Microvascular Research 88 (2013) 12–18

Contents lists available at SciVerse ScienceDirect

Microvascular Research

j ourna l homepage: www.e lsev ie r .com/ locate /ymvre

Regular Article

Antiangiogenic activity and direct antitumor effect from asulfated polysaccharide isolated from seaweed

Celina Maria P. Guerra Dore a, Monique Gabriela C. Faustino Alves a, Nednaldo D. Santos a,Ana Katarina M. Cruz a, Rafael Barros G. Câmara b, Allisson Jonathan G. Castro a, Luciana Guimarães Alves a,Helena B. Nader b, Edda Lisboa Leite a,⁎a Department of Biochemistry, Federal University of Rio Grande do Norte (UFRN), Sen. Salgado Filho 3000, Lagoa Nova, Natal, RN, Brazilb Department of Biochemistry Federal University of São Paulo (UNIFESP), Rua dos Otonis 567, Vila Clementino, São Paulo, SP, Brazil

⁎ Corresponding author. Fax: +55 84 3215 34 15.E-mail address: eddaleite@cb.ufrn.br (Edda L. Leite)

0026-2862/$ – see front matter © 2013 Elsevier Inc. Allhttp://dx.doi.org/10.1016/j.mvr.2013.03.001

a b s t r a c t

a r t i c l e i n f o

Article history:Accepted 7 March 2013Available online 16 March 2013

Angiogenesis is a dynamic proliferation and differentiation process. It requires endothelial proliferation, migrationand tube formation. In this context, endothelial cells are a preferred target for several studies and therapies. Anionicpolysaccharides (SV1 and PSV1) from brown seaweed Sargassum vulgarewere fractionated (SV1), purified (PSV1)and displayed with high total sugars, sulfate content and very low level of protein. The antiangiogenic efficacy ofpolysaccharides was examined in vivo in the chick chorioallantoic membrane (CAM) model by using fertilizedeggs. Decreases in the density of the capillaries were assessed and scored. The results showed that SV1 and PSV1have an inhibitory effect on angiogenesis. These results were also confirmed by the inhibition of tubulogenesis inrabbit aorta endothelial cell (RAEC) in matrigel. These compounds were assessed in an apoptosis assay (AnnexinV-FITC/PI) and cell viability by MTT assay of RAEC. These polysaccharides did not affect the viability and did nothave apoptotic or necrotic action. RAEC cellwhen incubatedwith SV1andPSV1 showed inhibitionof VEGF secretion,observed when compounds were incubated at 25, 50 and 100 μg/μL. The VEGF secretion with the RAEC cell line for24 h was more effective for PSV1 at 50 μg/μL (71.4%) than for SV1 at 100 μg/μL (75.9%). SV1 and PSV1 had anantiproliferative action (47%) against tumor cell line HeLa. Our results indicate that these sulfated polysaccharideshave antiangiogenic and antitumor actions.

© 2013 Elsevier Inc. All rights reserved.

Introduction

Angiogenesis, the formation of newcapillary vessels frompre-existingvasculature, is regulated by complex interactions between stimulatingand inhibitory factors, including those involving growth, cytokines,proteolytic enzymes, integrins and extracellular matrix components(Burgermeister et al., 2002). Some diseases are determined by the persis-tent angiogenic response caused by an increase in angiogenic mediatorsor inhibitor deficiency, such as neoplasias, metastases, psoriasis andrheumatoid arthritis.

The activating factors of angiogenesis interact with glycosaminogly-cans and proteoglycans present in the intracellular matrix, basal laminaand cell surface receptors, regulating growth, proliferation, migration,differentiation and endothelial cell survival in a variety of cell types(Chipperfield et al., 2002; Solimene et al., 1999). One of themost specificand important factors involved in angiogenesis is the vascular endothe-lial growth factor (VEGF), which links specifically to differentmembranereceptors of endothelial cells (Dvorak et al., 1995; Ferrara et al., 2003).In addition to stimulating angiogenesis, VEGF is also important for

.

rights reserved.

maintaining the integrity and permeability of blood vessels (Keck etal., 1989). Thus it was suggested that algal anionic polysaccharidescould link growth factors to specific regions of competent receptors,modulating or antagonizing responses when forming complexes withgreater affinity than those endogenously modulated by heparan sulfate(Dias et al., 2008).

