Flufenamic acid was purchased from Sigma-Aldrich (USA). BrdU monoclonal antibody was obtained from Neomarkers (USA). Texas-Red-conjugated goat anti-mouse secondary antibody was from Molecular Probes. Growth factor-free Matrigel was from BD Biosciences (USA). All cell culture media and reagents were obtained from Invitrogen (Carlsbad, CA, USA).

The HUVECs purchased from Sciencell (USA) were grown in ECM supplemented with 5% FBS, ECGs and PS in a humidified incubator with 5% CO₂ at 37°C. The cells were trypsinized with 0.15% trypsin-EDTA. Passages 3–10 were used for experiments.

Cell proliferation was determined by both counting the cell numbers and the BrdU incorporation assay. For counting the cell numbers, HUVECs were seeded at an initial density of 2×10⁵ per well in 6-well plates. The FFA was applied at a dose of 20, 50 and 100 μM. Cells were harvested and counted 24, 48 and 72 h after the treatment. Cell numbers were read in a Beckman Counter. For BrdU incorporation assay, 100 μM FFA was applied for 24 h. Then, cells were incubated in the medium with 10 μM BrdU for 3 h and stained with a monoclonal antibody against BrdU at 4°C for 12 h. A goat anti-mouse IgG labeled with Texas Red was used as secondary antibody. The results were expressed as the percentage of BrdU-positive cells over all the cells.

The Matrigel was applied to each well of a 24-well plate and incubated at 37°C for 60 min. The 6×10⁴ of endothelial cells was then seeded into each well with the medium containing 0.8% FCS with or without FFA (100 μM). Images of representative 10× fields were taken and endothelial cell tubes were quantified by counting length and branches.

The cells were washed twice with PBS and total cellular protein was then extracted in lysis buffer containing 62.5 mM Tris-HCl, 2% SDS, 10% glycerol with freshly added proteinase inhibitor cocktail (Sigma, Zwijndrecht, The Netherlands). The protein concentrations were determined by BCA assay (Pierce, Waltham, MA, USA). The protein lysates (40 μg/lane) were separated by SDS-PAGE and transferred onto nitrocellulose membranes. After blocking with 3% bovine serum albumin in phosphate-buffered saline, the membranes were incubated with antiphospho-AMPKα and-AMPKβ, or antiphospho-ACC antibody. After washing, the membranes were probed with horseradish peroxidase-conjugated anti-rabbit IgG and the bands were visualized using an ECL-Plus chemiluminescence detection system (GE Healthcare, NJ, USA). To confirm equal loading of proteins, the membranes were probed for β-actin protein.

HUVECs were lysed with TRIzol Reagent (Invitrogen) and total RNA was extracted according to the manufacturer’s instructions. First-strand cDNA was synthesized from 1.5 μg of total RNA using Moloney murine leukemia virus reverse transcriptase (M-MLV RT) according to the manufacturer’s instructions (Invitrogen). cDNA was amplified by PCR according to the manufacturer’s instructions with Taq DNA polymerase (Invitrogen) under the following conditions: 94°C for 5 min, followed by 40 cycles of 94°C for 30 sec, 56°C for 30 sec and 72°C for 90 sec, with a final elongation step of 10 min at 72°C. The primer informations are as follows: VEGF: forward, 5′-CTACCTCCACCATGCCAAGT-3′; reverse, 5′-TTT CTTGCGCTTTCGTTTTT-3′; AAMP: forward, 5′-CTTTGC ATTGCACTCAGCAT-3′; reverse, 5′-CAGTCACCATTCGGG ACTTT-3′; e-NOS: forward, 5′-GGCTCCCTCCTTCCGG CTG-3′; reverse, 5′-TAGCCGCACGACGCCCT-3′; GAPDH: forward, 5′-AGCCACTGCTGTGCTTTTAAG-3′; reverse, 5′-CCAAAACCAATGATCTCATCC-3′. The products were electrophoresed in 2% agarose gel and stained with ethidium bromide.

Angiogenesis was assayed using the chick embryo CAM assay according to the method described previously (29). Briefly, fertilized chicken eggs were treated with ethanol (70%) and then incubated at 37°C. On day 3, 2–3-ml albumen was aspirated at the acute pole using a sterile 25-G hypodermic needle in order to allow detachment of the developing CAM from the eggshell. After the removal of albumen, we cut a square window ∼10×10 mm into the shell and sealed the window with transparent tapes. Eggs were then incubated in a horizontal position. Six days later, we opened the window and implanted a 1-mm³ sterilized gelatin sponge containing DMSO or FFA onto the CAM. On day 12, the embryos of CAM were fixed by Bouin’s fluid and the distribution and density of CAM vessels next to the site of grafting were analyzed.

