Periostin-mediated NOTCH1 activation between tumor cells and HSCs crosstalk promotes liver metastasis of small cell lung cancer

Figures

• 

  Fig. 1. Secreted protein POSTN is aberrantly upregulated in small cell lung cancer. A POSTN expression levels in normal lung tissues and lung cancer tissues from GPL570 platform and GPL96 platform; B POSTN expression levels in various normal tissues (n = 40) and SCLC tissues (n = 23) in dataset GSE43346; POSTN levels in normal lung tissues (n = 5) and SCLC tissues (n = 9) in dataset GSE6044; C Altered copy numbers of POSTN in patients with primary lung cancers (n = 176) and metastatic lung cancers (n = 203) in Thoracic Cancer & Thoracic PDX datasets; D POSTN expression levels in normal lung tissues (n = 43), primary SCLC tissues (n = 21) and metastatic SCLC tissues (n = 4) in GSE40275 dataset; E–F Pearson’s correlation analysis of the expression levels of POSTN and WNT2/MMP2 in SCLC tumor samples from the cBioPortal database; G IHC images of POSTN in SCLC patient tissue array. Scale bars = 200 μm and 50 μm, respectively; H POSTN positive H-Score analyzed by AIpathwell software; I Relative protein levels of POSTN between NSCLC and SCLC cell lines from Cancer Cell Line Encyclopedia (Broad, 2019); J Volcano map of different proteins in conditioned medium (CM) between human SCLC cell lines…

• 

  Fig. 2. POSTN promotes the progression and metastasis of small cell lung cancer. A-B The mRNA levels (A) and protein levels (B) of POSTN in POSTN-knockdown H128 and H69 cells; C-D Sphere formation assay in POSTN-knockdown H128 and H69 cells (C), and the sphere numbers were quantified (D). Scale bar = 400 μm; E–F The mRNA levels (E) and protein levels (F) of POSTN in POSTN-overexpression H466 cells; G-H Cell migration in POSTN-overexpression H466 cells was detected by wound healing assay (G). Quantification of relative migration area was shown (H). Scale bar = 400 μm; I-J The invasion ability of POSTN-overexpression H466 cells was detected by transwell invasion assay (I). Quantification of invasion cell number was shown (J). Scale bar = 200 μm; K The protein levels of Snail and MMP9 in POSTN-knockdown H128 and H69 cells; L The protein levels of Snail and MMP9 in POSTN-overexpression H466 cells; M The protein levels of POSTN, Snail and MMP9 in mouse xenografts with inducible knockdown of POSTN. *P < 0.05, **P ≤ 0.01, ***P < 0.001, ****P ≤ 0.0001. n = 3. Student's t-test. All data were shown as means ± sem

• 

  Fig. 3. Cancer-derived POSTN affects the activation of hepatic stellate cells. A Representative tumor images of liver metastasis in SCLC subcutaneous graft model; B H&E, IHC, and Masson staining of liver without metastasis and liver with metastasis. Scale bar = 200 μm. LM stands for liver metastasis; C The mRNA levels of fibrosis markers in LX-2 cells treated with SCLC cells CM for 48 h; D The protein levels of fibrosis markers in LX-2 cells treated with SCLC cells CM for 48 h; E–F The protein levels of fibrosis markers in LX-2 cells treated with indicated SCLC cells CM for 48 h; G-H Immunofluorescence detection of α-SMA expression levels in LX-2 cells treated with indicated SCLC cells CM for 48 h. Scale bar = 10 μm. I-J The invasion ability of H466 cells co-cultured with qLX-2 and aLX-2 cells (quiescent, named qLX-2 or activated, named αLX-2) was detected by transwell invasion assay (I). Quantification of invasion cell number was shown (J). Scale bar = 200 μm. *P < 0.05, **P ≤ 0.01, ***P < 0.001, ****P ≤ 0.0001. n = 3. Student's t-test. All data were shown as means ± sem

