Decylubiquinone suppresses breast cancer growth and metastasis by inhibiting angiogenesis via the ROS/p53/ BAI1 signaling pathway
Jinghua Cao1 · Xiaohua Liu1 · Yang Yang1 · Bo Wei3 · Qianming Li1 · Guanquan Mao1 · Yajun He1 · Yuanyuan Li1 · Lingyun Zheng1 · Qianqian Zhang1 · Jiangchao Li1 · Lijing Wang1 · Cuiling Qi1,2
Received: 12 July 2019 / Accepted: 22 January 2020
© Springer Nature B.V. 2020
Abstract
Breast cancer is one of the most common cancers worldwide with a rising incidence, and is the leading cause of cancer- related death among females. Angiogenesis plays an important role in breast cancer growth and metastasis. In this study, we identify decylubiquinone (DUb), a coenzyme Q10 analog, as a promising anti-breast cancer agent through suppressing tumor-induced angiogenesis. We screened a library comprising FDA-approved drugs and found that DUb significantly inhib- its blood vessel formation using in vivo chick embryo chorioallantoic membrane (CAM) and yolk sac membrane (YSM) models. DUb was further identified to inhibit angiogenesis in the rat aortic ring and Matrigel plug assay. Moreover, DUb was found to suppress breast cancer growth and metastasis in the MMTV-PyMT transgenic mouse and human xenograft tumor models. To explore whether the anticancer efficacy of DUb was directly corrected with tumor-induced angiogenesis, the MDA-MB-231 breast cancer assay on the CAM was performed. Interestingly, DUb significantly inhibits the angiogenesis of breast cancer on the CAM. Brain angiogenesis inhibitor 1 (BAI1), a member of the G protein-coupled receptor (GPCR) adhesion subfamily, has an important effect on the inhibition of angiogenesis. Further studies demonstrate that DUb sup- presses the formation of tubular structures by regulating the reactive oxygen species (ROS)/p53/BAI1 signaling pathway. These results uncover a novel finding that DUb has the potential to be an effective agent for the treatment of breast cancer by inhibiting tumor-induced angiogenesis.
Keywords : DUb · Breast cancer · Tumor growth · Metastasis · Tumor angiogenesis · BAI1 · P53
Introduction
Breast cancer is one of the most common cancers world- wide, with a rising incidence, and is the leading cause of cancer-related death among females [1, 2]. Although the survival time of patients with breast cancer has been sig- nificantly prolonged by systemic therapy, the treatment of breast cancer remains extremely unfavorable and requires the development of novel strategies to suppress breast cancer growth and metastasis. It is well known that angiogenesis is one of the essential hallmarks of cancer, especially breast cancer, and aids in supplying nutrients to promote tumor growth. Angiogenesis in breast cancer is a predictor of poor prognosis and a key risk factor for breast cancer metasta- sis [3]. Thus, there is an urgent need to explore effective antiangiogenic compounds capable of significantly inhibit- ing breast carcinogenesis.
Chick embryo chorioallantoic membrane (CAM) and yolk sac membrane (YSM), two in vivo angiogenesis models, are commonly utilized to investigate tumor angiogenesis and screen angiogenic agents [4, 5]. We envisioned that many novel anticancer agents aimed at tumor angiogenesis may be discovered through high-throughput screening (HTS) of small molecule libraries using the CAM and YSM models. The coenzyme Q10 analog decylubiquinone (DUb) is a mitochondrial permeability transition pore (PTP) inhibitor and can trigger the opening of PTPs. Mitochondrial perme- ability transition (MPT) plays a key role in apoptotic and necrotic cell death. DUb-regulated PTP opening can vary depending on the tissue and cell lines used [6]. Pharmaco- logical agents have been shown to induce cell death by pro- longing PTP opening [7]. Amira Hosni-Ahmed et al reported that the combination treatment of DUb, an X-linked inhibitor of the anti-apoptotic protein (XIAP), and EDL-360 signifi- cantly inhibited glioma growth by inducing significant cell death, which shows that DUb has anticancer activity [8]. Furthermore, combination treatment of DUb with thialy- sine significantly suppressed the viability of human acute leukemia Jurkat T cells [9]. DUb also suppressed thialysine- induced Jurkat T cell apoptosis [9]. The findings indicate that DUb could exert an important effect on cancer progres- sion and treatments associated with the inhibition of angio- genesis. Furthermore, whether DUb inhibits breast cancer growth and metastasis has never been investigated, and its role in suppressing breast carcinogenesis remains unclear.
Brain angiogenesis inhibitor 1 (BAI1) is a member of the G protein-coupled receptor (GPCR) adhesion subfamily and has a long extracellular N terminus (120 kDa) that contains an integrin-binding Arg-Gly-Asp (RGD) motif, a putative hormone binding site and five antiangiogenic thrombospon- din type-1 repeats (TSRs) [10]. Vasculostatin (Vstat120), a soluble 120 kDa antiangiogenic factor, is released by cleaving BAI1 at a conserved proteolytic cleavage site [10]. BAI1 was initially found to be widely expressed in brain tissues [11, 12]. Nonetheless, more recent studies have demonstrated that BAI1 is also expressed in the neoplastic and/or nonneoplastic tissues of colon, lung, stomach, renal, stomach, pancreas and that its expression is closely related to microvessel density [13, 14]. It was also suggested that BAI1 overexpression suppresses tumor growth by inhibiting tumor-induced angiogenesis [15, 16]. Furthermore, BAI1 had an important effect on the inhibition of angiogenesis by acting as a mediator of p53, a tumor suppressor gene [17]. Some studies have also demonstrated that BAI1 is transcrip- tionally regulated by p53 [18, 19]. DUb effectively blocks mitochondrial reactive oxygen species (ROS) production by inhibiting cytochrome bc1 [20]. Further studies are neces- sary to investigate whether DUb inhibits angiogenesis by regulating the ROS/p53/BAI1 signaling pathway.
In this study, DUb was found to significantly suppress angiogenesis after a drug library was screened using CAM and YSM models. Further studies show that DUb suppresses blood vessel formation based on the rat aortic ring and the Matrigel plug assay. Herein, we observed the inhibitory role of DUb in breast cancer growth and metastasis in the sponta- neous breast cancer of MMTV-PyMT transgenic mouse and in transplanted human tumors. Further studies have dem- onstrated that DUb does not suppress vascular endothelial cell apoptosis but suppresses angiogenesis by regulating the ROS/p53/BAI1 signaling pathway, thereby inhibiting breast cancer development.
