1. Introduction
Angiogenesis is the process of developing new vessels from the pre-existing vasculature and it occurs throughout the entire lifespan of an organism under both physiological and pathological conditions [
1]. Since the vascular network is mostly quiescent in adulthood, physiological angiogenesis mainly occurs in developmental contexts such as embryonic blood vessels [
2] and endochondral bone formation [
3]; the exception is the female reproductive tract such as cycling ovary and uterus during pregnancy, where the formation of new vessels is required [
4]. However, in numerous pathological conditions such as tumors or inflammation, the resting vasculature can be reactivated [
5].
Angiogenesis takes place in a number of sequential steps: endothelial cell activation, cell sprouting, proliferation, migration, lumen formation, cell remodeling, and maturation. It is initiated in normally quiescent endothelial cells (ECs) upon receiving pro-angiogenic factors such as Vascular Endothelial Growth Factor (VEGF), the principal angiogenic factor. When VEGF contacts the VEGF receptor (VEGFR), the receptor generates the signals to develop a vascular system in the embryo or to generate blood vessels and lymphatic vessels in adults [
6]. Among various VEGFs (VEGFA, VEGFB, VEGFC, and VEGFD) and VEGFRs (VEGFR1, VEGFR2, and VEGFR3), VEGFA/VEGFR2 signaling prominently mediates angiogenesis processes of blood vessels. Diverse downstream signaling pathways including MAPK pathways: p38, ERK, phospholipase γ (PLCγ), Src, and FAK pathways are reported to regulate the cell survival, proliferation, migration, and cellular attachment of endothelial cells [
7,
8].
VEGF/VEGFR signaling also has an influence on a variety of other pathways. Notch signaling is an evolutionally conserved pathway involved in the determination of fate of many tissues and cell types including endothelial cells. Notch receptors and their ligand, the delta-like ligand 4 (Dll4), are responsible for tip and stalk cell differentiation from endothelial cells (ECs) in nascent vascular sprouts, while the migration and proliferation of tip and stalk cells governs the formation, elongation, and maturation of new vessels eventually [
5,
9].
As angiogenesis is regulated by the balance between pro-angiogenic and anti-angiogenic signals, any disturbance in that balance would lead to pathological conditions. While an insufficient vascular supply results in ischemic diseases, excess neovascularization is a hallmark of solid tumors [
10,
11]. The hypoxic and acidic nature of the tumor microenvironment (TME) induces the secretion of VEGFs from cells such as macrophage, fibroblasts, endothelial cells, and tumor cells, rendering TME to be angiogenic, which in turn enhances the tumor growth and metastasis [
12]. Accordingly, the inhibition of angiogenesis has been one of the therapeutic options in cancer and there are a variety of approved anti-angiogenic drugs; Bevacizumab, a neutralizing monoclonal antibody directed against VEGF, and tyrosine kinase inhibitors (RTKi) such as sorafenib, sunitinib, cediranib, and axitinib are the most commonly used ones in this category [
13,
14]. Moreover, the monoclonal antibody against Dll4 also exhibited inhibitory properties on breast tumor growth [
15].
Sesamin (
Sesamun indicum L.), a lignan found in sesame seeds and oil, is reported to have biological effects including chemoprevention, anti-inflammation, and anti-oxidant properties [
16,
17,
18]. The effect of sesamin on angiogenesis under physiological and pathological conditions has been studied by different scientists, however, the findings were contradicting and confusing. A study has shown that sesamin increases the in vitro and in vivo angiogenic processes including endothelial cell proliferation, migration, tube formation, and neovascularization via activation of signaling modulators: ERK, Akt, focal adhesion kinase (FAK) and p38 mitogen-activated protein kinase (MAPK) [
19]. In contrast, sesamin was shown to inhibit angiogenesis in other studies [
20,
21]. In addition to these controversial findings for the effect of sesamin on angiogenesis, the underlying signaling mechanisms and its effect on Notch signaling have not yet been elucidated enough.
This study thus aimed to investigate the effect of sesamin on angiogenesis in an in vivo model using the chick chorioallantoic membrane (CAM) [
22] and in EA.hy926, a human endothelial cell line. Although sesamin did not affect either in vitro or in vivo physiological angiogenesis, it could inhibit VEGFA-induced pathological angiogenesis in the CAM model. This might be due to its ability to decrease endothelial cells proliferation and migration via inactivation of Src and FAK signaling. Moreover, inactivation of those pathways by sesamin might also result in a lower expression of NOTCH, eventually inhibiting Dll4-Notch signaling.
