Characterization of ulvan extracts to assess the effect of different steps in the extraction procedure

Article abstract of DOI:10.1016/j.carbpol.2011.12.041

An effective application development of the polysaccharide ulvan requires a comprehensive knowledge about the influence of the extraction process on composition of the extracts and in ulvan itself. In this context, the two main objectives of the present work are (1) the establishment of an efficient extraction process for ulvan and (2) development of an accurate characterization methodology to evaluate the extract composition and ulvan content. Three ulvan-rich extracts obtained by different schemes of extraction were studied. The methodology for the analysis was improved and a detailed analysis of extracted ulvan was provided. The polysaccharide is rich in ulvanobiuronic acid 3-sulfate type A [→4)-β-d-GlcAp-(1 → 4)-α-l-Rhap 3S-(1→], with minor amounts of ulvanobiuronic acid 3-sulfate type B [→4)-α-l-IdoAp-(1 → 4)-α-l-Rhap 3S-(1→]. The extract with the higher degree of purification is a high molecular weight polysaccharide (790 kDa) composed of rhamnose (22.4%), glucuronic acid (22.5%), xylose (3.7%), iduronic acid (3.1%) and glucose (1.0%). It is highly sulfated (32.2%) and contains 1.3% of proteins and 10.3% of inorganic material. Applying simple extraction scheme it was possible to obtain an extract from green algae with high content of ulvan without affecting the overall chemical structure of the polysaccharide.

Full text of DOI:10.1016/j.carbpol.2011.12.041

Carbohydrate Polymers 88 (2012) 537–546
Contents lists available at SciVerse ScienceDirect
Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol
Characterization of ulvan extracts to assess the effect of different steps in the
extraction procedure
Carina Costaa, Anabela Alvesb,c,∗, Paula R. Pintoa,∗∗, Rui A. Sousab,c, Eduardo A. Borges da Silvaa,
Rui L. Reisb,c, Alírio E. Rodriguesa
a Laboratory of Separation and Reaction Engineering – LSRE, Associate Laboratory LSRE/LCM, Department of Chemical Engineering, Faculty of Engineering,
University of Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal
b 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, 4806-909 Taipas, Guimarães, Portugal
c ICVS/3B’s – PT Associated Laboratory, Guimarães, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
An effective application development of the polysaccharide ulvan requires a comprehensive knowledge
about the influence of the extraction process on composition of the extracts and in ulvan itself. In this
context, the two main objectives of the present work are (1) the establishment of an efficient extraction
process for ulvan and (2) development of an accurate characterization methodology to evaluate the
extract composition and ulvan content.
Received 30 September 2011
Received in revised form
22 December 2011
Accepted 24 December 2011
Available online 3 January 2012
Three ulvan-rich extracts obtained by different schemes of extraction were studied. The methodology
for the analysis was improved and a detailed analysis of extracted ulvan was provided. The polysaccharide
is rich in ulvanobiuronic acid 3-sulfate type A [→4)-␤-d-GlcAp-(1 → 4)-␣-l-Rhap 3S-(1→], with minor
amounts of ulvanobiuronic acid 3-sulfate type B [→4)-␣-l-IdoAp-(1 → 4)-␣-l-Rhap 3S-(1→]. The extract
with the higher degree of purification is a high molecular weight polysaccharide (790 kDa) composed
of rhamnose (22.4%), glucuronic acid (22.5%), xylose (3.7%), iduronic acid (3.1%) and glucose (1.0%). It is
highly sulfated (32.2%) and contains 1.3% of proteins and 10.3% of inorganic material. Applying simple
extraction scheme it was possible to obtain an extract from green algae with high content of ulvan without
affecting the overall chemical structure of the polysaccharide.
Keywords:
Ulvan
Green algae
Ulva lactuca
Extraction
Purification
Polysaccharide characterization
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
polymers for different technological and industrial-related appli-
cations is increasing. These are believed to be a better alternative
The emergent attention on renewable resources for chemi-
cals and polymers has been leading to a special focus on marine
biomass, in particular on green algae. Ulvan is one of the main
polysaccharides composing the algae belonging to Ulvaceae genera,
representing a potential source to be explored. Ulvan has already
proven to be a remarkable polysaccharide with different attractive
properties that make it suitable for a wide range of applications
(Castro et al., 2006; Castro, Zarra, & Lamas, 2004; Cluzet et al.,
2004; Harada & Maeda, 1998; Kaeffer, Benard, Lahaye, Blottiere, &
Cherbut, 1999; Mao, Zang, Li, & Zhang, 2006; Pengzhan, Ning, et al.,
2003; Pengzhan, Quanbin, et al., 2003; Qi, Zhang, et al., 2005; Qi,
Zhao, et al., 2005; Warrand, 2006). In fact, the relevance of natural
to the use of fully synthetic materials in various contexts (Mano
et al., 2007).
The major repeating structure (designated as aldobiuronic acid)
of ulvan is the ulvanobiuronic acid 3-sulfate composed by O-3-
sulfate rhamnose and glucuronic acid (type A, A3S). In other typical
repeating structure, iduronic acid replaces glucuronic acid (type B,
B3S) (Lahaye, Brunel, & Bonnin, 1997; Lahaye, Inizan, & Vigouroux,
1998):
A3S : → 4)␤-D-GlcAp(1 → 4)-␣-L-Rhap 3S(1 →
B3S : → 4)␣-L-IdoAp(1 → 4)-␣-L-Rhap 3S(1 →
(1)
(2)
where p is a indicative of a pyranoside form of the compound.
Xylose (O-2 sulfated or not) can also occur linked to sulfated rham-
nose. The designation are ulvanobioses U3S and U2ꢀS,3S (Lahaye,
1998):
∗
Corresponding author. Tel.: +351 2 53 51 09 00.
Corresponding author. Tel.: +351 2 20 41 36 06; fax: +351 2 25 08 14 49.
E-mail addresses: anabela.pinto@dep.uminho.pt (A. Alves), ppinto@fe.up.pt
∗∗
U3S : → 4)␤-D-Xylp(1 → 4)␣-L-Rhap 3S(1 →
(3)
(P.R. Pinto).
0144-8617/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2011.12.041

538
C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
U2ꢀS,3S : → 4)ˇ-D-Xylp 2S(1 → 4)␣-L-Rhap 3S(1 →
(4)
Other repeating structures (involving the same residues), some
of them branched with GlcA, are reported in the literature as
belonging to ulvan structure (Lahaye, 1998; Lahaye & Robic, 2007).
Glucuronans, xyloglucans, cellulose and proteins are cell-wall com-
ponents found in close association with ulvan (Lahaye, Jegou, &
Buleon, 1994; Lahaye & Ray, 1996; Lahaye & Robic, 2007). In
the extractive process of ulvan, these associations would lead
to co-extraction of a fraction of these components, affecting the
purity in ulvan and the biological responses of the extract (Cannell
et al., 1998; Hernández-Garibay, Zertuche-González, & Pacheco-
Ruíz, 2011). In this context, accurate analysis of the carbohydrate
composition of the extracts is required.
Concerning the chemical characterization, the depolymeriza-
tion of polysaccharides is often performed by acid-catalyzed
hydrolysis of glycosidic bonds, as Saeman hydrolysis (Kamerling &
Gerwig, 2007; Saeman, 1945). However, this procedure is not effec-
tive for linkages such as aldobiuronic (Conrad, 1980; McKinnell
& Percival, 1962) due to the high stability of glycosidic linkages
involving uronic acids (Dahlman, Jacobs, Liljenberg, & Olsson, 2000;
De Ruiter, Schols, Voragen, & Rombouts, 1992) and monosaccha-
rides, once released, are highly susceptible to further degradation
(Biermann, 1988; Willför et al., 2009). An alternative is the depoly-
merization by acid methanolysis (HCl in anhydrous methanol)
(De Ruiter et al., 1992; Sundberg, Sundberg, Lillandt, & Holmbom,
1996). In this process, both neutral and acid monosaccharides are
released and converted to methyl glycosides, while the carboxyl
groups of uronic acids are converted to methyl esters. One advan-
tage of this method is their good performance in the cleavage of
glucuronosyl bonds and the improvement of the stability of result-
ing monosaccharides (Willför et al., 2009). Besides this benefit, it
is still possible to detect and quantify some resistant disaccha-
ride. The improvement of conditions (time, temperature and HCl
concentration) in order to obtain the maximum yield has been
discussed in the literature concerning commercial gelling algae
(Quemener, Lahaye, & Metro, 1998) and lignocellulosic polysac-
charides (Bertaud, Sundberg, & Holmbom, 2002; Sundberg et al.,
1996).