Polysaccharides extracted from brown seaweed exhibit antithrom-botic (Mauray et al., 1995), anticoagulant (Cumashi et al., 2007), antioxi-dant, antiproliferative (Costa et al., 2011) and antivasculogenic activities(Dias et al., 2008), among others. In this study, we investigated the invitro antiangiogenic effect on apoptosis and antitumor activity, addition-ally to the inhibition of the VEGF secretion by sulfated polysaccharidesextracted from the brown seaweed Sargassum vulgare. Such an approachallowed us to investigate the antivasculogenic activity using the chickenembryo Chorioallantoic Membrane Assay (CAM).

Materials and methods

Cell culture

Rabbit aortic endothelial cells (RAEC)weremaintained at 37 °C understress of 2.5% CO2, in HAM-F12 medium (Sigma-Aldrich, St. Louis, MO,

Table 1Chemical characterization of SV1 and PSV1.

Total sugar(%)

Proteins(%)

Phenolic compounds(%)

Sulfate(%)

SV1 62.9 3.4 1.0 30.0PSV1 50.2 n.d. 0.6 20.3

Table 2Score values for the antiangiogenic effect on the chorioallantoic membrane of fertilizedchicken eggs.

Score value Effect observed

0 No effect –

0.5 Very weak effect No capillary-free area.Area with reduce capillary density around thepellet not larger than the area of the pellet.

1.0 Weak–mediumeffect Small capillary-free area or area withsignificantly reduced capillary density.Effect not larger than double the size of the pellet.

2.0 Strong effect Capillary-free area around the pellet at leastdouble the size of the pellet.

13C.M.P. Guerra Dore et al. / Microvascular Research 88 (2013) 12–18

USA), supplementedwith 10% fetal bovine serum. HeLa and B16 cell lineswere maintained at 37 °C under stress of 5% CO2, in DMEM medium(Sigma-Aldrich, St. Louis, MO, USA), supplemented with 10% fetal bovineserum.

Extraction and purification of the polysaccharides

The brown seaweed Sargassum vulgare C. Agardh (Phaeophyceae)was collected at Buzios beach, on the south coast of the state of RioGrande do Norte, Brazil. Seaweeds were stored in our laboratoryand oven dried at 50 °C under ventilation, ground in a blender andincubated with acetone to eliminate lipids and pigments. About 50 gof powdered algae was suspended with five volumes of 0.25 M NaCland the pHwas adjusted to 8.0 with NaOH. Tenmilligrams ofmaxataze,an alkaline protease from bacillus spores (BioBra, Montes Claros, MG,Brazil), was then added to the mixture for proteolytic digestion. Afterincubation of 24 h at 60 °C, under agitation and periodical pH adjust-ments, the mixture was filtered with a cheesecloth and precipitatedwith increasing amounts of ice cold acetone (0.3, 0.5, 1.0 and 2.0 v)under gentle agitation at 4 °C. The formed precipitates were collectedby centrifugation at 8000 g for 20 min and vacuum dried. The obtainedvolume fraction with 1.0 v was denominated SV1, and part of theincome column was eluted in Sepharose CL4-B chromatography(Sigma-Aldrich, St. Louis, MO, USA) and purified by gel filtration. Poly-saccharides were subjected to gel-permeation chromatography onSepharose CL-4B (140 × 1.8 cm) using 0.2 M acetic acid as eluent(Dore et al., 2013). The elution was monitored for total sugar (Duboiset al., 1956). The eluted polysaccharides were dialyzed against water,freeze-dried and used in the assays. The obtained purified polysaccha-rides were named PSV1.

Chemical composition and infrared spectroscopy

The total sugars were determined using a phenol–H2SO4 reaction,with D-fucose as standard (Sigma, St Louis, MO, USA) as previouslydescribed (Dubois et al., 1956). The sulfate content was measuredafter acid hydrolysis (HCl 6 N, 6 h, 100 °C) using the turbidimetricmethod (Dodgson and Price, 1962). The protein content was quanti-fied with Coomassie Brilliant Blue reagent and bovine serum albumin(Sigma, St. Louis, MO, USA), as standard (Bradford, 1976). The IRspectrum of the polysaccharides fractions SV1 and PSV1 S. vulgarewas carried out using a Fourier transform infrared spectrophotometer(FTIR, Bruker, Germany) equippedwith OPUS 3.1 software. The fractionwas ground with KBr powder (Merck, Darmstadt, Germany) and thenpressed into pellets for FTIR measurement in the frequency range of4000–500 cm−1.

