Limits...
Suppression of β3-integrin in mice triggers a neuropilin-1-dependent change in focal adhesion remodelling that can be targeted to block pathological angiogenesis.

Ellison TS, Atkinson SJ, Steri V, Kirkup BM, Preedy ME, Johnson RT, Ruhrberg C, Edwards DR, Schneider JG, Weilbaecher K, Robinson SD - Dis Model Mech (2015)

Bottom Line: Anti-angiogenic treatments against αvβ3-integrin fail to block tumour growth in the long term, which suggests that the tumour vasculature escapes from angiogenesis inhibition through αvβ3-integrin-independent mechanisms.The simultaneous genetic targeting of both molecules significantly impairs paxillin-1 activation and focal adhesion remodelling in endothelial cells, and therefore inhibits tumour angiogenesis and the growth of already established tumours.These findings provide a firm foundation for testing drugs against these molecules in combination to treat patients with advanced cancers.

View Article: PubMed Central - PubMed

Affiliation: School of Biological Sciences, University of East Anglia, Norwich Research Park, Norwich, NR4 7TJ, UK.

No MeSH data available.


Related in: MedlinePlus

Paxillin activation and focal adhesion disassembly are sensitive to NRP1 disruption in β3-integrin-heterozygous endothelial cells. (A) ECs of the indicated genotypes were seeded overnight on FN-coated glass coverslips. After 3 h of starvation, cells were treated ±VEGF for 10 min in serum-free medium, then fixed and immunostained for total paxillin (PXN; green) and neuropilin-1 (NRP1; red). Split-channel close-ups are shown to depict PXN/NRP1 colocalisation. Scale bar: 20 μm. The box and whisker plot shows Pearson's correlation coefficient of PXN/NRP1 colocalisation in each of the indicated genotypes in the indicated regions as determined using the ImageJ™ coloc2 plugin (means±interquartile ranges and extreme values; n≥5 cells per genotype, over ≥three independent experiments). (B) Left panel: ECs of the indicated genotype were seeded overnight on FN. They were then starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were lysed and western blotted (WB) for levels of phosphorylated (p) and total (t) PXN. GAPDH served as a loading control. Data are representative of three independent experiments. Right panel: the graph shows densitometry of pPXN relative to tPXN, as determined by WB, over an extended VEGF time course (means±s.e.m. from ≥three independent experiments). (C) ECs were seeded overnight onto FN-coated glass coverslips and stained for pPXN. ECs were starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were fixed and immunostained for pPXN (green). Scale bar: 10 μm. (D) ECs were transfected with a PXN-GFP construct and seeded at a low density on FN-coated coverslips. 72 h later, cells were starved and then treated with VEGF in reduced-serum medium. Representative cells were then imaged live (every 2 min) on an inverted fluorescence microscope for 1 h to monitor focal adhesion (FA) remodelling. The front and back ends of individual FAs were tracked over this period to measure FA assembly and disassembly, using the ImageJ™ MTrack2 plugin. The box and whisker plot shows the rate of FA assembly or disassembly for each of the indicated genotypes (means±interquartile ranges and extreme values; n≥20 FAs per genotype, from ≥two independent experiments). Asterisks indicate statistical significance: *P<0.05; **P<0.01; nsd, not significantly different. Unpaired two-tailed t-test.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4582102&req=5

