receptor tyrosine kinase activation results in SH2-mediated interactions leading to a multiplicity of downstream signaling events including the activation of the MAPK activities (van der Geer et al., 1994). In addition, a large body of evidence has shown that both the PDGF and EGF receptors activate PI3K and MAPK activities through these mechanisms (van der Geer et al., 1994; Claesson-Welsh 1995; Boonstra et al., 1995). However, there is also evidence to suggest substantial qualitative differences between downstream signaling events after receptor activation between different classes of receptor tyrosine kinases. For instance, it has been shown that activation of PI3K activity after stimulation of the PDGF receptor may be due to an allosteric activation of the enzyme mediated through an SH2-mediated interaction of the p85 subunit with the activated receptor (Kazlauskas and Cooper 1989; Klippel et al., 1992; Carpenter et al., 1993). In contrast, there is evidence that the activation of PI3K activity associated with the EGF receptor may not be mediated through such a direct mechanism (Hu et al., 1992; Raffioni and Bradshaw 1992) but may be activated through more indirect mechanisms including the formation of heterocomplexes between the EGF receptor and related receptors such as Her-2, Her-3 and Her-4, to which PI3K has been shown to bind (Peles et al., 1992; Fedi et al., 1994; Sepp-Lorenzino et al., 1996).

The signals that result from receptor activation can lead to a progression of cells from the G1 to the S phase of the cell cycle. This progression has been shown to be accompanied by the activation of cyclin-associated kinase activities. In these studies, we have used small synthetic inhibitors of PDGF and EGF receptor tyrosine kinases to abrogate receptor-mediated cell signaling leading to mitogenesis. The use of such inhibitors is becoming more common as new inhibitors become available. However, such tools cannot be used indiscriminantly since their selectivity is usually relative. Thus, the goal of our studies was to demonstrate that to observe good selectivity for inhibition of receptor kinase activity and downstream signaling events, it is critical to carefully choose the correct concentration of inhibitor for the specific event under observation.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion Summary References

Materials. Compounds AG1296 (6,7-dimethoxy-2-phenylquinoxaline) and AG1478 (4(3-chlorophenyamino)-6,7-dimethoxyquinazoline) were obtained from Dr. Alex Levitzki and synthesized as previously described (Gazit et al., 1996; Barker 1993). The NIH 3T3 cell lines for these studies overexpressing approximately 1,000,000 receptors/cell (Ullrich A, personal communication) were provided by Dr. Axel Ullrich (Max-Planck Institute, Munich, Germany). MAP2 protein was purified from pig brain as described (Pang et al., 1992). ECL was purchased from Amersham (Arlington Heights, IL). [³²P]ATP (3000 Ci/mmol) was purchased from Du Pont/NEN (Boston, MA). DMEM and calf serum were purchased from Gibco BRL (Gaithersburg, MD). EGF was purchased from TOYOBO (Osaka, Japan). PDGF was purchased from Boehringer Mannheim (Indianapolis, IN). The antiphosphotyrosine monoclonal antibody was purchased from Upstate Biotechnology Inc. (Lake Placid, NY). All other reagents were obtained from Sigma Chemical Co. (St. Louis, MO).

Cell culture. NIH 3T3 mouse fibroblasts overexpressing human EGF receptor or the -form of the human PDGF receptor (3T3/EGFR cells and 3T3/PDGFR cells, respectively), were grown in DMEM containing 10% calf serum and 2 mM L-glutamine at 37°C under an atmosphere of 5% CO₂. Before ligand stimulation, cells were made quiescent by a combination of contact inhibition and serum-deprivation by overnight incubation of 80% confluent cultures in serum-free medium containing 2 mM L-glutamine. Unless otherwise indicated, cells were stimulated for 5 min with 100 ng/ml of PDGF (3.3 nM) or EGF (16.7 nM), which are high concentrations that gave maximal cellular responses.

Antiphosphotyrosine immunoblotting. After 5 min of ligand treatment, cells were washed three times with ice-cold PBS followed by the addition of 100 µl of SDS sample buffer (Laemmli 1970). Samples were heated at 95°C for 5 min, applied to 7% SDS-polyacrylamide gels and transferred to nitrocellulose paper. Immunoblots were treated with anti-phosphotyrosine antibody (diluted 1:3000), and developed by standard ECL methods.

