Glycobidogy vol. 7 no. 2 pp. 253-263, 1997
The unusual tetrasaccharide sequence GlcApi-3GalNAc(4-suIfate)($l4GlcA(2-sulfate)pl-3GalNAc(6-sulfate) found in the hexasaccharides prepared by
testicular hyaluronidase digestion of shark cartilage chondroitin sulfate D
Satomi Nadanaka and Kazuyuki Sugahara1
Department of Biochemistry, Kobe Pharmaceutical University,
Higashinada-ku, Kobe 658, Japan
'To whom correspondence should be addressed
Eight hexasaccharide fractions were isolated from commercial shark cartilage chondroitin sulfate D by means of
gel nitration chromatography and HPLC on an aminebound silica column after exhaustive digestion with sheep
testicular hyaluronidase. Capillary electrophoresis of the
enzymatic digests as well as one- and two-dimensional 500
MHz 'H-NMR spectroscopy demonstrated that these hexasaccharides share the common core saccharide structure
GIcApi-3GalNAcpi-4GlcApl-3GalNAcpi-4GlcApi3GalNAc with three, four, or five sulfate groups in different
combinations. Six structures had the same sulfation profiles as those of the unsaturated hexasaccharides isolated
from the same source after digestion with chondroitinase
ABC (Sugahara etaL, Eur. J. Biochem., 293,871-880,1996)
and the other two have not been reported so far. In the new
components, a D disaccharide unit, GlcA(2-sulfate)pi3GalNAc(6-suIfate), characteristic of chondroitin sulfate D
was arranged on the reducing side of an A disaccharide
unit, GlcApi-3GaJNAc(4-sulfate), forming an unusual A-D
tetrasaccharide sequence, GlcApi-3GalNAc(4-sulfate)pl4GIcA(2-sulfate)pi-3GaINAc(6-sulfate) which is known
to be recognized by the monoclonal antibody MO225.
These findings support the notion that the tetrasaccharide sequence, GlcApi-3GalNAc(4-sulfate)pi^lGlcApi3GalNAc(6-sulfate) is included in the acceptor site of a
hitherto unreported 2-O-sulfotransferase responsible for its
synthesis. The sulfated hexasaccharides isolated in this
study will be useful as authentic oligosaccharide probes and
enzyme substrates in studies of sulfated glycosaminoglycans.
Key words: sulfated hexasaccharides/chondroitin sulfate
D/hyaluronidase/1 H-NMR
Introduction
Chondroitin sulfate (CS) proteoglycans are composed of CS
chains covalently bound to a protein core, and are widely distributed among tissues exhibiting a variety of biological functions. The molecular organization of CS chains consists of
repeating disaccharide units, each containing a glucuronic acid
(GlcA) and an N-acetylgalactosamine (GalNAc) residue.
Chemical heterogeneity resulting from diversity in estersulfation reactions accounts for structural differences in CS
chains. The expression of CS isoforms with concomitant spe© Oxford University Press
cific spatio-temporal patterns in various tissues suggests that
CS isoforms differing in sulfation position and degree perform
distinct functions in development (Mark et al., 1989; Sorrell et
al., 1990). The disappearance of specific antibody epitopes
upon enzymatic molecular digestion of CS chains in proteoglycans indicates that chemical heterogeneity can occur in discrete domains located at different sites along the chains (Sorrell
et al., 1993). Thus, structural analysis of CS chains is of critical
importance in studying the functions of CS proteoglycans.
The monoclonal antibody (mAb) 473HD recognizes CS or
dermatan sulfate (DS) proteoglycans, which are located on the
surface of immature glia cells in the central nervous system and
which promotes neurite outgrowth (Faissner et al., 1994).
Chondroitin sulfate C (CS-C) from shark cartilage suppresses
the interaction of mAb 473HD with the neural CS/DS proteoglycans. Shark cartilage CS chains possess a unique disulfated
disaccharide unit GlcA(2S)-GalNAc(6S) (D-unit) (Suzuki,
1960), which may form part of the epitope structure. The Dunit has been identified not only in shark cartilage, but also in
mouse mast cells derived from immune lymph nodes (Davidson et al., 1990) and in the basement membrane of mouse tooth
germ for a brief period (Mark et al., 1990). Thus, the structural
arrangement of the D disaccharide unit along CS chains, its
function and biosynthetic mechanism are of particular interest.
Determination of the precise structure of glycosaminoglycans is a difficult task in general since the structure is usually
heterogeneous and polydisperse. This difficulty is primarily
due to the lack of a systematic analytical procedure. We have
been investigating the detailed structures of various CS isoforms. So far, oligosaccharides such as tri-, tetra-, and hexasaccharides have been prepared from various CS isoforms,
including shark cartilage CS-D using the bacterial eliminase
chondroitinase ABC (Sugahara et al., 1994a,b, 1996a) and
structurally analyzed by a combination of an enzymatic digestion and 'H-NMR spectroscopy. CS chains are depolymerized
with specific bacterial chondroitinases ABC and AC into oligosaccharides bearing an unsaturated uronic acid residue at
their nonreducing terminus. However, unsaturated oligosaccharides are limited in their usefulness as enzyme substrates,
especially for biosynthetic enzymes, due to the unnatural unsaturated bond. Thus, saturated oligosaccharides will be indispensable as probes for elucidating the biological functions of
CS chains or as substrates for assaying CS-synthesizing enzymes.
A new analytical procedure has been developed for characterizing saturated oligosaccharides, in which reducing saccharides are derivatized with 2-aminoacridone (AMAC) and the
products are separated by means of capillary electrophoresis
(CE) (Kitagawa et al., 1995a). In this study, we isolated eight
hexasaccharides from shark cartilage CS-D after hyaluronidase
digestion. We used them to investigate the sequential arrange253
S.Nadanaka and K.Sugahara
ment of the D disaccharide units in the polysaccharide and to
prepare oligosaccharide probes for studying CS chains. Their
structures were determined by fluorophore-tagging in combination with CE and by 'H-NMR spectroscopy.
Results
Isolation of the hexasaccharides
A commercial preparation of shark cartilage CS-D was digested
with sheep testicular hyaluronidase, then fractionated by gel filtration into tetra-, hexa-, octa-, deca-, dodeca, and tetradecasaccharides using a column containing Bio-Gel P-10 as reported
(Sugahara et al., 1996b). They represented 22.5, 28.2, 20.2, 13.3,
9.1, and 6.7% of the resultant oligosaccharides, respectively,
based on the absorbance at 210 run. The hexasaccharide fraction
was subfractionated into fractions 1 to 1A by HPLC on an aminebound silica column as indicated in Figure 1. They were further
subfractionated and purified by rechromatography. The molar ratios of the subtractions were as follows: fractions 2-1:2-2, 25:75;
fractions 3-1:3-2, 19:81; fractions 5-1:5-2, 41:59; fractions 6-1:62, 77:23; fractions 1-1:1-2:1-3:1 A, 11:19:62:8. Among these, the
major fractions 1, 2-2, 3-2, 4, 5-2, 6-1, and 7-3 were reasonably
pure as examined by HPLC (not shown) and CE (Figure 2), and
structurally analyzed. The yields of the major fractions are summarized in Table I.