DMM019927F5: Paxillin activation and focal adhesion disassembly are sensitive to NRP1 disruption in β3-integrin-heterozygous endothelial cells. (A) ECs of the indicated genotypes were seeded overnight on FN-coated glass coverslips. After 3 h of starvation, cells were treated ±VEGF for 10 min in serum-free medium, then fixed and immunostained for total paxillin (PXN; green) and neuropilin-1 (NRP1; red). Split-channel close-ups are shown to depict PXN/NRP1 colocalisation. Scale bar: 20 μm. The box and whisker plot shows Pearson's correlation coefficient of PXN/NRP1 colocalisation in each of the indicated genotypes in the indicated regions as determined using the ImageJ™ coloc2 plugin (means±interquartile ranges and extreme values; n≥5 cells per genotype, over ≥three independent experiments). (B) Left panel: ECs of the indicated genotype were seeded overnight on FN. They were then starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were lysed and western blotted (WB) for levels of phosphorylated (p) and total (t) PXN. GAPDH served as a loading control. Data are representative of three independent experiments. Right panel: the graph shows densitometry of pPXN relative to tPXN, as determined by WB, over an extended VEGF time course (means±s.e.m. from ≥three independent experiments). (C) ECs were seeded overnight onto FN-coated glass coverslips and stained for pPXN. ECs were starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were fixed and immunostained for pPXN (green). Scale bar: 10 μm. (D) ECs were transfected with a PXN-GFP construct and seeded at a low density on FN-coated coverslips. 72 h later, cells were starved and then treated with VEGF in reduced-serum medium. Representative cells were then imaged live (every 2 min) on an inverted fluorescence microscope for 1 h to monitor focal adhesion (FA) remodelling. The front and back ends of individual FAs were tracked over this period to measure FA assembly and disassembly, using the ImageJ™ MTrack2 plugin. The box and whisker plot shows the rate of FA assembly or disassembly for each of the indicated genotypes (means±interquartile ranges and extreme values; n≥20 FAs per genotype, from ≥two independent experiments). Asterisks indicate statistical significance: *P<0.05; **P<0.01; nsd, not significantly different. Unpaired two-tailed t-test.

Mentions: Therefore, as a marker of FAs, we turned our attention to examining interactions between NRP1 and PXN. As well as demarcating FAs, PXN plays an important role in EC motility and can be regulated by NRP1 (Raimondi et al., 2014). Immunocytochemistry of cells plated overnight on FN showed a predicted colocalisation of total PXN and NRP1 in non-VEGF-stimulated β3-WT ECs. This colocalisation was maintained after 10 min of VEGF-stimulation (Fig. 5A). NRP1/PXN colocalisation was also apparent in non-VEGF-stimulated β3-HET ECs, but it was lost upon VEGF-stimulation. A similar pattern was observed in β3-HET;NRP1Δcyto ECs, with β3-WT;NRP1Δcyto ECs showing only a small loss of NRP1/PXN colocalisation after VEGF-stimulation.Fig. 5.


Suppression of β3-integrin in mice triggers a neuropilin-1-dependent change in focal adhesion remodelling that can be targeted to block pathological angiogenesis.

Ellison TS, Atkinson SJ, Steri V, Kirkup BM, Preedy ME, Johnson RT, Ruhrberg C, Edwards DR, Schneider JG, Weilbaecher K, Robinson SD - Dis Model Mech (2015)

Paxillin activation and focal adhesion disassembly are sensitive to NRP1 disruption in β3-integrin-heterozygous endothelial cells. (A) ECs of the indicated genotypes were seeded overnight on FN-coated glass coverslips. After 3 h of starvation, cells were treated ±VEGF for 10 min in serum-free medium, then fixed and immunostained for total paxillin (PXN; green) and neuropilin-1 (NRP1; red). Split-channel close-ups are shown to depict PXN/NRP1 colocalisation. Scale bar: 20 μm. The box and whisker plot shows Pearson's correlation coefficient of PXN/NRP1 colocalisation in each of the indicated genotypes in the indicated regions as determined using the ImageJ™ coloc2 plugin (means±interquartile ranges and extreme values; n≥5 cells per genotype, over ≥three independent experiments). (B) Left panel: ECs of the indicated genotype were seeded overnight on FN. They were then starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were lysed and western blotted (WB) for levels of phosphorylated (p) and total (t) PXN. GAPDH served as a loading control. Data are representative of three independent experiments. Right panel: the graph shows densitometry of pPXN relative to tPXN, as determined by WB, over an extended VEGF time course (means±s.e.m. from ≥three independent experiments). (C) ECs were seeded overnight onto FN-coated glass coverslips and stained for pPXN. ECs were starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were fixed and immunostained for pPXN (green). Scale bar: 10 μm. (D) ECs were transfected with a PXN-GFP construct and seeded at a low density on FN-coated coverslips. 72 h later, cells were starved and then treated with VEGF in reduced-serum medium. Representative cells were then imaged live (every 2 min) on an inverted fluorescence microscope for 1 h to monitor focal adhesion (FA) remodelling. The front and back ends of individual FAs were tracked over this period to measure FA assembly and disassembly, using the ImageJ™ MTrack2 plugin. The box and whisker plot shows the rate of FA assembly or disassembly for each of the indicated genotypes (means±interquartile ranges and extreme values; n≥20 FAs per genotype, from ≥two independent experiments). Asterisks indicate statistical significance: *P<0.05; **P<0.01; nsd, not significantly different. Unpaired two-tailed t-test.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4582102&req=5

