SNAI1, an endothelial–mesenchymal transition transcription factor, promotes the early phase of ocular neovascularization
Ocular neovascularization is a comprehensive process involved in retinal vascular development and several blinding diseases such as age-related macular degeneration and retinopathy of prematurity, with vascular endothelial growth factor (VEGF) regarded as the master regulator. However, the qualified effect of anti-VEGF therapy reveals that the underlying mechanisms are still not clearly identified. To initialize angiogenesis, endothelial cells undergo a phenotype switching to generate highly migratory and invasive cells. This process shares certain similar characters observed in endothelial–mesenchymal transition (EndMT). Here, we found that SNAI1, an EndMT transcription factor, was expressed by endothelial cells in both physiological and pathological ocular neovascularization. SNAI1 overexpression triggered cell morphological change and enhanced cell motility, while loss of SNAI1 attenuated migration, invasion and sprouting. RNA sequence analysis further revealed that SNAI1 knockdown decreased the expression of genes related to cytoskeleton rearrangement and ECM remodeling. Moreover, intravitreal injection of small interfering RNA of SNAI1 suppressed new vessel formation in developing retina as well as mice model of choroidal neovascularization and oxygen-induced retinopathy. Therefore, we propose that the EndMT transcription factor SNAI1 promotes the early phase of ocular neovascularization and may provide a potential therapeutic target.
KeywordsAngiogenesis SNAI1 Choroidal neovascularization Oxygen-induced retinopathy Endothelial–mesenchymal transition
Ocular angiogenesis is a highly coordinated process involved in retinal vasculature development and several ocular diseases such as age-related macular degeneration (AMD), proliferative diabetic retinopathy (PDR) and retinopathy of prematurity (ROP) [1, 2]. During neovascularization, sprouting endothelial cells (ECs) degrade the basement membrane (BM) with a series of diffusing and membrane-bound proteolytic enzymes, which are mainly involved in two classes: plasminogen activator (PA)/plasmin system and matrix metalloproteinases (MMPs) . Then, ECs begin to proliferate, invade and migrate along components of the extracellular matrix (ECM) and take part in the formation of an immature capillary structure and deposition of a new complex BM . This process involves complex and highly dynamic interactions between ECs and their environment . Though extensive researches have suggested that multiple factors participate into this growth program, with vascular endothelial growth factor (VEGF) signaling pathway emerging as a master regulator [6, 7], the underlying cellular and molecular processes still need to be fully identified.
Endothelial cell plasticity is critical for development and multiple biological processes including angiogenesis . During angiogenesis, activated ECs act more invasively compared to quiescent phase, including formation of lamellipodia and filopodia, degradation of BM, migration into ECM and participation in ECM remodeling. These functional changes share several similarities with endothelial–mesenchymal transition (EndMT), in which process ECs undergo a phenotype switching to generate elongated, spindle-shaped mesenchymal-like cells which are highly migratory and invasive [9, 10, 11]. These common points implied a possible role for EndMT transcription factors in angiogenesis.
The recognized EndMT transcription factors include the zinc-finger binding transcription factors Snail family and several basic helix-loop-helix (bHLH) factors such as ZEB1, ZEB2 and TWIST . Among them, the Snail family plays the most critical role. In vertebrates, the Snail family contains three members, characterized up to now, which are SNAI1 (Snail), SNAI2 (Slug) and SNAI3 . SNAI1 and SNAI2 were reported to trigger EndMT under certain conditions [14, 15, 16, 17, 18]. Apart from its functions in EndMT, evidences have indicated that Snail family had a role in angiogenesis. ECs-specific deletion of SNAI1 was embryonic lethal, while inducible SNAI1 loss of function caused vessel defects . Both SNAI1 and SNAI2 were reported to be increased in angiogenic ECs in vitro, and decreased as lumen stabilizing, which suggested their roles in initializing angiogenesis [20, 21]. Therefore, it would be intriguing to investigate if Snail family members have a role in ocular angiogenesis. Previous expression array data showed that SNAI1 was abundant in retina among the three family members . Thus, in this study, we focused on SNAI1 expression and function during physiological and pathological ocular angiogenesis. We further tested its function under VEGF stimulus and evaluated its therapeutic potential as well.
Materials and Methods
Pups and 8–10-week-old male C57/BL6 mice were maintained under specific pathogen-free conditions. All animal experiments were approved and followed the guidelines issued by the Animal Experiment Administration Committee of the Fourth Military Medical University and also complied with the ARVO Statement.