Disaccharide composition analysis of the isolated
hexasaccharides by CE
The disaccharide composition of the isolated hexasaccharides
was determined by our micro analytical procedure (Kitagawa
1
4
E
2
3 4 5 6 7
\\i±l(
c
o
et al., 1995a). The chondroitinase AC-II digest of each hexasaccharide was labeled with the fluorophore, AMAC, then disaccharides were analyzed by CE. Chondroitinase AC-II,
which is a bacterial eliminase, should degrade a CShexasaccharide to yield 2 mol of unsaturated disaccharide units
and one mole of a saturated disaccharide unit derived from the
non-reducing terminus. The high resolution CE quantitatively
resolves each authentic unsaturated disaccharide and the corresponding saturated disaccharide with a sulfation profile identical to that of the former. Therefore, this procedure not only
gives the disaccharide composition, but also identifies the disaccharide unit located at the nonreducing terminus (Kitagawa
et al., 1995a).
CE analysis of the chondroitinase AC-II digest of fraction
2-2 after the AMAC-derivatization demonstrated AMACderivatives of Di-6S, ADi-6S, and ADi-4S in a molar ratio of
1.00:1.00:0.90 (Figure 3A). The molar ratio was determined
using the integrated peak areas reported for each standard CSdisaccharide (Kitagawa et al., 1995a). The disaccharides, Di6S and ADi-6S, which share the same sulfation profile and
differ only in the nonreducing terminal uronic acid residues in
terms of saturation or unsaturation, were distinctly separated as
described. We judged that the saturated disaccharide unit Di6S was derived from the nonreducing terminus. The sequential
arrangement of the two unsaturated disaccharide units, ADi-6S
and ADi-4S, was determined using chondro-4-sulfatase and
chondro-6-sulfatase. The latter enzyme removes only a sulfate
group on C6 of the GalNAc residue at the reducing ends of
oligosaccharides (Sugahara et al., 1994a; Sugahara and Kojima, 1996). Although the former enzyme can remove sulfates
from internal GalNAc residues as well as from a GalNAc
residue at the reducing end, it acts preferentially on the latter
under standard incubation conditions (Sugahara and Kojima,
1996). Fraction 2-2 was resistant to chondro-6-sulfatase, but
sensitive to chondro^4-sulfatase under standard incubation conditions (data not shown), indicating that the GalNAc residue at
the reducing end is not 6-sulfated, but 4-sulfated. Hence, the
following structure is proposed for the compound in fraction
2-2:
GlcA^ l-3GalNAc(6S)P l ^ G l c A P l-3GalNAc(6S)P 14GlcApi-3GalNAc(4S)
CE analysis of the chondroitinase AC-II digest of fraction 4
after the AMAC-derivatization showed AMAC-derivatives of
Di-diSD and ADi-6S in a molar ratio of 1.00:2.29 (Figure 3B).
Di-diS D was judged to be derived from the nonreducing terminus. Hence, the following structure is proposed for the major
hexasaccharide in fraction 4:
Glc A(2S)P l-3GalNAc(6S)P l ^ G l c A0 l-3GalNAc(6S)P 14GlcA£l-3GalNAc(6S)
20
30
40
50
TIME (min)
60
O
O
Fig. 1. HPLC fractionation of the hexasaccharides. The hexasaccharide
fraction obtained by gel filtration on Bio-Gel P-10 (Sugahara et al, 1996b)
was fractionated by HPLC on an amine-bound silica column using a linear
gradient of NaH-jPO, as indicated by the dashed line. The elution positions
of the authentic unsaturated CS-hexasaccharides (Sugahara et al., 1996a) are
indicated by arrows at the top of the column.
1, AU-G<6S)-U-G{6S)-U-G(6S); 2, AU-G{6S)-U-G{6S)-U-G(4S);
3, AU-G(6S)-U-G(4S)-U-G(4S); 4, AU(2S)-G<6S)-U-G(6S)-U-G(oS);
5, AU(2S)-G{6S)-U-G(6S>-U-G(4S); 6, AU(2S)-G(6S)-U-G(4S)-U-G{4S);
7, AU-G(4S)-U(2S)-G(6S)-U-G(4S). AU, A" J HexA; U, GlcA; G, GalNAc.
254
CE analysis of the chondroitinase AC-II digest of fraction
7-3 after AMAC-derivatization showed AMAC-derivatives of
Di-diSD, ADi-diSD, and ADi^S in a molar ratio of 1.00:1.00:
0.87 (Figure 3C). Fraction 7-3 was sensitive to chondro-6sulfatase, but resistant to chondro-4-sulfatase under standard
incubation conditions (data not shown). The enzyme specificities indicated that one of the GalNAc(6S) residues was located
at the reducing terminus. Hence, the following structure is
proposed for the compound in fraction 7-3:
GlcA(2S)pi-3GalNAc(6S)pi^GlcApi-3GalNAc(4S)pi4GlcA(2S)P l-3GalNAc(6S)
Hexasaccharides from shark cartilage chondroitin sulfate D
Fr. 1
Fr. 2-2
Fr. 3-2
Fr. 4
Fr. 5-2
Fr. 6-1
Fr.7-3
E
c
in
CO
ID
O
<
CO
CE
O
en
CO
<
J
_JJ
10.5
115
10.5
115 10.5
115
10.5
115 10.5
115 10.5
115 10.5
115
TIME (mln)
Fig. 2. CE analysis of the isolated hexasaccharides. Each hexasaccharide fraction (0.5 nmol) obtained by HPLC was analyzed by CE as described in
Materials and methods
CE analysis of the chondroitinase AC-II digest of fraction 1
after the AMAC-derivatization yielded AMAC-derivatives of
Di-6S and ADi-6S in a molar ratio of 1.00:1.90 (Table I).
