Fig 2: Epithelial-mesenchymal transition increases tumor cell adherence to collagen through a1ß1 integrin.(A) Cell adherence was quantified 1 hour after the indicated KP cells were seeded on wells coated with collagen 1 (Col1), collagen IV (Col4), fibronectin, BSA, or nothing, and OD595 values are shown in the heatmap. E, epithelial; M, mesenchymal. n = 4. (B) Cell spreading assay on Col1. Cells were plated in Col1-coated plates for 1 hour, and percentages of spread cells were quantified. Scale bar: 50 µm. (C) Quantitative reverse transcription PCR analysis of the mRNA levels of ZEB1 and Col1 receptors in KP cell lines (Pearson correlation). (D) Western blot analysis of Itga1 and Itgb1 in KP cell lines classified as epithelial or mesenchymal. ß-Actin was used as a loading control. (E) Heatmap illustration of gene expression levels in lung adenocarcinoma and squamous cell carcinoma data sets (n = 1016 tumors) in The Cancer Genome Atlas. An epithelial-mesenchymal transition score was correlated with the levels of collagen receptors using the Pearson coefficient (r value). (F) Cell adherence was quantified 1 hour or 2 hours after the indicated cells were seeded on uncoated wells or on wells coated with Col1. n = 4. **P < 0.01. (G) Cell spreading assay on Col1. Percentages of spread cells were quantified. Scale bar: 100 µm. ***P < 0.001. (H) Quantitative reverse transcription PCR analysis of Col1 receptors. (I) Western blot analysis of Itga1 in 393P_Vec and 393P_ZEB1 cells. Data are shown as the mean ± SEM from a single experiment incorporating biological replicate samples (n = 3, unless otherwise indicated) and are representative of at least 2 independent experiments. *P < 0.01; ***P < 0.001. Two-tailed Student’s t test.

Fig 3: EB1-pulldown following TTL-modulation. (A) MDCK cell lysates were incubated with anti-EB1 antibodies followed by precipitation with agarose beads. Precipitates were analyzed by immunoblot using antibodies directed against ß1-integrin (CD29), acetyl-, detyr- and tyr-tubulin, ILK, GAPDH and EB1. n = 3 independent experiments. PC, Pre-clearing using agarose beads; IP+, Immunoprecipitation using EB1 antibodies and agarose beads; IP-, Immunoprecipitation using unspecific IgG antibodies and agarose beads. Representative results. (B) MDCK?TTL cell lysates were incubated with anti-EB1 antibodies followed by precipitation with agarose beads. Precipitates were analyzed by immunoblot using antibodies directed against ß1-integrin, detyr-tubulin, ILK, GAPDH and EB1. n = 3 independent experiments. PC, Pre-clearing using agarose beads; IP+, Immunoprecipitation using EB1 antibodies and agarose beads; IP-, Immunoprecipitation using unspecific IgG antibodies and agarose beads. Representative results. (C–E) Quantification of MDCK and MDCK?TTL co-precipitation efficiencies of ß1-integrin (C), detyr-tubulin (D) and ILK (E) from three independent experiments. The efficiencies were normalized by the total quantities of each polypeptide in the input. Mean ± s.d., n = 3. Statistical significance was tested with Student`s unpaired t-test (*p < 0.05; **p < 0.01).

Fig 4: ZEB1 increases tumor cell adherence to Col1 through a1ß1 integrin.(A) Western blot analysis of Itga1 in 393P_ZEB1 cells transfected with Itga1 shRNAs (shItga1) or control shRNA (shCTL). (B) Cell adhesion to Col1 or plastic (no coating) at indicated time points. n = 4. *P = 0.01; ***P < 0.001. (C) Percentages of spread cells on collagen at indicated time points. ***P < 0.001. (D) Western blot analysis of Itga1 in 393P cells transfected with Itga1 expression vector or control vector (Vec). (E) Cell adhesion to Col1 or plastic at indicated time points. n = 4. *P = 0.03; **P = 0.005. (F) Percentages of spread cells on Col1 at indicated time points. ***P < 0.001. (G–I) Cell adhesion to collagen or plastic. H1299 cells transfected with ITGA1 shRNA or control shRNA (shCTL) and Itga1 expression vector or empty vector (Vec) (G), 344SQ cells transfected with shRNA against Itgb1 or control shRNA and siRNA against Itga1 or control siRNA (siCTL; H), and 393P cells transfected with Iga1 expression vector or control vector and siRNA against Itgb1 or siCTL (I) are shown. n = 4. **P < 0.01; ***P < 0.001. (J) Schema showing tail vein injection of red fluorescent protein–labeled (RFP-labeled) 344SQ cells into mice. (K) Percentage of RFP-positive cells in mouse lungs at 4 hours after injection. (L) Green fluorescent protein–labeled (GFP-labeled) and RFP-labeled 344SQ transfectants were coinjected into mice and the resulting micrometastases were imaged and quantified. Data are shown as the mean ± SEM from a single experiment incorporating biological replicate samples (n = 3, unless otherwise indicated) and are representative of at least 2 independent experiments. Two-tailed Student’s t test for 2-group comparisons; 1-way ANOVA test for multiple comparisons.

Fig 5: MiR-140-3p promotes bone formation in DM rats with bone defects. A Representative micro-CT images within a region of interest (ROI) comprising the bone defect after 2 and 8 weeks following surgery in the bone defect of normal rats and those with diabetes mellitus after treatment with N-Exos, DM-Exos, and Exos overexpressing miR-140-3p. n = 5 in each group. B Quantitative analysis of new bone formation from micro-CT imaging in DM rats with bone defects after 2 and 8 weeks after transplantation with N-Exos, DM-Exos, or Exos overexpressing miR-140-3p. BV: bone volume; TV: tissue volume; Tb,N: trabecular number; Tb.Th: trabecular thickness; Tb.Sp: trabecular Spacing. n = 5 in each group. C–E Histological analysis of decalcified sections of bone defects in diabetic rats treated with N-Exos, DM-Exos, or Exos overexpressing miR-140-3p. HE/Masson’s/Safranin O/fast green staining of the new bone formation within the defect. Black arrows indicate new bone stained in light pink/blue/green within the 3 mm-diameter defect marked as a circle. Quantitative analysis of new bone formation was performed using ImageJ software. n = 5 in each group. F–H Immunohistological analysis of osteoblasts and BMSCs in diabetic rats treated with N-Exos, DM-Exos, or Exos overexpressing miR-140-3p. Collagen-1 and CD29/Sca-1 staining of osteoblasts and BMSCs in sections containing the bone defect. Black arrows indicate osteoblasts and BMSCs on the bone surface. n = 5 in each group; *p < 0.05; **p < 0.01; ***p < 0.001. Data represent means ± SD