Small interfering RNA (siRNA) of mouse SNAI1 (siSNAI1) for in vivo studies was selected among 3 sets (RiboBio) and manufactured as in vivo modified STABLE siSNAI1 by RiboBio. 5 nmol Cy3-labeled negative control (NC) or siSNAI1 was dissolved in 25 μL RNA-free phosphate-buffered saline (PBS) to prepare working solution.
CNV model was induced by laser photocoagulation as described before [23, 24]. Briefly, mice were anesthetized with pupils dilated. Laser photocoagulation was performed on both eyes of mice using a 532 nm laser (Iris Radiation Systems, USA; spot size 75 μm, power 100 mw, duration 0.1 s) through a contact lens. Five laser spots were made uniformly between the major retinal vessels, 1.5–2 disk diameters from the optical disk. Disruption of Bruch’s membrane was confirmed when a cavitation bubble appeared without hemorrhage. Modified NC or siSNAI1 were injected intravitreally at 0.1 nmol in either eye of adult mice on day 1 (d1) after photocoagulation if needed. Mice were killed at d7.
Oxygen-induced retinopathy (OIR) model was induced as described before . The pups and their mothers were placed in a high-oxygen chamber (75 ± 2%) by post-natal day 7 (P7) for 5 days and then returned to room air (normoxic) conditions. Modified NC or siSNAI1 were injected intravitreally at 0.1 nmol in either eye of pups on P12 if needed. Pups were killed at P17.
Cell culture, adenovirus infection and siRNA transfection
Human umbilical vein endothelial cells (HUVECs) were isolated from human umbilical cord biopsies, which were obtained from the Department of Gynecology and Obstetrics of Xijing Hospital, Fourth Military Medical University, by collagenase treatment. Informed consent was obtained from all individual participants included in the study. Cells at passages 2–6 were used and cultured in EC medium (ECM) (ScienCell) supplemented with 5% FBS, EC growth supplements (ECGS), 100 U/mL penicillin and 100 μg/mL streptomycin in a humidified atmosphere with 5% CO2 at 37 °C. For cytokine stimuli, HUVECs were starved in ECM with 0.5% FBS and without ECGS for 24 hours (24 h) and then treated with VEGF-165 (Sino Biological) at 25, 50 or 100 ng/mL for 6 h. Cells were treated with 20 μM cyclohexane (CHX) or 50 μM MG132 if necessary. MRC5 and bEnd.3 cells (ATCC) were cultured in Dulbecco’s modified eagle medium (DMEM) (Gibco) containing 10% FBS.
HUVECs were infected with SNAI1 overexpression adenovirus (Ad-SNAI1) and GFP-labeled ctrl adenovirus (Ad-ctrl) (Vigene) at a multiplicity of infection (MOI) of 50 according to the operation procedures. HUVECs or bEnd.3 cells were transfected with NC or siSNAI1 at a concentration of 50 nM using Lipofectamine 2000 reagent (Invitrogen), according to the manufacturer’s instructions. siRNA for human SNAI1 (RiboBio) was selected from 2 sets.
Retinal flat mounts were performed on P7, P14 or P17 for retinal vascular development analysis, while choroidal flat mounts on d7 for CNV analysis, as described previously [23, 25, 26]. Retinal and RPE–choroid–sclera complex tissues were post-fixed in 4% paraformaldehyde (PFA) for 6 h at 4 °C and then blocked and permeabilized in PBS containing 1% bull serum albumin (BSA) and 0.5% Triton X-100 overnight. Isolectin B4 (IB4) (1:100, FL-1201 or HL-1102, Vector) and anti-SNAI1 antibody (1:80, MABE167, Millipore or 1:200, ab180714, Abcam) in PBS containing 1% BSA and 0.5% Triton X-100 were incubated overnight at 4 °C, followed by incubation with the following secondary antibodies in PBS overnight at 4 °C: Alexa Fluor 488-conjugated goat anti-mouse IgM mu chain secondary antibody (1:400, ab150121, Abcam) or DyLight 488-conjugated goat anti-rabbit IgG(H + L) secondary antibody (1:400, A-23220, Abbkine). Each step was followed by three washes in PBS for 10 minutes (10 min). The retinal or RPE–choroid–sclera complex tissues were flat-mounted on a glass slide and then covered under a dissecting microscope (Olympus). All flat mounts were examined and photographed under a confocal laser scanning microscopy (FV1000, Olympus).
