Genetic Ablation of the MET Oncogene Defines a Crucial Role of the HGF/MET Axis in Cell-Autonomous Functions Driving Tumor Dissemination

Abstract

Cancer cell dissemination is sustained by cell-autonomous and non-cell-autonomous functions. To disentangle the role of HGF (Hepatocyte Growth Factor) and MET ligand/receptor axis in this complex process, we genetically knocked out the MET gene in cancer cells in which MET is not the oncogenic driver. In this way, we evaluated the contribution of the HGF/MET axis to cancer cell dissemination independently of its direct activities in cells of the tumor microenvironment. The lack of MET expression in MET^-/- cells has been proved by molecular characterization. From a functional point of view, HGF stimulation of MET^-/- cancer cells was ineffective in eliciting intracellular signaling and in sustaining biological functions predictive of malignancy in vitro (i.e., anchorage-independent growth, invasion, and survival in the absence of matrix adhesion). Cancer cell dissemination was assessed in vivo, evaluating: (i) the ability of MET^-/- lung carcinoma cells to colonize the lungs following intravenous injection and (ii) the spontaneous dissemination to distant organs of MET^-/- pancreatic carcinoma cells upon orthotopic injection. In both experimental models, MET ablation affects the time of onset, the number, and the size of metastatic lesions. These results define a crucial contribution of the HGF/MET axis to cell-autonomous functions driving the metastatic process.

Keywords: HGF; MET; invasion; metastasis; pancreatic cancers.

Figure 1

Generation and validation of a MET^−/− NSCLC cell line. (a) Flow cytometry analysis of MET expression at the surface of wild-type (wt) and MET^−/− A549 cells. (b) Immunoblotting analysis of MET expression in lysates from wild-type (wt) and MET^−/− A549 cells. Actin was used as a loading control. (c) HGF-dependent phosphorylation of MET and activation of downstream transducers in lysates from wild-type (wt) and MET^−/− A549 cells. Vinculin was used as a loading control. p145 MET: MET receptor β chain; p145 P-MET: phosphorylated MET receptor β chain; p190 MET: precursor form of the MET receptor; p60AKT: AKT; p60 P-AKT: phosphorylated AKT; p44/42ERK: ERK; p44/42 P-ERK: phosphorylated ERK; p40 actin: Actin; p117 Vinculin: vinculin. Data reported in the figure are representative of at least two experiments. The uncropped Western blots are shown in Supplementary Materials File S1.

Figure 2

In vitro analysis of growth properties of wild-type and MET^−/− A549 cells. (a) Analysis of viability by quantitation of metabolically active wild-type (wt) or MET^−/− A549 cells treated or not with HGF (25 ng/mL) after two or four days of culture. Graph reports the fold increase in cell number with respect to day one. Each point is the mean of triplicate values; bars represent SD. (b) Analysis of cell proliferation by EdU incorporation in wild-type (wt) or MET^−/− A549 cells treated or not with HGF (25 ng/mL) after 48 h of culture in serum-free medium. Graph reports the percentage of EdU-positive cells. Each point is the mean of duplicate values; bars represent SD. (c) Cell cycle analysis of wild-type (wt) or MET^−/− A549 cells treated or not with HGF (25 ng/mL) after 48 h of culture in 2% serum medium. Graph reports the percentage of cells in each cell cycle phase. Each point is the mean of triplicate values; bars represent SD. (d) The clonogenic ability of wild-type (wt) or MET^−/− A549 cells after 10 days of culture in medium with 10% serum. Graph reports the percentage of plating efficiency. Each point is the mean of 10 values; bars represent SD. ns: not significant.

Figure 3

In vitro analysis of HGF-driven biological responses in wild-type and MET^−/− A549 cells. (a) Analysis of anchorage-independent growth by soft agar assay with wild-type (wt) or MET^−/− A549 cells. Graph reports the number of colonies obtained after 14 days of culture in the presence or in the absence of HGF (25 ng/mL). Each point is the mean of quadruplicate values. Each quadruplicate value is obtained by calculating the mean of the colony number present in at least three different fields of the same sample; bars represent SD. (b) Analysis of cell invasion by transwell assay with wild-type (wt) or MET^−/− A549 cells. Graph reports the fold increase in cells migrated through a Matrigel layer during 24 h of culture in the presence of HGF (25 ng/mL), with respect to the untreated counterpart. Each point is the mean of triplicates. (c) Analysis of apoptosis induced by loss of cell/matrix interaction (Anoikis) in wild-type (wt) or MET^−/− A549 cells. Graph reports the percentage of annexin-positive cells (early and late stages of apoptosis, i.e., Annexin V positive/7AAD negative plus Annexin V positive/7AAD positive cells), treated or not with HGF (25 ng/mL) after 24 h of culture in the absence of matrix adhesion. Each point is the mean of triplicate values; bars represent SD. ***, p ≤ 0.001; **, p ≤ 0.01; *, p ≤ 0.05; ns: not significant. Data reported in the figure are representative of at least two experiments.