Hence, the following structure is proposed for the compound in
fraction 1:
molar ratio of 1.00:0.05:0.17:1.70:0.27:0.05 (Table I), indicating that this fraction contains minor impurities and that the
disaccharide units derived from the major compound in this
fraction are Di-6S and ADi-4S. The major compound in fraction 3-2 was resistant to chondro-6-sulfatase, but sensitive to
chondro-4-sulfatase. Hence, the following structure is proposed for the major compound in fraction 3-2:
GlcA3 l-3GalNAc(6S)P l ^ G l c A p l-3GalNAc(6S)P 14GlcA(3 l-3GalNAc(6S)
l-3GalNAc(6S)3 l-4GlcAp l-3GalNAc(4S)(314GlcApi-3GalNAc(4S)
CE analysis of the chondroitinase AC-II digest of fraction
3-2 after the AMAC-derivatization yielded AMAC-derivatives
of Di-6S, Di-4S, Di-diSD, ADi-4S, ADi-6S, and ADi-diSD in a
CE analysis of the chondroitinase AC-II digest of fraction
Table I. Disaccharide composition analysis of the isolated hexasaccharides
Disaccharides formed (molar ratio)b
Fr.
nmol'
Saturated
Unsaturated
Proposed structures of the major components0
1
2-2
381
562
Di-6S (1.00)
Di-6S (1.00)
U-G(6S)-U-C(6S)-U-G(6S)
U-G(6S)-U-G(6S)-U-G<4S)
3-2
346
4
5-2
398
635
Di^S (0.05)
Di-6S (1.00)
Di-diSD (0.17)
Di-diSD (1.00)
Di-6S (0.57)
Di-diSD (1.00)
6-1
7-3
318
335
ADi-6S (1.90)
ADi-4S (0.90)
ADi-6S (1.00)
ADMS (1.70)
ADi-6S (0.27)
ADi-diSD (0.05)
ADi-6S (2.29)
ADi-4S (1.41)
ADi-6S (1.26)
ADi-diSD (0.40)
ADMS (1.80)
ADi-4S (0.87)
ADi-diSD (1.00)
Di-diSD (1.00)
Di-diS D (1.00)
U-G(6S)-U-G(4S)-U-G(4S)
U(2S)-G(6S)-U-G(6S)-U-G<6S)
U(2SHX6S)-U-G(6S)-U-G(4S)
and
U-G(6S)-U-G<4S)-U(2S)-G(6S)
U(2S)-G<6S)-U-G(4S)-U-G(4S)
U(2S)-G(6S>U-G(4S)-U(2SK3(6S)
The hexasaccharide fractions were digested with chondroitinase AC-IL The products were labeled with AMAC, then analyzed by CE as described under
Materials and methods.
•Yield from 100 mg of CS-D.
"The molar ratios of the resultant disaccharides were calculated based upon the peak area of the AMAC-disacchandes obtained in CE and are expressed as
ratios relative to the major saturated disaccharide.
n j , G, 2S, 4S, and 6S represent GIcA, GalNAc, 2-, 4-, and 6-O-sulfate, respectively.
255
S.Nadanaka and K-Sugahara
1
to
CM
B
to
5-2 after the AMAC-derivatization demonstrated AMACderivatives of Di-6S, Di-diSD, ADMS, ADi-6S, and ADi-diSD
in a molar ratio of 0.57:1.00:1.41:1.26:0.40 (Table I). Since we
found AMAC-derivatives of two saturated disaccharides, Di6S and Di-diSD, we judged that fraction 5-2 contained two
compounds corresponding to the subtractions 5-2A and -2B in
a molar ratio of 36:64, bearing these disaccharide units at the
nonreducing ends. The results indicated that the disaccharide
units derived from the compound in the major fraction 5-2A
were Di-diSD, ADi-4S, and ADi-6S whereas those derived
from the compound in the minor fraction 5-2B are EH-6S,
ADi-4S, and ADi-diSD. Since chondro-6-sulfatase or chondro4-sulfatase removed a sulfate group from 27 or 68% of fraction
5-2, respectively, as judged by HPLC (data not shown), the
compounds in fractions 5-2A and 5-2B have 4- and 6-sulfated
GalNAc residues at the reducing terminus, respectively. Based
upon these results, the following structures are proposed for
these compounds in fractions 5-2A and 5-2B:
Fraction 5-2A:
Q
m
o
GlcA(2S)P l-3GalNAc(6S)P l ^ G l c A p l-3GalNAc(6S)P 14GlcApi-3GalNAc(4S)
m
<
Fraction 5-2B:
CO
GlcApi-3GalNAc(6S)pi^GlcApi-3GalNAc(4S)pi4GlcA(2S)P l-3GalNAc(6S)
O
CO
m
O
<
0
10
20
30
40
TIME (min)
Fig. 3. Disaccharide composition analysis of fractions 2-2, 4, and 7-3 by
CE. Each fraction (1 nmol) was digested with chondroitinase AC-II. The
digest was derivatized with AMAC, then analyzed by CE as described in
Materials and methods. The elution positions of the AMAC derivatives of
the authentic CS-<lisaccharides are indicated at the top of (A). (A), fraction
2-2, (B) fraction 4, (C) fraction 7-3.
6-1 after the AMAC-derivatization showed AMAC-derivatives
of Di-diS D and ADMS in a molar ratio of 1.00:1.80 (Table I).
This suggested that Di-diS D is derived from the nonreducing
end of a hexasaccharide. Hence, the following structure is proposed for the compound in fraction 6-1:
GlcA(2S)pi-3GalNAc(6S)pi^GlcApi-3GalNAc(4S)pi4GlcAp l-3GalNAc(4S)
CE analysis of the chondroitinase AC-II digest of fraction
256
500 MHz 'H-NMR spectroscopy
The hexasaccharides in fractions 1, 2-2, 3-2, 4, 6-1, 7-3 were
characterized by 500 MHz ' H-NMR spectroscopy to confirm
the structures proposed above. The ID spectra of fractions 4
and 7-3 are shown in Figure 4, A and B. The resonances
observed between 8 4.4 and 5.3 ppm are characteristic of anomeric protons, whereas those at around 8 2.0 ppm are characteristic of the 7V-acetyl group protons of GalNAc. Signals found
in the anomeric proton region between 4.4 and 5.3 ppm were
identified as H-l resonances of the constituent saccharide residues by comparison with the NMR spectra of saturated CStetrasaccharides (Sugahara et al., 1996b) and unsaturated CShexasaccharides (Sugahara et al., 1996a). Since the H-l resonances of f}GalNAc-l_ and the H-4 resonances of 4-0-sulfated
GalNAc residues of the hexasaccharides in all the fractions
were hidden under the HOD signal when recorded at 26°C, the
spectra were also recorded at 15°C (for example, see the inset
of Figure 4B).
