Flavonoids suppress human glioblastoma cell growth by inhibiting cell metabolism, migration, and by regulating extracellular matrix proteins and metalloproteinases expression

Accepted Manuscript Flavonoids suppress human glioblastoma cell growth by inhibiting cell metabolism, migration, and by regulating extracellular matrix proteins and metalloproteinases expression Balbino L. Santos, Mona N. Oliveira, Paulo.L.C. Coelho, Bruno.P. S. Pitanga, Alessandra B. da Silva, T. Adelita, Victor Diógenes A. Silva, Maria de F.D. Costa, Ramon S. El-Bachá, Marcienne Tardy, Hervé Chneiweiss, Marie-Pierre Junier, Vivaldo Moura-Neto, Silvia L. Costa PII:

S0009-2797(15)30028-4

DOI:

10.1016/j.cbi.2015.07.014

Reference:

CBI 7429

To appear in:

Chemico-Biological Interactions

Received Date: 23 December 2013 Revised Date:

5 June 2015

Accepted Date: 24 July 2015

Please cite this article as: B.L Santos, M.N Oliveira, P.L.C Coelho, B.P.S Pitanga, A.B da Silva, T. Adelita, V.D.A Silva, M.d.F.D Costa, R.S El-Bachá, M. Tardy, H. Chneiweiss, M.-P. Junier, V. MouraNeto, S.L Costa, Flavonoids suppress human glioblastoma cell growth by inhibiting cell metabolism, migration, and by regulating extracellular matrix proteins and metalloproteinases expression, ChemicoBiological Interactions (2015), doi: 10.1016/j.cbi.2015.07.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Flavonoids suppress human glioblastoma cell growth by inhibiting cell metabolism, migration, and by regulating extracellular matrix proteins and metalloproteinases expression. Balbino L Santos1*, Mona N Oliveira1*, Paulo L C Coelho1, Bruno P S

RI PT

Pitanga1, Alessandra B da Silva1, Taís Adelita T de Almeida1, Victor Diógenes A Silva1, Maria de F D Costa1, Ramon S El-Bachá1, Marcienne Tardy1, Hervé Chneiweiss2, Marie-Pierre Junier2, Vivaldo Moura-Neto3, Silvia

SC

L Costa1§.

1

Laboratório de Neuroquímica e Biologia Celular, Instituto de Ciências da

M AN U

Saúde, Universidade Federal da Bahia (UFBA), Av. Reitor Miguel Calmon s/n, Vale do Canela, 40110-902 Salvador-BA, Brazil; 2

Neuroscience Paris Seine INSERM U 1130, CNRS UMR 8246, UPMC UM

CR18, Université Pierre et Marie Curie, Campus Jussieu, 9 Quai Saint-Bernard, Batiments A-B, 75005 Paris. 3

Instituto de Ciências Biomédicas, Universidade Federal do Rio de Janeiro,

TE D

CCS - Bloco F, 21949-590, Rio de Janeiro, Brazil.

§

EP

*These authors contributed equally to this work Corresponding author: Tel.: +55 71 3283 8919; fax: +55 71 3283 8927. E-mail

AC C

address: [email protected]

ACCEPTED MANUSCRIPT Flavonoids suppress human glioblastoma cell growth by inhibiting cell metabolism, migration, and by regulating extracellular matrix proteins and metalloproteinases expression. Balbino L Santos1*, Mona N Oliveira1*, Paulo L C Coelho1, Bruno P S

RI PT

Pitanga1, Alessandra B da Silva1, Taís Adelita T de Almeida1, Victor Diógenes A Silva1, Maria de F D Costa1, Ramon S El-Bachá1, Marcienne

1

SC

Tardy1, Vivaldo Moura-Neto2, Silvia L Costa1§.

Laboratório de Neuroquímica e Biologia Celular, Instituto de Ciências da

Saúde, Universidade Federal da Bahia (UFBA), Av. Reitor Miguel Calmon s/n,

M AN U

Vale do Canela, 40110-902 Salvador-BA, Brazil; 2

Instituto de Ciências Biomédicas, Universidade Federal do Rio de Janeiro,

CCS - Bloco F, 21949-590, Rio de Janeiro, Brazil.

§

TE D

*These authors contributed equally to this work Corresponding author: Tel.: +55 71 3283 8919; fax: +55 71 3283 8927. E-mail

AC C

EP

address: [email protected]

1

ACCEPTED MANUSCRIPT Abstract

The malignant gliomas are very common primary brain tumours with poor prognosis, which require more effective therapies than the current used, such as with chemotherapy drugs. In this work, we investigated the effects of several

RI PT

polyhydroxylated flavonoids namely, rutin, quercetin (F7), apigenin (F32), chrysin (F11), kaempferol (F12), and 3',4'-dihydroxyflavone (F2) in human GL15 glioblastoma cells. We observed that all flavonoids decreased the number of viable cells and the mitochondrial metabolism. Furthermore, they damaged

SC

mitochondria and rough endoplasmic reticulum, inducing apoptosis. Flavonoids also induced a delay in cell migration, related to a reduction in filopodia-like structures on the cell surface, reduction on metalloproteinase (MMP-2)

M AN U

expression and activity, as well as an increase in intra- and extracellular expression of fibronectin, and intracellular expression of laminin. Morphological changes were also evident in adherent cells characterized by the presence of a condensed cell body with thin and long cellular processes, expressing glial fibrillary acidic protein (GFAP). Therefore, these flavonoids should be tested as

TE D

potential antitumor agents in vitro and in vivo in other malignant glioma models.

Keywords: flavonoids, glioblastoma, fibronectin, metalloproteinase, invasion

AC C

EP

and differentiation.

2

ACCEPTED MANUSCRIPT 1. Introduction The malignant gliomas are very common primary brain tumors responsible for about 40% of all primary tumors and 78% of all malignant tumors of the central nervous system (Louis et al., 2007). Over 80% of these

RI PT

tumors are considered high-grade (grades III and IV) when diagnosed according to the current classification of the World Health Organization (WHO). Glioblastoma (GB) is the most aggressive form of gliomas that affect the brain, is highly infiltrative, and is morphologically very heterogeneous. Currently, the

SC

protocol adopted for the treatment of patients with glioblastoma is based on surgery followed by radiation therapy and chemotherapy with temozolomide (TMZ) (Stupp et al., 2007). However, despite advances in treatment, the

M AN U

average life expectancy of patients with GB is approximately 14 months, and only a small number of patients can survive for up to 5 years after diagnosis. The excessive proliferation, diffuse ability to infiltrate the surrounding brain tissue and the suppression of anti-tumor immune response are crucial biological aspects that contribute to the malignant phenotype of glioblastomas and limit the success of current treatment protocols. The tumor invasion is a

TE D

complex process, in which the neoplastic cells initiate the migration on the primary tumor site, adhere to the extracellular matrix (ECM), degrade its components through proteolytic enzyme activities, and invade the normal

EP

tissue. Therefore, the crucial elements for the invasion of tumor include adhesion and migration that involve interactions between tumor cells and the ECM components surrounding the tumor (Demuth and Berens, 2004; Singh et

AC C

al., 2010).

The ECM is an organized and complex network of molecules that are

typically composed of large glycoproteins, including fibronectins, collagens, laminins and proteoglycans, which assemble fibers or other macromolecular arrangements. The ECM acts as a reservoir of growth factors and fluids, and plays

an

important

role

in

the

organization

of

tissues,

cellular

microenvironments and niches of stem cells. It is tissue specific and adapts to changes in the development, age and disease (Singh et al., 2010). 3

ACCEPTED MANUSCRIPT The enzyme matrix metalloproteinases (MMPs) are a family of endopeptidases, which selectively degrade extracellular matrix components (EMC). MMPs are involved in the migration and invasion of tumor cells, because they contribute for proteolytic degradation of basement membrane. Among MMPs, MMP2 (Gelatinase A, EC 3.4.24.24) has been widely studied

RI PT

because it is involved in the malignancy of tumor cells (Nakada et al., 1999). Flavonoids constitute a group of polyphenolic components originated from different plant species that have a low molecular weight, and present in a variety of fruits, vegetables, cereals, tea, wine, and fruit juices [Harborne, 1986].

SC

They exhibit a variety of biological activities such as anti-inflammatory, antioxidant, antiviral and antitumor actions (Havsteen, 2007). Inhibition of the growth of cancer cells by flavonoids in vitro and in vivo has been reported in

M AN U

several studies. Flavonoids such as quercetin and rutin, commonly found in the diet, have shown antiproliferative and apoptotic effects in glioma cells in vitro (Braganhol et al., 2006; Nguyen et al., 2004; Santos et al., 2011). Quercetin and other flavonoids have also been associated with inhibition of migration and invasion of some types of tumors (Lin et al., 2008; Shen et al., 2010; Wang et al., 2010). Despite this, there are few studies evaluating the mechanisms by

TE D

which different flavonoids inhibit tumor growth and migration of central nervous systems tumors. Understanding the role of flavonoids on the growth, viability and the processes of migration and invasion of gliomas may help to elucidate

EP

the antitumor activity of these natural products and contribute to generate new potential candidates for therapy. Therefore, in this study, we investigated the effects of polyhydroxylated flavonoids on cell viability, phenotypic changes,

AC C

migratory and invasive capacity in human glioblastoma cells of the GL-15 lineage (Bocchini et al., 1993).

