Expression and mutational analysis of MET in human solid cancers

Abstract

MET receptor tyrosine kinase and its ligand hepatocyte growth factor (HGF) regulate a variety of cellular functions, many of which can be dysregulated in human cancers. Activated MET signaling can lead to cell motility and scattering, angiogenesis, proliferation, branching morphogenesis, invasion, and eventual metastasis. We performed systematic analysis of the expression of the MET receptor and its ligand HGF in tumor tissue microarrays (TMA) from human solid cancers. Standard immunohistochemistry (IHC) and a computerized automated scoring system were used. DNA sequencing for MET mutations in both nonkinase and kinase domains was also performed. MET was differentially overexpressed in human solid cancers. The ligand HGF was widely expressed in both tumors, primarily intratumoral, and nonmalignant tissues. The MET/HGF likely is functional and may be activated in autocrine fashion in vivo. MET and stem cell factor (SCF) were found to be positively stained in the bronchioalevolar junctions of lung tumors. A number of novel mutations of MET were identified, particularly in the extracellular semaphorin domain and the juxtamembrane domain. MET-HGF pathway can be assayed in TMAs and is often overexpressed in a wide variety of human solid cancers. MET can be activated through overexpression, mutation, or autocrine signaling in malignant cells. Mutations in the nonkinase regions of MET might play an important role in tumorigenesis and tumor progression. MET would be an important therapeutic antitumor target to be inhibited, and in lung cancer, MET may represent a cancer early progenitor cell marker.

Figures

Figure 1

(A), Human solid cancer tumor tissue microarrays (TMA): Breast cancer (breast adenocarcinoma: n=40); Lung cancer (non-small cell lung carcinoma: n=40); Colon cancer (adenocarcinoma, n=30; adenocarcinoma, metastatic, n=10); Kidney cancer (renal cell, n=1; renal cell, metastatic, n=3; renal cell papillary, n=6; renal cell papillary, metastatic, n=1; renal cell chromophobes, n=3; renal cell, clear cell, n=21, renal cell, clear cell, metastatic, n=3; renal cell, clear cell, scarcomatoid, n=2); Ovarian cancer (36 adenocarcinomas, 1 small cell and 3 endometroid carcinomas); (B), Non-neoplastic tissues (multi-organ array including 2 tissue cores of each of 25 organs). Examples of 12 non-neoplastic tissues cores with MET IHC staining are shown here. (C), Automated Cellular Imaging System (ACIS). The ACIS system measures the intensity of the staining based on three related color parameters: the color defined by hue, the "darkness" defined as luminosity, and the density of the color defined as saturation. ACIS software for TMA was programmed by an experienced user-pathologist by setting the color-specific thresholds to determine and calculate staining intensity and the ratio of positively stained cells to the entire area of selection. (D), Expression of MET, HGF and phospho-MET (Y1003) in frozen sections of lung adenocarcinomas (cases 1&2) and squamous cell carcinoma (case 3). Original magnification x200. (E), Comparative IHC on the same tumor sample using 3 different c-MET antibodies (magnification x100). (a) MET from Zymed (clone 3D4) shows strong signal only within tumor tissue. (b) MET from Santa Cruz (rabbit polyclonal) shows moderately strong signal with non-specific reactivity in lung macrophages and inflammatory stromal cells. (c) MET staining with another Zymed antibody (Rabbit polyclonal) exhibit weak signal in lung tumor and moderate non-specific positivity of necrotic areas and alveolar macrophages. Simultaneous expression of MET (d), HGF (e) and phospho-MET (f) in non-small cell lungg carcinoma (magnification x200).

