Content uploaded by Albino Martins
Author content
All content in this area was uploaded by Albino Martins on Jan 26, 2017
Content may be subject to copyright.
Macromolecular
Bioscience
Full Paper
wileyonlinelibrary.com
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (1 of 13) 1600340
DOI: 10.1002/mabi.201600340
There is an urgent need for antitumor bioactive agents with minimal or no side effects over
normal adjacent cells. Fucoidan is a marine-origin polymer with known antitumor activity.
However, there are still some concerns about its application due to the inconsistent experi-
mental results, specifically its toxicity over normal cells and the mechanism behind its action.
Herein, three fucoidan extracts (FEs) have been tested over normal and breast cancer cell
lines. From cytotoxicity results, only one of the extracts shows selective antitumor behavior
(at 0.2 mg mL−1), despite similarities in sulfation degree and carbohydrates composition.
Although the three FEs present different molecular weights,
depolymerization of selected samples discarded Mw as the
key factor in the antitumor activity. Significant differences in
sulfates position and branching are observed, presenting FE 2
the higher branching degree. Based on all these experimental
data, it is believed that these last two properties are the ones
that influence the cytotoxic effects of fucoidan extracts.
The Key Role of Sulfation and Branching
on Fucoidan Antitumor Activity
Catarina Oliveira, Andreia S. Ferreira, Ramon Novoa-Carballal,
Cláudia Nunes, Iva Pashkuleva, Nuno M. Neves, Manuel A. Coimbra,
Rui L. Reis, Albino Martins, Tiago H. Silva*
C. Oliveira, Dr. R. Novoa-Carballal, Dr. I. Pashkuleva,
Prof. N. M. Neves, Prof. R. L. Reis, Dr. A. Martins, Dr. T. H. Silva
3B’s Research Group–Biomaterials
Biodegradables and Biomimetics
University of Minho
Headquarters of the European Institute of Excellence
on Tissue Engineering and Regenerative Medicine
AvePark, Parque de Ciência e Tecnologia
4805-017 Barco, Guimarães, Portugal
E-mail: tiago.silva@dep.uminho.pt
C. Oliveira, Dr. R. Novoa-Carballal, Dr. I. Pashkuleva,
Prof. N. M. Neves, Prof. R. L. Reis, Dr. A. Martins, Dr. T. H. Silva
ICVS/3B’s
PT Government Associate Laboratory
Braga/Guimarães, Portugal
A. S. Ferreira, Dr. C. Nunes, Prof. M. A. Coimbra
QOPNA, Department of Chemistry
University of Aveiro
Campus de Santiago 3810-193, Aveiro, Portugal
Dr. C. Nunes
CICECO, Department of Chemistry
University of Aveiro
Campus de Santiago 3810-193, Aveiro, Portugal
1. Introduction
Cancer can have various origin-factors, being characterized
by uncontrolled cell growth and spread.[1] Cancer thera-
peutics aim to increase the survival time and the quality-
of-life of the patients. The goal of cancer treatment is the
extinction of tumor cells, ideally with minimal damage to
healthy tissues. The toxic effects over normal cells often
limit the current chemotherapeutic agents used in cancer
treatments, which conduct to reduced dosage and, conse-
quently, efficacy of the treatment.[2,3]
Breast cancer is the second most frequent diag-
nosed cancer and the first among females.[4] If breast
cancer is detected in an early stage, there is a possi-
bility to be treated and removed surgically. However,
the treatment of the advanced stage breast cancer is
often followed by reoccurrence and can become fatal,
even when chemotherapeutic agents are administered.
There are several factors such as tumors heterogeneity,
drug resistance, side effects, and toxicity to healthy tis-
sues, that diminish the efficacy and usefulness of this
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
C. Oliveira et al.
Macromolecular
Bioscience
www.mbs-journal.de
1600340 (2 of 13) © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com
treatment.[5,6] Therefore, there is a strong interest in
developing better-tolerated anticancer agents and treat-
ment modalities.[7]
Because of the side effects of many current treatments,
the use of natural substances with low toxicity is of
great interest.[8] Natural-derived polymers are of special
interest due to their biological and chemical similari-
ties to natural tissues composition. Marine organism are
valuable sources of materials with intriguing properties
and characteristics.[9] Among marine-origin materials,
fucoidan is a polysaccharide that consists of sulfated
fucose residues and is possible to be extracted from
brown seaweeds of different species such as Fucus vesicu-
losus, Laminaria japonica, and Undaria pinnatifida.[10–12]
Fucoidan has been reported to have various biological
activities including antibacteria, antioxidant, antiviral,
and antitumor.[13,14]
Fucoidan antitumor behavior has been demonstrated
in vitro and in vivo for different type of cancers such as
breast, lung, prostate, colon, and melanoma.[7,15–18] It
retards tumor development, eradicating tumor cells by
targeting key apoptotic molecules. Furthermore, the syn-
ergistic effect of fucoidan with current anticancer agents
has also been reported.[8,19] Despite the promising results
about its anticancer effects, the fucoidan mechanism
of action is not clearly understood and thus, fucoidan
has not yet been developed as a regulated therapeutic
agent.[20,21] The main barriers to its utilization in clinic
are: (i) its toxicity to normal cells, and (ii) the variable and
contradictory results in fucoidan usage.[8,22] In addition,
there are a huge variety of algae fucoidan sources, which
implies various extraction and purification methods that
may influence fucoidan’s intrinsic properties and its bio-
activity such as molecular weight, carbohydrates compo-
sition, and sulfation. Molecular weight has been reported
as one of the main factors influencing fucoidan anti-
tumor behavior. From previously reported works, lower
molecular weight fucoidan presents higher antitumor
effects whereas higher sulfation degree has been related
with enhanced bioactivity responses.[23,24] However, most
studies analyzing antitumor activity did not characterize
in detail the composition of the used fucoidan and this is
probably the cause of some contradictory findings.[10,14,25]
An effective cancer therapeutic strategy is character-
ized by the ability to eliminate tumors without damaging
healthy tissues. Due to the different properties affecting
fucoidan bioactivity, there is the need to evaluate the
antitumor effects of a certain extract and characterize
which feature(s) is (are) playing the major role. For that
purpose, herein, three different extracts from the same
species (i.e., F. vesiculosus) were tested with human breast
cancer cells, and normal endothelial and fibroblastic cells.
From these biological data we were able to observe three
completely different bioactive responses and, in that
sense, we decided to carry out structure–activity relation-
ship studies.
2. Results
2.1. Cytotoxicity of Fucoidan Extracts
2.1.1. Fucoidan Extract 1 (FE 1) Does Not Present
Toxic Effects over Cancer Cells and Shows Toxicity
for Normal Cells
FE 1 presents cytostatic effects at day 2 and day 3 for MDA-
MB-231 cell line as demonstrated by the inhibited cells
growth when compared to the control and lower concen-
trations (Figure 1A). Regarding MCF-7 cell line, fucoidan
presents toxic effects only at day 3 and at concentration of
5 mg mL−1 (Figure 1B). Cytotoxic assays for the normal cells
showed cytotoxic effect at day 2 for concentrations above
2 mg mL−1 (Figure 1C,D).