Figure 4

Lung colonization assay with MET^−/− A549 cells. Luciferase-expressing wild-type and MET^−/− A549 cells were injected into the tail vein of hHGF-KI mice. (a) IVIS analysis of mice performed 4 h post-injection (day zero) and then after 4-8-16 days. Each time point represents the mean value of the group. Bars represent SEM. (b) IVIS analysis of lungs excised from mice at day 36. Each dot represents the value of the lungs excised from one mouse. Black and red lines: average value for each group. Bars represent SEM. The blue line indicates the threshold (10⁴) below which IVIS values are considered negative. *, p ≤ 0.05; **, p ≤ 0.01. The data reported in the figure are representative of two experiments.

Figure 5

Generation and validation of a MET^−/− pancreatic carcinoma cell line. (a) Flow cytometry analysis of MET expression at the surface of wild-type (wt) and MET^−/− Capan-I cells. (b) HGF-dependent phosphorylation of MET and activation of downstream transducers in lysates from wild-type (wt) and MET^−/− Capan-I cells. Vinculin was used as a loading control. p145 MET: MET receptor β chain; p145 P-MET: phosphorylated MET receptor β chain; p190 MET: precursor form of the MET receptor; p60AKT: AKT; p60 P-AKT: phosphorylated AKT; p44/42ERK: ERK; p44/42 P-ERK: phosphorylated ERK; p117 Vinculin: vinculin. The uncropped Western blots are shown in Supplementary Materials File S1.

Figure 6

In vitro analysis of HGF-driven biological responses in wild-type and MET^−/− Capan-I cells. (a) Analysis of viability by quantitation of metabolically active wild-type (wt) or MET^−/− Capan-I cells treated or not with HGF (25 ng/mL) after two or four days of culture. Graph reports the fold increase in cell number with respect to day one. Each point is the mean of triplicate values; bars represent SD. (b) Cell cycle analysis of wild-type (wt) or MET^−/− Capan-I cells treated or not with HGF (25 ng/mL) after 48 h of culture in 2% serum medium. Graph reports the percentage of cells in each cell cycle phase. Each point is the mean of triplicate values; bars represent SD. (c) Analysis of cell invasion by transwell assay with wild-type (wt) or MET^−/− Capan-I cells. Graph reports fold increase in cells migrated through a matrigel layer during 24 h of culture in the presence of HGF (25 ng/mL), with respect to the untreated counterpart. Each point is the mean of duplicates. (d) Analysis of apoptosis induced by loss of cell/matrix interaction (Anoikis) in wild-type (wt) or MET^−/− Capan-I cells. Graph reports the percentage of annexin-positive cells (early and late stages of apoptosis, i.e., Annexin V positive/7AAD negative plus Annexin V positive/7AAD positive cells) after 48 h of culture in the absence of matrix adhesion, treated or not with HGF (25 ng/mL). Each point is the mean of triplicate values; bars represent SD. ***, p ≤ 0.001; *, p ≤ 0.05; ns: not significant. Data reported in the figure are representative of at least two experiments.

Figure 7

In vivo analysis of MET^−/− Capan-I tumors and metastasis. Luciferase-expressing wild-type and MET ^−/− Capan-I cells were injected into the pancreas of hHGF-KI mice. (a) IVIS analysis of mice performed 3, 21, and 35 days post-injection. Each time point represents the mean value of the group. Bars represent SEM. (b–d) IVIS analysis of isolated organs (pancreas, livers, and lungs) excised from mice at day 35. Each dot represents the value of the organ excised from one mouse. The blue lines indicate the threshold (10⁴) below which IVIS values are considered negative. ****, p ≤ 0.0001; ***, p ≤ 0.001; **, p ≤ 0.01. The data reported in the figure are representative of two experiments.