Extended Data Fig. 9: Combined inhibition of AXL shedding and AXL activity synergistically reduced cancer cell proliferation, migration, and tumour growth.

a, Inhibition of AXL shedding by the broad-spectrum MMP inhibitor BB-94. β-actin was run on the same gel as loading control. The bar graph is statistical analysis of sAXL level in PANC1 CM. b, Inhibition of AXL shedding in primary KPCP tumour cells by the broad-spectrum MMP inhibitor BB-94. Cells were maintained in FBS-free medium containing 10 μm BB-94 for 24 h, and secretome in CM was profiled by LC-MS/MS analysis. The LFQ intensity of AXL was extracted for comparison between control and BB-94 treatment. c, Statistic analysis of the WB results in Fig. 6e, showing relative band intensities of p-AKT after normalization to total AKT. d, PRM analysis of two AXL pTyr peptides after enrichment of phosphopeptides. The treatment of PANC1 cells was the same as in Fig. 6e. The left panel of each pTyr peptide is representative transition peaks. The dark arrow indicates the retention time and mass error for the highest transition peak. e, Schematic workflow for testing synergistic effect between BB-94 and R428 on PDOs. f, Histological characterization of PDAC tumour tissues and corresponding PDOs and PDOs-derived xenografts. PDOs and PDOs-derived xenografts formed typical glandular tubular structures similar to the corresponding patient tumours, and had similar expression levels of the epithelial cell marker KRT19 and SOX9, and cell proliferation marker Ki67. g,h, Solo or combined drug treatments on xenograft tumours generated by one case of PDO (DAC-71) with significant synergistic response (g) between BB-94 and R428 (statistics are presented in Fig. 6g). i–k, Validation of synergistic effect on xenograft models generated by one case of PDO (DAC-18) with no synergistic response (i). Data are mean + s.d. of n = 8 xenograft tumours per group (j,k). l, Orthotopic tumours of PDAC under different drug treatment. Orthotopic model was constructed by injecting KPCP primary cancer cells into pancreata of NSG mice. After tumour formation, mice were daily administrated with indicated drugs for 14 days. The dark red tissues were spleens, which were removed before measuring tumour weights. Tumour weights are presented as mean ± s.d. of n = 9 mice per condition. m, Correlation of MMP1 and MMP11 mRNA expression levels in 66 pancreatic PDO lines with drug combination response of BB-94 and R428. Box plots are as defined in Fig. 2c. n,o, Colony formation assay of MIA PaCa2 cells with stable knockdown of MMP1 or MMP11 by shRNAs. Cells were treated with DMSO as control or 0.5 μM R428 for testing of inhibition efficiency (n). Knockdown efficiency was measured by real-time qRT–PCR (o). p, Transwell migration assays of MIA PaCa2 cells under shedding inhibition and/or AXL inhibition. Shown are representative crystal violet-stained images of transwell migration assays of MIA PaCa2 cells under different treatments. Quantification of cell migration determined by counting the number of migratory cells per field and relative numbers of migratory cells were calculated by normalizing to control cells. q, Expression of EMT markers in cancer cells upon inhibition of shedding by BB-94. Due to similar molecular weight of proteins, samples were run on separate gels, with β-actin as sample processing control. All the bar graphs are mean ± s.d. of n = 3 biological replicates (the data points in p are relative cell numbers from 2 random views of each biological replicates). P values are from two-tailed unpaired Student’s t-test (b,d,k,l,m,n,o,p) or one-tailed paired Student’s t-test (a,c,q).

Source Data