Hydroxytyrosol targets extracellular matrix remodeling by endothelial cells and inhibits both ex vivo and in vivo angiogenesis

Elsevier Editorial System(tm) for Food Chemistry Manuscript Draft Manuscript Number: FOODCHEM-D-15-03601R1 Title: Hydroxytyrosol targets extracellular matrix remodeling by endothelial cells and inhibits both ex vivo and in vivo angiogenesis Article Type: Research Article (max 7,500 words) Keywords: angiogenesis; aortic ring assay; BAEC; CAM assay; hydroxytyrosol; MMP; TIMP; uPA Corresponding Author: Dr. Miguel Angel Medina, Corresponding Author's Institution: University of Málaga First Author: Javier A García-Vilas Order of Authors: Javier A García-Vilas; Ana R Quesada; Miguel Ángel Medina Abstract: The health benefits of olive oil are attributed to their bioactive compounds, such as hydroxytyrosol. Previously, we demonstrated that hydroxytyrosol inhibits angiogenesis in vitro. The present study aimed to: i) get further insight into the effects of hydroxytyrosol on extracellular matrix remodeling; and ii) test whether hydroxytyrosol is able to inhibit angiogenesis ex vivo and in vivo. Hydroxytyrosol induced a shift toward inhibition of proteolysis in endothelial cells, with decreased expression of extracellular matrix remodeling-enzyme coding genes and increased levels of some of their inhibitors. Furthermore, this work demonstrated that hydroxytyrosol, at concentrations within the range of its content in virgin olive oil that can be absorbed from moderate and sustained virgin olive oil consumption, is a strong inhibitor of angiogenesis ex vivo and in vivo. These results suggest the need for translational studies to evaluate the potential use of hydroxytyrosol for angio-prevention and angiogenesis inhibition in clinical setting. This is the preprint version of our manuscript, corresponding to the article that has been published in final form at FOOD CHEMISTRY with DOI: 10.1016/j.foodchem.2016.10.111 1 Hydroxytyrosol targets extracellular matrix remodeling by endothelial cells and inhibits 1 both ex vivo and in vivo angiogenesis 2 Javier A. García-Vilasa,b, Ana R. Quesadaa,b and Miguel Ángel Medinaa,b * 3 aUniversidad de Málaga, Andalucía Tech, Departamento de Biología Molecular y Bioquímica, 4 Facultad de Ciencias, and IBIMA (Biomedical Research Institute of Málaga) 5 bCIBER de Enfermedades Raras (CIBERER), E-29071 Málaga, Spain 6 7 8 *Corresponding author: Dr. Miguel Ángel Medina, Departamento de Biología Molecular y 9 Bioquímica, Facultad de Ciencias, Universidad de Málaga, E-29071, Málaga. Phone: +34-10 952137132. Fax. +34-952131674. E-Mail: [email protected] 11 12 13 Running title: Antiangiogenic hydroxytyrosol 14 15 16 2 Abstract 17 The health benefits of olive oil are attributed to their bioactive compounds, such as 18 hydroxytyrosol. Previously, we demonstrated that hydroxytyrosol inhibits angiogenesis in vitro. 19 The present study aimed to: i) get further insight into the effects of hydroxytyrosol on 20 extracellular matrix remodeling; and ii) test whether hydroxytyrosol is able to inhibit 21 angiogenesis ex vivo and in vivo. Hydroxytyrosol induced a shift toward inhibition of 22 proteolysis in endothelial cells, with decreased expression of extracellular matrix remodeling-23 enzyme coding genes and increased levels of some of their inhibitors. Furthermore, this work 24 demonstrated that hydroxytyrosol, at concentrations within the range of its content in virgin 25 olive oil that can be absorbed from moderate and sustained virgin olive oil consumption, is a 26 strong inhibitor of angiogenesis ex vivo and in vivo. These results suggest the need for 27 translational studies to evaluate the potential use of hydroxytyrosol for angio-prevention and 28 angiogenesis inhibition in clinical setting. 