Angiopreventive efficacy of pure flavonolignans from milk thistle extract against prostate cancer: targeting VEGF-VEGFR signaling

Abstract

The role of neo-angiogenesis in prostate cancer (PCA) growth and metastasis is well established, but the development of effective and non-toxic pharmacological inhibitors of angiogenesis remains an unaccomplished goal. In this regard, targeting aberrant angiogenesis through non-toxic phytochemicals could be an attractive angiopreventive strategy against PCA. The rationale of the present study was to compare the anti-angiogenic potential of four pure diastereoisomeric flavonolignans, namely silybin A, silybin B, isosilybin A and isosilybin B, which we established previously as biologically active constituents in Milk Thistle extract. Results showed that oral feeding of these flavonolignans (50 and 100 mg/kg body weight) effectively inhibit the growth of advanced human PCA DU145 xenografts. Immunohistochemical analyses revealed that these flavonolignans inhibit tumor angiogenesis biomarkers (CD31 and nestin) and signaling molecules regulating angiogenesis (VEGF, VEGFR1, VEGFR2, phospho-Akt and HIF-1α) without adversely affecting the vessel-count in normal tissues (liver, lung, and kidney) of tumor bearing mice. These flavonolignans also inhibited the microvessel sprouting from mouse dorsal aortas ex vivo, and the VEGF-induced cell proliferation, capillary-like tube formation and invasiveness of human umbilical vein endothelial cells (HUVEC) in vitro. Further studies in HUVEC showed that these diastereoisomers target cell cycle, apoptosis and VEGF-induced signaling cascade. Three dimensional growth assay as well as co-culture invasion and in vitro angiogenesis studies (with HUVEC and DU145 cells) suggested the differential effectiveness of the diastereoisomers toward PCA and endothelial cells. Overall, these studies elucidated the comparative anti-angiogenic efficacy of pure flavonolignans from Milk Thistle and suggest their usefulness in PCA angioprevention.

Figure 1. Flavonolignans inhibit PCA DU145 xenograft growth through targeting proliferation and apoptosis.

DU145 xenografts were initiated and mice were administered either vehicle (CMC) or 50 and 100 mg/kg body weight doses of each diastereoisomer. (A) Tumor volume was measured and plotted as a function of time (days). Each value in the curves is mean ± SEM of 10–12 mice. (B–D) Xenograft tissues were analyzed for PCNA, TUNEL and cleaved caspase-3 (CC3) by IHC. The data shown in the bar diagrams is the mean± SEM of 4–5 samples. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; *, p ≤ 0.001; #, p ≤ 0.01; $, p ≤ 0.05.

Figure 2. Flavonolignans inhibit angiogenesis in vivo.

DU145 xenograft tissues were analyzed for CD31, nestin, VEGF and VEGFR2 by IHC. Quantitative analyses were performed using Zeiss Axioscope 2 microscope (Carl Zeiss, Germany) and photographs were originally captured (at 400x) with a Carl Zeiss AxioCam MrC5 camera with Axiovision Rel 4.5 software. The data shown in the bar diagrams is the mean ± SEM of 4–5 samples. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; *, p ≤ 0.001.

Figure 3. Flavonolignans decrease VEGFR1, HIF-1α, phosphorylated and total Akt levels in DU145 xenografts.

DU145 xenograft tissues were analyzed for VEGFR1, HIF-1α, phosphorylated Akt^ser473 and total Akt levels by IHC. Quantitative analyses were performed using Zeiss Axioscope 2 microscope (Carl Zeiss, Germany) and photographs were originally captured (at 400x) with a Carl Zeiss AxioCam MrC5 camera with Axiovision Rel 4.5 software. The data shown in the bar diagrams is the mean ± SEM of 4–5 samples. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; *, p ≤ 0.001.

Figure 4. Feeding of pure flavonolignans did not affect angiogenesis and normal histology in non-target organs.