2.1.2. FE2 Induces Cancer Cells Death but Does Not Affect
the Viability of Normal Cells at 0.2 mg mL−1
We began to test FE2 at the same concentrations as the FE1.
However, a significant effect of fucoidan over the breast
cancer cell lines was revealed at low concentrations: at
day 2 and at concentration of 0.2 mg mL−1, the fucoidan
induces cell death for both MDA-MB-231 and MCF-7 cell
lines (Figure 2A,B). This effect become more pronounced at
higher fucoidan concentrations. The two types of normal
cells show distinct behavior: cytostatic effect for the
endothelial cells was observed at day 2 and at concentra-
tion of 0.5 mg mL−1 (Figure 2C), while, for fibroblastic cells
this effect was seen above concentration of 0.3 mg mL−1
(Figure 2D).
These results demonstrated that FE2 induces apoptosis
for the two types of tumor cells above 0.2 mg mL−1 and
maintains the viability of normal endothelial and fibro-
blastic cells at the same concentration.
2.1.3. FE3 Is Toxic to Both Cancer and Normal Cells
Similar to FE2, FE3 induces cancer cells death at day 1 for
concentrations above 0.2 mg mL−1 (Figure 3A,B). However,
fucoidan extract 3 (FE 3) affects also the normal cells at
this concentration and thus, its effect differs from FE2.
Stronger cytostatic effect was observed for endothelial
cells (Figure 3C): significantly diminished proliferation
was measured for the cells at 0.1 mg mL−1 concentration
and significant cytotoxic effect is observed above concen-
tration of 0.3 mg mL−1. Concerning the fibroblastic cells,
fucoidan has severe consequences over this cell type and
the cytotoxic effect is observed for concentration above
0.2 mg mL−1 (Figure 3D).
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
The Key Role of Sulfation and Branching on Fucoidan Antitumor Activity
Macromolecular
Bioscience
www.mbs-journal.de
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (3 of 13) 1600340
www.advancedsciencenews.com
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
Figure 1. Cytotoxicity of FE 1, at different concentrations and time-points (1, 2, and 3 days), over human cell lines A) MDA-MB-231, B) MCF-7,
C) HPMEC-ST1.6R, and D) MRC-5. Data were considered statistically different if p < 0.05. * indicates significant differences when compared
to + control; +, when compared to 0.1 mg mL−1; x, when compared to 0.5 mg mL−1; o, when compared to 1 mg mL−1; &, when compared to
2 mg mL−1, and $, when compared to 3 mg mL−1.
Figure 2. Cytotoxicity of FE 2, at different concentrations and time-points (1, 2, and 3 days), over human cell lines A) MDA-MB-231, B) MCF-7,
C) HPMEC-ST1.6R, and D) MRC-5. Data were considered statistically different if p < 0.05. * indicates significant differences when compared
to + control; +, when compared to 0.1 mg mL−1; x, when compared to 0.2 mg mL−1; o, when compared to 0.3 mg mL−1, and &, when compared
to 0.4 mg mL−1.
C. Oliveira et al.
Macromolecular
Bioscience
www.mbs-journal.de
1600340 (4 of 13) © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com
2.2. Physicochemical Characterization of the Different
Fucoidan Extracts
2.2.1. Molecular Weight
The gel permeation chromatography (GPC) analysis
showed that the three extracts have significantly different
molecular weight. A smallest retention volume (longer
chain) was determined for FE 1 (Figure 4A,B). The molec-
ular weight of FE 1 is two times higher when compared
with FE 3. All the three extracts have a very similar Mw/Mn
(polydispersity).
5-([4,6-Dichlorotriazin-2-yl]amino)fluorescein hydro-
chloride (DTAF)-Labeled Fucoidans’ Cellular Internaliza-
tion: An internalization assay was performed to assess
if cancer cells were able to internalize the FE based on its
molecular weight. In that sense fucoidan extract 2 (FE 2)
(the one with selective bioactivity) was compared with
FE 1 (no bioactive activity and higher molecular weight).
No green labeling is observed for the positive controls and
DTAF (Figure 5) showing that DTAF alone is not internal-
ized by breast cancer and endothelial cells. Endothelial
cells cultured with labeled FE 1 and FE 2 do not present
any green fluorescence, i.e., DTAF-labeled fucoidan was
not internalized by these cells. On the other hand, breast
cancer cells internalized the labeled fucoidan: green
staining localized near to the nucleus is visible in Figure 5
for both FE 1 and FE 2.
FE 1 Hydrolysis: The above results (Figures 4 and 5) did not
lead straightforward to any conclusions about the signifi-
cance of molecular weights on fucoidans bioactivity. How-
ever, since many authors related the Mw with the antitumor
activity we performed further analysis about its influence.
The FE 1 (the one with higher molecular weight) was hydro-
lyzed by an acidic reaction in boiling water to lower mole-
cular weight polymer and we further
analyzed it. We obtained a polymer with
molecular weight of 40 kDa (Figure 6A).
This molecular weight is similar to the
one of FE 3 that has toxic effect over both
cancer and normal cells. The toxicity of
this new polymer (FE 1*) was evaluated
with breast cancer and endothelial cells
and demonstrated no significant differ-
ences as compared with the initial FE1
(Figure 6C,D vs Figure 1B,C).
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
Figure 4. A) Molecular weight (Mw) and Mw/Mn of FE 1, FE 2, and FE 3 measured by GPC,
B) molecular weight chromatograms.
Figure 3. Cytotoxicity of FE 3, at different concentrations and time-points (1, 2, and 3 days), over human cell lines A) MDA-MB-231, B) MCF-7,
C) HPMEC-ST1.6R, and D) MRC-5. Data were considered statistically different if p < 0.05. * indicates significant differences when compared
to + control; +, when compared to 0.1 mg mL−1; x, when compared to 0.2 mg mL−1; o, when compared to 0.3 mg mL−1, and &, when compared
to 0.4 mg mL−1.
The Key Role of Sulfation and Branching on Fucoidan Antitumor Activity
Macromolecular
Bioscience
www.mbs-journal.de
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (5 of 13) 1600340
www.advancedsciencenews.com
2.2.2. Carbohydrates Composition
The obtained results for the monomeric composition of the
analyzed sulfated-polysaccharides (Table 1) showed that
the predominant sugar was the fucose (71.2%–79.1% mol),
as expected. The following sugars are the uronic acids
(9.8%–15.3% mol), xylose (3.9%–8% mol), galactose
(3.5%–5.5% mol), and minor amounts of mannose and
glucose. It seemed that there are not much differences
between the analyzed fucoidan extracts. When comparing
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
Figure 5. Internalization of DTAF-labeled FE 1 and FE 2 by HPMEC-ST-1-6R and MCF-7 cell lines. In MCF-7 cultures, the nucleus staining by
DAPI was omitted in face of the colocalization of internalized DATF-labeled fucoidans.
