J Am Oil Chem Soc (2015) 92:571–587

DOI 10.1007/s11746-015-2624-5

ORIGINAL PAPER

Antioxidant Activity of Seaweed Extracts: In Vitro

Assays, Evaluation in 5 % Fish Oil‑in‑Water Emulsions
and Characterization
K. H. Sabeena Farvin · Charlotte Jacobsen 

Received: 9 January 2015 / Revised: 19 February 2015 / Accepted: 21 February 2015 / Published online: 12 March 2015
© AOCS 2015

Abstract  In this study the antioxidant activity of abso- Introduction

lute ethanol, 50 % ethanol and water extracts of two spe-
cies of seaweeds, namely Fucus serratus and Polysipho‑ The oxidative degradation of lipids or oils in raw or pro-
nia fucoides, were evaluated both in in vitro assays and in cessed food is important with respect to quality deteriora-
5 % fish oil-in-water (o/w) emulsions. The 50 % ethanolic tion as it imparts unwanted off-flavours and also deterio-
extracts of P. fucoides showed higher antioxidant activity rates the vital nutrients such as lipids, proteins and vitamins
both in in vitro assays and in 5 % oil-in-water emulsion in in food [1]. Lipid oxidation is mediated by a free radical
the presence or absence of iron. In spite of the higher phe- mechanism both in food systems and in biological systems
nolic content and very good antioxidant activity in some of and also contributes to several disease conditions such as
the in vitro assays, the absolute ethanol extracts of both the carcinogenesis, arteriosclerosis and the ageing process
species showed a pro-oxidative tendency in 5 % fish oil-in- in humans [2–4]. One way to prevent oxidation in food
water emulsion in the presence or absence of iron. In order products is by the addition of antioxidants. Antioxidants
to investigate the reason for the higher antioxidant activ- inhibit oxidation of lipids by transforming free radicals/
ity of 50 % ethanolic extracts of P. fucoides, these extracts peroxy radicals into non-radicals by donating electrons
were further fractionated into polyphenol-rich, protein-rich, and hydrogen or by chelating transition metals [5]. Several
polysaccharide-rich and low-molecular-weight fractions. synthetic antioxidants are available in the market. Their use
These fractions were tested both in in vitro and in 5 % oil- is restricted because of their reported carcinogenic effects
in-water emulsions. The results of the present study showed [6] and also the pathological damage in kidney and other
that the main effect was due to the phenolic compounds. organs [7, 8]. This leads to the search for new natural anti-
In conclusion, the 50 % ethanolic extracts of P. fucoides oxidants with multifunctional potential as an alternative to
can be a potential source of natural antioxidants as these synthetic antioxidants to prevent lipid oxidation in foods.
extracts have antioxidant activities similar to those of syn- Research in natural products of marine macroalgae (sea-
thetic antioxidants such as BHT. weed) has made significant advances in recent years and
marine macroalgae have been shown to produce a variety
Keywords  Fucus serratus · Polysiphonia fucoides · of compounds. Some of them have been demonstrated to
Total phenolic content · Antioxidant · possess biological activity of potential medicinal value [9,
5 % oil-in-water emulsion 10]. Marine macroalgae are reported to be rich sources of
various natural antioxidants such as catechins (e.g. gal-
locatechin, epicatechin and catechingallate), flavonols
and flavonol glycosides, which have been identified from
methanol extracts of red and brown algae [11–13]. A group
K. H. Sabeena Farvin (*) · C. Jacobsen  of phenolic compounds found in brown algae called phlo-
Division of Industrial Food Research, National Food Institute
rotannins, which are polymers of phloroglucinol, have
(DTU‑Food), Technical University of Denmark, B. 221,
SøltoftsPlads, 2800 Kgs Lyngby, Denmark been reported to possess strong antioxidant activity and
e-mail: sabeenafarvin@gmail.com; safa@food.dtu.dk their free radical potential is more potent than that of other

13

572 J Am Oil Chem Soc (2015) 92:571–587

polyphenols derived from terrestrial plants [14]. In addi- standards for identification of phenolic acids were
tion, sulphated polysaccharides and carotenoid pigments obtained from Sigma Aldrich (Steinheim, Germany).
including astaxanthin and fucoxanthin have also been dem- Chloroform and methanol were of HPLC grade (Lab-
onstrated to possess excellent antioxidant potential [15– Scan, Dublin, Ireland). All the other reagents were of ana-
17]. Most of the studies on the antioxidant activity of the lytical grade and obtained from Merck (Darmstadt, Ger-
seaweeds have been performed in simple in vitro assays. many). Cod liver oil was kindly donated by Maritex A/S
Therefore studies in more complex systems are necessary (Sortland, Norway). The fatty acid composition (unsatu-
to investigate their potential application as natural antioxi- rated only) of oil used was palmitoleic acid (16:1) 8.2 %;
dants in foods. oleic acid (18:1) 21.2 %; linoleic acid (18:2) 1.9 %;
In our earlier study on 16 species of marine macroal- α-linolenic acid (18:3) 0.84 %; stearidonic acid (18:4)
gae from the Danish coast, we found that the Polysipho‑ 2.5 %; eicosenoic acid (20:1) 11.6 %; eicosapentaenoic
nia fucoides and all the Fucus species tested (F. serratus, acid (20:5) 9.3 %; erucic acid (22:1) 0.8 %; docosapen-
F. vesiculosus, F. distiches, F. spiralis) showed the highest taenoic acid (22:5), 1.1 %; docosahexaenoic acid (22:6)
radical-scavenging activity, reducing power, inhibition of 11.6 %; saturated fatty acid content 14.1 %. The perox-
oxidation in a liposome model system and in bulk fish oil ide value (PV) of the oil used was 0.07 ± 0.03 meq/kg.
[18]. In real food systems, lipids usually exist in the emul- The seaweeds, Fucus serratus and Polysiphonia fucoides,
sion form instead of bulk form. As the emulsion is a mul- were collected from Houhavn (55°54′39″N 10°14′59″E)
tiphase system with different chemical environments in the and Limfjorden (56°45′00.04″N 8°50′12.52″E), Denmark,
different phases, the behaviour of antioxidants in emulsion respectively, between April and September 2009. Fresh
is more complex than that in bulk fish oil and therefore this seaweeds were washed with distilled water and their hold-
needs to be further explored. Moreover, antioxidants will fasts and epiphytes were removed. Immediately thereaf-
behave differently in the presence and absence of iron in the ter the rinsed seaweeds were placed in a freezer (−40 °C)
food system. For example, caffeic acid, a well-known anti- until further use. After thawing, the seaweed samples
oxidant [19], was reported to have a pro-oxidative effect in were freeze-dried for 2 days, ground to a fine powder and
emulsions in the presence of iron [20]. Hence, the overall passed through a 0.5-mm sieve to obtain a uniform pow-
aim of the present study was to examine the utilisation of F. der. The powdered seaweeds were stored at −80 °C under
serratus and P. fucoides as a source of natural antioxidants vacuum packing.
for retarding lipid oxidation in 5 % fish oil-in-water emul-
sion in the presence and absence of iron. A second aim was Preparation of the Algal Extract
to investigate whether the antioxidant activity of the sea-
weed extracts could be attributed to specific compounds/ Absolute ethanol, 50 % ethanol and water were used as
fractions of the extracts. In our earlier study [18], we used extraction solvents. For the preparation of extracts, 10 g of
water and 96 % ethanol for the extraction. However, in the powdered seaweed was extracted overnight with 100 ml of
present study we performed the extraction with water, 50 % 99 % ethanol or 50 % ethanol or water at room temperature
ethanol or absolute ethanol in order to investigate whether and centrifuged at 1,665 g for 10 min. The supernatant was
the concentration of ethanol affected the antioxidant activ- collected in a separate bottle after passing through What-
ity in vitro. Subsequently, those extracts, which were effi- man no. 4 filter paper. The residue was re-extracted three
cient in in vitro assays, were tested in 5 % fish oil-in-water times under the same conditions as described above. The
emulsions with or without iron addition. Finally, extracts combined filtrate of 99 % ethanol or 50 % ethanol extracts
were further fractionated into polyphenol-rich, protein-rich, was evaporated in a rotary evaporator (BUCHI, Switzer-
polysaccharide-rich and low-molecular-weight fractions. land) below 40 °C. The extract obtained after evaporation
These fractions were then tested both in in vitro and in 5 % of the organic solvent was weighed for calculation of the
oil-in-water emulsions. extraction yield and used as a natural antioxidant in 99 %
ethanol extracts. The water extracts and remaining aqueous
phase in 50 % ethanol extracts after evaporation of etha-
Materials and Methods nol were freeze-dried for 24 h. These extracts were kept
at −80 °C until analysis. The extraction procedures were
Materials repeated and the resulting extracts were pooled until suf-
ficient quantity of materials for the emulsion experiments
l-α-Phosphatidyl choline (PC), 1,1-diphenyl-2-pycryl- was obtained. This also minimised the effect of batch-to-
hydrazyl (DPPH), 3-(2-pyridyl)-5,6-diphenyl-1,2,4- batch variation in the extraction procedure. The freeze-
triazine-4′,4″-disulphonic acid monosodium salt (Fer- dried extracts were dissolved in the corresponding solvents
rozine), Folin-Ciocalteu’s phenol reagent and external when used for analysis.

