JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M310877200 on December 29, 2003

J. Biol. Chem., Vol. 279, Issue 11, 9765-9776, March 12, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/11/9765    most recent
M310877200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bao, X.
Right arrow Articles by Sugahara, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bao, X.
Right arrow Articles by Sugahara, K.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

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*

Xingfeng Bao{ddagger}§, Shuji Nishimura{ddagger}, Tadahisa Mikami{ddagger}, Shuhei Yamada{ddagger}, Nobuyuki Itoh¶, and Kazuyuki Sugahara{ddagger}||

From the {ddagger}Department of Biochemistry, Kobe Pharmaceutical University, Higashinada-ku, Kobe 658-8558 and the Department of Genetic Biochemistry, Kyoto University of Graduate School of Pharmaceutical Sciences, Kyoto 606-8501, Japan

Received for publication, October 2, 2003 , and in revised form, December 24, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
We have shown that over-sulfated chondroitin sulfate/dermatan sulfate (CS/DS) chains from various marine organisms exhibit growth factor binding activities and neurite outgrowth-promoting activities in embryonic mouse hippocampal neurons in vitro. In this study we demonstrated that CS/DS hybrid chains purified from embryonic pig brain displayed marked neuritogenic activity and growth factor binding activities toward fibroblast growth factor 2 (FGF2), FGF10, FGF18, pleiotrophin, and midkine, all of which exhibit neuroregulatory activities in the brain. In contrast, the CS/DS preparation from adult pig brain showed considerably less activity to bind these growth factors and no neuritogenic activity. Structural analysis indicated that the average size of the CS/DS chains was similar (40 kDa) between these two preparations, but the disaccharide compositions differed considerably, with a significant proportion of L-iduronic acid (IdoUA)-containing disaccharides (8~9%) in the CS/DS chains from embryos but not in those from adults (<1%). Interestingly, both neurite outgrowth-promoting activity and growth factor binding activities of the CS/DS chains from embryos were abolished by digestion not only with chondroitinase ABC but also with chondroitinase B, suggesting that the IdoUA-containing motifs are essential for these activities. These findings imply that the temporal expression of CS/DS hybrid structures containing both GlcUA and IdoUA and binding activities toward various growth factors play important roles in neurogenesis in the early stages of the development of the brain.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Chondroitin sulfate proteoglycans (CS-PGs)1 are complex macromolecules consisting of a protein core and at least one covalently linked CS glycosaminoglycan (GAG) chain and are ubiquitous components in the extracellular matrices of connective tissues and at the surfaces of many cell types (for reviews, see Refs. 1–3). In the mammalian brain CS-PGs are common extracellular matrix components with a highly regulated spatiotemporal expression (4–8). Although the CS chains attached to CS-PGs have attracted little attention until recently compared with heparan sulfate (9), recent advances in the structural biology of CS chains have suggested important biological functions in the development of the brain (10). Several studies have demonstrated that the composition of CS chains changes with aging and normal brain maturation and that some CS epitopes are only found in specific sections of the avian and mammalian brain (7, 8, 11). The developmentally regulated expression and tissue-specific distribution of CS variants suggest that CS chains differing in the degree and profile of sulfation exhibit distinct functions during the development of the brain.

The functions of CS-PGs and CS chains in the central nervous system can be categorized as the regulation of cell adhesion and migration, neurite formation, polarization of neurons, synaptic plasticity, survival of neurons, etc. (for reviews, see Ref. 1 and 4). Concerning the neurite formation, studies have shown that CS-PGs and CS chains exhibit primarily inhibitory effects on neurite outgrowth on defined growth-promoting substrata (4) and axon guidance in vivo (12, 13). The inhibitory function of CS chains was supported by recent studies demonstrating that degradation of CS chains by in vivo injection of chondroitinase ABC permitted axonal regeneration after spinal cord injury (14) and reactivation of ocular dominance plasticity in the adult visual cortex (15).

By contrast, Faissner et al. (16) report that DSD-1-PG, a CS-PG derived from neonatal mouse brain, displayed significant neurite outgrowth-promoting activity in vitro toward hippocampal neurons from embryonic day 18 (E18) rats, and this activity was neutralized by the monoclonal antibody 473 HD, which recognizes a structure embedded in the CS chains (referred to as the DSD-1 epitope). Clement et al. (7) show that the CS side chains of DSD-1-PG contain a small but significant proportion of an over-sulfated D disaccharide unit (GlcUA[2-O-sulfate]{beta}1–3GalNAc[6-O-sulfate]). Over-sulfated CS-D from shark cartilage contains a high proportion (~20%) of the D unit and exhibits neurite outgrowth-promoting activity, whereas CS-A and CS-C do not (17). Clement et al. (18) further demonstrate that another over-sulfated CS variant CS-E from squid cartilage, which contains a high proportion (more than 60%) of the E unit, GlcUA{beta}1-3GalNAc(4,6-O-disulfate), promotes neurite outgrowth in a DSD-1 epitope-independent fashion, giving a long prominent neurite with an axon-like morphology.