2. Methods

2.1. Flavonoids and treatment The flavonoids 3,3′,4′,5,7-pentahydroxyflavone-3-rutinoside (rutin) and 3,3′,4′,5,6-pentahydroxyflavone

(quercetin

or

F7)

were

extracted

and 4

ACCEPTED MANUSCRIPT characterized from Dimorphandra mollis, and 4’,5,7-trihydroxyflavone (apigenin or F32) was extracted and characterized from Croton betulaster in the Laboratory of Organic Chemistry and Natural Products (Institute of Chemistry, UFBa) and in the Laboratory of Research in Materia Medica (Faculty of Pharmacy, UFBa). 5,7-dihydroxyflavone (chrysin or F11; Aldrich, St Louis, MO),

RI PT

3,4',5,7-tetrahydroxyflavone (kaempferol or F12; Sigma, St Louis, MO), and 3',4'-dihydroxyflavone or F2 (Extra synthese -R&D Chemicals) were kindly provided by Prof. Guy G. Chabot from Laboratoire de Pharmacologie Chimique et Génétique, Université Paris Descartes. Structures of these flavonoids are

SC

shown in Table 1. All flavonoids were dissolved in dimethylsulfoxide (DMSO, Sigma, St Louis, MO) at a concentration of 20 mM and stored in the dark at –20ºC. When applied to cells, flavonoids were dissolved in the medium at a final

M AN U

concentration of 50 µM and incubated for 24 or 48 hours. Control cells were treated with the same volume of DMSO (0.5%) that was used as a vehicle for flavonoids, and it did not show any significant effect on the analyzed parameters when compared to cultures that were not exposed to this solvent.

2.2. Cell line and cultures

TE D

GL-15 (Bocchini et al., 1993) and U251 cells (kindly provided by Vivaldo Moura-Neto) derived from human glioblastomas were used between 20 passages, maintained in a humidified atmosphere of 95% air and 5% CO2 at

EP

37°C in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (GIBCO BRL, Grand Island, NY), a nutrient mixture (7 mM glucose, 2 mM glutamine, 0.011 g/l pyruvate) and antibiotics (100 IU/ml

AC C

penicillin G, 100 µg/ml streptomycin). Cells were grown in 100 mm diameter tissue culture plates (TTP, Switzerland) containing 10 ml medium, which was replaced three times a week. Stock cultures were subcultured into new plates every 3-4 days and cells for experiments were seeded into polystyrene culture plates as needed. After 24 h, cells were trypsinized and subcultured in other plates (96-, 24- or 6-wells) according to the analytical procedures to be performed. Dissociated sphere-derived TG-1 cells (Patru et al., 2010, kindly provided by Hervé Chneiweiss) were cultured in 75 cm2 tissue culture flasks plated at 2500-5000 cells/cm2 in DMEM: F-12 medium (1:1) containing the N2, 5

ACCEPTED MANUSCRIPT G5 (containing FGF and EGF) and B27 supplements (all from Invitrogen, France). They were dissociated in single-cell suspension each week with a renewal of two third of their culture medium.

RI PT

2.3. Cell viability and proliferation tests

2.3.1. Trypan blue staining

Membrane integrity and cell viability were evaluated after Trypan blue staining in the control group and treated cells seeded on 40 mm polystyrene

SC

culture dishes (1.6 x 104 cells/cm2). Both adherent and floating cells were obtained after trypsinization and were centrifuged for 10 minutes at 1,300 x g. Cells were then suspended in 200 µl DMEM without supplements and stained

M AN U

with Trypan blue at a final concentration of 0.1% (p/v). Three replicate experiments were performed for each analysis; the number of viable and nonviable cells/µl was determined after 48 h exposure to flavonoids by counting four 10 µl samples of cell suspension for each experiment in a Burker chamber

2.3.2. MTT Test

TE D

(Boehringer Mannheim).

Flavonoids were tested for its cytotoxicity towards GL-15, U251 and dissociated

TG-1

cells

using

the

3-(4,5-dimethylthiazol-2-yl)-2,5-

EP

diphenyltetrazolium bromide (MTT; Sigma, St Louis, MO) test in 96 well plates (TPP Switzerland) after cells had become confluent (95%). Cells were seeded at a density of 5.0 x 104 cells/cm2. The cells were exposed to flavonoids for 48

AC C

hours and 2 hours before the end of exposure time, the culture medium was replaced by a solution of MTT (1mg/ml) diluted in DMEM and then the plate was incubated for 2 hours. Thereafter, cells were lysed with 20% (w/v) SDS, 50% (v/v) DMF, pH 4.7 adjusted with a solution of 80% (v/v) acetic acid, 2.5% (v/v) 1M HCl, and plates were kept overnight at 37°C in order to dissolve formazan crystals. The cell cytotoxicity was quantified based on the conversion of yellow MTT to purple MTT formazan by mitochondrial dehydrogenases of living cells. The optical density of each sample was measured at 580 nm using a BIO-RAD 6

ACCEPTED MANUSCRIPT 550 PLUS spectrophotometer. Eight replicate wells were used for each

RI PT

analysis.

2.3.3. BrdU cell proliferation assay.

Proliferation was evaluated using the BrdU cell proliferation assay

SC

(Sigma-Aldrich, Inc). After 48 h of exposure to flavonoids, BrdU (10 µM) was added to wells of the plate. Cells were incubated for 2 hours in a humidified

M AN U

atmosphere of 95% air and 5% CO2 at 37°C. Cells were fixed and DNA was denatured by treatment with denaturing solution (2N HCl) for 20 min at room temperature. Mouse anti-BrdU monoclonal antibody (1/200, Sigma-Aldrich, Inc) diluted in PBS, was pipetted into the wells and allowed to incubate for 1 h. Unbound antibody was washed away and cells were incubated with conjugated antibodies specific for mouse IgG Alexa Fluor 488 (1/5000, Invitrogen

TE D

corporation) diluted in PBS-T, for 1 h under slow agitation at room temperature. After incubation, the cell nuclei were stained with DAPI (5 µg/ml) for 10 minutes at room temperature. All reagents were provided with the kit and used in accordance with the manufacturer’s instructions. Experiments were performed

EP

in triplicate. Thereafter, cells were analyzed using an epifluorescence microscope (Olympus BX-70) and photographed. Quantification was analyzed

AC C

with ImageJ 1.33u (Wayne Rasband, National Institute of Health, USA).

2.3.4. Transmission electron microscopy (TEM) The ultrastructural changes in GL-15 cells treated with flavonoids were

investigated by transmission electron microscopy (TEM). The cells were grown in 50ml culture bottles to form confluent monolayers. After reaching confluence, the monolayer was treated with the flavonoids studied pre-diluted in DMEM free from fetal calf serum (FCS) for 48 hours. The culture medium was discarded and the cell was fixed (2.0% glutaraldehyde, 2% paraformaldehyde in 0.1M 7

ACCEPTED MANUSCRIPT sodium cacodylate buffer, pH7.4) at room temperature (RT) for 2 hours. After this period, the samples were scraped from the bottle, centrifuged at 4,000 RPM for 5 minutes. The supernatant was discarded, and samples washed (3 times in 0.1M sodium cacodylate buffer) at intervals of 5 minutes and centrifuged at 4,000 RPM. Then, the samples were dehydrated in acetone at concentrations

RI PT

30 to 100% at intervals of 10 minutes for each concentration. After dehydration, the material passed through the substitution step in 100% acetone+epoxyresin (1:1, v/v) for 24 hours, then this was included in 100% epoxyresin (Polybed) mounted in BEEM© capsules. The histological sections were made in Leica

SC

ultramicrotome (Germany) and then stained with uranyl acetate and counterstained with lead citrate. The material was then analyzed and documented in a transmission electron microscope (Zeiss), on Electron Laboratory

of

the

Gonçalo

Moniz

M AN U

Microscopy

Research

Center,

CPqGM/Fiocruz, Salvador -Bahia.

2.4. Invasion assay

2.4.1. Scratch assays in monolayers

TE D

The potential of flavonoids in inhibiting cell migration was investigated by a monolayer single lesion test or Scratch assays. The test was performed in 6well culture plates. After GL-15 glioma cells were allowed to attach and reach

EP

confluence, a scratch (2 mm) was made through the culture dish. The cells were washed twice with phosphate-buffered saline (PBS) to remove detached cells for injury before their subsequent incubation with culture medium in the

AC C

absence (control) or presence of flavonoids studied. Cells were photographed at the time intervals 0, 12, 24 and 48 hours during treatment to assess the reduction in the length of the lesion areas.

2.4.2. Investigation of filopodias by scanning electron microscopy (SEM) After treatment with flavonoids, the culture medium was discarded and cells were fixed (2% glutaraldehyde, sodium cacodylate buffer 0.1M, pH 7.4) for 12 hours at 4 °C. Then, the cells were washed with sodium cacodylate buffer 0.1 M at pH 7.4 (3 times for 5 min at RT). After the final washing, cells were 8

ACCEPTED MANUSCRIPT fixed with osmium tetroxide solution (1% osmium tetroxide, 2% sodium cacodylate buffer 0.2M, v/v) and incubated for 1 hour at room temperature (RT). After fixation, the cells were washed with sodium cacodylate buffer 0.1 M, pH 7.4 (3 times for 10 minutes), dehydrated in alcohol solution at concentrations 30 - 100%, for 10 minutes at room temperature. Then, cells were subjected to

RI PT

critical point CPD030 Leica equipment IN and then coated with gold, using a Denton Desk Vacuun IV equipment. The cells were then analyzed in a scanning electron microscope (JSM 6390LV SEM), on the Electron Microscopy Laboratory of the Gonçalo Moniz Research Center – CPqGM/FIOCRUZ,

M AN U

2.5. Morphological and structural analysis

SC

Salvador-Bahia.

