Human umbilical vein endothelial cells (HUVECs) (passages 5–8; Lonza) were cultured in M199 (Gibco) containing 20% fetal bovine serum (FBS) (Lonza), bFGF (3 ng/ml; Invitrogen), heparin (5 U/ml) and 1% penicillin/streptomycin (both from Gibco) (5). HEK293a cells (ATCC CRL-1573) were cultured in Dulbecco's modified Eagle's medium (DMEM) (Gibco) containing 10% FBS. For hypoxic condition, cells were incubated in a Forma hypoxia chamber (Forma Scientific), which is an anaerobic system that strictly regulates oxygen levels; cells were maintained at low oxygen tension (1% O₂, 5% CO₂ and balanced with N₂) to simulate hypoxia.

Total RNA was isolated using the QIAshredder and RNeasyPlus Mini kits (Qiagen Inc.). The PrimeScript™ First Strand cDNA Synthesis kit (Takara) was used to synthesize cDNA from 1 µg of total RNA according to the manufacturer's instructions. Real-time PCR was performed using the SYBR-Green PCR Master Mix (Roche), using primers for human FGF11 as follows: forward, 5′-TGTCGCTTTAAGGAGTGCGT-3′ and reverse, 5′-AGAGAAGGCTCCCGGTACAT-3′. Real-time PCR data were acquired using an ABI PRISM-7500 sequence detection system (Applied Biosystems). The 18S rRNA gene was used as a positive control and for normalization. End-point PCR for FGF11 was also performed. GAPDH was used for normalization.

Oligonucleotide primers for PCR were designed as follows: FGF11 forward, 5′-GTCACCATCCAGAGTGCCAA-3′ and FGF11 reverse, 5′-CACTGTGGAGAGAAGGCTCC-3′; GAPDH forward, 5′-CATGACAACTTTGGCATTGTG-3′ and GAPDH reverse, 5′-GTTGAAGTCGCAGGAGACAAC-3′. The PCR products were analyzed using a 1.2 % agarose gel.

Full-length human FGF11 was synthesized by PCR and cloned into the pcDNA3.1/HA vector (Invitrogen). Transfection was carried out using Metafectene Pro (Biontex). For western blotting, cells were harvested and lysed with lysis buffer containing protease inhibitors (Roche). Total protein (20–30 µg) was immunoblotted with antibodies specific to FGF11 (R&D Systems), zonula occludens-1 (ZO-1) (Invitrogen), occludin (Invitrogen) or claudin-5 (Abcam). α-tubulin (Calbiochem) was used as an internal control. Quantification of band intensity was analyzed using ImageJ (NIH).

The tube formation assay was performed as previously described (19). Briefly, 200 µl of growth factor-reduced Matrigel (BD Biosciences) was pipetted into a well of a 24-well culture plate and polymerized for 30 min at 37°C. After transfection, HUVECs (1×10⁴ cells/well) were seeded onto polymerized Matrigel and incubated in M199 containing 2% FBS and heparin (10 U/ml). Every hour up to 16 h, the cultures were photographed with an Olympus TH4-200 microscope. Capillary-like tube networks were observed and the branch point number was counted.

HUVECs were transfected with an FGF11-overexpression plasmid or control mock plasmid. After one day, cells were plated on 60-mm culture dishes and the migration assay was performed as previously described (20). Briefly, confluent HUVECs were wounded and incubated in M199 media with 2% FBS and 1 mM thymidine. After 16 h, HUVECs were fixed with absolute methanol for 2 min and stained with Giemsa solution for 3 min. Migration activity was quantitated by counting the number of cells that moved beyond the reference line (20).