Inhibition of the angiogenesis

Chorioallantoic membrane assay (CAM)The angiogenesis inhibition activity was determined using the

Chorioallantoic Membrane Assay (CAM) (Deocaris et al., 2005) withmodifications. Fertilized chicken eggs (n = 6/treatment) by polysaccha-rides solution in agarose (100–1000 μg/egg)were kept in an incubator at37 °C and 55% relative humidity. Malformed or dead embryos were ex-cluded. On the fourth day, 5 mL of albumin was aspirated and a windowof 1.5 cm2 was opened on the fifth day. The window was closed andincubation was continued until the seventh day by adding different

concentrations of SV1 and PSV1 (100 and 1000 μg/egg) into an agarosesolution under the CAM. Spironolactone and heparin (Sigma Chemical,St. Louis, MO, USA) at a concentration of 10 μg/egg were used with pos-itive and negative controls, respectively. Agarose was employed as blankand the neovascular areas of CAM were photographed using a Nikondigital camera and analyzed by the software IMAGE PRO PLUS, using ascoring system to express results (Burgermeister et al., 2002). Six eggsper group were used in assessment. Our results were double checkedby two independent observers to minimize technical problems such asstandardization and interobserver variability. Representative areas ofthe vessel density were photographed using a Nikon digital camera.The images were analyzed by the software ImageJ. Each disk area wasanalyzed twice for assessment by score. The same area of the disk wasused for all evaluated eggs.

Matrigel tube formation assayIn the in vivo environment, the basal surface of endothelial cells is

in contact with a thin layer of a highly specialized extracellular matrixknown as the basement membrane, which forms a layer around thesecells and maintains the tubular structure characteristics of bloodvessels (Arnaoutova et al., 2009). Matrigel (BD Biosciences, Palo Alto,CA, USA) 250 μL was placed in each well of an ice-cold 24-well plate.The plate was allowed to sit at room temperature for 15 min and thenit was incubated at 37 °C for 30 min to permit the matrigel to solidify.The different groups were tested with four replicates per group. Weused 24-well plates and inserted 5000 cells/well. All wells were exam-ined under a microscope, performing an analysis of all areas of thecapillary. The length of formed cords and number of junctions werecompared among various groups. The biological activity of this matrixhad been demonstrated for over 20 years, when it was observed thatendothelial cells added to the reconstitution of the artificial basementmembrane adhered and rapidly organized into capillary-like structures(Kubota et al., 1988). For this test, 1 × 105 rabbit aortic endothelial cellsper well were added to the wells of the culture plate, which had beenpreviously added and incubated with 200 μL aliquots of a reconstitutedbasement membrane solution, at a concentration between 10 and15 mg/mL protein, in an atmospheric CO2 incubator (2.5% at 37 °C).The cells were incubated in eight groups. A) control group; B)group incubated with heparin; C) three groups incubated withPSV1 (25, 50 and 100 μg); and D) three groups incubated with SV1(25, 50 and 100 μg). The cells were incubated for 24 h, observedand photographed using a digital camera coupled to a phase interfer-ence microscope.

Antiproliferative activity (MTT assay)The MTT (Sigma, St. Louis, MO, USA) assay was carried out using

rabbit aortic endothelial cells, melanoma B16 and HeLa cell line, as de-scribed by Mosmann (1983). The culture was exposed to 25, 50 and100 μg of SV1 and PSV1 (triplicate) and incubated for 24 h at 37 °C.After incubation, 100 μL of F-12 medium (Sigma, St. Louis, MO, USA),containing MTT (final concentration of 5 mg/mL) was added to eachwell and themicroplatewas incubated for 4 h at 37 °C. The supernatant

Fig. 1. CAM assay to determine the antiangiogenic potential of SV1 and PSV1. A: Blank (agarose); B — Heparin (10 μg/mL); C — SV1 100; D — SV1 1000; E: PSV1 100; F: PSV1 1000.

14 C.M.P. Guerra Dore et al. / Microvascular Research 88 (2013) 12–18

was then removed and 100 μL of ethanol was added to each well to sol-ubilize the formazan crystals. After mixture homogenization, absor-bance was measured at 570 nm by a microplate reader. To control thereaction (100% proliferation), PSV1 and SV1 were replaced by an F-12medium.

Apoptosis assay (Annexin V-FITC/PI) and cell cycle distribution of rabbitaortic endothelial cells (RAEC)

Apoptosis assay (Annexin V/PI)RAEC cell death was evaluated using annexin V-fluorescein isothio-

cyanate (detection of phosphatidylserin in the cellular membrane) andpropidium iodide (PI), which can enter cells that have compromisedcell membrane integrity, constituting late apoptotic cells or necroticcells as identified by the BD Pharmingen Annexin V-FITC detection kit(BD Biosciences, NJ, USA). RAEC cells were harvested in a 6-well plate(100,000 cells per well) in an F-12 medium at 37 °C under 2.5% CO2

atmosphere for 24 h, until reaching confluence. After incubation, cellswere kept in a serum-free medium for 18 h and then treated with50 μg of SV1 or PSV1 for 24 h. Cells were then incubated with 3 μL ofannexin V and 5 μL of PI. These were analyzed in a FACSCalibur flowcytometer (Becton Dickinson & Company, NJ, USA). The assay wasperformed in duplicate.

