Roles of tetrahydrobiopterin in promoting tumor angiogenesis

Abstract

Nitric oxide (NO), which is derived from endothelial NO synthase (eNOS), provides crucial signals for angiogenesis in the tumor microenvironment. Tetrahydrobiopterin (BH4) is an absolute requirement for eNOS activity. In this study, we investigated whether this activation is both maintained by a wild-type Ras/phosphatidylinositol 3-kinase (PI3K)/Akt-positive feedback loop in endothelial cells and affects tumor angiogenesis. We found that supplementation of BH4 (via the pterin salvage pathway with Sep) increased Akt/eNOS phosphorylation in both human eNOS-transfected COS-7 cells and endothelial cells concomitant with increases in NO production, cell proliferation, migration, and tube formation. This augmentation was abrogated by a PI3K inhibitor. Sepiapterin (Sep) also increased GTP-bound wild-type Ras and PI3K/Akt/eNOS activation, which was prevented by the eNOS inhibitor, Nω-Nitro-L-arginine methyl ester (L-NAME). Furthermore, expression of GTP cyclohydrolase I (the rate-limiting enzyme in de novo BH4 synthesis) under doxycycline control potentiated in vivo tumorigenesis, tumor cell proliferation, as well as angiogenesis. Conversely, both switching off GTP cyclohydrolase I expression as well as inhibiting its enzymatic activity significantly decreased eNOS expression and tumor vascularization. This study demonstrates an important role for BH4 synthesis in angiogenesis by the activation of eNOS for NO production, which is maintained by a PI3K/Akt-positive feedback loop through effects on wild-type Ras in endothelial cells. Our findings suggest that BH4 synthesis may be a rational target for antiangiogenesis therapy for tumors.

Figure 1

Phosphorylation of eNOS and Akt by BH4 synthesis depends on PI3K signaling in COS-7 cells and HUVEC. After overnight culture without FBS, human eNOS-transfected COS-7 cell monolayers were incubated for 30 minutes with Sep at various doses (0, 2, 5, and 10 μmol/L) or the control (DMSO). Lysates were fractionated by SDS-PAGE and immunoblotted with monoclonal antibody to p-eNOS (Ser¹¹⁷⁷), p-Akt (Ser⁴⁷³), and GAPDH (A). HUVEC were cultured and pre-treated in Endothelial Cell Growth Medium-2 (EGM-2) without FBS. Lysates were analyzed as for COS-7 cells (B). Bands intensity of p-eNOS and p-Akt were quantified on ImageJ software. Data are shown as the mean ± SEM (*P < 0.05 vs. the control or Sep+LY, n = 4) (ANOVA).

Figure 2

Sep-induced NO production and cell proliferation, migration, and tubule formation are dependent on PI3K signaling in HUVEC. A: Cells were cultured and pretreated as for Figure 1. B–D: For proliferation (B), migration (C), and tubule formation at × 20 magnification (D), cells were incubated with Sep (10 μmol/L) ± LY (10 μmol/L) or the control (DMSO) for 24 hours. Data are shown as the mean ± SEM (*P < 0.05 vs. the control or Sep+LY, n = 6) (ANOVA).

Figure 3

Sep-activated wild-type Ras cascade is mediated by NO in HUVEC. A: HUVEC were cultured and pretreated ± Nω-Nitro-L-arginine methyl ester (L-NAME; 10 μmol/L) or LY (10 μmol/L) for one hour. Lysates were then incubated with GST-Raf-RBD. After washing, the pull-down proteins by agarose beads were fractionated by SDS-PAGE and immunoblotted with pan-Ras antibody to GTP-bound Ras and GST-Raf-RBD. B and C: Lysates were also immunoblotted and analyzed for p85α/p110α, p-eNOS (Ser¹¹⁷⁷), p-Akt (Ser⁴⁷³), and GAPDH. D: Schematic diagram of proposing signaling.

Figure 4

GTPCH overexpression and BH4 synthesis promote angiogenesis and tumor formation in mice. GCHtet-off cells (1 × 10⁷) were injected s.c. into flanks of 6- to 8-week-old female BALB/c SCID mice (n = 5) and Tet-off-EV as control. Tumor volumes were monitored two to three times per week (A), when they reached to maximum permitted volumes, mice were sacrificed, and tumors were sectioned and stained immunohistochemically for HA-GTPCH (red arrow-tumor cells) (A), Ki-67 (B), and CD34 (C) at × 20 magnification. Data are shown as the mean of five sections per animal ± SEM (*P < 0.05 vs. Tet-off-EV, n = 5) (ANOVA).

Figure 5

Both switch-off of GTPCH expression and inhibition of BH4 synthesis decrease angiogenesis in tumor stroma. HA-GTPCH expression is highly regulated by Dox in GCHtet-off cell line (A). GCHtet-off cells (1 × 10⁷) were injected as for Figure 4. When tumors reached to 100 mm³, mice were fed with Dox in the drinking water or i.p. injected with DAHP (300 mg/kg of body weight) daily. After seven days, tumor sections were stained for HA-GTPCH (B) and CD34 (C). Data are shown as the mean of five sections per animal ± SEM (*P < 0.05 vs. Dox or DAHP, n = 5) (ANOVA) (C). Tissue homogenates were analyzed for eNOS, HA-GTPCH, and GAPDH (*P < 0.05 vs. Dox or DAHP, n = 5) (ANOVA) (D). HUVEC were incubated with graded doses of DAHP overnight, and lysates were assayed for p-eNOS (Ser¹¹⁷⁷)/eNOS and GAPDH (representative of three experiments) (E).