(A–B) Lungs, liver and kidneys from each mouse were collected and analyzed for CD31 immunoreactivity as well as for histopathological analyses. Quantitative analyses were performed using Zeiss Axioscope 2 microscope (Carl Zeiss, Germany) and photographs were originally captured (at 400x) with a Carl Zeiss AxioCam MrC5 camera with Axiovision Rel 4.5 software. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B.

Figure 5. Flavonolignans inhibit angiogenesis in ex vivo and in vitro models.

(A) Flavonolignans inhibit angiogenesis ex vivo . Mouse aortas were plated on matrigel and treated with flavonolignans. The arrows in the picture mark the emerging vessels from the aortas. (B) Flavonolignans inhibit VEGF-induced tube formation in HUVEC. HUVEC were plated on matrigel and effect of diastereoisomers treatment on VEGF-induced tube formation was analyzed. Representative tubular network photomicrographs are shown at 100x (top panel). Tube length was quantified as detailed in ‘Methods’ (bottom panel). Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; *, p ≤ 0.001.

Figure 6. Flavonolignans inhibit VEGF-induced proliferation and invasion in HUVEC.

(A) HUVEC were induced with VEGF and treated with each flavonolignan in 0.5% serum media, and total cell number was analyzed after 24 h. (B) HUVEC were plated in the upper chamber with DMSO or individual diastereoisomer, while VEGF was added in the lower chamber and HUVEC invasion was studied. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; *, p ≤ 0.001; #, p ≤ 0.01; $, p ≤ 0.05.

Figure 7. Effect of flavonolignans on viability, cell cycle distribution and apoptosis in HUVEC.

(A–B) HUVEC were treated with DMSO or individual flavonolignan and analyzed for total cell number and cell cycle distribution. (C) HUVEC were treated with flavonolignans, and 24 h later, total cell lysates were prepared and analyzed for cell cycle regulators. The densitometry values presented below the bands are ‘fold change’ compared to control after loading control (α-tubulin) normalization. (D) HUVEC were treated with flavonolignans (at 30 µM dose) for 36 h and analyzed for morphology (representative photomicrographs are shown at 100x), levels of cPARP, cleaved caspase 3 and 9, and percentage apoptotic cells. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; *, p ≤ 0.001; #, p ≤ 0.01; $, p ≤ 0.05.

Figure 8. Effect of flavonolignans on VEGF-induced signaling cascade in HUVEC.

HUVEC were serum starved for 22 h, treated with diastereoisomers for 2 h and stimulated with VEGF (10 ng/ml) for 10 minutes. Total cell lysates were prepared and analyzed for mentioned signaling molecules. The densitometry values presented below the bands are ‘fold change’ compared to control after loading control (β-actin) normalization.

Figure 9. Differential effect of flavonolignans on human PCA DU145 cells and endothelial HUVEC in cell culture assays.

(A) Effect of flavonolignans (at 90 µM dose) on the three dimensional growth of DU145 cells was studied as detailed in ‘Materials and Methods’. Representative spheroids photomicrographs are shown at 100x and 400x. (B) In Transwell invasion assay, either DU145 cells (plated in the lower chamber) or HUVEC (plated in the upper chamber) were treated with individual diastereoisomers (at 30 µM dose), and HUVEC invasiveness was measured. (C) DU145 cells were treated with individual diastereoisomers (at 30 µM dose) and conditioned media was collected. HUVEC were plated on matrigel along with 0.5% FBS or conditioned media from different treatment groups; and tube formation was analyzed. Representative photomicrographs of tubular network are shown at 100x. Abbreviations: Sil A: Silybin A; Sil B: Silybin B; Iso A: Isosilybin A, Iso B: Isosilybin B; CM: Conditioned media; *, p ≤ 0.001; #, p ≤ 0.01.

Figure 10. Diastereoisomers exhibit anti-angiogenic effects through targeting signaling molecules in both prostate cancer cells and endothelial cells.

Silybin A, silybin B, isosilybin A and isosilybin B target angiogenesis in prostate tumors through targeting signaling molecules in PCA cells as well as in endothelial cells, the important component of PCA microenvironment.