Cycle cell distribution assayTo determine the effect of fucans on the cell cycle, RAEC cells

(2–3 × 105) were submitted to the same methodology as above. Aftertreatment with 50 μg of SV1 or PSV1, cells were fixed with 2% formalde-hyde and incubated for 30 min at 4 °C. After the incubation, they werewashed with PBS buffer (150 mM, pH 7.4) and 0.01% saponin, and a4 mg/mL RNase (Sigma Aldrich, St Louis, MO, USA) solution was addedto 0.05 M sodium acetate buffer, containing 0.02 M MgSO4. Cells werethen incubated for 1 h at 37 °C. Next, 5 μL of PI (Sigma Aldrich, StLouis, MO, USA) solution at 5 mg/mL was added at room temperature,after which cells were analyzed in a FACSCalibur flow cytometer (BectonDickinson & Company, NJ, USA). The assay was performed in duplicate.The DNA content was determined by flow cytometry. The cell cycle dis-tributionwas evaluated by comparing it with the one of the control cells.

Assay to determine vascular endothelial growth factor (VEGF) secretedby RAEC cells

The Human VEGF ELISA (Enzyme-Linked Immunosorbent Assay) kit(RayBio Inc., Norcross, GA, USA) is an in vitro enzyme-linked immuno-sorbent assay for the quantitative measurement of VEGF cell culturesupernatants. This assay employs an antibody that is specific forhuman. The VEGF was coated on a 96-well plate. Standards and samplesare pipetted into the wells and the VEGF present in a sample is boundto the wells by the immobilized antibody. The wells are washed andbiotinylated antihumanVEGF antibody is added. Afterwashing away un-bound biotinylated antibody, HRP-conjugated streptavidin is pipetted tothe wells. The wells are again washed and a TMB substrate solution isadded to the wells. The color develops in proportion to the amount ofbound VEGF. The Stop Solution changes the color from blue to yellowand the intensity of the color is measured at 450 nm.

Thus, the culturewas exposed to 25, 50 and 100 μg in 50 μL/mediumSV1 and PSV1 (triplicate), and incubated for 24 h at 37 °C. After incuba-tion, the culture medium was aspirated. VEGF in the RAEC cell mediumwas determined using the ELISA Kit, following the manufacturer's pro-tocols (Calbiochem USA, VEGF Elisa Kit).

Statistical analysis

We used the Graph Instat software. Values were expressed asmean ± SEM. The Analysis of variance (ANOVA) and the Tukey test, in-cluding the Kolmagorov test with evidence for a normality behavior,were used to assess biological activity data, with P b 0.05 establishedas statistically significant. Differences in the apoptosis assay andbetween treatment and control were compared using an unpairedStudent's t test.

Results

Chemical characterization and infrared spectroscopy

Four heteropolysaccharide fractions (0.3, 0.5, 1.0 and 1.5 v) wereobtained from the brown seaweed S. vulgare by acetone precipitationin several volumes (Dietrich et al., 1995; Dore et al., 2013). The fraction

Fig. 2. Effects of SV1 and PSV1 on capillary tube formation of rabbit aortic endothelial cells (RAEC). RAECs (1 × 105 per well) inmedium containing 10% serumwere seeded intoMatrigelpre-coated 24-well plates and treatedwith different concentrations of SV1 and PSV1 (A— Control; B—Heparin (10 μg/mL); C— SV1 25; D— SV1 50; E— SV1 100; F— PSV1 25; G— PSV150; H — PSV1 100). Representative phase contrast photomicrographs (100× magnification).

15C.M.P. Guerra Dore et al. / Microvascular Research 88 (2013) 12–18

denominated SV1 was eluted with 1.0 v of acetone and chosenaccording to its high yield (63.1%) compared with other factions. TheHPLC of this fraction revealed the existence of only one polysaccharide,which after acidic hydrolysis showed a monosaccharide composition,characterized by units of L-fucose, D-mannose, D-galactose, D-xylose,D-glucose and D-glucuronic acid. The SV1 was fractioned and eluted inSepharose CL-4B gel filtration chromatography (PSV1). The elution pro-file of the fraction, when compared to the known standard molecularweight profile, allowed us to estimate its molecular weight, which is

Fig. 3. Antiproliferative activity (MTT assay) from PSV1 and SV1 in rabbit aortic endothelial c100 μg/μL of PSV1 and SV1 (triplicate) and incubated for 24 h. Cells treated with F-12 mediStatistical analysis (a): P b 0.001; (b): P b 0.05; (c): P b 0.01.

around 160 kDa. Polysaccharides were eluted in 1 mL fractions andmonitored by total sugar levels.