2. Materials and Methods
2.1. Chemicals
Methylthiazoletetrazolium (MTT) (PubChem CID: 64965), Dimethyl sulfoxide (DMSO) (PubChem CID: 679), and Resazurin (PubChem CID: 11077) were obtained from Sigma Chemical, Inc. (St. Louis, MO, USA). Recombinant human vascular endothelial growth factor A (Srf21-derived, Catalog number 293-VE) was purchased from R&D systems® (Minneapolis, MN, USA). Illustra RNAspin Mini RNA Isolation Kit and the primers for real-time RT-PCR were acquired from GE Healthcare Europe GmbH (Freiburg, Germany) and BioDesign (Bangkok, Thailand), respectively. A Tetro cDNA Synthesis kit and a SensiFAST™ SYBR® No-ROX Kit were purchased from BIOLINE (London, UK). Angiogenesis Antibody Sampler Kit, Notch1 (D1E11) and β-Actin Rabbit monoclonal antibody were procured from Cell Signaling Technology® (Beverly, MA, USA).
2.2. Sesamin Preparation
Sesamin seeds were procured from the Lampang province of Thailand, and the voucher specimens (BKF no. 138181) were approved by the National Park, Wildlife and Plant Conservation Department, Ministry of Natural Resources and Environment, Bangkok, Thailand. Sesamin was extracted from the seeds by using the method reported in the previous study [
23]. The sesamin obtained was first dissolved in DMSO to prepare the stock solution (100 mM) before being diluted with the culture media to the required concentrations. All the experiments were carried out in compliance with the relevant guidelines of the institution.
2.3. In Vivo Angiogenesis Model: Chick Chorioallantoic Membrane (CAM) Assay
Fertilized chicken eggs were incubated at 37 °C and approximately 50–60% humidity. On day 3 of incubation, 5–6 mL of albumen was aspirated to detach the developing CAM from the top part of the shell. On day 8, a window of around 1.5 cm
2 was gently opened on the wide end of the egg without damaging the embryo. A plastic ring was placed directly on the top of the CAM. Various concentrations of sesamin and/or 20 ng/mL of VEGFA were added directly onto the plastic rings. The eggs were transferred back into the incubator and the numbers of vascular branches were counted at day 10 of incubation by photographing the CAM area of each egg. Images were analyzed by counting the number of branching vessels in the plastic ring. Scores of primary-, secondary-, tertiary-, and quaternary-branching vessels are denoted by 1, 2, 3, and 4, respectively. Further branching vessels with more than quaternary-branching were scored as 5. Scores from each group were averaged by the number of eggs and the average score of each group was normalized to that of the control group [
22,
24].
2.4. Cell Line and Culture
EA.hy926, a human umbilical vein endothelial cell line, was purchased from ATCC
® (CRL2922™). This cell line was derived by fusing human umbilical vein endothelial cells with the permanent human cell line, A549. This study used EA.hy926 because this cell line shows endothelial characteristics and more cells can be obtained than from the primary cells [
25]. The cells were cultured as a confluent monolayer in Dulbecco’s Modified Eagle’s Medium (DMEM), containing 10% fetal bovine serum and 2% HAT (100 μM hypoxanthine, 0.4 μM aminopterin, and 16 μM thymidine). Cells were maintained in a humidified incubator with 5% CO
2 at 37 °C. A cell passage of six or seven was used in the experiments.