Materials and methods
Chick embryo chorioallantoic membrane (CAM) assay
The CAM assay was performed to evaluate in vivo angi- ogenic activity as described in our previous reports [4]. Briefly, fertilized eggs were incubated at 38 °C and 36.5–38.5% humidity for 9 days and then windowed. After removal of the shell membrane, either 1.6 and 3.2 µg of DUb (cat. no.: D7911, CAS NO.:55486-00-5, Molecular for- mula: C19H30O4, Molecular weight: 322.44 g/mol, Solvent: DMSO, Sigma-Aldrich, St Louis, MO, USA) or DMSO (1.6‰ or 3.2‰ in volume) was added onto the CAM of 9-day-old embryos in the absence or presence of VEGF (10 ng/chick embryo). The eggs were incubated at 37.8 °C and 60% humidity until the embryos were 11 days old. After 48 h incubation, the vascular plexus of the CAM was pho- tographed using an inverted microscope. The blood vessel density (percentage of blood vessel area over the whole area under a microscopic field) was assessed using the image analysis program IPP 6.0 (Image-Pro Plus, version 6.0, Media Cybernetics).
Chick embryo yolk sac membrane (YSM) assay
The ability of DUb to stimulate angiogenesis was further determined using the YSM model as previously described [4]. Pathogen-free fertilized chicken eggs were incubated at 38 °C for 96 h and then distributed into sterilized dishes. The blood vessels of the YSMs were laid upward, and two silastic rings with a 9.5 mm inner diameter and 12 mm out- side diameter were symmetrically placed in the blood vessel position of the YSM. Next, 40 µL of DUb (1.6 and 3.2 µg) or DMSO (0.8‰ or 1.6‰ in volume) was added into the rings on top of the well-developed vessels. The blood ves- sels within the rings were photographed at 0 h, 12 h and 24 h after treatment using an inverted microscope. The blood vessel density in the images was quantified using an Image- Pro Plus 6.0 system. High‑throughput drug screening One of small molecule compounds (2 µg, FDA-approved drug library) or DMSO (control) was added onto the CAM of 9-day-old embryos. After 48 h incubation of the embryos, the effect of the compounds on the vascular plexus formation of the CAM was evaluated. The potential antiangiogenic compounds were preliminarily screened using CAM assay. YSM assay was further utilized to screen the most potential antiangiogenic compound.
MMTV‑PyMT transgenic mice and DUb treatments
DMSO or DUb (5 mg/kg, once every 3 days for 24 days) were intravenously injected into 9-week-old MMTV-PyMT mice, which spontaneously develop breast carcinoma. The breast tumors were measured every 2 days, and the tumor volumes were calculated using the following formula:
0.52 × length × width2. The mice were sacrificed after 24 days, and then the tumors and lungs were isolated to meas- ure the weight, count pulmonary metastatic foci and perform histological analysis.
MDA‑MB‑231 and MCF‑7 xenograft tumor models
Human breast cancer MDA-MB-231 cells (1 × 106/mouse) or MCF-7 cells (5 × 106/mouse) were subcutaneously injected into the mammary fat pads of male athymic nude mice (6 weeks old). After the tumors were visible, DMSO (1% in volume) or DUb (5 mg/kg, every day) was injected into the tail veins of the athymic nude mice. The length and width of the tumors were measured, and the tumor volume was calculated using the formula: tumor vol- ume = 0.52 × length × width2 [4]. The tumor-bearing mice were sacrificed, and the tumors were isolated for analysis.
Ethics approval
All animal experiments were conducted according to rel- evant national and international guidelines. All animal procedures were approved by the Medical Research Animal Ethics Committee of Guangdong Pharmaceutical University.
Immunohistological staining
The tumor tissues were embedded and sectioned. Immuno- histochemical staining was performed on the 3-µm sections. After dewaxed and hydrated, the sections were incubated anti-CD31 (cat. no. ab28364, Abcam, Cambridge, CB, UK), anti-Ki67 (cat. no. ab15580, Abcam) or anti-BAI1 (cat. no. NB110-81586, Novus, CO, UK) primary antibodies over- night. The next day, the sections were incubated HRP-con- jugated secondary antibodies. All the sections were stained with diaminobenzidine (DAB), then counterstained with hematoxylin. The number of CD31+ vessels or the percent- age of Ki67+ cells was counted in a 200 × field or 400 × field, respectively.
Matrigel plug assay
The Matrigel plug assay was performed as previously described [21]. Briefly, five- to six-week-old female Balb/c mice (Guangdong Medical Laboratory Animal Center, Guangzhou, China) were subcutaneously injected with 500 µL of Matrigel (cat. no. 356230, BD Biosciences, Becton Dickinson, San Jose, CA) containing heparin (60 units, Cisen Pharmaceutical Company, Shandong, China), FGF-2 (150 ng/mL, cat. no. 3139-FB/CF, R&D systems, Minne- apolis, MN, USA) and either DMSO or DUb (155.09 µM). The Matrigel plugs were harvested after 7 days and pho- tographed using a Canon Power shot G10 digital camera. Matrigel plugs were fixed, embedded in paraffin, and sec- tioned. Hemoglobin content was detected using enzyme- linked immunosorbent assay (ELISA, cat. no. L160803083, Cloud-Clone Corp., Wuhan, Hubei, China). Immunofluores- cence for CD31 and hemoglobin content were conducted to analyze angiogenesis.
Rat aortic ring assay
The rat aortic ring assay was performed as previously described [21]. Briefly, aortas isolated from 7- to 9-week- old male Sprague-Dawley rats were cut into approximately 1 mm long rings. The rings were then placed into 48-well plates containing 100 µL of Matrigel and incubated at 37 °C in 5% CO2 for 30–45 min. EGM medium (cat. no. CC-2935, LONZA, Walkersville, MD, USA) supplemented with DUb (77.54 µM) or DMSO (2‰) was added to the wells and incubated at 37 °C. After 7 days, the aortic rings were fixed with 4% formalin and photographed using the inverted microscope. The number of microvessels emerging from the aortic rings was counted using the image analysis program IPP 6.0.