Chemical shifts of the H-l signals of GalNAc-3 and GalNAc-5 of each hexasaccharide could not be readily discriminated, although their chemical shift values were significantly different. The H-l signal that has a chemical shift value
closer to that of GalNAc-3 H-l of the corresponding reference tetrasaccharide (Sugahara et al., 1996b) was assigned to
GalNAc-5 H-l of each hexasaccharide. For example, in the
hexasaccharide in fraction 7-3, the H-l signals found at 8 4.585
and 4.553 were assigned to GalNAc-3 and GalNAc-5, respectively (Figure 4B). Anomerization effects of otpGalNAc-1.
doubled the anomeric proton signal of the second constituent
PGlcA-2. For example, the anomeric resonances of PGlcA-2
of the hexasaccharide in fraction 4, which exists in equilibrium
of a- and fi-anomers, were observed at 8 4.566 and 4.511
(Figure 4A), and assigned to pGlcA-2 H-l of the a- and
P-hexasaccharides, respectively, by comparison with the
Hexasaccharides from shark cartilage choadroitin sulfate D
Fr. 4
5.4
4.6
4.8
52
Fr. 7-3
A
2-Sulfate 6-Sulfate
6-Sulfate
6-Sulfate
i -3GalN Acpi -4GlcApi -3GalN Acpi -4GlcApi -3GalNAc
4.4
42
4.0
3.6
3.6
3.4
3.2
2.2
2.0 ppm
2rSulfate 6-Sulfate
4-Sulfate 2-Sulfate 6-Sulfate
GlcApi -3GalNAcpi -4GlcApi -3GalNAcpi -4GlcApi -3GalNAc
fi
5
4
3
2
1
•nomerie protons
Fig. 4. One-dimensional 500 MHz 'H NMR spectra of fractions 4 and 7-3. The 500 MHz 'H-NMR spectra of fractions 4 (A) and 7-3 (B) were recorded in
D2O at 26°C and 15°C, respectively. The inset in (B) is the spectrum recorded at 15°C. The numbers and letters in the spectra refer to the corresponding
residues in the structures. For abbreviations, see the caption to Figure 1.
chemical shifts of GlcA-2 H-1 of the tetrasaccharide GlcAfil3GalNAc(6S)pi^GlcApi-3GalNAc(6S) (Sugahara et al.,
1996b).
Other proton signals in the ID spectrum were assigned using
the 2D HOHAHA and COSY spectra starting with the H-1
resonances of the sugar residues of each hexasaccharide as described for CS oligosaccharides (Sugahara et al., 1994a,b,
1996a). The NMR data are summarized in Tables II and HI
257
S.Nadanaka and K.Sugahara
together with those of the reference compounds. As representatives, the COSY spectra of fractions 4 and 7-3 are shown in
Figure 5, A and B. In fraction 4, beginning at 5 5.215 for the
H-l proton of otGalNAc-J_, we found a cross peak showing
connectivity to the H-2 resonance at 8 4.281. Connectivity of
the H-2 resonance to that of H-3 at 8 4.035 was also established. Continuation of this process allowed localization of the
H-4, H-5, H-6, and 6' resonances, hi a similar fashion, starting
with the H-l resonance of each of the other sugar residues,
most of other proton signals were assigned. Although no cross
peak was detected between the GalNAc-3 H-4 and H-5 or
between the GalNAc-5 H-4 and H-5, H-6 signals were found at
8 4.23 in the ID spectrum by comparison with the corresponding signals belonging to GlcA(31-3GalNAc(6S)pi^GlcA|313GalNAc(6S) and GlcA(2S)(31-3GalNAc(6S^l-4GlcA£l3GalNAc(6S) (Sugahara et al., 1996b). Likewise, most of the
proton signals were also assigned.
Modification by 0-sulfation causes downfield shifts of protons bound to the O-sulfated carbon atoms by approximately
0.4—0.7 ppm (Harris and Turvey, 1970). Thus, the sulfation
positions of saccharide residues were determined by comparison with the 'H-NMR data of nonsulfated saccharide residues
(Yamada et al., 1992). In the spectrum of fraction 4 (Figure
4A), we found a downfield shift of H-6, 6' of GalNAc-J_, 3 and
Table n. ' H-Chemical sniffs of structural-reporter groups of the constituent monosaccharides of the isolated hexasacchandes
Fr. 1
U-G-U-G-U-G
1
1 1
6S 6S 6S
GalNAc-i
GlcA-2
H-l
H-2
H-3
H^
H-5
H-6
H-6'
NAc
H-l
H-2
H-3
H-4
H-5
GalNAc-3
GlcA-4
H-l
H-2
H-3
H-4
H-5
H-6
H-6'
NAc
H-l
H-2
H-3
GalNAc-5
GlcA-6
(AHexA-6)
H^l
H-5
H-l
H-2
H-3
H-4
H-5
H-6
H-6'
NAc
H-l
H-2
H-3
H^t
H-5
Fr. 2-2
U-G-U-G-U-G
1
1 1
6S 6S 4S
Rl
AU-G-U-G-U-G
1
1
1
6S 6S 4S
R2
AU-G-U-G-U-G
1
1
1
6S 4S 4S
Fr. 3-2
U-G-U-G-U-G
1
1 1
6S 4S 4S
a
P
a
P
a
P
a
P
a
P
5.215
4.281
4.022
4.681
4.00
3.83
5.209
4.326
4.19
4.810"
4.254
3.78
3.71
4.710*
4.02
N.D.
4.74'
N.D.
N.D.
N.D.
5.209
4.324
4.19
4.811"
4.253
3.78
3.71
4.711"
4.02
N.D.
4.741*
N.D.
N.D.
NX).
5.209
4.328
4.188
4.808"
4.255
4.710*
4.04
3.99
4.740*
N.D.
5.209
4.326
4.19
4.807*
4.256
3.78
3.70
4.708*
4.02
N.D.
4.74*
N.D.
N.D.
N.D.
4.482
3.412
4.531
3.409
4.482
3.413
4.521
4.476
4.522
4.23"
4.342
4.22
4.14
2.015
4.566
2.017
4.508
3.382
3.598
3.70
N.D.
4.535
4.02
3.83
4.17"
3.93 c2
4.22
N.D.
2.015
4.491
3.365
3.579
3.686
N.D.
4.535
3.99
3.83
4.22"
3.9T2
4.22
4.21
2.021
4.494
3.315
3.475
3.686
N.D.
4.530
3.408
3.596
3.756
N.D.
4.533
3.971
3.83
4.178
N.D.
4.22
4.21
2.017
4.497
3.365
3.580
3.719
N.D.
4.551
4.020
3.83
4.22
N.D.
4.22
4.21
2.022
4.491
3.314
3.475
3.681
N.D.
N.D.
N.D.
2.022^
2.017
3.594
N.D.
N.D.
N.D.
4.550
4.02
3.835
4.180
N.D.
4.22
4.21
2.017
4.497
3.366
3.576
N.D.
N.D.
4.568
4.02
3.935
4.171
3.99
4.22
4.21
2.052
5.180
3.782
4.100
5.877
—
3.397
3.582
3.780
3.662
4.570
4.04"
4.0"
4.740*
N.D.
N.D.
N.D.
2.025 c3
4.465
3.379
3.582
3.735
3.662
4.554
4.020
3.840
4.230
3.97
4.22
N.D.
2.029*3
4.495
3.316
3.477
3.680
N.D.
2.022c4
4.476
3.397
3.58
3.76
3.66
4.570
4.01
3.93
4.74*
N.D.
3.78
3.78
2.03 l c4
4.464
3.379
3.57
3.75
3.66
4.588
4.05
3.99
4.210
N.D.