2.5.1. Phase microscopy and immunocytochemistry

The cell cultures were exposed to flavonoids (50µM) for 48 hours and after the exposure, the morphological aspects were analyzed and photographed in an optic phase microscope (Nikon TS-100) using a digital camera (Nikon E-

experiments.

TE D

4300). Ten pictures were analyzed for each treatment in three independent

Morphological changes and astroglial differentiation were also studied by immunocytochemistry for the cytoskeletal GFAP (glial fibrillary acidic protein),

EP

an astrocyte marker. All control and treated cells seeded on 24 well polystyrene culture plates (6 x 104 cells/well) were rinsed three times with PBS and fixed with cold methanol at -20oC for 10 minutes. Cells were incubated with rabbit

AC C

polyclonal antibodies specific for GFAP (1/200, Santa Cruz) diluted in PBS-T, for 1 h under slow agitation. After three washes with PBS, cells were incubated with antibodies specific for rabbit IgG conjugated to Alexa Fluor 488 (Invitrogen Corporation) diluted in PBS-T (1/5000), for 1 h under slow agitation at room temperature. After incubation, the cell nuclei were stained with DAPI (5µg/ml) for 10 minutes at room temperature, and the cytoplasm stained with Evans Blue dye (10mg/mL -Fluka) for 10minutes. Thereafter, cells were analyzed using an epifluorescence microscope (Olympus BX-70) and photographed. 9

ACCEPTED MANUSCRIPT 2.6. Expression of ECM components

2.6.1. Immunocytochemistry After a 48 h exposure to flavonoids, all control and treated cells seeded on 24 well polystyrene culture plates (6 × 104 cells/well) were rinsed three times

RI PT

with PBS and fixed with cold methanol at -20 oC for 10 minutes to investigate the intracellular synthesis. Cells were incubated with rabbit polyclonal antibodies specific for fibronectin (1/500, Sigma-Aldrich, Inc) and laminin (1/25, Sigma-Aldrich, Inc) diluted in PBS-T, for 1 h under slow agitation. After three

SC

washes with PBS, cells were incubated with rhodamine-conjugated antibodies specific for rabbit IgG (1/5000, Invitrogen corporation) diluted in PBS-T, for 1 h under slow agitation at room temperature. After incubation, the cell nuclei were

M AN U

stained with DAPI (5µg/ml) for 10 minutes at room temperature. Thereafter, cells were analyzed using an epifluorescence microscope (Olympus BX-70) and photographed. For extracellular labeling, the procedure was similar to that reported above, however fixation in methanol was performed after labeling with primary and secondary antibodies. Quantification was analyzed with ImageJ

2.6.2. Western blot

TE D

1.33u (Wayne Rasband, National Institute of Health, USA).

To investigate the expression of ECM proteins, the GL-15 cells were

EP

seeded in six well polystyrene culture plates (1.5 × 105 cells/well) and treated with flavonoids for 48 h. After a 48 h exposure, cells were rinsed twice with PBS, lysed and harvested in a 2% (w/v) SDS, 2 mM EGTA, 4 M urea, 0.5%

AC C

(v/v) Triton X-100, 62.5 mM Tris-HCl buffer (pH 6.8), supplemented with 1 µl/ml protease inhibitor cocktail (Sigma, St Louis, MO). Protein content was determined by the method of Lowry, Rosebrough, Farr, and Randall (1951) using a protein assay reagent kit (Bio-Rad, Hercules, CA). Thirty micrograms of total protein extract were loaded onto a discontinuous 4% stacking and 8% running SDS polyacrylamide gel (SDS–PAGE). Electrophoresis was performed at 200 V for 45 min. Proteins were then transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA, EUA) at 100 V for 1h. Equal protein loading was confirmed by staining membranes with Ponceau Red (Sigma). Membranes 10

ACCEPTED MANUSCRIPT were then blocked for 1 h at room temperature in 20 mM PBS (pH 7.5) containing 0.05% Tween 20 (PBS-T) and 5% powdered skim milk. Subsequently, membranes were incubated with rabbit anti-fibronectin (1:2000, Sigma-Aldrich, Inc) or rabbit anti-laminin (1:1000, Sigma-Aldrich, Inc ) diluted in TBS-T containing 1% powdered skim milk for 1 h. Conjugated alkaline

RI PT

phosphatase goat anti-rabbit IgG (1:5000 in TBS-T, Bio-Rad, Hercules, CA) was used as a secondary antibody. Immunoreactive bands were visualized using an alkaline phosphatase (AP)-conjugated substrate kit (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions. Quantification was

SC

performed by scanning densitometry (ScanJet4C – HP) of three independent experiments and analyzed with ImageJ 1.33u (Wayne Rasband, National

M AN U

Institute of Health, USA).

2.7. Expression and activity of MMP

2.7.1. Western blot

To investigate the expression of MMP proteins, the GL-15 cells were seeded in six well polystyrene culture plates (1.5 × 105 cells / well) and treated

TE D

with flavonoids for 48 h. The active (63kD) and inactive (72kD) forms of gelatinase MMP-2 were investigate. After a 48 h exposure, cells were rinsed twice with PBS, lysed and harvested in a 2% (w/v) SDS, 2 mM EGTA, 4 M urea,

EP

0.5% (v/v) Triton X-100, 62.5mM Tris-HCl buffer (pH 6.8), supplemented with 1µl/ml protease inhibitor cocktail (Sigma, St Louis, MO). Protein content was determined by the method of Lowry, Rosebrough, Farr, and Randall (1951)

AC C

using a protein assay reagent kit (Bio-Rad, Hercules, CA). Thirty micrograms of total protein extract were loaded onto a discontinuous 4% stacking and 8% running SDS polyacrylamide gel (SDS–PAGE). Electrophoresis was performed at 200 V for 45 min. Proteins were then transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA, EUA) at 100 V for 1h. Equal protein loading was confirmed by staining membranes with Ponceau Red (Sigma). Membranes were then blocked for 1 h at room temperature in 20 mM TBS (pH 7.5) containing 0.05% Tween 20 (TBS-T) and 5% powdered skim milk. Membranes were then blocked for 1 h at room temperature in 20 mM TBS (pH 7.5) 11

ACCEPTED MANUSCRIPT containing 0.05% Tween 20 (TBS-T) and 5% powdered skim milk. Subsequently, membranes were incubated with rabbit anti-MMP-2 (1:500, Santa Cruz) diluted in TBS-T containing 1% powdered skim milk for 1 h. Conjugated alkaline phosphatase goat anti-rabbit IgG (1:5000 in TBS-T, BioRad, Hercules, CA) was used as a secondary antibody. Immunoreactive bands

RI PT

were visualized using an AP-conjugated substrate kit (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions. Quantification was performed by scanning densitometry (ScanJet4C – HP) of three independent experiments and analyzed with ImageJ 1.33u (Wayne Rasband, National Institute of Health,

SC

USA).

2.7.2. Immunocytochemistry

M AN U

Control and treated cells seeded on 24 well polystyrene culture plates (6 × 104 cells/well) were rinsed three times with PBS and fixed with cold methanol at -20o C for 10 minutes. Then, cells were incubated with rabbit polyclonal antibodies specific for MMP-2 (1/500, Santa Cruz) diluted in PBS-T, for 1 h under slow agitation. After three washes with PBS, cells were incubated with conjugated antibodies specific for rabbit IgG ALEXA FLUOR 488 (1/5000,

TE D

Invitrogen corporation) diluted in PBS-T, for 1 h under slow agitation at room temperature. After incubation, the cell nuclei were stained with DAPI (5µg/ml) for 10 minutes at room temperature, and the cytoplasm stained with Evans Blue

EP

dye (10mg/mL - Fluka) for 10minutes. Thereafter, the cells were washed with PBS and analyzed using an epifluorescence microscope (Olympus BX-70) and

AC C

photographed.

2.7.3. Gelatinase zymography of MMP-2 and MMP-9 Matrix metalloproteinases (MMP) released into conditioned media were

determined by gelatinase zymography according to the method of Planchenault et al. (2001) and Vincent et al. (2001) with minor modifications. Briefly, medium was collected from control and treated cultures, their protein concentration was determined by colorimetric method of Lowry. 30 µg of protein were applied per well on 10% polyacrylamide gels (SDS-PAGE) containing 1mg/ml gelatin (Sigma–Aldrich). After electrophoresis, the gels were renatured in 2.5% Triton 12

ACCEPTED MANUSCRIPT X-100 (3 times for 10 min), then incubated overnight at 37°C in reactivating enzyme buffer (50mM Tris–HCl, pH 7.5; 10mM CaCl2; 0.2M NaCl). The gels were stained with 0.5% Coomassie Brilliant Blue R-250 (Sigma–Aldrich) for 30 minutes and destained (50% methanol solution; 10% glacial acetic acid) for 15 minutes until the appearance of the emerging bands (areas of degradation of

RI PT

gelatin).The gels were scanned (ScanJet4C -HP), and densitometric analysis of the bands were evaluated using the software Image J 1.33u (Wayne Rasband, National Institute of Helth, USA).