In general, ulvan is extracted with hot aqueous solutions (Lahaye
& Robic, 2007); however, the final step for removing contaminants
may vary considerably. One of the objectives of the present work
is the validation of a previously developed extraction methodology
to obtain ulvan from green algae (Alves, Caridade, Mano, Sousa,
& Reis, 2010). For this, the composition of the extracts produced
by the three schemes of extract preparation is assessed, as well
as the structural characteristics of the polysaccharide fraction. The
method of acid methanolysis was improved and tailored for the
maximum yield on typical monosaccharides for the analysis of
ulvan-rich extracts and results were compared with those from
enzymatic hydrolysis. Complementary HPSEC and 13C NMR spec-
troscopic data on the extracts were provided.
Fig. 1. Scheme of the ulvan extraction from green algae Ulva lactuca.
100 g dry algae per 3000 mL of water, for a total of 7 h (3 h the
first extraction and 2 h the following two extractions) at 75–85 ◦C,
on a boiling-water bath, under continuous stirring. After filtration
through a cotton cloth, the aqueous extracts were centrifuged, and
the liquid supernatant was filtered. The water extract was concen-
trated until 10–20% of its initial volume, in a rotary evaporator.
The derived solution was then subjected to enzymatic hydroly-
sis to remove contaminating starch and proteins. Afterwards, the
solution was decolorized and deodorized by adsorption on acti-
vated charcoal. The water extract was centrifuged, filtered and
precipitated with 4 vol. of absolute ethanol. Finally, the recov-
ered precipitate was freeze dried. Some deviations to this general
method are introduced to obtain different ulvan extracts (referred
to as samples) to be further characterized. A minimum of three
extractions were performed for each sample.
Sample I: crude ulvan is obtained by a reduction in the general
method, i.e., without further purification. In this sense, after Soxhlet
extraction of dried U. lactuca followed by hot-water extraction, the
water extract was concentrated until 10–20% of its initial value, in
a rotary evaporator, precipitated and freeze dried.
Sample II: obtained by applying the general procedure without
the use of an enzyme to remove starch.
Sample III: obtained by applying the general procedure. Enzy-
matic hydrolysis is achieved by applying both ␣-amylase and
proteinase K to remove starch and proteins, respectively.
A representative scheme of the ulvan extraction to obtain the
three samples is shown in Fig. 1. Algae used on this study belong
to U. lactuca species originated from the same batch, in order to
eliminate any variability associated with the biomass source.
2.2. Ash content
Ash content was determined by thermogravimetric analysis
(TGA). For TGA, a small amount (10–15 mg) of powder was taken
for analysis in a Perkin Elmer TGA7. The samples were heated from
30 to 900 ◦C at a rate of 20 ◦C/min in air atmosphere.
2. Materials and methods
2.1. Extraction of ulvan polysaccharide from green algae Ulva
lactuca
2.3. Quantification of sulfates
A general procedure was employed to extract ulvan, based on
the procedure developed by Alves et al. (2010).
Sulfate groups were released after hydrolysis of 10–12 mg of
polysaccharide ulvan with 3 mL trifluoroacetic acid (TFA, 99% Pro-
labo), 2 M, for 3 h at 100 ◦C, according to the method described by
Ray and Lahaye (1995). For sulfate groups in extract ashes, a sam-
ple of 10 mg was dissolved in 10 mL of deionized water with 0.1 mL
of TFA. Sulfate quantification was performed by using a Dionex
ion chromatograph (Dionex DX-120) equipped with a suppressed
conductivity detector. The column was a DionexIonpac AS9-HC
All reagents were supplied by Sigma–Aldrich. Green algae U. lac-
tuca were supplied by Setalg (France). General procedure: Soxhlet
extraction of dried U. lactuca with dichloromethane and ethanol
removed most of the lipids and coloring matter which otherwise
would contaminate the water extract. The residual off-white weed
was dried and subjected to three hot-water extractions, typically

C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
539
(250 mm × 4 mm I.D.) operated at room temperature. The eluent
thermally deactivated as above described. Control experiments
were performed to evaluate the presence of monosaccharides as
contaminant in enzymatic formulations.
was 9 mM Na2CO3 aqueous solution at a flow rate of 1.0 mL/min.
2.4. Protein quantification
The resulting hydrolysates were freeze-dried and analyzed by
GC-FID and GC–MS after trimethylsilylation as describe in Section
2.7.3. Myo-inositol (≥99.0% Sigma–Aldrich) was used as internal
standard (1 mL of a methanolic solution 0.1 mg/mL).
Protein present in the extracts was quantified by a colorimetric
protein assay, using BCATM Protein Assay Kit (Pierce), follow-
ing manufacturer instructions. Briefly, polysaccharide solution was
exposed to a working solution containing bicinchoninic acid (BCA)
as detection reagent for Cu1+ and optical density, at 562 nm, was
measured using a microplate reader (Redinbaugh & Turley, 1986). A
standard curve was created and sample protein values were read off
from the standard graph. Albumin solutions were used as standards
(0–2000 ␮g/mL).
2.7.2. Acid methanolysis: kinetic and improvement of the method
The ulvan extracts (10–12 mg, dried in vacuum over phos-
phoric pentoxide (≥98.5%, Sigma–Aldrich) were submitted to
acid methanolysis as described in the literature (Sundberg et al.,
1996) in 2 mL of HCl 2 M prepared by dilution with anhydrous
methanol (>99%, Sigma–Aldrich) of a commercial solution 3 M HCl
in methanol (Supelco). Methanolysis reaction was performed at
100 ◦C, initially for 4 h. For kinetic of acid methanolysis of ulvans in
two HCl concentrations, 2 M and 3 M, the reaction was carried out
for 3, 4, 6, 24, 48 and 72 h at 100 ◦C. The final goal was to establish
the parameters leading to the maximum released of monosaccha-
rides with minimal degradation. The analyses were performed as
described in Section 2.7.3.
2.5. Elemental analysis
Carbon (C), hydrogen (H), sulfur (S), nitrogen (N) and oxygen
(O) contents were quantified using a Thermo Scientific Flash 2000
Organic Elemental Analyser.
2.6. High pressure size exclusion chromatography (HPSEC)
2.7.3. Analysis of the partially methylated monosaccharide:
identification and quantification
Ulvan extracts, about 4 mg, were dissolved in 1 mL of elu-
ent solution −50 mM NaNO3 (99%, Panreac) containing 0.02%
NaN3 (99%, Sigma–Aldrich). The solutions (filtered by membrane
0.22 ␮m) were analyzed on a HPLC (Gilson, France) equipped with
a Shodex OH-pack SB HQ 804 (Shodex, Japan) and a refractive index
(RI) detector (model 131, Gilson). Elution was performed at 25 ◦C
with a solution of 50 mM NaNO3 (0.02% NaN3) at a flow rate of
0.7 mL/min. Pullulans (PSS, Germany) with molecular weight (Mw)
values between 342 Da and 805 kDa were the standards used for
calibration of RI signal. A linear relationship between retention time
of standards with the logarithmic function of Mw was obtained
(r2 = 0.982). Protein elution was followed in parallel by ultra-violet
(UV) detector (model 117, Gilson) at 280 nm.
After acid methanolysis, pyridine (≥99.0% Sigma–Aldrich) was
added to methanolysates to neutralize the remaining HCl, followed
by addition of 1 mL of sorbitol (>98.0%, Sigma–Aldrich) methano-
lic solution (0.1 mg/mL) and evaporation to dryness in a rotavapor.