C. Oliveira et al.
Macromolecular
Bioscience
www.mbs-journal.de
1600340 (6 of 13) © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com
one of the previous referred extracts with FE 1 (no toxicity
over cancer cells) we often find comparable values, with
at least one of the extracts (FE 2 and FE 3). The total sugar
mass per extract mass is also very similar for the three
extracts. Having a closer look into the quantification of
neutral sugars and uronic acids, and having in mind that
FE 2 and FE 3 are the ones that present toxicity to cancer
cells, it is not possible to find any correlation between
these sugar components and the anticancer activity.
2.2.3. NMR Spectroscopy
The differences in the biological activity observed in FE
seems to be unrelated to the Mw and sugars composition
therefore we decided to analyze in detail the sulfation and
branching of the samples. The fastest way to compare the
three extracts and obtain a general idea of the differences
in composition is 1H NMR. The 1H NMR of the three sam-
ples at 400 MHz is included in Figure 7.
The spectra correspond to the previously described
for fucoidan samples with the isolated regions of the
fucose methyl protons (H6, 1–1.5 ppm), acetyl protons
(at 2.2 ppm with a small proportion 0.5 to 1 acetyl groups
per every 10 fucose residues) and anomeric (5.0–5.5 ppm)
and a highly overlapped region corresponding to all other
signals in the fucose 1H NMR[26] The strong overlap ham-
pers a very detailed information but the three extracts
show enough differences to expect considerable different
composition. Especially interesting was difference the
region around 3.9–4.3 ppm that contains signals attrib-
uted to branched fucose.[27] In light of these results a
detailed study of sulfation position and branching was
performed by methylation analysis.
2.2.4. Sulfation Degree and Methylation Analysis
Sulfation is another key factor in fucoidan bioactivity:
higher sulfation degree is often associated with greater
bioactivity.[28] Quantification results demonstrate that the
studied extracts have a very similar total percentage of
sulfates, which varies between 28.0% and 29.3%. The anal-
ysis of partially methylated alditol acetates before (native)
and after desulfation (desulfated) of fucoidans allow to
know the substitution of sugar residues, corresponding to
Table 1. Carbohydrates composition of the three fucoidan extracts.
% total sugar mass/
extract mass
% molar of neutral sugars % molar of uronic acids
Fucose Xylose Mannose Galactose Glucose
FE1 52.5 ± 2.5 73.1 ± 0.4 8.0 ± 0.1 1.3 ± 0.2 3.5 ± 0.3 0.7 ± 0.1 13.3 ± 0.3
FE2 52.2 ± 1.9 79.1 ± 0.6 3.9 ± 0.0 0.8 ± 0.0 5.5 ± 0.1 0.8 ± 0.0 9.8 ± 0.5
FE3 50.5 ± 2.7 71.2 ± 0.5 5.3 ± 0.1 1.5 ± 0.3 5.4 ± 0.0 1.3 ± 0.9 15.3 ± 0.8
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
Figure 6. A) Molecular weight (Mw) and Mw/Mn measured by GPC, B) molecular weight chromatogram. Cytotoxicity of FE 1* over C) MCF-7
and D) HPMEC-ST-1.6R cell lines. No statistical significant differences were observed.
The Key Role of Sulfation and Branching on Fucoidan Antitumor Activity
Macromolecular
Bioscience
www.mbs-journal.de
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (7 of 13) 1600340
www.advancedsciencenews.com
branching points and the substitution by sulfate esters.
When a position is acetylated in the native polysaccharide
and become methylated after desulfation, it is indication
that this position contains a sulfate residue. However,
when a position is acetylated even after desulfation of the
polysaccharide, it can be inferred that it is a branching
point. The results are shown in Table 2.
In native fucoidan extracts, 2,3,4-Fuc was the major
residue for the three extracts, followed by 2,4- and 2,3-Fuc
(coeluated peaks) residues, indicating a high percentage
of substituted fucose. After desulfation it was observed
the decrease of these specific residues (75%–80% for
2,3,4-Fuc and 46%–61% for 2,4-Fuc + 2,3-Fuc), meaning
that part of the substituents was sulfate esters that were
removed with the desulfation procedure. An increase
mainly of the 4-Fuc and 3-Fuc residues after desulfation
allowed to infer that the fucoidan backbone was con-
stituted by these linkages, which is in accordance with
literature, reporting that F. vesiculosus fucoidan main
chain is composed of alternating 4-O and 3-O fucose link-
ages.[29,30] These results allowed also to conclude that sul-
fate esters were mainly linked in 2-O fucose. However, the
observation of the 2-O linkage in the desulfated sample
shows that this linkage should be also a characteristic of
the fucoidan structure. The increase of 3,4-Fuc residues
after desulfation, as well as the maintenance of about
10% of 2,3,4-Fuc, 2,4-Fuc, and 2,3-Fuc branched residues,
corroborate the presence of a branched fucoidan structure
mainly in 3-O or 4-O positions. As the sum of all branched
residues is equal to the sum of terminal residues of Fuc,
Xyl, and UA (GlcA and GalA), it can be inferred that the
fucoidan had a polymeric nature (no extra Fuc terminal
residues), containing fucose, xylose, and uronic acids as
side chains terminal residues.
The branching degree (sum of branched Fuc/(total Fuc
minus the terminal Fuc), indicates the frequency (%) of
side chain residues present in the backbone. Therefore,
this structural feature of the polysaccharides was deter-
mined for FE 1, FE 2, and FE 3 and allowed to show that
FE 2 was more branched (83.4%) than FE 1 (67.5%) or FE 3
(60.4%) (Table 2).
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
Figure 7. 1H NMR spectra (400 MHz, D2O, 25 °C) of the fucoidan
extracts: FE1 bottom, FE 2 middle, and FE 3 top.
Table 2. Sulfation degree (%), glycosidic-substitution composition (%) of fucoidan samples in fractions FE 1, FE 2, and FE 3, before and after
desulfation.
Sulfation degree (% sulfate mass/extract mass) FE 1 FE 2 FE 3
Sugar residues and substitution positions 28.0 ± 1.4 29.3 ± 2.8 28.6 ± 1.9
Native Desulfated Native Desulfated Native Desulfated
t-Fuc 5.5 14.2 4.2 13.1 4.7 13.4
4-Fuc 0.9 10.6 1.9 7.6 1.5 11.0
3-Fuc 1.5 9.3 1.2 7.3 1.5 13.6
2-Fuc 5.1 8.8 3.9 5.2 4.1 5.7
3,4-Fuc 3.9 12.3 7.3 11.2 5.8 9.2
2,4-Fuc + 2,3-Fuc 19.1 10.3 24.8 10.6 27.6 10.7
2,3,4-Fuc 42.8 9.1 42.9 11.3 34.8 6.8
Total Fuc 78.8 74.6 86.3 66.3 79.9 70.4
t-Xyl 6.8 9.1 3.3 7.7 4.4 8.2
t-GlcA + t-GalA 14.5 16.3 10.4 26.0 15.6 21.4
Branching degree [%] 67.5 83.4 58.9
Sulphate [%] 4-O linkage 30.5 27.4 38.5
3-O linkage 34.9 29.4 27.7
2-O linkage 55.6 50.9 60.3
C. Oliveira et al.
Macromolecular
Bioscience
www.mbs-journal.de
1600340 (8 of 13) © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com
The percentage of sulfate present in each fucose posi-
tion (4-,3-, and 2-O linkage) was determined based on
the increase of each fucose linkage after desulfation in
relation to the native fucoidan (Table 2). The major per-
centage of sulfate esters was present in the 2-O-Fuc for
all extracts (51%–60%), which is in agreement with lit-
erature.[29,30] In addition, it is also possible to observe
the presence of sulfates in 3-O position (28%–35%) and
4-O position (27%–36%). The FE 2 showed the presence of
fewer sulfate esters in 4-O and 2-O positions when com-
pared with the other extracts, in opposite to FE 3 that had
a higher percentage of sulfates in 4-O and 2-O positions.