Figure 1

(A), Human solid cancer tumor tissue microarrays (TMA): Breast cancer (breast adenocarcinoma: n=40); Lung cancer (non-small cell lung carcinoma: n=40); Colon cancer (adenocarcinoma, n=30; adenocarcinoma, metastatic, n=10); Kidney cancer (renal cell, n=1; renal cell, metastatic, n=3; renal cell papillary, n=6; renal cell papillary, metastatic, n=1; renal cell chromophobes, n=3; renal cell, clear cell, n=21, renal cell, clear cell, metastatic, n=3; renal cell, clear cell, scarcomatoid, n=2); Ovarian cancer (36 adenocarcinomas, 1 small cell and 3 endometroid carcinomas); (B), Non-neoplastic tissues (multi-organ array including 2 tissue cores of each of 25 organs). Examples of 12 non-neoplastic tissues cores with MET IHC staining are shown here. (C), Automated Cellular Imaging System (ACIS). The ACIS system measures the intensity of the staining based on three related color parameters: the color defined by hue, the "darkness" defined as luminosity, and the density of the color defined as saturation. ACIS software for TMA was programmed by an experienced user-pathologist by setting the color-specific thresholds to determine and calculate staining intensity and the ratio of positively stained cells to the entire area of selection. (D), Expression of MET, HGF and phospho-MET (Y1003) in frozen sections of lung adenocarcinomas (cases 1&2) and squamous cell carcinoma (case 3). Original magnification x200. (E), Comparative IHC on the same tumor sample using 3 different c-MET antibodies (magnification x100). (a) MET from Zymed (clone 3D4) shows strong signal only within tumor tissue. (b) MET from Santa Cruz (rabbit polyclonal) shows moderately strong signal with non-specific reactivity in lung macrophages and inflammatory stromal cells. (c) MET staining with another Zymed antibody (Rabbit polyclonal) exhibit weak signal in lung tumor and moderate non-specific positivity of necrotic areas and alveolar macrophages. Simultaneous expression of MET (d), HGF (e) and phospho-MET (f) in non-small cell lungg carcinoma (magnification x200).

Figure 1

(A), Human solid cancer tumor tissue microarrays (TMA): Breast cancer (breast adenocarcinoma: n=40); Lung cancer (non-small cell lung carcinoma: n=40); Colon cancer (adenocarcinoma, n=30; adenocarcinoma, metastatic, n=10); Kidney cancer (renal cell, n=1; renal cell, metastatic, n=3; renal cell papillary, n=6; renal cell papillary, metastatic, n=1; renal cell chromophobes, n=3; renal cell, clear cell, n=21, renal cell, clear cell, metastatic, n=3; renal cell, clear cell, scarcomatoid, n=2); Ovarian cancer (36 adenocarcinomas, 1 small cell and 3 endometroid carcinomas); (B), Non-neoplastic tissues (multi-organ array including 2 tissue cores of each of 25 organs). Examples of 12 non-neoplastic tissues cores with MET IHC staining are shown here. (C), Automated Cellular Imaging System (ACIS). The ACIS system measures the intensity of the staining based on three related color parameters: the color defined by hue, the "darkness" defined as luminosity, and the density of the color defined as saturation. ACIS software for TMA was programmed by an experienced user-pathologist by setting the color-specific thresholds to determine and calculate staining intensity and the ratio of positively stained cells to the entire area of selection. (D), Expression of MET, HGF and phospho-MET (Y1003) in frozen sections of lung adenocarcinomas (cases 1&2) and squamous cell carcinoma (case 3). Original magnification x200. (E), Comparative IHC on the same tumor sample using 3 different c-MET antibodies (magnification x100). (a) MET from Zymed (clone 3D4) shows strong signal only within tumor tissue. (b) MET from Santa Cruz (rabbit polyclonal) shows moderately strong signal with non-specific reactivity in lung macrophages and inflammatory stromal cells. (c) MET staining with another Zymed antibody (Rabbit polyclonal) exhibit weak signal in lung tumor and moderate non-specific positivity of necrotic areas and alveolar macrophages. Simultaneous expression of MET (d), HGF (e) and phospho-MET (f) in non-small cell lungg carcinoma (magnification x200).

Figure 1

(A), Human solid cancer tumor tissue microarrays (TMA): Breast cancer (breast adenocarcinoma: n=40); Lung cancer (non-small cell lung carcinoma: n=40); Colon cancer (adenocarcinoma, n=30; adenocarcinoma, metastatic, n=10); Kidney cancer (renal cell, n=1; renal cell, metastatic, n=3; renal cell papillary, n=6; renal cell papillary, metastatic, n=1; renal cell chromophobes, n=3; renal cell, clear cell, n=21, renal cell, clear cell, metastatic, n=3; renal cell, clear cell, scarcomatoid, n=2); Ovarian cancer (36 adenocarcinomas, 1 small cell and 3 endometroid carcinomas); (B), Non-neoplastic tissues (multi-organ array including 2 tissue cores of each of 25 organs). Examples of 12 non-neoplastic tissues cores with MET IHC staining are shown here. (C), Automated Cellular Imaging System (ACIS). The ACIS system measures the intensity of the staining based on three related color parameters: the color defined by hue, the "darkness" defined as luminosity, and the density of the color defined as saturation. ACIS software for TMA was programmed by an experienced user-pathologist by setting the color-specific thresholds to determine and calculate staining intensity and the ratio of positively stained cells to the entire area of selection. (D), Expression of MET, HGF and phospho-MET (Y1003) in frozen sections of lung adenocarcinomas (cases 1&2) and squamous cell carcinoma (case 3). Original magnification x200. (E), Comparative IHC on the same tumor sample using 3 different c-MET antibodies (magnification x100). (a) MET from Zymed (clone 3D4) shows strong signal only within tumor tissue. (b) MET from Santa Cruz (rabbit polyclonal) shows moderately strong signal with non-specific reactivity in lung macrophages and inflammatory stromal cells. (c) MET staining with another Zymed antibody (Rabbit polyclonal) exhibit weak signal in lung tumor and moderate non-specific positivity of necrotic areas and alveolar macrophages. Simultaneous expression of MET (d), HGF (e) and phospho-MET (f) in non-small cell lungg carcinoma (magnification x200).