13
J Am Oil Chem Soc (2015) 92:571–587 573

Determination of Total Phenolic Content Iron (Fe2+)‑Chelating Activity

Total phenolic content in the extracts was determined by The Fe2+-chelating activity of the fractions was estimated
the method described by Singleton and Rossi [21]. An ali- by the method of Farvin et al. [23]; 100 µl of the extracts at
quot (100 μl at a concentration of 2.5 mg/ml) of extract a concentration of 0.25, 0.5 and 1 mg/ml were used for the
was mixed with 0.75 ml of Folin-Ciocalteu reagent (1:10 analysis. EDTA at 1 mg/ml concentration was also evalu-
diluted) and allowed to stand at room temperature for ated for comparison. The chelating capacity was calculated
5 min. Sodium bicarbonate (6 %, 0.75 ml) was added to the as follows.
mixture and incubated at room temperature for 90 min. The
absorbance was measured at 725 nm using a spectropho- Iron chelating activity (%)
tometer (Shimadzu UV mini 1240, Duisburg, Germany). A
 
1 − (Abs of sample − Abs. of sample control)
= × 100
standard curve was plotted using different concentrations of Abs of Blank
gallic acid and the amount of total phenolics was calculated
as gallic acid equivalents in mg/100 g of dried seaweed. Reducing Power

Identification and Quantification of Phenolic Acids The reducing power was measured according to the method
by HPLC of Farvin et al. [23] for analysis in a micro plate reader;
100 µl of extracts at a concentration of 0.25, 0.5 and 1 mg/
Identification and quantification of phenolic acid was done ml was used for analysis. Increased absorbance (A700) of
by a modified method of Onyeneho and Hettiarachchy [22]. the reaction mixture indicated increased reducing power.
The extract was passed through a 0.45-μm filter (Milli- Ascorbic acid (AA) at 1 mg/ml concentration was used for
pore, Westboro, MA) before being injected into the HPLC. comparison.
Reverse phase HPLC was performed with an Agilent 1100
series HPLC (Agilent Technologies, Palo Alto, CA, USA), Inhibition of Lipid Peroxidation in a Liposome Model
equipped with a diode array detector (Agilent G13158). System
The column used was a ZORBAX Eclipse® XDB C18 ana-
lytical column (150 mm × 4.6 mm) (Agilent, USA) with 5 Liposomes were prepared from soybean phosphatidyl cho-
μm packing material. Elution was performed with an iso- line according to the method of Farvin et al. [23]. Lipid
cratic mixture of methanol and 10 mM ammonium acetate oxidation was performed in a model system containing
buffer, pH 5.4 (12:88 v/v), at a flow of 1 ml/min. Detec- 0.1 mg of phosphatidyl choline liposomes per ml of phos-
tion was done using a diode array detector with reference phate buffered saline (PBS) (3.4 mM Na2HPO4–NaH2PO4,
wavelength of 280 nm. Retention times and peak areas 0.15 M NaCl, pH 7.0) and extracts at concentrations of
were monitored and computed automatically by a Chem32 0.63, 1.25 and 2.5 mg/ml. Lipid oxidation was initiated by
integrator (Agilent, USA). Individual phenolic acids were iron redox cycling using 50 μM FeCl3 and 100 μM ascor-
identified by the retention time of sample chromatographic bate. The order of addition was buffer, extracts, liposome,
peaks being compared with those of authentic standards ferric chloride and ascorbic acid. The reactants were mixed
using the same HPLC operating conditions. by vortexing for 2 s and incubated at 37 °C in a water bath
for 1 h. The liposome assay solution with distilled water
Screening of the Extracts for Antioxidant Activity instead of sample was used as control. Lipid oxidation was
measured by determining the concentrations of thiobarbitu-
DPPH Radical‑Scavenging Activity ric acid reactive substance (TBARS) formed according to
the method of Farvin et al. [23]. BHT at a concentration of
The scavenging effect on α,α-diphenyl-β-picrylhydrazyl 0.2 mg/ml was also tested in a similar way for comparison.
(DPPH) free radical was measured by the method of Farvin The amount of TBA-reactive substances (MDA) released/
et al. [23]. The extracts at concentrations of 0.25, 0.5 and mg phospholipid (PL) was calculated using the molar
1 mg/ml were used for the analysis. BHT at 0.5 mg/ml con- extinction coefficient of MDA as 1.56 × 105. The  % inhi-
centration was also used for comparison. The radical-scav- bition of TBARS formation was calculated as follows
enging capacity was calculated as follows.  
Tc − T s
% inhibition = × 100
Tc
DPPH radical scavenging capacity (%) where Tc is the µmoles of MDA released by the control
(liposome alone) and Ts is the µmoles of MDA released by
 
1 − (Abs of sample − Abs. of sample control)
= × 100
Abs. Blank the samples.

13

574 J Am Oil Chem Soc (2015) 92:571–587

Antioxidant Effect of Algal Extracts in 5 % Fish‑Oil‑ 225 mg of Tenax GR (60–80 mesh, Varian, Middelburg,
in‑Water Emulsion The Netherlands).
Thermal Desorption and GC-MS An ATD-400 auto-
Five percent oil-in-water emulsion was prepared with 1 % matic thermal desorber with a Tenax GR-packed cold-
citrem (Danisco Dupont, Grindsted, Denmark) as emulsi- trap (Perkin-Elmer) was used for thermally desorbing the
fier. In brief, 5 g of citrem and 25 g fish oil were weighed collected volatiles. Helium was used as carrier gas with a
into a glass beaker and mixed together by a magnetic stirrer. flow of 1.3 ml/min. The transfer line of the ATD was con-
Then 470 ml of buffer (imidazole:acetate, 10 mM, pH 7) was nected to a 5890 IIA gas chromatograph (Hewlett-Pack-
measured into a one litre beaker and the algal extracts (Abso- ard, Palo Alto, CA) equipped with a DB 1701 column
lute ethanol and 50 % ethanol) at a concentration of 500 mg/ (30 m × 0.25 mm × 1.0 μM, J&W Scientific, Folsom, CA)
kg were dissolved into the buffer. A pre-homogenisation was coupled to an HP 5972A mass-selective detector. The tem-
done for 3 min using an Ultra-Turrax (T1500, Ystral, Dot- perature programme used was as follows: 45 °C for 5 min,
tingen, Germany) by adding the oil/citrem mixture slowly raised from 45 to 55 at 1.5 °C/min, raised from 55 to 90
over 1 min; then mixing was continued for a further 2 min. at 2.5 °C/min, raised from 90 to 220 °C at 12 °C/min and
After the pre-homogenisation, the emulsion was prepared by finally held at 220 °C for 4 min. The GC-MS transfer line
using a microfluidiser (total pressure of 9 MPa, Microfluid- temperature was kept at 280 °C. The ionisation energy of
ics, Newton, MA) by passing four times. A control without the mass spectrometer was set at 70 eV in the EI mode, and
antioxidant and an emulsion with 200 mg/kg of BHT were the detector operated with a mass range between 30 and
also made for comparison. After making the emulsions, 250 with a scan rate at 3.35 scans/s.
400 ml each of the emulsions were poured into 500 ml sterile Identification and Quantification of Secondary Vola‑
blue-capped bottles in duplicate. Emulsions with and with- tile Oxidation Compounds Secondary volatile oxida-
out added iron were prepared. For the emulsions with iron, tion compounds were identified by MS library searches
FeSO4 solution (100 μM) was added in order to induce oxi- (Wiley 138 K, John Wiley and Sons, Hewlett-Packard)
dation and the emulsions were stored at 20 °C for 3 days on and by comparing retention time and spectra with MS runs
a magnetic stirring plate in the dark. The emulsions without of external standards. The external standards used were
added iron were stored at 50 °C for 5 days on a magnetic 1-butanol-3-methyl, 1-butanol-2-methyl, 1-penten-3-one,
stirring plate in the dark. The sampling was done from the 1-penten-3-ol,1-pentanol, pentanal, hexanal, 1-hexanol,
same bottle after 0, 12, 24, 48 and 96 h and the 5th day. heptanal, 4-heptenal, 2,4-hexadienal, 2,4-heptadienal, octa-
The samples for chemical analysis were transferred to sepa- nal and nonanal. The individual compounds were quan-
rate brown glass bottles, flushed with nitrogen and stored at tified through calibration curves made by adding 1 μl of
−80 °C until analyses. PV, volatiles, tocopherols and sensory standards of volatiles at five concentration levels to Tenax
evaluation was used to assess antioxidant activity. GR tubes directly. Quantitation of the compounds released
from the emulsion systems were performed by selected ion
Analysis of Peroxide Value (PV) monitoring. The target ion represents a specific MS frag-
mentation ion of each compound. The target ions were
Lipids from the emulsions were extracted by chloroform: verified on the basis of two or three qualifier ions and the
methanol (1:1 v/v) as described by Bligh and Dyer [24]. PV chromatographic retention time. Measurements were made
was measured directly on the Bligh and Dyer extract by col- in triplicate on each sample.
orimetric determination of iron-thiocyanate according to the
method described by the international IDF standards [25]. Determination of Tocopherol Content

Dynamic Headspace Analysis of Volatile Secondary Tocopherol content was determined using an Agilent 1100
Oxidation Products in Emulsions series HPLC (Agilent Technologies, Palo Alto, CA, USA),
equipped with a fluorescence detector. About 2 g of the
Headspace Sampling The volatile compounds were sam- chloroform extract from the Bligh and Dyer extraction was
pled by the dynamic headspace technique. Four grams of evaporated under nitrogen and dissolved in 2 ml n-heptane;
emulsion, 30 mg of 4-methyl-1-pentanol (internal standard) from this, 1-ml samples were taken into separate vials
and 1 ml of synperonic (antifoam) were purged with nitro- before injection of an aliquot (40 µl) on a Spherisorb s5w
gen at 150 ml/min for 30 min at 45 °C. Antifoam was used column (250 mm × 4.6 mm) (Phase Separation Ltd., Dee-
to prevent the formation of foam, which would lead to con- side, UK). Elution was performed with an isocratic mixture
densation of water and contamination from the sample on of n-heptane/2-propanol (100: 0.4; v/v) at a flow of 1 ml/
the Tenax GR material. The volatile compounds were col- min. Detection was done using a fluorescence detector with
lected on traps (Perkin-Elmer, Hartford, CT) packed with excitation at 290 nm and emission at 330 nm and according