MDA‑MB‑231 breast cancer assay on CAM
To assess the direct effect of DUb on angiogenesis, an MDA- MB-231-bearing CAM assay was performed as previously described [4]. Briefly, after incubation at 37 °C for 10 days, the eggs were windowed in the shell. MDA-MB-231 breast cancer cells were added to the CAM of 10-day-old chick embryos. Three days later, 40 µL of DUb (1.6 or 3.2 µg) or DMSO was loaded in the silastic rings on the top of the CAM. The window was resealed with adhesive tape. The eggs were incubated at 37 °C for 3 days and cut out along the axis of the median line. The blood vessels of the tumor on the CAM were photographed using a stereomicroscope. The tumors were isolated from the CAM, the length and width of the tumors were measured, and the tumor volume was calculated (length × width2 × 0.52). The antiangiogenic effect of DUb was assessed by comparing the number of second- and third-order blood vessels on the CAM treated with DMSO (control).
Quantitative real‑time PCR
After HUVECs were treated with DUb (25 µM) or DMSO (0.16‰) for 48 h, total RNA was extracted from the cells. A quantitative real-time PCR (qRT-PCR) array of angi- ogenesis-related genes was performed according to the TaqMan-based qPCR assay kit (cat. no. PAMM-024A, SA Biosciences). All qRT-PCR assays were performed using an ABI PRISM 7000HT Sequence Detection System (Applied Biosystems).
Western blotting
HUVECs, human dermal microvascular endothelial cells (HDMECs), or tumor tissues treated with DUb (25 µM) or DMSO (0.32‰) were lysed in RIPA lysis buffer (Thermo Scientific, Scotts Valley, CA, USA) to extract total protein. Proteins were separated on SDS-polyacrylamide gels and transferred onto nitrocellulose membranes (Millipore, Bill- erica, MA, USA). The membranes were blocked in skim milk and incubated with primary antibodies targeting BAI1 and GAPDH at 4 °C overnight followed by horseradish per- oxidase-coupled IgG. The protein bands were detected using an Odyssey Infrared Imager (LICOR Bioscience, Lincoln, NE, USA).
Colony formation assay
MD-MBA-231 cells (1.5 × 103) and MCF-7 cells (2 × 103) were seeded into the 6-well plates. After the plates were treated with DUb (25 µM or 50 µM) or DMSO (0.32‰ or 0.64‰) for 7 days at 37 °C, the colonies were fixed with 4% paraformaldehyde and stained with 1% crystal violet solution for 20 min. The colonies were photographed and counted.
Cell migration assay
Cell migration assays were performed to determine the ability of DUb to suppress HUVEC and breast cancer cell migration. HUVECs, MDA-MB-231 cells or MCF-7 cells were suspended in serum-free EBM or DMEM medium con- taining DUb (25 µM or 50 µM) or a corresponding dose of DMSO (0.32‰ or 0.64‰). The HUVEC suspension was added to the upper chambers of the transwells, and culture medium containing 20% FBS plus various growth factors (SDF1, HB-EGF, VEGF and HGF) was introduced to the lower chambers. MDA-MB-231 cell or MCF-7 cell suspen- sion was added to the upper chambers of the transwells, and culture medium containing 50% FBS was introduced to the lower chambers. The HUVEC cells, MDA-MB-231 cells or MCF-7 cells on the upper side of the membrane were care- fully removed with a cotton swab after 12 h, 16 h or 20 h of incubation, respectively. After the cells were fixed in 4% paraformaldehyde overnight, the cells were stained with 1% crystal violet for 20 min, and the number of migratory cells was counted under an inverted light microscope.
Transwell invasion assay
Cell invasion assay was performed to determine the abil- ity of DUb to suppress breast cancer cell invasion. MDA- MB-231 cells or MCF-7 cells were suspended in serum- free DMEM medium containing DUb (25 µM or 50 µM) or a corresponding dose of DMSO (0.32‰ or 0.64‰). The cell suspension was added to the upper chambers of the transwells pre-coated with Matrigel (BD Biosciences), and culture medium containing 50% FBS was introduced to the lower chambers. After being respectively incubated for 24 h or 44 h, the cells on the upper side of the membrane were carefully removed with a cotton swab. The invaded cells were fixed and stained with 1% crystal violet for 20 min, and the number of invaded cells was counted.
Tube formation assay
The role of DUb in vascular tube formation was confirmed using HDMECs and HUVECs. Ninety-six-well plates were coated with 30 µL of growth factor-reduced Matrigel (BD Pharmingen) and incubated for 1 h at 37 °C. HDMECs or HUVECs (1.5 × 105 cells/mL) in EBM in the presence or
absence of VEGF (20 ng/mL) were seeded into 96-well plates. After the plates were incubated for 12 h at 37 °C, the presence of tube-like structures was photographed and ana- lyzed using a phase contrast inverted microscope. The total tube length was calculated using Image-Pro Plus software (Media Cybernetics).