FE 1 had a higher content of sulfate in 3-O position.
3. Discussion
Fucoidan has been reported to inhibit the growth of
cancer cells, having a great interest in the development
of new cancer therapies.[24] When considering a cancer
therapy, the ultimate goal is to affect cancer cells
without damaging the surrounding healthy environ-
ment, having non or minimal side-effects.[31] The tumor
microenvironment comprises not only the cancer cells
but also the noncancerous cells which includes endothe-
lial cells, fibroblasts, and circulating immune cells.[32,33]
For this reason, it is of utmost importance to test
whether fucoidan can only damage the tumor cells and
not the healthy surrounding tissues. The three fucoidan
extracts were tested with endothelial, fibroblastic and
breast cancer cells. When designing a blood adminis-
tered drug, there is a need to assure that its circulation
will not affect the blood vessels. Therefore, we tested the
cytotoxicity of fucoidan over human endothelial cells
(HPMEC-ST-1.6R cell line). The healthy tissue around
the tumor should not be affected when the drug enters
into the interstitial fluid. Therefore, fucoidan should not
affect this type of cells as well, herein represented by
human fibroblasts (MRC-5 cell line). Breast cancer is one
of the most frequent cancers and, as many other types of
cancer, presents heterogeneous behavior. We used two
different breast cancer cells lines to depict this tumor
heterogeneity. MCF-7 cell line is an ideal cell model to
study hormone responsive tumors since it is estrogen
receptor (ER)-positive cells and forms tumors in the
presence of estrogen.[34,35] On the other hand, MDA-MB-
231 cell line is ER-negative and have been shown to be
tumorigenic.[34]
As previously stated, the desired effect of a cancer
therapeutic strategy is that fucoidan has toxicity over
cancer cell and no or diminished effects over noncancer
cells. Some publications only show the effects of fucoidan
over cancer cells.[19,36] However, this can lead to some
misinterpretation of fucoidans’ antitumor activity, since
it could also have toxic effects over normal cells. In the
present study, we denoted that not all fucoidan extracts
present this desired behavior. Specifically, FE 1, despite
the wide range of concentrations, showed toxicity to
normal cells at lower concentrations than cancer cells,
resulting in an extract without antitumor features and
not suitable for cancer therapies. On the other hand, both
FE 2 and FE 3 presented toxicity to cancer cells, although
FE 3 also presented toxicity to normal fibroblast cells at
the same concentrations. Therefore, FE 3 has to be used
with extreme careful and in a more target and precise
way to try to affect only the cancer cells and diminished
the toxic effects over noncancer cells, by not affecting the
surrounding environment.[37] The FE 2 is the one with
the desired antitumor behavior, since it showed toxicity
effects over cancer cells at 0.2 mg mL−1 and, at this con-
centration, neither endothelial nor fibroblastic cells were
affected. From this first group of results, we conclude
that not all fucoidans present desirable features, leading
to different toxic profiles when in contact with the same
cells.
After this biological screening, the next step was
focused on the physicochemical characterization of the
three extracts, trying to understand which feature(s)
play(s) a pivotal role in this antitumor behavior. As pre-
viously reported, the physicochemical factors that may
influence fucoidan bioactivity are the molecular weight,
monosaccharide composition, sulfates degree, and sul-
fates position. Nevertheless, despite not being directly
related with fucoidan intrinsic structure the source and
extraction method have also been described to influence
its bioactive behavior.[10,14,28,38]
The effect of molecular weight has been reported as
the crucial factor for the fucoidan bioactivity. Previous
in vitro studies showed that lower molecular weight
fucoidan significantly increased the anticancer activity of
fucoidan.[23,36,39] Furthermore, it has been described that
even with higher amounts of sulfation, molecular weight
plays a more decisive role.[24] Among the fucoidans
studied in the present work, FE 1 has the highest molec-
ular weight when compared with FE 2 and FE 3, but the
latter are the ones presenting toxicity to breast cancer
cells. Molecular weight is known to be associated with
cell internalization, among other factors, from which we
hypothesized that when comparing FE 2 with FE 1, only
FE 2 (lower molecular weight) would be internalized by
cancer cells and exert its antitumor action.[40,41] However,
as illustrated in Figure 5, no differences were observed in
the cellular internalization of FE 1 and FE 2. Thus, these
results do not give any conclusions regarding the relation-
ship between cellular internalization and the molecular
weight for the studied system.
It has been described that hydrolyzed fucoidan exhib-
ited a higher percentage of anticancer activity.[24,38] In
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
The Key Role of Sulfation and Branching on Fucoidan Antitumor Activity
Macromolecular
Bioscience
www.mbs-journal.de
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (9 of 13) 1600340
www.advancedsciencenews.com
particular, hydrolyzed fucoidan under mild conditions
(in boiling water with HCl) showed higher antitumor
activity whereas hydrolyzed fucoidan generated under
harsh conditions (microwave) slightly enhanced the
anticancer effects.[24] In this sense, FE 1 was hydrolyzed
by acidic reaction to assess if the resulting fucoidan,
FE 1*, with lower molecular weight, would present tox-
icity over cancer cells. However, as seen in Figure 6, this
lower molecular weight extract did not present tox-
icity to normal and cancer cells. It has been described
that acidic hydrolysis can reduce also sulfate content
and depolymerization could change monosaccharides
ratio, as, e.g., decreasing the level of uronic acids.[39,42]
Besides, by comparing FE 1* and FE 3 it is possible to
conclude that despite having similar molecular weight
the two extracts present significant differences in their
bioactivity response. These results suggest that the
antitumor behavior may be related with other factors
and not directly associated with fucoidans’ molecular
weight.
The studied fucoidan extracts have a total percentage
of carbohydrates between 50.5 and 52.5 and of sulfates
between 28.0 and 29.3, which means that the total
amount of polysaccharides present in the samples varies
between 78.5% and 81.8%. These results confirm that
fucose and sulfates are the main components of fucoidan
from F. vesiculosus, as described elsewhere.[25,43] The
uronic acids content has been determined in fucoidan
composition, although we need to be cautious in
assuming that the full uronic acid measured is contained
within the fucoidan polymers, as has been described else-
where.[44] The carbohydrates composition did not indicate
any clear structure-activity relationship (SAR).
