Selective hydrolysis of the 3,6-Anhydrogalacotosidic linkage in red algal galactan: A combination of reductive acid hydrolysis and anhydrous mercaptolysis

Article abstract of DOI:10.1271/bbb.100306

Here we report a simple method for the structural analysis of red algal galactan containing 3,6-anhydroga-lactose. Structural heterogeneity in the galactan was demonstrated by this method. For selective hydrolysis of 3,6-anhydrogalactosidic linkages in the galactan, conditions for reductive mild acid hydrolysis were examined by characterizing the resulting oligosaccha-ride alditols by anhydrous mercaptolysis. Residues other than alditols at the reducing ends, including labile 3,6-anhydrogalactose, were liberated quantitatively as di-ethyl dithioacetal derivatives, whereas alditols at the reducing ends were not derivatized and were liberated as alditols intact. The liberated sugars were then separated and measured quantitatively by gas-liquid chromatog-raphy. Heating of agarose in reductive hydrolysis with 0.3 m trifluoroacetic acid in the presence of an acid-stable reducing agent, 4-methyl morpholine borane, at 80 °C for 90min and for 90 °C for 45min was found to be optimum for the selective hydrolysis of 3,6-anhydrogalactosidic bonds, without detectable cleavage of other glycosidic bonds.

Full text of DOI:10.1271/bbb.100306

Biosci. Biotechnol. Biochem., 74 (9), 1895–1900, 2010
Selective Hydrolysis of the 3,6-Anhydrogalacotosidic Linkage
in Red Algal Galactan: A Combination of Reductive Acid Hydrolysis
and Anhydrous Mercaptolysis
y
Yoichiro HAMA,1; Aya TSUNEOKA,1 Ryoji MORITA,1 Osamu NOMOTO,1 Kenshi YOSHINAGA,1
1
Hideo HATATE,2 Toshihisa SUMI,3 and Hiroki NAKAGAWA
1Department of Applied Biochemistry and Food Sciences, Faculty of Agriculture, Saga University,
1 Honjo, Saga 840-8502, Japan
2Department of Biological Production and Environmental Science, Faculty of Agriculture, University of Miyazaki,
1-1 Gakuen Kihanadai-nishi, Miyazaki 889-2192, Japan
3Department of Bioengineering, Kumamoto National College of Technology,
2627 Hirayama Shin-machi, Yatsushiro, Kumamoto 866-8501, Japan
Received April 21, 2010; Accepted June 1, 2010; Online Publication, September 7, 2010
[doi:10.1271/bbb.100306]
Here we report a simple method for the structural
analysis of red algal galactan containing 3,6-anhydroga-
lactose. Structural heterogeneity in the galactan was
demonstrated by this method. For selective hydrolysis
of 3,6-anhydrogalactosidic linkages in the galactan,
conditions for reductive mild acid hydrolysis were
examined by characterizing the resulting oligosaccha-
ride alditols by anhydrous mercaptolysis. Residues other
than alditols at the reducing ends, including labile 3,6-
anhydrogalactose, were liberated quantitatively as di-
ethyl dithioacetal derivatives, whereas alditols at the
reducing ends were not derivatized and were liberated as
alditols intact. The liberated sugars were then separated
and measured quantitatively by gas-liquid chromatog-
raphy. Heating of agarose in reductive hydrolysis with
0.3 M trifluoroacetic acid in the presence of an acid-stable
reducing agent, 4-methyl morpholine borane, at 80 ꢀC
for 90 min and for 90 ꢀC for 45 min was found to be
optimum for the selective hydrolysis of 3,6-anhydroga-
lactosidic bonds, without detectable cleavage of other
glycosidic bonds.
Stevenson and Furneaux5) reported a double-hydroly-
sis procedure in which mild acid hydrolysis was
employed to cleave only 3,6-anhydrogalactosidic link-
ages, followed by reduction with sodium borohydride.