Statistical analysis
Data analyses were performed using the GraphPad Prism 5 software package (GraphPad Software, CA), and the statisti- cal charts were drawn using SigmaPlot software. The results were compared using a two-tailed Student’s t-test, as the prerequisites (independence and normal distribution) were satisfied. Data included tumor volume, IHC results, tumor weight, and tube formation. Differences between the groups were considered significant at p < 0.05. Results DUb suppresses angiogenesis in the CAM and YSM models To discover potential and novel antiangiogenic molecules, we screened a library comprising 96 FDA-approved drugs utilizing the CAM and YSM assays. Three potential antian- giogenic compounds were initially found to inhibit angio- genesis in the CAM model and were subjected to an inves- tigation of the role in suppressing angiogenesis in the YSM model. After the YSM assays, DUb (its chemical structure shown in Fig. 1a) was considered to be the most potential antiangiogenic compound (Supplemental Table S1).We further investigated the effect of DUb on angiogen- esis using the CAM and YSM models. DUb (1.6 or 3.2 µg) or DMSO was added to the vascular plexus of the CAM. Images of the vascular plexus of the CAM were taken after 48 h (Fig. 1b). Compared with the control, DUb significantly decreased the density of CAM blood vessels in a concentra- tion-dependent manner (Fig. 1c). Furthermore, DUb also significantly repressed VEGF-induced vessel formation of CAM in a dose-dependent manner (Supplemental Fig. 1). The YSM model was further utilized to determine the effect of DUb on angiogenesis. Chick embryos were transferred into sterilized culture dishes and treated with DUb (1.6 and 3.2 µg) or DMSO for 24 h within plastic rings. Pictures of the vascular beds of the YSM within the plastic rings were taken at 0 h, 12 h and 24 h after treatment (Fig. 1d). Compared with the control, DUb significantly decreased the density of YSM blood vessels (Fig. 1e, f). These results demonstrated that DUb significantly impairs angiogenesis during the development of chick embryos. Fig. 1 DUb suppresses angio- genesis in the CAM and YSM models. CAM and YSM were treated with DUb (1.6 or 3.2 µg) or DMSO to confirm the effect of DUb on angiogenesis. a The chemical structure of DUb. b Images of the blood vessels of chick embryos treated with 1.6 or 3.2 µg of DUb or DMSO (1.6‰ or 3.2‰ in volume) for 48 h. The lower panels show the magnification of the blood ves- sel plexus in the upper image. c The relative blood vessel density on the CAM treated with 1.6 or 3.2 µg of DUb or DMSO. d To determine the growth rate of blood vessels, the images of the vascular beds of the YSM within plastic rings were captured at 0 hr. Images of the vascular beds of the YSM within plastic rings at 0 h, 12 h and 24 h after treatment. e Bar chart demonstrates the relative blood vessel density on the YSM following treatment with 1.6 or 3.2 µg of DUb or DMSO (0.8‰ or 1.6‰ in volume) for 12 h. f Bar chart demonstrates the relative blood vessel density for control and DUb treatment for 24 h. Data are presented for at least three independent exper- iments. *p < 0.05; **p < 0.01; ***p < 0.001. DUb inhibits microvessel formation in Matrigel plugs and outgrowth of aortic ring microvessels To further evaluate the antiangiogenic properties of DUb, we performed Matrigel plug and rat aortic ring assays. The Matrigel plug assays were performed by subcutaneous injec- tion of Matrigel containing heparin, FGF-2, or DUb (155.09 µM) into Balb/c mice. Seven days later, the Matrigel plugs were removed and photographed. Matrigel plugs of the blank control without FGF-2 and DUb were white. Plugs containing FGF-2 and DMSO present as deep red, indicat- ing that a large number of new blood vessels formed in the plugs. The plugs containing FGF-2 and DUb present as light red, demonstrating that there were very few blood vessels in the plugs (Fig. 2a). Next, the hemoglobin content in the plugs was detected by ELISA. The hemoglobin content of the plugs treated with DUb was significantly decreased compared with that in the DMSO-treated plugs (Fig. 2b). Immunofluorescent staining of blood vessels in the plugs demonstrated that the area of blood vessels from the plugs treated with DUb was significantly smaller than that from plugs treated with DMSO (Fig. 2c, d). The rat aortic ring assay was also performed to determine the effect of DUb (77.54 µM) on angiogenesis. As shown in Fig. 2e and f, treatment of aortic rings with DUb significantly inhibited capillary outgrowth. Fig. 2 Effects of DUb on Matrigel plug angiogenesis and microves- sel outgrowth of the aortic rings. a Representative pictures of the blank control group without FGF-2, DMSO group containing DMSO (0.5‰ in volume) and FGF-2, and DUb group containing DUb (155.09 µM) and FGF-2. b The hemoglobin content in the plugs. c After the plugs were embedded and sectioned, the sections were immunostained with CD31 antibodies. Representative images of immunostaining are shown. d The effect of DUb on the microvessel density. e Representative images of sprouts from aortic rings treated with DMSO or DUb. f The effect of DUb on microvessel outgrowth from aortic rings. Scale bars = 75 µm in c. *p < 0.05, ***p < 0.001. DUb suppresses breast cancer growth and metastasis in MMTV‑PyMT transgenic mice Because DUb suppressed angiogenesis in the chick embryo, Matrigel plugs and aortic rings, we supposed that DUb may also inhibit tumor-induced angiogenesis, thereby impeding tumor growth and metastasis. To address this, we detected the effect of DUb on the growth and metastasis of breast cancer using MMTV-PyMT mice. MMTV-PyMT mice were randomly divided into the control and DUb treat- ment groups. DUb (5 mg/kg) was injected into 9-week-old MMTV-PyMT mice every 3 days for 24 days. The tumor volumes were measured every other day. The tumor vol- umes of DUb-treated MMTV-PyMT mice were smaller than those of control mice at the corresponding times (Fig. 3a). The mice were sacrificed, and the tumors were removed and weighed 24 days after DUb treatment. The tumor weights in DUb-treated mice were significantly decreased compared with