29 30 31 Keywords: angiogenesis; aortic ring assay; bovine aorta endothelial cells (BAEC); 32 chorioallantoic membrane (CAM) assay; hydroxytyrosol; matrix metalloproteinase (MMP); 33 tissue inhibitor of metalloproteinase (TIMP); urokinase-type plasminogen activator (uPA) 34 35 3 1. Introduction 36 The phenolic compounds in extra virgin olive oil are bioactive compounds with well-37 documented beneficial effects (Fortes, García-Vilas, Quesada & Medina, 2012; Owen et al., 38 2000) these compounds are not essential in the sense that nutrients are. In fact, the European 39 Union has authorized a health claim for polyphenols, based on consumption of olive oil 40 polyphenols, for the protection of blood lipids from oxidative stress ("Commission Regulation 41 (EU) 432/2012 of 16 May 2012 establishing a list of permitted health claims made on foods, 42 other than those referring to the reduction of disease risk and to children’s development and 43 health. O.J. 2012, L136/1.," 2012). Hydroxytyrosol or 3,4-dihydroxyphenyl ethanol is claimed 44 to be the most important health-related phenolic compound of virgin olive oil (Lopez de Las 45 Hazas, Rubio, Kotronoulas, de la Torre, Sola & Motilva, 2015). Along with its cardioprotective 46 effects (Mnafgui et al., 2015; Samuel, Thirunavukkarasu, Penumathsa, Paul & Maulik, 2008), 47 beneficial effects of hydroxytyrosol for human health include its antifungal (Zoric et al., 2013), 48 antidiabetic (Tutino, Orlando, Russo & Notarnicola, 2015; Zheng, et al., 2015), neuroprotective 49 (De La Cruz, et al., 2015; Gallardo, Madrona, Palma-Valdes, Espartero & Santiago, 2015), anti-50 inflammatory (Fuccelli, Fabiani, Sepporta & Rosignoli, 2015; Persia, Mariani, Fogal & Penissi, 51 2014; Silva et al., 2015) and antitumoral (Granados-Principal et al., 2014; Sirianni et al., 2010; 52 Sun, Luo, & Liu, 2014; Zhao et al., 2014) activities. In 2012, our group added the first 53 description of hydroxytyrosol as an anti-angiogenic compound able to inhibit several key steps 54 in the angiogenic process. In fact, we demonstrated that hydroxytyrosol (but not tyrosol) 55 induced endothelial cell apoptosis and changes in cell cycle distribution as well as inhibiting 56 endothelial cell proliferation, migration and differentiation into "capillary-like" tubes (Fortes et 57 al., 2012). Furthermore, that work also identified matrix metalloproteinase 2 (MMP-2) as one of 58 the molecular targets of the anti-angiogenic action caused by hydroxytyrosol. Since the 59 publication of that article, additional data have been published regarding the roles of 60 hydroxytyrosol as an anti-angiogenic compound. Several molecular targets have been identified 61 contributing to the anti-angiogenic effects of hydroxytyrosol, including inhibition of MMP-9, 62 4 cyclooxygenase 2 and vascular endothelial growth factor receptor-2 (VEGFR-2) 63 phosphorylation (Lamy, Ouanouki, Beliveau & Desrosiers, 2014; Scoditti et al., 2012; Scoditti 64 et al., 2014). Furthermore, hydroxytyrosol has been recently shown to have protective effects 65 against rheumatoid arthritis, an angiogenesis-dependent disease (Silva et al., 2015). 66 The identification of MMPs as molecular targets for hydroxytyrosol suggests that this 67 compound can alter extracellular matrix remodeling (Fortes, García-Vilas, Quesada & Medina, 68 2012; Scoditti et al., 2014), a key step in the angiogenic process. Therefore, the first aim of the 69 present work was to study the effects of hydroxytyrosol on key extracellular matrix remodeling 70 enzymes expressed in endothelial cells. On the other hand, in spite of the recent efforts to get 71 deep insights on the anti-angiogenic potential of hydroxytyrosol there is still no data available 72 demonstrating its potential inhibitory effects in ex vivo or in vivo models of angiogenesis. The 73 second aim of this work was to test the anti-angiogenic potential of hydroxytyrosol in the ex 74 vivo aortic ring and the in vivo CAM angiogenesis models. 