Glycobiology Advance Access originally published online on December 29, 2004
Glycobiology 2005 15(6):593-603; doi:10.1093/glycob/cwi036
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Glycobiology vol. 15 no. 6 © Oxford University Press 2004; all rights reserved.
Structural characterization of the epitopes of the monoclonal antibodies 473HD, CS-56, and MO-225 specific for chondroitin sulfate D-type using the oligosaccharide library
2 Department of Biochemistry, Kobe Pharmaceutical University, 4-19-1 Motoyamakita-machi, Higashinada-ku, Kobe 658-8558, Japan; 3 Department of Cell Morphology and Molecular Neurobiology, Ruhr-University, 44801 Bochum, Germany; 4 Department of Biotechnology, Faculty of Engineering, Kyoto Sangyo University, Kita-ku, Kyoto 603-8555, Japan; 5 CREST, JST, 4-1-8 Honcho Kawaguchi, Saitama, Japan
1 To whom correspondence should be addressed; e-mail: k-sugar{at}kobepharma-u.ac.jp
Received on August 21, 2004; revised on December 20, 2004; accepted on December 22, 2004
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
The variation in the sulfation profile of chondroitin sulfate (CS)/dermatan sulfate (DS) chains regulates central nervous system development in vertebrates. Notably, the disulfated disaccharide D-unit, GlcUA(2-O-sulfate)-GalNAc(6-O-sulfate), correlates with the promotion of neurite outgrowth through the DSD-1 epitope that is embedded in the CS moiety of the proteoglycan DSD-1-PG/phosphacan. Monoclonal antibody (mAb) 473HD inhibits the DSD-1-dependent neuritogenesis and also recognizes shark cartilage CS-D, which is characterized by the prominent D-unit and is also recognized by two other mAbs, CS-56 and MO-225. We investigate the oligosaccharide epitope structures of these CS-D-reactive mAbs by ELISA and oligosaccharide microarrays using lipid-derivatized CS oligosaccharides. CS-56 and MO-225 recognized the octa- and larger oligosaccharides, though the latter also bound one unique hexasaccharide D-A-D, where A denotes the disaccharide A-unit GlcUA-GalNAc(4-O-sulfate). The octasaccharides reactive with CS-56 and MO-225 shared a core A-D tetrasaccharide, whereas the neighboring structural elements located on the reducing and/or nonreducing sides of the A-D gave a differential preference additionally to the recognition sequence for each antibody. In contrast, 473HD reacted with multiple hexa- and larger oligosaccharides, which also contained A-D or D-A tetrasaccharide sequences. Consistent with the distinct specificity of 473HD as compared with CS-56 and MO-225, the 473HD epitope displayed a different expression pattern in peripheral mouse organs as revealed by immunohistology, extending the previously reported CNS-restricted expression. The epitope of 473HD, but not of CS-56 or MO-225, was eliminated from DSD-1-PG by digestion with chondroitinase B, suggesting the close association of L-iduronic acid with the 473HD epitope. Despite such supplemental information, the integral epitope remains to be isolated for identification and comprehensive analytical characterisation. Thus novel information on the sugar sequences containing the A-D tetrasaccharide core was obtained for the epitopes of these three useful mAbs.
Key words: chondroitin sulfate / dermatan sulfate / epitope / monoclonal antibody / oligosaccharide library
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Chondroitin sulfate (CS) and dermatan sulfate (DS) are widely distributed in extracellular matrices and at cell surfaces as proteoglycans (PGs), in which glycosaminoglycan (GAG) chains are covalently attached to a variety of core proteins. Growing evidence suggests the biological significance of CS and DS in neural development as neuritogenic molecules (for reviews, see Oohira et al., 2000
; Sugahara and Yamada, 2000; Sugahara et al., 2003) and also as major inhibitors of axonal regeneration in the injured central nervous system (Bradbury et al., 2002; Moon et al., 2001; Morgenstern et al., 2002).
The CS chain backbone consists of repetitive disaccharide units containing D-glucuronic acid (GlcUA) and N-acetyl-D-galactosamine (GalNAc) residues, while DS is a stereoisomeric variant of CS with varying proportions of L-iduronic acid (IdoUA) in place of GlcUA. In mammalian tissues, these chains are often found as CS/DS hybrid structures. Further structural variability of CS/DS chains is produced by divergent sulfation in the repeating disaccharide units. The monosulfated disaccharide A-unit [GlcUA-GalNAc(4S)] and C-unit [GlcUA-GalNAc(6S)] are common and major components of mammalian CS chains, whereas small yet significant proportions of the disulfated disaccharide D-unit [GlcUA(2S)-GalNAc(6S)] and E-unit [GlcUA-GalNAc(4S,6S)] are detected in the mammalian brain and other tissues (Bao et al., 2004; Saigo and Egami, 1970; Ueoka et al., 2000; Zou et al., 2003) and some isolated neuronal CS-PG molecules (Clement et al., 1998; Maeda et al., 2003; Tsuchida et al., 2001) (2S, 4S, and 6S represent 2-O-, 4-O-, and 6-O-sulfate group, respectively). Proportions of these units in the chick, mouse, and pig brain change with development (Bao et al., 2004; Kitagawa et al., 1997; Maeda et al., 2003), suggesting that CS/DS chains differing in the degree and profile of sulfation may be involved in the functional diversity during the brain’s development.
DSD-1-PG/phosphacan purified from neonatal mouse brains exhibits neurite outgrowth-promoting activity toward cultured rodent hippocampal neurons (Faissner et al., 1994; Hikino et al., 2003). The neuritogenic activity is strongly inhibited by the monoclonal antibody (mAb) 473HD, which recognizes a unique CS/DS hybrid structure (referred to as the DSD-1 epitope) present in the GAG moiety of DSD-1-PG (Faissner et al., 1994; Hikino et al., 2003). Interestingly, the 473HD-reactive epitope is also embedded in shark cartilage CS preparations, CS-C and CS-D, which contain significant proportions (9.6% and 21.2%, respectively) of the D-unit, and the CS-D preparation itself also possesses neurite outgrowth-promoting activity (Clement et al., 1998, 1999; Hikino et al., 2003; Nadanaka et al., 1998), suggesting a correlation between the neuritogenic DSD-1 epitope and the CS disaccharide D motif.