Frozen sections of eyeballs were made as described before . Sections were dried for 2 h at room temperature and then washed in PBS and blocked with PBS containing 1% BSA and 0.5% Triton X-100 at room temperature for 1 h. Slides were incubated with primary antibodies overnight at 4 °C and secondary antibodies for 2 h at room temperature and then counterstained with Hoechst (94403, Sigma-Aldrich) for 15 min at room temperature. Each step was followed by three washes in PBS for 10 min. Slides were closed with coverslips. Primary antibodies included anti-CD31 antibody (1:400, 102502, BioLegend) and anti-SNAI1 antibody (1:80, MABE167, Millipore or 1:200, ab180714, Abcam). Secondary antibodies were Alexa Fluor 594-conjugated donkey anti-Rat IgG (H + L) secondary antibody (1:400, A-21209, Invitrogen), Alexa Fluor 488-conjugated goat anti-mouse IgM mu chain secondary antibody (1:400, ab150121, Abcam) and DyLight 488-conjugated goat anti-rabbit IgG (H + L) secondary antibody (1:400, A-23220, Abbkine). Nuclei were stained with Hoechst (94403, Sigma-Aldrich). All photographs were taken under a confocal laser scanning microscopy (FV1000, Olympus).
Cells were fixed with 4% PFA for 30 min and blocked with PBS containing 1% BSA and 0.5% Triton X-100 for 30 min at room temperature. F-actin was stained with FITC-conjugated Phalloidine (Yeasen), and nuclei were stained with Hoechst (94403, Sigma-Aldrich). Each step was followed by three washes in PBS for 10 min. Slides were covered and analyzed with a fluorescence microscope (BX51, Olympus).
Quantitative reverse transcription polymerase chain reaction (qRT-PCR)
Primer sequences used in the study
Mouse β-actin forward
Mouse β-actin reverse
Mouse SNAI1 forward
Mouse SNAI1 reverse
Human β-actin forward
Human β-actin reverse
Human SNAI1 forward
Human SNAI1 reverse
Human MMP2 forward
Human MMP2 reverse
Human MMP9 forward
Human MMP9 reverse
Human FURIN forward
Human FURIN reverse
Human CTSK forward
Human CTSK reverse
Human MMP10 forward
Human MMP10 reverse
Human CTSS forward
Human CTSS reverse
Human FN forward
Human FN reverse
Human LAMA4 forward
Human LAMA4 reverse
Human LAMB1 forward
Human LAMB1 reverse
Human LAMC1 forward
Human LAMC1 reverse
Human COL4A1 forward
Human COL4A1 reverse
Human COL4A2 forward
Human COL4A2 reverse
Cells were harvested in radio immunoprecipitation assay (RIPA) buffer (Beyotime) containing phenylmethylsulfonyl fluoride (PMSF) (Sigma-Aldrich) and then separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). Immunoblotting was performed with primary antibodies including: anti-SNAI1 antibody (1:1000, 3879, Cell Signaling Technology), anti-MMP2 antibody (1:1000, 10373-2-AP, Proteintech), anti-MMP9 antibody (1:1000, 10375-2-AP, Proteintech), anti-FURIN antibody (1:1000, 18413-1-AP, Proteintech), anti-CTSK antibody (1:1000, 11239-1-AP, Proteintech), anti-CTSS antibody (1:1000, ab134157, Abcam), anti-FN antibody (1:1000, 15613-1-AP, Proteintech), anti LAMB1 antibody (1:1000, 23498-1-AP, Proteintech), anti COL IV antibody (1:1000, 55131-1-AP, Proteintech) and anti-β-ACTIN antibody (1:5000, A5316, Sigma-Aldrich) on polyvinylidene fluoride (PVDF) membranes followed by incubation of horseradish peroxidase (HRP)-linked goat anti-rabbit IgG or horse anti-mouse IgG secondary antibody (1:2000, 7074 or 7076, Cell Signaling Technology). Each step was followed by three washes in PBS containing 0.1% Tween-20 (PBST) for 10 min. Membranes were observed using an enhanced chemoluminescence system (Clinx Science Instruments).