Assay of phosphatidylinositol-3 kinase activity. PI3K activity was measured as previously described (Ohmichi et al., 1992) with the following modifications. Lysates were prepared from confluent cells (100-mm plates) that had been treated with ligands as described above. Cells were lysed in 500 µl of Nonidet P-40 lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 5 mM EDTA, 1% NP-40, 1 mM VO₄³, 100 ng/ml aprotinin and 100 ng/ml leupeptin) and lysates were treated with rabbit IgG and Pansorbin. A total of 5 µl of anti-phosphotyrosine antibody was then added and the samples were incubated overnight. Immune complexes were isolated using 50 µl of protein-G/A agarose. PI3K activity was assayed by dissolving the immunoprecipitate in a 50-µl reaction buffer containing 10 mM Tris (pH 7.5), (0.2) mM EGTA, 100 mM NaCl and 20 mM MgCl₂, 0.4 µg phosphatidylinositol and 10 µM [-³²P]ATP (10 µCi). The reaction was then incubated for 20 min at room temperature and terminated by the addition of a three-part mixture containing 100 µl of chloroform:methanol:hydrochloric acid (100:200:2), 100 µl chloroform and 100 µl of water. The lower organic phase was dried under nitrogen to completion and the material was dissolved in 25 µl of a chloroform:methanol mixture (1:1). Samples were then applied to a Silica Gel 60 thin-layer chromatography plate and developed for 2 hr using a solvent containing chloroform:methanol:water:ammonium hydroxide (43:38:7:5). Radioactivity incorporated into phosphoinositides was estimated by autoradiography.

MAP kinase activity. MAPK activity in ligand-stimulated 3T3/EGFR and 3T3/PDGFR cells was assayed from cell lysates as described previously (Pang et al., 1992). Specifically, 10-µl aliquots of cell lysates were incubated with pig brain MAP-2 at 0.2 µg/ml for 15 min at 30°C in a reaction containing a final volume of 25 µl including 50 mM Tris-HCl (pH 7.4), 2 mM EGTA, 10 mM MgCl₂ and 40 µM total ATP containing [-³²P]ATP (1 µCi). The reaction was stopped by the addition of SDS sample buffer (Laemmli 1970). MAP-2 protein was resolved by (7%) SDS-PAGE and stained with Coomassie blue. Stained MAP-2 was excised from the gels and incorporated radioactivity was measured by Cerenkov counting or estimated by autoradiography.

Cyclin-associated cdk activity. Quiescent cells were stimulated with ligand in the absence or presence of the indicated compound (25 µM) and then washed with PBS and lysed in a solution of 50 mM HEPES, pH 7.0, 150 mM NaCl, 0.1% Triton X-100, 10% glycerol, 1.5 µM phenylmethylsulfonyl fluoride, 1 mM EDTA, 5 mM Na₃VO₄, for 5 min at room temperature. Particulate matter was removed from the lysate by centrifugation, and the supernatant was rapidly frozen at -80°C until analysis.

Cell cycle analysis. Quiescent cells were stimulated with either EGF (4 nM) or PDGF (3.8 nM) and incubated for 20 hr at 37°C in the absence or presence of various concentrations of AG1296 or AG1478. Stimulated cells were collected in PBS, fixed by the treatment of the cell suspension with 70% ice-cold methanol and stained with propidium iodide. The DNA content was measured using a FACScan flow cytometer, and cell cycle phase distribution was estimated using CellFIT software (Becton-Dickinson, San Jose, CA).

BrdU incorporation assay. 3T3/EGFR cells were plated on glass coverslips, grown to 80% confluence and then made quiescent by serum-deprivation for 24 hr. Quiescent cells were stimulated with EGF (2 nM) or PDGF (3.8 nM) in the absence or presence of AG1296 (10 µM) or AG1478 (0.1 µM) for 20 hr. BrdU was added for a 2-hr labeling period and the cells were fixed and stained for DNA content with Hoescht 33258 (Sigma) and for BrdU incorporation with biotinylated anti-BrdU and Texas Red coupled to streptavidin (Boehringer Mannheim). Total nuclei and nuclei stained for BrdU were counted in seven frames to determine the percentage of cells in S phase, and representative frames were photographed.

Data analysis. All experiments were performed at least twice. IC₅₀ values were visually estimated from autoradiographs and ECL exposed film. When quantitative data were available, IC₅₀ values were determined by nonlinear regression using the data analysis and presentation program PRISM (Graphpad Inc., San Diego, CA). Statistical analysis of differences (two tailed t test) was also performed with PRISM.