In addition, several lines of evidence suggest important roles for particular CS structures containing D- and/or E-units in neuroregulatory signal transduction (for reviews, see Maeda, 2003; Muramatsu, 2001; Sugahara et al., 2003). Among the reported brain CS-PGs (Bandtlow and Zimmermann, 2000), a cell membrane–tethered receptor-type tyrosine phosphatase, PTP
/RPTPß, which is a splicing variant of DSD-1-PG/phosphacan, binds heparin-binding growth factors, such as pleiotrophin (PTN) and midkine (MK) through its CS moiety as in the case of phosphacan (Maeda et al., 1996, 1999; Milev et al., 1998). Low-density lipoprotein receptor-related protein (Muramatsu et al., 2000; Sakaguchi et al., 2003) LRP6 (Sakaguchi et al., 2003) and PTP (Qi et al., 2001; Sakaguchi et al., 2003) have been identified as components of the MK receptor(s) (Muramatsu et al., 2000), and anaplastic lymphoma kinase (Stoica et al., 2002) appears to be another receptor. The binding to PTP and biological activities of PTN and MK are strongly inhibited by CS-D and CS-E from squid cartilage, which is characterized by a predominance of the E-unit (Maeda et al., 1996; Maeda and Noda, 1998; Tanaka et al., 2003). Recently, Maeda et al. (2003) reported that the binding affinity of phosphacan for PTN depends on the structure of its CS chains, especially on the presence of the D-unit. Thus, the neuroregulatory CS/DS structures containing the D-unit are of special interest.
To date, various CS-specific mAbs have been raised and widely used for immunohistochemical studies of various tissues. CS-56 (Avnur and Geiger, 1984) and MO-225 (Yamagata et al., 1987) also preferentially recognize CS-D from shark cartilage among typical CS variants, showing a specificity for CS in vitro similar to that of 473HD. However, the structural features of these CS epitopes are only poorly understood. In this study, we characterize the epitope structures recognized by these three anti-CS antibodies using an enzyme-linked immunosorbent assay (ELISA) and an oligosaccharide microarray system with an oligosaccharide library derived from CS-D to clarify the structural differences among the 473HD, CS-56, and MO-225 epitopes.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Minimum size requirement of the immunogenic CS epitopes involving the D-unit
The mAb 473HD recognizes the characteristic CS structure named the DSD-1 epitope that contains the D-unit (Clement et al., 1998
, 1999; Nadanaka et al., 1998). The commercially available mAb CS-56, which has been reported to react with CS-A and CS-C (Avnur and Geiger, 1984), also reacts preferentially with CS-D polysaccharides from shark cartilage, weakly with CS-C, and not at all with other CS variants A, B, or E, whereas the mAb MO-225 (Yamagata et al., 1987) reacts strongly with CS-D, moderately with CS-C and CS-E, and weakly with CS-A (manufacturer’s technical information, www.seikagaku-hit.com/bio/02tech/a02_kou/01pr/p02--2.htm).
At first, to determine the minimum sizes of the immunogenic CS structures required for recognition by these three antibodies, even-numbered CS oligosaccharide fractions, which were prepared by partial digestion of CS-D from shark cartilage with sheep testicular hyaluronidase (Sugahara et al., 1996b), were utilized. These oligosaccharide fractions were chemically coupled to the amino group of dipalmitoylphosphatidylethanolamine (DPPE) using their reducing terminal aldehyde groups by reductive amination (Stoll et al., 1988), without disturbing the sulfation pattern–dependent antigenic structures embedded in the CS oligosaccharides. The resultant neoglycolipid probes were immobilized on microtiter wells, which were then used in the ELISA system. Alternatively, the neoglycolipids were immobilized onto nitrocellulose membranes to prepare oligosaccharide microarrays (Fukui et al., 2002).
473HD reacted markedly with the hexa- and larger oligosaccharide fractions derived from CS-D with apparently comparable efficacy (Figure 1A). The degree of binding of 473HD to hexa- and octasaccharides was increased in a dose-dependent fashion, reaching a plateau at 200 and 100 pmol of the oligosaccharides, respectively (Figure 2A). Although the tetrasaccharide fraction seemed to react slightly with the mAb 473HD in ELISA compared with the unsaturated disaccharide control (Figure 1A), its reactivity was negligible in another evaluation system using the oligosaccharide microarray, where 25 pmol of each starting oligosaccharide was immobilized (data not shown), suggesting that 473HD recognizes CS-D-derived hexa- and larger oligosaccharides.
|
1-3GalNAc(6-O-sulfate)] in the presence (+) or absence (–) of each primary antibody.
|
CS-56 reacted preferentially with the octa- and larger oligosaccharide fractions from CS-D (Figure 1B). The degree of binding of CS-56 to the CS-D octasaccharide fraction was proportional to the amount (50–1000 pmol) of oligosaccharides, whereas the binding to the hexasaccharide fraction was weak (Figure 2B). Strong MO-225-reactivity was also observed toward octa- and larger oligosaccharide fractions, being proportional to the amount of lipid-derivatized octasaccharides (Figures 1C and 2C). The hexasaccharide fraction also appeared to react moderately with MO-225 (Figure 1C) and showed weak dose-dependent reactivity with MO-225 compared with the octasaccharide fraction (Figure 2C). Similar trends for CS-56 and MO-225 were observed on the microarrays, where tetra-, hexa-, and octasaccharide fractions were spotted (data not shown). These results suggest that the minimum size for recognition by MO-225 is hexa- or octasaccharides, whereas that for CS-56 is octasaccharides, which is consistent with the previous findings that the minimum recognition size for CS-56 is unsaturated decasaccharides prepared by digestion of CS-A and CS-C by chondroitinase ABC (Fukui et al., 2002).
Differential reactivities of mAbs 473HD, CS-56, and MO-225 toward structurally defined CS oligosaccharides
To further characterize these CS-D-reactive mAbs, their immunoreactivities toward an oligosaccharide library, composed of a series of structurally defined tetra-, hexa-, and octasaccharides prepared from CS-D of shark cartilage (Nadanaka and Sugahara, 1997; Nadanaka et al., 1998; Sugahara et al., 1994, 1996b), were assessed in an oligosaccharide microarray system, due to the limited availability of these oligosaccharides. The structurally defined oligosaccharides forming the library used in this study are summarized in Table I. It has been demonstrated that specific binding of anti-CS antibodies and other carbohydrate-binding proteins to CS oligosaccharide chains is detectable on the microarray, even under conditions where only a few picomoles of CS-lipid derivatives are used for immobilization (Fukui et al., 2002). Therefore, the neoglycolipids of the structurally defined oligosaccharides derived from CS-D were spotted onto a nitrocellulose membrane in a low picomole range (1–10 pmol per spot as a starting oligosaccharide before the lipid derivatization).