Then, the dried methanolysates were dissolved in 150 ␮L of pyri-
dine, 150 ␮L of bis(trimethylsilyl)-trifluoroacetamide (≥99% Fluka)
and 50 ␮L of chlorotrimethylsilane (≥99.0% Fluka), maintaining the
reaction mixture at 80 ◦C for 30 min.
For quantification, the products were analyzed by gas chro-
matography with flame ionization detector (GC-FID) on a DANI
GC 1000 chromatograph, with a capillary column ValcoBond VB1
(30 m × 0.32 mm I.D., 0.25 ␮m film thickness), using the following
temperature program: 100–175 ◦C at 3 ◦C/min and 175–290 ◦C at
8 ◦C/min. The temperature of the detector and the injector were
kept at 290 ◦C and 260 ◦C, respectively. External calibration was
performed for rhamnose (99%, Merk), xylose (99%, Merk), glu-
curonic acid (98%, Alfa Aesar), glucose (99.5%, Sigma–Aldrich) and
iduronic acid (98%, Carbosynth, UK). The GC–MS analyses were per-
formed in a Trace Gas Chromatograph 2000 Series equipped with
a Finnigan Trace MS mass spectrometer (EI), using helium as car-
ries gas (35 cm/s). The chromatographic conditions, including the
column, were the same as described for GC-FID with a transfer-line
temperature of 290 ◦C and split ratio of 1:100.
2.7. Assessing to the carbohydrate fraction of the ulvan extracts
2.7.1. Enzymatic approach
Two different approaches using enzymatic preparations were
tested aiming to obtain the maximum yield of released
monosaccharides: 1. Combining partial acid-hydrolysis with and
␤-glucuronidase from Helix pomatia (G-0751, Sigma–Aldrich) –
chemical-enzymatic method; 2. Applying a pool of enzymes, kindly
supplied by Novozymes (North America), containing: 25% cellulase
complex (NS22074, 1.0 EGU/g), 25% xylanase (NS22036, 1.0 FXU/g),
25% ␤-glucosidase (NS50010, 250 CBU/g) and 25% ␤-glucanase and
xylanase mixture (NS22002, 45 FBG/g). The activities are expressed
as EGU – Endo-Glucanase Unit, FXU – Fungal Xylanase Unit, CBU –
Cellobiase Unit and FBG – Fungal ␤-Glucanase Unit. The enzymatic
mixture is hereafter designated as EM.
Acid pre-hydrolysis and enzymatic hydrolysis with ␤-
glucuronidase was performed as described in the literature
(Quemener, Lahaye, & Bobin-Dubigeon, 1997) with some mod-
ifications. Partial hydrolysis of ulvan extracts, about 10 mg, was
carried out with 3 mL trifluoroacetic acid, 2 M, for 45 min at 120 ◦C.
The pre-hydrolysate was evaporated at low pressure and tem-
perature, resuspended in 5 mL of acetate buffer pH 5 containing
500–600 U of ␤-glucuronidase and incubated at 37 ◦C for 16 h. The
reaction was stopped by immersing the reaction tubes in boiling
water for 5 min.
Methanolysis of polysaccharides leads to the cleavage of the gly-
coside bonds with the production of the ␣- and ␤-anomers and
pyranose (p) and furanose (f) ring forms of the monosaccharides.
The identification was performed based on the number of glycoside
peaks, their relative retention time and signal intensity proportion
(Bleton, Mejanelle, Sansoulet, Goursaud, & Tchapla, 1996; Doco,
O’Neill, & Pellerin, 2001), and mass spectra of trimethylsilyl (TMS)
methyl glycoside derivatives using spectral data reported in the
literature (Bleton et al., 1996; Doco et al., 2001) and standards.
2.7.4. 13C NMR
13C NMR spectra were recorded on a BRUKER AVANCE III 400
spectrometer operating at 400 MHz. 30 mg of dried ulvan sample
were dissolved in 1.0 mL of D2O (99.9%) and data were acquired at
100.62 MHz for 13C at 70 ◦C. Standard parameters for 13C NMR mea-
surements were used: simple 1D pulse sequence, recycling time
of 2.0 s, spectral width of 250 ppm, and 1D sequence with power-
gated decoupling using 30◦ flip angle.
For the second approach, the EM supplied by Novozymes was
used as received. Each sample (20 mg) and the same weight of
the EM were dissolved in sodium acetate buffer pH 5 (4 mL). The
hydrolysis was carried out at 50 ◦C and after 5 h the enzymes were

540
C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
Table 1
9.0
RI
Non-carbohydrate components of ulvan-rich extracts (as % w/w dry extract mate-
rial) and elemental analysis.
6.0
3.0
Parameter
Sample
I
II
III
Ashes
Total
Sulfate-free
Sulfate (as SO3
Proteins
//
0.0
43.2 0.6
17.5 1.2
29.9 1.1
2.4 0.2
38.3 1.4
17.0 1.2
25.9 1.5
1.1 0.1
26.2 0.3
10.3 0.5
32.2 0.4
1.3 0.1
200
−
UV
)
150
100
50
Elemental analysis
%O
%N
%C
%H
%S
48.7
1.2
16.5
3.5
49.6
0.4
18.8
4.0
40.5
1.0
22.8
4.3
0
11.7
11.0
8.4
0
6
8
10
12
14
16
18
20
Data are presented as mean standard deviations (SD) of triplicate essays.
Values of elemental analysis: average of two essays.
Time (min)
Fig. 2. Typical HPSEC profile of ulvan-rich extracts (presented for sample III). Refrac-
tive index (RI) and ultraviolet (280 nm) detection (UV).
3. Results and discussion
were identified by Scanning Electron Microscopy-Energy Disper-
sive Spectroscopy (SEM-EDS) and include sodium, magnesium and
potassium (data not shown).
The main objective of the present research work was to better
understand the influence of the different steps of the extraction
methodology on composition and characteristics of ulvan. For this,
different ulvan extracts were obtained and characterized, taking
into consideration their non-carbohydrate and carbohydrate frac-
tions.
Proteins have already been described as potential contaminants
of cell wall polysaccharides, mostly because they are part of the
structure of cell walls and closely associated with polysaccharides
(Robic, Rondeau-Mouro, Sassi, Lerat, & Lahaye, 2009; Robic, Sassi,
& Lahaye, 2008). This is clear by the presented results. In sample I
(minimal purification), 2.4% of proteins are co-extracted with ulvan.
However, the extract was deproteinased to some extent by enzy-
matic hydrolysis with proteinase K, as can be seen for samples II
and III (about 1% of protein).
Concerning the elemental analysis results shown in Table 1 and,
in particular, the elements N and S, the values follow the same
pattern of variation between samples as described for protein and
sulfate, respectively.
3.1. Non-carbohydrate fraction of the extracts
As part of any characterization work on natural origin polysac-
charides obtained by extraction process involving different steps,
one has to be aware of the presence of different non-carbohydrate
components of the polysaccharide, namely inorganic compounds
and protein. Table 1 describes the percentages of non-carbohydrate
fraction determined for the ulvan samples. The results of elemental
analysis are also presented.
A common feature among all samples was the high content of
inorganic material quantified as ash (26–43% of dry extract). The
ash content clearly decreases with the increase of the number of
steps in the extraction procedure: from sample I, with minimal
or without purification (43%), to sample III, with two purification
steps (26%). It is recurrently reported in the literature the high con-
tent of ash in polysaccharides of marine origin, namely for sulfated
polysaccharides, including ulvan (Lahaye & Jegou, 1993; Pengzhan,
Quanbin, et al., 2003; Quemener et al., 1997). Considering that ulvan
is a sulfated polysaccharide, the sulfate content of ashes was quan-
tified and the results shown that more than an half of ashes weight
are composed by sulfates. As result, the sulfate-free ash content
was found to be in range 10–18% of extract weight; therefore, the
high content of inorganics can be attributed to the high content
of sulfate in the polysaccharide, as demonstrate the values found
directly from the extracts.