75 76 2. Materials and methods 77 2.1. Chemicals 78 Supplements and other chemicals not listed in this section were obtained from Sigma 79 Chemicals Co. (St. Louis MO, USA). Cell culture media, penicillin, streptomycin and 80 amphotericin were purchased from Biowhittaker (Walkersivlle, MD, USA). Fetal bovine serum 81 (FBS) and human serum (HS) were products of Harlan-Seralb (Belton, United Kingdom). 82 Plastics for cell culture were supplied by NUNC (Rockilde, Denmark) and VWR (West Chester, 83 Pennsylvania, USA). Hydroxytyrosol was supplied by Extrasynthèse (Lyon, France). 84 85 2.2. Cell culture 86 Most of the procedures described in Materials and methods have been previously used by 87 our research group for other studies (see, for instance Garcia-Vilas et al. 2013, Martínez-Poveda 88 5 et al. 2013). Bovine aorta endothelial cells (BAEC) were isolated from bovine aortic arches, as 89 previously described (Gospodarowicz & Moran, 1975), and maintained in Dulbecco´s modified 90 Eagle´s medium (DMEM) containing glucose (1 g/L), glutamine (2 mM), penicillin (50 91 IU/mL), streptomycin (50 mg/ L), amphotericin (1.25 mg/L), 10% fetal bovine serum. 92 93 2.3. RNA isolation and purification and cDNA synthesis 94 Cells at 80% of confluence in 6-well plates were treated with or without 1mM of 95 hydroxytyrosol for 24 h. After incubation, cells were harvested and washed (PBS). Total RNA 96 was isolated with the GeneElute Mammalian Total RNA Miniprep Kit (Sigma-Aldrich) 97 according to the purchaser's instructions. 98 cDNA synthesis was carried out with the iScript cDNA synthesis kit (BioRad). 99 100 2.4. qPCR 101 For quantitative RT-PCR (qPCR), total RNA isolation and complementary DNA synthesis 102 were performed as described above and PCR reactions were done using KAPA SYBR Fast 103 Master Mix (2x) Universal (KAPA Biosystems) in an Eco Real-Time PCR System. qPCR was 104 performed in triplicate for each sample in keeping with the manufacturer´s instructions. All 105 qPCR data were normalized to GAPDH expression (Martínez-Poveda et al. 2013). Primers, 106 amplicon size, and qPCR conditions for each gene are shown in Table 1. 107 108 2.5. Animals and ethical statement 109 All experimental procedures with animals were conducted in accordance with the Spanish 110 Legislation (Real Decreto 53/2013, BOE, 34/-11421, 2013) in compliance with the European 111 Community Directive 2010/63/EU regulating the use and care of laboratory animals. The 112 6 protocols were approved by the Ethics Committee for Animal Experiments of the University of 113 Málaga. 114 115 2.6. Ex vivo rat aortic ring assay 116 Thoracic aortas were removed from 12 week old rats and immediately transferred to a 117 culture dish containing DMEM. The perioaortic fibroadipodise tissue was carefully removed 118 with fine microsdissecting forceps and iridectomy scissors paying special attention not to 119 damage the aortic wall. Afterwards, 1-mm aortic rings were sectioned and embedded in a rat tail 120 interstitial collagen gel (1.5 mg/mL) prepared by mixing 7.5 volumes of 2 mg/mL collagen, 1 121 volume of 10x HBBS, 1.5 volumes of 186 mM NaHCO3 and 0.1 volumes of 1 M NaOH to 122 adjust the pH to 7.4. Collagen gels containing the aortic rings were polymerized in cylindrical 123 agarose wells and kept in triplicate