2.5. MTT Assay
To determine the toxicity on cells and select the optimal concentrations of sesamin, a MTT assay was performed. EA.hy926 cells were placed in a 96-well-plate (1000 cell/well) and incubated overnight. After cells were treated with various sesamin concentrations for 24 h, the culture media were discarded and replaced with 100 μL of MTT (0.5 mg/mL) solution for 4 h. Then, the MTT agent was discarded and 100 μL dimethyl sulfoxide (DMSO) was added into each well to solubilize the formazane crystals. The absorbance was measured at 540 nm using a microplate reader and the percentage cell survival compared to the controls was calculated as follows:
2.6. In Vitro Wound Healing Assay
To determine the migration ability of EA.hy926 cells, an in vitro wound healing assay was performed. EA.hy926 cells were seeded into 24 well plates (200,000 cells/well) and incubated. After 24 h of incubation, a line was scraped through the monolayer of the cells using a 250 μL-pipette tip (wounding). The cells were then washed with PBS and treated with various concentrations of sesamin and/or 20 ng/mL VEGFA. The wounds were photographed at different time points under a light microscope (40× magnification). The measurements of the scratched region were calculated using AxioVision Analytic Software version 4.7 from Carl Zeiss (Jena, Germany). [
20] The migration ability was calculated as follows:
2.7. AlamarBlue Assay
The extent of cell proliferation was assessed using the AlamarBlue assay. EA.hy926 cells were seeded into 96 well plates (1000 cells/well) and incubated at 37 °C, 5% CO2 for 24 h. The culture media was then discarded, and the cells were treated with various concentrations of sesamin and/or 20 ng/mL VEGFA for 1 week. Every 24 h, the culture media was discarded and replaced with 10% (v/v) Alarmar Blue fluorescent dye in media for 4 h at 37 °C. The absorbance of the media was read at 540 and 620 nm using a microplate reader spectrophotometer. The absorbance was measured at wavelengths of 540 and 620 nm. The percentages of the differences in reduction were calculated, and the data represented as percentages of the differences in reduction relative to those of the control group.
2.8. Gene Expression by Real-Time Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
Real time RT-PCR was used to examine the gene expression in EA.hy926 in response to treatment with various concentrations of sesamin and/or 20 ng/mL VEGFA. After 24- or 48-h incubation periods, cells were lysed, and the total RNA was isolated using the RNA extraction kit (GE Healthcare) following the manufacturer’s protocol. Samples were treated with DNase before washing and elution steps. After extraction, the amount of total RNA in samples was measured using nanodrop spectrophotometer (Thermo Scientific, Waltham, MA, USA), and 200 ng of total RNA were converted to cDNA using a Tetro cDN A Synthesis Kit in a final volume of 20 μL. Reaction conditions were set as suggested by the manufacturer. After reverse-transcription, the product cDNA was diluted with RNase free water in 1:10 dilution. Then, 8 μL of the cDNA mixture was used for real-time PCR experiments. Real time PCR was performed using a SensiFastTM SYBR
® No-ROX kit on Chromo4TM Four-Color Real-Time Detector from Bio-Rad (Hercules, CA, USA) in a final volume of 20 μL. The primer concentration was 500 nM. Forty cycles of PCR amplification were performed at 95 °C for 5 s and 60 °C for 10 s. The primers used are shown in
Table 1. The relative expression for each gene was normalized to that of GAPDH and against the control group by the 2
ΔΔCT method [
26]. Additional information on normalized gene expression (ΔCт) of each target gene against GAPDH gene was described in
Supplementary Materials.
2.9. Western Blotting Analysis
EA.hy926 cells were treated with various sesamin concentrations and/or VEGFA and incubated at 37 °C, 5% CO2. At the indicated times, the cells were lysed with lysis buffer containing 50 mM Tris-HCL, pH7.4, 250 mM NaCl, 0.5% NP-40, 5 mM EDTA, and 50 mM NaF along with protease and phosphatase inhibitors (Roche Diagnostics GmbH, Mannheim, Germany). An equal volume of whole cell lysate was electrophoresed and transferred to a nitrocellulose membrane. After blocking with 5% skim milk in 0.05% PBS-TWEEN, the membranes were incubated with specific primary antibodies. After the incubation of the secondary antibodies and washing, specific protein bands were developed using Supersignal West Femto Substrate (Thermo Scientific, Rockford, IL, USA) and were photographed using the molecular chemidoc XRS system (Bio-Rad, Hercules, CA, USA). The band density was analyzed using TotalLab TL120 software version 2006 and calculated in relation to the control sample. The beta actin was used as an internal protein control.
2.10. Statistical Analysis
All data are given as mean ± standard error of mean (SEM) from triplicate samples of three or two independent experiments. One-way analysis of variance (one-way ANOVA) and student’s t-test were used to compare the treatment and control conditions using data from three or two independent experiments, respectively. Statistical significance was assumed at p < 0.05.