|
These lipid derivatives were immobilized on the membrane with comparable efficacy, as demonstrated by the similar color intensity given by each neoglycolipid spot detected with so-called stub antibodies (data not shown), which recognize the enzymatically generated, unsaturated disaccharide unit 
A [4,5HexUA-GalNAc(4S)] or C [4,5HexUA-GalNAc(6S)] at the nonreducing ends of CS oligosaccharides, where 4,5HexUA represents unsaturated hexuronic acid. None of the three CS-D-specific mAbs reacted with any lipid derivatives of the tetrasaccharides listed in Table I (data not shown), but they showed differential preferences toward the structurally defined hexa- and octasaccharides (Figure 3), being consistent with the results shown in Figure 1. All oligosaccharide sequences bound by 473HD, CS-56, and/or MO-225 contained at least the D-unit, whereas the hexa- and octasaccharides devoid of the D-unit (i.e., C-C-C, C-C-A, C-A-A, and C-C-C-C) were not bound by these mAbs, where denotes the unsaturation of the nonreducing terminal HexUA in the oligosaccharide) (Figure 3). In contrast, control experiments using the nonimmune mouse IgM gave no positive staining, supporting the size-dependent reactivities of these mAbs (data not shown).
|
473HD reacted preferentially with two structurally defined hexasaccharides D-A-D and D-A-A, which contain the D-A tetrasaccharide sequence. A hexasaccharide fraction, which contains two separate sequences D-C-A and C-A-D in a molar ratio of 6:4 (Nadanaka and Sugahara, 1997), also reacted modestly with 473HD. In spite of the presence of the D-unit, D-C-C was not recognized (Figure 3A). In addition, the structurally defined octasaccharides in the spotted fractions 10–13, which contain an A-D tetrasaccharide sequence, showed positive staining by 473HD with comparable efficacy (Figure 3D). The observed binding preference suggests the importance of the A-D and/or D-A tetrasaccharide sequences for the recognition by 473HD. However, because neither A-D nor D-A tetrasaccharide (Table I) reacted with 473HD (data not shown), a minimum hexasaccharide size in addition to A-D and/or D-A sequences appears to be required for the effective binding of 473HD.
Both CS-56 and MO-225 reacted with the octasaccharides that contained a D-unit but not with the hexasaccharides, except for D-A-D (for MO-225) (Figures 3B, C, E, and F), which is basically in agreement with the size-dependent reactivities observed in Figures 1B and C. These positive octasacchatides shared the A-D tetrasaccharide sequence in common, suggesting its importance as an antigenic core domain for recognition by CS-56 and MO-225. However, relative reactivities of CS-56 and MO-225 toward the four positive octasaccharides were clearly different (Figures 3E and F). The relative staining intensity obtained with CS-56 was in the following order: C-A-D-C > A-A-D-C > A-A-D-A > C-A-D-A, suggesting that CS-56 prefers the hexasaccharide sequence A-D-C to A-D-A in the octasaccharide sequences. Thus, the disaccharide unit on the reducing side of the A-D sequence influences the recognition by CS-56. On the other hand, MO-225 reacted strongly with C-A-D-C and C-A-D-A, and weakly with A-A-D-C and A-A-D-A, implying that the sulfated position in the disaccharide unit located on the nonreducing side of the A-D sequence (i.e., C-A-D> A-A-D) contributes appreciably to the different degrees of recognition by MO-225. The reactivity of MO-225 toward the D-A-D hexasaccharide may be accounted for by a conformation common to GlcUA(2S)-GalNAc(6S)-A-D and 4,5HexUA-GalNAc(6S)-A-D. Thus, in addition to the A-D tetrasaccharide sequence, the neighboring sequences located on the reducing and/or nonreducing sides of the A-D sequence and their arrangements appear to affect the efficient and differential binding of CS-56 and MO-225.
Immunohistochemical localization of the epitope of 473HD in the adult mouse organs
The findings indicate that the three CS-D-reactive mAbs recognize overlapping yet distinct structures in CS chains. It is remarkable that the binding specificity of 473HD revealed using oligosaccharides embraces a larger array of structures than that of CS-56 and MO-225, because this contrasts at first sight the restricted distribution of its epitope. In fact, the DSD-1-epitope displayed a preponderance in brain and no detectability in various other adult mouse organ tissue extracts in a previous report (Clement et al., 1998), clearly different from the wide distribution of the epitopes of CS-56 (Yamamoto et al., 1995) and MO-225 (Maeda et al., 2003; Mark et al., 1990; Uchimura et al., 2002). To reconcile the seemingly contradictory findings, an immunohistochemical analysis of 473HD reactivity in adult mouse tissues from several organs was performed. On cryosections of kidney, colon, lung, testis, and adrenal gland a very discrete, faint expression pattern was observed, which was confined to various organ-specific basal laminae (Figures 4A, B, E, F, and G). In addition, the epitope was demonstrable in the cortex of the adrenal gland but not in association with basal lamina structures, which differed from the pattern recorded in other organs (Figure 4G). In contrast, no immunoreactivity for 473HD was detected in liver or heart muscle (Figures 4C and D). Thus, the immunohistological analysis revealed the selective and specific expression of the CS structure recognized by 473HD, which appeared much weaker and more localized in all peripheral organs when compared with the molecular and granular layers of the cerebellum, where it was abundant and strongly expressed (Figure 4H).
|
The expression in the peripheral organs was unexpected in the light of a previous study (Clement et al., 1998), where 473HD was found strongly reactive with cerebellar, weakly with cerebral, and not with various peripheral organ extracts by western blot, suggesting a central nervous system–specific expression and association of the epitope with DSD-1-PG/phosphacan. Consistent with this report, the strongest signals were again obtained with the cerebellum (Figure 4H). The association with the basal lamina and the extracellular matrix might have withstood to some extent the octylglucoside-based extraction buffer used, which, in conjunction with the lower overall expression levels of the epitope, was presumably at the base of the negative results collected in the previous study (Clement et al., 1998). The observed expression of the DSD-1 epitope in subgroups of epithelial cells is consistent with increasing evidence that phosphacan and/or PTP/RPTPß are also expressed outside the nervous system, for example in the epithelia of the gastric mucosa (Fujikawa et al., 2003). Whether PTP/RPTPß isoforms represent the corresponding core proteins in this case and in the peripheral organs is, however, unknown and remains to be clarified.