Higher sulfation is related with greater molecular bio-
activity and thus researchers have produced over-sul-
fated fucoidans to enhance its biological properties.[29]
It has been suggested that over-sulfation causes higher
negative charge in the molecule, which facilitate forma-
tion of fucoidan-protein complexes involved in cell pro-
liferation.[31] Sulfate mass quantification of the three
fucoidan extracts did not present significant variation.
Despite these results, sulfates can still play a role in the
antitumor activity, since the same sulfation degree can
correspond to different sulfates position, i.e., sulfation
distribution along fucoidan backbone, presenting dis-
tinct biological response. Thus, both sulfates position and
polymer branching were assessed to better understand
the role of the chemical structure on the exhibited biolog-
ical activity. The FE 2 was the polysaccharide with higher
percentage of fully branched chains (together in 3-O and
4-O-Fuc) and, as a consequence, it is a more branched
structure than the other ones. Also, as all hydroxyl groups
are linked, they could not have sulfates in these residues.
The sulfates occur mainly in 2-O-Fuc. In comparison with
FE 2, FE 1 had higher sulfation in 4-O-Fuc and 3-O-Fuc
while FE 3 had higher sulfation in 4-O-Fuc. Consequently,
the sulfates position is different for the three fucoidans.
It is possible that these structural differences may influ-
ence their cytotoxicity response, particularly regarding
the breast cancer cell lines tested.
It is thus shown here that the antitumor behavior is
not ubiquitous, as could be inferred from the published
reports, but dependent on their chemical structure,
being also important to highlight that the same fucoidan
extract can present different effects over different types
of cells and cancers.[15]
As referred before, the source (original species) and
extraction method are two factors that may affect fucoidan
intrinsic properties and that are commonly associated
with fucoidans’ bioactivity. Fucoidan preparations isolated
from different sources have shown differential anticancer
effects in vivo due to corresponding structural proper-
ties.[22] Since the source of fucoidan is the same for the
three extracts (i.e., F. vesiculosus), the extraction method
may be the factor influencing fucoidan intrinsic properties
and thus the different antitumor behavior. Even from the
same company, we have different batches that can pre-
sent different behaviors. The preservation of the fucoidan
molecules’ structural integrity essentially depends on the
extraction methodology which has a crucial, but partly
ignored, significance for obtaining the relevant structural
features required for specific biological activities and for
elucidating structure–function relations.[14]
To be suited to a regulated product, fucoidan extracts
must be defined and reproducible. Sustainable, clean,
and regulated harvesting, or culturing of a single type of
seaweed are required. In order to meet therapeutic regu-
latory requirements, common extraction methods and
distributions of fully characterized fucoidans need to be
taken into consideration. This is particularly relevant
when trying to take a product from a preclinical concept
to clinical trials. The development and use of such con-
sistent extraction procedures would also help in achieving
a better understanding of structure–activity relationship
of fucoidan extracts.
4. Conclusions
Despite the promising results about the anticancer
effects of fucoidan, some variability impedes its utiliza-
tion in the clinic. Specifically, contradictory experimental
results influenced by endogenous and exogenous factors
in fucoidan usage are the main barriers. It is also impor-
tant to have in mind that the ultimate goal of an effec-
tive cancer therapy is to damage cancer cells without
negatively affect the surrounding healthy environment.
This distinctive action mode was only observed with the
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
C. Oliveira et al.
Macromolecular
Bioscience
www.mbs-journal.de
1600340 (10 of 13) © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com
FE 2 being important to characterize the physicochemical
properties of the different extracts. The present results
allow to infer that the molecular weight, monosaccharides
composition, and content of sulfates could not be related
with the cytotoxicity of FE 1, FE 2, and FE 3 extracts. From
our results, we can conclude that the branching degree of
the fucoidans may be the most obvious cause of the dif-
ferent biological behavior of the three extracts.
In this sense, more focus should be direct to the opti-
mization and standardization of the extraction and puri-
fication processes to obtain consistent protocols that
account for the biodiversity of fucoidan extracts, from dif-
ferent seaweeds, and to retain the structural features of
significance for specific bioactivity of fucoidan extracts.
Furthermore, the determination and clarification of the
structural characteristics responsible for antitumor activi-
ties of fucoidan will be essential for its potential as a
marine-origin drug.
5. Experimental Section
5.1. Materials
FE 1 was purchased from Marinova, whereas Fucoidan Extract
2 (FE 2 – batch SLBC6348V) and Fucoidan Extract 3 (FE 3 – batch
081M7672V), were purchased from Sigma-Aldrich, all used as
received. The human vascular endothelial growth factor ELISA
kit was purchased from Peprotech (Rochy Hill, NJ, USA) and kept
at 4 °C until used. Phalloidin-Tetramethylrhodamine B isothiocy-
anate (Texas Red – Phalloidin), DTAF were both purchased from
Sigma. 4,6-Diamidino-2-phenyindole, dilactate (DAPI) was pur-
chased from Biotium (Hayward, CA, USA).
5.2. Biological Assays
5.2.1. Cell Expansion
Human fibroblasts (MRC-5 cell line) and pulmonary microvas-
cular endothelial cells (HPMEC-ST1.6R cell line) were used as
noncancer cells. Fibroblast cells were cultured in D-MEM low
glucose medium (Sigma-Aldrich) supplemented with 10% FBS
(Alfagene) and Pen/Strep (100 U/100 g mL−1; Life Technologies).
Fibroblasts were used at passages 14–16. Endothelial cells were
cultured in M199 medium (Sigma-Aldrich) supplemented with
20% FBS (Alfagene), 2 × 10−3 m Glutamax (Life Technologies),
Pen/Strep (100 U/100 g mL−1; Life Technologies), endothelial
cell growth supplement (ECGS – 25 µg mL−1; Becton Dickinson).
Endothelial cells were used at passages 38–40.
Human breast adenocarcinoma cells (MCF-7 and MDA-MB-231
cell lines) were used to assess the effect of different fucoidan’s
concentrations over these cancer cells models. Both cell lines
were cultured in D-MEM high glucose medium (Sigma-Aldrich)
supplemented with 10% Fetal Bovine Serum (FBS) (Alfagene),
Pen/Strep (100 U/100 g mL−1; Life Technologies) and 1% MEM
sodium pyruvate solution 100 × 10−3 m (Alfagene). The MCF-7 cell
line was used at passages 16–18, whereas the MDA-MB-231 cell
line was used at passages 42–44.
The four types of cells were incubated at 37 °C in a humidified
5% CO2 atmosphere. Media were exchanged every 2–3 days until
cells reached a 90% confluence.
5.2.2. Cell Culture
Noncancer and cancer cells were harvested and 15 000 cells were
cultured in 24 well-plates. The cells were left to adhere for 4 h and,
after that, fucoidan extracts were added to adherent cells. Fucoidan
extracts were dissolved in the culture medium at different concen-
trations: for FE 1 the concentrations tested were 0.1, 0.5, 1, 2, 3, 4,
and 5 mg mL−1, whereas for FE 2 and FE 3 the concentrations were
0.1, 0.2, 0.3, 0.4, and 0.5 mg mL−1. For all the assays a positive control
was performed (no fucoidan in the culture medium). Each experi-
mental condition was tested in triplicate and two independent
assays were performed for each type of cells and fucoidan extracts.
