Angiotensin II increases angiogenesis by NF-κB-mediated transcriptional activation of angiogenic factor AGGF1

Abstract

Angiogenic factor with G-patch and FHA domains 1 (AGGF1) is involved in vascular development, angiogenesis, specification of hemangioblasts, and differentiation of veins. When mutated, however, it causes Klippel-Trenaunay syndrome, a vascular disorder. In this study, we show that angiotensin II (AngII)-the major effector of the renin-angiotensin system and one of the most important regulators of the cardiovascular system-induces the expression of AGGF1 through NF-κB, and that AGGF1 plays a key role in AngII-induced angiogenesis. AngII significantly up-regulated the levels of AGGF1 mRNA and protein in HUVECs at concentrations of 10-40 μg/ml but not >60 μg/ml. AngII type 1 receptor (AT1R) inhibitor losartan inhibited AngII-induced up-regulation of AGGF1, whereas AT2R inhibitor PD123319 further increased AngII-induced up-regulation of AGGF1. Up-regulation of AGGF1 by AngII was blocked by NF-κB inhibitors, and p65 binds directly to a binding site at the promoter/regulatory region of AGGF1 and transcriptionally activates AGGF1 expression. AngII-induced endothelial tube formation was blocked by small interfering RNAs (siRNAs) for RELA (RELA proto-oncogene, NF-κB subunit)/p65 or AGGF1, and the effect of RELA siRNA was rescued by AGGF1. AngII-induced angiogenesis from aortic rings was severely impaired in Aggf1^+/- mice, and the effect was restored by AGGF1. These data suggest that AngII acts as a critical regulator of AGGF1 expression through NF-κB, and that AGGF1 plays a key role in AngII-induced angiogenesis.-Si, W., Xie, W., Deng, W., Xiao, Y., Karnik, S. S., Xu, C., Chen, Q., Wang, Q. K. Angiotensin II increases angiogenesis by NF-κB-mediated transcriptional activation of angiogenic factor AGGF1.

Keywords: AT1R; AT2R; p65.

Figure 1

AngII up-regulates the expression of AGGF1 in HUVECs. A) HUVECs were starved for 12 h and subsequently treated with AngII for 12 h at different concentrations: 10 μg/ml (9.56 μM), 20 μg/ml (19.12 μM), 40 μg/ml (38.24 μM), 60 μg/ml (57.36 μM), and 80 μg/ml (76.48 μM). Cells were harvested and used for Western blot analysis with an anti-AGGF1 antibody or control anti-GAPDH. B) HUVECs were starved for 12 h and then treated with AngII for 12 h at different concentrations: 5 μg/ml (4.78 μM), 10 μg/ml (9.56 μM), 15 μg/ml (14.34 μM), 20 μg/ml (19.12 μM), and 30 μg/ml (28.68 μM). Cells were harvested and used for Western blot analysis. C) HUVECs were treated as in B and used for real-time RT-PCR analysis. D) HUVECs were starved for 12 h, treated with 10 μg/ml (9.56 μM) of AngII with or without AT1R inhibitor losartan (IC50 = 20 nM) or AT2R inhibitor PD123319 (IC50 = 34 nM) for 12 h, and were subsequently used for Western blot analysis. *P < 0.05, **P < 0.01, ****P < 0.0001 (n = 3).

Figure 2

AngII regulates the expression of AGGF1 through the NF-κB pathway. A) Effect of JNK inhibitor SP600125 (20 μM) on Ang II-induced expression of AGGF1. B) Effect of p38 MAPK inhibitor SB203580 (10 μM) on Ang II-induced expression of AGGF1. C) Effect of NF-κB inhibitor PDTC (10 μM) on Ang II-induced expression of AGGF1. D) Effect of ERK inhibitor PD98059 (20 μM) on Ang II-induced expression of AGGF1. E) Effect of NF-κB inhibitor PDTC (10 μM) and AT1R inhibitor losartan (IC50 = 20 nM) on increased p65 phosphorylation and Ang II-induced expression of AGGF1. NS, not significant; PDTC, pyrrolidine dithiocarbamate. **P < 0.01, ***P < 0.001, ****P < 0.001 (n = 3).

Figure 3

NF-κB inhibitor QC inhibits the expression of AGGF1. A) HUVECs were treated with different concentrations of QC for 12 h, lysed, and used for Western blot analysis along with a phosphorylated p65 antibody, an anti-AGGF1 antibody, and a control anti-GAPDH antibody. B) HUVECs were treated as in A and used for real-time RT-PCR analysis to determine RELA and AGGF1 expression. C). HeLa cells were transfected with an AGGF1-promoting luciferase reporter (pGL3-AGGF1p) for 24 h, treated with different concentrations of QC for 12 h, and used for luciferase assays. The AGGF1 promoter luciferase reporter (pGL3-AGGF1p) contains the AGGF1 promoter and regulatory region (−636 bp from the translation start site) fused to the luciferase gene. ***P < 0.001 (n = 3).