those in the control mice (Fig. 3b). Interestingly, the number of metastatic foci of the lung surface in DUb-treated mice was fewer than that in DMSO-treated mice (Fig. 3c). The results of immunohistological staining against CD31 showed that the microvessel density in DUb-treated tumors was lower than that in the control tumors (Fig. 3d). Immu- nohistological staining for Ki67 was performed on the tumor sections. There was a significant reduction in breast cancer cell proliferation in DUb-treated tumor tissues as compared with the control (Supplemental Fig. 2a). DUb inhibits MDA‑MB‑231 and MCF‑7 xenograft tumor growth Given that DUb is capable of inhibiting mouse-derived tumor growth, we speculated that DUb may also suppress human-derived tumor growth by inhibiting angiogenesis. To this end, we established a xenograft tumor model in nude mice using MDA-MB-231 and MCF-7 human breast cancer cells. Interestingly, compared to DMSO, DUb significantly reduced the tumor volume of xenografted human breast cancer cells (Figs. 4a, 5a). After the mice were sacrificed 9 days or 25 days after DUb treatment, the tumors were removed and weighed. We found that the tumor weights in DUb-treated mice were lighter than those in the control mice (Figs. 4b, 5b). The results of immu- nohistological staining against CD31 antibodies showed that the microvessel density in DUb-treated tumors was significantly reduced compared with that in the control tumors (Figs. 4c, 5c). Immunohistological staining for Ki67 was performed on MDA-MB-231 xenograft tumor sections. There was a significant reduction in breast cancer MDA-MB-231 cell proliferation in DUb-treated tumor as compared with the control (Supplemental Fig. 2b). Col- ony formation, transwell migration and invasion assays were further performed to investigate the effect of DUb on the breast cancer MD-MBA-231 and MCF-7 cells. We found that DUb significantly suppressed breast cancer cell colony formation (Supplemental Fig. 3), migration and invasion (Supplemental Fig. 4). DUb, a PTP inhibitor, has an important effect on cell necrosis. To determine the effect of DUb on breast cancer necrosis, the MCF-7 xeno- graft tumors were sectioned and stained with H&E. The necrotic and total areas of the MCF-7 xenograft tumors were measured. There is no difference in the percentage of necrotic tumor area between DMSO group and DUb group (Supplemental Fig. 5). Fig. 3 DUb inhibits tumor growth and metastasis in spontaneous breast cancer. Female MMTV-PyMT mice (9 weeks of age) were intravenously treated with DMSO (1% in volume) or DUb (5 mg/ kg) once every 3 days for 24 days. a The tumor volumes were meas- ured every other day. DUb-treated MMTV-PyMT mice developed smaller tumors than did the DMSO-treated mice. b After the mice were sacrificed, the breast tumors were isolated and weighed. The results show that DUb reduced the weight of the tumors. c After the mice were sacrificed and lungs were harvested, the metastatic foci on the lung surfaces were counted. The results show that DUb-treated MMTV-PyMT mice developed fewer lung metastatic foci than did DMSO-treated mice. d Representative immunohistological images from MMTV-PyMT mice treated with DUb or DMSO. The statistical chart shows that DUb suppressed tumor-induced angiogenesis. MVD microvascular density. Scale bars = 50 µm in d. *p < 0.05; **p < 0.01; ***p < 0.001. Fig. 4 DUb inhibits angio- genesis and MDA-MB-231 xenograft breast cancer growth in nude mice. a, b Effects of DUb (5 mg/kg) on human breast cancer growth. Breast cancer- bearing mice were intravenously administered DUb. The tumor volume (a) and weight (b) were assessed. c Immunohistochemi- cal images of CD31-positive endothelial cells in tumor tissue sections were detected and showed that the microvas- cular density was decreased in the tumor tissues treated with DUb. Scale bars = 50 µm in c. *p < 0.05; **p < 0.01; ***p < 0.001. Fig. 5 DUb suppresses MCF-7 xenograft tumor growth and angiogenesis in nude mice. a, b Effects of DUb (5 mg/kg) on MCF-7 xenograft breast cancer growth. Breast cancer-bearing mice were administered DUb.The tumor volume (a) and weight (b) were assessed. DUb- treated breast cancer-bearing mice developed smaller and lighter tumors than did DMSO- treated mice. c Representative immunohistological images from MCF-7 xenograft tumor- bearing mice treated with DUb or DMSO. The statistical chart demonstrated that DUb signifi- cantly inhibited tumor-induced angiogenesis. Scale bars = 50 µm in c. *p < 0.05; **p < 0.01; ***p < 0.001. DUb suppresses breast cancer growth and angiogenesis on the CAM To further investigate the direct correlation between DUb and tumor-induced angiogenesis, a breast cancer assay on the CAM was established and treated with DUb (1.6 and 3.2 µg) to exert its role in tumor angiogenesis. Images of the vascular plexus of cancer xenografts on the CAM (Fig. 6a, b) are shown. Compared with the control, DUb significantly decreased the tumor volume and the vascular density growth rate of cancer xenografts on the CAM in a concentration- dependent manner (Fig. 6c, d). These results show that DUb significantly inhibits breast cancer growth by decreasing tumor-induced angiogenesis. DUb suppresses HUVEC proliferation, migration and tube formation Because DUb impaired angiogenesis in the CAM and YSM assays, we explored whether DUb inhibits HUVEC migration and tube formation. We firstly found that DUb significantly suppressed the HUVEC proliferation (Fig. 7a). We found that DUb markedly inhibited HUVEC migra- tion inside Boyden Chambers in a dose-dependent manner (Fig. 7b, d). Meanwhile, HUVEC tube formation was evalu- ated, and DUb was demonstrated to significantly prevent HUVECs from developing spider-like microvascular capil- laries (Fig. 7c, e). These results showed that DUb suppressed angiogenesis by inhibiting the migration and tube formation ability of vascular endothelial cells. Fig. 6 DUb inhibits breast cancer-induced angiogenesis. MDA-MB-231 breast cancer cells were introduced onto the CAM. Breast cancers on the CAM were treated with 1.6 or 3.2 µg of DUb or DMSO (0.8‰ or 1.6‰ in volume) for 3 days. a Representative images of the blood