5.2.3. Cell Viability
The metabolic activity of noncancer and cancer cells, cultured
at different fucoidan extracts’ concentrations and time points,
was determined by the MTS assay (CellTiter 96 AQueous One
Solution, Promega). The MTS assay is a colorimetric method
commonly used for cytotoxicity assays or for determining the
number of viable cells in proliferation. Basically, the quantity
of formazan product is directly proportional to the number of
living cells in culture.[45] At days 1, 2, and 3, the culture medium
was removed and the testing conditions were rinsed with sterile
Phosphate-Buffered Saline (PBS). A mixture of culture medium
(without FBS and phenol red) and MTS reagent (5:1 ratio) was
added to each well and left to incubate for 3 h, at 37 °C, in a
humidified 5% CO2 atmosphere. Thereafter, the absorbance of
the MTS reaction medium from each sample was read in tripli-
cate at 490 nm in a microplate reader (Synergy HT, Bio-TEK). All
experiments were performed in triplicate.
5.2.4. DTAF-Labeled Fucoidan and Morphological
Observation
DTAF-labeled fucoidan was prepared as described elsewhere.[46,47]
Briefly, DTAF was reconstituted in methanol and kept at 4 °C
until further use. A 0.2 mg mL−1 solution of FE 1 and FE 2 was left
to stir with DTAF for 3 h (20 µg mL−1). The procedure described
in Sections 5.2.2 and 5.2.3 were followed, but was added DTAF-
labeled FE 1 and FE 2 instead of the pure FE. At days 1, 2, and 3,
samples were washed with PBS, fixed for 30 min with 10% for-
malin, washed again with PBS and kept in PBS at 4 °C.
To evaluate whether and where fucoidan has been internalized
by the different cell types, the nucleus and cytoskeleton of
those cells were also fluorescent labeled. First, a blocking step
was performed with 3% Bovine Serum Albumin (BSA) in PBS
for 30 min. Then, the BSA solution was removed and cells were
washed with PBS. After that, Phalloidin-Tetramethylrhodamine
B isothiocyanate (1:200 in PBS) was added and left incubating
for 45 min. Another washing step was performed and the same
samples were incubated with 4,6-Diamidino-2-phenyindole,
dilactate (1:1000 in PBS) for 15 min. Cell were washed with PBS
and observed in a transmitted and reflected light microscope
with Apotome 2 (Axio Imager Z1m, Zeiss).
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
The Key Role of Sulfation and Branching on Fucoidan Antitumor Activity
Macromolecular
Bioscience
www.mbs-journal.de
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (11 of 13) 1600340
www.advancedsciencenews.com
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
5.3. Fucoidan Characterization
5.3.1. Molecular Weight Determination by Gel
Permeation Chromatography
GPC measurements were performed with a Malvern Viscotek
TDA 305 with refractometer, right angle light scattering and vis-
cometer detectors on a set of four columns: precolumn Suprema
5 µm 8 × 50 S/N 3111265, Suprema 30 Å 5 µm 8 × 300 S/N
3112751, Suprema 1000 Å 5 µm 8 × 300 S/N 3112851 PL, and
Aquagel-OH MIXED 8 µm 7.5 × 300 S/N 8M-AOHMIX-46-51,
with refractive index detection (RI-Detector 8110, Bischoff).
The system was kept at 30 °C using 0.1 m NaN3, 0.01 m NaH2PO4
(pH = 6.6) as eluent at rate of 1 mL min−1. The elution times of
the RI detector signal were calibrated with a commercial polysac-
charide set from Varian that contains 10 Pullulans with narrow
polydispersity and Mp (molecular mass at the peak maximum)
ranging from 180 Da to 708 kDa.
5.3.2. Depolymerization of Fucoidan Extract FE1
by Acid Hydrolysis
FE 1 was hydrolyzed to obtain lower molecular weight fucoidan,
following a procedure described previously.[38] Briefly, partially
hydrolyzed fucoidan was obtained by hydrolyzing FE 1 (20 mg)
with 1 mL of 0.01 m HCl at 100 °C for 10 min, followed by neu-
tralization with 1 mL of 0.01 m NaOH. The resulting hydrolyzed
fucoidan FE1* was analyzed by GPC. A control was performed by
dissolving FE 1 (20 mg) in 1 mL of 0.01 m HCl without heat treat-
ment, followed by neutralization with 1 mL of 0.01 m NaOH. The
procedures described in Sections 5.2.2 and 5.2.3 were followed,
but, instead of FE 1, hydrolyzed fucoidan was added at dif-
ferent concentrations (0.2, 0.5, and 1 mg mL−1). A positive control
with untreated cells and controls with HCl and NaOH without
fucoidan were also performed.
5.3.3. Carbohydrates Composition: Determination
of Neutral and Acidic Monosaccharides
Neutral monosaccharides were determined as alditol acetates
as described elsewhere.[48] Briefly a prehydrolysis of fucoidans
was performed, with 72% sulfuric acid for 3 h at room tempera-
ture (RT). Afterward, the fucoidan extracts were submitted to a
hydrolysis with sulfuric acid 1 m at 100 °C for 2.5 h. 2-Deoxyglu-
cose was used as an internal standard. Monosaccharides were
reduced with sodium borohydride and acetylated by acetic anhy-
dride using methylimidazole as a catalyst. The formed alditol
acetate derivatives were analyzed by gas chromatography (GC)
with a 30 m column DB-225 (J&W Scientific, Folsom, CA, USA),
internal diameter, and film thickness of 0.25 and 0.15 mm,
respectively, and using a flame ionization detector (Perkin Elmer,
Clarus 400). The hydrolysis of all samples was performed in
duplicate and each one was injected twice. A third analysis was
done for the few samples with higher variability. Uronic acids
were quantified by a modification of the 3-phenylphenol col-
orimetric method.[48] Samples were prepared by hydrolysis with
72% sulfuric acid for 3 h at RT followed by 1 h in sulfuric acid 1 m
at 100 °C. A calibration curve was made with d-galacturonic acid.
The hydrolysis and analysis of the samples was done in triplicate.
5.3.4. NMR Spectroscopy
NMR experiments were acquired on a Bruker AVANCE 400 spec-
trometer using D2O as solvent. The residual HOD signal was used
as reference for the chemical shifts that are reported in ppm.