All experiments were repeated three times independently. Data were presented as mean ± SEM. or as percentage of control. Statistical comparisons between groups were performed using the Student’s t-test. p<0.05 was considered statistically significant.

To test whether FFA plays a role in angiogenesis, we first investigated the effect of FFA on HUVEC growth. The cells were incubated with DMSO, 20, 50 and 100 μM FFA for 24, 48 and 72 h before the cell numbers were determined. As shown in Fig. 1, FFA at the concentration of 20 μM had no effect on HUVEC growth at any time-point. FFA at the concentration of 50 μM had weak effect on HUVEC growth. The cell number was slightly reduced at the concentration of 50 μM after 72 h, however, there was no significant difference compared with DMSO control. When 100 μM FFA was applied to HUVEC, the numbers of the HUVEC started to reduce after 48 h and was greatly reduced after 72 h. No apoptosis was found at this concentration (data not shown).

It has been reported that FFA inhibits cell proliferation in other cell types (30–32). We next determined whether the effect of FFA on HUVEC cell number was due to its influence on cell proliferation. To address this question, we applied the BrdU incorporation assay. HUVECs were treated with 100 μM FFA for 24 h before they were incubated with 10 μM BrdU for another 3 h. BrdU is an analog of DNA precursor thymidine. When cells are proliferating, BrdU can be incorporated into DNA similarly to thymidine. In this way, the amount of the BrdU in the cells reflects the proliferation rate of the cells. As shown in Fig. 2, the percentage of BrdU-positive cell in DMSO and FFA-treated group was 12±0.86 and 8.65±0.49%, respectively. There was a significant reduction of the percentage of BrdU-positive cells in the FFA treatment group compared with the control group (p<0.05).

In vitro assays of tube formation using endothelial cells are commonly used to study critical steps of angiogenesis (33). We then examine the effect of FFA on tube formation using HUVECs cultured on Matrigel. The tube-like structure appeared 12 h after the HUVECs were seeded. The total tube length and the number of branching points were analyzed as indexes of angiogenesis.

We applied FFA to HUVECs when they were seeded on Matrigel. As shown in Fig. 3A, after 12 h, FFA greatly increased the formation of tube structure. Both total tube length and branching points were significantly increased in FFA treatment group. The relative total tube length of FFA-treated group increased 50.4% compared with that of control group (Fig. 3B). The average branching points per field of FFA-treated group increased 135% compared with that of control group (Fig. 3C). Together, these results suggest that FFA promotes angiogenesis without promoting endothelial cell proliferation.

FFA is one of the Nary-anthranilic acid derivatives, belonging to the fenamate group of NSAIDs (34). To examine whether FFA regulates angiogenesis through AMPK activation, phospho-AMPKα and-AMPKβ and phospho-ACC were measured by western blotting. As shown in Fig. 4, incubation of HUVECs with FFA resulted in increased levels of phosphorylated AMPKα and AMPKβ, which was associated with paralleled elevation of phosphorylated ACC, one of the AMPK substrates (35). Compared with control, FFA-treated group had significantly higher phosphorylated AMPK levels (p<0.05).

Vascular endothelial growth factor (VEGF), endothelial NO synthase (e-NOS), the angio-associated migratory cell protein (AAMP) are three angiogenesis related genes, which were strongly expressed in endothelial cells (36). The mRNA levels of VEGF, e-NOS and AAMP were measured by RT-PCR. Expression of all three angiogenesis related genes is shown in Fig. 5. There was a significant difference between FFA group and control, FFA group had a significantly higher mRNA accumulation level of all the three angiogenesis related genes (p<0.05).

We further examined the possible effect of FFA on angiogenesis in vivo. In order to explore the role of FFA in vivo, we applied the chicken chorioallantoic membrane (CAM) assay. We implanted a 1-mm³ sterilized gelatin sponge which contained PBS or FFA onto the chorioallantoic membrane for ∼72 h for the blood vessel to grow into the sponge. Then the sponge was fixed and the distribution and density of CAM vessels next to the site of grafting were analyzed.

As shown in Fig. 6, the vessel density in the CAM implanted with the gelatin sponge containing PBS was 15.99±1.30. The vessel density in the CAM implanted with the gelatin sponge containing FFA was 23.74±1.82, which was a significant increase compared with the control group. These results suggest that FFA promotes angiogenesis in vivo.