Fucans are hygroscopic polysaccharides that absorb water and assuch, the sum of the components does not reach 100%. The chemicalcharacterization of SV1 and PSV1 is illustrated in Table 1. Gel filtrationchromatography favors oligosaccharide separation with molecularmasses and different sulfation patterns. Thus, sulfate and total sugarcontent decreases among samples due to the purification process(Dore et al., 2013). The 13C NMR spectrum signals between 105 and

ells (RAEC), melanoma B16 and HeLa cells lines. The culture was exposed to 25, 50, andum served as a negative control group. Values are expressed as mean ± SEM (n = 3).

Fig. 4. Flow cytometry analysis with double stain for annexin V and PI in rabbit aortic endothelial cells. Histogram: A and B: negative control (no treatment); C and D: cells treatedwith PSV1; E and F: Cells treated with SV1. Upper left quadrant: negatively stained for annexin V and positively for PI (non-viable cells–necrotic cells); Upper right quadrant:positively stained for annexin V and PI (non-viable cells–late apoptotic cells); Lower left quadrant: negatively stained for annexin V and PI (viable cells); Lower right quadrant:positively stained for annexin V and negatively for PI (non-viable cells–early apoptotic cells).

16 C.M.P. Guerra Dore et al. / Microvascular Research 88 (2013) 12–18

74 ppm correspond to sulfated carbons in the fucopyranose ring, whilethe group of signals at 71–66 ppm may arise from unsubstituted car-bons. Moreover, signals were recorded at 103.1 and 61.3 ppm forgalactoses and at 100.8 ppm for xylose. Data on these polysaccharideswere described by Dore et al. (2013).

Fig. 5. Flow cytometry analysis with annexin V and PI double staining in rabbit aorticendothelial cells. Bar graph: Viable cells: negatively stained for annexin V and PI;Non-viable cells: positively stained for annexin V and negatively for PI (early apoptoticcells); negatively stained for annexin V and positively for PI (necrotic cells); positivelystained for annexin V and PI (late apoptotic cells). Significance level related to control(untreated cells): (a) P > 0.05; (b) P b 0.001 Unpaired Student's t test. Values areexpressed as mean ± SEM (n = 2).

Angiogenesis inhibition assays

SV1 and PSV1 activities in the chorioallantoic membrane (CAM) assaySV1 and PSV1 fractions were tested in the CAM assay to assess its

angiogenic potential. A scoring system was established according tomethodology used by Burgermeister et al. (2002). A score of 0 representsnormal capillary growth and score 2 indicates a strong antiangiogeniceffect with a large capillary-free area (Table 2). SV1 and PSV1 were ad-ministered in agarose disks at 100 and 1000 μg/disk and low molecularweight heparin as standard. At the concentrations tested, SV1 and PSV1exhibited strong antiangiogenic activity, obtaining scores of 0.5 forheparin, 1.0 for SV1 100, SV1 1000 and PSV1 100 (Fig. 1), and an increasein antiangiogenic potential was observed in eggs treated with PSV11000, obtaining a score of 2.0.

Matrigel tube formation assayTubulogenesis involves several cellular processes, including differen-

tiation, polarization, shape change, proteolysis, growth, mitosis, death,motility, adhesion, signaling, ion fluxes, cytoskeletal organization andmembrane traffic (Hogan and Kolodziej, 2002; Zegers et al., 2003).Cells that make up tubular organs display apical–basal polarity, that is,an apical surface of the membrane directed toward a central lumenand a basal surface linked to an extracellular matrix layer (Hogan andKolodziej, 2002). Any event that interferes in this polarization coulddirectly affect tubulogenesis.We showed that SV1 and PSV1 significantlyinhibited capillary tube formation, see Fig. 2. Therewas no significant dif-ference between the concentrations tested, their effect being similar tothat of heparin. In plates treated with PSV1, a dose-dependent tendency

Fig. 6. Analysis of distribution in rabbit aortic endothelial cells (RAEC) in the cell cycle. Untreated endothelial cells (A) either, treated with PSV1 (B) or treated with SV1 (C) for 24 h.Cell cycle phases — G1: gap 1; G2: gap 2; S: synthesis phase. Values are expressed as mean ± SEM (n = 2).

