Blocking the angiopoietin-2-dependent integrin β-1 signaling axis abrogates small cell lung cancer invasion and metastasis

Abstract

Small cell lung cancer (SCLC) is the most aggressive lung cancer entity with an extremely limited therapeutic outcome. Most patients are diagnosed at an extensive stage. However, the molecular mechanisms driving SCLC invasion and metastasis remain largely elusive. We used an autochthonous SCLC mouse model and matched samples from patients with primary and metastatic SCLC to investigate the molecular characteristics of tumor metastasis. We demonstrate that tumor cell invasion and liver metastasis in SCLC are triggered by an Angiopoietin-2 (ANG-2)/Integrin β-1-dependent pathway in tumor cells, mediated by focal adhesion kinase/Src kinase signaling. Strikingly, CRISPR-Cas9 KO of Integrin β-1 or blocking Integrin β-1 signaling by an anti-ANG-2 treatment abrogates liver metastasis formation in vivo. Interestingly, analysis of a unique collection of matched samples from patients with primary and metastatic SCLC confirmed a strong increase of Integrin β-1 in liver metastasis in comparison with the primary tumor. We further show that ANG-2 blockade combined with PD-1-targeted by anti-PD-1 treatment displays synergistic treatment effects in SCLC. Together, our data demonstrate a fundamental role of ANG-2/Integrin β-1 signaling in SCLC cells for tumor cell invasion and liver metastasis and provide a potentially new effective treatment strategy for patients with SCLC.

Keywords: Lung cancer; Oncology.

Figure 1. ITGB1 correlates with late tumor stage and EMT in patients.

Human SCLC data were analyzed regarding ITGB1 correlations. (A) RNA-Seq of human SCLC material was analyzed for the expression of integrin genes (n = 81). (B) Association between ITGB1 expression and UICC tumor stage (n = 75) were tested using the nonparametric Jonckheere-Terpstra test. (C) A single-sample Gene Set Enrichment Analysis (ssGSEA) was performed using the “GSVA” R package against the MSigDB Hallmark EMT gene set from which ITGB1 was removed. (D) mRNA expression data for SCLC provided the “mesenchymal” gene set in human SCLC cell lines (n = 64) that was correlated to ITGB1 gene expression. The heatmap was generated by the heatmap tool provided by SCLC-CellMiner. (E) Pearson correlation coefficients with ITGB1 expression in human SCLC cell lines of all genes per “mesenchymal” and “epithelial” gene set (P values from 1-sample Wilcoxon-tests for median correlation coefficient equal to zero per gene set). Data of mRNA gene expression are listed in the Supplemental Table 1.

Figure 2. ITGB1 is increased in liver metastasis of patients with SCLC.

(A) FFPE lung tumor material of 6 patients with SCLC and matched liver metastasis was analyzed for the expression of ITGB1. (B) Schematic of matched SCLC patient samples obtained from a primary lung tumor and liver metastasis. Created with BioRender.com. (C and D) IHC stained slides were scanned and analyzed regarding ITBG1 expression and Vimentin expression using ImageScope and ImageJ. Representative images of ITGB1 for each patient (n = 6) are shown in A and for Vimentin in Supplemental Figure 4. Statistical analysis was done using 2 way ANOVA (*P < 0.05; data are shown as mean ± SEM). Clinicopathologic characteristics are listed in the Supplemental Table 2. Scale bars: 60 μm.

Figure 3. ITGB1 is increased on tumor cells upon extensive stage of disease and correlates with EMT in an autochthonous SCLC mouse model.

SCLC-bearing mice were analyzed for metastatic disease and the expression of the ANG-2 receptor ITGB1. (A) Schematic of the autochthonous Rb1-loss– and Trp53-loss–driven SCLC mouse model developing limited stage of disease (LD-SCLC) and extensive stage of disease (ED-SCLC). Created with BioRender.com. (B) Relative expression of ITGB1 on SCLC tumor cells isolated from mice (n = 6–9) normalized to IgG control, determined by flow cytometry. Origin of tumor cells, presence of metastasis, and stage are indicated. (C) Representative IHC results for E-cadherin, Vimentin, Snail, and ITGB1 in matched murine tumors obtained from the primary SCLC and SCLC liver metastasis. Scale bars: 200 μm. (D–G) Quantification of protein expression (n = 5–7) using the threshold method of ImageJ in matched samples. Statistical analysis was done using 2 way ANOVA (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM).

