Blood flow influences atherosclerosis by generating wall shear stress, which alters endothelial cell (EC) physiology. Low shear stress induces dedifferentiation of EC through a process termed endothelial-to-mesenchymal transition (EndMT). The mechanisms underlying shear stress-regulation of EndMT are uncertain. Here we investigated the role of the transcription factor Snail in low shear stress-induced EndMT. Studies of cultured EC exposed to flow revealed that low shear stress induced Snail expression. Using gene silencing it was demonstrated that Snail positively regulated the expression of EndMT markers (Slug, N-cadherin, α-SMA) in EC exposed to low shear stress. Gene silencing also revealed that Snail enhanced the permeability of endothelial monolayers to macromolecules by promoting EC proliferation and migration. En face staining of the murine aorta or carotid arteries modified with flow-altering cuffs demonstrated that Snail was expressed preferentially at low shear stress sites that are predisposed to atherosclerosis. Snail was also expressed in EC overlying atherosclerotic plaques in coronary arteries from patients with ischemic heart disease implying a role in human arterial disease. We conclude that Snail is an essential driver of EndMT under low shear stress conditions and may promote early atherogenesis by enhancing vascular permeability.
A correction to this article is available online at https://doi.org/10.1038/s41598-020-60955-x.
Endothelial-to-mesenchymal transition (EndMT) is characterised by multiple morphological and physiological changes including loss of cell polarity, disruption of intercellular junctions, increased proliferation, delamination and migration of cells into surrounding tissue[
Wall shear stress, imposed on the vessel lumen by the flow of blood, controls the localization of atherosclerotic lesions by regulating fundamental physiological activities in EC[
For studies of human cells, experiments were approved by the University of Sheffield Research Ethics Committee (reference SMBRER310) and all subjects gave informed consent. Studies using human cells and tissues were used in accordance to the standards set by the Declaration of Helsinki. For animal studies, all procedures were approved by the University of Sheffield ethics committee and performed in accordance with the UK Home Office Animals (Scientific Procedures) Act 1986 and in accordance with Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes.
Antibodies targeting human and murine Snail (ab180714, Abcam), human N-cadherin (ab12221, Abcam), human VE-cadherin (555661, BDPharmingen), murine CD31 (102514, Biolegend) and Ki67 (ab15580, Abcam) were obtained commercially. AlexaFluor-conjugated secondary antibodies and TO-PRO-3 were from Invitrogen.
Pig aortas from 4–6 month old animals (approximately 80 kg) were obtained immediately after slaughter from a local abattoir. They were cut longitudinally along the outer curvature to expose the lumen. EC exposed to high (outer curvature) or low (inner curvature) shear stress were harvested using collagenase (1 mg/ml for 10 minutes at room temperature) prior to gentle scraping. RNA was extracted using an RNeasy MiniKit (Qiagen) and concentrated using an RNeasy MinElute Cleanup kit (Qiagen) and the purity and integrity of total RNA samples was assessed using a Bioanalyser (Agilent).
Human umbilical vein EC (HUVEC) and porcine aortic EC (PAEC) were isolated using collagenase digestion and cultured. The purity of HUVEC was assessed by measuring the expression of VE-cadherin which was typically observed in >99% cells (Supplementary Fig. 1). It was confirmed that collagenase-isolated PAEC expressed PECAM-1 and had negligible expression of α-smooth muscle actin (SMA) or CD14 (monocyte/macrophage marker)[
Adherent endothelial cells were detached from culture flasks with trypsin, washed and resuspended in PBS containing 7% foetal calf serum. Cells were then incubated with anti-VE-cadherin antibodies conjugated to PeCy7 (clone 16B1, eBioscience) or IgG1K-PeCy7 (eBioscience) for 30 min at 4 °C. After 2 washes, fluorescence was measured using a LSRII BD Bioscience flow cytometer.
