Fig. 4.

Knockdown of EGFL9 inhibits distant tumor metastasis in vivo. a MRI examination identified peritumor lymph nodes metastasis in 4T1/NT injected mice (bottom panels), but not in 4T1/EGFL9-sh2 injected mice (top panels). Five cases of 4T1 tumor xenograft mouse model were shown. Yellow arrow showed the peritumor localization in each case. b MRI examination identified axillary lymph nodes and lung metastasis in 4T1/EGFL9-NT-injected mice (bottom panels), but not in 4T1/EGFL9-sh2-injected mice (top panels). Five cases of 4T1 tumor xenograft mouse model were shown. Yellow arrow shows the localization of axillary lymph nodes metastasis and red arrow showed the lung metastasis in each case. c Quantitative tumor size summary by MRI detection. No significant difference was observed in two groups of tumors (NS, P > 0.05). Each bar represents the mean ± SEM (standard error of the mean, n = 5). P value was determined by unpaired two-tailed t-test. d The frequencies of metastasis identified by MRI in peritumor lymph nodes (5/5 vs 2/5), axillary lymph nodes (5/5 vs 0/5), and lungs (4/5 vs 1/5) of BALB/c mice (n = 5 in each group) with control 4T1 (blue bar) and 4T1/EGFL9-sh2 (red bar) cells. e Comparison of metastatic lesion numbers in two groups of tumors. Average metastatic lesion numbers from five series sections were counted for each tumor. ***P < 0.001. f Comparison of metastasis lesion area in two groups of tumors. Relative metastatic lesion areas (tumor lesion/lung section) from sections with highest metastatic lesion number in each tumor were compared. ***P < 0.001. For both e and f panels, P values were determined by unpaired two-tailed t-test (n = 5). g Representative H&E staining of lung sections from mice. IHC demonstrated significant metastatic lesions in NT (non-targeting shRNA)-transfected 4T1 cells (top panels) in comparison with EGFL9-sh2-transfected 4T1 (bottom panels). Magnification: ×40 (left panels) and ×100 (right panels). Scale bar: 200 µm