AXL degradation in combination with EGFR-TKI can delay and overcome acquired resistance in human non-small cell lung cancer cells

Abstract

Acquired resistance to epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) has been a major obstacle in the treatment of non-small cell lung cancer (NSCLC) patients. AXL has been reported to mediate EGFR-TKIs. Recently, third generation EGFR-TKI osimertinib has been approved and yet its acquired resistance mechanism is not clearly understood. We found that AXL is involved in both gefitinib and osimertinib resistance using in vitro and in vivo model. In addition, AXL overexpression was correlated with extended protein degradation rate. We demonstrate targeting AXL degradation is an alternative route to restore EGFR-TKIs sensitivity. We confirmed that the combination effect of YD, an AXL degrader, and EGFR-TKIs can delay or overcome EGFR-TKIs-driven resistance in EGFR-mutant NSCLC cells, xenograft tumors, and patient-derived xenograft (PDX) models. Therefore, combination of EGFR-TKI and AXL degrader is a potentially effective treatment strategy for overcoming and delaying acquired resistance in NSCLC.

Fig. 1. Overexpression of AXL is observed in EGFR-TKI resistant NSCLC cells.

NSCLC cells (PC9, PC9-gef, HCC827, HCC827-gef, and HCC827-osi) were treated with gefitinib or osimertinib for 72 h. Cell viability was measured by MTT assay as described in Methods (a). IC₅₀ values and subculture conditions were stated (b). The protein expressions of major RTKs, p-AKT, and p-ERK were determined by Western blotting (c). Cells were incubated on the confocal dish for 24 h, fixed, and stained with C-term AXL (red) and DAPI (blue) for visualization of cellular localization (d). Data are presented as the mean fold changes ± SD of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005 by t-test

Fig. 2. The total AXL protein degradation rate is extended in EGFR-TKI resistant cells.

The efficiency of siAXL was confirmed using real-time PCR and Western blotting (a). The effects of gefitinib or osimertinib on the cell viability were evaluated after cells were transfected with siNC or siAXL compared with control (no transfection) in HCC827-gef cells (b) and HCC827-osi cells (c). Cycloheximide (CHX) was treated to stop the protein production and the protein expression of AXL (140 kda) was detected by Western blotting over the indicated time points. The expressions of AXL and β-actin were quantified by densitometry using ImageJ. AXL expressions were normalized to β-actin and further compared with 0 h expressions (d). mRNA expression of AXL and GAS6 were detected by real-time PCR using β-actin as a reference gene (e). Data are presented as the mean fold changes ± SD of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005 by t-test

Fig. 3. YD can regulate the degradation of AXL and effectively inhibit the growth of both EGFR-TKI sensitive and resistant cells.

Chemical structure of yuanhuadine (YD) (a). Indicated cell lines were treated with YD for 72 h andviability was measured. (b). The protein expression of AXL (140 kda) was detected after co-treatmentof CHX and YD by Western blotting analysis over the indicated time points to test the effects of YD onthe protein degradation. The expressions of AXL and β-actin were quantified by densitometry usingImageJ. AXL expressions were normalized to β-actin and further compared with 0 h expression (c). Cells were treated with indicated concentrations of YD for 24 h (d). Data are presented as the mean foldchanges ± SD of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005 by t-test

Fig. 4. YD and EGFR-TKIs synergistically inhibit the growth of gefitinib and osimertinib cross-resistant cells in in vitro and in vivo.

Cell viability was measured after combined treatment of YD and gefitinib or osimertinib for 72 h in HCC827-gef (a) and HCC827-osi (b), respectively. Based on the cell viability results, CI values were calculated to demonstrate the combination effects in each cell line. The effects of combination treatment compared with single treatment were accessed using western blot. Cells were treated with YD (10 nM) and/or gefitinib (1 μM) in HCC827-gef cells, and YD (10 nM) and/or osimertinib (500 nM) in HCC827-osi cells (c). HCC827-gef cells were subcutaneously implanted into the flanks of Balb/c-nude mice (n = 5). Mice were orally administered with compounds 6 times per week for 21 days and doses indicated are 1 mg/kg for YD, 10 mg/kg for gefitinib. HCC827-osi cells were subcutaneously implanted into the flanks of Balb/c-nude mice (n = 5). Mice were orally administered with compounds 6 times per week for 21 days and doses indicated are 0.5 mg/kg for YD, 5 mg/kg for osimertinib (d). Data are presented as the mean fold changes ± SD of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005 by t-test

Fig. 5. YD and EGFR-TKIs synergistically inhibit the growth in acquired gefitinib-resistant PDX models.

When tumor volumes reached 200 to 250 mm3, mice engrafted with YHIM-1009 tumors were segregated into appropriate treatment groups (three to five mice per group). Mice were orally administered with compounds every day for 31 days and doses indicated are 1 mg/kg for YD, 10 mg/kg for gefitinib (a). *P < 0.05, **P < 0.01, ***P < 0.005 by t-test

Fig. 6. The combined administration of YD and gefitinib can delay the emergence of resistance and inhibit AXL overexpression in a long-term in vivo xenograft model.

HCC827-Luc cells were subcutaneously implanted into the flanks of Balb/c-nude mice (five mice per group). Drugs were orally administered 6 times per week for 90 days and doses indicated are 0.5 mg/kg for YD, 10 mg/kg for gefitinib (a). The graph represents tumor volumes of indicated days which were normalized to the initial tumor volume for comparison (b). Bioluminescence images of mice at the final day before sacrifice were measured and mice were injected with Firefly D-luciferin before imaging (c). Immunohistochemical analysis of AXL was performed in tumor tissue (d). The schematic diagram illustrating the significance of combined YD with EGFR-TKI treatment in NSCLC (e). *P < 0.05, **P < 0.01, ***P < 0.005 by t-test