	Results

Top Abstract Introduction Materials & Methods Results Discussion Summary References

AG1478 (see fig. 1 for its structure) has been shown to be a selective inhibitor of the EGF receptor kinase (Osherov and Levitzki, 1994; Levitzki and Gazit, 1995) and AG1296 (see fig. 1 for its structure) has been shown to be a selective inhibitor of the PDGF receptor kinase (Kovalenko et al., 1994; Levitzki and Gazit, 1995; Gazit et al., 1996). To evaluate the efficacy of these compounds in NIH 3T3 cells overexpressing either EGF receptors or PDGF receptors, cells were preincubated with increasing concentrations of either compound for 2 hr before the addition of ligands for their respective overexpressed receptors. After 5 min of cell stimulation, lysates were prepared and autophosphorylation of the receptor was measured (fig. 2). At 100 µM, each compound significantly inhibited auto-phosphorylation of its cognate receptor kinase. At 10 and 100 µM, AG1478 inhibited autophosphorylation of both PDGF and EGF receptors. However, at 1 and 0.1 µM, AG1478 significantly inhibited EGF receptor autophosphorylation without much inhibition of PDGF receptor autophosphorylation. In contrast, AG1296 significantly inhibited PDGF receptor kinase activity between 1 and 100 µM with no observed inhibition of EGF receptor kinase activity. The ability of these compounds to inhibit receptor autophosphorylation is likely to result from direct inhibition of the kinase, rather than inhibition of ligand binding to its receptor, because inhibitory activity is also observed in in vitro kinase assays in which no ligand is present (data not shown).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Chemical structures of 6,7-dimethyl-2-phenylquinoxaline (AG1296) and 4-(3-chlorophenylamino)-6,7-dimethoxyquinazoline (AG1478).

View larger version (68K): [in this window] [in a new window]   Fig. 2.   Specific inhibition of PDGF and EGF-stimulated receptor tyrosine autophosphorylation by AG1296 and AG1478. Cells overexpressing either the human PDGF or EGF receptors were pretreated with various concentrations of AG1296 or AG1478 for 2 hr before ligand stimulation. Cell lysates were prepared and subjected to SDS-PAGE followed by immunoblotting with antiphosphotyrosine antisera. An autoradiograph of the immunoblot is shown. The top panels represents lysates derived from PDGF receptor-expressing cells and the bottom panels represents lysates derived from EGF receptor-expressing cells. Arrows denote tyrosine phosphorylated forms of PDGF and EGF receptors.

The detection of tyrosine phosphorylation by immunoblotting represents an integration of all tyrosine phosphorylation events in the immunoprecipitated receptor complex. Therefore, the inhibitory effects of these compounds represent a general effect on trans-phosphorylation of the receptor complex after dimerization. However, it is possible that the compounds preferentially inhibit specific phosphorylation sites, because the activity of the kinase against any given site will reflect the accessibility of that site to the catalytic kinase core and how closely the primary amino acid sequence surrounding the tyrosine corresponds to the preferred substrate for the kinase (Songyang et al., 1995). In fact, it has recently been demonstrated that AG1296 preferentially inhibits the phosphorylation of Y857 of PDGFr (Kovalenko et al., 1997). Thus, it was unclear whether suboptimal concentrations of these inhibitors might preferentially affect specific signaling pathways of the receptors by selective inhibition of some tyrosine phosphorylation sites. To measure these effects, the relative ability of these compounds to inhibit downstream EGF and PDGF receptor activation events were compared for these compounds.

For example, several studies have suggested that activation of PI3K may be essential for receptor kinase-mediated induction of mitosis (Fantl et al., 1992; Valius and Kazlauskas, 1993; van der Geer and Hunter, 1993). To evaluate the efficacy of the kinase inhibitors against activation of PI3K, 3T3/EGFR or 3T3/PDGFR cells were pretreated with increasing concentrations of each compound, followed by stimulation with their respective ligands and cell lysis for measurement of PI3K activity (fig. 3). As described previously (Kaplan et al., 1987), the induction of PI3K activity correlated with the appearance of receptor-associated phosphotyrosine. Furthermore, each compound inhibited PI3K activation induced by its cognate receptor. That is, treatment of 3T3/PDGFR cells with 10 or 100 µM AG1296 blocked PDGF-stimulated PI3K activation, but exhibited no activity in 3T3/EGFR cells stimulated with EGF. Similarly, 0.1 to 100 µM AG1478 inhibited EGF-stimulated PI3K activity, but were inactive in PDGF-stimulated 3T3/PDGFR cells.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Selective inhibition of PDGF and EGF-induced PI3K activity. Cells overexpressing either human PDGF or EGF receptors were pretreated for 2 hr with various concentrations of AG1296 or AG1478 prior to stimulation of cells by PDGF or EGF. Cell lysates were prepared and immunoprecipitated with antiphosphotyrosine antisera. PI3K activity was assayed in the immunoprecipitated material and analyzed by thin layer chromatography for the presence of 32P-labeled phosphoinositides. Autoradiographs of the thin layer chromatographs are shown.