Mnova Software 9.0 (Mestrelab Research) was used for spectral
processing. Sulfation position and branching was qualitatively
analyzed based on the chemical shifts described previously.[26,27]
5.3.5. Sulfation Degree: Ester Sulfate Determination
The content of ester sulfates in the fucoidan extracts was deter-
mined by the turbidimetric method proposed by Dodgson
and Price.[49,50] The fucoidan extracts were accurately weighed
(usually 2–4 mg) and dissolved in the respective amount of
N-hydrochloric acid 1 m. The mixture was submitted to a hydrol-
ysis at 105–110 °C for 5 h. A portion (0.2 mL) was transferred to
a tube containing 3.8 mL of 3% (w/v) trichloroacetic acid. Barium
chloride-gelatin reagent (1 mL) was added and, after mixing,
the whole was kept at RT for 15–20 min. The barium chloride-
gelatin reagent was previously prepared by mixing gelatin (1 g)
with 200 mL of hot water (60–70 °C) and was allowed to stand at
4 °C overnight. Barium chloride (1 g) was dissolved in the semi-
gelatinous fluid and the resultant cloudy solution was allowed
to stand for 2–3 h before use. The solution was then analyzed at
360 nm (Jenway 6405 UV/Vis) against reagent blank containing
distilled water instead of sample.
A second 0.2 mL portion of the hydrolysate was mixed with
3.8 mL of trichloroacetic acid, as described above, and with
1 mL of gelatin solution (i.e., containing no barium chloride).
The extinction of this “control” solution was then measured at
360 nm against a reagent blank consisting on distilled water
instead of sample and 1 mL of gelatin solution. The concentration
of sulfate esters was determined by building a K2SO4 calibration
curve, containing between 20 and 200 µg of SO42− ion.
5.3.6. Desulfation
For polysaccharide desulfation, 10 mg were dissolved in 1.8 mL of
dried dimethyl sulfoxide. Then, 0.1 mL pyridine was added, fol-
lowed by 13 mg of pyromellitic acid, 12 mg of NaF, and 0.2 mL
of pyridine. The mixture was stirred at 120 °C for 3 h, cooled and
poured into 1 mL of 3% of NaHCO3 aqueous solution. The solu-
tion containing the desulfated polysaccharide was dialysed and
freeze-dried. The procedure was repeated in order to guarantee
the complete desulfation. Afterward, the desulfated polysaccha-
ride was submitted to methylation analysis.[51,52]
5.3.7. Methylation Analysis
Glycosidic-substitution analysis was determined by gas chroma-
tography-quadrupole mass spectrometry (GC-qMS) of the par-
tially methylated alditol acetates based on Ciucanu and Kerek[53]
and Coelho et al.[54] The native and desulfated samples (1–2 mg)
were dissolved in 1 mL of anhydrous dimethylsulfoxide, and then
powdered NaOH (40 mg) were added under an argon atmosphere.
The samples were methylated with CH3I (80 µL) during 20 min
with stirring, following by a second and third addition of 80 µL
CH3I and stirring for another 20 min. CHCl3/MeOH (1:1, v/v, 3 mL)
C. Oliveira et al.
Macromolecular
Bioscience
www.mbs-journal.de
1600340 (12 of 13) © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
was added, and the solution was dialyzed (membrane with a
pore diameter of 12–14 kDa) against 50% EtOH. The dialysate
was evaporated to dryness and the material was remethylated
using the same procedure. The remethylated material was hydro-
lyzed with 2 m TFA (1 mL) at 120 °C for 1 h, and then reduced and
acetylated as previously described for neutral sugar analysis
(using NaBD4 instead of NaBH4). The partially methylated alditol
acetates were separated and analyzed by GC-qMS (GC-2010 Plus,
Shimadzu). The GC was equipped with a DB-1 (J&W Scientific,
Folsom, CA, USA) capillary column (30 m length, 0.25 mm of
internal diameter, and 0.10 µm of film thickness). The samples
were injected in “split” mode with the injector temperature at
250 °C. The temperature program used was as follows: (1) an
initial temperature of 80 °C; (2) an increase of 7.5 °C min−1 until
140 °C and a hold time of 5 min; (3) an increase of 0.2 °C min−1
until 143.2 °C; (4) an increase of 12 °C min−1 until 200 °C; (5) an
increase of 50 °C min−1 until 250 °C and a hold time of 5 min. The
helium carrier gas had a total flow rate of 8.5 mL min−1. The GC
was connected to GCMS-QP 2010 Ultra Shimadzu mass quadru-
pole selective detector operating with an electron impact mode
at 70 eV and scanning the range m/z 50–700 in a 1 s cycle in a
full scan mode acquisition.
5.4. Statistical Analysis
Statistical analysis was performed using Graph Pad Prism
Software. Differences between the different conditions of
the cellular assays were analyzed using nonparametric test
(Kruskal–Wallis test) and a p < 0.05 was considered significant.
Data are presented as mean ± standard deviations.
Acknowledgements: The authors would like to thank the
funding from projects 0687_NOVOMAR_1_P, cofunded by
INTERREG 2007–2013/POCTEP, CarbPol_u_Algae (EXPL/
MAR-BIO/0165/2013), and IF/00376/2014/CP1212/CT0015,
funded by the Portuguese Foundation for Science and
Technology, FCT, and ComplexiTE (ERC-2012-ADG 20120216-
321266), funded by the European Research Council under the
European Union’s Seventh Framework Programme for Research
and Development. The authors would also like to thank FCT,
Portugal, for the scholarship of A.S.F. (SFRH/BD/102471/2014),
fellowship of C.N. (SFRH/BPD/100627/2014), Investigator
grants of A.M. (IF/00376/2014), R.N.-C. (IF/00373/2014), and I.P.
(IF/00032/2013) and the financial support to CICECO- Aveiro
Institute of Materials (POCI-01-0145-FEDER-007679, FCT UID/
CTM/50011/2013) and QOPNA (UID/QUI/00062/2013), through
national founds and cofinanced by the FEDER, within the PT2020
Partnership Agreement.
Received: August 8, 2016; Revised: November 16, 2016;
Published online: ; DOI: 10.1002/mabi.201600340
Keywords: antitumor activity; fucoidan; polymer branching;
structure–activity relationship; sulfation
[1] G. M. Cooper, The Cell - A Molecular Approach, Sinauer
Associates, Sunderland, MA 2000.
[2] K. Sak, Chemother. Res. Pract. 2012, 2012, 282570.
[3] A. Gennari, M. D’amico, D. Corradengo, Ther. Adv. Med.
Oncol. 2011, 3, 229.
[4] C. DeSantis, J. Ma, L. Bryan, A. Jemal, CA. Cancer J. Clin. 2014,
64, 52.
[5] National Collaborating Centre for Cancer, U. National
Collaborating Centre for Cancer, UK 2009.
[6] M. M. Gottesman, Annu. Rev. Med. 2002, 53, 615.
[7] H.-Y. Hsu, T.-Y. Lin, Y.-C. Wu, S.-M. Tsao, P.-A. Hwang,
Y.-W. Shih, J. Hsu, Oncotarget 2014, 5, 7870.
[8] F. Atashrazm, R. M. Lowenthal, G. M. Woods, A. F. Holloway,
J. L. Dickinson, Mar. Drugs 2015, 13, 2327.
[9] T. H. Silva, a. Alves; B. M. Ferreira, J. M. Oliveira, L. L. Reys,
R. J. F. Ferreira, R. a Sousa; S. S. Silva, J. F. Mano, R. L. Reis,
Int. Mater. Rev. 2012, 57, 276.