4. Discussion
The angiogenesis process occurs throughout the entire life of vertebrates and plays an important role in both the physiological and pathological conditions [
5,
6]. Abnormal angiogenesis, either insufficient or excess, may creating implications in a variety of diseases such as placental insufficiency, ischemic heart disease, or tumor growth [
11]. Having an increased vascular supply is essential for tumors not only to match with its high demand of nutrients but also to enhance its dissemination and metastasis [
10]. Consequently, VEGF signaling inhibitors have been used as one of the therapeutic options in cancer [
13]. Many phytochemicals have also been investigated for their effect on angiogenesis; previous studies have reported the effect of the phytochemical sesamin (
Sesamun indicum L.), with known anti-inflammatory and anti-oxidant activities on the angiogenesis process. Nevertheless, their findings were contradicting as Tsai et al. reported sesamin as being anti-angiogenic while Chung et al. stated that sesamin induced angiogenesis [
19,
20]. Moreover, the effect of sesamin on the NOTCH signaling pathway, the key signaling in different stages of angiogenesis, has not yet been elucidated. Hence, this study attempted to confirm the effect of sesamin on angiogenesis and the underlying signaling pathways in both in vitro and in vivo using the human endothelial cell line, EA.hy926, and the chick chorioallantoic membrane (CAM) models, respectively.
We first investigated the effect of sesamin on in vivo angiogenesis in the CAM model and the formation of vascular branches in CAMs was not affected by sesamin, reflecting its lack of influence on the physiological angiogenesis. This finding is contrary to that of Chung et al., in which sesamin significantly increased neo-vascularization using the Matrigel model in male BALB/c mice [
14]. It appears that different in vivo methods used to observe the formation of vascular branching may yield different results. We then examined the effect of sesamin on angiogenesis under pathological conditions using vascular endothelial growth factor A (VEGFA) as an instigator in the same CAM model. VEGF, an inflammatory cytokine, is the key driver of sprouting angiogenesis and is well-known to be overexpressed in tumor cells, via induction of HIF-1α under hypoxic conditions, promoting proliferation and migration of vascular endothelial cells and blood vessels outgrowth, thereby assisting the growth and metastasis of tumors [
10,
29]. Hence, we treated the CAM with VEGFA, the major VEGF involved in the growth of blood vessels, to mimic the pathological conditions of tumors, and it was found that VEGFA doubled the vascular branches formation in the CAM. This effect of angiogenesis induction by VEGFA was significantly reduced by cotreatment with sesamin, prompting the role of sesamin in the inhibition of pathological angiogenesis mediated by VEGF.
Next, we further studied the effect of sesamin on angiogenesis and its underlying mechanisms, including its effect on NOTCH signaling, using the in vitro cell model, human endothelial cells EA.hy926. During the sprouting process of a new vessel, quiescent endothelial cells become activated and secrete enzymes that degrade extracellular matrix (ECM) proteins in basement membranes, allowing endothelial cells to migrate through. Then, endothelial cells in a nascent sprout are differentiated into two cell types, namely tip cells and stalk cells: tip cells are highly migratory and responsible for sensing the directional cues, thereby defining the route of the new sprouts and stalk cells, which highly proliferate and support sprout elongation [
9]. Hence, the effect of sesamin on the proliferation and migration of endothelial cells was studied in EA.hy926 endothelial cells. VEGFA indeed induced the proliferation and migration of EA.hy926 as expected, in accordance with the findings of previous studies in endothelial cells, HUVEC and EA.hy926 [
19,
30,
31]. Sesamin was found to inhibit the induced proliferation and migration significantly, although it had no significant influence on the uninduced EA.hy926 cells. Agreeing with the results of in vivo CAM, the findings in the EA.hy926 cells once again indicated the ability of sesamin on hindering the pathological angiogenesis. This study endorsed the findings of Tsai et al. stating that sesamin can inhibit the VEGFA-induced proliferation and migration of HUVECs endothelial cells. However, sesamin was claimed to inhibit the proliferation and migration of uninduced HUVEC as well in that study, contradicting our results [
20]. The different endothelial cell types may respond in a different way to sesamin, accounting for the distinct results.