In addition to PI3K activation, EGF- and PDGF-receptor signaling leads to stimulation of MAPK activity (for reviews, see Claesson-Welsh 1994; Boonstra et al., 1995; Pawson 1995). This induction has been shown to result from the binding of Grb2-SOS complexes to tyrosine phosphorylated receptors or receptor-associated Shc (Pelicci et al., 1992; Lowenstein et al., 1992; Rozakis-Adcock et al., 1992; Yokote et al., 1994). In order to evaluate whether AG1296 or AG1478 might preferentially inhibit induction of MAPK activity, cells were treated as before and lysates were assayed for MAPK activities using MAP2 as an in vitro substrate for the enzyme (fig. 4A). As previously reported, we have found that both EGF and PDGF stimulation of their cognate receptors resulted in a substantial increase in MAPK activity. Pretreatment of cells with 100 µM AG1296 completely blocked stimulation of the enzyme by PDGF, and significant inhibition was also observed in the presence of 10 µM AG1296. For EGF receptor signaling, nearly complete inhibition of MAPK activation was observed at 10 and 100 µM AG1478. As was the case with PI3K, AG1296 did not inhibit EGF-induced activation of MAPK and AG1478 did not inhibit PDGF-induced activation of MAPK (fig. 4A). Neither compound inhibited the catalytic activity of MAPK when tested in vitro (fig. 4B).

View larger version (49K): [in this window] [in a new window]   Fig. 4.   Selective inhibition of PDGF and EGF-induced MAPK activity. A, Cells overexpressing either human PDGF or EGF receptors were pretreated for 2 hr with various concentrations of AG1296 or AG1478 before stimulation of cells by PDGF or EGF. Lysates were prepared and MAPK activity was assayed as described using MAP-2 as a substrate. Autoradiographs denoting phosphorylated MAP-2 protein are shown. B, Lysates were prepared from exponentially growing 3T3 cells and an in vitro MAP kinase assay was performed in the presence of various concentraions of AG1296 or AG1478. An autoradiograph denoting phosphorylated MAP-2 protein is shown.

The ability of these compounds to inhibit these two prominent short-term signaling events after receptor activation strongly suggested that such compounds would also inhibit ligand-induced cell proliferation. To begin testing this hypothesis, 3T3/EGFR cells were stimulated with either EGF or PDGF for 20 hr and then tested for BrdU incorporation (fig. 5). Approximately 7% of unstimulated cells incorporated BrdU during the 2-hr labeling period. Stimulation with either EGF or PDGF increased the number of cells in S phase to about 75%. AG1478 (0.1 µM) inhibited EGF-stimulated BrdU incorporation, and AG1296 (10 µM) was much less effective. Conversely, AG1296 inhibited PDGF-stimulated BrdU incorporation into nuclear DNA, and AG1478 was much less effective (fig. 5).

View larger version (54K): [in this window] [in a new window]   Fig. 5.   Inhibition of BrdU incorporation into cells by AG1296 or AG1478. Serum-deprived 3T3/EGFR cells (unstimulated) that had been grown on coverslips were stimulated with 2 nM EGF or 3.8 nM PDGF in the absence or presence of AG1296 (10 µM) or AG1478 (0.1 µM) for 20 hr. The cells were then exposed to BrdU for 2 hr, fixed and stained for total DNA (cells) or BrdU incorporation. Representative frames are shown.

To confirm these observations by an independent method, cell cycle analysis was performed on 3T3/EGFR cells and 3T3/PDGFR cells stimulated with EGF or PDGF, respectively (fig. 6). Pretreatment of cells with AG1296 or AG1478 reduced the number of cells in S-phase (fig. 6), which resulted from an accumulation of cells in G1 (not shown). Complete growth inhibition of 3T3/EGFR cells stimulated with EGF was observed after treatment with 1 to 100 µM AG1478, and significant inhibition of cell cycle progression occurred in the presence of 0.1 µM AG1478. For 3T3/PDGFR cells stimulated with PDGF, AG1478 exhibited a partial inhibition of cell cycle progression at 10 µM and complete inhibition at 100 µM. AG1296 inhibited PDGF-induced cell cycle progression of 3T3/PDGFR cells at 100 µM and significant inhibition at 10 µM. AG1296 had no significant effect on EGF-induced cell cycle progression of 3T3/EGFR cells at any concentration up to 100 µM (fig. 6).