Possible involvement of the CS/DS hybrid structure containing IdoUA in 473HD-reactivity
Recent studies have demonstrated that CS/DS hybrid chains with a significant proportion of IdoUA play important roles in promoting the outgrowth of neurites and in the binding of several growth factors (Bao et al., 2004; Hikino et al., 2003). The DSD-1 epitope, which is recognized specifically by 473HD, in the CS chains of mouse DSD-1-PG/phosphacan has also been postulated to be constituted of a CS/DS hybrid structure in addition to the D-unit (Clement et al., 1998; Faissner et al., 1994; Hikino et al., 2003). Thus the native CS structure of the 473HD epitope may involve an additional structural element, such as a CS/DS hybrid structure containing IdoUA in addition to the A-D and/or D-A tetrasaccharide sequences. Therefore, to investigate whether an IdoUA-containing CS/DS hybrid structure is involved in the reactivity toward 473HD, the immunoreactivity of 473HD toward DSD-1-PG preparations pretreated with various CS lyases, including chondroitinases ABC, AC-I, AC-II and B, was examined and compared with the results obtained in parallel experiments performed with CS-56 and MO-225.
Chondroitinase ABC cleaves nearly all the galactosaminidic linkages in CS/DS chains, whereas chondroitinases AC-I and AC-II specifically cleave the galactosaminidic linkages bound to GlcUA in an endolytic and exolytic fashion, respectively (Yoshida et al., 1993). Since chondroitinase AC-II tends not to act beyond the IdoUA-containing building blocks in CS/DS chains (Yanagishita and Hascall, 1979), the selective resistance to chondroitinase AC-II can be used as an important criterion to identify the IdoUA-containing structures (Yoshida et al., 1993). In contrast, chondroitinase B cleaves the GalNAc-IdoUA linkage in an endolytic fashion but not the GalNAc-GlcUA linkage in DS or CS/DS chains (Sugahara et al., 1995). Pretreatment of the DSD-1-PG preparation with these chondroitinases even under harsh conditions (see Materials and methods) resulted in no significant changes in the reactivity toward the mAb that recognizes the core protein of DSD-1-PG, excluding the possibility of the undesired detachment of the CS chains of DSD-1-PG from the ELISA plates.
The DSD-1-PG preparation treated with chondroitinase ABC or AC-I did not react with 473HD, CS-56, or MO-225 (Figure 5) as expected, most likely due to the complete removal of the antigenic CS/DS moiety from DSD-1-PG. The treatment of DSD-1-PG with chondroitinase AC-II also completely eliminated the reactivity toward CS-56 and MO-225, whereas 
40% of the reactivity with 473HD was retained even after the enzyme treatment (Figure 5), consistent with a previous finding (Faissner et al., 1994) and suggesting that the enzyme did not remove completely the CS/DS chains, presumably due to its exolytic action (Yanagishita and Hascall, 1979; Yoshida et al., 1993), which can be blocked by an IdoUA residue. These results suggest that an IdoUA-containing CS/DS structure is associated in some way with the 473HD epitope but not with the CS-56 or MO-225 epitope. It should be noted, however, that chondroitinase AC-II could eliminate the reactivity with 473HD if harsh incubation conditions were used most likely due to the endolytic activity of the enzyme (Hiyama and Okada, 1976).
|
The treatment with chondroitinase B removed partially the epitopes recognized by these three mAbs under the standard incubation conditions for the enzyme (see Materials and methods), with 40–60% of the reactivity toward these mAbs retained (Figure 5), most likely because of the significant yet incomplete elimination of the CS/DS chains by the endolytic action of the enzyme. The enzyme’s actions on the immobilized CS/DS chains may be partially interfered with by the possible steric hindrance of the immobilized CS/DS chains. In support of this view, these immunoreactivities were completely abolished by digestion with chondroitinase B under the harsh incubation conditions (see Materials and methods) using 10 times the enzyme protein (data not shown). Taken together, the differential sensitivity of the 473HD epitope to the treatments by various chondroitinases suggests that the epitope in the CS chains of DSD-1-PG may include an IdoUA-containing CS/DS hybrid structure in addition to the A-D or D-A sequence. However, the possibility cannot be excluded that an IdoUA residue(s) may be outside the 473HD epitope being located on the reducing side of the epitope, and hence the epitope is eliminated by chondroitinase B.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
In this study, the structural characteristics of the CS epitopes recognized by the mAb 473HD and widely used commercial mAbs CS-56 and MO-225 were examined in detail using the CS oligosaccharide library. All these antibodies react with CS-D from shark cartilage and show immunohistochemically overlapping but differential tissue staining patterns.
Accumulating evidence suggests that the proportion of the D-unit and the presence of IdoUA in CS/DS chains are associated with neuritogenesis (Sugahara and Yamada, 2000; Sugahara et al., 2003). In addition, primary hippocampal neurons cultured on a substratum coated with CS-D rich in the D-unit or the oversulfated DS preparations containing the characteristic iD disaccharide unit [IdoUA(2S)-GalNAc(6S)], sprout multiple neurites of dendritic nature (Clement et al., 1998, 1999; Faissner et al., 1994; Hikino et al., 2003; Nadanaka et al., 1998), where i stands for IdoUA. An exogenous application of CS-D gives rise to an aberrant morphogenesis of Purkinje cell dendrites, possibly mediated by PTN-PTP signaling (Tanaka et al., 2003). In the developing cerebellum, the CS-56 and MO-225 epitopes containing the D-unit are abundantly distributed in the molecular layer, where multiple Purkinje cell dendrites expand extensively (Maeda et al., 2003). Therefore, the CS-56 and MO-225 epitopes may also be partly involved in the promotion of neurite outgrowth.
The application of the lipid derivatives of these oligosaccharides to solid-phase binding assay systems such as ELISA and the recently developed oligosaccharide microarray has provided efficient and sensitive analytical systems, giving novel structural information regarding the minimum sizes and sequences of the CS epitopes of mAbs 473HD, CS-56, and MO-225. All structurally defined octasaccharides that reacted with 473HD, CS-56, and MO-225 contained the A-D tetrasaccharide core sequence (Figures 3B, C, E, and F). These octasaccharides are major components accounting for as much as 58% (w/w) of the octasaccharide fractions isolated by partial digestion of CS-D from shark cartilage with chondroitinase ABC (Nadanaka et al., 1998), and the A-D sequence has also been found in one tetrasaccharide and a few hexasaccharides isolated from CS-D (Nadanaka and Sugahara, 1997; Sugahara et al., 1996a,b). Therefore, the preferential reactivity of these mAbs with CS-D among typical CS variants appears to be attributable to the abundant A-D tetrasaccharide sequence characteristic of CS-D.