Figure 1

(A), Human solid cancer tumor tissue microarrays (TMA): Breast cancer (breast adenocarcinoma: n=40); Lung cancer (non-small cell lung carcinoma: n=40); Colon cancer (adenocarcinoma, n=30; adenocarcinoma, metastatic, n=10); Kidney cancer (renal cell, n=1; renal cell, metastatic, n=3; renal cell papillary, n=6; renal cell papillary, metastatic, n=1; renal cell chromophobes, n=3; renal cell, clear cell, n=21, renal cell, clear cell, metastatic, n=3; renal cell, clear cell, scarcomatoid, n=2); Ovarian cancer (36 adenocarcinomas, 1 small cell and 3 endometroid carcinomas); (B), Non-neoplastic tissues (multi-organ array including 2 tissue cores of each of 25 organs). Examples of 12 non-neoplastic tissues cores with MET IHC staining are shown here. (C), Automated Cellular Imaging System (ACIS). The ACIS system measures the intensity of the staining based on three related color parameters: the color defined by hue, the "darkness" defined as luminosity, and the density of the color defined as saturation. ACIS software for TMA was programmed by an experienced user-pathologist by setting the color-specific thresholds to determine and calculate staining intensity and the ratio of positively stained cells to the entire area of selection. (D), Expression of MET, HGF and phospho-MET (Y1003) in frozen sections of lung adenocarcinomas (cases 1&2) and squamous cell carcinoma (case 3). Original magnification x200. (E), Comparative IHC on the same tumor sample using 3 different c-MET antibodies (magnification x100). (a) MET from Zymed (clone 3D4) shows strong signal only within tumor tissue. (b) MET from Santa Cruz (rabbit polyclonal) shows moderately strong signal with non-specific reactivity in lung macrophages and inflammatory stromal cells. (c) MET staining with another Zymed antibody (Rabbit polyclonal) exhibit weak signal in lung tumor and moderate non-specific positivity of necrotic areas and alveolar macrophages. Simultaneous expression of MET (d), HGF (e) and phospho-MET (f) in non-small cell lungg carcinoma (magnification x200).

Figure 2

Tumor microarray expression analysis of MET in human solid cancers. Tumor microarrays from human solid cancers were used for immunnostaining with anti-total-MET and also phosphospecific MET antibodies as described in the Materials and Methods. Cytoplasmic and membranous expression of MET and phospho-MET was quantified manually in each core at x400 magnification by two pathologists (AM, AH) and with Automated Cellular Imaging System (ACIS, Clarient, CA) as described in the Materials and Methods. MET and phospho-MET intensity measurements were translated into the 4-tier system as negative (0), weak (1+), moderate (2+), or strong (3+) staining and compared with manual scoring. Combined quantitative analyses of the expression of total-MET and phospho-MET in different human solid cancers are shown here. (A), Total-MET expression. (B), Tumor specific localization of MET expression. (C), Phospho-MET expression.

Figure 2

Tumor microarray expression analysis of MET in human solid cancers. Tumor microarrays from human solid cancers were used for immunnostaining with anti-total-MET and also phosphospecific MET antibodies as described in the Materials and Methods. Cytoplasmic and membranous expression of MET and phospho-MET was quantified manually in each core at x400 magnification by two pathologists (AM, AH) and with Automated Cellular Imaging System (ACIS, Clarient, CA) as described in the Materials and Methods. MET and phospho-MET intensity measurements were translated into the 4-tier system as negative (0), weak (1+), moderate (2+), or strong (3+) staining and compared with manual scoring. Combined quantitative analyses of the expression of total-MET and phospho-MET in different human solid cancers are shown here. (A), Total-MET expression. (B), Tumor specific localization of MET expression. (C), Phospho-MET expression.