Figure 4. ITGB1-KO reduced the promigratory and proinvasive gene signatures in SCLC.

(A) Schematic of ITGB1-KO cell line generation obtained from murine liver metastasis. Created with BioRender.com. (B) Relative expression of the ITGB1 SCLC tumor cells before and after ITGB1-KO as mean fluorescence intensity (MFI) normalized to IgG control, determined by flow cytometry with representative histograms. Statistical analysis was done using 2 way ANOVA (***P < 0.001; data are shown as mean ± SEM). (C) Volcano plot of RNA-Seq data of ITGB1-KO versus controls (n = 3 per group). DEGs downregulated upon ITGB1-KO are depicted in blue, and those upregulated are depicted in red. GO enrichment analysis listed in Supplemental Table 4 revealed genes of the pathway GO:0030334 (regulation of cell migration) enriched, indicated in green. (D and E) Significantly deregulated genes were classified as proinvasive/promigratory, not linked to invasion/migration, antiinvasive/antimigratory, or pseudogene listed in Supplemental Table 3 and compared regarding their enrichment upon ITGB1-KO.

Figure 5. ITGB1 signaling is needed for intravasation of SCLC cells.

(A) Schematic of orthotopic injection of SCLC cells to determine intravasation capacity. Created with BioRender.com. (B) Representative μCT of WT SCLC clones and ITGB1-KO injected into the lungs and growing in vivo with representative images of murine lungs and liver postmortem (n = 5 per group). (C) Representative images of ITGB1 IHC stain of SCLC WT and ITGB1-KO orthotopically injected into the lung (n = 5 per group). Scale bars: 100 μm. (D) The capability of WT SCLC clones and ITGB1-KO to form tumors after orthotopic injection was determined by IHC based on H&E and NCAM and quantified for lungs and livers, respectively (n = 5).

Figure 6. ANG-2–dependent ITGB1 signaling triggers SCLC cell migration.

(A and B) The area of migrated SCLC tumor cells isolated from liver metastasis was determined by scratch assay (n = 6). Cells were stimulated with ANG-2 for 24 hours. Images after 0 and 24 hours were analyzed using ImageJ. Statistical analysis was done using the Student’s 2-tailed t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM). (C and D) The area of migrated SCLC tumor cells with Crispr ITGB1-KO was determined by scratch assay (n = 6). Cells were stimulated with ANG-2 for 24 hours. Statistical analysis was done using the 2-tailed Student’s t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM).

Figure 7. ANG-2–dependent ITGB1 signaling triggers FAK-SRC signaling in SCLC.

(A) Intracellular FAK-SRC signaling was determined by Western blot upon ANG-2 stimulation of cultured SCLC cells isolated from liver metastasis. Representative experiment out of 3. (B) Western blots were quantified according to total protein levels and normalized to β-actin. Statistical analysis was done using the 2-tailed Student’s t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM). (C) EMT proteins, components of the FAK-SRC signaling were determined by Western blot upon ANG-2 stimulation and simultaneous treatment by 100 nM Saracatinib. Representative experiment out of 3. (D) Western blots were quantified according to total protein levels and normalized to β-actin. Statistical analysis was done using the 2-tailed Student’s t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM).

Figure 8. Blockade of ANG-2–dependent ITGB1 signaling abrogates SCLC liver metastases in vivo.

(A) Scheme of SCLC-bearing mice treated for 2 weeks with anti–ANG-2 antibody or corresponding IgG control. Created with BioRender.com. (B and C) Liver tissue was harvested and the average number of microscopic liver metastases per 2 mm² was counted by scanned H&E slides. IgG (black; n = 10); anti–ANG-2 monotherapy (aANG-2, light blue; n = 6). Representative images for relevant conditions are shown. Scale bars: 500 μm (IgG) and 100 μm (aANG). Metastases are indicated by black arrows and dashed lines, respectively. (D) The role of ITGB1 in intra- and extravasation. Adapted from “Overview of Metastatic Cascade,” by BioRender.com (2022). Retrieved from https://app.biorender.com/biorender-templates

Figure 9. Synergistic treatment effects by combined blockade of ANG-2/ITGB1, VEGFR, and anti–PD-1 in mice with SCLC.

