Oncogenic transformation by inhibitor-sensitive and -resistant EGFR mutants

Abstract

Background: Somatic mutations in the kinase domain of the epidermal growth factor receptor tyrosine kinase gene EGFR are common in lung adenocarcinoma. The presence of mutations correlates with tumor sensitivity to the EGFR inhibitors erlotinib and gefitinib, but the transforming potential of specific mutations and their relationship to drug sensitivity have not been described.

Methods and findings: Here, we demonstrate that EGFR active site mutants are oncogenic. Mutant EGFR can transform both fibroblasts and lung epithelial cells in the absence of exogenous epidermal growth factor, as evidenced by anchorage-independent growth, focus formation, and tumor formation in immunocompromised mice. Transformation is associated with constitutive autophosphorylation of EGFR, Shc phosphorylation, and STAT pathway activation. Whereas transformation by most EGFR mutants confers on cells sensitivity to erlotinib and gefitinib, transformation by an exon 20 insertion makes cells resistant to these inhibitors but more sensitive to the irreversible inhibitor CL-387,785.

Conclusion: Oncogenic transformation of cells by different EGFR mutants causes differential sensitivity to gefitinib and erlotinib. Treatment of lung cancers harboring EGFR exon 20 insertions may therefore require the development of alternative kinase inhibition strategies.

Figure 1. Mammalian Cells Expressing the Lung Cancer-Derived Mutant EGFR Grow in an Anchorage-Independent Manner

(A) NIH-3T3 cells infected with retroviruses encoding the indicated wild-type or mutant EGFR were selected in the presence of 2.5 μg /ml puromycin for 4 d. In the top photomicrographs, 1 × 10⁵ cells were suspended in soft agar for a colony formation assay and photographed after 3 wk incubation at 37 °C. Expression of lung cancer-derived missense EGFR mutants, but not wild-type or kinase-inactive D837A EGFR, induces colony formation in soft agar. In the bottom photomicrographs, samples were identical, but 20 ng/ml EGF was added to the top agar. Representative photomicrographs are shown. (B) Anti-EGFR immunoblot analysis of pooled stable NIH-3T3 cells infected as described in (A). All EGFR constructs are expressed at similar levels. pBp, pBabe-Puro vector; wt, wild-type EGFR. (C) Lysates from 4 × 10⁴ cells from the human lung adenocarcinoma cell line H3255, harboring the L858R mutation in EGFR, or the wild-type or L858R EGFR-overexpressing NIH-3T3 cells, were immunoblotted for total EGFR levels. Although total protein levels per cell are lower for the H3255 than the NIH-3T3 cells, EGFR expression levels are slightly higher in the H3255s. (D) NIH-3T3 cells infected with retroviruses encoding the mutant EGFR were selected in the presence of 2 μg/ml puromycin for 9 d. Selected cells (1 × 10⁵) were suspended in soft agar for a colony formation assay and photographed after 3 wk incubation at 37 °C. Expression of the deletion and insertion EGFR mutants induced formation of colonies in soft agar with higher efficiency than expression of L858R. Representative photos are shown. Polyoma mT, NIH-3T3 cells infected with positive control pBabe-Puro retrovirus encoding the polyoma middle T antigen. (E) hTBE cells expressing the SV40 early region and hTERT were infected with control virus pBabe-Puro (pBp) or with viruses encoding the indicated EGFR alleles. Cells were plated in 0.4% Noble agar, and colonies were counted with an automated imager at 6 wk. Mean ± standard deviation is shown for three independent determinations. Control cells (pBp) formed many microscopic colonies, but colonies formed by cells expressing EGFR mutants were more numerous and larger. del, L747_E749del A750P mutated EGFR; ins, D770_N771insNPG mutated EGFR; pBp, pBabe-Puro vector; RasV12, V12 H-Ras; wt, wild-type EGFR. (F) Anti-EGFR immunoblot analysis of hTBE cells infected as described in (E). All EGFR constructs were expressed at similar levels. pBp, pBabe-Puro vector; wt, wild-type EGFR.

Figure 2. Ligand-Independent Activation of the Mutant EGFR

(A) Cells expressing the wild-type or mutant EGFR were lysed and immunoblotted with antibodies to total EGFR or antibodies that recognize specific phosphorylation sites in the EGFR C-terminal tail as labeled. All four mutant EGFR proteins, representative of the four classes of EGFR mutations observed in lung adenocarcinoma tumor DNA, exhibited constitutive phosphorylation on the indicated C-terminal autophosphorylation sites. Note that the nomenclature for the anti-phospho-EGFR antibodies reflects elimination of the 24-amino acid signal peptide. Due to difficulties in isolating clonal cell lines with the same levels of mutant EGFR expression, G719S is expressed at higher levels and D770_N771ins NPG at lower levels than the other mutant EGFR. del, L747_E749del A750P; ins, D770_N771insNPG; pBp, pBabe-Puro vector control; wt, wild-type EGFR. (B) Cells expressing the wild-type or L858R EGFR were placed in media containing 0.5% CS for 24 h. A combination of three neutralizing antibodies (anti-EGF, anti-TGFα, and anti-EGFR) was added 3 h prior to EGF stimulation and lysis. Upper row of blots show the anti-phospho-EGFR Y1068 immunoblots. The lower row shows anti-EGFR immunoblots. No inhibition of L858R EGFR autophosphorylation was observed upon treatment with a combination of three neutralizing antibodies (“neutr Ab”) sufficient to prevent EGF stimulation of autophosphorylation of the wild-type EGFR.