Table 3Analysis of distribution in rabbit aortic endothelial cells (RAEC) in the cell cycle. Endothelialcells were either untreated, treated with PSV1, or treated with SV1 for 24 h. Cell cyclephases — G1: gap 1; G2: gap 2; S: synthesis phase. Values are expressed as mean ± SEM(n = 2).

Cycle cellphase

Untreatedcells (%)

Cells treatedwith PSV1 (%)

Cells treatedwith SV1 (%)

G1 83.79 ± 0.00 65.60 ± 2.77 64.83 ± 2.55G2 7.81 ± 0.00 6.33 ± 1.01 6.27 ± 1.47S 8.41 ± 0.00 28.26 ± 2.04 28.89 ± 1.08G2/G1 2.00 ± 0.00 2.00 ± 0.00 2.00 ± 0.00

17C.M.P. Guerra Dore et al. / Microvascular Research 88 (2013) 12–18

was observed with greater contact inhibition efficiency between cellswhen theywere incubatedwith higher concentrations of the compound.

Antiproliferative activity (MTT assay)Considering the effect against cancer cells, compounds that exhibit

cytotoxic activity show interesting effects in chemotherapy only whentheir toxicity is higher for cancer cells than for non-cancerous cells.Thus, the abilities of these polysaccharides (PSV1 and SV1) in tumorand non-tumor cells were tested. Cell viability was determined in differ-ent lines using the MTT test. RAEC is a normal endothelial cell line, orig-inated in the rabbit aorta; HeLa and B16 lines are tumor cells, the formeroriginated in the human cervix and the latter being a rat melanoma. Theeffect of the analyzed samples showed that both exhibit antiproliferativeactivity in tumor cells (P > 0.05), with no statistically significant differ-ences. However, PSV1 and SV1 were more effective as proliferationinhibitors in HeLa cells. It was observed that PSV1, at a concentration of50 μg/μL, inhibited around 47% of proliferation in these cells, demon-strating direct antitumor activity. When SV1 and PSV1 samples weretested with normal endothelial cells, cell proliferation was not inhibited(Fig. 3). This selectivity against specific cancer cell lines is one of therequirements for the development of new drugs in order to lower theextent of side-effects.

It became clear, therefore, that these polysaccharides have differenteffects on the various tested cell lines. Fucans show different effects onsome cell lines, suggested by their different structural composition inrelation to branching, types of sugars, and sulfate and sugar content.They inhibit or do not proliferate some tumor and non-tumor cellssuch as fibroblasts (Almeida-Lima et al., 2010).

Apoptosis assay (Annexin V-FITC/PI) and cycle cell distribution of rabbitaortic endothelial cells (RAEC)

Flow cytometry analysis demonstrated that RAEC cells treatedwith PSV1 have a similar profile to non-treated cells (Figs. 4–6).There was no statistically significant difference between non-treatedcells and those treated with PSV1 (Fig. 4). However, this is not observedin cells containing SV1. According to staining, viable cells were annexinV (−)/PI (−); early apoptotic cells were annexin V (+)/PI (−); lateapoptotic cells were annexin V (+)/PI (+), and necrotic cells wereannexin V (−)/PI (+). Thus, PSV1 exhibits no apoptotic activity anddid not cause necrosis. However, SV1 behaves differently, causingearly and late apoptosis in treated cells (Fig. 5).

Flow cytometry analyses of RAEC cells treated with PSV1 and SV1for 24 h showed a changed proportion between cell cycle phases,when compared to non-treated cells (Fig. 6). Thus, PSV1 and SV1 pro-moted an increase in the number of cells in the S phase of the cellcycle, blocking progression to the G2 phase and indicating that thecompounds are cytostatic (Table 3).

Assay to determine vascular endothelial growth factor (VEGF) secretedby RAEC cells

PSV1 and SV1 significantly inhibited VEGF secretion in the culturemedium of RAEC cells, incubated with the samples at concentrationsof 25, 50 and 100 μg/μL. There was no statistically significant differ-ence between the samples. However, there was a significant differ-ence between PSV1 25 (60.4%) and PSV1 50 (71.4%) concentrations(P b 0.001), PSV1 25 (60.4%) and PSV1 100 (69.3%) (P b 0.01), SV125 (66.9%) and SV1 100 (75.9%) (P b 0.01) and SV1 50 (68.3%) andSV1 100 (P b 0.01). The inhibition of VEGF secretion was more effec-tive with PSV1 50 (71.4%) and SV1 100 (75.9%) (Fig. 7).