The sulfate content of the ulvan extracts (26–32%) is within
the values published in the literature (Hernández-Garibay et al.,
2011; Quemener et al., 1997). Sample III is that with the highest
amount of sulfate groups in both dry material and ashes (and low
value of sulfate-free ashes) confirming that the purification steps
were effective in elimination of inorganic material other than sul-
fate; these results substantiate that sulfate groups are linked to the
polysaccharide chain since they were not preferentially removed
in the purification steps. Hence, the sulfate content found is closely
related with the degree of substitution by sulfate in the polysac-
charide chain. The remaining inorganic material in this sample
is probably coming from counterions of sulfate groups or car-
boxylic groups (Lahaye, 1991; Ray & Lahaye, 1995). Counterions
3.2. Molecular weight distribution
Another important issue in polysaccharide characterization is
its molecular weight (Mw) distribution. The effect of flow rate
(0.5–1.0 mL/min), temperature (25–50 ◦C) and concentration of
suppressor (0.05–0.5 M) in the chromatographic profiles of HPSEC
were evaluated to guarantee the best resolution of ulvan chro-
matograms. The best resolution was found to flow rate 0.7 mL/min.
Regarding temperature and NaNO3 concentration, no effect on elu-
tion profiles was observed and, therefore, the chosen temperature
was 25 ◦C and the lowest value of concentration in the range was
used (50 mM) for HPSEC analysis.
Fig. 2 shows the typical HPSEC profile of the ulvan-rich extracts,
represented by sample III. The three samples revealed similar
elution profiles on RI with multimodal distribution. It should be
emphasized that due to the charge density of ulvan, and even-
tual branching, the hydrodynamic volume may differ of those of
pullulans (used as standards). Therefore, the estimated should be
interpreted as comparative data between samples. For the three
samples, in the region of high molecular weight (lower retention
time) the peaks are wide and not well resolved. This may be due
to the above referred tendency of ulvan to undergo aggregation. In
general, samples are mostly composed of two broad macromolec-
ular fractions, centered at 775 kDa (rt 8.3 min) and at 150 kDa (rt
9.8 min) and a third intense peak at 1.5 kDa (rt 14 min). This last
could be attributed to an ulvan fraction or co-extracted saccharides
with low polymerization degree.
The UV chromatograms (Fig. 2, typical chromatogram is shown
for sample III) evidenced similar elution profiles, with several

C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
541
900
800
700
600
500
400
300
200
100
0
fractions of material absorbing at 280 nm, probably proteins.
Of the three samples, samples II and III showed lower relative
intensity in UV which is in accordance with the lower protein
contents already stated in Table 1. The elution profiles on RI
and UV shows that these protein fractions are not co-eluting
with polysaccharides suggesting that these components are not
chemically bonded. This observation is important if further purifi-
cation of ulvan is considered, since it contributes to understand
the nature of contaminants association in the extracts. Differ-
ent molecular weight distributions and molecular weights of
polysaccharide ulvan have been found by other authors (Lahaye
& Robic, 2007; Robic, Gaillard, Sassi, Lerat, & Lahaye, 2009; Robic
et al., 2008). The aggregation phenomenon of ulvan polysaccha-
ride as polyelectrolyte is the most probable reason for the large
variation of Mw obtained by the different techniques in the litera-
ture.
d
d
0
18 20 22 24 26 28 30 32 34 36 38 40 42
Time (min)
3.3. Analysis of the carbohydrate composition of the ulvan
extracts
Fig. 3. Gas chromatogram (GC-FID) of TMS methyl glycosides of monosaccha-
rides obtained by acid methanolysis of sample II. Conditions: 2 M HCl/MeOH, 4 h
of reaction, 100 ◦C (␣-Xylp: ␣-xylopyranoside; ␤-Xylp: ␤-xylopyranoside; ␣-Rhap:
␣-rhamnopyranoside; ␤-Rhap: ␤-rhamnopyranoside; ␣-Glcp: ␣-glucopyranoside;
␤-Glcp: ␤-glucopyranoside; ␣-GlcAp: ␣-glucuronic acid pyranoside; ␤-GlcAp: ␤-
glucuronic acid pyranoside; ␤-GlcAf: ␤-glucuronic acid furanoside (6,3 lactone);
IdoA: iduronic acid (lactone form); d: disaccharides; Sorbitol (IS – internal stan-
dard)).
The characterization of the polysaccharides in the extracts based
on the identification and quantification of composing monosaccha-
rides assumes a crucial role in the study of the effect of different
steps in the extraction procedure. Monosaccharide in each sam-
ple was determined by (1) enzymatic hydrolysis following two
approaches and (2) acid methanolysis with GC-FID and GC–MS
analysis.
ulvan. As general trend, one point out that proportion of glucose
and xylose obtained are higher for EM approach.
The sum of the sulfate-free ashes, sulfate and protein contents
of the three sample fails to complete the compositional analysis:
for the chemical-enzymatic approach, the overall sum of the com-
ponents is 75%, 72% and 82% for samples I, II, and III, respectively;
EM hydrolysis gave overall sums of 78%, 64% and 78% for samples
I, II and III, respectively. In spite of the observed improvement by
the chemical-enzymatic method, these results prompt us to test
different analysis approach to overcome the hydrolysis resistance
demonstrated by enzymatic hydrolysis used in this work.
3.3.1. Enzymatic essays
The enzymatic hydrolysis approach using EM was performed to
provide a wide range hydrolytic activity, since it is expected that
other cell-wall polysaccharides could be co-extracted with ulvan.
␤-Glucuronidase approach over the TFA hydrolysates (chemical-
enzymatic method) was also performed, in this case, aiming to
improve the hydrolysis of aldo-glucuronosyl linkages (Lahaye &
Robic, 2007; Quemener et al., 1997; Willför et al., 2009). The results
are presented in Table 2.
In both approaches, the glucuronic acid and rhamnose are
the main residues in the extracts. GlcA liberated by chemical-
enzymatic method accounts for about 45% of the total of
polysaccharides, almost the double of the value obtained by
EM (see also % mol in Table 2). Confirming the effectiveness
of chemical-enzymatic method toward aldo-glucuronosyl link-
ages in comparison to EM, the content of glucuronic acid found
(12–19 wt%) is within the range reported in the literature (6–19%)
(Lahaye & Robic, 2007; Quemener et al., 1997), while for EM the
values for GlcA is considerable lower (5–9 wt%). In accordance, the
disaccharide Rha-GlcA was not detected after the treatment with
␤-glucuronidase, confirming its effective activity toward glycosidic
linkage of aldobiuronic acids that remain after TFA treatment. Rha
content found by the two methods is similar (7–13 wt%) but con-
siderable lower than the range of values of the literature (17–45%)
(Lahaye & Robic, 2007).
The other residues are iduronic acid, xylose and glucose.
Iduronic acid, a typical residue found in ulvan from U. lactuca (in
ulvanobiuronic acid type B) accounts for 2–3%, close to values found
in the literature for U. lactuca (Lahaye & Robic, 2007; Quemener
et al., 1997). Quemener et al. (1997) have detected iduronic acid
for the first time in algae polysaccharides based on hydroly-
sis improvement using the TFA hydrolysis and ␤-glucuronidase
(chemical-enzymatic method), referring to it as the most favorable
for the quantitative release of iduronic acid. In accordance, moder-
ately higher content for iduronic acid was found for samples II and
III by chemical-enzymatic method. Xylose is referred to as part of
ulvan structure in ulvanobioses moieties and glucose is frequently
associated with glucans or amorphous cellulose co-extracted with
3.3.2. Conventional acid methanolysis 4 h, 2 M: results and
identified limitations
Methanolysis of ulvan-rich extracts was carried out for 4 h
at 100 ◦C, with 2 M HCl/MeOH, the conditions referred to as the
best compromise between efficient cleavage of glucuronosyl link-
ages and low degradation of the monomers released (Bertaud
et al., 2002; Rabemanolontsoa, Ayada, & Saka, 2011), although
Quemener et al. (1997) have pointed out that this method were
not efficient to quantitatively release uronic acids. Fig. 3 depicts a
typical chromatogram of TMS methyl glycosides obtained by acid
methanolysis. The quantification results are presented in Table 3.