13
J Am Oil Chem Soc (2015) 92:571–587 575

to the AOCS Official method [26]. Two extractions were extracts of P. fucoides were analysed for chemical compo-
made from each sample and the measurement was per- sition. The total phenolic content of the PopRF was deter-
formed in duplicate and quantified by authentic standards. mined by the Folin-Ciocalteu method as described previ-
Results are expressed in µg tocopherol per g of lipid. ously [21].
The soluble protein content of the LMW aqueous frac-
Sensory Evaluation tion and ProRF was determined according to the Bradford
micro-assay protocol with bovine serum albumin (BSA) as
The sensory evaluation was performed by an expert panel the calibration standard [28].
composed of four persons. The emulsions were evaluated Total soluble carbohydrates and reducing sugars were
for colour and odour with the descriptors fishy/rancid and analysed colorimetrically for the LMW aqueous factions
others. For “other” the panel discussed and agreed upon and PosRF by the modified phenol-sulfuric acid [29] and
which attributes could be used to describe the odour or col- dinitrosalicylic acid (DNS) [30] methods, respectively, with
our of the specific emulsion. Samples were evaluated on a glucose as the standard.
continuous intensity scale ranging from zero intensity to a
maximum intensity of nine. The panel members assessed Analysis of Monosaccharides in Fractions
the samples on an individual basis. Subsequently the panel
agreed on an average score for each descriptor. Samples of The monosaccharide content in the PosRF and LMW aque-
50 ml were served in randomised order after incubation for ous fractions was determined by high-performance anion-
1 h at 10 °C. exchange chromatography coupled with pulsed electro-
chemical detection (HPAEC-PAD). In brief, 400 µl of 2 M
Fractionation of Extracts TFA was added to 2 mg of lyophilised sample in a screw-
cap vial. Each vial was tightly sealed and heated to 121 °C
About 3.3 g of freeze-dried 50 % ethanolic extracts of P. for 2 h in a drying oven. After hydrolysis the vials were
fucoides (prepared by the same procedure described above) cooled under tap water. The hydrolysates were lyophilised
were suspended in 100 ml distilled water, extracted three and kept at −20 °C under nitrogen until analysis. Prior to
times with 100 ml ethyl acetate (EtOAc). The EtOAc phase analysis by HPAEC-PAD, the hydrolysates were redis-
was collected into a separate bottle. The remaining aque- solved in 5 ml of double-deionised water containing 0.1 %
ous layer was adjusted to pH 5 and again extracted three of sodium azide to prevent microbial growth. Just before
times with 100 ml EtOAc. The remaining aqueous layer injection for HPAEC analysis each hydrolysate was filtered
was adjusted to pH 2 and extracted three times with 100 ml through a 0.22-µm GH polypro Acrodis ® filter (Pall Life
EtOAc. The EtOAc fractions were pooled into the same Sciences, Ann Arbor, MI, USA). The recovery value of the
bottle and evaporated off the solvent to get the polyphenol- monosaccharides was estimated by exposing a mixture of
rich fraction (PopRF). The aqueous layer was centrifuged monosaccharide standards l-(+)-fucose, l-(+)-rhamnose,
at 10,000 rpm for 10 min and the precipitate obtained d-(−)-arabinose, d-(+)-galactose, d-(+)-glucose, d-(+)-
after pH adjustment was taken as the protein-rich fraction xylose, d-(+)-mannose, l-(−)-guluronic acid and d-(+)-
(ProRF). The supernatant was precipitated with 99.5 % mannuronic acid to the above-described acid hydrolysis
ethanol (EtOH) (1:3) v/v and left at 4 °C for 5 h, followed conditions.
by centrifugation at 10,000 rpm for 10 min. The supernatant
obtained was concentrated to a small volume with a rotary Total Amino Acid Composition of Low‑Molecular‑Weight
evaporator and the resulting solution was freeze-dried and Fractions
considered the low-molecular-weight fraction. The precipi-
tate was dissolved in distilled water, treated with Sevag rea- A sample of 200 mg of LMW aqueous fractions was hydro-
gent [27] several times to remove any protein materials and lysed overnight in 2 ml of 6 M HCl in sealed ampoules. The
again precipitated with three volumes of ethanol and centri- samples were appropriately diluted and filtered through a
fuged at 10,000 rpm for 10 min. The precipitate was dried to 0.2 µm membrane filter and subsequently the amino acids
yield the polysaccharide-rich fraction (PosRF). The yield of were derivatised using an EZ: Fast kit from Phenomenex
each fraction is expressed as g % (g/100 g) of the extracts. A/S (Allerød, Denmark). The amino acid content was ana-
lysed by LC-MS as described by Farvin et al. [31].
Characterization of Fractions with Respect to Main
Constituents Antioxidant Screening of the Fractions

Four sub-fractions obtained, namely PopRF, PosRF, ProRF The antioxidant properties of the fractions were evaluated
and LMW aqueous fraction, derived from 50 % ethanolic using the four in vitro assays mentioned above, viz. DPPH

13

576 J Am Oil Chem Soc (2015) 92:571–587

radical-scavenging activity, Fe2+-chelating activity, reduc- compounds into the solution. An earlier study by Wang
ing power and inhibition of lipid oxidation in a liposome et al. [32] on extraction of polyphenolic compounds from
model system. In addition the different fractions were eval- seaweed also found that water was inferior to polar organic
uated for their antioxidant activity in 5 % oil in water emul- solvents in extracting polyphenolic compounds.
sions as previously described.
Individual Phenolic Content as Determined by HPLC
Statistical Analyses
The individual phenolic compounds showed wide vari-
The data obtained from in vitro assays, PV, volatiles and ation with the extracting solvent as well as with species
tocopherol were analysed by one- or two-way analysis of (Table 1). The absolute ethanol extracts showed higher lev-
variance (ANOVA). The statistical comparisons among els of protocatechuic acid. Absolute ethanolic extracts of F.
the samples were performed with the Bonferroni multiple serratus showed only protocatechuic acid and gallic acid,
comparison test using the Graph Pad Prism 4 statistical while the absolute ethanolic extracts of P. fucoides showed
package programme (Graph Pad Software Inc., San Diego, high levels of caffeic acid in addition to protocatechuic
CA, USA). A P value <0.05 was considered statistically acid and low levels of gallic, gentisic, vanillic and syringic
significant. acids. In the case of 50 % ethanolic extracts, P. fucoides
showed significantly high amounts of caffeic acid. The
level of protocatechuic acid was significantly lower than in
Results and Discussion the absolute ethanol fractions. The 50 % ethanolic extracts
of P. fucoides also contained minor quantities of gallic,
Yield and Total Phenolic Content hydroxybenzoic and chlorogenic acids. The 50 % etha-
nolic extracts of F. serratus contained only four phenolic
The yields of the various extracts (as % of dried algal pow- acids and their order of occurrence was gentisic > protocat-
der) varied with the solvent used for the extraction and it echuic > gallic > hydroxybenzoic acid. Water extracts of
increased in the order 99 % ethanol < water < 50 % eth- P. fucoides contained significantly high amounts of caffeic
anol. The yield of the absolute ethanol (δ  = 25) fraction acid and minor quantities of gentisic and gallic acids, while
was the lowest and was 8.4 ± 0.8 % for P. fucoides and the water extract of F. serratus was found to contain only
9.9  ± 0.1 % for F. serratus. The yield of water (δ  = 80) gallic acid. Even though the TPC of the absolute ethanol
extracts was 13.2 ± 1.6 % for P. fucoides and 29.5 ± 0.1 % extract of F. serratus was high, it contained fewer indi-
for F. serratus. The yield of 50 % ethanol (δ  = 52.2) vidual phenolic acids and a lower total concentration of
extracts was found to be the highest and was 21.7 ± 0.1 % phenolic acids than the 50 % ethanol extract. This might
and 37.4 ± 1.1 % for P. fucoides and F. serratus, respec- be due to the presence of high contents of phlorotannins in
tively. Polar solvents such as water can extract sugars, pro- this fraction and the fact that most of the compounds quan-
teins, salts and mucus. In 50 % ethanol extracts both the tified in this study were simple phenolic acids. Phlorotan-
polar (δ > 50) and semi-polar (δ = 20–50) compounds can nins are polymers of phloroglucinol derived entirely from
be extracted, which might be the reason for the higher yield acetate and are abundant in brown algae [33].
of these extracts.
In contrast to the yield, the total phenolic content (TPC) Antioxidant Activity of Extracts in In Vitro Systems
of the extracts increased with a decrease in solvent polar-
ity. Absolute ethanol had the highest total phenolic con- In order to study the antioxidant mechanisms of these
tent, which was followed by 50 % ethanol and water. In extracts, four in vitro assays, namely DPPH radical-scav-
absolute ethanol extracts, P. fucoides was found to have enging activity, iron-chelating activity, reducing power
the highest TPC (15.7 ± 0.6 g/100 g extract) followed and ability to inhibit oxidation in liposomes, were used
by F. serratus (12.2 ± 0.8 g/100 g extract). In 50 % eth- (Fig. 1a–d).
anol extracts, the highest TPC was found for F. serratus At the lowest concentration tested (0.25 mg/ml), the
(10.6 ± 1.9 g/100 g extract) and the TPC of P. fucoides was DPPH radical-scavenging activity was the highest in all the
only 4.4 ± 0.7 g/100 g of extract. Water extracts showed extracts and any increase in concentration above 0.25 mg/
the lowest levels, which ranged from 1.2 ± 0.1 g/100 g of ml resulted in either flattening or a decrease in radical-
extract (P. fucoides) to 6.7 ± 1.2 g/100 g extract (F. serra‑ scavenging activity (Fig. 1a). The decrease in DPPH-scav-
tus). The reason for the higher TPC of the absolute ethanol enging activity as the concentration increased might be
extracts and the decreasing TPC with increasing polarity due to the interference of carotenoids or other co-extracted
might be that ethanol precipitates proteins and polysaccha- compounds at higher concentrations. The interference of
rides, which release some of the reversibly bonded phenolic carotenoids in DPPH radical-scavenging activity has been