The resulting oligosaccharide alditols with 3,6-anhydro-
galactitol at the reducing ends were fully hydrolyzed for
compositional analysis by gas-liquid chromatography
(GLC). They also developed a reductive hydrolysis
procedure in which mild acid hydrolysis is carried out in
the presence of an acid-stable reducing agent, 4-methyl
morpholine borane (MMB), to obtain oligosaccharide
alditols with 3,6-anhydrogalactitol at the reducing
ends.5) Since then, these methods have been modified
by several other groups.6–8) However, the degree of
hydrolysis of the 3,6-anhydrogalactosidic bonds in the
galactans was evaluated by determining the resulting
AG-ol, because AG, unlike AG-ol, cannot be measured
directly by these methods. Therefore, the behavior of the
remaining AG residues during the reaction period
remains substantially unknown.
We have reported a method for the determination of
component monosaccharides of red algal galactans by
which all the component sugars including AG were
liberated quantitatively as diethyl dithioacetal (diethyl
mercaptal) derivatives by anhydrous mercaptolysis9,10)
(Fig. 1A). For structural analysis of red algal galactans,
we planned to design a simple method including
reductive mild acid hydrolysis and anhydrous mercap-
tolysis. Using this method, reducing end residues and
residues other than reducing ends should be measurable
independently as alcohols and diethyl dithioacetals
respectively in each monosaccharide, including AG
(Fig. 1B). Therefore, not only the resulting AG-ol at the
reducing ends but also native AG can be mearured
simultaneously. Conditions for the selective hydrolysis
of 3,6-anhydrogalatosidic bonds were examined by
tracing the amounts of AG and AG-ol, as well as other
component sugar derivatives, during the reaction period.
Key words: 3,6-anhydrogalactose; algal galactan; hetero-
geneity; mercaptolysis; reductive hydrolysis
Various biological activities have been found in
marine algal polysaccharides.1–3) However, since algal
polysaccharides in nature are generally heterogeneous in
size and composition, the structure-activity relationships
of these saccharides are not fully established yet.
Red algal galactans such as agar, carrageenans, and
porphyran contain 3,6-anhydrogalactose (AG) as a
major component monosaccharide.4) Since AG is easily
destroyed by harsh acidic conditions such as acid
hydrolysis and methanolysis, which are generally used
in compositional and structural analysis of polysaccha-
rides, the details of the structure of these algal galactans
remain to be elucidated.
y
To whom correspondence should be addressed. Tel/Fax: +81-952-28-8783; E-mail: hamay@cc.saga-u.ac.jp
Abbreviations: AcOH, acetic acid; AG, 3,6-anhydrogalactose; AG-ol, 3,6-anhydrogalactitol; EtSH, ethanethiol; Gal-ol, galactitol; GLC, gas-liquid
chromatography; HMDS, 1,1,1,3,3,3-hexamethyldisilazane; Man-ol, mannitol; 2-O-Me-AG, 2-O-methyl-3,6-anhydrogalactose; 6-O-Me-Gal, 6-O-
methyl-galactose; 6-O-Me-Gal-ol, 6-O-methyl-galactitol; MMB, 4-methyl morpholine borane; TFA, trifluoroacetic acid; TMCS, trimethylchlor-
osilane

1896
Y. HAMA et al.
A
B
Fig. 1. Scheme for the Structural Analysis of Red Algal Galactan Containing 3,6-Anhydrogalactose by Reductive Acid Hydrolysis and Anhydrous
Mercaptolysis.
A, Component monosaccharides of the galactan, including labile 3,6-anhydrogalactose, were liberated quantitatively as diethyl dithioacetal
derivatives by anhydrous mercatolysis. B, 3,6-Anhydrogalactosidic linkage in the algal galactan was cleaved by reductive mild acid hydrolysis.
The resulting oligosaccharide alditols were subjected to anhydrous mercaptolysis. Reducing end residues and residues other than at reducing
ends were liberated as alcohols and diethyl dithioacetals respectively, in the monosaccharides. After trimethylsilylation, these sugar deivatives
were separated and determined quantitatively by GLC. When the hydrolysis of 3,6-anhydrogalactosidic linkage but not galactosidic linkage is
completed, Gal diethyl dithioacetal and AG-ol are detected, as in B, with the formation of neither Gal-ol nor AG diethyl dithioacetal. Generally,
part of D-Gal is 6-O-methylated, and L-Gal is either 6-O-sulfated or in the 3,6-anhydride form. Gal6S, galactose 6-sulfate.