Cell proliferation assay
EdU incorporation was performed as described before . HUVECs were seeded on 48-well plates and incubated with 50 μM EdU (5-ethynyl-2′-deoxyuridine) (RiboBio) in complete ECM for 2 h and then fixed with 4% PFA at room temperature for 30 min, followed by staining with Apollo 567 and Hoechst according to manufacturer’s instructions. Images were captured under a fluorescence microscope (BX51, Olympus).
Cell migration assay
Cell migration ability was tested by wound healing assay as described before . In brief, a scratch was created with a p20 pipette tip, and the medium was then replaced by ECM supplemented with 0.5% FBS and without ECGS. Wound closure was observed under a microscope (CKX41, Olympus) with a CCD camera (DP70, Olympus) and measured at 12 h after the scratch was made.
Cell invasion assay
Cell invasion ability was observed by spheroid collagen type I invasion assay and tested by Transwell invasion assay based on published protocol [29, 30]. For spheroid collagen type I invasion assay, HUVECs at a total number of 5.0 × 104 in 100 μL EGM-2 medium (Lonza) were seeded on 96-well ultra-low attachment surface plates (Corning) for 2 days to generate cell spheres. Collagen type I (C8062, Solarbio) solution with a final concentration of 1 mg/mL was prepared according to the manufacturer’s instructions. The spheres were then embedded in the collagen gel, and 100 μL complete ECM with or without additional VEGF at 50 ng/mL was added. Invasion of spheres was observed the next day under a microscope (CKX41, Olympus) with a CCD camera (DP70, Olympus). For Transwell invasion assay, the upper chamber of Transwell chamber (Corning) was pre-coated with 100 μL Matrigel Basement Membrane Matrix (matrigel) (1:20, BD Biosciences) overnight at 37 °C and then cultivated with HUVECs at a total number of 1.0 × 105 in 200 μl ECM containing 0.5% FBS without ECGS. 500 μL complete ECM with 50 ng/ml VEGF-165 (Sino Biological) was added to the lower chamber. Cells were cultured for 12 h and were fixed by 4% PFA and stained with crystal violet. Number of cells invading to the lower chamber was observed under a microscope (CKX41; Olympus) with a CCD camera (DP70; Olympus).
Tube formation assay
Assessment of tube formation ability of HUVECs was performed as described previously . HUVECs (1 × 105 cells per well) were loaded onto 200 μL matrigel (1:1, BD Biosciences) which was pre-coated in 48-well plates and incubated at 37 °C for 4 h. The number of loops and total length of cell cords in the enclosed lumen structures were determined.
Fibrin beads assay
A fibrin beads sprouting assay was conducted, as described previously, using a fibrin beads assay kit (Amersham-Pharmacia Biotech) according to the supplier’s instructions [31, 32]. Briefly, HUVECs were transfected and changed medium into EGM-2 medium (Lonza) the day before beading. Cytodex 3 microcarrier beads (Amersham) were incubated with HUVECs (400 cells per bead) at 37 °C for 4 h and then transferred to 12-well plates and left overnight. The beads were embedded in fibrinogen (Sigma-Aldrich) containing 0.625 U/mL thrombin (Sigma-Aldrich) at a density of 100 beads/mL in a 48-well plate the next day, and MRC5 cells (2 × 105 cells per well) in 0.5 mL EGM-2 medium were added. The medium was changed every other day. Sprouting was quantified by measuring the number and length of sprouts.
Aortic ring sprouting assay
Aortic ring assay was performed following previous protocol . In brief, aortas were removed from adult mice, cleaned and transfected with 200 nM siSNAI1 or NC using Lipofectamine 2000 reagent (Invitrogen) overnight. The aortas were then embedded in 80 μL matrigel in a 48-well plate containing 500 μL opti-MEM Reduced Serum Media (Gibco) supplemented with 2.5% FBS and 30 ng/mL VEGF-164 (Sino Biological). Endothelial sprouts were allowed to grow over 4 days with medium changing every other day. The number and length of sprouts were measured.