View larger version (51K): [in this window] [in a new window]   Fig. 6.   Selective inhibition of cell cycle progression by AG1296 or AG1478, after ligand stimulation. Various concentrations of AG1296 or AG1478 were added to 3T3 cells overexpressing either PDGF or EGF receptors and incubated for 2 hr. The cells were then stimulated with EGF (4 nM) or PDGF (3.8 nM) and incubated for 20 hr. After incubation, cells were collected and subjected to DNA content analysis using a FACScan flow cytometer. The cell cycle phase distribution was estimated using CellFIT software (Becton-Dickinson Immunocytometry Systems). Results are expressed as the percentage of cells in S-phase ± % cv. Control samples that were not exposed to the indicated compound are indicated as +E or +P for cells stimulated with EGF or PDGF, respectively, while those unstimulated are designated as -E or -P. Statistically significant differences from control cells stimulated with ligand are indicated by: *P < .05; **P < .001; ***P < .0001.

Because these compounds arrest the cells in the G1 phase of the cell cycle, their ability to inhibit cyclin E-associated cyclin-dependent kinase (E/cdk) activity was examined (fig. 7). PDGF-induced E/cdk activity initiated about 10 hr after stimulation of 3T3/PDGFR cells, and was maximal 24 hr after cell stimulation. In contrast, EGF-induced E/cdk activity appeared 4 hr earlier, and reached maximal levels by 8 to 10 hr after cell stimulation. AG1296 effectively inhibited PDGF- but not EGF-induction of E/cdk activity. In contrast, AG1478 inhibited induction of E/cdk activity by both EGF and PDGF.

View larger version (44K): [in this window] [in a new window]   Fig. 7.   Inhibition of cyclin E-associated cyclin dependent kinase activity by AG1296 or AG1478. Cells overexpressing PDGF or EGF receptors were made quiescent, treated with AG1296 or AG1478, and then stimulated with PDGF or EGF. Cell lysates were harvested at the indicated times after ligand stimulation, and subjected to immunoprecipitation using anticyclin E antibodies. E/cdk activity in the immunoprecipitates was assayed using histone H1 as a substrate.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion Summary References

Protein phosphorylation has emerged as one of the key mechanisms for regulating signal transduction, enzyme activity, protein-protein interactions and cell proliferation in eukaryotic cells. Transmembrane signaling through many receptor systems involves activation of a protein tyrosine kinase that is either an intrinsic part of the receptor or a protein that associates with it. In general, ligand binding to the receptor induces receptor dimerization, which leads to kinase transphosphorylation and enhancement of kinase activity (for a review, see Ullrich and Schlessinger 1990). Tyrosine phosphorylation of the receptor and/or substrates then induces a cascade of protein interactions and enzyme activations through multiple pathways that lead to cellular responses such as transcriptional activation and cell proliferation (for a review, see van der Geer et al., 1994).

Effect of specific kinase inhibitors on PDGF and EGF receptor signal transduction pathways. We have used AG1296, a PDGF receptor kinase inhibitor (Kovalenko et al., 1994; Levitzki and Gazit, 1995; Gazit et al., 1996), and AG1478, an EGF receptor kinase inhibitor (Osherov and Levitzki, 1994; Levitzki and Gazit, 1995) as prototype compounds to characterize signal transduction pathways induced by ligand binding to PDGF or EGF receptors. Both PDGF and EGF receptors are known to initiate signal transduction by autophosphorylation (transphosphorylation in a receptor dimerized by ligand binding; Ullrich and Schlessinger 1990). As previously reported, AG1296 is a potent inhibitor of PDGF receptor kinase (fig. 2). Furthermore, it exhibits strong selectivity for inhibition of PDGF over EGF receptor autophosphorylation. This suggests that it should be a good tool for examining signal transduction pathways mediated by ligand binding to PDGF receptors. AG1478 is very potent at inhibiting EGF receptor autophosphorylation (fig. 2), but appears to be less selective than AG1296 in that it also inhibits PDGF receptor autophosphorylation at concentrations of more than 1 µM.