SC

2.8. Statistical analysis

Each experimental variable was analyzed in three independent experiments and results are expressed as mean ± standard deviation. One-way

M AN U

ANOVA followed by the Student-Newmann-Keuls test was used to determine the statistical differences among groups differing in only one parameter.

3. Results

3.1. Flavonoids were cytotoxic, inhibited proliferation and induced

TE D

apotosis in glioblastoma cells

The effect of flavonoids on the metabolic activity of human glioblastoma cell lines GL-15, U251 and TG-1 cells after 48 hours of treatment was

EP

investigated by the MTT test, which measures the dehydrogenases function in the mitochondrial metabolism. When compared to control (DMSO 0.1%), it was observed that the flavonoids chrysin, apigenin and 3',4'-dihydroxyflavone were

AC C

cytotoxic (p < 0.05) to GL-15 cells (Fig.1a and b), inducing significant decreases on mitochondrial activities. However, kaempferol, quercetin, and rutin (50 µM) showed no effect on mitochondrial dehydrogenase functions of GL-15 cells, even induced significant reduction on the number of viable cells after 48 hours treatment and staining with trypan blue dye (Fig.1 c). This assay demonstrated a significant decrease (p <0.05) in the number of viable cells exposed to 50 µM flavonoids chrysin, 3',4'-dihydroxyflavone, apigenin, kaempferol and rutin. The flavonoid 3',4'-dihydroxyflavone was the most potent. In the proliferation assay 13

ACCEPTED MANUSCRIPT after BrdU incorporation, we observed a significant reduction in BrdU incorporation for all treatments (Fig. 1 d). The analysis of images obtained from transmission electron microscopy (TEM) showed that in control conditions group (Fig.1e), the GL-15 cells showed a dense cytoplasm, with nuclei of different sizes and shapes and

RI PT

prominent nucleoli (Fig.1d-A) with cytoplasmic extensions filopodia like (Fig. 1eB). The mitochondria visualized presented different shapes and sizes, sometimes with swollen appearance and disorders in the crests (Fig. 1e-D). The endoplasmic reticulum and the Golgi complex showed no visible changes (Fig.

SC

1e-B, C and D). For treated group, the cells treated with flavonoid 3',4'dihydroxyflavone showed some ultrastructural changes characteristic of apoptosis as dilated cisternae of the Golgi complex an drough endoplasmic

M AN U

reticulum (Fig. 1e-A, B and C), nuclei with aspect retracted and condensation of chromatin material (Fig. 1e-A). It was also observed electrodense mitochondria with different size and shape, presenting some aspect disarranged crests type cristolysis (Fig.1e-D), the presence of several concentric lamellae type myelin figures, vacuolar structures in the cytoplasm (Fig. 1e-C), and structures on the

TE D

periphery of the plasma membrane similar to blebs (budding).

3.2. Flavonoids interfere in cellular migration and increase ECM components

EP

To determine the anti-invasive potential of flavonoids in glioblastoma cells, initially we sought to determine the migration of cells in assays of single lesion in monolayers of GL-15 cells. We found that the flavonoids induce a

AC C

delay in migration of GL-15 cells to sites of injury, particularly during the first 12 hours after treatment (Fig. 2A) when compared to control conditions (0.1% DMSO). It was also felt an accumulation of cells apparently falling of the carpet cells after treatment with the flavonoid rutin. Filopodia is a finger-like actin protrusion in plasmatic membrane associated to potential of migration and invasion of cells. To investigate inhibition of invasion we evaluating the presence of filopodia in control condition and treated cells. The scanning electron microscopy of GL-15 cells in the control condition (DMSO 0.1%) showed heterogeneity in cell shape and 14

ACCEPTED MANUSCRIPT cytoplasmatic prolongations, including the presence of extensive filopodia (Fig. 2b - A and B). Cells were observed displayed extensive cytoplasmatic projections, as illustrated in Figure 2b. In these projections, it is possible to observe the presence of numerous filopodia on the cell surface, with concentration at the end of the cytoplasmatic projection of the cell (Fig. 2b - B

RI PT

and C). In cells treated with flavonoids was evident the presence of bipolar cells with retracted cytoplasm (Fig.2b - D) and thin and long cytoplasmic extension shaving heterogeneity with respect the presence of filopodia (Fig.2b - E). In the treatment with rutin, apigenin and kaempferol, for example, bipolar cells was

SC

observed with aspect radial glial type, emitting long cytoplasmic processes with limited surface filopodia (Fig.2b – D, E and F, respectively), compared to control

dihydroxyflavone and chrysin.

M AN U

(0.1% DMSO). A similar aspect was observed for the flavonoids 3',4'-

The anti-invasive potential was also determined by analyzing the expression of ECM components, as well as the expression and activity of MMPs. Immunostaining for ECM proteins, laminin and fibronectin was performed in the extracellular and cytoplasmatic compartment. In control conditions, GL-15 cells expressed low fibronectin and laminin, both extracellular

TE D

and intracellular level (Fig. 3a e 3b). After treatment with the flavonoids has been observed an increase in the labeling of fibronectin extracellular which was more evident in the cultures treated with apigenin (Fig. 3a/pos-fixed). This

EP

increased expression of fibronectin induced by flavonoids was also evidenced by marking the protein in the intracellular and extracellular compartments (Fig. 3a/pre-fixed), indicating that all flavonoids tested induced an upregulation in the

AC C

synthesis of fibronectin. We not observed marking for lamininin the control conditions (0.1% DMSO) and treatments in pos-fixed cells (Fig. 3b/pos-fixed). However, analysis of expression of laminin after fixation showed that flavonoids induced protein synthesis, especially apparent in the perinuclear region (Fig. 3b/pre-fixed). The expression of ECM proteins were also evaluated by western blot technique (Fig. 3c). In control conditions was observed low expression of fibronectin, which was induced in cultures treated with flavonoids chrysin, kaempferol, apigenin and especially rutin. The expression of laminin by this technique was only evident in the cultures treated with rutin. 15

ACCEPTED MANUSCRIPT MMPs are expressed and secreted as inactive precursors that are activated by removal of a pro-peptide N-terminus. The latent and active forms of the enzymes can be differentiated according to their molecular weights. The expression of MMP-2 was assessed by western blot and immunocytochemistry. The expression of MMP-2 (Fig. 4a – 4c) was clearly observed in cultures on

RI PT

condition control (0.1% DMSO). Moreover, the cell cultures treated with 3', 4'dihydroxyflavone showed a weak labeling, which was not evident for the other treatments (Fig. 4a and 4b). The analysis of MMP expression by western blot analysis (Fig. 4c) also revealed a reduction inexpression of MMP2 (63kDa) at

SC

lower levels after treatment with chrysin, kaempferol and quercetin, confirming the findings of the immunocytochemical analysis (Fig. 4a). No alterations was noted by western blot analysis in active MMP-2 expression when cells were

M AN U

treated with the flavonoids 3',4'-dihydroxiflavone; apigenin and rutin. Assessment of the activity of gelatinases by zymography showed a reduction in metalloproteinase activity after 24 hours treatment with flavonoids tested. However, cells treated with the flavonoid 3',4'-dihydroxyflavone, chrysin, kaempferol and quercetin, showed amost expressive reduction of gelatinases

TE D

(Fig.4d).

3.3. Effect of flavonoids on the morphology and structure of GL-15 cells Morphological and structural changes in the GL-15 cells after treatment

EP

with the flavonoids were analyzed by phase contrast microscopy as well as by immunocytochemistry for the glial fibrillary acidic protein GFAP. GL-15 cells in control conditions (0.1% DMSO) form a monolayer of cells with bipolar

AC C

phenotype and volume of the cell body slightly enlarged. In addition, the flavonoids proved to be morphogenic for GL-15 cells, mainly apigenin, quercetin and rutin. It was noted that the treated cells were more elongated, with the emission of fine and long cytoplasmic processes and cytoplasmic retraction (Fig. 5/phase). A decrease in cellularity was also observed after treatment with flavonoids, especially with the flavonoid rutin. The immunodetection of GFAP demonstrated a heterogeneous protein labeling restricted to a small population of GL-15 cells in control conditions (0.1% DMSO). In the cultures treated with flavonoids was possible to observe 16

ACCEPTED MANUSCRIPT some cells stained for GFAP, especially treatment with apigenin, quercetin and rutin in which the distribution of the staining was evident in the cellular body and throughout the extent of the thin cytoplasmic process emitted by GL-15 cells,

RI PT

acquired a phenotype pattern of astrocytic differentiation (Fig. 5/GFAP).

4. Discussion

In our studies on the antitumor potential of rutin (Santos et al., 2011), we observed that this flavonoid inhibited the proliferation and viability, proportionate

SC

to dosage, in glioblastoma cells of the GL-15 lineage, as well as inducing apoptosis when treated in concentrations of 10 to 100 µM . The cellular lineage GL-15, established by Bocchini et al (1993), corresponds with highly

M AN U

proliferating human glioblastoma cells. In the interests of continuing the evaluation of the antitumor potential of flavonoids, the present study investigated the antitumor properties of polihydroxylated flavonoids (3',4'dihydroxyflavone, quercetin, chrysin, kaempferol, apigenin and rutin) - extracted from plants of Northeastern Brazil’s semi-arid region - in terms of cytotoxicity, proliferation, viability, morphological and ultra-structural alterations, and cellular

TE D

migration in human glioma cell culture.