RNA sequence analysis
HUVECs were transfected with siSNAI1 or NC as mentioned above and cultured in complete ECM for 24 h. Cells were then starved in ECM with 0.5% FBS and without ECGS for 24 h following VEGF-165 stimulus at 50 ng/mL for 6 h. RNA was extracted, sequenced and analyzed by RiboBio. In brief, total RNA was isolated from cells using TRIzol reagent (Invitrogen) according to the manufacturer’s protocol. RNA purity was assessed using the ND-1000 Nanodrop. Each RNA sample had an A260:A280 ratio above 1.8 and A260:A230 ratio above 2.0. RNA integrity was evaluated using the Agilent 2200 TapeStation (Agilent Technologies), and each sample had the RINe above 7.0. mRNAs were isolated from total RNA and fragmented to approximately 200 bp and subsequently subjected to first strand and second strand cDNA synthesis following by adaptor ligation and enrichment with a low-cycle according to instructions of TruSeq® RNA LT/HT Sample Prep Kit (Illumina). The purified library products were evaluated using the Agilent 2200 TapeStation and Qubit®2.0 (Life Technologies) and then diluted to 10 pmol/L for cluster generation in situ on the HiSeq 3000 pair-end flow cell followed by sequencing (2 × 150 bp) on HiSeq 3000. Prior to sequencing, the raw data were filtered to produce high-quality clean data. All the subsequent analyses were performed using the clean data.
The reference genome and gene model annotation files were downloaded directly from the NCBI library. The clean reads were aligned to the reference genome using TopHat (Broad Institute). During read alignment, a maximum of two mismatches and a gap length of 2 bps were allowed. The functional annotations of genetic variants were generated using ANNOVAR . Read alignments were processed by Cufflinks software package  to determine the differential expression of genes. Gene expression values were quantified as reads per kilobase of transcript per million mapped reads (RPKM). Original data were deposited in the Gene Expression Omnibus (GEO) database (accession # GSE111386).
Retinal vascular development, CNV and OIR lesion assessment
The assessment of retinal vessels were based on published protocol . Digitized whole retina images were reconstructed using Photoshop CS6 software (Adobe Systems). Average distance from optic disk to the edge of vasculature was assessed as the overall situation of vascular development. For vasculature density, capillary plexus area situated between an artery and a vein was selected and branching points were defined as junctions of capillary segments. Tip cells were counted per endothelial vessel length of 1000 μm in the angiogenic front. For each time point, six pups were treated in two independent experiments and four to six fields per retina were analyzed.
CNV lesion was analyzed as described formerly [23, 24]. Briefly, horizontal optical sections. (2 μm step) were obtained under a confocal microscopy from flat-mount RPE–choroid–sclera complex. The deepest focal plane in which the surrounding choroidal vascular network connecting to the lesion could be identified was judged to be the floor of the lesion. Any IB4 positive vessels in the laser-treated area and superficial to this reference plane was judged as CNV. Images of each section were digitally stored, followed three-dimensional (3D) reconstruction by IMARIS software (Bitplane). The summation of whole fluorescent area in each horizontal section was used as an index for the volume of CNV. Six mice were treated in two independent experiments, and five CNV spots per eye were analyzed.
OIR lesion was evaluated according to the proposed protocol . The neovascular (NV) and total areas were manually outlined and quantified by a masked observer, and the ratio of NV to total retinal area was then calculated. Three pups were analyzed.
Statistical analysis was performed with the Image-Pro Plus 6.0 and GraphPad Prism 5 software. All quantitative data were presented as mean ± SEM. Statistical significance was calculated using unpaired or paired Student’s t test. P < 0.05 was considered as significant.
SNAI1 was expressed in sprouting vessels in both physiological and pathological ocular neovascularization
SNAI1 overexpression improved cell motility and invasion in vitro
VEGF increased SNAI1 expression and generated invasive cells
Loss of SNAI1 inhibited angiogenesis due to repressive cell motility and invasion in vitro
RNA sequence analysis showed that SNAI1 knockdown decreased genes involved in cytoskeleton arrangement, ECM degrading and remodeling
SNAI1 knockdown by siSNAI1 intravitreal injection suppressed ocular neovascularization in vivo
Findings from this study showed that (a) endothelial cells could express SNAI1 in both physiological and pathological ocular neovascularization; (b) SNAI1 expression, partly by VEGF stimulus, improved ECs motility and invasion; and (c) intravitreal injection of siSNAI1 suppressed new vessel formation in developing retina, CNV and OIR model. Taken together, our data suggested that SNAI1, an EndMT transcription factor, promoted the early phase of ocular neovascularization and might provide a potential therapeutic target.