Gene silencing was performed using siRNA against Snail1 (OnTargetPlusSMARTPOOL, L-010847-01, Dharmacon) using the Lipofectamine RNAiMAX transfection system as described[
The levels of human or porcine transcripts were assessed by quantitative real time PCR (qRT-PCR) using gene-specific primers as described[
The expression levels of proteins were assessed by immunofluorescent staining using antibodies against Snail, N-cadherin or Ki67 and AlexaFluor488- or Alexafluor568-conjugated secondary antibodies. Nuclei were identified using DAPI (Sigma). Image analysis was performed using Image J software (1.49p) to calculate average fluorescence. Isotype controls or omission of the primary antibody was used to control for non-specific staining.
Wounds were created on confluent EC monolayers using a pipette tip and migration of EC into the wounded area was visualised using time-lapse confocal microscopy (Leica AF6000 Timelapse microscope) as described[
The permeability of EC monolayers exposed to flow was determined using rhodamine-labelled albumin as described[
Mice were housed under specific-pathogen free conditions. Mice with EC deletion of Twist1 (Tie2-Twist1
Coronary arteries from patients undergoing cardiac transplantation for ischaemic heart disease or dilated cardiomyopathy were obtained under the Sheffield Teaching Hospitals Institutional Review Board authorisation STH16346 (12/NW/0036). Sections made from formalin-fixed, paraffin-embedded tissues were incubated in xylene for 10 min, hydrated by sequential exposure to decreasing concentrations of ethanol (100% to 50%) and water. Heat-mediated antigen retrieval was carried out in tris-sodium citrate in a heated water bath. The sections were then blocked for endogenous peroxidase activity and incubated in 1% milk prior to overnight incubation with primary antibodies followed by HRP-conjugated secondary antibodies and substrate (VECTASTAIN Elite ABC kit; Vector laboratories and Dako). Sections were then counterstained with hematoxylin and visualized by bright field microscopy.
Differences between samples were analysed using an unpaired or paired Student's t-test or ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001).
All data generated or analysed during this study are included in this published article (and its Supplementary Information files).
To determine whether shear stress regulates the expression of mesenchymal genes we exposed cultured EC to flow. qPCR revealed that the expression of several mesenchymal markers (Snail, Slug, N-cadherin and αSMA) was elevated in HUVEC exposed to low, oscillatory shear compared to high shear stress using a parallel plate apparatus (Fig. 1a). Similarly, PAEC exposed to low shear using an orbital system expressed higher levels of Snail, Slug and N-cadherin compared to cells under high shear (Fig. 1b). Thus we conclude that low shear stress promotes mesenchymal marker expression in cultured EC. Similarly, immunofluorescent staining (Fig. 1c) and Western blotting (Fig. 1d) revealed that N-cadherin protein was expressed in a proportion of cells exposed to low shear stress but not in cells exposed to high shear.
Graph: Figure 1 Low shear stress induced mesenchymal genes via Snail. (a) HUVEC were exposed to low oscillatory (+/−4 dyn/cm2) or high (13 dyn/cm2) wall shear stress (WSS) using a parallel plate system. (b) PAEC were exposed to orbital flow to generate low (5 dyn/cm2) or high (15 dyn/cm2) wall shear stress. (a,b) After 72 h, levels of EndMT marker transcripts and VE-cadherin transcripts were quantified by qRT-PCR. The expression level at the low WSS site is presented relative to the expression at the high WSS site (normalised to 1; dotted line). Data were pooled from six independent experiments using cells from different donors and mean levels +/− SEM are shown. (c–e) HUVEC were exposed to orbital flow to generate low (5 dyn/cm2) or high (15 dyn/cm2) WSS for 72 h. (c) Expression of N-cadherin (green) and VE-cadherin (red) was determined by immunofluorescent staining and co-staining using DAPI (blue). Scale bar, 50 μm. The proportion of cells that expressed N-cadherin or VE-cadherin was measured. (d) The expression levels of N-cadherin (left) and VE-cadherin (right) were assessed by Western blotting using specific antibodies and anti-PDHX antibodies were used to control for total protein levels. Representative blots are shown. Bands were quantified by densitometry. (e) Expression of Snail (green) was determined by immunofluorescent staining and co-staining using DAPI (blue). Scale bar, 50 μm. Fluorescence intensity was quantified in multiple cells. (f) HUVEC were transfected with siRNA targeting Snail or with scrambled sequences and exposed to orbital flow for 72 h. Cells exposed to low WSS (5 dyn/cm2) were collected and transcript levels of Slug, N cadherin and α-SMA were quantified by qRT-PCR. (c–f) Data were pooled from three independent experiments using cells from different donors and mean levels +/− SEM are shown.