Agarose was used as a representative AG-containing galactan, because
its structure has been established to be a linear polymer of alternating
In addition, a method involving mild acid hydrolysis
followed by reduction with sodium borohydride was
also evaluated.
3-linked ꢀ-D-Gal and 4-linked ꢁ-L-Gal in which part of D-Gal is 6-O-
methylated and all of L-Gal occurs as L-AG. The degrees of hydrolysis
of anhydrogalactosidic bonds and other glycosidic bonds in the
hydrolyzate were evaluated by quantitative determination of the
Materials and Methods
component sugar diethyl dithioacetals (Gal, 6-O-Me-Gal, and AG:
Reagents. The reagents used in this study were of highest grade,
unless stated otherwise. The following reagents were purchased from
internal and non-reducing end residues) and their alcohols (Gal-ol,
6-O-Me-Gal-ol, and AG-ol: reducing end residues).
commercial sources: acetic acid (AcOH), trifluoroacetic acid (TFA),
ethanethiol (EtSH, first grade), MeOH, sodium borohydride (NaBH4,
Selective hydrolysis and simultaneous reduction with MMB (method
I). To an agarose sample (1 mg) dissolved in distilled water (0.9 ml),
MMB (8.1 mg/0.1 ml of distilled water) and 3 M TFA (0.11 ml) were
for amino acid analysis), 1,1,1,3,3,3-hexamethyldisilazane (HMDS),
and trimethylchlorosilane (TMCS) from Wako Pure Chemical Indus-
tries (Tokyo); hydrogen chloride-methanolic reagent 10 (for esterifi-
added. The resulting 0.3 M TFA solution containing MMB was heated
at 80 or 90 ꢀC for an appropriate period of time, and the hydrolyzate
cation) from Tokyo Chemical Industry (Tokyo); pyridine (silylation
grade) from Pierce (Rockford, IL); agarose I (for electrophoresis) from
Dojin (Kumamoto); and D-mannitol (D-Man-ol), D-Gal, D-galactitol
was neutralized with 1 M NH3. The solution was then separated by gel
filtration on Toyopearl HW-40 fine (1:0 ꢁ 30 cm ꢁ 2, Tosoh, Tokyo)
(D-Gal-ol), 6-O-methyl-D-galactose (6-O-Me-D-Gal), 3,6-anhydro-D-
galacstose (D-AG), and 4-methyl morpholine borane (MMB) from
with distilled water as the eluent in order to remove the excess reagents
in the solution. Elution of the carbohydrate components was monitored
Sigma-Aldrich (St. Louis, MO). 6-O-Methyl-D-galactitol (6-O-Me-D-
Gal-ol) and 3,6-anhhydro-D-galactitol (D-AG-ol) were prepared by
with a refractive index detector (Shimadzu RID-6A, Shimadzu, Kyoto,
Japan) and by TLC of the eluted fractions. Fractions containing
reducing 6-O-Me-D-Gal and D-AG with NaBH4, respectively. Each of
these sugar alchlols was purified by gel filtration on Sephadex G-15
(GE Healthcare, Tokyo) before use.
carbohydrate components were collected and lyophilized. The sample
was then redissolved in 1 ml of distilled water. A 100-ml aliquot was
taken, mixed in a glass vial with D-Man-ol (10 mg) as internal standard,
and lyophilized for GLC analysis.
Mild acid hydrolysis and reduction. The conditions for the
hydrolysis of 3,6-anhydrogalactosidic linkages in method I and method
II described below were essentially those of Stevenson and Furneaus.5)
Selective hydrolysis and subsequent reduction with NaBH4 (method
II). An agarose sample (100 mg) was mixed with D-Man-ol (10 mg) as

Selective Hydrolysis of 3,6-Anhydrogalactosidic Linkage
internal standard in a glass vial, and the sample was heated with 0.5 ml
1897
of 0.1 M TFA at 80 ꢀC for an appropriate period of time. The reactant
was then lyophilized to remove acid and solvent. The resulting
oligosaccharides were reduced to their oligosaccharide alditols by the
addition of 0.5 mg-NaBH4/0.1 ml-1 M NH3. After the solution was left
for 1 h at room temperature, excess NaBH4 was destroyed by the
addition of 3 M AcOH. The sample solution was then quickly passed
through a Dowex 50-x8 (Hþ form, Dow Chemical, Midland, MI)
column at 4 ꢀC, and the effluent collected was lyophilized. The boric
acid in the solid material was removed by repeated codistillation with
MeOH.