The responsible signal transductions for the inhibitory effect of sesamin on VEGFA mediated angiogenesis were also investigated. We focused on the activation of p38, ERK, PLCγ1, Src, and FAK pathways, given that activation of ERK through PLCγ1 was reported to associate with the regulation of endothelial cell survival and proliferation while that of p38 and FAK was to control the migration of endothelial cells and that of Src was to regulate cell-cell contacts, proliferation, and migration [
7,
32]. Although the study by Byung-Hee Chung showed that 30 μM of sesamin could induce the activation of ERK, Akt, and FAK in uninduced HUVECs [
19], we did not find the activation of these pathways by sesamin in EA.hy926. In contrary, VEGFA obviously induced all signaling pathways related to the activation of angiogenesis and sesamin pretreatment significantly impeded the activation of Src and FAK by VEGFA but not that of ERK, p38, and PLCγ1. Previous studies in HUVECs reported that total saponins and carvedilol could inhibit VEGFA-induced angiogenesis by the inhibition of FAK and Src activation, respectively [
30,
33], implying that Src and FAK signaling might be responsible for the proliferation and migration abilities of endothelial cells. The results of this study and previous studies altogether indicated the role of Src and FAK pathways in mediating the inhibitory effect of sesamin on angiogenesis under high VEGFA conditions.
As mentioned above, endothelial cells in the form of tip or stalks cells work together in a harmonious way in sprouting angiogenesis. While the spear head tip cells migrate and lead the sprout towards the stimulant, a high VEGF gradient, they suppress the tip cell phenotype in the adjacent cells, which then become stalk cells. Stalk cells highly proliferate and facilitate the guiding of tip cells, resulting in elongation of the new vessels. Notch signaling pathway, which plays a critical role in cell fate determination, is involved in the specification of endothelial cells into tip and stalk cells [
9]. When VEGF interacts with its main receptor VEGFR2, also known as KDR, the expression of the Dll4 ligand is upregulated in tip cells, which in turn activates Notch signaling in the neighboring stalk cells, resulting in increased proliferation but suppression of tip cell-behavior in these cells (lateral inhibition). Notch activation also reduces VEGFR2 expression in the stalk cells. In contrast, VEGFR2 expression is increased in tip cells as they have low activation of Notch signaling [
5,
9]; consequently, tip and stalk cells exhibit distinctive gene expression profiles. In this study, we studied the expression of angiogenic genes in VEGFA-induced EA.hy926 cells and the effect of sesamin on them. The results showed that when cells were induced with VEGFA, the
NOTCH1 expression was significantly induced, whereas
Dll4, and
VEGF expressions were not significantly changed.
KDR expression was reduced, though not significant, by VEGFA treatment. This profile of high Notch and low KDR expressions proposed that endothelial cells in this study represent stalk cells, although it cannot be definitely distinguished since we used monolayer endothelial cell line lacking 3D interaction. Interestingly, sesamin was found to reduce the Notch1 induction by VEGFA significantly in both gene and protein levels. Previous studies found that inhibition of Notch signaling leads to an increased formation of non-functional vascular sprouts in tumor vasculature, which was explained by the de-repression of tip-cell behavior in endothelial cells [
34]. Since sesamin was shown to reduce vascular branching in the previous experiment using the CAM model, the finding of Notch expression inhibition by sesamin was found to be contradicting. However, Liu et al. reported that vascular network formation was partially inhibited by blocking Notch signaling on the other hand [
28]. These conflicting results might be explained by the plausible mechanism: inhibition of Notch 1 signaling will hinder the proliferation of stalk cells, a necessity for the elongation of new sprouts in angiogenesis. The results of inhibition of VEGFA-induced proliferation and migration of EA.hy926 cells also support this explanation.
Activation of alternative angiogenic signaling pathways is contributing to the resistance against VEGF blockade therapy in cancer patients. Angiopoietin 2 (Ang2), being the antagonist of Ang1 that stabilizes nascent vessels by recruiting mural cells, is implicated in the formation of unstable and leakier vessels and it is upregulated in many cancers [
13]. Hence, we examined the effect of sesamin on this pathway; specifically the expressions of
Ang1,
Ang2, and their receptor
Tie2, in EA.hy926 cells. Nevertheless, no conclusive finding was obtained. Sesamin increased the expression of
Ang1 under VEGF influence of 12 h-duration, suggesting its potential to help the newly formed vessels to become stabilized, but the finding was not confirmed as the effect was not significant and disappeared in the 24-h period. On the other hand, treatment with sesamin alone reduced the expression of
Ang1 and increased that of
Ang2 in EA.hy926 cells without VEGF induction, although it was not significant. Therefore, the effect of sesamin on angiopoietins is yet to be confirmed in future studies using other endothelial cell types or a different dose and duration of the sesamin treatment.