Luteolin (≥98% pure) was purchased from Sigma-Aldrich (St. Louis, MO, USA), dissolved in DMSO and stored at −20°C. Fetal bovine serum (FBS) was obtained from Life Technology (Carlsbad, CA, USA). Pentobarbital sodium was purchased from Merck (Darmstadt, Germany). Matrigel was purchased from BD Biosciences (Franklin Lakes, NJ, USA). The antibodies against Axl, PI3K, p-PI3K^Tyr458, Akt, p-Akt^Thr308, mTOR, p-mTOR^Ser2448, p70S6k, p-p70S6k^Thr389, heat shock protein 90 (HSP90), matrix metalloproteinase 2 (MMP-2), matrix metalloproteinase 9 (MMP-9), and donkey anti-Rabbit IgG were purchased from Cell Signaling Technology (Danvers, MA, USA). The p-Axl^Y779 antibody and recombinant human Gas6 were obtained from R&D Systems (Minneapolis, MN, USA). The pericyte medium (PM) was obtained from ScienCell Research Laboratories (San Diego, CA, USA). The fertilized chicken eggs were purchased from South China Agricultural University (Guangzhou, China). PKH 26, PKH 67 and other reagents were purchased from Sigma-Aldrich.

Human microvascular endothelial cells (HMEC-1s) were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in RPMI-1640 medium (Gibco, Invitrogen Corp., Carlsbad, CA, USA) supplemented with 10% FBS. HBVPs were obtained from ScienCell Research Laboratories and cultured in PM. All cells were cultured at 37°C under humidified atmosphere containing 5% CO₂.

Adult male Sprague Dawley rats (weighing 180–220 g) were obtained from the Guangdong Medical Experimental Animal Center (Guangzhou, China). The animals were maintained in a specific pathogen-free room with free access to water and standard laboratory chow. All animal experiments were approved by the laboratory animal ethics committee of Jinan University (Guangzhou, China).

The effect of luteolin on the viability of HMEC-1s was assessed using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay. HMEC-1s were seeded in 96-well plates at 1×10⁴ cells per well and cultured for 24 h. Then, the cells were treated with various concentrations of luteolin. After incubating for 24 and 48 h, the cell viability was determined using the MTT assay. The Gas6-induced HMEC-1s proliferation was measured by the MTT assay as previously described (21). HMEC-1s were seeded in 96-well plates at 1×10⁴ cells per well and cultured for 24 h. Adherent cells were starved with serum-free RPMI-1640 medium for 6 h, then, the cells were treated with various concentrations of luteolin, and Gas6 (100 ng/ml) was simultaneously added. After incubating for 24 and 48 h, the cell viability was determined using the MTT assay.

HMEC-1s were grown to 100% confluence in 6-well plates. The cells were starved with serum-free RPMI-1640 medium for 6 h, and then scratched with pipette tips. The cells were cultured with 1% FBS RPMI-1640 medium containing various dilutions of luteolin in the presence or absence of Gas6 (100 ng/ml) for 8 h. Images of the cells were taken 0 and 8 h after wounding.

The in vitro migration and invasion were assessed using a Transwell assay with or without Matrigel as previously described (22). Briefly, 2×10⁴ HMEC-1s were suspended in 100 µl of the serum-free RPMI-1640 medium with or without various concentrations of luteolin (10 and 20 µM) added to the upper chamber insert, the bottom chamber was filled with 600 µl of fresh RPMI-1640 medium supplemented with or without Gas6 (100 ng/ml). After 24-h incubation, the upper chamber was fixed with 4% paraformaldehyde for 30 min and the cells were stained with 0.1% crystal violet. The non-migrated HMEC-1s that remained on the inner side of the upper chamber were removed with a cotton swab. The cells on the lower surface were photographed under an Olympus IX70 inverted microscope. For the cell invasion assay, the upper chamber was pre-coated with 35 µl of diluted Matrigel (Matrigel was diluted to a ratio of 3:1 with PBS) for 1 h at 37°C, and the assay was performed according to the procedures described above. The migrated and invasive cells were quantified using Image-Pro Plus 6.0.