13
J Am Oil Chem Soc (2015) 92:571–587 577

Table 1  The major phenolic acids in the extracts of seaweeds identified by HPLC (mg/g extract)
Phenolic compounds (mg/g extract) F. serratus P. fucoides
AE 50E W AE 50E W

Gallic 0.45 ± 0.0 3.83 ± 0.1 2.6 ± 0.1 0.77 ± 0.0 0.61 ± 0.1 0.9 ± 0.0

Protocatechuic 9.21 ± 0.8 4.21 ± 0.0 – 14.01 ± 0.1 2.82 ± 0.1 –
Gentisic – 4.86 ± 0.0 – 0.22 ± 0.1 – 1.2 ± 0.1
Hydroxybenzoic – 0.27 ± 0.0 – – 0.15 ± 0.0 –
Chlorogenic – – – – 0.90 ± 0.0 –
Vanilic – – – 0.39 ± 0.0 – –
Syringic – – – 0.25 ± 0.0 – –
Caffeic – – – 2.04 ± 0.0 20.43 ± 0.0 15.0 ± 0.1
Salicylic – – – – – –
Coumaric – – – – – –
Ferulic – – – –

Values are mean ± SD (n = 3)

AE absolute ethanol, E ethanol extracts, W water extracts
When monophenolic acids were not detectable this is indicated by “–”

(a) (b)
100 100
% DPPH Radical scavenging

FAE FAE
% Fe2+ Chelating activity

PAE 80 PAE
80
F 50E F 50E
60 60 P 50E
P 50E
FWE FWE
40 40
PWE PWE
20 20

0 0
0.25 0.5 1.0 0.25 0.5 1.0
concentration in mg concentration in mg

(c) (d)
% inhibition of TBARS formation
Reducing power (OD at 700nm)

1.5 80
FAE P 50E
PAE F 50E
60
1.0 F 50E PAE
P 50E FAE
FWE 40 PWE
0.5 PWE FWE
20

0.0 0
0.25 0.5 1.0 0.63 1.25 2.5
concentration in mg concentration in mg

Fig. 1  Effect of concentration of seaweed extracts on (a) the DPPH lute ethanol (AE), 50 % ethanol (50E) and water (W) extracts of F.
radical-scavenging capacity, (b) Fe2+-chelating capacity, (c) reducing serratus (F) and P. fucoides (P). Values are expressed as mean ± SD
power, (d) antioxidant activity in the liposome model system of abso- (n = 3)

reported elsewhere [34]. At 0.25 mg/l concentration, all with 76.8 % radical-scavenging activity. At this concen-
the extracts except water extract of F. serratus had a higher tration, the water extracts of F. serratus showed signifi-
DPPH radical-scavenging activity than BHT at 0.5 mg/ml cantly (P < 0.05) lower DPPH radical-scavenging activity

13

578 J Am Oil Chem Soc (2015) 92:571–587

(72.2 %) when compared to the other extracts. There was high iron-chelating activity of these extracts. The absolute
no significant difference between the other extracts at the ethanol extracts showed lower iron-chelating activity. This
concentration of 0.25 mg/l. is in accordance with other studies where the metal chela-
Most of the reports on the antioxidant activity of sea- tion was reported to play a minor role in the overall activity
weeds have shown that there is a direct correlation between of the polyphenols [32, 41].
the DPPH radical-scavenging activity and total phenolic Irrespective of the stage in the oxidative chain in which
compounds in the extracts [32, 35]. In the present study, the antioxidant action is assessed, most non-enzymatic
higher radical-scavenging activity of the absolute and antioxidative activity is mediated by redox reactions [42].
50 % ethanolic extracts of F. serratus and lower radical- The antioxidant effects are related to the development of
scavenging activity of the water extracts of this species reductants and these reductants were reported to be ter-
are in accordance with these studies. Radical-scavenging minators of free radical chain reactions [43]. The reduc-
activity of phenolic compounds also depends upon their ing capacity of a compound may serve as a significant
unique phenolic structure and the number and location of indicator of its potential antioxidant activity. The reduc-
the hydroxyl groups [36]. Compounds with the second ing power of the absolute ethanol extracts of F. serratus
hydroxyl group in the ortho or para position have higher showed a concentration dependency and increased with an
activity than when it is in the meta position. Previous increase in concentration (Fig. 1c). The reducing power
studies have reported that caffeic acid with two hydroxyl of ascorbic acid at 1 mg/ml was found to be 1.56, which
groups is a more efficient antiradical compound than its was higher than those of all the extracts at this concentra-
monohydroxyl counterpart coumaric acid. Likewise gallic tion. Among the extracts, at 1 mg/ml the highest reducing
acid, a trihydroxyl phenol, is more potent than protocat- power was shown by absolute ethanol extracts of F. ser‑
echuic and gentisic acid, its dihydroxyl counterparts [36]. ratus and P. fucoides, which was followed by the water
Thus, the higher radical-scavenging activity of water and and 50 % ethanolic extracts of F. serratus and 50 % etha-
50 % ethanolic extracts of P. fucoides might be attributed nolic extracts of P. fucoides (Fig. 1c). Water extracts of P.
to the higher levels of caffeic acid in the extracts (Table 1). fucoides showed the lowest reducing power. Our results
In addition, red algae of the family Rhodomelaceae, espe- are in agreement with those of Jiménez-Escrig et al. [44]
cially the genera Rhodomela, Odonthalia and Polysiphonia, who also found a high ferric iron reducing power of Fucus
were reported to contain several kinds of bromophenols species. Interestingly, the red seaweed P. fucoides showed
[37]. Yan et al. [38] found that 2-bromo-3,4-dihydroxyl high reducing power only in ethanolic extracts, reveal-
benzaldehyde, a normal constituent in some red algae like ing that some compounds extracted in ethanol are giving
Rhodomela teres and Polysiphonia ureolata, showed higher reducing properties to this species. This might be due to
DPPH radical-scavenging activity than phloroglucinol. the presence of bromophenols, which could be extracted
The high DPPH radical-scavenging activity of the ethanolic into the ethanolic extracts. Bromophenols are reported to
extracts of P. fucoides also might be contributed by the have high reducing power [42].
bromophenols present in the extract. The ability of the extracts to prevent lipid oxidation in
The iron-chelating activity showed a different pattern a liposome model system is shown in Fig. 1d. There was
than the DPPH-scavenging activity and is shown in Fig. 1b. a decrease in inhibition of oxidation as the concentration
The iron-chelating activity showed a concentration depend- increased in most of the extracts except the water and
ency in most of the extracts and increased with an increase absolute extracts of P. fucoides. The 50 % ethanolic extract
in concentration. Water and 50 % ethanolic extracts of F. of P. fucoides showed the highest inhibition of lipid oxi-
serratus showed the highest iron-chelating activity of dation in the liposome model system and was as effective
more than 90 % when tested in the highest concentration. as BHT, which inhibited TBARS formation by 57.8 % at
At 1 mg/ml concentration water extracts of F. serratus a 0.2 mg/ml concentration. Water and absolute ethanolic
showed 99.7 % and the 50 % ethanolic fraction of F. ser‑ extracts of P. fucoides showed the lowest inhibition at the
ratus 92.3 % iron-chelating activity, which was higher than lowest concentration tested. However, as the concentra-
EDTA (90.1 %) at this concentration. The water and 50 % tion increased their inhibition of lipid oxidation was also
ethanolic extracts of P. fucoides showed intermediate iron- improved. Water, 50 % ethanolic and absolute ethanol
chelating activity. The higher iron-chelating activity of the extracts of F. serratus showed moderate inhibition at the
water extracts was also reported by Wang et al. [32]. Water beginning and decreased with an increase in concentration
extracts were low in monophenolic acids such as caffeic and 50 % ethanolic extracts showed the lowest inhibition
acids and the high metal-chelating activity was most likely at the highest concentration. The decrease in inhibition of
due to other compounds than monophenolics. As reported TBARS formation with an increase in concentration might
earlier [39, 40], polysaccharides such as fucoidan/alginates be due to the pro-oxidative activity of some of the co-
extracted in water extracts might be responsible for the extracted substances.