We investigate by MTT test if the flavonoids interfere in the mitochondrial function of glioblastoma cells. Through this technique we demonstrate that

EP

amongst the flavonoids tested in a 50 µM concentration, 3',4'-dihydroxyflavone, apigenin and chrysin reduce the mitochondrial function of tumor cells. On the other hand, we observed that the flavonoids kaempferol, rutin and quercetin did

AC C

not have a significant effect on mitochondrial functioning, even though they interfered in the metabolism. These findings show that the flavonoids tested are capable of altering the cellular metabolism of glioblastoma cells of the GL-15 lineage.

The effect of flavonoids on the viability and proliferation in GL-15 cells was observed after a period of 48 hours of treatment. The proliferation was significantly reduced when the cells were treated with chrysin, 3',4'dihydroxyflavone, apigenin, kaempferol, quercetin and rutin. Amongst the flavonoids studied, we highlight the flavonoids 3',4'-dihydroxyflavone and rutin 17

ACCEPTED MANUSCRIPT for having reduced most significantly the number of viable cells. On the one hand, the flavonoid 3',4'-dihydroxyflavone showed itself to be more cytotoxic, significantly reducing the number of viable cells. On the other hand, the flavonoid rutin, even though it did not demonstrate significant cytotoxicity, also reduced the number of viable cells after 48 hours of treatment.

RI PT

Many investigations have showed that flavonoids interfere in the metabolic activity of tumors. Investigations are demonstrated with total extract of Scutellaria baicalenses that flavonoids present in this extract inhibited in a dose-dependent way cellular viability in the lineage of human glioblastoma, as

SC

well as inhibiting growth and inducing apoptosis, indicating the flavonoids of this extract as promising assistants in the treatment of malignant gliomas (Scheck et al., 2006). Other study showed that the flavonoid quercetin inhibited cellular

M AN U

proliferation in glioblastomas of the U138MG lineage, causing inhibition in cellular viability, inducement of apoptosis and stoppage of the cellular cycle in the G2 phase (braganhol et al., 2006).

Amongst the flavonoids used in this study, some antitumor properties of the flavonoids rutin, quercetin and apigenin have been studied. The flavonoid apigenin showed in vitro inhibition of the proliferation and viability of cancer cells

TE D

in the breast, melanomas, uterine cervix and in lung adenocarcinoma (Cardenas et al., 2006). Yang et al (2000), by studying in vivo with rats induced with intestinal dysplasia, observed that after treatment with a diet rich in rutin

EP

and quercetin there was an increase in the number of cells in apoptosis, indicating a possible inhibitory effect in colon carcinogenicity. In other study, were observed that quercetin presented an anti-proliferative and apoptotic effect

AC C

in lung cancer of the A549 lineage, with apoptosis associated with the activation of the MEK-ERK pathway (Nguyen et al., 2004). On the other hand, other studies have demonstrated that inhibition of proliferation and the apoptosis in tumor cells treated with flavonoids are associated with a reduction in ERK activation. Gliomas treated with kaempferol and rutin inhibited growth related to an apoptotic effect due to diminishing ERK activation (Jeong et al., 2009; Santos et al., 2011). Thus we have different classes of flavonoids presenting similar effects in relation to cell growth and death. This strengthens knowledge on the potential of flavonoids in their regulation of the growth and death of cells, 18

ACCEPTED MANUSCRIPT by different signaling mechanisms. By that very fact we emphasize the importance of more studies to characterize the effect of flavonoids on mechanisms and signaling pathways in the proliferation of tumors, and especially in human glioblastoma cells. The capacity of the flavonoids to induce change in the morphology and

RI PT

differentiation of GL-15 cells was studied under the phase contrast microscope, by GFAP immunocytochemistry and scanning electron microscopy. We have observed that all flavonoids studied here induced morphological modifications after 48 h of treatment in the 50 µM concentration. The GL-15 cells presented -

SC

whilst exposed to the flavonoids 3',4'-dihydroxyflavone, chrysin, kaempferol, apigenin, quercetin and rutin - contraction in the cell body and acquisition of a bipolar phenotype emitting cytoplasmic extensions characteristic of astroglial

M AN U

differentiation. These morphological modifications in the GL-15 cells were also evident under the scanning electron microscopy, where we could observe with more detail the morphological changes induced by the flavonoids. In the immunostaining of the GFAP protein, we observed a heterogeneous expression in the GL-15 cells, better observed in cells which presented a more fusiform and bipolar morphology. GFAP is a more important protein in the intermediate

TE D

filaments in astrocytes, considered a specific protein which distinguishes these cells. Some works have related that the increase in the expression of GFAP is associated with differentiation of malignant gliomas cells (Santos et al., 2011).

EP

This suggests that gliomas treated with flavonoids would be assuming structural characteristics similar to those in astrocyte cells. The immunocytochemical analysis done by Bocchiniet al (1993) with GL-15 cells revealed that these cells and

homogeneously

AC C

constitutively

express

the

protein

vimentin,

and

heterogeneously the GFAP protein. These same authors displayed a densitydependent expression for GFAP in GL-15 cells. Our results demonstrated that the flavonoids evaluated were capable of inducing phenotype alterations in glioblastoma cells of the GL-15 lineage, alterations which suggest a morphogenic potential. To migrate, the cells characteristically need to acquire a polarized morphology in response to various extracellular signals with the formation of cytoplasmatic lamellipodia and filopodia type extensions which are orchestrated 19

ACCEPTED MANUSCRIPT by the actin filaments. In the extremity of the lamellipodia the cells form adhesions which connect the extracellular matrix to the actin cytoskeleton, permitting the anchorage of the cytoplasmatic extensions and consequently facilitating the dragging of the cellular body (Le Clainche & Carlier, 2008). In tumor cells and transformed cells, the metalloproteases (MMPs) associated with

RI PT

the lamellipodia mediate the proteolysis of the constituents of the ECM, including fibronectin, laminin and collagen, contributing in this way to the process of cellular migration and invasion. Recent research has suggested that MMPs have a central role in the physiopathology of brain tumors and that the

SC

presence of these enzymes seems to be strongly associated with the tumor’s aggressiveness (Shen et al., 2010; Wang et al., 2010). The migration and invasion of gliomas is a process which occurs in various stages, where as the

M AN U

tumor separates itself from the primary lesion it establishes new contacts with the extracellular matrix (ECM), degrading and remodeling this matrix in order to disperse itself through the normal tissues of the brain (Tzu et al., 2008). In our study we investigate the anti-migratory and anti-invasive effect of flavonoids in GL-15 cells by test of simple lesion through the cellular monolayer, analyzing the presence and activity of metal proteases MMP-2 and MMP-9 and the

TE D

expression of ECM components (laminin and fibronectin). We also analyze, by electron transmission microscopy and scanning electron microscopy, the effect of flavonoids with regard to the changes induced in the formation of lamellipodia

EP

and filopodia extensions, structures which are also associated with the invasive character of tumors. Our studies showed that all of the flavonoids evaluated diminished the activity of the gelatinase MMP-2 when analyzed by zymographic highlighting

AC C

method,

the

flavonoids

3',4'-dihydroxyflavone;

3,3',4',5,6-

pentahydroxyflavone (quercetin); 5,7-dihydroxyflavone (chrysin) and 3,4',5,7tetrahydroxyflavone (kaempferol). We also showed that flavonoids reduced the expression of MMP-2, highlighting quercetin, chrysin, kaempferol and rutin. It was also shown that GL-15 cells are very agile in their capacity of migration, and that cells treated with polihydroxylated flavonoids presented retardation during migratory activity. These findings suggest that the flavonoids considered presented an anti-invasive and anti-migratory effect in cells of the GL-15 lineage 20

ACCEPTED MANUSCRIPT with regard to the potential of these composites in regulating the expression and activity of MMPs. The action of some flavonoids in the regulation of migration and invasion of tumoral cells has also been observed. Wang et al. (2001) showed that the flavonoid baicalein inhibited the expression and secretion of the metal

RI PT

proteases MMP-2 and MMP-9, as well as repressing adhesion to the fibronectin base and migration in human brain carcinomas of the MDA-MB-231 lineage. Evaluation of the antitumor effect of the acacetin in cells of prostate cancer D145, also showed that the flavonoid could inhibit the ability for these cells to

SC

migrate and invade by the reduction of MMP-2 and MMP-9, which was associated with the inhibition of phosphorylation of the protein p38 MAPK (Shen et al., 2010). In another investigation was shown that the flavonoid quercetin

M AN U

presented an inhibiting effect in the MMP-9 of breast carcinoma cells of the MCF-7 lineage (Lin et al., 2008). These authors observed further still that this effect was associated with the antioxidant activity of quercetin, an activity being attributed to the hydroxylate groups (OH) localized in both carbons C3’ and C4’; and additionally, an OH group localized in one of the carbons C3, C5 or C7 enabled the inhibiting effect in the MMP-9 expression and consequent inhibition

TE D

of the invasive potential of MCF-7 cells. In our own study, besides the quercetin (3,3’,4’,5,6-pentahydroxyflavone),

the

flavonoids

3',4'-dihydroxyflavone,

kaempferol, apigenin and rutin also present OH groups in carbons C3’ or C4’,

EP

with the exception of chrysin which presents OH groups in carbons C5 and C7. Rutin, besides presenting OH groups in both carbons C3’ and C4’, also presents hydroxyl groups in carbons C3 and C5. The flavonoids which

AC C

presented the OH group in C3’ carbon and in the C4’ carbon, or in both, positively regulated the GFAP expression, and presented a regulating effect in the activity and expression of MMPs as well in the retardation of cellular migration. The flavonoid kaempferol, in spite of presenting OH only in C4’ carbons, in GL-15 cells showed an expressive effect in the inhibition of the expression of MMP-2 and in retarding cellular migration. In this way, our results corroborate with data observed in other studies with differing lineages of tumors. 21

ACCEPTED MANUSCRIPT The protein laminin corresponds with a glycoprotein family involved in maintaining the integrity of the architecture of basal membranes throughout various tissues. This glycoprotein group participates in the formation of the structure of the extracellular matrix (ECM), interacting with different types of normal or pathological cells, influencing their proliferation, differentiation,

RI PT

adhesion and migration, hence their importance in the regulation of cellular migration, principally in neoplastic cells. The presence of some classes of receptors for laminin in the cell membrane is crucial for this regulation. An important group of receptors for laminin are the integrins which recognize and

SC

link themselves to a specific site in one of the laminin chains α, β or γ, unchaining a series of cellular effects (Tzu et al., 2008). Other proteins which are also associated with cellular adhesion to the ECM are the fibronectins.

