γ-Actin plays a key role in endothelial cell motility and neovessel maintenance

Abstract

Background: Angiogenesis plays a crucial role in development, wound healing as well as tumour growth and metastasis. Although the general implication of the cytoskeleton in angiogenesis has been partially unravelled, little is known about the specific role of actin isoforms in this process. Herein, we aimed at deciphering the function of γ-actin in angiogenesis.

Methods: Localization of β- and γ-actin in vascular endothelial cells was investigated by co-immunofluorescence staining using monoclonal antibodies, followed by the functional analysis of γ-actin using siRNA. The impact of γ-actin knockdown on the random motility and morphological differentiation of endothelial cells into vascular networks was investigated by timelapse videomicroscopy while the effect on chemotaxis was assessed using modified Boyden chambers. The implication of VE-cadherin, VEGFR-2 and ROCK signalling was then examined by Western blotting and using pharmacological inhibitors.

Results: The two main cytoplasmic isoforms of actin strongly co-localized in vascular endothelial cells, albeit with some degree of spatial preference. While β-actin knockdown was not achievable without major cytotoxicity, γ-actin knockdown did not alter the viability of endothelial cells. Timelapse videomicroscopy experiments revealed that γ-actin knockdown cells were able to initiate morphological differentiation into capillary-like tubes but were unable to maintain these structures, which rapidly regressed. This vascular regression was associated with altered regulation of VE-cadherin expression. Interestingly, knocking down γ-actin expression had no effect on endothelial cell adhesion to various substrates but significantly decreased their motility and migration. This anti-migratory effect was associated with an accumulation of thick actin stress fibres, large focal adhesions and increased phosphorylation of myosin regulatory light chain, suggesting activation of the ROCK signalling pathway. Incubation with ROCK inhibitors, H-1152 and Y-27632, completely rescued the motility phenotype induced by γ-actin knockdown but only partially restored the angiogenic potential of endothelial cells.

Conclusions: Our study thus demonstrates for the first time that β-actin is essential for endothelial cell survival and γ-actin plays a crucial role in angiogenesis, through both ROCK-dependent and -independent mechanisms. This provides new insights into the role of the actin cytoskeleton in angiogenesis and may open new therapeutic avenues for the treatment of angiogenesis-related disorders.

Keywords: Actin; Angiogenesis; Cytoskeleton; ROCK signalling; Vascular endothelial cells.

Figure 1

Localization of β- and γ-actin in vascular endothelial cells. Representative photographs of HMEC-1 (left) and BMH29L (right) endothelial cells stained with β-actin (top) and γ-actin (middle) antibodies. The merged photographs (bottom) show β-actin in green, γ-actin in red and DNA (DAPI) in blue. Scale bar, 20 μm.

Figure 2

γ-actin knockdown in vascular endothelial cells. (A) Histogram showing γ-actin relative gene expression following treatment with control (white) and γ-actin siRNA (black) for 24 h as assessed by quantitative RT-PCR. β2-microglobulin was used as housekeeping gene. Columns, means of at least three individual experiments; bars, SE. Statistics were calculated by comparing γ-actin relative expression in control and γ-actin siRNA-treated HMEC-1 cells; **, p < 0.01. (B) Representative immunoblots of HMEC-1 (left) and BMH29L (right) cell lysates following treatment with control and γ-actin siRNA for 72 h. Membranes were probed with anti-β-actin, anti-γ-actin and anti-GAPDH (loading control) antibodies. (C) Histogram showing the relative protein expression γ-actin as determined by densitometry after normalization with GAPDH, following treatment with control (white) and γ-actin siRNA (black) for 72 h. Columns, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing γ-actin expression level in control and γ-actin siRNA-treated cells; ***, p < 0.001.

Figure 3

Effect of γ-actin knockdown on the localization of β- and γ-actin. Representative photographs of HMEC-1 endothelial cells treated for 72 h with control (left) or γ-actin siRNA (right) and stained with β-actin (top) and γ-actin (middle) antibodies. The merged photographs (bottom) show β-actin in green, γ-actin in red and DNA (DAPI) in blue. Scale bar, 20 μm.

Figure 4

Effect of γ-actin knockdown on the formation vascular networks in vitro. (A) Representative photographs of HMEC-1 (left) and BMH29L cells (right) incubated for 8 h on Matrigel™. Cells were treated either with control (top) or γ-actin siRNA (bottom) for 72 h. Scale bar, 250 μm. (B) Histogram showing the surface occupied by vascular networks following treatment with control (white) and γ-actin siRNA (black) for 72 h. Columns, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing the mean surface occupied by vascular networks per view field (at least 10 view fields per condition) for control siRNA- and γ-actin siRNA-treated cells. ***, p < 0.001.