A full EndMT involves reduced expression of VE-cadherin, a component of EC adherens junctions that plays a central role in vascular homeostasis[
We tested the hypothesis that Snail regulates mesenchymal genes under low shear stress. Consistent with this, immunofluorescent staining revealed that Snail protein expression was significantly higher in EC exposed to low compared to high shear stress (Fig. 1e). The function of Snail was studied using Snail-targeting siRNA that reduced Snail mRNA and protein expression (Supplementary Fig. 2). Silencing of Snail significantly reduced the expression of Slug, αSMA and N-cadherin in EC exposed to low shear stress (Fig. 1f). Thus we conclude that Snail drives EndMT in response to low shear stress by inducing mesenchymal genes.
Since Snail was regulated by flow, we studied its function in sheared endothelium by gene silencing. Firstly, we studied the role of Snail in EC migration because this function is a key characteristic of EndMT. Scratch assays revealed that EC pre-conditioned under low shear stress migrated with higher velocity than cells pre-exposed to high shear (Supplementary Fig. 3a). Silencing of Snail significantly reduced migration of cells exposed to low shear stress (Fig. 2, lower panels) but did not influence cells exposed to high shear (Fig. 2, upper panels) or static conditions (Supplementary Fig. 3b).
Graph: Figure 2 Snail enhanced migration under low shear stress. HUVEC were treated with siRNA targeting Snail, or with scrambled non-targeting siRNA (or were untreated as a control). They were then cultured in 6 well plates prior to exposure to low or high wall shear stress (WSS) for 72 h using an orbital plate system. To assess cell migration, a scratch wound was made in the monolayer, and cells were imaged for 20 h. Representative images are shown (scale bar 100 μm). The distance migrated at multiple time points (lower left) and average velocity (lower right) was determined and mean values +/− SEM are shown. Data were pooled from four independent experiments using cells from different donors and mean levels +/− SEM are shown. Differences between means were assessed using a two-way ANOVA.
We next studied the influence of Snail on cell division and observed that the rate of proliferation of EC exposed to low shear stress was significantly reduced by silencing of Snail (Fig. 3a), indicating that Snail is a positive regulator of this process. By contrast, silencing of Snail did not influence proliferation of EC exposed to high shear (Fig. 3a) or static conditions (Supplementary Fig. 4). The permeability of endothelial monolayers to macromolecules was also studied because this is tightly linked to mitosis rate[
Graph: Figure 3 Snail enhanced proliferation and permeability in EC exposed to low shear stress. HUVEC were treated with siRNA targeting Snail or with scrambled (Scr) non-targeting siRNA. (a) Cells were subsequently cultured in 6 well plates prior to exposure to orbital flow to generate low or high wall shear stress (WSS) for 72 h. Cell proliferation was quantified by immunofluorescent staining using anti-Ki67 antibodies and co-staining using DAPI. Representative images are shown (Scale bar, 50 μm). The % Ki67-positive cells were calculated for multiple fields of view in four independent experiments using cells from different donors and mean levels +/− SEM are shown. Differences between means were assessed using a 2-way ANOVA. (b) Cells cultured on Transwell inserts were exposed to orbital flow (low WSS) for 72 h prior to assessment of endothelial permeability under static conditions using rhodamine (Rd)-albumin as a tracer. A schematic is shown (left panel). The concentration of Rd-albumin in the lower compartment was measured. Data were pooled from four independent experiments using cells from different donors and mean values +/− SEM are shown (right panel). Differences between means were assessed using an unpaired t-test.