vessels of breast cancer on CAM treated with DMSO or DUb for 3 days. The tumors are shown in the blue dotted squares. b The panels show the higher magnification of the black dotted squares in a. c, d The effects of DUb on the tumor volume and the vascular density growth rate. Data are presented for at least three independent experiments. Scale bars = 500 µm in a; 200 µm in b. *p < 0.05; **p < 0.01; ***p < 0.001. Fig. 7 DUb suppresses vascular endothelial cell proliferation and migration and tube formation. a The number of proliferative cells was significantly decreased in the presence of DUb. b, d The migration ability of HUVECs was evaluated using a transwell assay. HUVECs treated with DMSO (0.16‰ or 0.32‰ in volume) or various concen- trations of DUb (25 µM or 50 µM) were added to the upper chamber and allowed to migrate through the membrane for 12 h. The number of migrated cells was photographed and counted. The number of migrated cells was significantly decreased in the presence of DUb. c, e HUVECs were introduced into the Matrigel in the presence of the indicated concentration of DUb or DMSO, and tube formation was observed and recorded for 5 h after treatment. Data are presented for at least three independent experiments. Scale bars = 100 µm in d and e. *p < 0.05, ***p < 0.001. DUb suppresses angiogenesis through the ROS/p53/ BAI1 signaling pathway To analyze angiogenesis-associated genes associated with DUb, a qRT-PCR array was performed. We found that BAI1 expression was significantly upregulated (Fig. 8a). qRT-PCR was further performed to determine whether BAI1 expres- sion was upregulated in HUVECs treated with DUb. BAI1 expression was found to be upregulated in HUVECs treated with DUb (Fig. 8b). Furthermore, western blotting results demonstrated that p53 and BAI1 expression was increased in HUVECs and HDMECs treated with DUb (Fig. 8c–f). To further determine the effect of DUb on the protein expression of BAI1 in the breast cancer tissues, we extracted the total protein of MDA-MMTV-PyMT-driven tumor and found that DUb significantly increased the BAI1 protein expression in the tumor of MMTV-PyMT mice (Fig. 8g). We also detected BAI1 protein expression in the MDA-MB-231 and MCF-7 xenograft tumors using IHC staining and found that BAI1 expression was significantly increased in DUb-treated tumor tissues, whereas BAI1 was expressed not only in the blood vessels but also in the tumor tissues (Supplemental Fig. 6). Next, we measured the effect of DUb on ROS generation and found that DUb significantly induced ROS generation in HUVECs (Fig. 8h). P53 silencing also decreased BAI1 expres- sion in HDMECs treated with DUb (Fig. 8i, j). We evaluated the importance of BAI1 in angiogenesis by detecting whether silencing BAI1 can promote vascular structure formation in HDMECs. These results indicated that microvascular tube for- mation increased after BAI1 expression was silenced (Fig. 8k, l). Furthermore, HDMEC tube formation was rarely affected by DUb after silencing BAI1 expression (Fig. 8k, l). Although DUb is a PTP inhibitor and plays a key role in apoptotic cells, DUb did not affect HUVEC apoptosis (Supplemental Fig. 7). Fig. 8 DUb suppresses angiogenesis through the ROS/p53/BAI1 signaling pathway. a qRT-PCR array analysis shows that BAI1 expression is upregulated in HUVECs treated with DUb (25 µM). b qRT-PCR was further performed to confirm the qRT-PCR array results. c, d DUb (25 µM) significantly induced the protein expression of p53 and BAI1 in HUVECs. e, f DUb (25 µM) significantly induced the protein expression of p53 and BAI1 in HDMEC cells. g DUb sig- nificantly increased the protein expression of BAI1 in the tumors of MMTV-PyMT mice. h DUb (25 µM) induced ROS generation. i, j p53 knockdown significantly suppresses BAI1 protein expression in HDMECs. k, l BAI1 silencing with and without DUb (25 µM) signifi- cantly promoted HDMEC tube formation. Furthermore, the ability of BAI1-silenced HDMECs to form tubular structures was not affected after DUb treatment. Data are presented for at least three independent experiments. Scale bars = 100 µm in j. **p < 0.01; ***p < 0.001. Discussion In this study, we present an important finding that DUb, a coenzyme Q10 analog, can significantly inhibit breast can- cer growth and metastasis by suppressing tumor-induced angiogenesis. DUb suppresses angiogenesis via the ROS/ p53/BAI1 signaling pathway in vascular endothelial cells. Therefore, DUb results in a significant reduction in the microvessel density, which attenuates breast cancer growth and metastasis in the spontaneous breast carcinoma MMTV- PyMT mice and in xenograft mouse models with MDA- MB-231 and MCF-7 human breast carcinoma cells. Breast cancer is one of the most diagnosed cancers in females worldwide, with a rapidly increasing trend [22]. Although the incidence of breast cancer is high, very few effective drugs are available. The embryonic CAM and YSM assays are often used for drug screening because they are inexpensive, simple and highly reproducible [4]. Consequently, we used CAM and YSM assays to screen a drug library and found that DUb has a potent inhibitory role in vascular structure formation within the CAM and YSM models. In addition, we further found that DUb (1.6 µg or 3.2 µg) significantly inhibits vessel formation within the CAM and YSM. Given that DUb is a small molecular compound, 1.6 µg or 3.2 µg DUb for a chick embryo may be high. It is worth detecting the effect of fewer dosage of DUb than ever on angiogenesis within the CAM and YSM. DUb also significantly suppressed microvessel formation in an in vivo Matrigel plug assay and an ex vivo rat aortic ring assay. As such, we speculated that DUb might be an effective drug to inhibit tumor progression by suppressing angiogenesis. On the other hand, one recent study showed that the combination treatment of DUb with either EDL-360 or thialysine could significantly suppress the growth of glioma or human acute leukemia [8]. However, the effect of DUb on cancer growth and metastasis has never been elucidated. We found that DUb significantly attenuates breast cancer growth and metastasis of MMTV-PyMT-driven breast car- cinogenesis and