4.211
4.211
2.056
5.183
3.774
4.103
5.877
—
Chemical shifts are given in ppm downfield from internal sodium 4,4-dimethyl-4-silapentane-l-sulfonate, but were actually measured indirectly to acetone in
2
HjO (5 2.225) at 26°C. Chemical shifts were determined from the clearly resolved signals in ID 'H-NMR spectra or from cross-peaks in the COSY spectra.
They are summarized with those of reference unsaturated hexasaccharides isolated from shark cartilage CS-D (Sugahara el al., 1996a):
R1, AHCAAC l-3GalNAc(6S)P 1-4G1CAP 1 -3GalNAc(6S)01 -4GIcAP 1 -3GalNAc(4S);
R2, AHexAal-3GalNAc(6S)pl-4GlcApi-3GalNAc(4S)pi-4GlcApl-3GalNAc(4S). The estimated error for the values is ±0.001 ppm.
'Values determined at 15°C.
"The estimated error for the values to two decimal places is ±0.01 ppm because of partial overlap of signals.
cl 4
~ Tbe two respective values may be interchanged.
ND, Not determined.
—, Not occurring.
258
Hexasaccharides from shark cartilage chondroitin sulfate O
Table m.
'H-Chemical shifts of structural-reporter groups of the constituent monosaccharides of the isolated hexasaccharides
Fr. 4
2S
R3
2S
Fr. 6-1
2S
1
I
GalNac-1
GlcA-2
GalNAc-3
GlcA-4
GalNAc-5
GlcA-6
(AHexA-6)
H-l
H-2
H-3
H^*
H-5
H-6
H-6'
NAc
H-l
H-2
H-3
H-4
H-5
H-l
H-2
H-3
H-4
H-5
H-6
H-6'
NAc
H-l
H-2
H-3
H-4
H-5
H-l
H-2
H-3
H-4
H-5
H-6
H-6'
NAc
H-l
H-2
H-3
H^
H-5
1
U-G-U-G-U-G
1
1 1
6S 4S 4S
AU-G-U-G-U-G
1
1 1
6S 6S 4S
R4
2S
|1
AU-G-U-G-U-G
1
1 1
6S 4S 4S
7-3
2S
I
I
U-G-U-G-U-G
1
1 1
6S 6S 6S
FT.
2S
i
I
I
i
U-G-U-G-U-G
1
1 1
6S 4S 6S
a
P
a
P
a
P
a
P
a
P
5.215
4.281
4.035
4.234
4.347
4.220
4.14"
4.683
4.00
3.84
N.D.
N.D.
N.D.
N.D.
5.199*
4.324"
4.184*
N.D.
N.D.
3.78*
3.70*
4.71"
4.00*
4.06*
4.74"
N.D.
N.D.
N.D.
5.209
4.321
4.18
4.807"
4.253
3.78
3.70
4.707*
4.02
N.D.
4.738"
N.D.
N.D.
N.D.
5.324
4.186
4.11
4.291
4.321
4.21
4.15
4.721
4.130
3.82
N.D.
N.D.
N.D.
N.D.
5.208
4.326
4.19
4.805"
4.255
3.78
3.71
4.707*
4.02
N.D.
4.74"
N.D.
N.D.
N.D.
2.015
4.566
4.511
3.383
3.594
3.71
N.D.
4.537
4.015"
3.83
4.17c2
N.D.
4.23
4.23
2.015
4.494
3.359
3.574
3.71
N.D.
4.537
3.970"
3.83
4.23 s2
3.97
4.23
4.23
2.036
4.716
4.077
3.731
3.594
N.D.
2.016^
4.51"
3.3841
3.58*
3.77*
3.65*
4.556"
4.00*
N.D.
4.74*
N.D.
N.D.
N.D.
2.027MJ
4.459*
3.365*
3.58"
3.726*
3.64*
4.54*
3.94"
N.D.
4.262"
N.D.
N.D.
N.D.
2.035**3
4.72"
4.069"
3.726"
3.56'
N.D.
2.016
4.531
3.410
4.485
3.414
3.593
N.D.
N.D.
4.549
4.01
3.83
4.18
N.D.
4.20
4.20
2.016
4.497
3.362
3.576
N.D.
N.D.
4.581
4.02
3.96
3.97
3.99
4.20
4.20
2.084
5.515
4.465
4.18
6.028
—
2.040
4.84
4.130
3.82
3.765
N.D.
4.585
4.03
3.98
4.80
N.D.
N.D.
N.D.
2.034
4.463
3.373
3.586
3.73
3.649
4.553
3.97
3.852
4.25
4.733*
4.22
4.22
2.040
4.711"
4.078
3.73
3.586
N.D.
2.021°"
4.516"
4.473*
3.388*
3.78
3.68
4.557*
4.02"
N.D.
4.74"
N.D.
N.D.
N.D.
2.029°*
4.46"
3.371"
3.576c3
3.75
3.68
4.589"
4.02*
3.96"
N.D.
4.00
4.19
4.19
2.088
5.518
4.464
4.177
6.030
—
Chemical shifts are given in ppm downfield from internal sodium 4,4-dimethyl-4-silapentane-1 -sulfonate, but were actually measured indirectly to acetone in
2
H 2 O (8 2.225) at 26°C. Chemical shifts were determined from the clearly resolved signals in ID 'H-NMR spectra or from cross-peaks in the COSY spectra.
They are summarized with those of reference unsaturated hexasaccharides isolated from shark cartilage CS-D [Sugahara et al., 1996a]:
R3, AHexA(2S)al-3GalNAc<6S)P 1 -4GlcAp 1 -3GalNAc(6S)P 1 -4GlcAP l-3GalNAc(4S);
R4, AHexA(2S)al-3Ga]NAc(6S)pi-4GlcApi-3GalNAc(4S)pi-4GlcApi-3GalNAc(4S). The estimated error for the values is ±0.001 ppm.
•Values determined at 15°C.
"The estimated error for the values to two decimal places is ±0.01 ppm because of a partial overlap of signals.
cl 3
" The two respective values may be interchanged.
ND, Not determined.
—, Not occurring.
5, and of H-2 of GlcA-6 by approximately A0.5 and 0.8 ppm,
as compared with the corresponding nonsulfated structures
(Yamada et al., 1992), indicating the 6-O-sulfation of GalNAc1., 3 and 5, and the 2-0-sulfation of GlcA-6. In the spectrum of
fraction 7-3 (Figure 4B), downfield shifts were found for H-4
of GalNAc-3 (A0.6 ppm), H-6 of GalNAc-i and 5 (A0.4 ppm),
and H-2 of GlcA-2 and 6 (A0.8 ppm), indicating O-sulfation of
the corresponding carbon atoms. Thus, the NMR data were
consistent with the structure proposed above based on the enzymatic analysis. Likewise, the structures of the other hexasaccharide fractions were similarly confirmed (Tables n, Iff).