at 37˚C in 60 mm diameter Petri dishes (bacteriological 124 polysterene, Falcon, Becton Dickinson, Lincoln Park, New Yersey). Each Petri dish contained 6 125 mL of MCDB131 medium supplemented with 1% L-glutamine, 25 mM NaHCO3, 100 U/mL 126 penicillin, 100 µg/mL streptomycin, in presence of hydroxytyrosol or etanol (vehicle). Cultures 127 were kept at 37˚C, 5% CO2 in a humidified environment, and photographs were taken after 6, 9 128 and 14 days. The antiangiogenic response was quantified by microvessel counting according to 129 published criteria (Nicosia & Ottinetti, 1990). 130 131 2.7. In vivo chorioallantoic membrane (CAM) assay 132 Fertilized chick eggs were incubated horizontally at 38°C in a humidified incubator, 133 windowed by day 3 of incubation and processed by day 8. The indicated amount of 134 hydroxytyrosol was added to a 1% solution of methylcellulose in water, and 10 μL drops of this 135 solution were allowed to dry on a Teflon-coated surface in a laminar flow hood. Then, the 136 methylcellulose disks were implanted on the CAMs, and the eggs were sealed with adhesive 137 tape and returned to the incubator for 48 h (Garcia-Vilas et al. 2013). Negative controls were 138 7 always made with ethanol (vehicle) mixed with methylcellulose. After the reincubation, CAMs 139 were examined under a stereomicroscope. The assay was scored as positive when two 140 independent observers reported a significant reduction of vessels in the treated area. 141 142 2.8. Statistical analysis 143 For all of the assays, at least three independent experiments were carried out. Results are 144 expressed as means ± SD. Statistical significance was determined by using Student´s paired 145 simple test. Values of p<0.05 were considered to be significant. 146 147 3. Results 148 149 3.1. Hydroxytyrosol induces changes in the expression levels of genes involved in ECM 150 remodeling in endothelial cells 151 To fulfil the first goal of our work, we analyzed by qPCR the effects of 1 mM 152 hydroxytyrosol treatment of BAEC for 24 h on the expression levels of messenger 153 corresponding to a number of genes involved in ECM remodeling. The primers, annealing 154 temperature and amplicon sizes are summarized in Table 1. Table 2 summarizes the qPCR 155 quantitative data for the 9 tested potential targets. Both MMP-1 and MMP-2 mRNA expression 156 levels were drastically diminished by hydroxytyrosol, whereas three out of four tissue inhibitors 157 of metalloproteinases (namely, TIMP-1, -2 and -4) mRNA expression levels were increased 158 several fold. On the other hand, the mRNA levels for urokinase-type plasminogen activator 159 (uPA), another ECM remodeling enzyme, were also drastically diminished upon hydroxytyrosol 160 treatment, whereas the levels of the messenger for its specific receptor (uPAR) increased more 161 than 20-fold. Finally, we could not detect any signal for plasminogen activator inhibitor 1 (PAI-162 1) mRNA. 163 164 8 3.2. Hydroxytyrosol has a very potent inhibitory effect on the ex vivo rat aortic ring 165 angiogenesis assay 166 The second goal of this work was to test the anti-angiogenic potential of hydroxytyrosol in 167 the ex vivo aortic ring and the in vivo CAM angiogenesis models. Figure 1 clearly shows that 168 hydroxytyrosol was able to induce very strong inhibitory effects in the ex vivo aortic ring assay, 169 even at concentrations much lower than those used previously by us to show its in vitro anti-170 angiogenic effects (Fortes, García-Vilas, Quesada & Medina, 2012). At 0.25 mM 171 hydroxytyrosol the inhibition of microvessel outgrowth from the aortic rings was complete after 172 6, 9 and 14 days of incubation. At 0.125 mM hydroxytyrosol we observed outgrowing of 173 proliferative endothelial cells from the aortic ring but these cells were not forming microvessel 174 tubes. Furthermore, there were partial inhibitory effects for 62.5 µM hydroxytyrosol. 