xenografted human breast cancers. In addi- tion, the results of immunohistochemical staining against CD31 demonstrated that DUb attenuates breast cancer pro- gression by suppressing tumor-induced angiogenesis. We further investigated whether the anticancer capability of DUb is also attributed to its effect on the breast cancer cells. We found that DUb significantly inhibited breast cancer cell colony formation, migration and invasion, indicating that the anticancer effects of DUb are not just ascribed to its antiangiogenic properties. The results demonstrate that DUb restrains tumor-induced angiogenesis, thereby inhibit- ing breast cancer growth and metastasis, and have provided direct evidence of the cancer therapeutic effects of DUb. To better understand how DUb inhibits angiogenesis, we performed a qRT-PCR array to analyze DUb-treated vascular endothelial cells for changes in the expression of angiogenesis-related genes. We identified BAI1 as a potential target and found that DUb inhibits angiogenesis through the upregulation of BAI1 expression in vascular endothelial cells, including HUVECs and HDMECs. It has been reported that BAI1 is expressed in HEMECs but not HUVECs [16]. Our in vitro experiment also demonstrated that DUb did not affect HEMEC tube formation when BAI1 was silenced. DG Duda et al reported that BAI1 has a domain containing p53 functional binding sites and is trans- activated by p53 [15]. Meanwhile, BAI1 was considered a mediator in the p53 signaling pathways and an efficient target of tumor-related angiogenesis [23]. We also demon- strated that DUb significantly induced the protein expression of p53 and BAI1 in vascular endothelial cells. Furthermore, p53 silencing significantly decreased BAI1 protein expres- sion in HMDEC cells. Previous studies have demonstrated that adaphostin, a dihydroquinone derivative, can induce the generation of ROS as a potential anticancer agent [24, 25]. Many chemo- therapeutic and radiotherapeutic agents kill tumor cells by increasing ROS stress [26, 27]. The effect of ubiquinone analogs on ROS generation depends on the cell lines [20, 28]. Ubiquinone 0 induces ROS production in Clone-9 cells, while it reduces ROS production in MH1C1 cells. DUb inhibits ROS production in the fibroblast cell lines, Clone-9 and MHICI cell lines [20]. However, we found that DUb increased the ROS production in the HUVECs. There- fore, DUb induced ROS generation in a cell line dependent manner. ROS have been shown to be involved in angiogen- esis by targeting transcription factors or tumor suppressor genes, such as activator protein-1 (AP-1), hypoxia-inducible factor-1 alpha (HIF-1a), nuclear factor kappa-light-chain- enhancer of activated B cells (NF-ƙB), and p53 [29, 30]. Moreover, the ROS/p53/BAI1 signaling pathway has an important effect on tumor angiogenesis. Hence, the block- ade of ROS/p53/BAI1 signaling by DUb is an effective therapeutic strategy for suppressing tumor angiogenesis. We also found that DUb significantly enhanced the produc- tion of ROS. Thus, the accumulating evidence demonstrates that DUb may inhibit angiogenesis by targeting the ROS/ p53/BAI1 signaling pathway in vascular endothelial cells. The qRT-PCR array showed that the expression of another angiogenesis-related genes, such as EGF, F3 et al., was also changed in the DUb-treated HUVECs, indicating that DUb might not only signal through ROS/P53/BAI1 signaling pathway. It has been reported that DUb has a significant effect on the mitochondrial function by targeting cytochrome bc1 [20]. Diebold et al. found that suppression of respiratory chain complex III impaired endothelial cell proliferation, but not migration by decreasing the NAD+/NADH ratio [31]. Therefore, further research is needed to confirm the effect of DUb on angiogenesis by targeting the respiratory chain complex. Taken together, the data in this study have demonstrated that DUb is capable of suppressing breast cancer growth and metastasis in tumor xenograft and MMTV-PyMT transgenic mouse models. The inhibitory role was dominantly due to the ability of DUb to suppress tumor-induced angiogenesis. Furthermore, the ROS/p53/BAI1 signaling pathway has been identified as one of the dominant antiangiogenic sign- aling pathways of DUb. Finally, DUb may be a potential candidate for the effective treatment of breast cancer patients because of its antiangiogenic effect. Acknowledgements The authors thank Huiping Wang and Min Zhang for conducting preliminary experiments and providing techni- cal assistance. This work was supported by grants from the National Science Foundation of China (81773095, 31500966 to Cuiling Qi), the Science and Technology Planning Project of Guangdong Prov- ince (2015A020211029 to Cuiling Qi, 2017A010103009 to Bo Wei), the Science and Technology Planning Project of Guangzhou City (201607010135 to Cuiling Qi), and the Major Special Project of Guangdong Science and Technology Plan (2017B020227009 to Bo Wei). Author contributions CLQ and LJW contributed to the conception and design of the manuscript. CLQ, XHL, JHC, YY, BW, QML, GQM, XHL, QQZ, JCL and LJW evaluated the development of methodology. CLQ, XHL, YY and JHC analyzed and interpretated the data. CLQ wrote, reviewed and revised the manuscript. CLQ, XHL, BW, QML, GQM, XHL, QQZ, JCL and LJW supervised the study. All authors read and approved the final manuscript. Compliance with ethical standards Conflict of interest The authors declare that they have no conflicts of interest. References 1. Nagini S (2017) Breast cancer: current molecular therapeu- tic targets and new players. Anticancer Agents Med Chem 17(12):152–163 2. Ganz PA, Goodwin PJ (2015) Breast cancer survivorship: where are we today? Adv Exp Med Biol 862:1–8 3. Li Q, Kan X, Yin J, Sun L, Wang Y, Li Y, Yang Q, Xiao H, Chen Y, Weng X, Cai W, Zhu X (2018) Chamaejasmine B induces the anergy of vascular endothelial cells to VEGFA pro-angiogenic signal by autophagic regulation of VEGFR2 in breast cancer. Front Pharmacol 22:963–974 4. Qi C, Zhou Q, Li B, Yang Y, Cao L, Ye Y, Li J, Ding Y, Wang H, Wang J, He X, Zhang Q, Lan T, Lee KK, Li W, Song X, Zhou J, Yang X, Wang L (2014) Glipizide, an antidiabetic