Recently, Hikino et al. (19) have shown that the over-sulfated dermatan sulfate (DS) chains purified from various marine organisms also exhibit such activities in a sulfation pattern-dependent manner and that the alleged CS side chains of DSD-1-PG are actually of the DS-type, which implies the possible involvement of DS chains in the neuritogenesis during brain development. Lafont et al. (20) previously reported that soluble DS chains prepared from bovine mucosa increased dendrite growth in mesencephalic neurons. These findings suggest that not only over-sulfated CS but also over-sulfated DS chains are neuritogenic. Thus, CS/DS chains cannot simply be classified as either supportive or inhibitory of neurite outgrowth; the sulfation pattern and specific sugar sequence may define these properties (3, 10, 21). Compared with CS-PGs, DS-PGs are minor components of the adult mammalian brain, and so far only a few DS-PG species, such as decorin and biglycan, have been partially characterized (22). Thus, DS-type GAG chains appear to exist in the brain, but their distribution, developmental changes, and functions remain unclear.

The CS/DS variants, from marine organisms used in our previous studies, include CS-D, CS-E, and hagfish notochord-derived CS-H (the major H disaccharide unit is IdoUA{alpha}1–3GalNAc[4,6-O-disulfate]), all of which contain unusually high proportions of over-sulfated disaccharide units. However, they are only minor components in the brain (7, 8, 11, 23, 24). For example, CS-PGs from E18 rat brain contained 1.7% D unit and 1.2% E unit (23). To evaluate the contribution of the CS/DS chains to the neurite extension, it is essential to characterize CS/DS chains purified from mammalian brains. The developmentally regulated disaccharide composition of brain CS/DS chains suggests that at different developmental stages the chains may have distinct biological functions in neurite formation and growth factor signaling (8, 11, 25). In this study, we purified CS/DS chains from embryonic and adult pig brains and investigated their effects on neurite outgrowth and growth factor binding. We also characterized the structures responsible for these activities. Preliminary results were reported in abstract form (26).


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Animals—Embryonic pigs (body weight: 315 g) and adult pig brains were purchased from a local pig-raising company. ddY mice were used for the preparation of E16 hippocampal neurons.

Materials—The following enzymes were purchased from Seikagaku Corp. (Tokyo, Japan): chondroitinase ABC (EC 4.2.2.4 [EC] ), chondro-4-sulfatase (EC 3.1.6.9 [EC] ) and chondro-6-sulfatase (EC 3.1.6.10 [EC] ) from Proteus vulgaris; chondroitinase AC-I (EC 4.2.2.5 [EC] ), chondroitinase B (EC 4.2.2), heparitinase (EC 4.2.2.8 [EC] ), and heparinase (EC 4.2.2.7 [EC] ) from Flavobacterium heparinum; hyaluronidase (EC 4.2.2.1 [EC] ) from Streptomyces hyalurolyticus. Recombinant human (rh)-midkine (MK) expressed in Escherichia coli and rh-fibroblast growth factor 1 (FGF1) (acidic FGF) expressed in E. coli were purchased from PeproTech EC LTD (London, UK). rh-pleiotrophin (PTN) expressed in E. coli was from RELIA Tech GmbH (Braunschweig, Germany), and rh-heparin binding epidermal growth factor-like growth factor (HB-EGF) expressed in Sf 21 insect cells was from R&D systems (Minneapolis). rh-FGF2 (basic FGF) expressed in E. coli was from Genzyme Techne (Minneapolis, MN), and rh-FGF10 expressed in E. coli was provided by Takashi Katsumata (Sumitomo Pharmaceutical Research Center, Osaka, Japan). Recombinant mouse FGF18 was prepared as reported (27). Actinase E was obtained from Kaken Pharmaceutical Co. (Tokyo, Japan). Sep-Pak Plus AccellTM anion-exchange cartridges were obtained from Waters Corp. (Milford, MA), and a Superdex 75 HR column (10 x 300 mm), a Superdex Peptide HR column (10 x 300 mm), a Superdex 200 HR column (10 x 300 mm), and prepacked disposable PD-10 columns were from Amersham Biosciences. Size-defined dextrans (average Mr 65,500, 37,500, and 18,000), bovine intestinal mucosa HS (average Mr 7,500), and low molecular weight heparin from porcine intestinal mucosa (average Mr 6,000) were purchased from Sigma. All other chemicals and reagents were of the highest quality available.