Discussion
We isolated sulfated hexasaccharides from the repeating disaccharide region of shark cartilage CS-D after digestion with
testicular hyaluronidase. Hence, these structures (designated as
Hyase-hexasaccharides) were composed of the so-called A, C,
and D disaccharide units: A, GlcApl-3GalNAc(4-sulfate); C,
GlcA31-3GalNAc(6-sulfate); D, GlcA(2-sulfate)[313GalNAc(6-sulfate). They included three trisulfated (fractions
1, 2-2, 3-2), four tetrasulfated (fractions 4, 5-2A, 5-2B,and
6-1), and one pentasulfated (fraction 7-3) structure. The struc259
S.Nadanaka and ICSugahara
5.44
5.4
5.2
5.0
4.8
4.6 4.4 4.2
(ppm)
4.0 3.8
3.6
3.4
3.2
Fig. 5. COSY spectra of fractions 4 and 7-3. The 500 MHz COSY spectra of fractions 4 (A) and 7-3 (B) were recorded in D2O at 26°C. The numbers and
letters in the spectra refer to the corresponding residues in the structures. For abbreviations, see the caption to Figure 1.
260
Hexasaccharides from shark cartilage chondroitin sulfate D
tures found in fractions 1, 2-2, 3-2, 4, 5-2A, 6-1 have sulfation
profiles identical to the corresponding unsaturated hexasaccharides (designated as Chase-hexasaccharides) prepared using the
bacterial eliminase chondroitinase ABC (Sugahara et al.,
1996a), and hence are considered to be their saturated counterparts. However, those found in fractions 5-2B and 7-3 or
corresponding unsaturated Chase-hexasaccharides have not
been reported so far and these Hyase-hexasaccharides are unusual in that the D-unit is located at their reducing ends, which
is in contrast to the observation that Chase-oligosaccharides
tend to have a D disaccharide unit at the nonreducing end
(Sugahara et al., 1994a, 1996a). The structural differences
among the oligosaccharide products may reflect the specificities of the catalytic activities of chondroitinase ABC and hyaluronidase.
Both the hexasaccharides in fractions 5-2B and 7-3 contained an unusual A-D tetrasaccharide sequence, GlcApi3GalNAc(4S)pi^GlcA(2S)pi-3GalNAc(6S), at their reducing ends. This structural characteristic substantiates the notion
(Sugahara et al., 1996a) that a D disaccharide unit GlcA(2sulfate)pl-3GalNAc(6-sulfate) is arranged on the immediate
reducing side of an A disaccharide unit GlcA(31-3GaINAc(4sulfate), forming the characteristic A-D tetrasaccharide sequence, GlcAp 1 -3GalNAc(4-sulfate) P1 -4GlcA(2-sulfate) P1 3GalNAc(6-sulfate), in a CS-D polymer. Testicular hyaluronidase catalyzes transglycosilation as well as hydrolysis
(Weissmann, 1955; Highsmith et al., 1975). Although the disaccharide unit at the nonreducing ends of the Hyasehexasaccharides may have been added by transglycosilation, it
is unlikely that the A-D tetrasaccharide sequence on the reducing sides in the above hexasaccharides is an artifact produced
in this manner. CS-Tetra- and CS-hexasaccharides serve as
acceptor substrates for the transglycosilation reactions whereas
CS-disaccharides do not (Weissmann, 1955; Highsmith et al.,
1975). Hence, the A-D tetrasaccharide sequence in the hexasaccharides in fractions 5-2B and 7-3 is most likely a naturally
occurring structure. In fact, the A-D tetrasaccharide sequence
has been found in a saturated tetrasaccharide isolated from a
hyaluronidase digest of shark cartilage CS-D (Sugahara et al.,
1996b) and also in three minor unsaturated hexasaccharides
produced by chondroitinase ABC that does not have transglycosilation activity (Sugahara et al., 1996a). In addition, the
major unsaturated octasaccharides prepared by partial chondroitinase ABC digestion of shark cartilage CS-D contained
the A-D tetrasaccharide sequence (S.Nadanaka and K.Sugahara, manuscript in preparation).
The A-D tetrasaccharide sequence is recognized by the mAb
MO-225 (Yamagata et al., 1987). Its epitope is transiently
expressed and disappears during developmental maturation of
the mouse odontoblast, and it is considered to be biologically
functional (Mark et al., 1990). Hence, the biosynthetic mechanism and the substrate specificities of the sulfotransferases
responsible for synthesis of the A-D sequence or the D structure is of particular interest The A-D sequence may reflect the
substrate specificity of a hitherto unreported chondro-2-Osulfotransferase. In this regard, it is intriguing that no A-C
sequence has been found in the Chase-hexasaccharides (Sugahara et al., 1996a) or in the Hyase-hexasaccharides isolated in
this study. Even if it is transiently formed, it may be a suitable
recognition site for the chondro-2-O-sulfotransferase, and
might have been converted to the A-D sequence accepting a
third sulfate group at the C2 position of the internal GlcA
residue as discussed previously (Sugahara et al., 1996a).
Some other structural features are also noteworthy. Three
consecutive C disaccharide units were found in a row in fraction 1 whereas two consecutive D disaccharide units or three
consecutive A disaccharide units have not been detected in
CS-D oligosaccharides. These may also reflect the substrate
specificities of the involved sulfotransferases. Since chondroitin sulfate A preparations from many biological sources often
contain an A-unit as the major disaccharide component, they
are likely to have a long stretch of consecutive A-units. It is
conceivable that the chondro-4-O-sulfotransferase responsible
for the synthesis of this sequence is different from that involved in the synthesis of the CS-D structure. There may be
several chondro-4-O-sulfotransferases. To investigate CS biosynthetic enzymes including such sulfotransferases, structurally defined oligosaccharide substrates are indispensable. In
this regard, the Hyase-hexasaccharides prepared here have
practical importance. They will be utilized as enzyme substrates for biosynthetic enzymes including GalNAc transferases and sulfotransferases and also for catalytic hydrolases
such as glycosidases and sulfatases. In fact, they have been
applied as acceptor substrates for a novel serum a-GalNAc
transferase (Kitagawa era/., 1995b) and for a serum p-GalNAc
transferase that synthesizes the repeating disaccharide region
of CS chains, to characterize their substrate specificities
(Kitagawa et al., 1997).
The present study substantiated the notion about the specificities of chondro-6-sulfatase and chondro-4-sulfatase. It has
been proposed that these sulfatases specifically release a sulfate group only from the reducing GalNAc from unsaturated
tetrasaccharides and hexasaccharides prepared using chondroitinases, although the latter enzyme can remove sulfate
groups also from internal GalNAc residues under harsh incubation conditions (Sugahara et al., 1994a; Sugahara and Kojima, 1996). In this study these enzymes acted on saturated
hexasaccharides, indicating that the unsaturated bond at the
non-reducing terminus is not requisite for the enzymatic actions, and that the enzymes are useful tools for studying sulfated oligosaccharides irrespective of saturation or unsaturation
of their nonreducing termini. It will be interesting to test
whether they are applicable to oligosaccharides with a reducing
GalNAc(4- or 6-sulfate) residue that are derived from glycoproteins and glycolipids.