Ocular neovascularization is orchestrated by complicated cellular interactions and vascular growth factors. Above all, VEGF, due to its role in endothelial proliferation, survival and migration, has proved to be the master regulator . Elevated intraocular levels of VEGF have been reported within the setting of iris neovascularization, retinal vein occlusion (RVO), PDR, ROP and CNV, which led to the clinical development of a number of approved anti-VEGF therapies . Activation of ECs acts as the initial stage of angiogenesis. Responding to angiogenic stimuli, ECs undergo a compound morphogenetic process including polarization, lamellipodia and filopodia formation, cell–cell and cell–matrix contact rebuilding and ECM degradation and therefore migrate through the BM to form new vessels. The series of activities generate highly invasive phenotype, which have resemblance to EndMT process. Accordingly, it has been proposed that angiogenesis may involve a “partial EndMT”, as ECs display a fundamental transition in both morphology and functions but remain most endothelial characteristics . However, much more are needed to elucidate the correlation between angiogenesis and EndMT.
As a comparison of angiogenesis and EndMT reveals several similarities, it seems reasonable to assume that factors inducing EndMT may also play a role in ocular angiogenesis. First of all, we observed that SNAI1 was expressed in ECs in the early phase of physiological and pathological ocular angiogenesis. In retinal development model, SNAI1 mRNA level was highly up-regulated at P7 and remained high level till P14. The expression of SNAI1 in veins, venules and descending ECs was further revealed by immunofluorescence, which was consistent with previous report . In pathological angiogenesis, newly growing ECs also expressed SNAI1 in CNV and OIR model. We then found that both SNAI1 mRNA and protein level increased responding to VEGF stimulus. As SNAI1 is highly unstable and degraded by proteasome via phosphorylation by GSK-3β , we further tested whether this up-regulation was resulted from increased synthesis or decreased degradation by adding proteasome inhibitor MG132 or protein synthesis inhibitor CHX. Our data showed that VEGF stimulus promoted SNAI1 protein synthesis rather than restrained it degradation.
It was reported that SNAI1 could induce EndMT under certain conditions [14, 15, 17, 39, 40, 41, 42]. Besides, SNAI1 also had a role in embryonic vessel formation and in vitro angiogenesis [19, 20]. Recently, Park et al. have reported that ECM-induced SNAI1 expression contributed to vertical vessel sprouting via the regulation of VEGFR3 . However, more details about how SNAI1 functions in ocular neovascularization need to be elucidated. Here, by using SNAI1 overexpression adenovirus and SNAI1 siRNA, we found that SNAI1 expression in ECs mainly affected cell motility, migration and invasion. These activities are coincident with properties involving in EndMT. SNAI1 knockdown also partly blocked ECs response to VEGF stimulus. To our surprise, in lumen formation assay, total length of siSNAI1-treated HUVECs slightly increased compared with control. Previous data showed that SNAI1 expression rose up temporarily during ECs network formation and disappeared when it was completed . As the complicated mechanisms involved in ECs lumen formation in vitro, this result indicated, though need more support, that down-regulation of SNAI1 may stabilize vessel structure.
The downstream targets of Snail family in ECs are barely studied. Several studies pointed out that delta like canonical Notch ligand 4 (Dll4), MMP15, membrane type 1-matrix metalloproteinase (MT1-MMP) and VEGFR3 may mediated part of SNAI1 or SNAI2 function in ECs [14, 21, 42, 43]. Here, we performed RNA sequence and put forward that genes involving in cytoskeleton arrangement, ECM degradation and remodeling might account for the phenotype changes in SNAI1 activation. Firstly, ECs require remodeling of the F-actin cytoskeleton in order to migrate into perivascular areas. Previous studies have shown that Rho family GTPase could be activated by VEGF to generate protrusions in leading endothelial tip cells that guide newly formed sprouts [44, 45]. The best studied Rho family GTPase are Rho, Rac and Cdc42 . Here, we detected a decreased enrichment of genes involved in Rho and Rac1 pathway, which indicated that SNAI1 down-regulation impaired cell motility. Secondly, during angiogenic sprouting, the basal lamina is degraded and replaced by new matrix molecules like laminin, fibronectin, and collagen type IV . The former process required a series of diffusing and membrane-bound proteolytic enzymes, which mainly fell into two classes: PA and MMPs . Here, the decrease in genes related to ECM disassembly including MMP2, MMP9, FURIN, CTSK, MMP10 and CTSS accounted for the impaired migration, invasion and sprouting after SNAI1 knockdown. Besides, we noticed that genes involved in ECM component, including FN, LAMB1, LAMC1, COL4A1 and COL4A2 were also decreased in siSNAI1-treated HUVECs, which indicated that SNAI1 also played a role in following ECM synthesis. Therefore, our data inferred that SNAI1 up-regulation in activated ECs mainly promotes genes related to cytoskeleton arrangement and ECM remodeling. These changes on gene expression enabled ECs to form prominence, invade through the ECM and deposit new BM and might result in the observed SNAI1-induced pro-angiogenic characters.