Figure 5

Effect of γ-actin knockdown on VE-cadherin expression during morphological differentiation of endothelial cells into vascular networks. (A) Representative photographs of HMEC-1 cells at various time points of the morphological differentiation process on Matrigel™, following treatment with control (top) and γ-actin siRNA (bottom) for 72 h. Scale bar, 250 μm. (B) Representative immunoblots of HMEC-1 cell lysates obtained at different time points of the morphological differentiation process on Matrigel™, following treatment with control and γ-actin siRNA for 72 h. Membranes were probed with anti-VE-cadherin, anti-γ-actin and anti-GAPDH (loading control) antibodies. (C) Histogram showing the relative protein expression of VE-cadherin as determined by densitometry after normalization with GAPDH (loading control), following treatment with control (white) and γ-actin siRNA (black) for 72 h. Columns, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing VE-cadherin expression level in control and γ-actin siRNA-treated HMEC-1 cells; **, p < 0.01; ***, p < 0.001.

Figure 6

Effect of γ-actin knockdown on cell adhesion, migration and motility. (A) Histogram showing the relative adhesion of endothelial cells following treatment with control (white) and γ-actin siRNA (black) for 72 h. Fluorescently labeled HMEC-1 cells were allowed to adhere to various substrates for 1 h. Columns, means of at least three individual experiments; bars, SE. (B) Histogram showing the relative migration of endothelial cells towards FCS, VEGF, FGF_β and ECGF following treatment with control (white) and γ-actin siRNA (black) for 72 h. Fluorescently labeled HMEC-1 cells were allowed to migrate through 8 μm pore PET membrane and towards a chemo-attractant for 6 h. Columns, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing the fluorescence measured at 492/517 (Abs/Em) with control and γ-actin siRNA-treated cells; *, p < 0.05; **, p < 0.001. (C) Representative trajectories of 5 individual HMEC-1 cells, recorded by time-lapse videomicroscopy over 6 h, following treatment with control (left) and γ-actin siRNA (right) for 72 h. Scale, −130 μm to +130 μm for both x and y axes. (D) Representative photographs of control (top) and γ-actin siRNA-treated (bottom) BMH29L cells in wound healing experiments, taken 12 h after start of experiment. Broken lines show the position of the initial cell-free gap (at time 0) and solid lines highlight the position of the migration edge after 12 h. Inset, % of wound closure. (E) Graph showing the percentage of wound recovery as a function of time for control (black, solid line) and γ-actin siRNA-treated (red, broken line) BMH29L cells. Points, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing control and γ-actin siRNA-treated cells at specific time points; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 7

Effect of γ-actin knockdown on ROCK signalling. (A) Representative photographs of HMEC-1 endothelial cells treated for 72 h with control (top) or γ-actin siRNA (bottom) and co-stained with phalloidin (left) and anti-paxillin antibody. The merged photographs (right) show phalloidin in red, paxillin in green and DNA (DAPI) in blue. Inset shows a magnified view of paxillin staining in the lamellipodial region. Scale bar, 20 μm. (B-C) Scatter dot plots showing the thickness of actin stress fibres and the size of paxillin-containing adhesions in HMEC-1 endothelial cells treated for 72 h with control (o) or γ-actin siRNA (∆). Bars, means of at least 15 individual cells. (D) Representative immunoblots of HMEC-1 cell lysates following treatment with control and γ-actin siRNA for 72 h. Membranes were probed with anti-γ-actin, anti-GAPDH (loading control) and anti-phospho-myosin light chain 2 antibodies. (E) Histogram showing the relative levels of phosphorylated myosin light chain 2 in HMEC-1 cells following treatment with control (white) and γ-actin siRNA (black) for 72 h. Columns, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing control siRNA- versus γ-actin siRNA-transfected cells. **, p < 0.01; ***, p < 0.001.

Figure 8

Effect of ROCK signalling inhibition on γ-actin knockdown-induced vascular regression. (A) Graph showing the percentage of wound recovery as a function of time for control (black) and γ-actin knockdown (red) HMEC-1 cells, either untreated (left) or treated with ROCK inhibitors, H-1152 (middle) or Y-27632 (right) at 1 and 10 μM, respectively. Points, means of at least four individual experiments; bars, SE. (B) Representative photographs of HMEC-1 cells incubated for 8 h on Matrigel^TM. Cells were transfected with either control (top) or γ-actin siRNA (bottom) for 72 h and either untreated (left) or treated with ROCK inhibitors, H-1152 (middle) or Y-27632 (right) at 1 and 10 μM, respectively. Inset, % of angiogenesis inhibition as compared to untreated control cells. Scale bar, 250 μm. (C-D) Histograms showing the surface occupied by and the number of vascular networks formed by control and γ-actin siRNA-transfected HMEC-1 cells following treatment with ROCK inhibitors. Columns, means of at least four individual experiments; bars, SE. Statistics were calculated by comparing the mean surface occupied by vascular networks and the mean number of vascular networks per view field (at least 10 view fields per condition) for control siRNA- versus γ-actin siRNA-transfected cells unless indicated otherwise. *, p < 0.05; **, p < 0.01; ***, p < 0.001.