13
J Am Oil Chem Soc (2015) 92:571–587 579

Fig. 2  Effect of absolute ethanol (AE) and 50 % ethanol extracts content in 5 % fish oil-in-water emulsions without added iron during
(50E) of F. serratus (F) and P. fucoides (P) on (a) peroxide value storage at 50 °C. Values are expressed as mean ± SD (n = 3)
(PV), (b) 1-penten-3-one, (c) total volatiles and (d) total tocopherol

Antioxidant Effects of Extracts in 5 % Fish Oil‑in‑Water higher than the control emulsions. The emulsion contain-
Emulsions ing 50 % ethanolic extracts of F. serratus showed slightly
higher PV than the control up to 24 h. Thereafter the devel-
Preliminary experiments were performed to find the best opment of PV was slower than the control in the remaining
concentrations of the extracts to be used for the emul- part of the storage period.
sion experiments. It was found that water extracts did not For the emulsion with added iron (Fig. 3a), there was a
have antioxidant activity at the tested concentrations so we significant (P < 0.05) increase in the peroxide value in the
eliminated these extracts in the oil-in-water (o/w) emul- emulsion containing absolute ethanol extracts at day 0 itself
sion experiments in order to reduce the number of samples. and it ranged from 19.6 ± 2.3 to 24.5 ± 2.3 meq/kg oil for
Thus, a storage study was conducted by adding 500 mg/kg F. serratus and P. fucoides respectively. PV in P. fucoides
of the 50 % ethanolic extracts and absolute ethanol extracts absolute ethanol emulsion showed a decline after 24 h
of P. fucoides and F. serratus in 5 % fish oil-in-water emul- whereas it continued to increase in F. serratus absolute eth-
sion in the presence or absence of iron. anol-containing emulsions. It should be noted that the rate
The peroxide value (PV) is a chemical indicator of of progression of lipid oxidation was slower in absolute
how much of the oil is in the early stages of oxidation and ethanol extracts than in the control after the iron addition.
reflects the degree of oxidation. For the emulsion without This suggests that some compounds in the absolute etha-
added iron (Fig. 2a), at the beginning there was no signifi- nol extracts were making it pro-oxidative in the presence
cant difference (P > 0.05) in PV between the control and of iron. Even in the presence of iron P. fucoides 50 % etha-
emulsions containing different seaweed extracts. The PV at nolic extracts significantly (P < 0.05) inhibited the develop-
day 0 ranged from 0.1 ± 0.2 to 0.8 ± 0.6 meq/kg oil. The ment of PV during the entire storage period and were as
PV increased significantly during the storage period and effective as BHT at 200 mg/kg concentration. There was no
the control showed the highest PV of 51.9 ± 2.9 meq/kg significant difference (P > 0.05) between the control and
oil at the end of the storage period. The emulsion contain- the emulsion with F. serratus 50 % ethanol extract for 24 h,
ing 50 % ethanolic extracts of P. fucoides was as effective but thereafter F. serratus 50 % extracts reduced the devel-
as BHT at 200 mg/kg and showed significantly (P < 0.05) opment of PV compared to the control during the remain-
lower levels of PV throughout the storage period. The ing part of the storage period.
emulsions containing absolute ethanol extracts showed sig- The concentration of the following volatile second-
nificantly (P < 0.05) higher PV during the storage and was ary oxidation products was measured during the storage:

13

580 J Am Oil Chem Soc (2015) 92:571–587

(a) (b) 1-penten-3-one

80 25

FAE +Fe 20
PV (meq/kg oil)
60

ng/g emulsion
Control + Fe
15 control+Fe
40 F50E +Fe
PAE +Fe 10 F50E+Fe
BHT+Fe
20 P50E +Fe 5 FAE+Fe
BHT+Fe P50E+Fe
0 0 PAE+Fe
0 12 24 36 48 60 72 84 96 0 12 24 36 48 60 72 84 96
Storage time in hours Storage time in hours

(c) 1500 Total volatiles (d) Tocopherol

control+Fe 25
control+Fe
PAE+Fe PAE+Fe
20
ng/g emulsion

1000 FAE+Fe FAE+Fe

P50%E+Fe

ug/g oil
15 P50E+Fe
F50%E+Fe F50E+Fe
BHT+Fe 10
500 BHT+Fe
5

0 0
0 12 24 36 48 60 72 84 96 0 12 24 36 48 60 72 84 96
Storage time in hours Time in Hours

Fig. 3  Effect of absolute ethanol (AE) and 50 % ethanol extracts total tocopherol content in 5 % fish oil-in-water emulsions containing
(50E) of F. serratus (F) and P. fucoides (P) during storage at 20°C on iron as oxidation inducer. Values are expressed as mean ± SD (n = 3)
(a) peroxide value (PV), (b) 1-penten-3-one, (c) total volatiles and (d)

1-pentene-3-one, 1-pentene-3-ol, 2,4-heptadienal, 2,4-hex- storage period. There was a gradual increase in 1-penten-
adienal, 4-heptenal formed from n-3 fatty acids, pentanal 3-one concentrations in the control and emulsions with
and hexanal formed from n-6 fatty acids and heptanal, BHT, 50 % ethanolic and absolute ethanolic extracts of F.
octanal and nonanal formed from n-9 fatty acids. Many serratus up to 48 h, and thereafter a significant (P < 0.05)
of the detected volatiles have previously been identified in increase. At the end of the storage period, the control and
boiled fish [45] and fish oil [46] and have been shown to emulsion with 50 % ethanolic extract of F. serratus showed
correlate with the degree of oxidation in fish oil-enriched the highest level of 1-penten-3-one (Fig. 2b). The emulsion
emulsions [46, 47]. The content of 1-pentene-3-one and with 50 % ethanolic extract of P. fucoides showed signifi-
the total volatiles are shown in Figs. 2b, c and 3b, c. The cantly low 1-pentene-3-one content up to 96 h; thereafter
development of the volatiles showed different patterns in concentrations of 1-penten-3-on increased significantly
the presence and absence of iron. In the case of emulsions (P < 0.05) as the storage period progressed. At the end
without added iron (Fig. 2b, c), initially, there were no of the storage period, there was no significant difference
significant (P > 0.05) differences between samples for the (P > 0.05) between the effects of 50 % ethanolic extracts of
different volatiles and their level increased gradually dur- P. fucoides and absolute ethanol extracts of F. serratus on
ing the storage period. All emulsions with extracts showed 1-penten-3-one formation and both were found to be sig-
less total volatiles than the control throughout the storage nificantly (P < 0.05) lower than the control and BHT.
period (Fig. 2c). The emulsions with absolute ethanol and In emulsions added with iron (Fig. 3b, c), the emulsion
50 % ethanolic extract of P. fucoides showed low total vola- containing P. fucoides 50 % ethanolic extract showed low
tiles throughout the storage period. After 48 h of storage levels of all the volatiles analysed throughout the storage
there was an increase in total volatiles in the control, and period (Fig. 3b, c). There was no significant (P > 0.05)
in the emulsions with BHT, absolute and 50 % ethanolic difference between emulsions with P. fucoides 50 %
extracts of F. serratus and the control showed the highest extracts and BHT for all the volatiles (Fig. 3c) except for
total volatiles at the end of the storage period (Fig. 2c). 1-penten-3-one (Fig. 3b) for which BHT showed a signifi-
However, 1-penten-3-one showed a different pattern than cant (P < 0.05) increase after 48 h. The control showed a
total volatiles (Fig. 2b). Though the emulsion with an abso- steady increase in volatiles and showed the highest con-
lute ethanol extract of P. fucoides showed higher PV, it tent of volatiles at the end of the storage period. Both the
showed the lowest 1-penten-3-one content throughout the emulsion with the absolute ethanol extracts and the 50 %