[10] J. H. Fitton, D. N. Stringer, S. S. Karpiniec, Mar. Drugs 2015,
13, 5920.
[11] L. Yang, P. Wang, H. Wang, Q. Li, H. Teng, Z. Liu, W. Yang,
L. Hou, X. Zou, Mar. Drugs 2013, 11, 1961.
[12] N. Li, Q. Zhang, J. Song, Food Chem. Toxicol. 2005, 43, 421.
[13] B. Li, F. Lu, X. Wei, R. Zhao, Molecules 2008, 13, 1671.
[14] M. T. Ale, J. D. Mikkelsen, A. S. Meyer, Mar. Drugs 2011, 9, 2106.
[15] G. Moussavou, D. H. Kwak, B. W. Obiang-Obonou,
C. A. O. Maranguy, S.-D. Dinzouna-Boutamba, D. H. Lee,
O. G. M. Pissibanganga, K. Ko, J. I. Seo, Y. K. Choo, Mar. Drugs
2014, 12, 4898.
[16] M. T. Ale, H. Maruyama, H. Tamauchi, J. D. Mikkelsen,
A. S. Meyer, Int. J. Biol. Macromol. 2011, 49, 331.
[17] H.-J. Boo, J.-H. Hyun, S.-C. Kim, J.-I. Kang, M.-K. Kim,
S.-Y. Kim, H. Cho, E.-S. Yoo, H.-K. Kang, Phytother. Res. 2011,
25, 1082.
[18] H.-J. Boo, J.-Y. Hong, S.-C. Kim, J.-I. Kang, M.-K. Kim, E.-J. Kim,
J.-W. Hyun, Y.-S. Koh, E.-S. Yoo, J.-M. Kwon, H.-K. Kang, Mar.
Drugs 2013, 11, 2982.
[19] Z. Zhang, K. Teruya, T. Yoshida, H. Eto, S. Shirahata,
Mar. Drugs 2013, 11, 81.
[20] H. Lee, J. S. Kim, E. Kim, PLoS One 2012, 7, e50624.
[21] K. Senthilkumar, P. Manivasagan, J. Venkatesan, S.-K. Kim,
Int. J. Biol. Macromol. 2013, 60, 366.
[22] J.-Y. Kwak, Mar. Drugs 2014, 12, 851.
[23] A. Kasai, S. Arafuka, N. Koshiba, D. Takahashi, K. Toshima,
Org. Biomol. Chem. 2015, 13, 10556.
[24] F. Atashrazm, R. Lowenthal, G. Woods, A. Holloway,
J. Dickinson, Mar. Drugs 2015, 13, 2327.
[25] B. Li, F. Lu, X. Wei, R. Zhao, Molecules 2008, 13, 1671.
[26] M.-J. Clément, B. Tissot, L. Chevolot, E. Adjadj, Y. Du,
P. A. Curmi, R. Daniel, Glycobiology 2010, 20, 883.
[27] M. I. Bilan, A. A. Grachev, A. S. Shashkov, N. E. Nifantiev,
A. I. Usov, Carbohydr. Res. 2006, 341, 238.
[28] S. Koyanagi, N. Tanigawa, H. Nakagawa, S. Soeda,
H. Shimeno, Biochem. Pharmacol. 2003, 65, 173.
[29] L. Chevolot, B. Mulloy, J. Ratiskol, A. Foucault,
S. Colliec-Jouault, Carbohydr. Res. 2001, 330, 529.
[30] V. H. Pomin, P. A. S. Mourão, Glycobiology 2008, 18, 1016.
[31] S. Z. Moghadamtousi, H. Karimian, R. Khanabdali, M. Razavi,
M. Firoozinia, K. Zandi, H. A. Kadir, Sci. World. J. 2014, 2014,
768323.
[32] K. Wang, L.-H. Cai, B. Lan, J. J. Fredberg, Nat. Methods 2016,
13, 124.
[33] F. R. Balkwill, M. Capasso, T. Hagemann, J. Cell Sci. 2012, 125,
5591.
[34] D. L. Holliday, V. Speirs, Breast Cancer Res. 2011, 13, 215.
[35] A. S. Levenson, V. C. Jordan, Cancer Res. 1997, 57, 3071.
[36] R. Kimura, T. Rokkaku, S. Takeda, M. Senba, N. Mori,
Mar. Drugs 2013, 11, 4267.
The Key Role of Sulfation and Branching on Fucoidan Antitumor Activity
Macromolecular
Bioscience
www.mbs-journal.de
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (13 of 13) 1600340
www.advancedsciencenews.com
Macromol. Biosci. 2017, DOI: 10.1002/mabi.201600340
[37] R. V. J. Chari, Acc. Chem. Res. 2008, 41, 98.
[38] C. Yang, D. Chung, I.-S. Shin, H. Lee, J. Kim, Y. Lee, S. You,
Int. J. Biol. Macromol. 2008, 43, 433.
[39] J. I. Choi, H. J. Kim, Carbohydr. Polym. 2013, 97, 358.
[40] L. Zhang, S. Bhatnagar, E. Deschenes, M. Thurber Greg,
Sci. Rep. 2016, 6, 25424.
[41] M. M. Schmidt, K. D. Wittrup, Mol. Cancer Ther. 2009, 8, 2861.
[42] L.-E. Rioux, S. L. Turgeon, M. Beaulieu, Phytochemistry 2010,
71, 1586.
[43] T. Nishino, C. Nishioka, H. Ura, T. Nagumo, Carbohydr. Res.
1994, 255, 213.
[44] M. Nagaoka, H. Shibata, I. Kimura-Takagi, S. Hashimoto,
K. Kimura, T. Makino, R. Aiyama, S. Ueyama, T. Yokokura,
Glycoconj. J. 1999, 16, 19.
[45] T. M. Buttke, J. A. McCubrey, T. C. Owen, J. Immunol. Methods
1993, 157, 233.
[46] S. Prigent-Richard, M. Cansell, J. Vassy, A. Viron, E. Puvion,
J. Jozefonvicz, D. Letourneur, J. Biomed. Mater. Res. 1998, 40,
275.
[47] M. Ellouali, C. Boisson-Vidal, P. Durand, J. Jozefonvicz,
Anticancer Res. 1993, 13, 2011.
[48] C. Nunes, L. Silva, A. P. Fernandes, R. P. F. Guiné,
M. R. M. Domingues, M. A. Coimbra, Carbohydr. Polym. 2012,
87, 620.
[49] K. S. Dodgson, Biochem. J. 1961, 78, 312.
[50] K. S. Dodgson, R. G. Price, Biochem. J. 1962, 84, 106.
[51] I. J. Miller, J. W. Blunt, Carbohydr. Res. 1998, 309, 39.
[52] S. Geresh, S. M. Arad, O. Levy-Ontman, W. Zhang, Y. Tekoah,
R. Glaser, Carbohydr. Res. 2009, 344, 343.
[53] I. Ciucanu, F. Kerek, Carbohydr. Res. 1984, 131, 209.
[54] S. Vincenzi, A. Crapisi, A. Curioni, Food Hydrocoll. 2014, 34,
202.