Figure 2

Tumor microarray expression analysis of MET in human solid cancers. Tumor microarrays from human solid cancers were used for immunnostaining with anti-total-MET and also phosphospecific MET antibodies as described in the Materials and Methods. Cytoplasmic and membranous expression of MET and phospho-MET was quantified manually in each core at x400 magnification by two pathologists (AM, AH) and with Automated Cellular Imaging System (ACIS, Clarient, CA) as described in the Materials and Methods. MET and phospho-MET intensity measurements were translated into the 4-tier system as negative (0), weak (1+), moderate (2+), or strong (3+) staining and compared with manual scoring. Combined quantitative analyses of the expression of total-MET and phospho-MET in different human solid cancers are shown here. (A), Total-MET expression. (B), Tumor specific localization of MET expression. (C), Phospho-MET expression.

Figure 3

Tumor microarray expression analysis of HGF in human solid cancers. Tumor microarrays from human solid cancers were used for immunnostaining with anti-HGF antibody as described in the Materials and Methods. Similar to the analyses shown in Figure 2 above, expression of HGF was quantified manually in each core at x400 magnification by two pathologists (AM, AH) and with Automated Cellular Imaging System (ACIS, Clarient, CA). HGF intensity measurements were translated into the 4-tier system as negative (0), weak (1+), moderate (2+), or strong (3+) staining and compared with manual scoring. Combined quantitative analyses of the expression of HGF in different human solid cancers are shown here. (A), TMA analysis of HGF expression. (B), HGF expression in human solid cancers.

Figure 3

Tumor microarray expression analysis of HGF in human solid cancers. Tumor microarrays from human solid cancers were used for immunnostaining with anti-HGF antibody as described in the Materials and Methods. Similar to the analyses shown in Figure 2 above, expression of HGF was quantified manually in each core at x400 magnification by two pathologists (AM, AH) and with Automated Cellular Imaging System (ACIS, Clarient, CA). HGF intensity measurements were translated into the 4-tier system as negative (0), weak (1+), moderate (2+), or strong (3+) staining and compared with manual scoring. Combined quantitative analyses of the expression of HGF in different human solid cancers are shown here. (A), TMA analysis of HGF expression. (B), HGF expression in human solid cancers.

Figure 4

Mutations of MET in human solid tumors. MET receptor is shown in the schematic diagram highlighting different functional domains of the receptor: extracellular semaphorin (Sema) domain, PSI domain, the four IPT-repeats, transmembrane (TM) domain, juxtamembrane (JM) domain and cytoplasmic tyrosine kinase (TK) domain. The MET mutations identified in different human solid cancers in this study are represented in the top. Summary of various mutations of MET previously reported in human solid cancers, including renal cell carcinomas (both sporadic and hereditary), gastric carcinoma, hepatocellular carcinoma, glioma, squamous cell carcinoma of the head and neck, SCLC, NSCLC, mesothelioma and melanoma, are shown in the bottom for comparison (see references: Ma et al., 2003a, 2003b, 2005a; Jagadeeswaran et al., 2006; Puri et al., 2007).

Figure 5

MET expression at the bronchioalveolar junction of normal lung and lung tumors with implication of lung progenitor cells. (A), Relative location of MET expression in normal lung tissues was examined using immunostaining with anti-MET antibody on a normal lung tissue showing more proximal airways including the bronchioalveolar junction. The strong positive immuno-expression of MET selectively at the bronchioaleolvar junction (see arrows) may imply a role of MET in the biology of lung progenitor cells. (B), Identification of bronchioalveolar progenitor cells in lung TMA using stem cell factor marker (SCF, Rabbit monoclonal, Abcam). SCF positivity of regenerating epithelial cells in area of inflammation and fibrosis (a, x100; d – same, higher magnification (x400). Brochioalveolar lung carcinoma with almost identical staining pattern with SCF and MET markers (b – SCF, x100, c – MET, x100; e – SCF, x400, f – MET, x400).

Figure 5

MET expression at the bronchioalveolar junction of normal lung and lung tumors with implication of lung progenitor cells. (A), Relative location of MET expression in normal lung tissues was examined using immunostaining with anti-MET antibody on a normal lung tissue showing more proximal airways including the bronchioalveolar junction. The strong positive immuno-expression of MET selectively at the bronchioaleolvar junction (see arrows) may imply a role of MET in the biology of lung progenitor cells. (B), Identification of bronchioalveolar progenitor cells in lung TMA using stem cell factor marker (SCF, Rabbit monoclonal, Abcam). SCF positivity of regenerating epithelial cells in area of inflammation and fibrosis (a, x100; d – same, higher magnification (x400). Brochioalveolar lung carcinoma with almost identical staining pattern with SCF and MET markers (b – SCF, x100, c – MET, x100; e – SCF, x400, f – MET, x400).