Figure 4

Identification of a conserved p65 binding site at the AGGF1 promoter and regulatory region and overexpression of p65 activates the AGGF1 transcription. A) The position weight matrix of the p65 binding site at the AGGF1 promoter and regulatory region from the JASPAR Database. B) The p65 binding site is located in the region from −278 to −266 bp from the AGGF1 translation start site. C) The p65 binding site at the AGGF1 promoter (or regulatory region) shows high evolutionary conservation among different species. D) Schematic diagram of the AGGF1 promoter luciferase reporter (AGGF1-LUC WT) with the p65 binding site. A mutant report was created by mutating the p65 site from 5′-GGTCCTTCCC-3′ to 5′-GATCGTACCG-3′, resulting in a mutant luciferase reporter AGGF1-LUC MUT. E) Overexpression of p65 activates AGGF1-LUC WT, but not AGGF1-LUC MUT. HeLa cells were cotransfected with pFlag-RELA as well as the WT or MUT luciferase reporter and subsequently used for luciferase assays. ***P < 0.001 (n = 3).

Figure 5

Transcriptional factor p65 regulates transcriptional activation of AGGF1. A, B) HUVECs were transfected with pFlag-RELA (p65) for overexpression of p65 or siRNAs specific for the RELA gene for 48 h and used for Western blot analysis to determine the expression levels of p65 and AGGF1. C) HUVECs were treated as in A and B and used for real-time RT-PCR analysis to measure the expression level of AGGF1. D) HeLa cells were cotransfected with the AGGF1 promoter reporter AGGF1-LUC MUT, pRL-TK, pFlag-RELA (p65), or RELA siRNAs for 48 h, and then used for luciferase assays. **P < 0.01, ***P < 0.001 (n = 3).

Figure 6

Transcriptional factor p65 binds directly to the AGGF1 promoter. A) Schematic diagram of the 5′-UTR of the human AGGF1 gene. B) ChIP assays. The interaction between p65 and its binding site was detected using qPCR analysis with agarose gel electrophoresis. C) The data from B were quantified and plotted. D) EMSA. The p65–DNA complex can be detected when p65 is overexpressed (compare lanes 1 and 2). The p65–DNA complex was eliminated by addition of 200-fold excess of the EMSA competitor (lane 3). The p65–DNA complex can be supershifted by a specific anti-p65 antibody (lane 4). For probe design and preparation for EMSA, a pair of primers labeled with biotin (5′-TGCCGCTGGCGCCGTTGTTT-3′, 5′-TACAGAGACAGAGGGAGGAG–biotin-3′) were used to amplify a 200 bp DNA fragment containing the p65 binding site. The same pair of primers, unlabeled, was used to prepare the unlabeled DNA fragment containing the p65 binding site, which was used as the competitor in EMSA. SS complex, supershifted protein–DNA complex. NS, not significant. **P < 0.01 (n = 3).

Figure 7

AngII induces capillary tube formation by AGGF1. A) Capillary tube formation by HUVECs transfected with siNC. B) Capillary tube formation by HUVECs transfected with control siNC and treated with AngII (10 μg/ml or 9.56 μM). C) Capillary tube formation by HUVECs transfected with siRELA and treated with AngII (10 μg/ml or 9.56 μM). D) Capillary tube formation by HUVECs transfected with siAGGF1 and treated with AngII (10 μg/ml or 9.56 μM). E) AGGF1 restores the effect of siRELA on AngII-induced capillary tube formation. Capillary tube formation was observed in HUVECs transfected with siRELA and treated with AngII (10 μg/ml or 9.56 μM) and AGGF1. F) The images in (A-E) were quantified and plotted. Scale bars, 200 μm. **P < 0.01, ***P < 0.001 (n = 3).

Figure 8

AngII induces neovessels to sprout by activating AGGF1. A) Neovessels sprout from aortic rings of WT mice treated with control water. B) Neovessels sprout from aortic rings of WT mice treated with AngII (10 μg/ml or 9.56 μM). C) Neovessels sprout from aortic rings of Aggf1^+/− mice treated with control elution buffer. D) Neovessels sprout from aortic rings of Aggf1^+/− mice treated with AngII (10 μg/ml or 9.56 μM). E) Neovessels sprout from aortic rings of Aggf1^+/− mice treated with AngII (10 μg/ml or 9.56 μM) and AGGF1 (5 μg/ml). Scale bar, 200 μm . F) Quantitative data for sprout areas. ***P < 0.001, ****P < 0.0001 (n = 3).

Figure 9

AGGF1 signaling does not crosstalk with VEGF-A signaling. A) HUVECs were transfected with siRNA for VEGFR2 (siVEGFR2), treated with AngII (10 μg/ml/9.56 μM), and used for Western blot analysis to measure the level of AGGF1. B) HUVECs were transfected with siRNA for VEGFR2 (siVEGFR2), treated with AngII (10 μg/ml or 9.56 μM), and used for real-time RT-PCR analysis to measure the level of AGGF1 mRNA. C) HUVECs were transfected with siRNA for AGGF1 (siAGGF1), treated with AngII (10 μg/ml or 9.56 μM), and used for real-time RT-PCR analysis to measure the level of VEGFR2 mRNA. D) HUVECs were transfected with siRNA for AGGF1 (siAGGF1), treated with AngII (10 μg/ml or 9.56 μM), and used for real-time RT-PCR analysis to measure the level of VEGFA mRNA. NS, not significant. *P < 0.05, ***P < 0.001 (n = 3).