The development of ocular new vessels is companied by ECs sprouting and invasion. In developing retina, deep retinal vascular plexus emerge from venous vessels and penetrate the retina perpendicular to the primary plexus from P7 . In CNV, choroidal ECs break through the Bruch’s membrane and grow into the avascular subretinal space . In OIR, after transferred from hyperoxic to normoxic conditions, the retinal vessels are dilated and tufts of pathological neovasculature grows on the surface of the retina . Studies once revealed that systemically inhibition of SNAI1 limited retinal deep vascular formation without impacting superficial vessels . Therefore, we speculated that SNAI1 could provide a promising therapeutic target for the following reasons: (a) based on our results, activated ECs expressed SNAI1, and inhibition of SNAI1 suppressed angiogenesis in vitro; (b) VEGF was the master regulator in ocular neovascularization and SNAI1 suppression limited VEGF response; and (c) SNAI1 is quickly degraded and retain silence in mature vasculature, and thus, its inhibition should not cause undesirable side effects [17, 20]. To test this hypothesis, we performed siSNAI1 intravitreal injection in both pups and adult mice. We found that local SNAI1 inhibition could significantly reduce retinal deep sprouting as well as OIR and CNV formation. In summary, our data suggested a crucial role of SNAI1 in ocular angiogenesis initialization, which might partly attribute to its promotion on genes related to cell motility and ECM remodeling under VEGF stimulus, and thus, SNAI1 could be regarded as a novel target for ocular neovascularization diseases.
We thank Ms. Qing-Yi Wang for her kind help on correcting the galley proof.
JXS, GRD, YSW and HH designed experiments. JXS and TFC performed most of the studies and analyzed data. MHL, LJS and YL carried out CNV model. WQX and JBS carried out OIR model. XCY, ZYY and YL helped with in vitro experiments. LL and HH provided reagents and advised on experimental design. JXS wrote manuscript. GRD and YSW revised manuscript.
This study was funded by National Natural Science Foundation of China (Nos. 81770936, 81670863), Youth Science and Technology Nova Program of Shaanxi Province (No. 2016KJXX-19) and Foundation for the Author of National Excellent Doctoral Dissertation of the People’s Republic of China (No. 201481).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 12.Gonzalez DM, Medici D (2014) Signaling mechanisms of the epithelial–mesenchymal transition. Sci Signal 7(344):re8. https://doi.org/10.1126/scisignal.2005189
- 14.Lee SW, Won JY, Kim WJ, Lee J, Kim KH, Youn SW, Kim JY, Lee EJ, Kim YJ, Kim KW, Kim HS (2013) Snail as a potential target molecule in cardiac fibrosis: paracrine action of endothelial cells on fibroblasts through snail and CTGF axis. Mol Ther J Am Soc Gene Ther 21(9):1767–1777. https://doi.org/10.1038/mt.2013.146 CrossRefGoogle Scholar
- 16.Cano A, Perez-Moreno MA, Rodrigo I, Locascio A, Blanco MJ, del Barrio MG, Portillo F, Nieto MA (2000) The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol 2(2):76–83. https://doi.org/10.1038/35000025 CrossRefPubMedGoogle Scholar
- 18.Cobaleda C, Perez-Caro M, Vicente-Duenas C, Sanchez-Garcia I (2007) Function of the zinc-finger transcription factor SNAI2 in cancer and development. Annu Rev Genet 41:41–61. https://doi.org/10.1146/annurev.genet.41.110306.130146 CrossRefPubMedGoogle Scholar
- 20.Welch-Reardon KM, Ehsan SM, Wang K, Wu N, Newman AC, Romero-Lopez M, Fong AH, George SC, Edwards RA, Hughes CC (2014) Angiogenic sprouting is regulated by endothelial cell expression of Slug. J Cell Sci 127(Pt 9):2017–2028. https://doi.org/10.1242/jcs.143420 CrossRefPubMedPubMedCentralGoogle Scholar