(A) Schematic of SCLC-bearing mice treated with vehicle control (black; n = 10), IgG control (gray; n = 7), VEGFR inhibitor monotherapy (VEGFRi, dark blue; n = 6), anti–ANG-2 monotherapy (aANG-2, light blue; n = 6), anti–PD-1 monotherapy (aPD-1, orange; n = 10), and anti–ANG-2/VEGFR inhibitor/anti–PD-1 triple combination therapy (triple, pink; n = 10). Created with BioRender.com. (B and C) Overall survival (OS) and progression-free survival (PFS) were determined based on mouse adapted RECIST v1.1 criteria determined by serial μCT of mice from the 6 therapy cohorts. Statistical analysis was done using Mantel-Cox test (**P < 0.01). (D) Change in target lesion diameter from the start of therapy upon best response to treatment. The received therapy is indicated by the color code. PD, progressive disease; SD, stable disease; PR, partial response. (E and F) Tumor cells isolated from primary SCLC tumors under therapy were analyzed for PD-L1 and MHC class I expression by flow cytometry. (G) Tumor cells isolated from primary SCLC tumors under therapy were analyzed for PD-L1 and MHC class I expression under response and resistance by flow cytometry. Statistical analysis was done using the 2-tailed Student’s t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM).

Figure 10. Resistance to anti–PD-1 therapy–induced exhausted T cells in SCLC primary and liver metastases.

SCLC-bearing mice were analyzed for microvessel density and T cell infiltrates based on IHC and flow cytometry upon detection of progressive disease based on mouse adapted RECIST v1.1. (A–F) Corresponding to primary SCLC samples treated as follows: vehicle control (vehicle; n = 9), IgG control (IgG; n = 7), VEGFR inhibitor monotherapy (VEGFRi; n = 6), anti–ANG-2 monotherapy (aANG-2; n = 6), anti–ANG-2 and VEGFR inhibitor combination therapy (aANG-2/VEGFRi; n = 5), anti–PD-1 monotherapy (aPD-1; n = 6), and anti–ANG-2/VEGFR inhibitor/anti–PD-1 triple combination therapy (triple; n = 8). (G–L) Corresponding to SCLC liver metastases: vehicle (n = 7), IgG control (IgG; n = 6), VEGFR inhibitor monotherapy (VEGFRi; n = 6), anti–ANG-2 monotherapy (aANG-2; n = 3), anti–ANG-2 and VEGFR inhibitor combination therapy (aANG-2/VEGFRi; n = 5), anti–PD-1 monotherapy (aPD-1; n = 6), and anti–ANG-2/VEGFR inhibitor/anti–PD-1 triple combination therapy (triple; n = 5). (A and G) Microvessel density was determined based on IHC using ImageJ to measure CD31⁺ counts in a field of view (FOV) of 200 μm². Statistical analysis was done using the 2-tailed Student’s t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM). (B and H) The ratio of CD4⁺ T cells versus CD8⁺ T cells is compared in the different therapy cohorts based on flow cytometry. (C and I) The fraction of PD-1 and TIM-3 double-positive CD8⁺ T cells is determined for each therapy cohort based on flow cytometry. (D and J) The fraction of PD-1 and TIM-3 double-positive CD4⁺ T cells is determined for each therapy cohort based on flow cytometry. (E and K) Dot plots of flow cytometry of CD8⁺ T cells of 1 representative experiment are shown. (F and L) Dot plots of flow cytometry of CD4⁺ T cells of 1 representative experiment are shown. Statistical analysis was done using the 2-tailed Student’s t test (*P < 0.05; **P < 0.01; ***P < 0.001; data are shown as mean ± SEM). Representative IHC stains of primary SCLC and SCLC liver metastases are shown in Supplemental Figures 15–18.