The total content of monosaccharides obtained by acid methanol-
ysis (2 M, 4 h) is in the range 28–55%, which is a considerable high
value comparatively to that found by other authors using the same
conditions (19%) (Quemener et al., 1997). This could be attributed
to differences of the extract composition rather than quantifica-
tion issues. Notice that between samples I, II and III there are also
significant differences, suggesting that the samples has different
susceptibility to methanolysis conditions. GlcA and Rha are the
main residues in the extracts, accounting for 70–85% of total weight
of polysaccharide, in accordance with the results obtained by EM
and chemical-enzymatic method.
Comparatively to the other two, sample III shows significantly
higher values for both Rha and GlcA (about 22% of extract dry
weight). Considering the results of Table 1 (total content of non-
carbohydrate components – sulfate-free ashes, sulfate and proteins
– 44–50%) and the total content of polysaccharide depicted in
Table 3 (28–55%), the total quantified weighs are 77%, 73% and

542
C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
Table 2
Composition of the polysaccharide fraction in the ulvan-rich extracts accessed by chemical-enzymatic method and enzymatic hydrolysis using EM.
Sample I
% w/w
Sample II
% w/w
Sample III
% w/w
% mol
% mol
% mol
Chemical-enzymatic method
Rha
GlcA
IdoA
Glc
7.8 0.4
12.4 0.3
2.1 0.1
1.8 0.0
1.6 0.1
–
33
45
8
7
7
9.1 0.4
13.7 1.0
2.1 0.1
1.6 0.1
1.6 0.0
–
35
45
7
6
7
11.9 1.1
19.3 0.6
3.2 0.2
1.9 0.1
2.3 0.1
–
34
46
8
5
7
Xyl
Disaccharides
–
–
–
Total
25.6 0.2
100
28.0 1.4
100
38.5 2.1
100
Hydrolysis using EM
Rha
GlcA
IdoA
Glc
11.0 0.6
7.6 0.6
2.1 0.2
2.6 0.1
2.3 0.1
4.3 0.6
42
25
7
9
10
7
7.3 0.3
4.6 0.4
1.5 0.1
1.0 0.1
1.6 0.1
4.2 0.2
43
23
8
5
10
11
13.4 0.8
9.2 0.7
2.3 0.0
1.5 0.0
2.7 0.1
5.2 0.1
45
26
7
5
10
7
Xyl
Disaccharides
Total
29.9 0.6
100
20.2 1.1
100
34.3 1.5
100
% w/w of monosaccharide calculated as regular anhydropolymers.
Data are presented as mean standard deviations (SD) of triplicate essays.
98% for samples I, II and III, respectively. This verification revealed
an underestimation of the polysaccharide content of ulvan-rich
extracts (except for sample III) which could be attributed to
the chemical stability of glucuronosyl linkages (Biermann, 1988;
Conrad, 1980; McKinnell & Percival, 1962; Willför et al., 2009). As an
example (1 → 4)-␤-d-glucuronan have been reported as very resis-
tant toward hydrolysis with sulfuric acid (De Ruiter et al., 1992). The
underestimation of the quantification also can be also related with
other non-detected oligosaccharides present in the mixture. This
underestimation was already stated for the enzymatic methodolo-
gies referred in Section 3.3.1.
monosaccharides units released and therefore higher acid concen-
trations were not attempted. The variation of monosaccharides
percentages released from ulvan-rich extracts with reaction time
and acid concentration are shown Fig. 4A and B for sample I, C and
D for sample II and E and F for sample III.
The results show that the lower content of total released
monosaccharides was achieved after 4 h of reaction time for sample
I and sample II due to the loss of GlcA. This behavior can be related
to degradation of GlcA during the initial reaction time due to its par-
ticular instability in acid. Some authors claims that HCl is mostly
consumed in the first hours providing the advantage of a reasonable
stability of released methyl glycosides, up to 24 h (Bertaud et al.,
2002; Sundberg et al., 1996). Other authors (Quemener & Lahaye,
1998) have proven that the maximum yields for studied sugars
were obtained when methanolysis was performed with pH values
of 6–7, which is in accordance with the high stability of released
sugars when the acidity of the solution drops rather quickly (Willför
et al., 2009). All these assumptions could explain the release of more
labile monosaccharides until 3 h of reaction and their degradation
3.3.3. Kinetics of methanolysis and effect of HCl concentration:
method improvement
A study concerning the release of residues along reaction
time (3–72 h), with two HCL concentrations (2 M and 3 M) was
developed aiming to achieve maximum values for monosaccharide
content. According to Chambers and Clamp (1971), methanolic
4 M and 6 M HCl lead to the considerable destruction of the
Table 3
Monosaccharides composition of ulvan-rich extracts released after acid methanolysis, as % w/w dry material and % mol.
Sample I
Sample II
% w/w
Sample III
% w/w
% w/w
% mol
% mol
% mol
Acid methanolysis using 2 M HCl/MeOH, 4 h, 100 ◦
C
Rha
GlcA
IdoA
Glc
8.6 0.1
10.6 1.0
2.6 0.1
1.8 0.0
1.8 0.1
2.2 0.5
35
36
10
7
8
4
9.2 0.3
12.1 1.3
2.5 0.6
1.4 0.1
1.9 0.1
1.5 0.1
36
40
8
5
8
22.4 0.9
22.5 1.1
3.1 0.2
1.0 0.0
3.7 0.1
1.9 0.0
45
38
5
2
8
Xyl
Disaccharides
3
2
Total
27.6 1.7
100
28.6 1.4
100
54.6 2.3
100
Acid methanolysis using 3 M HCl/MeOH, 48 h, 100 ◦
C
Rha
GlcA
IdoA
Glc
11.1 0.3
21.7 0.1
2.6 0.2
1.7 0.0
2.1 0.1
–
32
52
6
4
6
16.1 0.7
20.4 0.8
3.0 0.1
1.6 0.0
2.6 0.1
–
40
43
6
4
7
13.5 1.1
24.2 1.7
2.6 0.1
0.9 0.0
1.6 0.1
–
45
38
5
2
8
Xyl
Disaccharides
–
–
2
Total
39.3 0.6
100
43.7 0.1
100
42.8 2.9
100
% w/w of monosaccharide calculated as regular anhydropolymers.
Data are presented as mean standard deviations (SD) of triplicate essays.

C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
543
25.0
20.0
15.0
10.0
5.0
25.0
20.0
GlcA
15.0
GlcA
Rha
IdoA
Xyl
Rha
10.0
IdoA
5.0
0.0
Xyl
Glc
Glc
0.0
3 H
4 H
6 H
24 H
48 H
3 H
4 H
6 H
24 H
48 H
A
HCl 2M
C
HCl 2M
25.0
20.0
15.0
10.0
5.0
25.0
20.0
15.0
10.0
5.0
GlcA
Rha
IdoA
Xyl
GlcA
Rha
IdoA
Xyl
0.0
0.0
Glc
Glc
3H
4H
6H
24H
48 H
72 H
3H
4H
6H
24H
48 H
72 H
HCl 3M
D
HCl 3M
B
30.0
25.0
20.0
15.0
10.0
5.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Sample I
GlcA
Rha
IdoA
Xyl
Sample II
Sample III
Glc
0.0
3 H
4 H
24 H
3 H
4 H
6 H
24 H
48 H
G
E
HCl 2M
HCl 2M
30.0
25.0
20.0
15.0
10.0
5.0
2.5
2.0
1.5
1.0
0.5
0.0
Sample I
Sample II
Sample III
GlcA
Rha
IdoA
Xyl
0.0
Glc
3H
4H
6H
24H
48 H
3H
4H
6H
24H
48 H
72 H
F
H
HCl 3M
HCl 3M
Fig. 4. Quantification of monosaccharides released by acid methanolysis for different reaction times and HCl concentrations in sample I (A, B), sample II (C, D) and sample
III (E, F), and disaccharides content in the three samples (G, H). A, C, E, G refer to HCl 2 M and B, D, F, H refer to HCl 3 M.
after this time. After 6 h, the cleavage proceeds and more resis-
tant/protected linkages are cleaved releasing the monosaccharides
of the ulvan structure and increasing the quantified value. However,
the destroyed monosaccharides are no longer recovered and does
not account for the total. Hence, the total content is always underes-
timates. The increase of Rha and GlcA percentage is accompanied by
a decrease of disaccharides (GlcA-Rha) content, as demonstrate Fig.