Anhydrous mercaptolysis. Anhydrous mercaptolysis of the samples
was carried out as described previously.9,10) Briefly, to the dried sample
(100 mg), 1.0 ml of 0.5 M HCl in EtSH:MeOH (2/1, v/v), which had
been prepared by mixing 1 volume of 1.5 M HCl/MeOH with 2
volumes of EtSH, was added. The sample was then heated at 60 ꢀC for
6 h, concentrated with a stream of nitrogen, and trimethylsilylated with
pyridine:TMCS:HMDS (5/1/1, v/v/v). Sugar composition was deter-
mined by GLC on a Shimadzu GC-14A (Shimadzu, Kyoto) equipped
with a DB-1 capillary column (0.25-mm thickness, 0:25 mm ꢁ 25 m,
J&W Scientific, Folsom, CA). Identification of sugar derivatives was
done by comparing their retention times on GLC with those of
authentic compounds and/or by EI-mass spectral analysis on GC-MS
(Shimadzu GC-MS QP-5000, Shimadzu).
Separation and characterization of oligosaccharide alditols. An
agarose sample (agarose I, 100 mg) was dissolved in 10 ml of distilled
water by heating in a boiling water bath. After cooling to around 40 ꢀC,
1.11 ml of 3 M TFA and 150 mg of MMB (solid) were added to form a
0.3-M TFA solution in the presence of MMB. The solution was heated
at 80 ꢀC for 90 min, followed by neutralization with AcOH. The
oligosaccharide alditols in the solution were concentrated by lyophi-
lization and redissolved in 2.0 ml of 0.1 M NH4HCO3. The sample was
fractionated by gel filtration on a Superdex 30 column (1:6 ꢁ 87:6 cm)
with 0.1 M NH4HCO3 as the eluent. An aliquot (10 ml) of the fractions
was spotted onto a Silicagel 60 TLC plate (Merck, Darmstadt,
Germany), and developed with 1-buthanol:AcOH:water (2:1:1, v/v/v).
The carbohydrates on the plate were visualized with diphenylamine
reagent.11,12) The hexose contents of the fractions were determined by
the anthrone method.13)
Results and Discussion
Reductive mild acid hydrolysis of agarose in the
presence of MMB (method I) and mild acid hydrolysis of
agarose followed by reduction with NaBH4 (method II)
Examples of typical chromatograms obtained by GLC
by method I (reductive hydrolysis) are shown in Fig. 2.
As described previously,9) since neither anomeric nor
structural isomers occur in diethyl dithioacetal deriva-
tives as in alditols, the components were detected as
single peaks (Fig. 2A). Therefore, the chromatograms
obtained were very simple as compared to those
obtained after methanolysis or acid hydrolysis. As can
be seen clearly in Fig. 2A, the AG in agarose was
detected as AG diethyl dithioacetal (peak e) along with
the dithioacetals of 6-O-Me-Gal and Gal (peaks f and g,
respectively). Only trace amounts of 2-O-Me-AG (peak
d) were detected. Figure 2B shows the results of
reductive hydrolysis at 80 ꢀC for 15 min. Even at this
early stage of hydrolysis, new peaks derived from AG
residues at the reducing ends appeared (peaks a and b,
2-O-Me-AG-ol and AG-ol respectively) along with
concomitant decreases in internal and non-reducing
end AG residues (peaks d and e, 2-O-Me-AG and AG
respectively). The ratio of hydrolyzed 3,6-anhydroga-
lactosidic linkages to total 3,6-anhydrogalactosidic link-
ages in the galactan can be calculated as follows:
Fig. 2. Compositional Analysis of the Reductive Hydrolyzate of
Agarose by GLC.