Figure 3. Shc and STAT3 Signaling Pathways Are Constitutively Activated in Cells Expressing the Mutant EGFR

(A) Cells expressing the wild-type or L858R mutant EGFR were placed in 0.5% calf serum for 24 h and left unstimulated or stimulated with 20 ng/ml EGF (“+EGF”) for 8 min. EGFR was immunoprecipitated from 200 μg of cell lysate, and eluted immune complexes were separated by SDS-PAGE and immunoblotted with anti-Shc. Shc constitutively coimmunoprecipitates with the L858R EGFR but not the wild-type EGFR. IP, immunoprecipitation; NL, no-lysate control immunoprecipitation. (B) Anti-phospho-Shc immunoblots (upper row of blots) and anti-Shc immunoblots (lower row) of whole cell lysates from the experiment in (A). All three Shc isoforms are constitutively phosphorylated in cells expressing the L858R EGFR. (C) Immunoblots of whole cell lysates with anti-phospho-STAT3 Y705 (upper row), total STAT3 (middle row), or actin as a loading control (lower row). STAT3 is constitutively phosphorylated in cells expressing any of the four mutant EGFR proteins, representative of the four classes of EGFR mutations observed in lung adenocarcinoma tumor DNA. del, L747_E749del A750P; ins, D770_N771insNPG; pBp, pBabe-Puro vector control; wt, wild-type EGFR. (D) M67 STAT3 luciferase reporter assay. NIH-3T3 cells expressing the indicated EGFR were transfected with a STAT-dependent reporter (m67-firefly luciferase) [25] and a control reporter expressing Renilla luciferase. STAT-dependent luciferase production was measured after 48 h and normalized to Renilla luciferase. The normalized luciferase values were divided by the values for cells expressing the wild-type EGFR to produce relative luciferase units. NIH-3T3 expressing mutant forms of EGFR exhibited elevated levels of STAT-dependent transcriptional activity relative to wild-type. ins, D770_N771insNPG EGFR; wt, wild-type EGFR. (E) Immunoblots of whole cell lysates with anti-phospho-Akt S473 (upper row), total Akt (middle row), or actin as a loading control (lower row). Akt is constitutively phosphorylated in cells expressing mutant EGFR. del, L747_E749del A750P; ins, D770_N771insNPG; wt, wild-type EGFR.

Figure 4. Sensitivity of Cell Transformation Induced by Expression of Mutant EGFR Characterized by Missense Mutation or Exon 19 Deletion, but not Exon 20 Insertion, to Gefitinib and Erlotinib

(A) Anchorage-independent growth of clonal NIH-3T3 cells transformed with mutant EGFR or EGF-stimulated wild-type EGFR treated with the indicated concentrations of erlotinib immediately prior to suspension in soft agar. Transformation induced by expression of L858R, G719S, and L747_E749del A750P EGFR, but not EGF-stimulated wild-type EGFR or D770_N771insNPG EGFR, was inhibited by 0.1 μM erlotinib. Representative photographs are shown. (B) Number of colonies formed in soft agar by clonal NIH-3T3 cells expressing L858R EGFR and D770_N771insNPG EGFR treated with the indicated concentrations of gefitinib or erlotinib immediately prior to suspension in soft agar. Transformation by cells expressing the L858R EGFR was inhibited by 0.1 μM gefitinib or erlotinib, whereas transformation by cells expressing the insertion mutant was resistant to low concentrations of these inhibitors. Colonies were quantitated by counting ten fields each of triplicate wells photographed with a 10× objective; mean ± standard deviation is shown. Ins, D770_N771insNPG EGFR. (C) Transformation induced by expression of D770_N771insNPG EGFR is inhibited 10-fold more efficiently by the irreversible EGFR inhibitor CL-387,785 [35]. Clonal NIH-3T3 cells expressing the insertion mutant were treated with the indicated concentrations of gefitinib, erlotinib, or CL-387,785 immediately prior to suspension in soft agar. This assay was not done in triplicate, but the results are representative of two independent experiments. The number of colonies was normalized to maximum colony formation for each treatment, and sigmoidal dose response curves were fitted to the data using Prism Graphpad software to determine IC50s.

Figure 5. Sensitivity of Mutant EGFR Autophosphorylation to EGFR Inhibitors Reflects Inhibition of Anchorage-Independent Growth

Cells expressing wild-type, L858R, or D770_N771insNPG EGFR were treated for 2 h with the indicated concentrations of gefitinib or CL-387,785. Cells expressing the wild-type EGFR were then stimulated for 10 min with 7 ng/ml EGF, and all plates were lysed. Whole-cell lysates were immunoblotted for phospho-EGFR Y1068 (upper row of blots), total EGFR (middle row), and actin as a loading control (lower row). Although compound concentrations necessary for inhibition of autophosphorylation do not exactly correspond to inhibition of anchorage-independent growth, the relative sensitivity of autophosphorylation of the wild-type and mutant EGFR to gefitinib or CL-387,785 mirrors the relative sensitivity of colony formation to these inhibitors.