Materials and methods
Materials
Shark cartilage CS-D (super special gTade), CS-disaccharide standards, chondroitinases ABC (EC 4.2.2.4) and AC-D (EC 4.2.2.5), chondn>4-sulfatase (EC
3.1.6.9), and chondro-6-sulfatase (EC 3.1.6.10) were purchased from Seikagaku Corp., Tokyo. Sheep testicular hyaluronidase (EC 3.2.1.35) was obtained from Sigma. Bio-Gel P-10 resin was obtained from Bio-Rad. Authentic
unsaturated CS-hexasaccharides were purified as described previously (Sugahara et al., 1996a).
Preparation of hexasaccharides
A commercial preparation of shark cartilage CS-D was exhaustively digested
with sheep testicular hyaluronidase in a total volume of 2.0 ml of 50 mM
sodium phosphate buffer (pH 6.0) including 150 mM NaCl at 37°C for 20 h
(Sugahara et al., 1992). The digest was separated into tetra-, hexa-, octa-,
deca-, dodeca, and tetradecasaccharide fractions by gel chromatography on a
column (1.6 x 95 cm) containing Bio-Gel P-10 as described previously (Sugahara et al., 1996a). The hexasaccharide fraction was desalted through Sephadex G-25, and then subfractionated by HPLC using an amine-bound silica
column as described below. Each peak was purified by rechromatography
261
S.Nadanaka and K-Sugahara
under the same conditions as the first step. Peaks were quantified by the
carbazole procedure using GlcA as a standard (Bitter and Muir, 1962).
Enzyme digestion of isolated oligosacchandes
The isolated hexasaccharides were digested with chondroitinase AC-D basically as described previously (Sugahara et al., 1994a). Each isolated oligosaccharide (1 nmol) was incubated with 5 mlU of the enzyme in a total volume
5 ml of 50 mM sodium acetate buffer, pH 6.0, at 37°C for 20 min. Reactions
were terminated by boiling for 1 min. Digestion with chondro-6-sulfatase
(Yamagata el al., 1968) was carried out according to Sugahara and Kojima
(1996) using 50 mlU of the enzyme in 34 mM Tris-HCl, pH 7.5, containing
34 mM sodium acetate and 0.01% (w/v) bovine serum albumin for 3 h. Digestion with chondro-4-sulfatase (Yamagata et al., 1968) was performed under
standard incubation conditions (Sugahara and Kojima, 1996) using 20 mlU of
the enzyme for 10 min in the same buffer as that used for chondro-6-sulfatase.
CE
The purity of each hexasaccharide fraction was judged by CE as described
using a Waters capillary ion analyzer (Sugahara et al., 1994c). Electrophoresis
proceeded for 30 min with 25 mM sodium phosphate buffer (pH 3.0), at a
constant voltage of 15 kV. The separated fractions were detected at 185 run. hi
a disaccharide composition analysis, the isolated hexasaccharides were digested with chondroitinase AC-D, derivatized with AMAC, and then analyzed
by CE as described previously (Kitagawa et al., 1995a).
HPbC
The hexasaccharide fractions were subfractionated and the enzyme digests
were analyzed by HPLC as described (Sugahara et al., 1989; Yoshida et al.,
1989). Chromatography proceeded on a 4.6 x 250 mm amine-bound silica
PA03 column (YMC Co., Kyoto) using a linear gradient from 16 to 800 mM
NaH 2 PO 4 over 90 min or 16-530 mM NaH 2 PO 4 over 60 min at a flow rate of
1.0 ml/min at room temperature. The eluates were monitored by absorbance at
232 run.
500 MHz 'H-NMR spectroscopy
The 500 MHz 'H-NMR spectra of the isolated hexasaccharides were measured
basically according to Vliegenthart et al. (1983) on a Varian VXR-500 at a
probe temperature of 15 or 26°C as reported previously (Sugahara et al.,
1994a). Chemical shifts are given relative to sodium 4,4-dimethyl-4silapentane-1-sulfonate, but were actually measured indirectly relative to acetone (8 2.225) in 2 H 2 O with intermediate lyophilization.
Acknowledgments
We thank Dr. Makiko Sugiura (Kobe Pharmaceutical University) for recording
the NMR spectra and Shoko Shimoji and Kazumi Higashihara for technical
assistance in preparation of the oligosaccharides. This work was supported in
part by the Science Research Promotion Fund from the Japan Private School
Promotion Foundation and Grants-in-Aid for Scientific Research on Priority
Areas (05274107) from the Ministry of Education, Science and Culture of
Japan.
Abbreviations
AMAC, 2-aminoacridone; CE, capillary electrophoresis; COSY, correlation
spectroscopy; HOHAHA, homonuclear Hartmann-Hahn; 1 or 2D, one- OT twodimensional; PAPS, 3'-phosphoadenosine 5'-phosphosulfate; IdoA, L-iduronic
acid; HexA, hexuronic acid; AHexA or A 4J HexA, 4-deoxy-a-/Areo-hex-4enepyranosyluronic acid; ADi-OS, A 4 5 H e x A a l - 3 G a l N A c ; ADi-4S,
A"-5HexAal-3GalNAc(4-sulfate); ADi-6S, A 4J HexAal-3GalNAc(6-sulfate);
ADi-diS D , A 4 3 HexA(2-sulfate)al-3GalNAc(6-sulfate); ADi-diS E ,
A 4 -'HexAal-3GalNAc(4,6-disulfate); ADi-triS, A 4 J HexA(2-sulfate)al3GalNAc(4,6-disulfate); CS, chondoritin sulfate; DS, dermatan sulfate; AU, G,
U, 2S, 4S or 6S denotes A 4J HexA, GalNAc, GlcA, 2-O-, 4-O-, or 6-O-sulfate,
respectively.
References
Bitter,M. and Muir,H. (1962) A modified uronic acid carbazole reaction. Anal.
Biochem., 4, 330-334.
262
Davidson^., GileadJ-, Amirajvi., Ginsburgjt and RazinJE. (1990) Synthesis
of chondroitin sulfate D and heparin proteoglycans in murine lymph nodederived mast cells. The dependence on fibroblasts. J. Biol. Chem., 265,.
12324-12330.
Faissner,A., Clement,A., Lochter^A., StreitA-, Mandl.C. and Schachner,M.
(1994) Isolation of a neural chondroitin sulfate proteoglycan with neurite
outgrowth promoting properties. J. Cell Biol., 126, 783-799.
HarrisJvdJ. and TurveyJ.R. (1970) Sulphates of monosaccharides and derivatives. Carbohydr. Res., 15, 57-63.