Discussion

Angiogenesis is a dynamic proliferation and differentiation process,which requires endothelial proliferation, migration and tube formation.Thus, endothelial cells are a preferential target for therapy (Chen et al.,2005). Tumors with strong angiogenic activity are related to a lowerpatient survival rate (Giatromanolaki et al., 1996). The uncontrolledangiogenesis process has been implicated in tumor growth and metas-tases, as well as the progression of disorders known as “angiogenicdiseases”, such as hemagiomas, psoriasis and rheumatoid arthritis(Carmeliet, 2003).

Besides the vertebrate glycosaminoglycans (GAG), marine organ-isms are rich sources for sulfated polysaccharides, such as carrageenansfrom red algae or fucoidans from brown algae. Thus, there is a growinginterest in developing new pharmacological strategies for pre-clinicaland clinical approaches to this process (Groth et al., 2009).

In this respect, the present study demonstrated that SV1 andPSV1, fractionated and purified polysaccharides respectively, isolatedfrom the brown seaweed Sargassum vulgare, exhibit angiogenesis inhib-itory activity when tested in hen embryos and on endothelial celltubulogenesis of rabbit aorta. In this cell line, polysaccharides do notaffect viability or display apoptotic or necrotic activity. We observed

Fig. 7. Inhibition of VEGF secretion. Values are expressed as mean ± SEM (n = 3).Statistical analysis (a): P b 0.05; (b): P > 0.05.

18 C.M.P. Guerra Dore et al. / Microvascular Research 88 (2013) 12–18

that RAEC cells, incubatedwith PSV1 and SV1, inhibited VEGF secretion.PSV1 and SV1 exhibit antiproliferative activity in the HeLa tumor cellline.

Angiogenesis in tumors has been intensely investigated for thelast decade, since it is an essential component in cancer growth andpropagation, and an important target for investigations related to itstherapeutic use (Folkman, 1990; Polverini, 1995). Advances in thisarea have been related to the discovery of growth factors, such asthe vascular endothelial growth factor (VEGF), which specifically reg-ulates endothelial cell proliferation (Ferrara, 2000). It is important toconsider that tumor vasculature not only is a simple supply line ofnutrients to tumors, but also governs tumor physiopathology, andconsequently its growth, metastasis and response to various therapies(Fukumura and Jain, 2007). In this connection, PSV1 and SV1 are com-pounds that act directly on ex vivo and in vitro angiogenesis inhibition.This activity is directly related to the capacity of these polymers toinhibit VEGF secretion in endothelial cells.

In experiments with fibroblasts (Orlandini and Oliviero, 2001), foundthat the VEGF expression is induced by cell-cell contact. Studies withpolysaccharide extracts from fungi have related the monosaccharidecomposition of extracts to their antiangiogenic activity. Extracts rich infucose, glucose and mannose are capable of interacting with cell mem-branes, impeding tube formation (Chen et al., 2005; Nierodzik andKarpatkin, 2006). Hogan and Kolodziej, 2002 observed that coagulationcascade factors are directly related to angiogenesis, and that TF/FVIIaand thrombin induce angiogenesis during cancer progression. By con-trast, FXa does not promote in vitro tubulogenesis. Recent studies inour laboratory demonstrated that SV1 directly inhibits thrombin activityand stimulates FXa (data not shown). Thus, it can be inferred that theantiangiogenic activity of SV1 and PSV1, polymers made up of units offucose, galactose and glucuronic acid, is related to their monosaccharidecomposition, acting directly on coagulation factors and inhibition ofVEGF secretion.

Therefore, SV1 and PSV1 are sulfated polysaccharides from S. vulgarewith potential antitumor therapeutic action, exhibiting an antiangiogenicand cytostatic effect, as well as cytotoxic activity in tumor cells.

References

Almeida-Lima, J., et al., 2010. Evaluating the possible genotoxic, mutagenic and tumorcell proliferation–inhibition effects of a non-anticoagulant, but antithromboticalgal heterofucan. J. Appl. Toxicol. 30, 708–715.

Arnaoutova, I., et al., 2009. The endothelial cell tube formation assay on basementmembrane turns 20: state of the science and the art. Angiogenesis 12, 267–274.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgramquantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.72, 248–254.

Burgermeister, J., et al., 2002. LaPSvS1, a (1 → 3)-beta-galactan sulfate and its effect onangiogenesis in vivo and in vitro. Carbohydr. Res. 337, 1459–1466.