The HMEC-1s were seeded in Matrigel-coated 96-well plates at a density of 2.5×10⁴ cells per well with or without the presence of the indicated concentrations of luteolin (10 and 20 µM), and Gas6 (100 ng/ml) was simultaneously added. After 8 h, capillary-like tubes were observed and photographed under an Olympus IX70 inverted microscope, and the number of tubes was calculated by Image-Pro Plus 6.0.

The dorsal aortas were isolated from the Sprague-Dawley rats as previously described (15). Then the dorsal aortas were cut into 1- to 1.5-mm-long rings. Next, the rinsed aortic rings were placed in Matrigel-coated 96-well plates and sealed with an overlay of 60 µl of Matrigel. After 2-h incubation, fresh RPMI-1640 medium containing Gas6 (300 ng/ml) was added, and the rings were incubated for 48 h. Then, the aortic rings were treated with luteolin (10 µM or 20 µM) for 7 days. The microvessels were photographed under an Olympus IX70 inverted microscope, and the microvessel structures that extend outward from the aortic ring were quantified using Image-Pro Plus 6.0.

The three-dimensional (3D) co-culture of the HMEC-1s and HBVPs was performed as previously described with slight modifications (23). Briefly, the HMEC-1s were labeled with PKH 26 (λex=551 nm, λem=567 nm) following the manufacturer's instructions. Next, 2×10⁴ HMEC-1s were seeded into Matrigel-coated 96-well plates and incubated for 2 h to allow the capillary network to form. Then, 2×10⁴ HBVPs, which were pre-treated with luteolin for 4 h and labeled with PKH 67 (λex=490 nm, λem=502 nm), were added to the capillary network and incubated for another 10 h to allow the HBVPs to adhere to the capillary network and form complicated and solid tubes. The capillary networks were observed and captured under an Olympus IX70 inverted microscope at 0, 2 and 10 h after the addition of HBVPs.

The CAM assay was performed as previously described with slight modifications (24). Fertilized chicken eggs were incubated at 37.8°C in a humidified incubator containing 60–65% humidity for 5 days (blunt end up). A small window approximately 1×1 cm, was carefully drilled in the eggshell on the gas chamber side to create an artificial gas chamber. Then, the shell and shell membrane were removed to expose the CAM. Next, a piece of filter membrane containing Gas6 (300 ng/ml) and the indicated concentrations of luteolin (10 and 20 µM) dissolved in 20 µl PBS, was placed in the center of the CAM. The window was sealed with transparent tape and the CAM was incubated for another 48 h. The microvessels in the CAM were observed under an Olympus SZX18 dissecting microscope, and the microvessels were quantified with Image-Pro Plus 6.0.

HMEC-1s were starved for 6 h in serum-free RPMI-1640 medium and HBVPs were treated with serum-free PM in the same way, then the cells were treated with luteolin (10 and 20 µM) for 4 h. After luteolin was removed, cells were stimulated with Gas6 (100 ng/ml) for another 1 h. Then, the cells were harvested and lysed with RIPA buffer at 4°C. Proteins were analyzed by electrophoresis on 10% SDS-polyacrylamide gels and then detected by western blot as previously described (25). Protein gray values were measured with ImageJ software (NIH, Bethesda, MD, USA).

All experiments were performed in triplicate. The data are expressed as the mean ± SEM, and the statistical analyses were performed with GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). Significant differences were evaluated with one-way ANOVA followed by Tukey's test. P<0.05 was considered statistically significant.

First, the non-toxic concentration of luteolin in HMEC-1s was determined by the MTT assay. Our results showed that luteolin had a negligible effect on the proliferation of HMEC-1s at a dose range of 10 to 20 µM (Fig. 1A). Therefore, the concentrations of 10 and 20 µM were identified as non-toxic doses and used in the subsequent in vitro, ex vivo and in vivo experiments. Then, we examined the effect of luteolin on the Gas6-induced proliferation of HMEC-1s. We found that Gas6 significantly promoted the proliferation of HMEC-1s. In addition, luteolin caused a concentration-dependent inhibitory effect on Gas6-induced growth of the HMEC-1s (Fig. 1B).