M AN U

Some research has showed a strong staining in ECM proteins, especially laminin and fibronectin in glioblastoma cellular lineages - in vitro and in vivo suggesting that the expression of ECM components through these cells would be involved in the regulation of invasion and migration of tumors. Planchenault et al. (2001), for example, showed that glioma cells of the 8MG and 42MG presented intracellular and extracellular imunoreactivity for fibronectin, while in

TE D

cells of the GL-15 lineage intracellular immunoreactivity was more punctual for fibronectin, localized in the perinuclear area in a restricted number of these cells, imunoreactivity not being observed in the extracellular medium. In this

EP

same work, the authors showed a punctual staining for intracellular laminin in the three cellular lineages. In our work, we observed a discreet distribution for laminin in the extracellular medium of GL-15 cells and a more evident

AC C

cytoplasmatic expression, principally in cells treated with 3',4'-dihydroxyflavone. Staining was also shown for fibronectin in the extracellular medium, highlighted through treatments with apigenin and rutin, whilst in the cytoplasm staining was strongly distributed in all treatments with flavonoids. The discreet distribution of laminin and the staining for fibronectin observed in extracellular behavior can be associated with an anti-migratory effect exercised by the flavonoids. The cytoplasmic staining observed for laminin and fibronectin is suggestive of regulation on the part of the flavonoids within the mechanism of migration and adhesion of tumor cells. Glioma cells express fibronectin and laminin, these 22

ACCEPTED MANUSCRIPT molecules being liberated in the ECM; however, more invasive gliomas, as in the case of the GL-15 lineage, express less fibronectin when compared to gliomas which have a lesser degree of invasiveness (Planchenault et al., 2001). The presence of elevated levels of fibronectin in the ECM of less invasive gliomas has been associated with cell-cell interactions between tumor and host

RI PT

cell, contributing to cell proliferation and migration. In spite of laminin forming the basal membrane and contributing to cellular anchorage, principally through the formation of hemidesmossomas, the specific linkage site for laminin with integrin molecules and the process of laminin chains by specific enzymes such

SC

as MMP2, MT1MMP, plasmin and BMP1 are able to unchain other cellular responses, such as the inducement of tumor migration and invasion (Tzu et al., 2008). Studies have shown, for example, that the processing of the α5 chain of

M AN U

laminin 511 is associated with a decrease in aggregation and rise in the migratory potential of prostate cancer (Bair et al., 2005). In another study, it was observed that the processing of the β3 chain of laminin 332 facilitated the migration of colon and prostate cancer cells (Remy et al., 2006). In this way, even having observed an increase in the cytoplasmatic expression of fibronectin and laminin in a highly invasive cellular lineage (GL-15), it is still important to

TE D

evaluate which cellular mechanisms of signaling and regulation would be involved in these ECM proteins in GL-15 cells. The analysis by transmission electron microscope showed the cells not

EP

treated with flavonoids displayed their RER and Golgi apparatus preserved, heterogeneity in nuclear morphology with evident nucleolus, filopodia type structures, with tumefied and electrolucent mitochondria. On the other hand,

AC C

cells treated with polihydroxylated flavonoids showed evidence of ultra structural alterations, indicative of a possible apoptotic effect caused by the flavonoids studied in this work. Amongst these alterations we observed an expressive dilation in the RER and Golgi complex, retracted nuclei and more electro

dense

mitochondria

of

varying

sizes

with

some

exhibiting

disarrangement in the structure of their crests and the presence of myelin figures. In glioma cells of A-172 lineage it was observed the presence of irregular morphology with voluminous and irregular nuclei, microvillusprojections on the surface of the cell, with preserved mitochondria and RER (Arismendi23

ACCEPTED MANUSCRIPT Morillo, 2010). On the other hand, these same authors observed that some glioma cells being treated with flavopiridol (a synthetic flavone) showed condensation of chromatinic material with collapsed and fragmented nuclei; vacuolization in the cytoplasm as possible dilations of the endoplasmic reticle; and mitochondria without visible ultra structural alterations as possible apoptotic

RI PT

activity induced by flavone. In ultra structural analyses the presence of intumesced mitochondria with partial or total disruptions in crests and cristolysis are common findings in glioma cells. In our work we observed the presence of spherical, intumesced and electro lucent mitochondria in GL-15 cells not treated

SC

with flavonoids, however cristolysis was not evident. On the other hand, cells treated with flavonoids showed the presence of electro dense mitochondria, of various sizes and some presenting distorted crests with the aspect of cristolysis.

M AN U

Once the enzymes involved in the oxidative phosphorylation are localized in the internal mitochondrial membrane, the surface area of the mitochondria and the numbers of crests are generally associated to the level of metabolic activity exhibited by the cell (Modica-Napolitano et al., 2002). Hence alterations involving mitochondrial size as well as the integrity of the crests will possibly have repercussions, altering cellular metabolism. The presence of integral and

TE D

electro dense mitochondria can produce energy by oxidative phosphorylation, whilst electro lucent mitochondrias which display partially or totally distorted crests and cristolysis are incapable of generating energy by oxidative

EP

phosphorylation (Arismendi-Morillo, 2010). In our findings, we observed that in treatments with quercetin and kaempferol the mitochondrias show a more electro dense aspect. MTT data for these same flavonoids showed elevated

AC C

percentages when compared to other flavonoids. These data allow us to suggest that there is a possible effect by quercetin and kaempferol in the metabolic activity of cells of the GL-15 lineage, even though these very same substances impair the viability of the lineage. The presence of concentric lamellas, similar to myelin figures and structures in the periphery of the plasma membrane similar to blebs were also visualized in cells treated with these and other flavonoids. According to Adrain and Martin (2001), the presence of blebs in the plasma membrane, cellular retraction, compaction of chromatin and the formation of apoptotic bodies are phenotypic characteristics of cells that die 24

ACCEPTED MANUSCRIPT through apoptosis under the regime of cysteine proteases (caspases). In ultra structural study done with glioma cells also related that the intense cytoplasmic vacuolization, the disintegration of the plasmatic membrane and the presence of myelin figures are ultra structural indicators suggestive of the death of cells (Pavon et al., 2010).

RI PT

The tumor microenvironment possesses innumerable molecules from the extracellular matrix in its constitution. In order to migrate, glioma cells need to modify their form and interact with the micro environment. The activity of actin filaments in the plasma membrane generates structures known as lamelipodias

SC

and digitate protrusions known as filopodias which are considered essential processes in cellular migration, principally in tumors. Filopodias function like sensors in the external micro environment and as extensions of plasma

M AN U

membrane which will go on to realize initial contact with the extracellular matrix (Le Clainche&Carlier, 2008). In our own research we observed through scanning electron microscopy that polyhydroxilated flavonoids notably reduced the number of filopodias on the surface of the cytoplasmic extensions emitted by GL-15 cells. In accordance with these results, we also observed by investigation of migration an evident retardation in the migration of these cells

TE D

during the first 12 hours of treatment. These data lead us to suggest an important anti-migratory and anti-invasive effect in these flavonoids with regard to glioma cells of the GL-15 lineage. This is an important finding in relation to

EP

the flavonoids studied taking into consideration that the invasive potential of gliomas is one of the characteristics which highlight the difficulty of using conventional treatments. Some research has shown in vitro that the reduction in

AC C

the number of filopodias and alterations in lamelipodias in invasive glioma cells are associated with the lessening of the migratory and invasive capacity of the tumor (Lin et al., 2008; Nakada et al., 1999). In a study with human hepatoma cells, for example, it was shown that the potential of the flavonoid luteolin (3’,4’,5’,7’-tetrahydroxyflavone) inhibited the formation of lamellipodia and filopodia, with a consequent reduction in the migratory and invasive potential of cells (Lee et al., 2006). In our study we observed that amongst the flavonoids studied, 3’,4’-dihydroxyflavone and 3,4',5,7-tetrahidroxiflavone (kaempferol) showed an expressive reduction in filopodias on the cellular surface. This leads 25

ACCEPTED MANUSCRIPT us to suggest, once again, a possible association between the hydroxylate groups in carbons C3’ and C4’ and the antitumor effect of polihydroxylate flavonoids. However, more studies are necessary to characterize in a more specific way the molecular mechanisms involved in this process. To conclude our results show that polihydroxylate flavonoids present the

RI PT

capacity to induce differentiation and inhibit the migratory and invasive activity in human glioblastoma cell cultures, as well as their inducing ultra structural modifications indicative of death by apoptosis. These results suggest that these phenolic composites can be considered positive candidates in aid of treatment

SC

of malignant gliomas. However, more studies will be necessary to better understand the action of these flavonoids in the mechanisms which have a role

Conflict of interest statement

Acknowledgements

TE D

There is no conflict of interest.