4G and H. This behavior is in accordance with the well-known aldo-
glucuronosyl linkage resistance assigned to ulvanobiuronic acid,
d-GlcA-(1,4)-l-Rha (Quemener et al., 1997).
Higher values were achieved, for samples I and II, using more
severe conditions: 48 h at 100 ◦C with 3 M HCl/MeOH (Table 3). In
these conditions, the GlcA percentage was remarkably high; Rha
also increase while for IdoA, Glc and Xyl no changes were notice-
able. These conditions represented the best compromise to achieve
the maximum quantity of released monosaccharides. After that
time, a decreased of the amount of monosaccharides was observed.
For sample III the highest yield on total monosaccharides, in par-
ticularly rhamnose, was achieved for 4 h, with 2 M HCl/MeOH. For
3 h of reaction, the cleavage of linkages involving Rha and GlcA was
greater than for the other samples; this probably led to higher acid
consumption in the first hours (protecting the liberated monosac-
charides from further degradation), however without affecting the
subsequent development of methanolysis since high yields were
achieved. This behavior could be attributed to the high content of
Rha in this sample (22% released in 4 h) comparatively to the others
samples, 8.6% for sample I and 9.2% for sample II using the same con-
ditions. The decrease in total sugar content in sample III for more
than 4 h of reaction time is almost exclusively due to it degradation
as depicted in Fig. 4E and F. In accordance, rhamnose is referred to
as a labile monosaccharide (Choy & Dutton, 1973).
In the optimized methanolysis conditions for each sample, the
percentage of saccharide fraction found was between 39% and 55%.
Rha and GlcA are main monosaccharides, but minor amounts of
glucose, iduronic acid and xylose were also detected in the range
of 1.0–3.7%. Sample III revealed the lowest amount of glucose,
which is in accordance with the enzymatic treatment applied as
an additional step in the extraction procedure to remove starch as
a contaminant. Considering the sum of all the identified compo-
nents of ulvan-rich extracts (carbohydrate fraction, ashes, protein
and sulfates) a total content close to 90% was obtained for samples
I and II while for sample III this value reached to 98%. These results
confirmed acid methanolysis as highly effective for ulvan, leading
to very good total yields of carbohydrates; but, more important,
this study reveals also that the conditions should be studied for
each polysaccharide extract.

544
C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
to establish a criterion for ulvan content in the algae extracts
(obtained by variants of the same methodologies of extraction)
based on the molar percentage of the major repeating units:
the ulvanobiuronic acids and ulvanobioses. Other basic repeating
sequence referenced, →4)-␤-d-GlcpA-(1 → 4)-␣-l-Rhap-(1 → 4)-
␤-d-Xylp (Lahaye & Ray, 1996) was also considered in this
approach. Gathered together the data of acid methanolysis at the
optimized conditions, repeating sequences were established as fol-
low resulting the empirical formulas that represent distribution of
residues and sulfate groups in each extract:
Sample I – [GlcA-Rha]20 [IdoA-Rha]6 [GlcA-Rha-Xyl]6 GlcA26 Glc4
−
[SO3
]
Fig. 5. Typical 13C NMR spectrum of ulvan-rich extracts (presented for sample III)
dissolved in D2O, acquired at 70 ◦C for a minimum of 48 h at 100.62 MHz. G and I cor-
respond to the carbon signals of the glucuronic acid and iduronic acid, respectively.
R and Rꢀ refers to rhamnose linked to glucuronic acid and iduronic acid, respectively.
Number refers to the carbon number in the structure of each residue. Peaks noted
by * are unattributed signals, most probably corresponding to carbon signals of the
protein (contaminant).
30
Sample II – [GlcA-Rha]27 [IdoA-Rha]6 [GlcA-Rha-Xyl]7 GlcA9 Glc4
−
[SO3
]
26
Sample III – [GlcA-Rha]25 [IdoA-Rha]6 [GlcA-Rha-Xyl]4 GlcA24 Glc2
−
[SO3
]
32
Considering the total sulfate groups in each ulvan-rich extract,
one can suggest that sulfated sequences represented more than
2/3 of the total that constitutes the polysaccharide chain. Glucose
can be associated with co-eluting contaminants, the glucans. Sam-
ple III showed the lowest proportion of glucose in accordance with
the treatment preformed in the extraction procedure. Considering
that no signal in 13C NMR spectra were assigned to 1,4-linked ␤-
d-glucuronan one can suggest that the additional GlcA reported
in the empirical formulas could be present linked to the chain, in
accordance with already reported branching at O-2 of rhamnose by
glucuronic acid unit (Lahaye & Ray, 1996).
Sample III had the highest content of total sugar and sulfate
groups, with the lowest amounts of proteins (1.3%) and inorganic
material (10%). This sample is composed by rhamnose (22%), glu-
curonic acid (23%), xylose (3.7%), iduronic acid (3.1%) and glucose
(1.0%) and it is highly sulfated (32%). These findings are in accor-
dance with previous reported features for ulvan of this alga species
(Quemener et al., 1997; Siddhanta, Goswami, Ramavat, Mody, &
Mairh, 2001). However, this particular ulvan is rich in glucuronic
acid and the ratio Rha/GlcA is approximately 1 for the three sam-
ples. This peculiarity may be an influence of different factors,
such as seasonality or extraction methodology (Hernández-Garibay
et al., 2011; Lahaye et al., 1999; Robic et al., 2008; Siddhanta et al.,
2001; Wong & Cheung, 2000). Ulvan obtained through the devel-
oped extraction procedure revealed a high molecular weight, with
apparent Mw in the range 770–790 kDa. Concerning the SEC pro-
files, sample III showed superior intensity of high molecular weight
fractions and lower contribution of protein, in accordance with
chemical characterization results.
The total content of carbohydrate obtained from the two dif-
ferent approaches of enzymatic hydrolysis was lower than the
obtained after acid methanolysis (at optimized conditions for each
ulvan-rich extracts). The hydrolytic activity is limited by the type of
enzymes in each approach and, consequently, the acid methanoly-
sis had led to a broader action toward the polysaccharides.
3.3.4. 13C NMR spectroscopy
The typical 13C NMR spectrum of the ulvan-rich extracts is
depicted in Fig. 5 for sample III. The spectra of the three samples
do not present significant differences, confirming that the overall
composition of ulvan structure was close whatever the extraction
procedure. The chemical shifts were attributed based on references
reporting assignments for ulvan and oligosaccharides (Lahaye et al.,
1997, 1998, 1999; Lahaye & Ray, 1996; Lahaye, Ray, Baumberger,
Quemener, & Axelos, 1996; Lahaye & Robic, 2007).
The signals assignments corresponding to carbons of rhamnose
(R1 – 97.6 ppm, R2 – 67.0 ppm, R3, R4 – 76.1 ppm, R5 – 66.1 ppm, R6
– 14.9 ppm) and glucuronic acid (G1 – 101.1 ppm, G2 – 71.8 ppm, G3
– 72.0 ppm, G4 – 76.8 ppm, G5 – 74.1 ppm) that constitute ulvanobi-
uronic acid 3-sulfate – type A were identified, as depicted in Fig. 5.
The chemical shifts attributed to iduronic acid (I2 – 68.3 ppm, I3
– 69.3 ppm and I4 – 76.8 ppm) and rhamnose 3-sulfate linked to
iduronic acid (Rꢀ1 – 98.8 ppm, Rꢀ2 – 67.0 ppm, Rꢀ3 – 76.8 ppm, Rꢀ4
– 74.6 ppm, Rꢀ5 – 65.8 ppm, Rꢀ6 – 14.8 ppm) in ulvanobiuronic acid
3-sulfate – type B, were also detected allowing to confirm the pres-
ence of this sequence. The C-6 of GlcA (and IdoA) was assigned at
173.0 ppm (not shown in the spectra).