• 

  Fig. 4. POSTN-enriched conditioned medium activated HSCs via NOTCH signaling axis. A Venn map of up-regulated genes in SCLC CM treated group and OE-POSTN group; B KEGG pathway enrichment analysis of significantly up-regulated genes in both SCLC CM group; C Cluster heat map of positive regulators in NOTCH pathway; D GSEA enrichment analysis of NOTCH pathway positive regulators in LX-2 cells treated with H69 CM; E The String analysis of the interaction between POSTN and NOTCH positive regulators; F The mRNA levels of indicated genes in LX-2 cells treated with SCLC CM for 48 h; G The indicated protein levels in LX-2 cells treated with SCLC CM for 48 h; H The mRNA levels of indicated genes in LX-2 cells treated with different SCLC CM for 48 h; I The indicated protein levels in LX-2 cells treated with different SCLC CM for 48 h. *P < 0.05, **P ≤ 0.01, ***P < 0.001, ****P ≤ 0.0001. n = 3. Student's t-test. All data were shown as means ± sem

• 

  Fig. 5. POSTN directly binds to membrane receptor NOTCH1 to promote the activation of hepatic stellate cells. A Schematic diagram of POSTN and NOTCH1 truncations; B-C Co-IP was used to analyze the interaction between POSTN and full-length NOTCH1 (B) or NOTCH1-NECD domain (C) in 293T cells; D Pull down FLAG to analyze the interaction between NOTCH1-NECD domain and different POSTN truncations in 293T cells; E Pull down HA to analyze the interaction between NOTCH1-NECD domain and different POSTN truncations in 293T cells; F-G The co-localization of POSTN full-length or truncations and NOTCH1-NECD domain in 293T cells detected by Immunofluorescence analysis (F). Quantification of the Pearson value was shown (G). Scale bar = 10 μm; H-I. I The co-localization of exogenous POSTN and NOTCH1-NECD domains in LX-2 cells treated with CM from POSTN-FLAG-overexpression H446 cells detected by Immunofluorescence analysis (H). Quantification of the fluorescence intensity was shown (I). Scale bar = 10 μm; J The indicated proteins level in LX-2 cells treated with SCLC CM and/or siNOTCH as indicated for 48 h; K The indicated proteins level in LX-2 cells treated with SCLC CM and/or DAPT (20 μM) as indicated…

• 

  Fig. 6. Targeting the POSTN-NOTCH1 signaling axis suppresses SCLC tumor growth in vivo. A Experimental schematic illustrating the establishment of co-injection model and the indicated drug treatment; B Tumor size from mice injected with SCLC cells and HSCs in 12 days; C Representative subcutaneous tumor images on day 12; D Statistical chart of the tumor weight in day 12; E Tumor growth curve of mice injected with H128 cells and LX-2 cells in 16-day drug administration; F The appearance of the subcutaneous tumor after 16 days of drug administration; G The subcutaneous tumor weight after 16-day drug administration. H Tumor growth curve of mice injected with H69 cells and LX-2 cells in 16-day drug administration; I The appearance of the subcutaneous tumor after 16 days of drug administration; J The subcutaneous tumor weight after 16-day drug administration. *P < 0.05, **P ≤ 0.01, ***P < 0.001, ****P ≤ 0.0001. n = 3. Difference between two groups was tested by two-sided Student’s t test; the analyses of experiments with multiple groups were performed by one-way analysis of variance (ANOVA). All data were shown as means ± sem

• 

  Fig. 7. Targeting the POSTN-NOTCH1 signaling axis inhibits SCLC liver metastasis in vivo. A Experimental schematic illustrating the establishment of liver metastasis model and the indicated drug treatment; B Representative bioluminescence images of liver metastasis models on day 24; C Number of mice with or without liver metastasis (LM); D Quantification of photon flux in 24 days; E–F Representative bioluminescence images of livers on day 24 (E). Quantification of liver average radiance was shown (F); G Quantification of liver weight (LW) / body weight (BW) ratios on day 24. H-I Serum AST activity (H) and ALT activity (I) of mice on day 24. *P < 0.05, **P ≤ 0.01, ***P < 0.001, ****P ≤ 0.0001. n = 3. Difference between two groups was tested by two-sided Student’s t test; the analyses of experiments with multiple groups were performed by one-way analysis of variance (ANOVA). All data were shown as means ± sem

• 

  Fig. 8. Summary illustration