Preparation and Purification of CS/DS Chain Fractions—Embryonic (9.7 g) and adult (108.0 g) pig brains were homogenized with 2.5 volumes of ice-cold phosphate-buffered saline (PBS) containing 20 mM EDTA (11). The homogenates were centrifuged at 15,000 x g for 25 min at 4 °C, and to the resultant supernatant 4 volumes of ethanol was added, which precipitated PBS-soluble CS/DS-PGs. The PBS-insoluble fraction was further homogenized with 1.5 volumes of acetone, and the acetone-insoluble materials were collected by centrifugation. After drying, both PBS-soluble and -insoluble CS/DS-PG-containing fractions from embryo and adult pig brains were extensively digested with actinase E at 60 °Cin0.1 M borate-sodium buffer, pH 8.0, containing 10 mM CaCl2. After incubation and subsequent treatment with 5% trichloroacetic acid, the GAG-containing fractions were recovered by ethanol precipitation. Each GAG fraction was desalted by gel filtration on a PD-10 column and then fractionated by anion-exchange chromatography on an Accell QMA Plus cartridge using stepwise elution with 0.3 M phosphate buffers, pH 6.0, containing 0.15, 1.0, and 1.5 M NaCl. All subfractions were desalted by gel filtration on a PD-10 column. The total amount of GAGs in each subfraction was evaluated by the carbazole reaction (28), and the CS/DS disaccharide composition was analyzed by chondroitinase ABC digestion followed by HPLC (29). The 1.0 M NaCl-eluted fractions from both PBS-soluble and -insoluble fractions, which contained more than 90% of the recovered CS/DS chains of the embryo or adult pig brain-derived sample, was used for further purification.

To remove hyaluronic acid (HA) PBS-soluble and -insoluble fractions obtained from both embryonic and adult pig brains were subjected to digestion with Streptomyces hyaluronidase at 60 °C in 20 mM acetate buffer, pH 5.0 (30). After 4 h of incubation, the digests were subjected to gel filtration chromatography on a column of Superdex 75 (10 x 300 mm) that was eluted with 0.2 M NH4HCO3 at a flow rate of 0.4 ml/min. Absorption at 210 nm was used to monitor GAG chains. The flow-through fractions, which contained GAG chains, were collected, pooled, and evaporated to dryness. The hyaluronidase-resistant GAGs were subsequently digested with a mixture of heparitinase and heparinase in 20 mM sodium acetate buffer, pH 7.0, containing 2 mM calcium acetate for 4 h (31). The polysaccharides resistant to heparinase and heparitinase were recovered by gel filtration as described above. After the removal of HA and heparan sulfate, the CS/DS-containing materials were passed through a C-18 cartridge to remove trace amounts of free peptides. The CS/DS fractions thus prepared from PBS-soluble and -insoluble fractions of embryonic pig brains were designated Es-CS/DS and Ei-CS/DS, whereas those of adult pig brains were As-CS/DS and Ai-CS/DS, respectively. The purity of all these preparations was checked by gel filtration after chondroitinase ABC digestion and by amino acid analysis after acid hydrolysis with 3 M HCl in the gas phase at 100 °C for 16 h.

Determination of Molecular Size—An aliquot (10.0 µg) of Es-, Ei-, As-, or Ai-CS/DS was chromatographed on a gel filtration column of Superdex 200 (10 x 300 mm) that had been calibrated using a series of size-defined commercial polysaccharides (32). The column was eluted with 0.2 M NH4HCO3 at a flow rate of 0.3 ml/min. Fractions were collected at 3-min intervals, lyophilized, and digested with 10 mIU of chondroitinase ABC as described below. The resultant digests were analyzed by anion-exchange HPLC as described previously.

Analysis of Disaccharide Composition—An aliquot (4.0 µg) of Es-, Ei-, As-, or Ai-CS/DS was incubated with 40 mIU of chondroitinase ABC in a 50mM Tris-HCl buffer, pH 8.0, containing 60 mM sodium acetate in a total volume of 40 µlat37 °C for 4 h (33). One-fourth of the digest was subsequently incubated with 40 mIU of chondro-4-sulfatase or chondro-6-sulfatase in a total volume of 50 µl at 37 °C for 1 h (34). Each digest corresponding to one-fourth of the starting materials was analyzed by anion-exchange HPLC on an amino-bound silica PA-03 column (4.6 x 250 mm, YMC Co., Kyoto, Japan) as reported (29). Identification and quantification of the resulting disaccharides were achieved by comparison with CS-derived authentic unsaturated disaccharides: {Delta}HexUA{alpha}1–3GalNAc, {Delta}HexUA{alpha}1–3GalNAc(6-O-sulfate), {Delta}HexUA{alpha}1–3GalNAc(4-O-sulfate) ({Delta}Di-4S), {Delta}HexUA(2-O-sulfate){alpha}1–3GalNAc(6-O-sulfate), {Delta}HexUA{alpha}1–3GalNAc(4,6-O-disulfate), and {Delta}HexUA(2-O-sulfate){alpha}1–3GalNAc(4,6-O-disulfate).

