Figure 1.
Harmine activates p53 by inducing phosphorylation of p53 and binding to MDM2.
(A) Chemical structure, molecular formula and 212.25 molecular weight of harmine. (B) MDM2-p53 interaction assay of harmine-treated HUVECs. After harmine treatment for 48 hours, the endogenous p53-MDM2 interaction was detected by a co-immunoprecipitation assay. (C) HUVECs were treated by harmine for 48 hours. The levels of phosphorylation of p53 at ser-15, ser-20 and ser-37 were assessed by western blot. (D) Quantitative analysis of the phosphorylation level of p53 was performed by Odyssey software. Harmine induced phosphorylation of p53 at ser-15 and ser-37, but not ser-20, in HUVECs. (E) Harmine bound to MDM2 partly, as determined by a ProteOn XPR36 Protein Interaction Array.
Table 1.
The parameters of the binding of harmine to MDM2.
Figure 2.
Harmine suppresses the degradation of p53 and increases the accumulation of p53 in the nucleus.
(A) Nuclear accumulation of p53 protein and MDM2 expression in HUVECs after harmine treatment for 48 hours. (B) Harmine prevented the degradation of p53 in HUVECs in the presence of CHX. (C) Harmine increased the nuclear levels of p53. Cells were treated with harmine for 48 hours, and then cytoplasmic and nuclear extractions were performed to detect the levels of p53 protein in the cytoplasm and nucleus by western blot. (D) Nuclear accumulation of p53 in the presence of various concentrations of harmine in HUVECs, as determined by immunofluorescence staining with an anti-p53 antibody. Cells were stained with the anti-p53 antibody (red) and nuclei were counterstained with DAPI (blue).
Figure 3.
Harmine induces the expression of p53-target genes.
(A) Protein expression levels of p21, cyclin B1, CDC2, cyclin A, CDK2, cyclin D1 and cyclin E in HUVECs treated by harmine for 48 hours. (B) Expression levels of endogenous angiogenesis inhibitors TSP-1 and Bai1 in HUVECs. Protein expression level of TSP-1 was detected followed by addition of harmine for 48 hours, the mRNA levels of angiogenesis inhibitors were detected after addition of harmine for 24 hours.
Figure 4.
Effects of Harmine on endothelial cell apoptosis
, G2/M arrest and proliferation. (A) The proportion of apoptotic cells was increased by harmine. Harmine induced cleavage of PARP and suppressed the expression of survivin and Bcl-2. The amount of cleaved PARP and the expression levels of survivin and Bcl-2 in harmine-treated HUVECs were detected by western blot analysis with specific antibodies. (B) Cell cycle arrest of HUVECs by various concentrations of harmine. Cells were treated with various concentrations of harmine for 48 hours, and then the cell cycle was analyzed by flow cytometry. (C) Growth inhibition of HUVECs by harmine, as determined by an MTS assay. All of the experiments were performed followed by addition of harmine for 48 hours. *, P<0.05; **, P<0.01; ***, P<0.001.
Table 2.
Representation of the percentage in different phases of the endothelial cell cycle distribution by harmine.
Figure 5.
Harmine inhibits HUVEC migration and tube formation.
(A) Representative images and quantitative data of a wound healing migration assay in the presence or absence of 20 µM harmine for 8 hours. (B) Results of transwell migration assays of HUVECs treated or untreated by harmine for 8 hours. Right panel indicates quantitative data. (C) Representative images of tube formation assays of HUVECs in the presence or absence of harmine (left panel) for 8 hours, and the quantitative data of tube formation assays (right panel). *, P<0.05; **, P<0.01; ***, P<0.001.
Figure 6.
Harmine inhibits angiogenesis ex vivo and in vivo.
(A) Representative images (left panel) and the optical density (right panel) of microvessels sprouting from rat thoracic aorta rings in the absence or presence of harmine. The microvessel density of the untreated group was considered as 100%. (B) Representative images of blood vessels in control eyes (left panel) and harmine-treated eyes (right panel). Black arrows indicate blood vessels. (C) The vessel length, clock number and area of blood vessels in control and harmine-treated groups. At day 7, mice were anesthetized, and images of blood vessels in control and harmine-treated eyes were recorded (n = 10). *, P<0.05; **, P<0.01.
Figure 7.
Harmine inhibits tumor growth and angiogenesis.
(A) Inhibition of solid tumor growth in harmine-treated xenograft mouse models. Images of solid tumors in control and experimental groups are shown. (B) The size of solid tumors was assessed in the absence or presence of harmine, and statistical results of solid tumors and the body weight of mice. (C) The blood vessels of xenografted tumors in control and harmine-treated groups. Blood vessels were stained using anti-vWF and anti-CD31 antibodies. Black arrows indicate positive blood vessels. (D) Growth inhibition of A549 lung cancer cells treated by harmine for 48 hours, as determined by an MTS assay. (E) The quantitative data of A549 lung cancer cells migration assays. A549 lung cancer cells were treated by harmine for 12 hours after the starvation overnight. *, P<0.05;**, P<0.01; ***, P<0.001.
Figure 8.
Proposed model for the inhibition of angiogenesis and tumor growth by harmine.
Harmine increases the activation of p53 and induces the accumulation of stable p53 in the nucleus. Stable p53 down-regulates its downstream target genes, including CDC2, cyclin B1, survivin and Bcl-2, leading to cell cycle arrest and apoptosis of HUVECs. In addition, p53 up-regulates the expression of TSP-1 and Bai1. All these events in endothelial cells result in inhibition of tumor angiogenesis and growth.