175 176 3.3. Hydroxytyrosol inhibits in vivo angiogenesis 177 Figure 2 shows that in untreated, control CAMs, blood vessels form a spatially-oriented and 178 dense network of vascular structures with progressively smaller diameters as they branch. 179 Hydroxytyrosol-treated CAMs showed inhibited angiogenesis, as revealed by an inhibition of 180 new vessel formation within the area covered by the methylcellulose discs, a centrifugal growth 181 of peripheral vessels, avoiding the treated area, and an overall decrease in vascular density. 182 Table 3 summarizes the results obtained with the CAM assay. Positive, inhibitory effects were 183 observed in 90% of eggs treated with 800 nmol of hydroxytyrosol. The inhibition was still 184 observed in more than half of the CAMs treated with 400 nmol of hydroxytyrosol. For 185 treatments with 200 and 100 nmol of hydroxytyrosol, 42% and 13% of the eggs scored positive. 186 No inhibitory effect was observed in eggs treated with 50 nmol of hydroxytyrosol. 187 188 4. Discussion 189 15 Table 1. Primers used for qPCR with indication of their respective annealing temperatures 361 and amplicon sizes. 362 363 Gene Primers Annealing temperature (°C) Amplicon size (bp) MMP-1 Forward: gacttagtccagaaatacctg 60 121 Reverse: caaagaattcctgcatttgc MMP-2 Forward: gacatacatctttgctggagac 60 200 Reverse: acgctcttcagactttggttct TIMP-1 Forward: gggacaccagaagtcaacca 63 81 Reverse: ggcttggaaccctttatacatc TIMP-2 Forward: aagcggtcagtgagaaggaa 65 112 Reverse: tctcaggccctttgaacatc TIMP-3 Forward: gcagcggaccacaacagcta 68 150 Reverse: ccggatcacgatgtcggagt TIMP-4 Forward: agggagagcctgaatcatca 65 68 Reverse: gcactgcatagcaagtggtg uPA Forward: cgccacacactgcttcatg 60 310 Reverse: ccccttgcgtgttggagtt uPAR Forward: gcccaatcctggagcttga 60 63 Reverse: tccccttgcagctgtaacact PAI-1 Forward: gcacaaccccacaggaaca 65 81 Reverse: gtcccgatgaaggcgtcttt 16 Table 2. Relative expression values of mRNAs for some extracellular matrix remodelling 364 enzymes and their inhibitors in BAEC treated with hydroxytyrosol. 365 qPCR was carried out as described in Materials and Methods. All qPCR data were 366 normalized with GAPDH expression levels. Data are given as means±S.D. and they are 367 percentages of expression taking the corresponding expression values in control, untreated cells 368 as 100%. 369 370 371 372 MMP-1 MMP-2 TIMP-1 TIMP-2 TIMP-3 TIMP-4 uPA uPAR PAI1 BAEC 16.9 ± 0.5 0.28 ± 0.2 3857.3 ±537.1 418.0 ±123.9 14.76 ± 1.47 377.6 ± 209.6 20.95 ± 12.1 2195.8 ± 433.0 - 17 Table 3. Inhibition of in vivo angiogenesis by hydroxytyrosol (HT) as determined by the 373 CAM assay. 374 HT (nmol/CAM) Positive/Total Inhibition (%) 0 0/8 0 50 0/6 0 100 1/8 13 200 5/12 42 400 5/8 63 600 5/7 71 800 9/10 90 375 376 18 Figure legends 377 378 Fig. 1. Hydroxytyrosol inhibits microvessel outgrowth in the ex vivo rat aortic ring assay. 379 (A) Representative samples of the aortic ring without or with treatment. (B) Quantification of 380 the area occupied by new microvessels in controls, controls treated with VEGF and rings treated 381 with 31.2 µM or 62.5 µM hydroxytyrosol and VEGF. The results are the mean ± SD of three 382 different assays. 383 384 Fig. 2. Hydroxytyrol inhibits angiogenesis in vivo in the CAM assay. Arrows point rebound 385 of vessels outward from the treated area. Asterisks indicate disrupted vessels. 386 387 388 389 390 391 392 Figure 1 Click here to download high resolution image Figure 2 Click here to download high resolution image