drug,suppresses tumor growth and metastasis by inhibiting angiogen- esis. Oncotarget 5(20):9966–9979 5. Tufan AC, Satiroglu-Tufan NL (2005) The chick embryo chorioal- lantoic membrane as a model system for the study of tumor angio- genesis, invasion and development of anti-angiogenic agents. Curr Cancer Drug Targets 5(4):249–266 6. Devun F, Walter L, Belliere J, Cottet-Rousselle C, Leverve X, Fontaine E (2010) Ubiquinone analogs: a mitochondrial perme- ability transition pore-dependent pathway to selective cell death. PLoS ONE 5(7):e11792 7. Armstrong JS (2006) Mitochondria: a target for cancer therapy. Br J Pharmacol 147(3):239–248 8. Hosni-Ahmed A, Sims M, Jones TS, Patil R, Patil S, Abdelsamed H, Yates CR, Miller DD, Pfeffer LM (2014) EDL-360: a potential novel antiglioma agent. J Cancer Sci Ther 6(9):370–377 9. Jun DY, Rue SW, Han KH, Taub D, Lee YS, Bae YS, Kim YH (2003) Mechanism underlying cytotoxicity of thialysine, lysine analog, toward human acute leukemia Jurkat T cells. Biochem Pharmacol 66(12):2291–2300 10. Kaur B, Cork SM, Sandberg EM, Devi NS, Zhang Z, Klenotic PA, Febbraio M, Shim H, Mao H, Tucker-Burden C, Silverstein RL, Brat DJ, Olson JJ, Van Meir EG (2009) Vasculostatin inhib- its intracranial glioma growth and negatively regulates in vivo angiogenesis through a CD36-dependent mechanism. Cancer Res 69(3):1212–1220 11. Hamann J, Hsiao CC, Lee CS, Ravichandran KS, Lin HH (2016) Adhesion GPCRs as modulators of immune cell function. Handb Exp Pharmacol 234:329–350 12. Moon SY, Shin SA, Oh YS, Park HH, Lee CS (2018) Understand- ing the role of the BAI subfamily of adhesion G protein-coupled receptors (GPCRs) in pathological and physiological conditions. Genes (Basel) 9(12):E597 13. Izutsu T, Konda R, Sugimura J, Iwasaki K, Fujioka T (2011) Brain-specific angiogenesis inhibitor 1 is a putative factor for inhibition of neovascular formation in renal cell carcinoma. J Urol 185(6):2353–2358 14. Mathema VB, Na-Bangchang K (2017) Regulatory roles of brain- specific angiogenesis inhibitor 1 (BAI1) protein in inflammation, tumorigenesis and phagocytosis: a brief review. Crit Rev Oncol Hematol 111:81–86 15. Duda DG, Sunamura M, Lozonschi L, Yokoyama T, Yatsuoka T, Motoi F, Horii A, Tani K, Asano S, Nakamura Y, Matsuno S (2002) Overexpression of the p53-inducible brain-specific angio- genesis inhibitor 1 suppresses efficiently tumour angiogenesis. Br J Cancer 86(3):490–496 16. Cork SM, Kaur B, Devi NS, Cooper L, Saltz JH, Sandberg EM, Kaluz S, Van Meir EG (2012) A proprotein convertase/MMP-14 proteolytic cascade releases a novel 40 kDa vasculostatin from tumor suppressor BAI1. Oncogene 31(50):5144–5152 17. Nam DH, Park K, Suh YL, Kim JH (2004) Expression of VEGF and brain specific angiogenesis inhibitor-1 in glioblastoma: prog- nostic significance. Oncol Rep 11(4):863–869 18. Nishimori H, Shiratsuchi T, Urano T, Kimura Y, Kiyono K, Tat- sumi K, Yoshida S, Ono M, Kuwano M, Nakamura Y, Tokino T (1997) A novel brain-specific p53-target gene, BAI1, containing thrombospondin type 1 repeats inhibits experimental angiogen- esis. Oncogene 15(18):2145–2150 19. Shiratsuchi T, Nishimori H, Ichise H, Nakamura Y, Tokino T (1997) Cloning and characterization of BAI2 and BAI3, novel genes homologous to brain-specific angiogenesis inhibitor 1 (BAI1). Cytogenet Cell Genet 79(1–2):103–108 20. Armstrong JS, Whiteman M, Rose P, Jones DP (2003) The Coenzyme Q10 analog decylubiquinone inhibits the redox- activated mitochondrial permeability transition: role of mitcohondrial [correction mitochondrial] complex III. J Biol Chem 278(49):49079–49084 21. Qi C, Li B, Yang Y, Yang Y, Li J, Zhou Q, Wen Y, Zeng C, Zheng L, Zhang Q, Li J, He X, Zhou J, Shao C, Wang L (2016) Glipizide suppresses prostate cancer progression in the TRAMP model by inhibiting angiogenesis. Sci Rep 6:27819 22. Ţenea-Cojan T, Macovei A, Păun I, Costin AI, Georgescu CV, Georgescu CC, Vladu IM, Ene CG, Radu L (2018) Assessment of hormonal receptor immunoexpression and Her2 status in inva- sive breast cancer after conservative and radical surgery. Rom J Morphol Embryol 59(3):763–772 23. Dai F, Chen Y, Song Y, Huang L, Zhai D, Dong Y, Lai L, Zhang T, Li D, Pang X, Liu M, Yi Z (2012) A natural small molecule harmine inhibits angiogenesis and suppresses tumour growth through activation of p53 in endothelial cells. PLoS ONE 7(12):e52162 24. Le SB, Hailer MK, Buhrow S, Wang Q, Flatten K, Pediadita- kis P, Bible KC, Lewis LD, Sausville EA, Pang YP, Ames MM, Lemasters JJ, Holmuhamedov EL, Kaufmann SH (2007) Inhi- bition of mitochondrial respiration as a source of adaphostin- induced reactive oxygen species and cytotoxicity. J Biol Chem 282(12):8860–8872 25. Devun F, Walter L, Belliere J, Cottet-Rousselle C, Leverve X, Fontaine E (2010) Ubiquinone analogs: a mitochondrial permeability transition pore-dependent pathway to selective cell death. PLoS ONE 5(7):e11792 26. Toler SM, Noe D, Sharma A (2006) Selective enhancement of cellular oxidative stress by chloroquine: implications for the treat- ment of glioblastoma multiforme. Neurosurg Focus 21(6):E10 27. Gupta SC, Hevia D, Patchva S, Park B, Koh W, Aggarwal BB (2012) Upsides and downsides of reactive oxygen species for can- cer: the roles of reactive oxygen species in tumorigenesis, preven- tion, and therapy. Antioxid Redox Signal 16(11):1295–1322 28. Yu-Wai-Man P, Soiferman D, Moore DG, Burté F, Saada A (2017) Evaluating the therapeutic potential of idebenone and related qui- none analogues in Leber hereditary optic neuropathy. Mitochon- drion 36:36–42 29. Lei Y, Huang K, Gao C, Lau QC, Pan H, Xie K, Li J, Liu R, Zhang T, Xie N, Nai HS, Wu H, Dong Q, Zhao X, Nice EC, Huang C, Wei Y (2011) Proteomics identification of ITGB3 as a key regula- tor in reactive oxygen species-induced migration and invasion of colorectal cancer cells. Mol Cell Proteom 10(10):M110.005397 30. Ushio-Fukai M, Nakamura Y (2008) Reactive oxygen species and angiogenesis: NADPH oxidase as target for cancer therapy. Can- cer Lett 266(1):37–52 31. Diebold LP, Gil HJ, Gao P, Martinez CA, Weinberg SE, Chandel NS (2019) Mitochondrial complex III is necessary for endothelial cell proliferation during angiogenesis. Usp22i-S02 Nat Metab 1(1):158–171.