Consistent with the notion that the DSD-1 epitope is structurally associated with the D-unit (Clement et al., 1998, 1999; Nadanaka et al., 1998), 473HD also reacted with several oligosaccharides containing the D-unit. The reactivity of 473HD was detected in some but not all hexa- and larger oligosaccharides (Figure 1A). This seemed unexpected at first sight in light of the previously reported restricted expression of the DSD-1-epitope in the central nervous system (Clement et al., 1998) in contrast to the more widely distributed epitopes of CS-56 and MO-225, which require octa- and larger oligosaccharides (Figures 1 and 2). Although common features of the hexa- and octasaccharides immunoreactive to 473HD were the tetrasaccharide A-D and/or D-A sequences, no additional specific structural element was detected in the sequences neighboring the tetrasaccharide core of the octasaccharide sequences examined (Figures 3A and D). These seemingly contradictory results can now be resolved by the findings that the 473HD epitope is expressed at low levels in discrete patterns in several peripheral mouse organs in addition to the brain.
Immunohistochemical analysis revealed the weak expression of the 473HD epitope mainly in the basal laminae of several peripheral mouse organs in addition to the strong expression in cerebellum, while the CS-56 epitope is also widely expressed (Yamamoto et al., 1995). However, comparison of the immunhistological data generated with these mAbs is difficult for several reasons. First, the relative location of the epitopes with regard to the long CS chains and the core protein are unknown, yet both factors may influence the accessibility in vivo. Second, CS-56 was generated against human materials (Avnur and Geiger, 1981) and 473HD against the mouse antigen (Faissner et al., 1994), which may cause subtle differences in affinity. Third, the data sets available in the literature have been collected on different species, which renders a comparison of histological patterns difficult. Taking these restrictions into account, it appears nevertheless that important differences can be identified on the level of distribution of the 473HD and CS-56 epitopes. Thus the CS-56 antibody clearly recognizes structures in the chicken heart, in correlation with trabecular and atrial septal formation, and in blood vessel walls (Capehart et al., 1999; Daugaard et al., 1991), although 473HD did not react with the heart in our study. A systematic comparison of histological results collected with both antibodies and identification of the core proteins of the different epitopes are, however, beyond the scope of the present study.
The neuritogenic activity of DSD-1-PG is inhibited by exogenously applied 473HD and by digestion with chondroitinase B, which cleaves the GalNAc-IdoUA linkage in the DS-type structure (Faissner et al., 1994; Hikino et al., 2003), suggesting that the neuritogenic motif of the DSD-1-PG epitope overlaps the 473HD epitope and may contain IdoUA. Analysis of the chondoitinase B–cleavable sites in the CS chains of DSD-1-PG suggests the presence of GlcUA-GalNAc(4S)-IdoUA/IdoUA(2S) and GlcUA-GalNAc(6S)-IdoUA/IdoUA(2S) trisaccharide sequences (Hikino et al., 2003). In addition, the importance of the iA-unit [IdoUA-GalNAc(4S)] in the neuritogenic activity of CS/DS hybrid chains of the embryonic pig brain was recently revealed (Bao et al., 2004). Furthermore, oversulfated DS preparations isolated from marine animals such as ascidians, which contain the characteristic iD disaccharide unit [IdoUA(2S)-GalNAc(6S)], sprout multiple dendritic neurites in cultured primary hippocampal neurons (Hikino et al., 2003). In view of these findings, the 473HD-reactive DSD-1 epitope may contain at least either one of the above-mentioned trisaccharides in addition to A-D and/or D-A sequences in longer sequences. Alternatively, a combination of the A-D and/or D-A sequences with either the iA [IdoUA-GalNAc(4S)] (Bao et al., 2004) or iD [IdoUA(2S)-GalNAc(6S)] unit may be present in the DSD-1 epitope, although the existence of the iD-unit has not been reported in the brain CS/DS. However, as discussed, it remains to be further investigated whether an IdoUA-containing structure is indeed a part of the 473HD epitope because the possibility exists that an IdoUA is not included in the epitope but located on its reducing side in the parent chain. Identification of the bona fide 473HD epitope awaits the isolation of 473HD-reactive oligosaccharides from mammalian brains.
The minimum octasaccharide size required for the effective binding of both CS-56 and MO-225 implies the importance of the neighboring structure surrounding the A-D sequence. In this regard, the relative intensity of the reactivity of CS-56 with the structurally defined octasaccharides (Figure 3B) suggests that the disaccharide unit A or C residing on the reducing side of the A-D sequence in the hexasaccharide sequences, A-D-A and A-D-C, respectively, confers the respective sequence specificity on CS-56. Their evaluated contributable effects were in the order of A-D-C > A-D-A (Figure 3E). In contrast, in the case of MO-225, the disaccharide unit on the nonreducing side of the A-D sequence has significant effects on the binding, with the C-A-D sequence more effective than the A-A-D sequence in the octasaccharide sequences X-A-D-Y, where X and Y represent an A-, C-, or D-unit. It remains to be determined whether MO-225 can bind the hexasaccharides themselves to investigate the possibility that the minimum recognition size is a hexasaccharide in view of its binding to the D-A-D sequence (Figure 3C). More A-D-containing hexasaccharides should also be isolated and tested. It should also be remembered that the E-D tetrasaccharide, in addition to the A-D tetrasaccharide, shows significant inhibitory effects on the binding of MO-225 to PG-H/versican as has been demonstrated by competitive ELISA (Yamagata et al., 1987). Although the involvement of the E-unit in the specific binding of MO-225 remains unclear in terms of the sugar sequence requirement, MO-225 reacts with several oligosaccharide fractions prepared from squid cartilage CS-E (Deepa et al., unpublished data) with a high proportion of the E-unit (56% of the total disaccharide units) (Kinoshita et al., 1997), supporting the notion of the involvement of the E-unit in addition to the D-unit. Some CS-E oligosaccharides may share conformations similar to those of CS-D oligosaccharides.
Although the binding specificity of the three mAbs toward the structurally defined CS oligosaccharides appears similar at the primary sugar sequence level, the effects of the treatments of the DSD-1-PG with various chondroitinases on the binding of these mAbs were clearly different, supporting the virtual structural divergence of these CS epitopes. Maeda et al. (2003) have reported the systematic immunohistochemical localization of several CS epitopes, including CS-56 and MO-225 epitopes in the developing postnatal mouse brain. The CS-56 epitope is highly expressed in the cerebral cortex early in the postnatal period, and its expression is markedly decreased at around 3 weeks after birth (Maeda et al., 2003). In contrast, the expression of the MO-225 epitope is weak in the cerebral cortex but rather strong in the other regions, including the cerebellum, during the developing stages. MO-225 reactivity in the cerebellum reaches a culmination at postnatal day 14 and decreases thereafter (Maeda et al., 2003). Interestingly, the 473HD epitope is abundantly distributed in the cerebellum but weakly in the residual regions of the adult mouse brain (Clement et al., 1998). These spatiotemporal expression patterns of the CS epitopes support their differential reactivity to the CS-D-derived oligosaccharides. In addition, analysis of the three dimensional structure of these oligosaccharides will help clarify the divergent specificity of these mAbs.