Agarose was hydrolyzed reductively in 0.3 M TFA in the presence
of MMB at 80 ꢀC for 0 min (A), 15 min (B), and 90 min (C). After
hydrolysis, the samples were mercaptolyzed anhydrously at 60 ꢀC
for 6 h. The monosaccharide derivatives thus obtained were
trimethylsilylated and analyzed by GLC. Column, DB-1; column
oven temperature, 180 ꢀC ! 240 ꢀC at 2 ꢀC/min; injection temper-
ature, 250 ꢀC; detector temperature, 280 ꢀC; carrier gas, He. Peaks a,
2-O-Me-AG-ol; b, AG-ol; c, Man-ol (internal standard); d, 2-O-Me-
AG; e, AG; f, 6-O-Me-Gal; g, Gal; ꢂ, phthalate (contaminated from
the screw cap of the sample vial).
[amount of AG-ol]/[sum of amounts of AG and AG-ol].
At the final stage of the reaction (Fig. 2C), most of AG
was detected as AG-ol (peak b), but AG did not
disappear completely (peak e). On the other hand, other
component sugars, such as Gal and 6-O-Me-Gal, were
detected dominantly as diethyl dithioacetals (peaks f and
g respectively) without the formation of corresoponding
alcohols throughout the reaction period.

1898
Y. HAMA et al.
300
300
A
B
200
100
0
200
100
0
0
30
60
90
120
300
600
900
0
30
60
90
120
Time (min)
Time (min)
Fig. 3. Time Course of the Reductive Hydrolysis of 3,6-Anhydrogalactosidic Linkages in Agarose.
Agarose was hydrolyzed reductively in 0.1 M TFA in the presence of MMB at 80 ꢀC (A) and at 90 ꢀC (B). After hydrolysis, the samples were
mercaptolyzed and analyzed by GLC as in Fig. 2. , AG-ol; , AG; , Gal; , Gal-ol; , 6-O-Me-Gal.
Table 1. Sugar Compositions of Agarose and Oligosaccharide Alditols from Agarose
6-O-Me-
Gal
2-O-Me-
AG
6-O-Me-
Gal-ol
2-O-Me-
AG-ol
Gal
AG
Gal-ol
AG-ol
Agarose
nmol
molar ratio
nmol
173.3
79.2
0.31
76.3
0.30
259.3
8.0
0.03
n.d.
—
n.d.
—
n.d.
—
n.d.
—
n.d.
—
0.67
171.8
0.68
1.00
11.3
0.04
Oligosaccharide
Alditols
n.d.
—
n.d.
—
252.2
1.00
7.7
0.03
molar ratio
Fr. 93 in Fig. 5
Fr. 87 in Fig. 5
molar ratio
molar ratio
0.66
1.36
0.34
1.12
n.d.
0.69
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
1.00
1.00
0.03
0.63
Oligosaccharide alditols were obtained by the reductive hydrolysis in 0.3 M TFA at 80 ꢀC for 90 min. A sample (100 mg of agarose or oligosaccharide alditols) was
mercaptolyzed anhydrously and analyzed as described in the text.
300
Changes in the GLC-determined amounts of the sugar
derivatives during the reaction period at 80 ꢀC are shown
in Fig. 3A. As the reaction period increased, the amount
of AG decreased, with concomitant increases in the
amount of AG-ol. After reductive hydrolysis for 15 min,
45% of the AG detected in the original sample
disappeared, whereas almost the same amount of new
AG-ol appeared (Fig. 3A, Fig. 2B). A plateau was
reached after 90 min, at which 95% of the AG was
detected as AG-ol (Fig. 3A, Fig. 2C). In contrast,
throughout the reaction period, the amounts of Gal and
6-O-Me-Gal remained basically unchanged, and the
corresponding alcohols were undetected (Fig. 2). These
data strongly suggest that anhydrogalactosidic bonds
were selectively hydrolyzed without hydrolyzing other
glycosidic bonds, and that the resulting oligosaccharides
with AG at the reducing end were reduced simulta-
neously to stable oligosaccharide alditols with AG-ol at
the reducing end. The sugar compositions of agarose and
the resultant oligosaccharide alditols obtained after
reductive hydrolysis at 80 ꢀC for 90 min are summarized
in Table 1. Based on the composition of the oligosac-
charide alditols, almost all of them were composed of
galactosyl-(3,6-anhydrogalactitol), in which one third of
Gal residues were 6-O-methylated.