13
J Am Oil Chem Soc (2015) 92:571–587 581

ethanol extracts of F. serratus showed higher levels of total for the control, both absolute ethanol and 50 % ethanolic
volatiles and 1-penten-3-one than the control in the initial extracts of F. serratus in the presence of iron. However, in
part of the storage period and hence showed a pro-oxida- the absence of iron, there was not much difference in the
tive tendency. The emulsion containing an absolute ethanol fishy/rancid odour between the start and end of the stor-
extract of P. fucoides showed a higher amount of volatiles age periods. BHT did not show much difference in fishy/
than the control but it was significantly (P < 0.05) lower for rancid odour at the start and end of the storage period in
1-penten-3-one during the entire storage period. either the presence or absence of iron. Thus, the results of
Changes in total tocopherol in emulsions during storage the present study suggest that 50 % ethanolic extracts of
in the presence or absence of iron are shown in Figs. 2d P. fucoides significantly prevented oxidation in 5 % oil-
and 3d, respectively. In the case of the emulsion without in-water emulsions and is as effective as BHT. The higher
added iron, at the first day of storage there was no signifi- oxidation of emulsions containing absolute ethanolic frac-
cant (P > 0.001) difference in the levels of all the tocoph- tions in the presence or absence of iron might be due to
erol homologues between the different treatment groups the interaction of some of the co-extracted components.
(Fig. 2d). However, as the storage progressed a significant Burlakova et al. [49] reported that the ability of phenolic
(P < 0.001) decrease was noticed in the levels of all toco- antioxidants to inhibit lipid oxidation was decreased in the
pherols in different treatment groups except for the emul- presence of phospholipids because of the formation of a
sion with BHT, which showed no significant (P > 0.001) complex between phenolic antioxidants and phospholip-
reduction in tocopherols throughout the storage period. ids. This might be one of the reasons for the pro-oxidative
The 50 % ethanolic extract of P. fucoides was more effec- tendency of the absolute ethanol fraction, as these fractions
tive in preventing loss of tocopherol than other extracts and might contain both phenolic antioxidants and phospholip-
there was no significant (P > 0.05) difference between this ids. Moreover, these absolute ethanol fractions are highly
extract and BHT at the end of the storage period. The con- coloured with lots of pigments, which may also have a pro-
trol and emulsion with an absolute ethanolic extract of P. oxidative effect at higher concentrations. The pro-oxidative
fucoides showed the lowest tocopherol level at the end of effects of pigments at higher concentration have already
the storage period. been established [50].
In the case of the emulsion with added iron, at the first
day of storage there was no significant (P > 0.001) dif- Approximate Chemical Composition and Antioxidant
ference in the levels of all the tocopherol homologues Activity of P. fucoides 50 % Ethanolic Extract Fractions
between the different treatment groups (Fig. 3d). How-
ever, as the storage progressed a significant (P < 0.001) In order to investigate which compounds in the 50 %
decrease was noticed in the levels of tocopherols in all the ethanolic extract were responsible for the highest anti-
groups including BHT. Also in this case the 50 % etha- oxidant activity both in the emulsion and in in vitro anti-
nolic extract of P. fucoides was more effective in prevent- oxidant assays, these extracts were further fractionated
ing loss of tocopherol than other extracts and there was no into the polyphenol-rich fraction (PopRF), polysaccha-
significant (P > 0.05) difference between this extract and ride-rich fraction (PosRF), protein-rich fractions (ProRF)
BHT at the end of the storage period. This was followed and lower-molecular-weight (LMW) aqueous fractions.
by P. fucoides absolute ethanol extracts. The control and The yields of each of these components were 5.0 ± 0.3,
both emulsions with extracts of F. serratus showed the 3.6  ± 0.1, 55.06 ± 2.5 and 20.9 ± 0.2 % respectively.
lowest tocopherol content at the end of the storage period. The total phenolic content of the PopRF was found to be
Tocopherols, through their chromanol moiety, can donate a 45.47  ± 1.52 mg GAE/g of extract, which is higher than
phenolic hydrogen to a lipid peroxy radical to form a reso- in the original 50 % ethanolic extracts. The levels of total
nance-stabilised chromanoxyl radical, which in turn reacts soluble carbohydrates, reducing sugars and soluble protein
with other radicals to form stable adducts and therefore ter- were analysed in the PosRF, ProRF and LMW aqueous
minates the free radical chain reactions [48]. Meanwhile, fractions, as shown in Table 3. Total soluble carbohydrate
tocopherols will be oxidised and consumed during these was found to be high for the LMW aqueous fraction and
processes, which may explain in part the loss of tocopherol this fraction also contains some proteins (Table 3). In order
in these emulsions as the oxidation progressed. to elucidate the composition of these fractions the mono-
Sensory analysis data (Table 2) also fit well with the PV, saccharide and amino acid compositions were also deter-
volatiles and tocopherol data. Fishy/rancid odour increased mined. The monosaccharide composition of both the PosRF
during storage. At the end of storage, the fishy/rancid odour and LMW aqueous fractions are shown in Table 4. When
was lowest for P. fucoides 50 % ethanolic extracts in the compared to PosRF, the LMW aqueous fraction was found
presence and absence of iron. At the end of the storage to contain significantly (P < 0.05) higher content of d-(+)-
period, there was an increase in the fishy or rancid odour fucose, d-(+)- glucose and particularly d-(+)-mannose.

13

582 J Am Oil Chem Soc (2015) 92:571–587

Table 2  The average scores Treatment groups Fishy/rancid Others odour Colour

for odour (attributes were fishy/
rancid and others) for 5 % oil Without iron (0 day)
in water emulsion treated with
 Control 2–3 1–2 (Paint) White
different seaweed extracts
 PAE 0–1 4–5 (Hay/citrous fruit/polish) Yellowish brown
 FAE 0 3–4 (Plant/fruit/seaweed) Yellowish green
 P50E 2–3 2–3 (Paint/polish) White
 F50E 0–1 3 (Bread/citrous fruit) Yellowish with light orange tint
 BHT 2–3 2–3 (Paint/polish) White
8th day
 Control 5 0 White
 P. Fuc AE 1–2 4 (Avacado/synthetic/paper) Yellowish brown
 F. ser AE 2 3 (Plant/fruit/hay) Yellowish green
 P. Fuc 50E 1–2 2–3 (Phenolic/polish) Light brown
 F. ser 50E 0–1 3 (Mango/banana/seaweed) Orangish brown
 BHT 2–3 2–3 (Fruit/hay) White
With iron (0 day)
 Control + Fe 3 2–3 (Forest/polish) White
 PAE + Fe 0–1 4–5 (Paper/fruit/polish) Muddy
 FAE + Fe 1–2 2–3 (Plant/fruit/hay) Yellowish green
 P50E + Fe 0 1 (Forest) Purple brown
 F50E + Fe 0–1 1–2 (Mango/banana) Orangish brown
 BHT + Fe 0 1 (Green) White
96 h
 Control + Fe 4 2–3 (Fruit/paint) White
 PAE + Fe 1–2 4–5 (Fruit/paper) Muddy
 FAE + Fe 4–5 2 (Plant/fruit/hay) Yellowish brown
 P50E + Fe 1–2 2–3 (Paper/polish/hay) Pale purple brown
 F50E + Fe 4 2–3 (Fruit/paint) Orangish brown
 BHT + Fe 0–1 3–4 (Flowery-pelagonia) White

Table 3  Content of total soluble carbohydrates, reducing sugars and soluble proteins in the polysaccharide-rich fraction (PosRF), protein-rich
fraction (ProRF) and low-molecular-weight aqueous fraction (LMWF) from 50 % ethanolic extracts of P. fucoides
Sample Total soluble carbohydrate (g Glu/kg) Reducing sugar (g Glu/kg) Protein (g/kg)

PosRF 96.3 ± 10.5 48.4 ± 1.14 Nd

LMWF 116.03 ± 7.5 48.8 ± 0.6 1.8 ± 0.1
ProRF Nd Nd 22.5 ± 1.7

Each value is expressed as mean ± SD (n = 3)

Glu glucose, Nd not detected

In contrast, the PosRF contained significantly (P < 0.05) in concentration (Fig. 4a). The PopRF showed the highest
higher amounts of d-(−)-arabinose and d-(+)-galactose DPPH radical-scavenging activity (100 % at 1 mg/ml con-
when compared to LMW aqueous fraction (Table 4). centration) and was higher than BHT (77.8 % at 1 mg/ml
The antioxidant activity of the different fractions was concentration). This was followed by the LMW aqueous
determined by the same four in vitro assays as used for fraction and crude protein fractions. The carbohydrate frac-
the original extracts (Fig. 4a–d). Unlike the 50 % etha- tions showed the lowest DPPH radical-scavenging activity.
nolic extracts of P. fucoides (Fig. 1a), the DPPH radical- The results of the present study show that the polyphenol
scavenging activity of the different fractions showed a fraction is the main component responsible for the higher
concentration dependency and increased with an increase radical-scavenging activity of extracts although the other

13
J Am Oil Chem Soc (2015) 92:571–587 583

Table 4  Content of monosacharides in the polysaccharide-rich fraction (PosRF) and low-molecular-weight aqueous fraction (LMWF) of P.
fucoides 50 % ethanolic extracts
µg/mg dry matter
Sample d-(+)- l-(+)- d-(-)- d-(+)- d-(+)- d-(+)- d-(+)- l-Guluronic d-mannuronic
Fucose Rhamnose Arabinose Galactose Glucose Xylose Mannose acid acid

PosRF 1.2 ± 0.1 0 2.6 ± 0.4 6.1 ± 0.2 2.6 ± 0.3 0.8 ± 0.1 0.71 ± 0.0 0.0 0.0

LMWF 2.0 ± 1.3 0.3 ± 0.05 0.22 ± 0.1 3.0 ± 0.2 6.1 ± 0.5 0.0 60.3 ± 2.5 0.0 0.0

Each value is expressed as mean ± SD (n = 3)

Fig. 4  Effect of the concentration of seaweed extracts on (a) the (ProRF), polysaccharide-rich fraction (PosRF) and low-molecu-
DPPH radical-scavenging capacity, (b) Fe2+-chelating capacity, (c) lar-weight aqueous fraction (LMWF) obtained from 50 % ethanol
reducing power, (d) antioxidant activity in the liposome model sys- extracts of P. fucoides. Values are expressed as mean ± SD (n = 3)
tem of the polyphenol-rich fraction (PopRF), protein-rich fraction

components also contribute to radical scavenging. Duan iron-chelating activity between the different fractions and
et al. [51] also reported a higher DPPH radical-scavenging it increased with an increase in concentration (Fig. 4b).
activity for the ethyl acetate extracts and fractions obtained LMW fractions showed the highest iron-chelating activ-
from a red algae Polysiphonia urceolata than BHT. The ity at the high concentration tested. There are also reports
relatively strong scavenging effects of the PopRF on DPPH that among the polysaccharide fractions, the low-molecu-
radicals confirmed the key role of algal polyphenols as free lar-weight fractions had a higher chelating activity than
radical scavengers and primary, chain-breaking antioxi- the higher molecular weight [52]. As the LMW aqueous
dants as reported earlier [18, 32]. fraction is a mixture of small molecules such as amino
The results of iron-chelating activity of fractions show acids, peptides and mono/oligosaccharides, the high iron-
(Fig.  4b) that the iron-chelating activity of the extracts chelating activity of these fractions may stem from these
(Fig. 1b) comes from the combined effect of the chelating components. It seems that the fucose and mannose contents
activity of the different fractions. At the lowest concentra- have a role in the iron-chelating activity as LMW aqueous
tion tested there was no significant difference (P > 0.05) in fractions that contained high fucose and mannose contents

