Fig 1.
Tumor development is delayed in α5(IV)-deficient mice.
(A) H&E staining of lungs of Kras/α5 WT and Kras/α5 KO mice. Arrows indicate hemorrhage lesions. (B-D) Quantification of tumor volume and numbers in H&E-stained lung sections from Kras/α5 WT (n = 6) and Kras/α5 KO (n = 6) mice. (E) Lung tumor sections were stained with anti-Ki67, anti-cleaved caspase-3, or anti-CD31 antibodies. No apoptotic signal was evident in both groups. (F and G) Quantitative proliferative indices (F) and microvascular density (MVD) (G) were measured. Data are presented as mean ± SEM. **P < 0.01, ***P < 0.001. Scale bars: 200μm.
Fig 2.
Epithelial α5(IV) supports tumor cell growth and tumorigenicity.
(A) RT-PCR analysis of α5(IV) knockdown efficiency in A549 cells. (B-D) Knockdown of α5(IV) in A549 cells significantly impaired cell proliferation (B), migration (C) and anchorage-independent cell growth (D). Data are presented as mean ± SD. (E) Growth kinetics of xenograft tumors from A549 cells expressing scramble control (shScram) (n = 5) or COL4A5 (shCOL4A5-2) (n = 6) shRNAs. (F) H&E, anti-Ki67, anti-cleaved caspase-3, or anti-CD31 immunohistochemical staining on xenograft tumor sections. No apoptotic signal was evident in both groups. (G and H) Quantitative proliferative indices (G) and microvascular density (MVD) (H) were measured. Data are presented as mean ± SEM. **P < 0.01, ***P < 0.001. NS: not significant. Scale bars: 200μm.
Fig 3.
Stromal α5(IV) is required for tumor growth and tumor angiogenesis.
(A) Growth kinetics of Lewis lung cancer cell (LLC) tumors transplanted in Col4a5 KO mice (n = 6) or the WT littermates (n = 7). (B) Anti-Ki67, anti-cleaved caspase-3 or anti-CD31 immunohistochemical staining on LLC transplanted tumor sections implanted in WT or Col4a5 KO mice. (C and D) Quantitative proliferative (C) and apoptotic (D) indices in LLC transplanted tumor sections on WT or KO mice (n = 5). (E-G) Quantitative microvascular density (MVD) (E), sinusoid microvessel number (VN) (F) and vascular diameter (G) in LLC tumor sections transplanted in WT or Col4a5 KO mice (n = 5). (H) In vivo Matrigel plug assay in 8-week-old WT or Col4a5 KO mice (n = 5). Dextran-FITC was injected through the tail vein to visualize the penetrating blood vessels (middle panels). Matrigel plugs were removed and fixed for CD31 staining (right panels). (I) Quantitative microvascular density (MVD) in Matrigel plugs implanted in WT or Col4a5 KO mice (n = 5). Data are presented as mean ± SEM. **P < 0.01, ***P < 0.001. Scale bars: 200 μm.
Fig 4.
α5(IV) is expressed in endothelial cells and required for endothelial cell proliferation and tubulogenesis.
(A) Immunofluorescent staining on lung sections shows partial colocalization of α5(IV) and CD31. Scale bar: 25μm. (B) RT-PCR analysis of α5(IV) knockdown efficiency in HMEC-1 cells. (C and D) Knockdown of α5(IV) impaired proliferation (C) and migration (D) of HMEC-1 cells, determined by BrdU incorporation and modified Boyden chamber assays, respectively. Data are presented as mean ± SD. ***P < 0.001. (E) In vitro tubulogenesis of HMEC-1 cells expressing control (shScram) or COL4A5 (shCOL4A5) shRNAs. Scale bar: 200 μm.
Fig 5.
α5(IV), but not α1(IV), deficiency results in impaired activation of ERK.
(A) Western blot analyses of phosphorylation levels of ERK, Akt, FAK and Src and α1(IV), α2(IV) and α5(IV) expression in α5(IV)- or α1(IV)-knockdown A549 cells. (B) Western blot analyses of phosphorylation levels of ERK in α5(IV)-knockdown A549 cells expressing constitutively active MEK1. (C-E) Expression of constitutively active MEK1 in α5(IV)-knockdown A549 cells rescued the defects in cell proliferation (C), migration (D), and anchorage-independent cell growth (E). Data are presented as mean ± SD. **P < 0.01.
Fig 6.
DDR1 is downregulated in α5(IV)-knockdown cells.
(A) Western blot analyses of DDR1 expression in α5(IV)- or α1(IV)-knockdown A549 cells. (B and C) DDR1 staining on lung sections from Col4a5+/Y and Col4a5LacZ/Y mice (B) or lung tumor sections from Kras/α5 WT and Kras/α5 KO mice (C). Scale bars: 200 μm. (D) Quantitative RT-PCR analysis of DDR1 expression in A549 cells expressing control (shScram) or COL4A5 (shCOL4A5) shRNAs. (E and F) A549 cells expressing control (shScram) or COL4A5 (shCOL4A5) shRNAs were treated with or without 100 μg/mL cycloheximide (CHX) for 0.5, 1, 2 or 4 hours. (E) DDR1 protein levels were analyzed by western blot. (F) Relative protein levels of DDR1 after cycloheximide treatment in (E). (G) Western blot analysis of DDR1 expression in A549 cells expressing control (shScram) or COL4A5 (shCOL4A5) shRNAs treated with or without 10 μM proteasome inhibitor MG132 for 4 hours or 50 mM lysosome inhibitor NH4Cl for 6 hours. (H) A549 cells expressing control (shScram) or COL4A5 (shCOL4A5) shRNAs were treated with 10 μM proteasome inhibitor MG132 for 4 hours. DDR1 was immunoprecipitated and ubiquitination levels were detected.
Fig 7.
DDR1 is required for ERK activation, cell proliferation and migration in lung cancer cells.
(A) Western blot analyses of phosphorylation levels of ERK, Akt, FAK and Src in A549 cells with DDR1 knockdown. (B-D) Knockdown of DDR1 in A549 cells significantly impaired cell proliferation (B), migration (C), and anchorage-independent cell growth (D). Data are presented as mean ± SD. **P < 0.01, ***P < 0.001.
Fig 8.
Constitutively active DDR1 rescued the defects of α5(IV)-deficient lung cancer cells.
(A) Full-length wild-type DDR1 was expressed in control or α5(IV)-knockdown A549 cells. DDR1 was immunoprecipitated and tyrosine phosphorylation levels were detected. (B) Div-DDR1 chimeric proteins were expressed in α5(IV)-knockdown A549 cells. DDR1 expression and tyrosine phosphorylation levels were detected. Div-DDR1 with mutations in the Div coil-coiled domain (mDiv-DDR1) or kinase domain (Div-DDR1 K655A) failed to activate and autophosphorylate. (C) Western blot analyses of phosphorylation levels of ERK, Akt, FAK and Src in α5(IV)-knockdown A549 cells expressing Div-DDR1 chimeric proteins. (D-F) Expression of Div-DDR1, but not mDiv-DDR1 or Div-DDR1 K655A in α5(IV)-knockdown A549 cells rescued the defects in cell proliferation (D), migration (E), and anchorage-independent cell growth (F). Data are presented as mean ± SD. ***P < 0.001. NS: not significant.