200
100
0
0
60
120
180
240
300
Time (min)
Fig. 4. Time Course for the Hydrolysis of Glycosidic Linkages in
Agarose, Followed by Reduction with NaBH4.
The detailed conditions are described in the text. Symbols are the
same as in Fig. 3.
Therefore, the optimum conditions for the selective
hydrolysis of 3,6-anhydrogalactosidic bonds were con-
cluded to be 90 min at 80 ꢀC and 45 min at 90 ꢀC in 0.3 M
TFA in the presence of MMB. However, it is noted that
part (4–5%) of the AG was not converted to AG-ol even
on heating in 0.3 M TFA at 90 ꢀC for 120 min. It is
unclear why all the anhydrogalactosidic bonds were not
hydrolyzed, but this might be related to the observation
that the sulfated AG residues were more resistant to
acidic conditions than the non-sulfated AG residues.5,14)
Figure 4 shows the results for method II, which
involved mild acid hydrolysis in 0.1 M TFA at 80 ꢀC
Similar results were obtained when reductive hydroly-
sis was carried out at 90 ꢀC (Fig. 3B). Most of the 3,6-
anhydrogalactosidic linkages were selectively hydro-
lyzed until 30 min, and the oligosaccharides alditols
produced were stable up to a reaction time of 120-min,
but an apparent decrease in AG-ol was observed when
the reaction time exceeded 300 min. The amounts of Gal
and 6-O-Me-Gal decreased gradually after 120 min.

Selective Hydrolysis of 3,6-Anhydrogalactosidic Linkage
1899
4.0
3.0
maltose
v0
P-10
P-5
A
2.0
1.0
0
30
40
50
60
70
80
90
100
110
120
Fraction number (1.5 ml/tube)
B
43 49 51 54 57 59 62 64 66 68 70 72 74 76 78 81 83 87 90 93 95 98 102 Gal
Fig. 5. Gel Filtration of Oligosaccharide Alditols Obtained by Reductive Hydrolysis of Agarose (A) and TLC of the Eluted Fractions (B).
The detailed conditions are described in the text. Arrows indicate the elution positions of pulluran molecular weight makers (v0, (void
volume), P-200 (MW 186 000); P-10 (MW 12 200); P-5 (MW 5 300)) and maltose (MW 342). The Degrees of polymerization of the
oligosaccharide alditols were deduced from the elution volumes of these markers.
followed by NaBH4 reduction. The highest amount of
AG-ol was obtained at 120-min reaction, and the amount
is almost identical to those obtained by method I at
80 ꢀC for 90 min and at 90 ꢀC for 45 min (Fig. 3). These
results indicate that the destruction of AG in the acidic
condition was negligible until a reaction time of 120 min
at 80 ꢀC. However, the amount of AG-ol obtained by
method II did not make a plateau: after it reached a
maximum, it decreased with prolonged heating in 0.1 M
TFA at 80 ꢀC. Moreover, Gal-ol formation was apparent
for method II. We expect this formation to be accom-
panied by the destruction of AG at the reducing ends.
However, this method should be useful when oligosac-
charides with native AG but not AG-ol at the reducing
ends are required.
Kazlowski has reported methods for the preparation
and separation of neoagaro- and agaro-oligosaccharides
from agarose by ꢀ-agarase digestion and by mild acid
hydrolysis.15) Similar results, including the formation of
even-numbered oligosaccharides from red algal galac-
tans, have been reported by several other research-
ers.16,17) On the other hand, Yu18) and Yang et al.19)
reported recently the formation of odd-numbered oligo-
saccharides by mild acid hydrolysis of ꢂ-carrageenan
and both ꢂ-carrageenan and agarose, respectively. In
their studies, residues at the reducing end were deter-
mined to be Gal, not AG. These results appear to have
resulted from the destruction of AG at the reducing end,
as observed for meethod II. The latter researchers,19)
however, reported the exclusive formation of odd-
numbered oligosaccharide alditols obtained by reductive
hydrolysis in the presence of MMB.
profile of the reductive hydrolyzate of agarose and TLC
of the eluted fractions. As expected, most of the
hydrolyzate was composed of dimers, but tetramers
and other oligomers were also detected (typically, Fr.