Highsmith.S., GarvinJ.H. Jr. and Chipman.D.M. (1975) Mechanism of action
of bovine testicular hyaluronidase. J. Biol. Chem., 250, 7473-7480.
Kitagawa.H., Kinoshita^A. and Sugaharajt. (1995a) Microanalysis of glycosaminoglycan-derived disaccharides labeled with the ftuorophore 2-aminoacridone by capillary electrophoresis and high-performance liquid chromatography. Anal. Biochem., 232, 114-121.
Kitagawa,H., Tanaka.Y., TsuchidaJC, Goto,F., Ogawa.T., Lidholt.K., Lindahl.U. and SugaharaJC- (1995b) A'-Acetylgalactosamine (GalNAc) transfer
to the common carbohydrate-protein linkage region of sulfated glycosaminoglycans. J. Biol. Chem., 270, 22190-22195.
Kitagawajl., TsutsumiJC., Ujikawa.M., Goto.F., TamuraJ., NeumannJC.,
Ogawa.T. and SugaharaJC. (1997) Regulation of chondroitin sulfate biosynthesis by specific sulfation: acceptor specificity of serum P-GalNAc
transferase revealed by structurally defined oligosaccharides. Gtycobiology,
7, in press.
Mark,M.P., Butler.W.T. and RuchJ.V. (1989) Transient expression of a chondroitin sulfate-related epitope during cartilage histomorphogenesis in the
axial skeleton of fetal rats. Dev. Biol., 133, 475-4S8.
MarkJd.P., BakerJ.R., KimataJC. and RuchJ.-V. (1990) Regulated changes in
chondroitin sulfation during embryogenesis: an immunohistochemical approach. Int. J. Dev. Biol., 34, 191-204.
Seno,N., Akiyama,F. and Anno,K. (1974) Substrate specificity of chondrosulfatases from Proteus vulgaris for sulfated tetrasaccharides. Biochim. Biophys. Acta, 362, 290-298.
SorrelU.M., Mahmoodian,F., SchaferJ.A., Davis,B. and Caterson.B. (1990)
Identification of monoclonal antibodies that recognize novel epitopes in
native chondroitin/dermatan sulfate glycosaminoglycan chains: their use in
mapping functionally distinct domains of human skin. J. Histochem. Cytochem., 38, 393-^02.
SorrellJ.M., CarrinoJDA. and Caplan^A.I. (1993) Structural domains in chondroitin sulfale identified by anti-chondroitin sulfate monoclonal antibodies.
Immunosequencing of chondroitin sulfates. Matrix, 13, 351-361.
SugaharaJC. and Kojima.T. (1996) Specificity studies of the bacterial sulfatases using the structurally defined sulfated oligasaccharides isolated from
shark cartilage chondroitin sulfate D. Eur. J. Biochem., 239, 865-870.
SugaharaJC, Okumura.Y. and YamashinaJ. (1989) The Engelbreth-HolmSwarm mouse tumor produces undersulfatcd heparan sulfate and oversulfated galactosaminoglycans. Biochem. Biophys. Res. Common , 162, 189—
197.
SugaharaJC, Yamada,S., Sugiura,M., TakedaJC., Yuen.R., Khoo.H.E. and
Poh,C.H. (1992) Identification of the reaction products of the purified hyaluronidase from stonefish (Synanceja horrida) venom. Biochem. J., 283,
99-104.
SugaharaJC., ShigenoJC, Masuda,M., FujiiJJ., Kurosalca^A. and Takeda.K.
(1994a) Structural studies on the chondroitinase ABC-resistant sulfated tetrasaccharides isolated from various chondroitin sulfate isomers. Carbohydr.
Res., 255, 145-163.
SugaharaJC., Takemura.Y., Sugiura.M., Kohno.Y., Yoshida,K., TakedaJC.,
Khoo,K.H., Morris.H.R. and Dell^A. (1994b) Chondroitinase ABC-resistant
sulfated trisaccharides isolated from digests of chondroitin/dermatan sulfate
chains. Carbohydr. Res., 255, 165-182.
Sugahara^C, Tohno-oka,R., Yamada,S., KhooJCH., MorrisJlJl. and Dell,A.
(1994c) Structural studies on the oligosaccharides isolated from bovine kidney heparan sulphate and characterization of bacterial heparitinases used as
substrates. Glycobiology, 4, 535-544.
SugaharaJC, Nadanaka.S., TakedaJC and KojimaJ. (1996a) Structural analysis of unsaturated hexasaccharides isolated from shark cartilage chondroitin
sulfate D: substrates for exolytic action of chondroitinase ABC. Eur. J.
Biochem., 239, 871-880.
SugaharaJC, Tanaka,Y. and Yamada.S. (1996b) Preparation of a series of
sulfated tetrasaccharides from shark cartilage chondroitin sulfate D using
testicular hyaluronidase and structure determination by 500-MHz 'H NMR
spectroscopy. Glycoconjugate J., 13, 609-619.
Suzuki.S. (1960) Isolation of novel disaccharides from chondroitin sulfates. / .
Biol. Chem., 235, 3580-3588.
VliegenthanJJ.G., Dorland,L. and Van Halbeek,R (1983) High-resolution
'H-nuclear magnetic resonance spectroscopy as a tool in the structural
Hexasaccharides from shark cartilage chondroitin sulfate D
analysis of carbohydrates related to glycoproteins. Adv. Carbohydr Chem.
Biochem., 41, 209-374.
WeissmaruvB- (1995) The transglycosylative action of testicular hyaluronidase. J. Biol. Chan.. 216, 783-794.
Yamada,S., YoshidaJC., Sugiurajvi. and Sugahara.K. (1992) One- and two
dimensional 'H-NMR characterization of two series of sulfated disaccharides prepared from chondroitin sulfate and heparan sulfate/heparin by bacterial eliminase digestion. / . Biochem., 112, 440-447.
YamagataX, Saito.H., Habuchi.O. and Suzuki.S. (1968) Purification and
properties of bacterial chondroitinase and chondrosulfatases. / . Biol. Chem ,
243, 1523-1535.
Yamagatajvt, Kimata.K., OiJce.Y., Tani,K., MaedaJ*)., Yoshida,K., Shimomura,Y., Yoneda^l. and Suzuld.S. (1987) A monoclonal antibody that specifically recognizes a glucuronic acid 2-sulfate-containing determinant in intact
chondroitin sulfate chain. / . Biol. Chem., 262, 4146-4152.
YoshidaJC., Miyauchi.S., Kikuchi.H., TawaoXA. and Tokuyasu.K. (1989)
Analysis of unsaturated disaccharides from glycosaminoglycuronan by
high-performance liquid chromatography. Anal. Biochem., 177, 327-332.
Received on July 4, 1996; revised on September 17, 1996; accepted on September 17, 1996
263