- 23.Dou GR, Li N, Chang TF, Zhang P, Gao X, Yan XC, Liang L, Han H, Wang YS (2016) Myeloid-specific blockade of notch signaling attenuates choroidal neovascularization through compromised macrophage infiltration and polarization in mice. Sci Rep 6:28617. https://doi.org/10.1038/srep28617 CrossRefPubMedPubMedCentralGoogle Scholar
- 24.Lambert V, Lecomte J, Hansen S, Blacher S, Gonzalez ML, Struman I, Sounni NE, Rozet E, de Tullio P, Foidart JM, Rakic JM, Noel A (2013) Laser-induced choroidal neovascularization model to study age-related macular degeneration in mice. Nat Protoc 8(11):2197–2211. https://doi.org/10.1038/nprot.2013.135 CrossRefPubMedGoogle Scholar
- 26.Dou GR, Wang YC, Hu XB, Hou LH, Wang CM, Xu JF, Wang YS, Liang YM, Yao LB, Yang AG, Han H (2008) RBP-J, the transcription factor downstream of Notch receptors, is essential for the maintenance of vascular homeostasis in adult mice. FASEB J Off Publ Fed Am Soc Exp Biol 22(5):1606–1617. https://doi.org/10.1096/fj.07-9998com CrossRefGoogle Scholar
- 27.Yan X, Yang Z, Chen Y, Li N, Wang L, Dou G, Liu Y, Duan J, Feng L, Deng S, Han H, Zhang P (2015) Endothelial cells-targeted soluble human Delta-like 4 suppresses both physiological and pathological ocular angiogenesis. Sci China Life Sci 58(5):425–431. https://doi.org/10.1007/s11427-015-4834-3 CrossRefPubMedGoogle Scholar
- 31.Yan XC, Cao J, Liang L, Wang L, Gao F, Yang ZY, Duan JL, Chang TF, Deng SM, Liu Y, Dou GR, Zhang J, Zheng QJ, Zhang P, Han H (2016) miR-342-5p Is a Notch downstream molecule and regulates multiple angiogenic pathways including notch, vascular endothelial growth factor and transforming growth factor beta signaling. J Am Heart Assoc. https://doi.org/10.1161/JAHA.115.003042 CrossRefPubMedPubMedCentralGoogle Scholar
- 35.Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL, Pachter L (2012) Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc 7(3):562–578. https://doi.org/10.1038/nprot.2012.016 CrossRefPubMedPubMedCentralGoogle Scholar
- 37.Connor KM, Krah NM, Dennison RJ, Aderman CM, Chen J, Guerin KI, Sapieha P, Stahl A, Willett KL, Smith LE (2009) Quantification of oxygen-induced retinopathy in the mouse: a model of vessel loss, vessel regrowth and pathological angiogenesis. Nat Protoc 4(11):1565–1573. https://doi.org/10.1038/nprot.2009.187 CrossRefPubMedPubMedCentralGoogle Scholar
- 39.Grande MT, Sanchez-Laorden B, Lopez-Blau C, De Frutos CA, Boutet A, Arevalo M, Rowe RG, Weiss SJ, Lopez-Novoa JM, Nieto MA (2015) Snail1-induced partial epithelial-to-mesenchymal transition drives renal fibrosis in mice and can be targeted to reverse established disease. Nat Med 21(9):989–997. https://doi.org/10.1038/nm.3901 CrossRefPubMedGoogle Scholar
- 40.Gras B, Jacqueroud L, Wierinckx A, Lamblot C, Fauvet F, Lachuer J, Puisieux A, Ansieau S (2014) Snail family members unequally trigger EMT and thereby differ in their ability to promote the neoplastic transformation of mammary epithelial cells. PLoS ONE 9(3):e92254. https://doi.org/10.1371/journal.pone.0092254 CrossRefPubMedPubMedCentralGoogle Scholar
- 42.Xu X, Tan X, Tampe B, Sanchez E, Zeisberg M, Zeisberg EM (2015) Snail is a direct target of hypoxia-inducible factor 1alpha (HIF1alpha) in hypoxia-induced endothelial to mesenchymal transition of human coronary endothelial cells. J Biol Chem 290(27):16653–16664. https://doi.org/10.1074/jbc.M115.636944 CrossRefPubMedPubMedCentralGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.