Relative potency for inhibition of receptor kinase autophosphorylation and downstream activities. Approximately 10- to 100-fold more inhibitor was needed for complete inhibition of ligand-induced PI3K or MAPK activation or cell proliferation than for inhibition of receptor autophosphorylation. This is likely to result from the fact that measurement of the receptor associated tyrosine phosphorylation represents an integration over many possible tyrosine phosphorylation sites (5-15, depending on the receptor tyrosine kinase), while activation of PI3K or MAPK is usually dependent on only one or two sites. Unless an inhibitor has the ability to specifically prevent phosphorylation of the receptor tyrosines that mediate activation of downstream signaling, it would be necessary to use a concentration of compound high enough to consistently block all receptor kinase activity. Therefore, it is not surprising that there is a requirement for a higher concentration of compound to inhibit activation of downstream signaling events.

Selectivity of receptor kinase autophosphorylation and downstream activities. Although AG1478 is a selective EGF receptor kinase inhibitor, it inhibited PDGF receptor phosphorylation at 10 and 100 µM (fig. 2), prevented PDGF-induced cell proliferation at 100 µM (fig. 6) and suppressed PDGF-induction of E/cdk activity (fig. 7). However, AG1478 was unable to inhibit PDGF-induced MAPK or PI3K activation at concentrations up to 100 µM. These data suggest that AG1478 may inhibit another enzymatic activity that is capable of mediating PDGF-induced cell proliferation without activation of PI3K or MAPK. A candidate for such an enzyme may be found in the sphingosine-1-phosphate signaling pathway, which appears to be critical for PDGF-induced but not EGF-induced mitogenesis (Rani et al., 1997).

View larger version (53K): [in this window] [in a new window]   Fig. 8.   Comparison of the concentrations of AG1296 or AG1478 needed to observe selective inhibition of different signaling events. The minimal concentration at which AG1296 completely inhibits PDGF-, but not EGF-induced signaling events and at which AG1478 completely inhibits EGF- but not PDGF-induced signaling events are shown. The signaling events are: RTK, receptor autophosphorylation; MAPK, activation of MAPK; PI3K, activation of PI3K; G1/S, inhibition of cell cycle progression from G1 to S phase.

Cyclin-dependent kinase activation after EGF and PDGF receptor activation. Cell cycle progression is currently thought to be regulated by the progressive activation and deactivation of cyclin-dependent kinases (see Sherr, 1993 for a review). The activity of cdks is regulated by their phosphorylation state and by the cyclical synthesis and degradation of cyclins and cdk inhibitors (see Morgan, 1995 for a review). E/cdk activity (the kinase activity of the cyclin E/cdk2 complex) increases during the latter half of the G₁ phase of the cell cycle and is essential for entry into S phase (Ohtsubo et al., 1995).

Receptor overexpression. In this report, we have used cell lines overexpressing EGF or PDGF receptors. This, of course, has advantages and disadvantages. It is easy to detect receptor phosphorylation with large numbers of receptors, and cells can be induced to proliferate with only one growth factor. This is in contrast to normal murine fibroblastoid cells which generally require two growth factors to initiate proliferation of serum-deprived and contact-inhibited cells (Stiles et al., 1979). However, inhibition of ligand-induced cell proliferation is more difficult to achieve, because it is necessary to inhibit signaling through the vast majority of receptors. In general, the IC₅₀ values for compound inhibition of ligand-induced BrdU incorporation is several-fold higher in cells with overexpressed receptors than in cells with endogenous levels of receptor. For example, in an ELISA-format BrdU incorporation assay, the IC₅₀ value for AG1296 inhibition of PDGF-induced BrdU incorporation was 2.4 µM in 3T3 cells with endogenous levels of PDGF receptors, but 7.5 µM in 3T3 cells overexpressing PDGF receptors. In examining 10 compounds that inhibit PDGF-induced BrdU incorporation, the IC₅₀ values in cells overexpressing PDGF receptors was an average of 3.5-fold more than in cells with endogenous levels of PDGF receptors. Thus, the apparent potency observed in the data is somewhat lower than would be observed for inhibition of EGF or PDGF receptors in cells expressing fewer receptors.

	Summary

Top Abstract Introduction Materials & Methods Results Discussion Summary References

In summary, we have demonstrated that potent and selective receptor tyrosine kinase inhibitors are useful tools that can be used to characterize ligand-inducible signal transduction events. However, it is critical to optimize the concentration of inhibitor used for two reasons: 1) different signaling events are differentially sensitive to inhibition and 2) selectivity tends to be relative, and may be lost at high concentrations of inhibitor.