78–90, Fig. 5). The occurrence of oligosaccharides other
than dimers clearly indicates the presence of structural
heterogeneity in agarose, although agarose is believed to
be a linear polymer of alternating 3-linked ꢀ-D-Gal and
4-linked ꢁ-L-Gal in which part of D-Gal is 6-O-
methylated and all the L-Gal occurs as L-AG. The molar
ratios of component sugars of Frs. 93 (dimer) and 87
(tetramer and others) are shown in Table 1. As described
above, sulfated AG residues have been reported to be
more resistant to acidic conditions than non-sulfated
AG ones,5,14) and the sugar composition in Fr. 87
suggests a D-Gal!L-AG (sulfated)!D-Gal!2-O-Me-
L-AG sequence for agarose, where half of D-Gal is 6-O-
methylated. The presence of substantial amounts of 2-O-
Me-AG-ol and native AG in Fr. 87 (tetramers and
others) but not in Fr. 93 (dimers) may be a clue to
understanding the detailed structure of agarose.
References
1) Campo VL, Kawano DF, Silva JrDBd, and Carvalho I,
Carbohydr. Polym., 77, 167–180 (2009).
2) Ghosh T, Chattopadhyay K, Marschall M, Karmakar P, Mandal
P, and Ray B, Glycobiol., 19, 2–15 (2009).
3) Arad SM, Ginzberg A, and Huleihel M, ‘‘Biomaterials from
Aquatic and Terrestrial Organizations,’’ eds. Fingerman M
and Nagabhusan R, Science Publishers, Enfield, pp. 37–62
(2006).
4) Yalpani M, ‘‘Polysaccharides: Synthesis, Modifications and
Structure/Property Relations,’’ Elsevier, Amsterdam, pp. 9–44
(1988).
5) Stevenson TT and Furneaux RH, Carbohydr. Res., 210, 277–
298 (1991).
Structural heterogeneity in agarose
The optimum conditions for the selective hydrolysis
of 3,6-anhydrogalactosidic bonds obtained in this study,
i.e., heating in 0.3 M TFA in the presence of MMB at
80 ꢀC for 90 min, was applied to agarose, and the
resulting oligosaccharide alditols were separated by gel
filtration on Superdex 30. Figure 5 shows the elution
6) Falshaw R and Furneaux RH, Carbohydr. Res., 252, 171–182
(1994).
7) Chiovitti A, Bacic A, Craik DJ, Munro SLA, Kraft GT, and Liao
ML, Carbohydr. Res., 299, 229–243 (1997).
8) Jol CN, Neiss TG, Penninkhof B, Rudolph B, and De Ruiter
GA, Anal. Biochem., 268, 213–222 (1999).

1900
Y. HAMA et al.
9) Hama Y, Nakagawa H, Kurosawa M, Sumi T, Xia X, and
15) Kazlowski B, Pan CL, and Ko YT, Carbohydr. Res., 343, 2443–
2450 (2008).
Yamaguchi K, Anal. Biochem., 265, 42–48 (1998).
10) Hama Y, Nakagawa H, Mochizuki K, Sumi T, and Hatate H,
J. Biochem., 125, 160–165 (1999).
16) Caram-Lelham N, Carbohydr. Res., 273, 71–76 (1995).
17) Chen HM, Zheng L, Lin W, and Yan XJ, Talanta, 64, 773–777
(2004).
11) Harris G and MacWilliam IC, Chem. Ind. Lond., 249 (1954).
12) Andersons K, Li SC, and Li YT, Anal. Biochem., 287, 337–339
(2000).
18) Yu G, Guan H, Ioanoviciu AS, Sikkander SA, Thanawiroon C,
Tobacman JK, Toida T, and Linhardt RJ, Carbohydr. Res., 337,
433–440 (2002).
13) Morris DL, Science, 107, 254–255 (1948).
14) Quemener B and Lahaye M, J. Appl. Phycol., 10, 75–81
(1998).
19) Yang B, Yu G, Zhao X, Jiao G, Ren S, and Chai W, FEBS J.,
276, 2125–2137 (2009).