DMM019927F5: Paxillin activation and focal adhesion disassembly are sensitive to NRP1 disruption in β3-integrin-heterozygous endothelial cells. (A) ECs of the indicated genotypes were seeded overnight on FN-coated glass coverslips. After 3 h of starvation, cells were treated ±VEGF for 10 min in serum-free medium, then fixed and immunostained for total paxillin (PXN; green) and neuropilin-1 (NRP1; red). Split-channel close-ups are shown to depict PXN/NRP1 colocalisation. Scale bar: 20 μm. The box and whisker plot shows Pearson's correlation coefficient of PXN/NRP1 colocalisation in each of the indicated genotypes in the indicated regions as determined using the ImageJ™ coloc2 plugin (means±interquartile ranges and extreme values; n≥5 cells per genotype, over ≥three independent experiments). (B) Left panel: ECs of the indicated genotype were seeded overnight on FN. They were then starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were lysed and western blotted (WB) for levels of phosphorylated (p) and total (t) PXN. GAPDH served as a loading control. Data are representative of three independent experiments. Right panel: the graph shows densitometry of pPXN relative to tPXN, as determined by WB, over an extended VEGF time course (means±s.e.m. from ≥three independent experiments). (C) ECs were seeded overnight onto FN-coated glass coverslips and stained for pPXN. ECs were starved for 3 h and treated ±VEGF for 10 min in serum-free medium. Cells were fixed and immunostained for pPXN (green). Scale bar: 10 μm. (D) ECs were transfected with a PXN-GFP construct and seeded at a low density on FN-coated coverslips. 72 h later, cells were starved and then treated with VEGF in reduced-serum medium. Representative cells were then imaged live (every 2 min) on an inverted fluorescence microscope for 1 h to monitor focal adhesion (FA) remodelling. The front and back ends of individual FAs were tracked over this period to measure FA assembly and disassembly, using the ImageJ™ MTrack2 plugin. The box and whisker plot shows the rate of FA assembly or disassembly for each of the indicated genotypes (means±interquartile ranges and extreme values; n≥20 FAs per genotype, from ≥two independent experiments). Asterisks indicate statistical significance: *P<0.05; **P<0.01; nsd, not significantly different. Unpaired two-tailed t-test.
Mentions: Therefore, as a marker of FAs, we turned our attention to examining interactions between NRP1 and PXN. As well as demarcating FAs, PXN plays an important role in EC motility and can be regulated by NRP1 (Raimondi et al., 2014). Immunocytochemistry of cells plated overnight on FN showed a predicted colocalisation of total PXN and NRP1 in non-VEGF-stimulated β3-WT ECs. This colocalisation was maintained after 10 min of VEGF-stimulation (Fig. 5A). NRP1/PXN colocalisation was also apparent in non-VEGF-stimulated β3-HET ECs, but it was lost upon VEGF-stimulation. A similar pattern was observed in β3-HET;NRP1Δcyto ECs, with β3-WT;NRP1Δcyto ECs showing only a small loss of NRP1/PXN colocalisation after VEGF-stimulation.Fig. 5.

Bottom Line: Anti-angiogenic treatments against αvβ3-integrin fail to block tumour growth in the long term, which suggests that the tumour vasculature escapes from angiogenesis inhibition through αvβ3-integrin-independent mechanisms.The simultaneous genetic targeting of both molecules significantly impairs paxillin-1 activation and focal adhesion remodelling in endothelial cells, and therefore inhibits tumour angiogenesis and the growth of already established tumours.These findings provide a firm foundation for testing drugs against these molecules in combination to treat patients with advanced cancers.

View Article: PubMed Central - PubMed

Affiliation: School of Biological Sciences, University of East Anglia, Norwich Research Park, Norwich, NR4 7TJ, UK.

No MeSH data available.


Related in: MedlinePlus