In this study, we extended the oligosaccharide microarray method (Fukui et al., 2002) to successfully show its applicability and usefulness to the study of mAb epitopes using an oligosaccharide library of structurally defined CS oligosaccharides. Now that GAG oligosaccharide libraries (Sugahara and Yamada, 2000; Yamada and Sugahara, 1998) are available for such sensitive microarrays, in which even a few picomoles of a lipid-derivatized CS oligosaccharide is detectable, a further screening using various CS oligosaccharide libraries, particularly those prepared from mammalian tissue sources, may provide bona fide bioactive oligosaccharide structures recognized by anti-CS antibodies. Furthermore, the oligosaccharide libraries are of course applicable to high-throughput detection and specificity assignment for investigating the functional domain sequences of CS, DS, and heparan sulfate chains for the binding of various functional proteins, including heparin-binding growth factors, CS/DS-binding growth factors, and cytokines.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Materials
The following sugar, enzymes, and antibodies were purchased from Seikagaku (Tokyo): shark cartilage CS-D (sodium salt, super special grade), conventional chondroitinase ABC (EC 4.2.2.4 [EC] ) from Proteus vulgaris, chondroitinase AC-I (EC 4.2.2.5 [EC] ) from Fravobacterium heparinum, chondroitinase AC-II (EC 4.2.2.5 [EC] ) from Arthrobacter aurescens, chondroitinase B from F. heparinum, and mAbs CS-56 (Avnur and Geiger, 1984
) and MO-225 (Yamagata et al., 1987), which more selectively recognize shark cartilage CS-D polysaccharide than the other CS variants A, B, C, D, and E (see the Results).
The mAb 473HD, which recognizes the brain-specific DSD-1 epitope of CS chains, was prepared as described previously (Faissner et al., 1994). DSD-1-PG was purified from postnatal days 1–15 mouse brains as described previously (Faissner et al., 1994). Even-numbered, saturated oligosaccharide (tetra-, hexa-, octa-, deca-, and dodecasaccharide) fractions were prepared by a partial enzymatic digestion of commercial CS-D from shark cartilage with sheep testicular hyaluronidase as described previously (Nadanaka and Sugahara, 1997). Structurally defined tetra-, hexa-, and octasaccharides listed in Table I were isolated from CS-D of shark cartilage as described previously (Nadanaka et al., 1998; Nadanaka and Sugahara, 1997; Sugahara et al., 1994, 1996b).
Lipid-derivatization of CS-derived oligosaccharides
Oligosaccharides derived from CS-D were coupled to DPPE (Wako Pure Chemical Industries, Osaka, Japan) as described previously (Stoll et al., 1988) with a slight modification. Briefly, a dried oligosaccharide fraction (50–1000 pmol as oligosaccharide) was dissolved in 2.5 µl water and mixed with 10 nmol DPPE in 47.5 µl chloroform/methanol (1:1, v/v). The reaction mixture was sonicated in a sonic bath, heated at 60°C for 10 min, mixed with 5 µl of a fresh preparation of 160 mM sodium cyanoborohydride in methanol, and incubated at 60°C for 16 h. The reaction mixture was directly used for immobilization without purification for ELISA. For the preparation of oligosaccharide microarrays, each resultant lipid-derivatized reaction mixture was dried in a vacuum concentrator and dissolved in chloroform/methanol/water (25:25:8, v/v/v) (Fukui et al., 2002).
ELISA with lipid-derivatized CS oligosaccharides
A whole mixture of each lipid-derivatized solution was added to each well of a 96-well microtiter plate (Nunc immuno Plate, PolySorp, Nalge Nunc, Tokyo). After evaporation of the solvent using a hair dryer, the lipid-derivative-coated wells were overlaid briefly with 0.08% polyisobutylmethacrylate in chloroform/hexane/methanol (21:4:100, v/v/v) to stabilize the immobilized lipid-derivatives (Taki et al., 1990). Each well was blocked with 3% (w/v) bovine serum albumin (BSA) in phosphate-buffered saline (PBS), pH 7.4, at room temperature for 4 h and incubated with a CS-specific mAb, 473HD, CS-56, or MO-225, which were diluted by 50-, 200-, and 500-fold, respectively, with PBS, pH 7.4, containing 1% (w/v) BSA at room temperature for 2 h. After being washed with Tris-buffered saline (TBS), pH 7.5, three times, each well was incubated at room temperature for 2 h with alkaline phosphatase–conjugated secondary antibodies, anti-mouse IgM + IgG (for CS-56 and MO-225) or anti-rat IgM (for 473HD), which had been diluted 5000-fold with TBS, pH 7.5, and washed as described. Finally, 0.05% (w/v) p-nitrophenyl phosphate in 50 mM carbonate buffer, pH 9.8, was added, and the yellow color produced was measured in a spectrophotometer at 415 nm.
CS oligosaccharide microarrays
The preparation of CS oligosaccharide microarrays was carried out as described previously (Fukui et al., 2002) with slight modifications. Briefly, the lipid-derivatized CS oligosaccharides in chloroform/methanol/water (25:25:8, v/v/v) were applied by jet spray using a sample applicator (Linomat V, Camag, Switzerland) at a spotting rate of 70 nl/sec, as 2-mm-width bands onto a 0.45-µm nitrocellulose membrane (Bio-Rad, Tokyo). Each membrane was blocked with 3% (w/v) BSA in PBS, pH 7.4, for 1 h and overlaid with appropriately diluted CS-specific mAbs (473HD, CS-56, or MO-225) or with control mouse IgM in 3% (w/v) BSA/PBS, pH 7.4, for 2 h. The membranes were rinsed several times with 3% BSA in PBS, pH 7.4, and overlaid for 1 h with horseradish peroxidase–conjugated anti-rat IgM (for 473HD) or anti-mouse IgM + IgG (for CS-56, MO-225, or control mouse IgM) diluted 500- or 3000-fold, respectively, with 3% (w/v) BSA in PBS, pH 7.4. Antibody binding was visualized using 3,3'-diaminobenzidine as a chromogen.