In previous studies, some authors performed spectroscopic
characterization of ulvan using 13C NMR analysis and noticed the
presence of carbon signals representing contiguous 1,4-linked ␤-d-
glucuronic acid units in polysaccharide structure and/or a separate
1,4-linked ␤-d-glucuronan (seaweed glucuronans) co-extracted
with ulvan, and occurring as a contaminant (Lahaye et al., 1998,
1999; Lahaye & Robic, 2007). However, the typical chemical shifts of
carbon in these structures were not detected in any of the samples.
4. Conclusion
The chemical composition and structural characteristics of
ulvan-rich extracts were assessed by complementary methodolo-
gies. Sugar analysis by acid methanolysis was improved and proved
to be the best mean for quantifying sugar content in ulvan extracts.
This is an important tool to assess the effect of the different steps
in the extraction procedures.
3.3.5. Effect of the extraction steps in the composition of the
extracts
The three modes of extraction and purification of ulvan proved
to impact the content of contaminants and the sugar analysis of the
extracts without affecting the overall chemical structure.
The extraction allowed obtaining a final extract with 10% of
inorganic compounds, and the enzymatic treatment steps were
effective on the reduction of both protein and glucose, to a maxi-
mum of 2% on the final extracts. There were no evidences for the
presence of glucuronans in the extracts and, therefore, all the car-
bohydrates should be related with the polysaccharide ulvan. In
The uncertainty related with the identification of the residues
belonging to the ulvan and those from to co-extracted polysaccha-
rides has been frequently referred in the literature (Elboutachfaiti
et al., 2009) representing a challenge in the evaluation of the
real content of ulvan in the extracts. Overcome this limita-
tion is essential regarding application studies where the starting
material should be carefully controlled. For this, our proposal is

C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
545
this sense, it is possible to extract a purified polysaccharide from
glucuronic acid oligosaccharides using
Research, 344, 1670–1675.
Harada, N., & Maeda, M. (1998). Chemical structure of antithrombin-active rhamnan
sulfate from Monostrom nitidum. Bioscience Biotechnology and Biochemistry, 62,
1647–1652.
Hernández-Garibay, E., Zertuche-González, J., & Pacheco-Ruíz, I. (2011). Isolation
and chemical characterization of algal polysaccharides from the green seaweed
Ulva clathrata (Roth) C. Agardh. Journal of Applied Phycology, 23, 537–542.
Kaeffer, B., Benard, C., Lahaye, M., Blottiere, H. M., & Cherbut, C. (1999). Biological
properties of ulvan, a new source of green seaweed sulfated polysaccharides,
on cultured normal and cancerous colonic epithelial tells. Planta Medica, 65,
527–531.
a
glucuronanlyase. Carbohydrate
green algae employing a simple methodology. The ulvan obtained
is a high molecular weight and highly sulfated polysaccharide, rich
in rhamnose and glucuronic acid (ulvanobiuronic acid A), with
minor amounts of iduronic acid, and xylose (ulvanobiuronic acid
B). Considering the overall results, the final assessment of ulvan
percentage in the sample III was 88%.
Acknowledgments
Kamerling, J. P., & Gerwig, G. J. (2007). Strategies for the structural analysis of carbo-
hydrates. In P. K. Johannis (Ed.), Comprehensive glycoscience (pp. 1–68). Oxford:
Elsevier.
Lahaye, M. (1991). Marine algae as sources of fibres: Determination of soluble and
insoluble dietary fibre contents in some sea vegetables. Journal of the Science of
Food and Agriculture, 54, 587–594.
Lahaye, M. (1998). NMR spectroscopic characterisation of oligosaccharides from two
Ulva rigida ulvan samples (Ulvales, Chlorophyta) degraded by a lyase. Carbohy-
drate Research, 314, 1–12.
This work was supported by the project IBEROMARE (‘Cen-
tro Multipolar de Valorizac¸ ão de Recursos e ResíduosMarinhos’),
approved by Operational Programme for Cross-border Coop-
eration: Spain – Portugal, 2007–2013 (POCTEP), with funding
contribution through the European Regional Development Fund
(ERDF co-funding) and POCTEP.
Lahaye, M., Alvarez-Cabal Cimadevilla, E., Kuhlenkamp, R., Quemener, B., Lognoné,
V., & Dion, P. (1999). Chemical composition and 13C NMR spectroscopic char-
acterisation of ulvans from Ulva (Ulvales, Chlorophyta). Journal of Applied
Phycology, 11, 1–7.
Anabela Alves is grateful for financial support from
Fundac¸ ãopara
a
Ciência
e
Tecnologia (FCT) through the
SFRH/BD/39359/2007 grant.
Lahaye, M., Brunel, M., & Bonnin, E. (1997). Fine chemical structure analysis of
oligosaccharides produced by an ulvan-lyase degradation of the water-soluble
cell-wall polysaccharides from Ulva sp. (Ulvales, Chlorophyta). Carbohydrate
Research, 304, 325–333.
Lahaye, M., Inizan, F., & Vigouroux, J. (1998). NMR analysis of the chemical struc-
ture of ulvan and of ulvan–boron complex formation. Carbohydrate Polymers, 36,
239–249.
Lahaye, M., & Jegou, D. (1993). Chemical and physical–chemical characteristics of
dietary fibres from Ulva lactuca (L.) Thuret and Enteromorpha compressa (L.) Grev.
Journal of Applied Phycology, 5, 195–200.
Lahaye, M., Jegou, D., & Buleon, A. (1994). Chemical characteristics of insoluble
glucans from the cell wall of the marine green alga Ulva lactuca (L.) Thuret.
Carbohydrate Research, 262, 115–125.
The authors thank Novozymes for kindly providing the enzy-
matic mixture. Authors are also grateful to Dra. Mariana Andrade
and Materials Center of the University of Porto (CEMUP) for the
acquisition of 13C NMR spectra (NMR spectrometer is part of
the National NMR Network and was purchased in the frame-
work of the National Program for Scientific Re-equipment, contract
REDE/1517/RMN/2005, with funds from POCI 2010 (FEDER) and
FCT).
References
Lahaye, M., & Ray, B. (1996). Cell-wall polysaccharides from the marine green alga
Ulva rigida (Ulvales, Chlorophyta) – NMR analysis of ulvan oligosaccharides.
Carbohydrate Research, 283, 161–173.
Alves, A., Caridade, S. G., Mano, J. F., Sousa, R. A., & Reis, R. L. (2010). Extraction and
physico-chemical characterization of a versatile biodegradable polysaccharide
obtained from green algae. Carbohydrate Research, 345, 2194–2200.
Bertaud, F., Sundberg, A., & Holmbom, B. (2002). Evaluation of acid methanoly-
sis for analysis of wood hemicelluloses and pectins. Carbohydrate Polymers, 48,
319–324.
Biermann, C. J. (1988). Hydrolysis and other cleavages of glycosidic linkages in
polysaccharides. In R. S. Tipson, & H. Derek (Eds.), Advances in carbohydrate
chemistry and biochemistry (pp. 251–271). San Diego, CA: Academic Press.
Bleton, J., Mejanelle, P., Sansoulet, J., Goursaud, S., & Tchapla, A. (1996). Char-
acterization of neutral sugars and uronic acids after methanolysis and
trimethylsilylation for recognition of plant gums. Journal of Chromatography A,
720, 27–49.
Cannell, R. J. P., Dufresne, C., Florence, A. J., Gailliot, F. P., Gibbons, S., Gray, A. I.,
et al. (1998). Natural products isolation. In R. J. P. Cannell (Ed.), Methods in
biotechnology (pp. 343–408). Totowa, NJ: Humana Press.
Castro, R., Piazzon, M. C., Zarra, I., Leiro, J., Noya, M., & Lamas, J. (2006). Stimulation
of turbot phagocytes by Ulva rigida C. Agardh polysaccharides. Aquaculture, 254,
9–20.
Lahaye, M., Ray, B., Baumberger, S., Quemener, B., & Axelos, M. A. V. (1996). Chemical
characterisation and gelling properties of cell wall polysaccharides from species
of Ulva (Ulvales, Chlorophyta). Hydrobiologia, 326–327, 473–480.