A partial genomic DNA sequence encompassing the human FGF11 promoter region bearing putative HREs was amplified by PCR and cloned into the luciferase pGL3 promoter vector (Promega). Primer information was as follows: forward, 5′-CTGCTAGCCCAACCTCTCCTTCCTACC-3′ (pGL3-FGF11-HREs); forward, 5′-GTGCTAGCGGGGCTGGTTAGATTGGAG-3′ (pGL3-FGF11-ΔHREs); and reverse, 5′-ATAGATCTACTAGGGCATGCTCTTGACG-3′. HEK293a cells were plated at a density of 2×10⁵ cells/well of a 6-well plate and transfected with various combinations of effector plasmids. Luciferase assays were performed using the luciferase assay system kit with a GloMax luminometer (both from Promega), according to the manufacturer's instructions. Relative luciferase activity was normalized to relative light units and β-galactosidase activity.

The data are expressed as means ± standard deviations (SD). The statistical differences between the groups were compared using the unpaired t-test or the one-way analysis of variance (ANOVA). P-values ≤0.05 were considered to indicate statistically significant results.

Solid tumor angiogenesis is initiated by hypoxic conditions that serve as a strong stimulus for new vessel formation (9,10). Since we are interested in identifying hypoxia-induced genes, we first investigated the mRNA expression of FGF homologous factors (FHFs; FGF11-FGF14) in HUVECs after exposure to hypoxia (1% O₂). FGF14 mRNA was not detected in HUVECs by real-time PCR. FGF12 and FGF13 expression increased slightly under hypoxia, yet their expression level was very low in HUVECs. Whereas FGF11 mRNA expression was relatively high in comparison to FGF12 and FGF13 expression (data not shown). FGF11 mRNA expression was significantly increased in response to hypoxic conditions (Fig. 1A and B). Western blotting results for the FGF11 protein suggests that the protein level was significantly increased under hypoxia (Fig. 1C).

To investigate the effect of FGF11 expression on angiogenesis, tube formation and cell migration were examined for HUVECs transfected with pFGF11 (Fig. 2A). FGF11 overexpression significantly stimulated tube formation compared to the control cells (Fig. 2B and C), yet did not stimulate migration activity (Fig. 2D). HUVECs treated with basic FGF (bFGF) as a positive migration control (14) migrated normally, thus we concluded that FGF11-overexpression did not affect endothelial migration activity. Instead, FGF11 may be involved in stabilizing capillary-like tube structures.

Since capillary tube formation in HUVECs was increased with FGF11 overexpression, we examined whether FGF11 overexpression in endothelial cells affects the expression of TJ proteins by western blotting (Fig. 3). TJ proteins play a role in stabilizing capillary structure by maintaining adhesive cell-cell interactions (21). We found that FGF11 overexpression markedly increased the levels of the TJ proteins, such as occludin, ZO-1, and claudin-5 (Fig. 3).

We further examined the novel finding that FGF11 expression was upregulated in response to hypoxic conditions by determining the mechanism through which hypoxia stimulates FGF11 expression, focusing on the FGF11 promoter. HIF-1α is a key transcription factor that activates genes involved in the hypoxic response by binding to HREs in the gene promoter region (2,16). Notably, the FGF11 promoter contains two HREs (5′-ACGTG-3′) (Fig. 4A).

We determined the effects of HIF-1 on the FGF11 promoter containing two HREs (FGF11-HREs; Fig. 4B and C) using a promoter luciferase assay. Reporter gene activity was significantly increased in response to hypoxia for the cells transfected with FGF11-HREs (WT). In contrast, reporter gene activity was not changed by hypoxia for cells transfected with the HRE-deletion-fragment (ΔHRE) (Fig. 4B), suggesting that the HREs in the FGF11 promoter region are sensitive to hypoxia. To determine whether FGF11-HREs are responsive to hypoxia via the HIF-1α transcription factor, we co-transfected cells with both WT FGF11-HREs and HIF-1α under normoxic conditions (Fig. 4C). To promote assembly of the functional HIF1 complex, cells were also co-transfected with the partner of HIF-1α, HIF-1β. With HIF-1α overexpression, reporter gene activity was high even under normoxic conditions, whereas the HRE-deletion-fragment (ΔHRE) was unaffected by HIF-1α overexpression (Fig. 4C), which indicates that FGF11 promoter induction occurs via HIF-1.