Fig. 2

Costunolide is a dual inhibitor of MEK1 and AKT1/2 that overcomes osimertinib resistance. A. Model of costunolide binding with MEK1. Left: Predicted binding between costunolide and MEK1. Right: Ligand Interaction Diagram (LID) of the binding. MEK1 structure is shown as a ribbon representation and costunolide is shown as a stick. B. Inhibitory effect of costunolide on MEK1. 100 ng active MEK1 kinase was pre-incubated with various concentrations of costunolide at RT for 15 min. Next, 200 ng inactive ERK2 and ATP buffer were added and the mixture was incubated at 30 ℃ for 30 min. Phosphorylated and total ERK were detected by Western blot. C. The same method was used to confirm the inhibition effect of costunolide on AKT1 and AKT2. Inactive GSK3β was used as the substrate. D. Binding between costunolide and MEK1.0.5 mg cell lysates from H1975, PC-9, and HCC827 were incubated with Sepharose-4B or costunolide-conjugated Sepharose-4B. The pulled down proteins were detected by Western blotting. E. The same method was used to analyze the binding between costunilode with AKT (D). F. Influence of costunolide on MEK and its downstream effectors. PC-9, HCC827, and H1975 cells were treated with costunolide at 0, 5, 10 and 20 µM concentrations for 6 h. Next, the cell lysates were loaded to detect phosphorylation of MEK1/2, ERK, and RSK2. G. HCC827 and H1975 cell lysates were used to determine phosphorylation of AKT and downstream GSK3β and NFκB. H. Costunolide enhanced the inhibitory effect of osimertinib in PC9-Osi, HCC827-Osi and H1975-Osi cells. Cells were treated with the indicated concentrations of costunolide and osimertinib or in combination for 48 h. Next, cell viability was measured using MTT assays. I. Alteration of MEK or AKT signaling. PC9-Osi cells were treated with 1 µM osimertinib, 10 µM costunolide or their combination. Cells were harvested 12 h later and lysed to detect the phosphorylation levels of MEK1, ERK and RSK2. HCC827-Osi cells were treated by costunolide, osimertinib or their combination. After treatment for 12 h, cells were harvested to determine phosphorylation of AKT and GSK3β. J. Effect of costunolide (20 mg/kg), osimertinib (10 mg/kg) and their combination on tumor growth. The osimertinib resistance PDX model, HLG57-OR, was inoculated into 4 groups. Drugs were orally administrated every day starting 1 week after transplantation. Tumor volume (length × width× height × 0.52) was measured twice per week. (n = 5) K. Image of tumor excised from PDX models. After mice were sacrificed, tumors were excised exfoliated from subcutaneous tissue. L. Protein expression level of MEK and AKT, as well as their downstream signaling effectors, in tumor tissues. A portion of each tumor tissue sample was ultrasonicated. 3 samples from each group were loaded to check labeled signaling proteins by Western blotting. M. Representative images of ki67, p-MEK1, p-ERK, p-AKT, p-GSK3β expression in tumor tissues. Tissue slides were stained with antibodies and analyzed using IHC. N. Expression of ki67, p-AKT1/2, p-GSK3β, p-MEK1, p-ERK in each group. IHC images in each group were analyzed with Image J and total IOD from each image was recorded to indicate the protein expression level. (n = 5) O. Diagram illustrating the functional mechanism of costunolide. Quantitative analysis of western blotting bands was calculated by Image J software in (B, C, F, G, I and L). Data was presented as mean ± SD with 3 independent experiments in (H). One-way ANOVA with a multiple comparisons test was applied in (H and N), unpaired t-test was used in (J). ns P > 0.05, **P < 0.01, ***P < 0.001