Carmeliet, P., 2003. Angiogenesis in health and disease. Nat. Med. 9.Chen, S.C., et al., 2005. Antiangiogenic activities of polysaccharides isolated from

medicinal fungi. FEMS Microbiol. Lett. 249, 247–254.Chipperfield, H., et al., 2002. Heparan sulfates isolated from adult neural progenitor

cells can direct phenotypic maturation. Int. J. Dev. Biol. 46, 661–670.Costa, L.S., et al., 2011. Antioxidant and antiproliferative activities of heterofucans from

the seaweed Sargassum filipendula. Mar. Drugs 9, 952–966.Cumashi, A., et al., 2007. A comparative study of the anti-inflammatory, anticoagulant,

antiangiogenic, and antiadhesive activities of nine different fucoidans from brownseaweeds. Glycobiology 17, 541–552.

Deocaris, C.C., et al., 2005. Screening for anti-angiogenic activity in shiitake mushroom(Lentinus edodes Berk) extracts. J. Med. Sci. 5, 43–46.

Dias, P.F., et al., 2008. A polysaccharide isolated from the brown seaweed Sargassumstenophyllum exerts antivasculogenic effects evidenced by modified morphogene-sis. Microvasc. Res. 75, 34–44.

Dietrich, C.P., et al., 1995. A new approach for the characterization of polysaccharidesfrom algae: presence of four main acidic polysaccharides in three species of theclass Phaeophycea. Plant Sci. 108, 143–153.

Dodgson, K.S., Price, R.G., 1962. A note on the determination of the ester sulphate contentof sulphated polysaccharides. Biochem. J. 84, 106–110.

Dore, C.M., et al., 2013. A sulfated polysaccharide, fucans, isolated from brown algaeSargassum vulgare with anticoagulant, antithrombotic, antioxidant and anti-inflammatory effects. Carbohydr. Polym. 91, 467–475.

Dubois, M., et al., 1956. Colorimetric method for determination of sugars and relatedsubstances. Anal. Chem. 28, 350–356.

Dvorak, H.F., et al., 1995. Vascular permeability factor/vascular endothelial growthfactor, microvascular hyperpermeability, and angiogenesis. Am. J. Pathol. 146,1029–1039.

Ferrara, N., 2000. Vascular endothelial growth factor and the regulation of angiogenesis.Recent Prog. Horm. Res. 55, 15–35.

Ferrara, N., et al., 2003. The biology of VEGF and its receptors. Nat. Med. 9, 669–676.Folkman, J., 1990. What is the evidence that tumors are angiogenesis dependent?

J. Natl. Cancer Inst. 82, 4–6.Fukumura, D., Jain, R.K., 2007. Tumor microvasculature and microenvironment: targets

for anti-angiogenesis and normalization. Microvasc. Res. 74, 72–84.Giatromanolaki, A., et al., 1996. Prognostic value of angiogenesis in operable non-small

cell lung cancer. J. Pathol. 179, 80–88.Groth, I., et al., 2009. Pharmacological profiles of animal- and nonanimal-derived

sulfated polysaccharides—comparison of unfractionated heparin, the semisyntheticglucan sulfate PS3, and the sulfated polysaccharide fraction isolated from Delesseriasanguinea. Glycobiology 19, 408–417.

Hogan, M., Kolodziej, P.A., 2002. Molecular mechanisms of tubulogenesis. Nat. Rev. 3,513–523.

Keck, P.J., et al., 1989. Vascular permeability factor, an endothelial cell mitogen relatedto PDGF. Science 246, 1309–1312.

Kubota, Y., et al., 1988. Role of laminin and basement membrane in the morphologicaldifferentiation of human endothelial cells into capillary-like structures. J. Cell Biol. 107.

Mauray, S., et al., 1995. Venous antithrombotic and anticoagulant activities of afucoidan fraction. Thromb. Haemost. 74, 1280–1285.

Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: applicationto proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63.

Nierodzik, M.L., Karpatkin, S., 2006. Thrombin induces tumor growth, metastasis, andangiogenesis: evidence for a thrombin-regulated dormant tumor phenotype. CancerCell 10, 355–362.

Orlandini, M., Oliviero, S., 2001. In fibroblasts Vegf-D expression is induced by cell–cellcontact mediated by cadherin-11. J. Biol. Chem. 276, 6576–6581.

Polverini, P.J., 1995. The pathophysiology of angiogenesis. Crit. Rev. Oral Biol. Med. 6,230–247.

Solimene, A.C.C., et al., 1999. Angiogênese em tumores sólidos. Sboc Revista, Porto Alegre,1 9–20.

Zegers, M.M., et al., 2003. Epithelial polarity and tubulogenesis in vitro. Trends Cell Biol.13, 169–176.