Immunohistochemistry
Adult NMRI mice (postnatal day 50) were anesthetized and perfused with 4% paraformaldehyde. The organs were removed, postfixed for 2 h with 4% paraformaldehyde in PBS at 4°C, and then cryoprotected in 30% sucrose in PBS overnight. The tissue was embedded in Tissue Freezing Medium (Leica Instruments GmbH, Heidelberg, Germany) and sectioned at 14 mm on a Leica cryostat. For immunohistochemical analysis, the tissue sections were rehydrated for 1 h at room temperature in a blocking solution composed of 1.7% w/v NaCl and 10% v/v normal goat serum in PBS, pH 7.4, and subsequently washed three times for 5 min in PBS. The sections were incubated with the mAb 473HD (rat IgM; diluted 500-fold) overnight at 4°C. After washing three times for 5 min in PBS, the sections were incubated with anti-rat IgM CY3 (diluted 500-fold with 0.1% BSA/PBS) for 2 h at room temperature. During incubation with secondary antibodies, the chromatin was labeled with bisbenzimide at a dilution of 1:105 (Sigma, Munich, Germany). After three further washes in PBS, the slices were mounted in Immumount. Expression of the 473HD epitope was analyzed using an epifluorescence microscope (Zeiss, Axiophot) with a 20x or 40x objective lens. Immunofluorescence images were captured with a digital camera (Axiocam) using the Axiovision 4.1 program (Zeiss).
Analysis of the effects of digestions with various chondroitinases on the immunogenicity of DSD-1-PG toward 473HD, CS-56, and MO-225
The DSD-1-PG preparation equivalent to 250 ng GlcUA was absorbed onto a 96-well microtiter plate (Nunc immuno Plate, MaxiSorp Nalge Nunc International) at 4°C for 10 h. After three washes with PBS containing 0.05% (w/v) Tween 20 (PBS-T), each well was blocked with 1% (w/v) BSA in PBS, pH 7.4, for at least 1 h. For enzymatic digestion of the CS moiety of DSD-1-PG with various chondroitinases, the microtiter wells were incubated with 0.1 mIU of chodroitinase ABC or AC-I, or 0.1 mIU (standard conditions) or 1.0 mIU (harsh conditions) of chondroitinase AC-II or B in a total volume of 50 µl of the appropriate buffer (Pojasek et al., 2001; Sugahara et al., 1995) including a protease inhibitor cocktail (catalog number P-8340, Sigma) at 37°C for 2 h as described previously. Thereafter, the plates were washed three times with PBS-T and processed for ELISA with the mAb 473HD, CS-56, or MO-225 as described. The wells were still reactive with the mouse anti-phosphacan mAb (Chemicon, Temecula, CA), diluted 500-fold in TBS, pH 7.5, after treatment with either one of the chondroitinases, excluding the possibility of an unexpected detachment of the CS chains of DSD-1-PG from microtiter wells due to the proteolytic degradation of the core protein by proteases, which may be present as contaminants in the chondroitinase preparations. For the mouse anti-phosphacan, alkaline phosphatase–conjugated anti-mouse IgM + IgG was used as a secondary antibody.
|
Acknowledgements |
|---|
We thank Atsuko Nakashima for preparation of even-numbered oligosaccharides from CS-D, and Christine Heinisch, Melanie Lenz and Anke Baar for the technical support. The work performed in Kobe was supported in part by the Science Research Promotion Fund from the Japan Private School Promotion Foundation, Grant-in-Aid for Exploratory Research 15659021 from MEXT, HAITEKU (2004–2008), and CREST, JST (to K. S.). Support by the German Research Council (DFG, SPP 1172) to A. F. is also greatly acknowledged.
|
Abbreviations |
|---|
BSA, bovine serum albumin; CS, chondroitin sulfate; DPPE, dipalmitoylphosphatidylethanolamine; DS, dermatan sulfate; ELISA, enzyme-linked immunosorbent assay; GAG, glycosaminoglycan; mAb, monoclonal antibody; MK, midkine; PBS, phosphate-buffered saline; PG, proteoglycan; PTN, pleiotrophin; PTP, receptor-type protein tyrosine phosphatase; TBS, Tris-buffered saline
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Avnur, Z. and Geiger, B. (1984) Immunocytochemical localization of native chondroitin-sulfate in tissues and cultured cells using specific monoclonal antibody. Cell, 38, 811–822.[CrossRef][ISI][Medline]
Bandtlow, C.E. and Zimmermann, D.R. (2000) Proteoglycans in the developing brain: new conceptual insights for old proteins. Physiol. Rev., 80, 1267–1290.
Bao, X., Nishimura, S., Mikami, T., Yamada, S., Itoh, N., and Sugahara, K. (2004) Chondroitin sulfate/dermatan sulfate hybrid chains from embryonic pig brain, which contain a higher proportion of L-iduronic acid than those from adult pig brain, exhibit neuritogenic and growth factor binding activities. J. Biol. Chem., 279, 9765–9776.
Bradbury, E.J., Moon, L.D., Popat, R.J., King, V.R., Bennett, G.S., Patel, P.N., Fawcett, J.W., and McMahon, S.B. (2002) Chondroitinase ABC promotes functional recovery after spinal cord injury. Nature, 416, 636–640.[CrossRef][Medline]
Capehart, A.A., Mjaatvedt, C.H., Hoffman, S., and Krug, E.L. (1999) Dynamic expression of a native chondroitin sulfate epitope reveals microheterogeneity of extracellular matrix organization in the embryonic chick heart. Anat. Rec., 254, 181–195.[CrossRef][Medline]
Clement, A.M., Nadanaka, S., Masayama, K., Mandl, C., Sugahara, K., and Faissner, A. (1998) The DSD-1 carbohydrate epitope depends on sulfation, correlates with chondroitin sulfate D motifs, and is sufficient to promote neurite outgrowth. J. Biol. Chem., 273, 28444–28453.
Clement, A.M., Sugahara, K., and Faissner, A. (1999) Chondroitin sulfate E promotes neurite outgrowth of rat embryonic day 18 hippocampal neurons. Neurosci. Lett., 269, 125–128.[CrossRef][ISI][Medline]
Daugaard, S., Strange, L., and Schiodt, T. (1991) Immunohistochemical staining for chondroitin sulphate and keratan sulphate. An evaluation of two monoclonal antibodies. Histochemistry, 95, 585–589.[CrossRef][ISI][Medline]
Faissner, A., Clement, A., Lochter, A., Streit, A., Mandl, C., and Schachner, M. (1994) Isolation of a neural chondroitin sulfate proteoglycan with neurite outgrowth promoting properties. J. Cell Biol., 126, 783–799.
