Fig. 3

MSI2 deficiency in CAFs inhibits NSCLC metastasis in vivo. A–C MSI2 deficiency in CAFs has no effect on tumor growth. Luciferase-labeled NSCLC NCI-H292 cells were subcutaneously injected with Ctrl CAFs or gMSI2 CAFs (ratio 1:1) into both flanks of NU/J mice. A Tumor volume was monitored over three-week period. Data are mean ± SD (n = 8; 2 tumors/mice). B Representative bioluminescence images and its quantified bioluminescence signals at primary sites (right) at 3 weeks post inoculation. Data are mean ± SD (n = 4). C Representative images of gross tumors (left) and weight of resected tumors at week 3 (right). Scale bar = 1 cm. Data are mean ± SD (n = 8). D Representative images of tumor xenografts showing local spread of tumor mass (arrowhead). Asterisk, primary tumor mass. Frequencies of mice with or without a local spread were compared between NSCLC with Ctrl or gMSI2 CAFs. ^*P < 0.05 versus NCI-H292 cells with Ctrl CAFs; chi-squared test. E Representative bioluminescence images of the dissected lungs and its quantified bioluminescence signals. Data are mean ± SEM (n = 4); ^*P < 0.05 versus NCI-H292 cells with Ctrl CAFs; Mann–Whitney U test, two-sided. F Representative micrographs of H&E-stained lung tissues. Scale bars = 500 μm or 100 μm. Relative lung metastasis area and lung metastatic foci were quantified from the H&E micrographs using ImageJ. Data are mean ± SD (n = 3); ^*P < 0.05, ^***P < 0.001 versus NCI-H292 cells with Ctrl CAFs; two-tailed Student’s t-test. G Representative IHC micrographs of primary tumors stained for Ki-67. Scale bar = 100 μm. Percentage of Ki-67-positive cells relative to total number of cells was assessed. Data are mean ± SD (n = 3)