13

584 J Am Oil Chem Soc (2015) 92:571–587

Table 5  Amino acid content of LMW aqueous fractions of P. manner. As PopRF has a relatively lower iron-chelating
fucoides 50 % ethanolic extracts activity, the reducing power dominates, and this explains
Amino acid LMW aqueous fraction (g/100 g) the decrease of inhibition of TBARS formation as the
concentration increases. This also explains the better per-
LYS 0.42 ± 0.1
formance of PosRF than ProRF and LMW fractions in the
ARG –
liposome system as it had a lower reducing property than
ALA 0.86 ± 0.07
these two fractions. Thus, the polyphenol fractions with
C–C 0.02 ± 0.01
higher reducing power were most likely the components
LEU 0.01 ± 0.00 responsible for the decrease in TBARS inhibition as the
MET 0.04 ± 0.00 concentration increased in the in vitro antioxidant tests of
PHE 0.03 ± 0.00 the original 50 % ethanolic extracts of P. fucoides (Fig. 1d).
PRO 0.04 ± 0.02 In order to investigate how these fractions behave in an
THR 3.11 ± 0.05 emulsion, a study was conducted with 500 mg/kg of each of
TYR 0.27 ± 0.01 these fractions in 5 % fish oil-in-water emulsion with added
ASP 0.03 ± 0.01 iron. The effectiveness of these fractions was assessed
SER 0.41 ± 0.02 by measuring the PV, AV and loss of tocopherol, which is
GLU 0.13 ± 0.02 shown in Fig. 5a–c. In general, the oxidation parameters
HYP – were significantly (P < 0.05) lower for emulsions with
VAL 0.54 ± 0.05 PopRF when compared to control and other fractions, and
HIS – it was comparable to the synthetic antioxidant BHT even
TRP 0.02 ± 0.01 in the presence of iron. The emulsions with ProRF and
ILE 0.08 ± 0.0 PosRF showed an increase in PV as the storage time pro-
GLY 0.12 ± 0.02 gressed; however, it was lower than the control throughout
Sum 6.14 ± 0.14 the storage period. Despite the high iron-chelating activity
and DPPH radical-scavenging activity, the low-molecular-
weight fractions showed significantly (P < 0.05) higher
showed good iron-chelating activity at higher concentra- PV, AV and tocopherol loss in 5 % oil-in-water emulsions
tion than PosRF. A higher antioxidant activity of green tea when compared to the control group. It should also be noted
polysaccharide over oogong and black tea polysaccharide that the antioxidant activity of this fraction measured by in
was also reported to stem from their difference in monosac- vitro assays might not accurately reflect the real antioxi-
charide composition [53]. The amino acid composition of dant potentials of LMW peptides/amino acids or the mono-/
the LMW fraction showed high threonine contents (50.7 % oligosaccharides in emulsions because some of these com-
of the total amino acids) followed by Ala (13.9 %), Val ponents may have antioxidative/pro-oxidative behaviours
(8.8 %), Lys (6.8 %), Ser (6.7 %), Tyr (4.3 %) and trace depending upon their location in the emulsion interface or
levels of other amino acids (Table 5). Several amino acids, in the water/lipid phase [61] and also in the presence of iron.
such as His, Phe, Tyr, Met, Lys, Pro and Cys, have been The results of the present study thus confirmed that the
reported to show antioxidant activity [54–57]. The primary phenolic components are the main antioxidants in 50 %
mechanism for this antioxidant activity is believed to be ethanolic extracts of P. fucoides even though other com-
inactivation of free radicals and metal chelation [58]. ponents such as polysaccharides and proteins also con-
The reducing power and inhibition of TBARS forma- tributed to the overall antioxidant activity. These species
tion were highest for the polyphenol-rich fraction (Fig. 4c, of seaweeds are reported to possess bromophenols, which
d). Unlike other fractions, inhibition of TBARS formation are strong antioxidants, especially radical scavengers [37,
was highest at the lower concentration for polyphenols and 38], which might be the main contributor of the antioxidant
it decreased with an increase in concentration (Fig. 4d). effect of these fraction. As only a few studies are available
The PosRF, which showed the lowest reducing power, was on the antioxidative activity of Polysiphonia and most of
found to perform better than the ProRF and LMW frac- them are in in vitro systems, it was difficult to compare our
tion in the liposome system by having higher inhibition of study with others.
TBARS formation. The molar ratios of ferric to ferrous iron
play an important role in the initiation of lipid peroxida-
tion [59, 60]. In the liposome model system, the oxidation Conclusions
is induced by Fe3+/ascorbate, as the PopRF fractions had a
very high reducing power; both ascorbic acid and PopRF This is the first detailed study of antioxidant activity of
might have reduced Fe3+ to Fe2+ in a dose-dependent algal extracts in 5 % oil-in-water emulsions in which the

13
J Am Oil Chem Soc (2015) 92:571–587 585

(a) (b)
60 35
LMWF 30

Anisidine Value
PV (meq/kg oil)
Cont 25 LMWF
40
PosRF 20 Cont
ProRF PopRF
15
20 BHT
10 ProRF
PopRF 5 PosRF
BHT
0 0
0 20 40 60 80 100 0 20 40 60 80 100
Time in hours Time in hours

(c)
Total Tocopherol (ug/g lipid)

25

20 BHT
15 PopRF
ProRF
10 PosRF
Cont
5
LMWF
0
0 12 24 36 48 60 72 84 96
Time in hours

Fig. 5  Effect of the polyphenolrich fraction (PopRF), protein-rich content in 5 % fish oil-in-water emulsions containing iron as an oxi-
fraction (ProRF), polysaccharide-rich fraction (PosRF) and low- dation inducer. Each fraction was tested at a concentration of 500 mg/
molecular-weight aqueous fraction (LMWF) obtained from 50 % kg and BHT at a concentration of 200 mg/kg. Values are expressed as
ethanol extracts of P. fucoides during storage at 20 °C on (a) the per- mean ± SD (n = 3)
oxide value (PV), (b) anisidine value (AV) and (c) total tocopherol

behaviour of these extracts in the presence and absence of ethanolic extracts of P. fucoides could be potential rich
iron was investigated. This study also provided information sources of natural antioxidants for food applications. How-
about the fractions responsible for the higher antioxidant ever, as we only studied simple phenols, further studies are
activity of the potent extracts. The results of the present needed to characterise the phenolic compounds respon-
study demonstrated that the type of extractant and the spe- sible for this higher antioxidant activity and to investigate
cies had a great impact on both the phenolic content and whether they have any positive or negative health effects.
antioxidant activity of seaweed extracts. It was found that
P. fucoides had better antioxidant activity than F. serratus. Acknowledgments  This work was financially supported by the
Danish Research Council for Technology and Production. The help
Although the phenolic compounds were extracted better provided by the technician Inge Holmberg with the HPLC analy-
with absolute ethanol extracts, 50 % ethanol extracts were sis of amino acids, Anis Arnous with the monosaccharide analy-
found to have better antioxidant properties in emulsions. sis and Susan Løvstad Holdt with collecting seaweeds is greatly
Absolute ethanol extracts showed a pro-oxidative tendency acknowledged.
in the presence or absence of iron, at least in the initial
part of the storage period; 500 mg/kg of 50 % ethanolic
extracts of P. fucoides was shown to have antioxidant activ- References
ity both in in vitro and in emulsions and is comparable to
200 mg/kg of the synthetic antioxidant BHT. Further frac- 1. Frankel EN (2005) Lipid oxidation. The Oily Press, Dundee
2. Yagi K (1987) Lipid peroxides and human diseases. Chem Phys
tionation of 50 % ethanolic extracts from P. fucoides fol- Lipids 45:337–351
lowed by evaluation of antioxidant efficacy demonstrated 3. Witztum JL (1994) The oxidation hypothesis of atherosclerosis.
that the phenolic fractions were the components responsi- Lancet 344:793–795
ble for the higher oxidation protection of 50 % ethanolic 4. Schwartz JL (1996) The dual roles of nutrients as anti-oxidants
and pro-oxidants: their effects on tumour cell growth. J Nutri
extracts of P. fucoides. However, the other components 126:1221S–1227S
also contributed to the antioxidative effects to some extent. 5. Frankel EN (1998). Antioxidants. Lipid oxidation, 1st edn. The
Hence, based on these results, we can conclude that 50 % Oily Press, Dundee, pp 129–166