M AN U

in the invasiveness of gliomas to determine specific pharmacological targets.

We gratefully acknowledge the research support provided by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação de

EP

Aparo à pesquisa do Estado da Bahia (FAPESB), Fundação Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), and by Programa de Pós-gradução em Ciência Animal nos Trópicos – Universidade Federal da

AC C

Bahia. We also gratefully acknowledge the Plataforma de Microscopia Eletrônica – Centro de Pesquisa Gonçalo Muniz, Fundação Oswaldo Cruz, Salvador, Brazil for ultramiscrocopy analysis, and Prof. Guy G. Chabot from Laboratoire de Pharmacologie Chimique et Génétique, Université Paris Descartes, by providing some flavonoids adopted in this study.

References 26

ACCEPTED MANUSCRIPT 1. Adrain, C. and S. J. Martin. The mitochondrial apoptosome: a killer unleashed by the cytochrome seas. Trends in Biochemical Sciences 2001 26(6): 390-397. 2. Arismendi-Morillo G. Electron microscopy morphology of the mitochondrial in

gliomas

and

their

vascular

microenvironment.

RI PT

network

BiochimBiophysActa. 2010 Nov 9.

3. Bair EL, Chen ML, McDaniel K, Sekiguchi K, Cress AE, Nagle RB, et al. Membrane type 1 matrix metalloprotease cleaves laminin-10 and promotes

SC

prostate cancer cell migration. Neoplasia. 2005 Apr;7(4):380-9.

4. Bocchini, V., T. Beccari, et al. Glial fibrillary acidic protein and its encoding mRNA exhibit mosaic expression in a glioblastoma multiform cell line of

M AN U

clonal origin. International Journal of Developmental Neuroscience 1993 11(4): 485-492.

5. Braganhol E, Zamin LL, Canedo AD, Horn F, Tamajusuku AS, Wink MR, et al. Antiproliferative effect of quercetin in the human U138MG glioma cell line. Anticancer Drugs. 2006 Jul;17(6):663-71.

6. Cardenas M, Marder M, Blank VC, Roguin LP. Antitumor activity of some

TE D

natural flavonoids and synthetic derivatives on various human and murine cancer cell lines. Bioorg Med Chem. 2006 May 1;14(9):2966-71. 7. Chintala SK, Gokaslan ZL, Go Y, Sawaya R, Nicolson GL, Rao JS. Role of

EP

extracellular matrix proteins in regulation of human glioma cell invasion in vitro. ClinExp Metastasis. 1996 Sep;14(4):358-66. 8. Demuth T and Berens ME. Molecular mechanisms of glioma cell migration

AC C

and invasion. J Neurooncol 2004 70 (2), 217–228.

9. Harborne JB. Nature, distribution and function of plant flavonoids. Prog Clin Biol Res 1986; 213: 15-24

10. Havsteen BH. The biochemistry and medical significance of the flavonoids. Pharmacol Ther. 2002 Nov-Dec;96(2-3):67-202. 11. Hwang JH, Smith CA, Salhia B, Rutka JT. The role of fascin in the migration and invasiveness of malignant glioma cells. Neoplasia. 2008 Feb;10(2):14959. 27

ACCEPTED MANUSCRIPT 12. Jeong JC, Kim MS, Kim TH, Kim YK. Kaempferol induces cell death through ERK and Akt-dependent down-regulation of XIAP and survivin in human glioma cells. Neurochem Res. 2009 May;34(5):991-1001. 13. Le Clainche C, Carlier MF. Regulation of actin assembly associated with protrusion and adhesion in cell migration. Physiol Rev. 2008 Apr;88(2):489-

RI PT

513. 14. Lee WJ, Wu LF, Chen WK, Wang CJ, Tseng TH. Inhibitory effect of luteolin on hepatocyte growth factor/scatter factor-induced HepG2 cell invasion involving both MAPK/ERKs and PI3K-Akt pathways. ChemBiol Interact.

SC

2006 Mar 25;160(2):123-33.

15. Lin, C.-W., W.-C. Hou, et al. Quercetin inhibition of tumor invasion via suppressing

PKCI/ERK/AP-1-dependent

matrix

metalloproteinase-9

M AN U

activation in breast carcinoma cells. Carcinogenesis 2008 29(9): 1807-1815. 16. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, et al. The 2007 WHO classification of tumours of the central nervous system. Acta N europathol. 2007 Aug;114(2):97-109.

TE D

17. Lowry, O.H., Rosebrough, NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951 Nov;193(1):265-75. 18. Modica-Napolitano JS and Singh K. Mitochondria as targets for detection and treatment of cancer. Expert RevnMol Med 2002 4(09): 1-19.

EP

19. Nakada M, Nakamura H., et al. Expression and tissue localization of membrane-type 1, 2, and 3 matrix metalloproteinases in human astrocytic

AC C

tumors. Am J Pathol. 1999 154(2): 417-28.

20. Nguyen TT, Tran E, Nguyen TH, Do PT, Huynh TH, Huynh H. The role of activated MEK-ERK pathway in quercetin-induced growth inhibition and apoptosis in A549 lung cancer cells. Carcinogenesis. 2004 May;25(5):647-

59.

21. Patru C, Romao L, Varlet P, Coulombel L, Raponi E, Cadusseau J et al. CD133, CD15/SSEA-1, CD34 or side populations do not resume tumorinitiating properties of long-term cultured cancer stem cells from human malignant glio-neuronal tumors. BMC Cancer. 2010 Feb 24;10:66. 28

ACCEPTED MANUSCRIPT 22. Pavon LF, Marti LC, Sibov TT, Malheiros SM, Oliveira DM, Guilhen DD, et al. The ultrastructural study of tumorigenic cells using nanobiomarkers. Cancer BiotherRadiopharm. 2010 Jun;25(3):289-98. 23. Planchenault, T., S. Costa, et al. Differential expression of laminin and fibronectin and of their related metalloproteinases in human glioma cell lines:

RI PT

relation to invasion. Neurosci Lett 2001 299(1-2): 140-4. 24. Remy L, Trespeuch C, Bachy S, Scoazec JY, Rousselle P. Matrilysin 1 influences colon carcinoma cell migration by cleavage of the laminin-5 beta3 chain. Cancer Res. 2006 Dec 1;66(23):11228-37.

SC

25. Santos, BL, A. R. Silva, et al. Antiproliferative, proapoptotic and morphogenic effects of the flavonoid rutin on human glioblastoma cells. Food Chemistry 2011 127(2): 404-411.

M AN U

26. Scheck AC, Perry K, Hank NC, Clark WD. Anticancer activity of extracts derived from the mature roots of Scutellariabaicalensis on human malignant brain tumor cells. BMC Complement Altern Med. 2006;6:2. 27. Shen KH, Hung SH, Yin LT, Huang CS, Chao CH, Liu CL, et al. Acacetin, a flavonoid, inhibits the invasion and migration of human prostate cancer DU145 cells via inactivation of the p38 MAPK signaling pathway. Mol Cell

TE D

Biochem. 2010 Jan;333(1-2):279-91. 28. Singh RD, Haridas N, Patel JB, Shah FD, Shukla SN, Shah PM, et al. Matrix metalloproteinases and their inhibitors: correlation with invasion and

EP

metastasis in oral cancer. Indian J Clin Biochem. 2010 25 (3):250-9. 29. Stupp R, Hegi ME, Gilbert MR, and Chakravarti A. Chemoradiotherapy in

AC C

malignant glioma: standard of care and future directions. J Clin Oncol. 2007 25 (26), 4127-4136.

30. Stylli SS, Kaye AH, Lock P. Invadopodia: at the cutting edge of tumour invasion. J ClinNeurosci. 2008 Jul;15(7):725-37.

31. Tzu J, Marinkovich MP. Bridging structure with function: structural, regulatory, and developmental role of laminins. Int J Biochem Cell Biol. 2008;40(2):199-214. 32. Udayakumar TS, Chen ML, Bair EL, Von Bredow DC, Cress AE, Nagle RB, et al. Membrane type-1-matrix metalloproteinase expressed by prostate 29

ACCEPTED MANUSCRIPT carcinoma cells cleaves human laminin-5 beta3 chain and induces cell migration. Cancer Res. 2003 May 1;63(9):2292-9. 33. Vincent L, Chen W, Hong L, Mirshahi F, Mishal Z, Mirshahi-Khorassani T, et al. Inhibition of endothelial cell migration by cerivastatin, an HMG-CoA reductase inhibitor: contribution to its anti-angiogenic effect. FEBS Lett.

RI PT

2001 Apr 27;495(3):159-66. 34. Wang L, Ling Y, Chen Y, Li CL, Feng F, You QD, et al. Flavonoid baicalein suppresses adhesion, migration and invasion of MDA-MB-231 human breast cancer cells. Cancer Lett. 2010 Nov 1;297(1):42-8.