Thus it was concluded that Snail influenced the physiological properties of EC exposed to low shear stress by promoting EC migration, proliferation and increased permeability to macromolecules, but it did not influence cells exposed to high shear stress or static conditions.
Since Snail was induced by low shear stress in vitro, we predicted that it would be expressed at atherosusceptible sites in the aorta. Consistent with this, qPCR analysis revealed that the expression of Snail and several of its downstream targets (Slug, N-cadherin, αSMA) was elevated in EC freshly-isolated from a low shear (inner curvature) region compared to those freshly-isolated from a high shear region (outer curvature) of the porcine aorta (Fig. 4A). By contrast, the expression of VE-cadherin was similar at low and high shear stress regions (Fig. 4a). En face staining revealed that Snail protein expression was enriched at a low shear stress site in the murine aorta (Fig. 4b). It is notable that Snail localized to the nucleus of EC at the low shear stress site suggesting that it is active under these mechanical conditions. We have previously demonstrated that Twist1 promotes EndMT and lesion development at low shear stress sites[
Graph: Figure 4 Snail was preferentially expressed at low shear atherosusceptible sites. (a) EC were freshly-isolated from low wall shear stress (WSS; inner curvature) and high WSS (outer curvature) regions of the aorta in six pigs. Levels of Snail, Slug, N-cadherin, α-SMA and VE-cadherin mRNA were quantified by qRT-PCR. The expression level at the low WSS site is presented relative to the expression at the high WSS site (normalised to 1; dotted line). Mean levels +/− SEM are shown. (b,c) EC at low WSS (susceptible) or high WSS (protected) regions of the aorta were studied by en face staining. (b) C57BL/6 mice (n = 5) were stained using anti-Snail antibodies (red), co-stained using anti-CD31 antibodies (green) and counterstained using TO-PRO-3 (DNA; blue). (c) TWIST1cKO or TWIST1fl/fl mice (n = 4 each group) were stained using anti-Snail antibodies (green) and counterstained using TO-PRO-3 (DNA; blue). Representative images (scale bar, 10 μm) and quantitation of Snail fluorescence levels (mean +/− SEM) are shown. Differences between means were assessed using a paired t-test (b) or two-way ANOVA (c).
Atherosusceptible regions of arteries are associated with altered mass transport, inflammation and other physiological features as well as low shear stress. Therefore, to examine directly whether altered shear stress can induce Snail expression in vivo we modified flow in the murine carotid artery using a constrictive cuff. This causes tapering of the lumen to generate higher shear at the stenosis and low oscillatory shear downstream[
Graph: Figure 5 Snail was induced by low oscillatory shear stress in murine carotid arteries. Flow-altering, constrictive cuffs were placed on the right carotid arteries of C57BL/6 mice. They generated anatomically distinct regions exposed to high or low oscillatory wall shear stress (WSS; as indicated). Contralateral arteries were used as an experimental control. Experimental arteries were harvested after 14 days and en face staining was performed using anti-Snail antibodies (red), anti-CD31 antibodies (green) and the nuclear counter stain TO-PRO-3 (blue). Representative images (scale bar, 10 μm) and quantitation of Snail expression (n = 3; mean +/− SEM) are shown.
Previous studies demonstrated that EC overlying human atherosclerotic plaques express mesenchymal markers[
Graph: Figure 6 Expression of Snail in endothelium overlying human coronary artery plaques. The expression of Snail in coronary arteries of patients with ischemic heart disease (IHD) or dilated cardiomyopathy (DCM) was studied by immunohistochemistry using specific primary antibodies and HRP-conjugated secondary antibodies (brown). Consecutive sections were stained for EC using anti-vWF antibodies (brown). Sections were counterstained using hematoxylin. Representative images are shown. The region boxed (upper panel, 10X magification) is shown at higher magnifications (centre and lower panels, 40X). Scale bar 20 μm. The proportion of EC that expressed Snail was calculated for each section. Data were pooled from six plaques and mean values +/− SEM are presented. Differences between means were assessed using an unpaired t-test.