13

586 J Am Oil Chem Soc (2015) 92:571–587

6. Linderschmidt R, Trylka A, Goad M, Witschi H (1986) The 28. Bradford MM (1976) A dye binding assay for protein. Anal Bio-
effects of dietary butylated hydroxytoluene on liver and colon chem 72:248–254
tumor development in mice. Toxicol 38:151–160 29. Masuko T, Minami A, Iwasaki N, Majima T, Nishimura S, Lee
7. Branen AL (1975) Toxicology and biochemistry of butyl- YC (2005) Carbohydrate analysis by a phenol-sulfuric acid
ated hydroxytoluene and butylated hydroxyanisole. J Am Oil method in microplate format. Anal Biochem 339:69–72
Chem Soc 52:59–63 30. Miller GL (1959) Use of dinitrosalicylic acid reagent for determi-
8. Ito N, Fukushima S, Hasebawa A (1983) Carcinogenicity of BHA nation of reducing sugar. Anal Chem 31:426–428
in F344 rats. J Natl Cancer Inst 70:343–352 31. Farvin KHS, Baron CP, Nielsen NS, Otte J, Jacobsen C (2010)
9. Fujihara M, Nagumo T (1993) An influence of the structure of Antioxidant activity of yoghurt peptides: part 2—characterisation
alginate on the chemotactic activity of macrophages and the anti- of peptide fractions. Food Chem 123:1090–1097
tumor activity. Carbohydr Res 243:211–216 32. Wang T, Jónsdóttir R, Ólafsdóttir G (2009) Total phenolic com-
10. Konig GM, Wright AD, Sticher O, Anghofer CK, Pezutto JM pounds, radical scavenging and metal chelation of extracts from
(1994) Biological activities of selected marine natural products. Icelandic seaweeds. Food Chem 116:240–248
Planta Medica 60:532–537 33. Ragan MA, Glombitza KW (1986) Physodes and the phenolic
11. Santoso J, Yoshie Y, Suzuki T (2002) The distribution and profile compounds of brown algae 4. Oligomeric polyphloroglucinols
of nutrients and catechins of some Indonesian seaweeds. Fish Sci from Fucus vesiculosus -Photoplate mass-spectrometric investi-
68:1647–1648 gation. Phytochem 21:2709–2711
12. Yoshie-Stark Y, Wang W, Petillo D, Suzuki T (2000) Distribution 34. Marxen K, Vanselow KH, Lippemeier S, Hintze R, Ruser A,
of catechins in Japanese seaweeds. Fish Sci 66:998–1000 Hansen UP (2007) Determination of DPPH radical oxidation
13. Yoshie-Stark Y, Hsieh YP, Suzuki T (2003) Distribution of flavo- caused by methanolic extracts of some microalgal species by
noids and related compounds from seaweeds in Japan. J Tokyo linear regression analysis of spectrophotometric measurements.
Univ Fish 89:1–6 Sensors 7:2080–2095
14. Ahn GN, Kim KN, Cha SH, Song CB, Lee J, Heo MS, Yeo IK, 35. Zhang Q, Zhang J, Shen J, Silva A, Dennis DA, Barrow CJ
Lee NH, JeeYH Kim JS, Heu MS, Jeon YJ (2007) Antioxidant (2007) A simple 96-well microplate method for estimation of
activities of phlorotannins purified from Ecklonia cava on free total polyphenol content in seaweeds. Eighteenth international
radical scavenging using ESR and H2O2-mediated DNA damage. seaweed symposium. Develop App Phycol 1:219–224
Euro Food Res Technol 226:71–79 36. Brand-Williams W, Cuvelier ME, Berset C (1995) Use of a free
15. Rupérez P, Ahrazem O, Leal JA (2002) Potential antioxi-
radical method to evaluate antioxidant activity. Food Sci Technol
dant capacity of sulfated polysaccharides from the edible 28:25–30
marine brown seaweed Fucus vesiculosus. J Agric Food Chem 37. Fujimoto K, Ohmura H, Kaneda T (1985) Screening for antioxy-
50:840–845 genic compounds in marine algae and bromophenols as effective
16. Kobayashi M, Sakamoto Y (1999) Singlet oxygen quenching
principles in a red algae Polysiphonia ulceolate. Bull Jap Soc Sci
ability of astaxanthin esters from the green algae Haematococcus Fish 51:1139–1143
pluvialis. Biotechnol Lett 21:265–269 38. Yan X, Nagata T, Fan X (1998) Antioxidative activities in some
17. Yan XJ, Chuda Y, Suzuki M, Nagata T (1999) Fucoxanthin as the common seaweeds. Plant Food Hum Nutr 52:253–262
major antioxidant in Hizikia fusiformis, a common edible sea- 39. Toth G, Pavia H (2000) Lack of phlorotannin induction in the
weed. Biosci Biotech Biochem 63:605–607 brown seaweed Ascophyllum nodosum in response to increased
18. Farvin KHS, Jacobsen C (2013) Phenolic compounds and antiox- copper concentrations. Mar Ecol Prog Ser 192:119–126
idant activities of selected species of seaweeds from Danish coast 40. Kuda T, Tsunekawa M, Hishi T, Araki Y (2005) Antioxidant prop-
138:1670–1681 erties of dried ‘kayamo-nori’, a brown alga Scytosiphon lomen‑
19. Gülçin I` (2006) Antioxidant activity of caffeic acid (3,4-dihy- taria (Scytosiphonales, Phaeophyceae). Food Chem 89:617–622
droxycinnamicacid). Toxicol 217:213–220 41. Rice-Evans CA, Miller NJ, Paganga G (1996) Structure-antioxi-
20. Sørensen A, Haahr A, Becker EM, Skibsted LH, Bergenståhl B, dant activity relationships of flavonoids and phenolic acids. Free
Nilsson L, Jacobsen C (2008) Interactions between iron, phenolic Rad Biol Med 20:933–956
compounds, emulsifiers, and ph in omega-3-enriched oil-in-water 42. Wang BG, Zhang WW, Duan XJ, Li XM (2009) In vitro anti-
emulsions. J Agric Food Chem 56:1740–1750 oxidative activities of extract and semi-purified fractions of the
21. Singleton VL, Rossi JA (1965) Colourimetry of total phenolics marine red alga, Rhodomela confervoides (Rhodomelaceae).
with phosphomolybdic-phosphotungtic acid reagents. Am J Enol Food Chem 113:1101–1105
Viticult 16:144–158 43. Dorman HJD, Kosar M, Kahlos K, Holm Y, Hiltunen R (2003)
22. Onyeneho SN, Hettiarachchy NS (1993) Antioxidant activity,
Antioxidant properties and composition of aqueous extracts from
fatty acid and phenolic acid composition of potato peels. J Sci Mentha species, hybrids, varieties, and cultivars. J Agric Food
Food Agric 62:345–350 Chem 51:4563–4569
23. Farvin KHS, Andersen LL, Nielsen HH, Jacobsen C, Jakob- 44. Jiménez-Escrig A, Jiménez-Jiménez I, Pulido R, Saura-Calixto
sen G, Johansson I, Jessen F (2014) Antioxidant activity of F (2001) Antioxidant activity of fresh and processed edible sea-
Cod (Gadusmorhua) protein hydrolysates: in vitro assays weeds. J Sci Food Agric 81:530–534
and evaluation in 5% fish oil-in-water emulsion. Food Chem 45. Milo C, Grosch W (1996) Changes in the odorants of boiled
149:326–334 salmonand cod as affected by the storage of the raw material. J
24. Bligh EG, Dyer WJ (1959) A rapid method of total lipid extrac- Agric Food Chem 44:2366–2371
tion and purification. Can J Biochem Physiol 37:900–917 46. Venkateshwarlu G, Let MB, Meyer AS, Jacobsen C (2004) Mod-
25. International IDF standards (1991) Section 74A: 1991. Interna- eling the sensory impact of defined combinations of volatile lipid
tional Dairy Federation, IDF-square Vergot 41. Brussels, Belgium oxidation products on fishy and metallic off-flavors. J Agric Food
26. AOCS (1992) official method Ce 8-89. Determination of toco- Chem 52:1635–1641
pherols and tocotrienols in vegitable oils and fats by HPLC. 47. Hartvigsen K, Lund P, Hansen LF, Hølmer G (2000) Dynamic
Champaign, AOCS headspace gas chromatography/mass spectrometry characteriza-
27. Staub AM (1965) Removal of proteins: sevag method. Methods tion of volatiles produced in fish oil enriched mayonnaise during
Carbohydr Chem 5:5–6 storage. J Agric Food Chem 48:4858–4867

13
J Am Oil Chem Soc (2015) 92:571–587 587

48. Kamal-Eldin A, Appelqvist L (1996) The chemistry and anti- 55. Marcuse R (1960) Antioxidative effect of amino acids. Nature
oxidant properties of tocopherols and tocotrienols. Lipids 186:886–887
31:671–701 56. Karel M, Schaich K, Roy RB (1975) Interaction of peroxidising
49. Burlakova EB, Mazaletskaya LI, Sheludchenko NL, Shishkina methyl linoleate with some proteins and amino acids. J Ag Food
LN (1995) Inhibitory effect of the mixtures of phenol antioxi- Chem 23:159–165
dants and phosphatidylcholine. Russ Chem Bull 44:1020–1041 57. Karel M, Tannenbaum SR, Wallace DH, Maloney H (1966) Anti-
50. El-Agamey A, Lowe GM, McGarvey DJ, Mortensen A, Phillip oxidation of methyl linoleate in freeze-dried model systems. III.
DM, Truscott TG, Young AJ (2004) Carotenoid radical chemistry Effects of added amino acids. J Food Sci 31:892–896
and antioxidant/pro-oxidant properties. Arch Biochem Biophys 58. Decker EA (1998) Antioxidant mechanisms. In: Akoh CC, Min
430:37–48 DB (eds) Food lipids: chemistry, nutrition and biotechnology.
51. Duan XJ, Zhang WW, Li XM, Wang BG (2006) Evaluation of Marcel Dekker, New York, pp 397–421
antioxidant property of extract and fractions obtained from a red 59. Miller DM, Aust SD (1989) Studies of ascorbic—depend-

alga, Polysiphonia urceolata. Food Chem 95:37–43 ent, iron catalysed lipid peroxidation. Arch Biochem Biophys
52. Qi H, Zhao T, Zhang T, Li Z, Zhao Z, Xing R (2005) Antioxi- 271:113–119
dant activity of different molecular weight sulfated polysaccha- 60. Minotti G, Aust SD (1987) The requirement of iron (III) in the
rides from Ulvapertusa Kjellm (Chlorophyta). J Appl Phycol initiation of lipid peroxidation by iron (II) and hydrogen perox-
17:527–534 ide. J Biol Chem 262:1098–1104
53. Chen H, Qu Z, Fu L, Dong P, Zhang X (2009) Physicochemi- 61. Mcclements DJ, Decker EA (2000) Lipid oxidation in oil-

cal properties and antioxidant capacity of 3 polysaccharides from in-water emulsions: impact of molecular environment on
green tea, oolong tea, and black tea. J Food Sci 74:469–474 chemical reactions in heterogeneous food systems. J Food Sci
54. Marcuse R (1962) The effect of some amino acids on the oxi- 65:1270–1282
dation of linoleic acid and its methyl ester. J Am OilChemSoc.
39:97–103

13