Fujikawa, A., Shirasaka, D., Yamamoto, S., Ota, H., Yahiro, K., Fukada, M., Shintani, T., Wada, A., Aoyama, N., Hirayama, T., Fukamachi, H., and Noda, M. (2003) Mice deficient in protein tyrosin phosphatase receptor type are resistant to gastric ulcer induction by VacA of Helicobacter pylori. Nat. Genetics, 33, 375–381.
Fukui, S., Feizi, T., Galustian, C., Lawson, A.M., and Chai, W. (2002) Oligosaccharide microarrays for high-throughput detection and specificity assignments of carbohydrate-protein interactions. Nat. Biotechnol., 20, 1011–1017.[CrossRef][ISI][Medline]
Hikino, M., Mikami, T., Faissner, A., Vilela-Silva, A.C., Pavão, M.S., and Sugahara, K. (2003) Oversulfated dermatan sulfate exhibits neurite outgrowth-promoting activity toward embryonic mouse hippocampal neurons: implications of dermatan sulfate in neuritogenesis in the brain. J. Biol. Chem., 278, 43744–43754.
Hiyama, K. and Okada, S. (1976) Action of chondroitinases. I. The mode of action of two chondroitinase-AC preparations of different origin. J. Biochem. (Tokyo), 80, 1201–1207.
Kinoshita, A., Yamada, S., Haslam, S.M., Morris, H.R. Dell, A., and Sugahara, K. (1997) Isolation and structural determination of novel sulfated hexasaccharides from squid cartilage chondroitin sulfate E that exhibits neuroregulatory activities. J. Biol. Chem., 272, 19656–19665.
Kitagawa, H., Tsutsumi, K., Tone, Y., and Sugahara, K. (1997) Developmental regulation of the sulfation profile of chondroitin sulfate chains in the chicken embryo brain. J. Biol. Chem., 272, 31377–3138.
Maeda, N. (2003) Signal transduction of proteoglycan-type protein tyrosine phosphatase, PTP. Connect. Tissue, 35, 19–23.
Maeda, N. and Noda, M. (1998) Involvement of receptor-like protein tyrosine phosphatase /RPTPß and its ligand pleiotrophin/heparin-binding growth-associated molecule (HB-GAM) in neuronal migration. J. Cell Biol., 142, 203–216.
Maeda, N., Nishiwaki, T., Shintani, T., Hamanaka, H., and Noda, M. (1996) 6B4 proteoglycan/phosphacan, an extracellular variant of receptor-like protein-tyrosine phosphatase /RPTPß, binds pleiotrophin/heparin-binding growth-associated molecule (HB-GAM). J. Biol. Chem., 271, 21446–21452.
Maeda, N., Ichihara-Tanaka, K., Kimura, T., Kadomatsu, K., Muramatsu, T., and Noda, M. (1999) A receptor-like protein-tyrosine phosphatase PTP/RPTPß binds a heparin-binding growth factor midkine. Involvement of arginine 78 of midkine in the high affinity binding to PTP. J. Biol. Chem., 274, 12474–12479.
Maeda, N., He, J., Yajima, Y., Mikami, T., Sugahara, K., and Yabe, T. (2003) Heterogeneity of the chondroitin sulfate portion of phosphacan/6B4 proteoglycan regulates its binding affinity for pleiotrophin/heparin binding growth-associated molecule. J. Biol. Chem., 287, 35805–35811.
Mark, M.P., Baker, J.R., Kimata, K., and Ruch, J.V. (1990) Regulated changes in chondroitin sulfation during embryogenesis: an immunohistochemical approach. Int. J. Dev. Biol., 34, 191–204.[Medline]
Milev, P., Chiba, A., Häring, M., Rauvala, H., Schachner, M., Ranscht, B., Margolis, R.K., and Margolis, R.U. (1998) High affinity binding and overlapping localization of neurocan and phosphacan/protein-tyrosine phosphatase-/ß with tenascin-R, amphoterin, and the heparin-binding growth-associated molecule. J. Biol. Chem., 273, 6998–7005.
Moon, L.D., Asher, R.A., Rhodes, K.E., and Fawcett, J.W. (2001) Regeneration of CNS axons back to their target following treatment of adult rat brain with chondroitinase ABC. Nat. Neurosci., 4, 465–466.[ISI][Medline]
Morgenstern, D.A., Asher, R.A., and Fawcett, J.W. (2002) Chondroitin sulphate proteoglycans in the CNS injury response. Prog. Brain Res., 137, 313–332.[ISI][Medline]
Muramatsu, H., Zou, K., Sakaguchi, N., Ikematsu, S., Sakuma, S., and Muramatsu, T. (2000) LDL receptor-related protein as a component of the midkine receptor. Biochem. Biophys. Res. Commun.,270, 936–941.[CrossRef][ISI][Medline]
Muramatsu, T. (2001) Roles of proteoglycans in reception of the midkine signal. Trends Glycosci. Glycotechnol., 13, 563–572.
Nadanaka, S. and Sugahara, K. (1997) The unusual tetrasaccharide sequence GlcA ß1-3GalNAc(4-sulfate)ß1-4GlcA(2-sulfate)ß1-3GalNAc(6-sulfate) found in the hexasaccharides prepared by testicular hyaluronidase digestion of shark cartilage chondroitin sulfate D. Glycobiology, 7, 253–263.
Nadanaka, S., Clement, A.M., Masayama, K., Faissner, A., and Sugahara, K. (1998) Characteristic hexasaccharide sequences in octasaccharides derived from shark cartilage chondroitin sulfate D with a neurite outgrowth promoting activity. J. Biol. Chem., 273, 3296–3307.
Oohira, A., Matsui, F., Tokita, Y., Yamauchi, S., and Aono, S. (2000) Molecular interactions of neural chondroitin sulfate proteoglycans in the brain development. Arch. Biochem. Biophys., 374, 24–34.[CrossRef][ISI][Medline]
Pojasek, K., Shriver, Z., Kiley, P., Venkataraman, G.,, and Sasisekharan, R. (2001) Recombinant expression, purification, and kinetic characterization of chondroitinase AC and chondroitinase B from Flavobacterium heparinum. Biochem. Biophys. Res. Commun., 286, 343–351.[CrossRef][ISI][Medline]
Qi, M., Ikematsu, S., Maeda, N., Ichihara-Tanaka, K., Sakuma, S., Noda, M., Muramatsu, T., and Kadomatsu, K. (2001) Haptotactic migration induced by midkine. Involvement of protein-tyrosine phosphatase . Mitogen-activated protein kinase, and phosphatidylinositol 3-kinase. J. Biol. Che