In addition to its established roles in development[
Our study demonstrated that Snail was expressed at sites of endogenous low shear stress in healthy murine and porcine arteries, and was induced by the imposition of low shear stress on carotid arteries. Thus we conclude that Snail is expressed at low shear stress regions that are predisposed to atherosclerosis and may therefore contribute to vascular injury and lesion formation. To determine the effects of Snail on EC function and its potential role in vascular dysfunction we carried out gene silencing experiments in vitro. This revealed that Snail expression enhanced the permeability to macromolecules in EC exposed to low shear stress. Thus our observation implies that Snail may promote the build-up of cholesterol-rich lipoproteins and other biomolecules in the vessel wall at sites of low shear stress and has obvious implications for the initiation of atherosclerosis, which involves the accumulation of lipoproteins in the intima. Excessive mitosis of EC at atheroprone sites leads to the transient formation of intercellular gaps which enhances the permeability of endothelial monolayers to macromolecules[
Although it is established that EndMT accompanies atherosclerosis, the extent to which this process contributes to lesion initiation and progression remains uncertain. Evrard et al. implicated EndMT in plaque progression because the proportion of endothelial-derived mesenchymal cells in human plaques correlated positively with the severity of disease[
Although it leads to a profound alteration in EC behaviour and appearance, EndMT represents part of a spectrum of EC phenotypes and an example of cellular plasticity[
To assess the role of Snail in EC responses to low shear stress we silenced it by transfecting cells with specific siRNA sequences. It is plausible that the transfection process per se can alter cell physiology because it involves electroporation and addition of nucleic acids. Indeed in this study we observed that cells transfected with a non-targeting control siRNA had reduced migration compared to untransfected cells (compare Fig. 2 with Supplementary Fig. 3a), suggesting that siRNA transfection itself can modulate EC function. Nevertheless cell migration under low shear stress was consistently reduced in cells transfected with siRNA targeting Snail compared to control non-targeting sequences. Thus we concluded that silencing of Snail had an effect above and beyond that observed with control sequences indicating that Snail is a positive regulator of migration under low shear conditions.
Another important technical consideration relates to the orbital in vitro system that was used to apply flow to cultured EC. Using this system, EC monolayers cultured in 6-well plates were exposed to orbital shaking which generates relatively low shear at the centre of the well and high shear at the periphery. The regions studied were defined previously using computational fluid dynamics to calculate the shear stress distribution over the orbited well[
Although our studies of the aortic arch in pigs and mice demonstrated that Snail expression was enriched in EC located at the inner curvature they did not reveal a causal relationship between shear stress and Snail expression. To address this we modified shear stress in the murine carotid artery using a constrictive cuff developed by Krams[
M.M.M., J.S.-C., S.F., C.S., R.X., S.H., M.A. generated, analysed and interpreted data. A.M., J.C., S.E.F., K.V.D.H., V.R. interpreted data and revised the manuscript for important intellectual content. P.C.E. designed the study, interpreted data and wrote the manuscript.
The authors declare that they have no competing interests.
Graph: SUPPLEMENTAL MATERIAL
Supplementary information accompanies this paper at doi:10.1038/s41598-017-03532-z
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
By Marwa M. Mahmoud; Jovana Serbanovic-Canic; Shuang Feng; Celine Souilhol; Rouyu Xing; Sarah Hsiao; Akiko Mammoto; Jing Chen; Markus Ariaans; Sheila E. Francis; Kim Van der Heiden; Victoria Ridger and Paul C. Evans
Reported by Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author