Lahaye, M., & Robic, A. (2007). Structure and functional properties of ulvan, a polysac-
charide from green seaweeds. Biomacromolecules, 8, 1765–1774.
Mano, J. F., Silva, G. A., Azevedo, H. S., Malafaya, P. B., Sousa, R. A., Silva, S. S., et al.
(2007). Natural origin biodegradable systems in tissue engineering and regen-
erative medicine: Present status and some moving trends. Journal of the Royal
Society Interface, 4, 999–1030.
Mao, W., Zang, X., Li, Y., & Zhang, H. (2006). Sulfated polysaccharides from marine
green algae Ulva conglobata and their anticoagulant activity. Journal of Applied
Phycology, 18, 9–14.
McKinnell, J. P., & Percival, E. (1962). Structural investigations on the water-soluble
polysaccharide of the green seaweed Enteromorpha compressa. Journal of the
Chemical Society (Resumed), 3141–3148.
Pengzhan, Y., Ning, L., Xiguang, L., Gefei, Z., Quanbin, Z., & Pengcheng, L. (2003). Anti-
hyperlipidemic effects of different molecular weight sulfated polysaccharides
from Ulva pertusa (Chlorophyta). Pharmacological Research, 48, 543–549.
Pengzhan, Y., Quanbin, Z., Ning, L., Zuhong, X., Yanmei, W., & Zhi’en, L. (2003).
Polysaccharides from Ulva pertusa (Chlorophyta) and preliminary studies on
their antihyperlipidemia activity. Journal of Applied Phycology, 15, 21–27.
Qi, H., Zhang, Q., Zhao, T., Chen, R., Zhang, H., Niu, X., et al. (2005). Antioxidant activity
of different sulfate content derivatives of polysaccharide extracted from Ulva
pertusa (Chlorophyta) in vitro. International Journal of Biological Macromolecules,
37, 195–199.
Qi, H., Zhao, T., Zhang, Q., Li, Z., Zhao, Z., & Xing, R. (2005). Antioxidant activity of
different molecular weight sulfated polysaccharides from Ulva pertusa Kjellm
(Chlorophyta). Journal of Applied Phycology, 17, 527–534.
Quemener, B., & Lahaye, M. (1998). Comparative analysis of sulfated galactans from
red algae by reductive hydrolysis and mild methanolysis coupled to two differ-
ent HPLC techniques. Journal of Applied Phycology, 10, 75–81.
Quemener, B., Lahaye, M., & Bobin-Dubigeon, C. (1997). Sugar determination in
ulvans by a chemical-enzymatic method coupled to high performance anion
exchange chromatography. Journal of Applied Phycology, 9, 179–188.
Quemener, B., Lahaye, M., & Metro, F. (1998). Assessment of methanolysis for the
determination of composite sugars of gelling carrageenans and agarose by HPLC.
Carbohydrate Research, 266, 53–64.
Rabemanolontsoa, H., Ayada, S., & Saka, S. (2011). Evaluation of different methods
to determine monosaccharides in biomass. In T. Yao (Ed.), Zero-carbon energy
Kyoto 2010 (pp. 123–128). Tokyo: Springer.
Ray, B., & Lahaye, M. (1995). Cell-wall polysaccharides from the marine green alga
Ulva ‘rigida’ (Ulvales, Chlorophyta). Chemical structure of ulvan. Carbohydrate
Research, 274, 313–318.
Robic, A., Gaillard, C., Sassi, J. F., Lerat, Y., & Lahaye, M. (2009). Ultrastructure of ulvan:
A polysaccharide from green seaweeds. Biopolymers, 91, 652–664.
Castro, R., Zarra, I., & Lamas, J. (2004). Water-soluble seaweed extracts mod-
ulate the respiratory burst activity of turbot phagocytes. Aquaculture, 229,
67–78.
Chambers, R. E., & Clamp, J. R. (1971). An assessment of methanolysis and other
factors used in the analysis of carbohydrate-containing materials. Biochemical
Journal, 125, 1009–1018.
Choy, Y.-M., & Dutton, G. G. S. (1973). Structure of the capsular polysaccharide of
Klebsiella K-type 56. Canadian Journal of Chemistry, 51, 3021–3026.
Cluzet, S., Torregrosa, C., Jacquet, C., Lafitte, C., Fournier, J., Mercier, L., et al. (2004).
Gene expression profiling and protection of Medicago truncatula against a fun-
gal infection in response to an elicitor from green algae Ulva spp. Plant, Cell &
Environment, 27, 917–928.
Conrad, H. E. (1980). The acid lability of the glycosidic bonds of l-iduronic acid
residues in glycosaminoglycan. Biochemical Journal, 191, 355–363.
Dahlman, O., Jacobs, A., Liljenberg, A., & Olsson, A. I. (2000). Analysis of carbohydrates
in wood and pulps employing enzymatic hydrolysis and subsequent capillary
zone electrophoresis. Journal of Chromatography A, 891, 157–174.
De Ruiter, G. A., Schols, H. A., Voragen, A. G. J., & Rombouts, F. M. (1992). Carbohydrate
analysis of water-soluble uronic acid-containing polysaccharides with high-
performance anion-exchange chromatography using methanolysis combined
with TFA hydrolysis is superior to four other methods. Analytical Biochemistry,
207, 176–185.
Doco, T., O’Neill, M. A., & Pellerin, P. (2001). Determination of the neutral and acidic
glycosyl-residue compositions of plant polysaccharides by GC–EI-MS analysis
of the trimethylsilyl methyl glycoside derivatives. Carbohydrate Polymers, 46,
249–259.
Elboutachfaiti, R., Delattre, C., Petit, E., El Gadda, M., Courtois, B., Michaud, P., et al.
(2009). Improved isolation of glucuronan from algae and the production of

546
C. Costa et al. / Carbohydrate Polymers 88 (2012) 537–546
Robic, A., Rondeau-Mouro, C., Sassi, J. F., Lerat, Y., & Lahaye, M. (2009). Structure and
interactions of ulvan in the cell wall of the marine green algae Ulva rotundata
(Ulvales, Chlorophyceae). Carbohydrate Polymers, 77, 206–216.
Robic, A., Sassi, J. F., & Lahaye, M. (2008). Impact of stabilization treatments of
the green seaweed Ulva rotundata (Chlorophyta) on the extraction yield, the
physico-chemical and rheological properties of ulvan. Carbohydrate Polymers,
74, 344–352.
Redinbaugh, M. G., & Turley, R. B. (1986). Adaptation of the bicinchoninic acid protein
assay for use with microtiter plates and sucrose gradient fractions. Analytical
Biochemistry, 153, 267–271.
Sundberg, A., Sundberg, K., Lillandt, C., & Holmbom, B. (1996). Determination
of hemicelluloses and pectins in wood and pulp fibres by acid methanol-
ysis and gas chromatography. Nordic Pulp and Paper Research Journal, 11,
216–219.
Warrand, J. (2006). Healthy polysaccharides – The next chapter in food products.
Food Technology and Biotechnology, 44, 355–370.
Willför, S., Pranovich, A., Tamminen, T., Puls, J., Laine, C., Suurnäkki, A., et al.
(2009). Carbohydrate analysis of plant materials with uronic acid-containing
polysaccharides
– A comparison between different hydrolysis and subse-
quent chromatographic analytical techniques. Industrial Crops and Products, 29,
Saeman, J. F. (1945). Kinetics of wood saccharification – Hydrolysis of cellulose
and decomposition of sugars in dilute acid at high temperature. Industrial &
Engineering Chemistry, 37, 43–52.
Siddhanta, A. K., Goswami, A. M., Ramavat, B. K., Mody, K. H., & Mairh, O. P. (2001).
Water soluble polysaccharides of marine algal species of Ulva (Ulvales, Chloro-
phyta) of Indian waters. Indian Journal of Marine Sciences, 30, 166–172.
571–580.
Wong, K. H.,
& Cheung, P. C. K. (2000). Nutritional evaluation of some sub-
tropical red and green seaweeds. Part I. Proximate composition, amino
acid profiles and some physico-chemical properties. Food Chemistry, 71,
475–482.