Fig 1.
Dimerization arm targeting strategy for inhibition of EGFR.
(a) EGF-induced activation of the extracellular receptor of EGFR. In the inactive state, the dimerization arm (dark blue) of the extracellular receptor is buried within domain IV (beige). In the active state, the receptor undergoes a conformational change to promote intermolecular interactions of the dimerization arm for receptor dimerization. Other features of the receptor include domain I (green), domain II (blue), domain III (brown) and EGF (dark grey). (b) The dimerization arm (dark blue) makes extensive contacts with domain II of the receptor binding partner (beige). (c) Triazolyl-bridged peptides were designed to mimic the dimerization arm, thereby blocking receptor dimerization and activation of the kinase (violet) through occlusion of the dimerization arm binding pocket. Additional features include the transmembrane domain (grey dashed lines). Structures were rendered using PyMol (PDB files: 1NQL, 3NJP, and 2GS6).
Fig 2.
Design and synthesis of EDA peptides.
(a) Peptide sequences were derived from the dimerization arm sequence of EGFR. The overall linker length and positioning of the azide and alkyne amino acids were varied. Non-natural amino acids are show in red and blue. (b) Dimerization arm mimics were synthesized by incorporating non-natural amino acids into the peptide sequence using solid phase peptide synthesis (SPPS). Peptides were cyclized on solid support via copper (I)-catalyzed azide-alkyne cycloaddition prior to resin cleavage. (c) Non-natural amino acids used for the triazole cross-link: N-Fmoc-L-propargylglycine (Pg), N-Fmoc-4-azido-L-homoalanine (Aha), N-Fmoc-5-azido-L-norvaline (Anv), N-Fmoc-6-azido-L-norleucine (Anl).
Fig 3.
Molecular dynamics simulations of EDA peptides.
(a) Molecular dynamics simulations were performed and the g_hbond program in Gromacs suite was used to determine the stability of the hydrogen bond characteristic of the β-loop conformation. A query of the number of frames in which the hydrogen bond between Asn271 and Tyr275 is present predicts that this hydrogen bond is largely maintained in EDA2 and EDA4. Data is plotted as the percent of molecular dynamics trajectory frames in which the hydrogen bond between Asn271 and Tyr275 is present. (b) Molecular dynamics simulations were performed to predict the overall structure of the EDA peptides and the cluster centers for EDA1–6 are shown. Red arrows indicate linkers that fold over the non-binding surface of the peptide. The black dashed line indicates the hydrogen bond between Asn271 and Tyr275. Widths between the β-sheets were measured between the Cα carbon of residues Tyr270 and Nle277 (red dashed line).
Fig 4.
EDA peptides are resistant to proteolytic degradation.
Proteolytic stability was measured in the presence of (a) a cocktail of immobilized chymotrypsin and trypsin over a time range of 0–4 hours and (b) 50% mouse serum over a time range of 0–16 hours. (c, d) CD spectra of the non-modified and EDA2 peptides were obtained on a Jasco J-710 CD Spectrometer at 25°C in 10 mM sodium phosphate buffer at pH 6.5 and 7.4. EDA2 maintains its structure under both conditions.
Fig 5.
EDA2 down-regulates activated EGFR and reduces cell viability.
(a) Cells were stimulated with 50 ng/mL EGF for 5 min in the presence or absence of EDA peptides, followed by western blotting. An apparent decrease in phosphorylated EGFR was observed when treated with EDA2. Vertical lines indicate non-adjacent samples from the same western blot. (b) Quantification of EGFR phosphorylated at Tyr1068, normalized to tubulin, shows that EDA2 reduces phosphorylated EGFR by greater than 60%, while the non-modified and scrambled controls do not. Data is plotted as the average of at least three experiments, where error bars represent SEM. * p < 0.05, ** p < 0.01 relative to the EGF-stimulated control. All remaining means are not significant (p > 0.05) relative to the EGF-stimulated control. (c) A panel of cell lines was dosed daily with EDA2 or gefitinib for 5 days, and viability was quantified using the Cell Titer Blue assay. EDA2 reduced viability by 50%, 72%, and 34% in MDA-MB-231, PC-3, and Bx-PC-3 cells, respectively. Data is plotted as the average of quadruplicates, where error bars represent SEM.
Fig 6.
EDA2 down-regulates EGFR dimers.
(a) MDA-MB-231 cells were stimulated with 10 ng/mL EGF for 5 min in the presence or absence of 5 μM EDA2 and the scrambled control peptide (EDA2-Scr). The dimer species of EGFR was detected using a fluorescent Duolink assay and is shown as an overlay of the PLA signal (orange) and DAPI (blue). When dimerized, one EGFR molecule may bind the plus probe while the other may bind the minus probe. The close proximity of the plus and minus PLA probes then allows for ligation and amplification, which can be detected as a punctate fluorescent signal. Images were obtained with a 40x objective and scale bars represent 25 μm scale. (d) The dimer signals of individual cells were measured for each condition tested (n = 500 cells per condition). Data is plotted as the average signal count per cell, where error bars represent SEM. **** p < 0.0001 relative to the stimulated control, #### p < 0.0001 relative to the scrambled control. Remaining comparisons do not differ significantly (p > 0.05).