The effect of luteolin on the Gas6-induced migration was evaluated by the wound-healing migration assay and Transwell assay. Confluent HMEC-1s were starved with serum-free RPMI-1640 medium for 6 h and scratched with pipette tips. Then, the cells were treated with or without luteolin and Gas6 (100 ng/ml). In the Gas6-treated group, the number of migrated HMEC-1s increased compared with the control group; however, the luteolin treatment suppressed the Gas6-induced migration of HMEC-1s (Fig. 2A). Similar results were observed in the migration chamber and invasion chamber assay. Gas6 facilitated the migration and invasion of HMEC-1s, and the stimulatory effect of Gas6 was attenuated by luteolin (Fig. 2B and C). Taken together, luteolin significantly suppressed the Gas6-induced motilities of HMEC-1s.

We performed a tube formation assay and an aortic ring assay to examine the anti-angiogenic effects of luteolin ex vivo. In the tube formation assay, we found that the capillary-like tube networks in the Gas6-treated group were more solid and complicated than those in the control group. In contrast, luteolin significantly inhibited Gas6-induced tube formation. The number of tubes was significantly decreased after luteolin exposure (Fig. 3A). Next, we used a mouse aortic ring angiogenesis assay to investigate whether luteolin suppressed the Gas6-induced outgrowth of microvessels. The outgrowth of microvessels in the aortic ring assay was remarkably enhanced following Gas6 treatment. In contrast, luteolin markedly suppressed Gas6-induced microvessel sprouting (Fig. 3B). These results suggested that luteolin inhibited Gas6-induced angiogenesis ex vivo.

Since pericytes are crucial for the maturation of neovessels during late-stage angiogenesis and Axl is overexpressed in pericytes (26,27), we conducted a 3D co-culture of HMEC-1s and HBVPs to evaluate the effect of luteolin on pericyte recruitment to the endothelial tubes. A large number of HBVPs migrated to the endothelial tubes within 2 h in the control group. Noteworthy, almost all Gas6-treated HBVPs migrated and adhered to the formed endothelial tubes. However, few HBVPs were recruited to the endothelial tubes in the luteolin-treated group. In addition, the tubes in the Gas6-treated group were more complex and solid after an additional 10-h incubation compared with those in the control group. However, only a few HBVP-supported tubes were observed after the luteolin treatment. These results suggested that luteolin inhibited Gas6-induced HBVP recruitment to the endothelial tubes (Fig. 4A). Then, the inhibitory effect of luteolin on the activation of Axl in HBVPs was examined by western blotting. Our results showed that luteolin significantly suppressed the Gas6-induced phosphorylation of Axl in HBVPs (Fig. 4B).

We then conducted a CAM assay to examine the anti-angiogenic effect of luteolin in vivo. The CAM assay is a widely used and accessible system in angiogenesis studies (28). We found that the blood vessels in the Gas6-treated group formed a dense and spatially-oriented branching network, and the number of blood vessels in the embryonic neovascularization increased significantly compared with control group. However, in the luteolin-treated group, the number of blood vessels was significantly decreased compared with Gas6 group (Fig. 5). These results suggested that luteolin suppressed the Gas6-induced angiogenesis in vivo.

The above mentioned results demonstrated that luteolin suppressed Gas6-induced angiogenesis in vitro, ex vivo and in vivo. Thus, we investigated whether this effect was associated with the suppression of the Gas6/Axl signaling pathway and explored its downstream molecular mechanisms. We found that luteolin significantly suppressed the Gas6-mediated phosphorylation of Axl, followed by the downregulation of Gas6/Axl mediated phosphorylation of the PI3K, Akt, mTOR and p70S6k (Fig. 6A and B). Luteolin also markedly inhibited the expression of MMP-2 and MMP-9 (Fig. 6B), which play crucial roles in various physiological processes, such as wound healing and vessel sprouting (29). In addition, luteolin also inhibited the expression of HSP90 (Fig. 6B), which has been shown to promote angiogenesis (30). These data indicated that luteolin inhibited Gas6-induced angiogenesis by inactivating the Gas6/Axl signaling pathway in HMEC-1s.