SC

35. Yang K, Lamprecht SA, Liu Y, Shinozaki H, Fan K, Leung D, et al. Chemoprevention studies of the flavonoids quercetin and rutin in normal and azoxymethane-treated mouse colon. Carcinogenesis. 2000 Sep;21(9):1655-

AC C

EP

TE D

M AN U

60.

30

ACCEPTED MANUSCRIPT Figure Legends

Figure 1- Analysis of proliferation, cytotoxicity and cell viability in GL-15

RI PT

cells. GL-15 cells were treated with the vehicle of drug dilution (0.1% DMSO) or flavonoids F2 (3',4'-dihydroxyflavone), F11 (5,7-dihydroxyflavone), F32 (4',5,7trihydroxyflavona),

F12

pentahydroxyflavona)

(3,4',5,7-tetrahydroxyflavone); or

rutin

(3-ramnoglicoside

F7

(3,3',4',5,6-

of

3,3',4',5,6-

SC

pentahydroxyflavone) at a concentration of 50 or 100 µM. The cell viability was assessed after 48h of treatment. A, b) analysis of the cytotoxicity of flavonoids by the MTT assay. Results are expressed as the mean percentage ±SD relative

M AN U

to the control group, which was considered as 100%. c) proportion of viable and non-viable cells after exposure at polyhydroxylated flavonoids by Trypan blue staining. Results are expressed as the mean number of viable and non-viable cells. (*) Statistically different, significance p<0.05. d) number of BrdU positive cells after 48h of treatment. e) ultrastructural analysis of GL-15 cells by using transmission electron microscopy. Control condition - Cells in the control

TE D

condition (DMSO 0.1%). A: cells with the nucleus along with dispersed chromatin and large nucleolus and the central nucleus (white arrow). B: presence of filopodia structure type on the membrane surface(black arrow) and

EP

RER with aspect preserved (white arrowheads). It is also noted the presence of mitochondria with small spherical appearance. C: Golgi apparatus with aspect preserved (white arrow). D: RER preserved(white arrowhead) and short

AC C

spherical and electrolucente mitochondria. Treated cells-cell line GL-15 treated with flavonoids. A: cells with core appearance retracted (thin white arrow) with dilated RER (white thick arrow). It is also noted the presence of cisterns of Golgi apparatus very enlarged (white arrowheads). B: Magnified image of the Golgi complex, showed dilatation of the cisterns (white arrowhead). C: dilated RER (white thick arrow) and presence of myelin figures (dark arrows). D: Mitochondria with varied morphology and size, and showing more electrodense and aspect of cristolysis (dashed white arrows). 31

ACCEPTED MANUSCRIPT Figure 2 – Invasion evaluation by scratch assay and scanning electron microscopy of GL-15 cells. GL-15 cells were treated with the vehicle of drug dilution (0.1% DMSO) or flavonoids F2 (3',4'-dihydroxyflavone), F11 (5,7dihydroxyflavone),

F32 F7

F12

(3,3',4',5,6-pentahydroxyflavona)

or

(3,4',5,7rutin

(3-

RI PT

tetrahydroxyflavone);

(4',5,7-trihydroxyflavona),

ramnoglicoside of 3,3',4',5,6-pentahydroxyflavone) at a concentration of 50 µM. a) Scratch assay of GL-15 cells. Phase-contrast microscope images were shown (100×). Indicated times (0, 12 and 48h) depict the duration of treatment

SC

after scratching of the cell monolayer. b) Scanning electron microscopy of cell line GL-15. Cells in control conditions (Fig. A, B and C). Treated cells (Fig. D, E and F). A: cells with morphological heterogeneity (small arrow), with emission of

M AN U

long cytoplasmic processes with filopodias present (thin white arrow). B: cells with appearance to spindle filopodias distributed throughout the extension of the prolongation (white arrow). C: Magnified image of FigB high lighting filopodias (white arrow) in the end of the extension cytoplasmic. D: cells with a fusiform appearance (thin arrow), and spherical cells (arrowhead). E: cells with a fusiform appearance, showing extensive cytoplasmic projections, with few

TE D

filopodias (thin arrow) and others with filopodias evident (arrowhead). F: bipolar

EP

cell, with no evidence of filopodias.

Figure 3 – ECM expression in glioblastoma GL-15 cell lines. GL-15 cells were treated with DMSO at 0.1% or flavonoids at 50 µM. F2 (3’,4’-

AC C

dihydroxyflavone), F11 (5,7-dihydroxyflavone), F32 (4’,5,7-trihydroxyflavona), F12 (3,4’,5,7-tetrahydroxyflavone); F7 (3,3’,4’,5,6-pentahydroxyflavona) or rutin (3-ramnoglicoside of 3,3’,4’,5,6-pentahydroxyflavone). Tests after 48h of treatment. Pos-fixed – fixed cells after treatment. Pre-fixed – fixed cells before treatment. a) Immunohistochemical staining for ECM protein fibronectin. Red marking, fibronectin; blue marking nucleus (DAPI-5µg/ml). 20x0.70objective. b) Immunohistochemical staining for ECM protein laminin. Red marking, laminin; blue marking nucleus (DAPI-5µg/ml). Intensity fluorescence analysis of expression of fibronectin and laminin are indicated in the left side of the figure. 32

ACCEPTED MANUSCRIPT 20x0.70objective. c) Expression of Fibronectin (230kD) and laminin (200kD) by western blot. The sample of protein (30µg) was subjected to SDS-PAGE (7.5%), followed by immunoblotting. Densitometric analysis of expression of the bands of fibronectin and laminin. Molecular mass marker in kD is indicated in

RI PT

the left side of the figure.

Figure 4 – Expression and activity of MMP. GL-15 cells were treated with DMSO at 0.1% or flavonoids at 50 µM. F2 (3',4'-dihydroxyflavone), F11 (5,7F32

tetrahydroxyflavone);

F7

(4',5,7-trihydroxyflavona),

F12

SC

dihydroxyflavone),

(3,3',4',5,6-pentahydroxyflavona)

or

(3,4',5,7-

rutin

(3-

ramnoglicoside of 3,3',4',5,6-pentahydroxyflavone). a) immunohistochemical

M AN U

staining for MMP-2 proteins in pre-fixed GL-15 cells, after 48h of treatment. Green marking, MMP-2; blue marking nucleus (DAPI -5µg/ml) and red marking citoplasmic. Intensity fluorescence analysis of expression of MMP is indicated in the left side of the figure. 20x0.70objective. b) proportion of MMP-2 positive cells after exposure at polyhydroxylated flavonoids. Results are expressed as the mean number of MMP-2 positive cells. (*) Statistically different, significance

TE D

p<0.05. c) Expression of active MMP-2 (63kD) proteins after 48h exposure to control and flavonoids was determined by western blot. The sample of protein (30µg) was subjected to SDS-PAGE (7.5%), followed by immunoblotting.

EP

Densitometric analysis of expression of the bands of MMP-2 was determined. Molecular mass marker in kD is indicated in the left side of the figure. d) MMP activities in conditioned medium of confluent monolayer cultures of GL 15 cell

AC C

lines analyzed by 1% gelatin zymography assay in control conditions (DMSO 0.5% ) and after 24h treatment with flavonoids. Densitometric analysis of bands from enzymatic degradation was determined. Molecular weight marker is indicated in kD on the left side of figure.

Figure 5 – Immunostaining for GFAP protein in cell line GL-15. GL-15 cells were treated with DMSO at 0.1% or flavonoids at 50 µM for 48 h. GFAP: glial fibrillary acidic protein (GFAP) immunostaining for GL-15 cells. Fixation before 33

ACCEPTED MANUSCRIPT marking. Cells have intracellular and extracellular marking. Red: cytoplasm; green: GFAP and blue: nucleus. DMSO: dimethylsulfoxide 0.1%; F2 (3',4'dihydroxyflavone), F11 (5,7-dihydroxyflavone), F32 (4',5,7-trihydroxyflavona), F12 (3,4',5,7-tetrahydroxyflavone); F7 (3,3',4',5,6-pentahydroxyflavona) or rutin (3-ramnoglicoside of 3,3',4',5,6-pentahydroxyflavone). 20x0.70 objective. The

RI PT

proportion of GFAP positive cells were determined after exposure at polyhydroxylated flavonoids. Results are expressed as the mean number of

AC C

EP

TE D

M AN U

SC

GFAP positive cells. Significance p<0.05.

34

ACCEPTED MANUSCRIPT Table 1- Flavonoids tested. Structural models from PubChem Compound NCBI

Flavonoide

Estructure

RI PT

F11 - Chrysin 5,7-dihydroxyflavone

TE D

F12 - Kaempferol 3,4',5,7-tetrahydroxyflavone

M AN U

F32 - Apigenine 4',5,7-tryhydroxiflavone

SC

F2 3',4'-dihydroxyflavone

AC C

EP

F7 - Quercetin 3,3',4',5,6-pentahydroxyflavone

Rutin 3,3′,4′,5,7-pentahydroxyflavone-3-rutinoside

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

e)

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT *Research Highlights 

Flavonoids reduced proliferation, viability, invasion and filopodias of GL-15 glioblastoma cells. Flavonoides induced a decrease in levels of MMPs expression and activity



Flavonoids regulated expression of MEC components fibronectin and laminin



Glioblastoma cells exposed to flavonoids presented mitochondrial damage, vacuolation and entered in apoptosis.

EP

TE D

M AN U

SC

Flavonoids induced astroglial differentiation in remaining glioblastoma cells.

AC C



RI PT