Analysis of the Glucuronate: Iduronate Ratio—An aliquot (2 µg) of Es-, Ei-, As-, or Ai-CS/DS was exhaustively digested with chondroitinases ABC (20 mIU), AC-I (20 mIU), or B (10 mIU) in the appropriate buffer, 50 mM Tris-HCl buffer, pH 7.3 (for AC-I) (19) or 100 mM Tris-HCl buffer, pH 8.0 (for B) (35) (for ABC, see above). Each digestion was initiated with half the amount of the enzyme described above and run at 37 °C for 12 h, after which the rest of the enzyme was added, and incubation was continued for another 2 h. Half of each digest was analyzed by anion-exchange HPLC for identification and quantification of the resultant unsaturated disaccharides as described above.

Analysis of the Iduronate Distribution—An aliquot (0.4 µg) of Es- or Ei-CS/DS was digested with chondroitinase AC-I (6 mIU) and another with chondroitinase B (6 mIU) as described above. The reaction mixtures were lyophilized and derivatized with a fluorophore 2-aminobenzamide (2AB), then the excess 2AB reagent was removed by paper chromatography (36). The 2AB derivatives were subjected to gel filtration on a column (10 x 300 mm) of Superdex Peptide using 0.025 M NH4HCO3 containing 7% 1-propanol as an effluent at a flow rate of 0.4 ml/min. Eluates were monitored by fluorescence with excitation and emission wavelengths of 330 and 420 nm, respectively. Identification and quantification of 2AB-labeled oligosaccharides were achieved by comparison with size-defined 2AB-labeled standard unsaturated CS disaccharides and sulfated oligosaccharides2 prepared by 2AB derivatization of CS-E oligosaccharides generated by testicular hyaluronidase digestion (37). The percentage of galactosaminidic bonds cleaved by each specific enzyme was calculated from the distribution of fluorescence intensity across peaks relative to the total amount of the starting material.

Delayed Extraction Matrix-assisted Laser Desorption Ionization Time-of-flight Mass Spectrometry (MALDI-TOF MS)—The chondroitinase AC-I-resistant 2AB-labeled oligosaccharides, which were obtained as described above, were mixed with a matrix of 2,5-dihydroxybenzoic acid and used for MALDI-TOF MS analyses in a positive ion mode as described previously (38). The MS spectra were recorded on a Voyager DE-RP/Pro (PerSeptive Biosystems, Framingham, MA) in the linear mode.

Preparation of Substrates for Cell Culture—An aliquot (2 µg) of Es-CS/DS or As-CS/DS was digested with 5 mIU of chondroitinases ABC, AC-I, or B in a total volume of 20 µl of the appropriate buffer at 37 °C (or 30 °C for chondroitinase B) for 120 min, and the digests were used to prepare substrates for cell cultures. Aliquots (2 µg) of the parent CS/DS fractions or their chondroitin lyase-treated preparations were individually coated onto plastic coverslips (10 x 10 mm) that had been precoated with poly-DL-ornithine (P-ORN) (Sigma, Tokyo, Japan) dissolved in 0.1 M borate buffer, pH 8.2, at a concentration of 1.5 µg/ml. Heat-inactivated chondroitin lyases with their buffers were used as controls.

Cell Culture and Neurite Outgrowth Promotion Assays—Cultures of mouse hippocampal neurons were established from E16 animals as described (7, 17) with some modifications (19). Briefly, hippocampi were dissected from mouse E16 embryonic brains and dissociated into a single cell suspension. The cells were seeded on coverslips coated with defined GAG substrates at a density of 10,000 cell/cm2 and cultivated in Eagle's modified essential medium containing N2 supplements (Invitrogen), 0.1 mM pyruvate, 0.1% (w/v) ovalbumin, 0.029% (w/v) L-glutamine, 0.2% (w/v) sodium hydrogen carbonate, and 5 mM HEPES. The cultures were incubated in a humidified atmosphere with 5% CO2 at 37 °C. After 24 h of incubation the cells were fixed with 4% (w/v) paraformaldehyde for 30 min at room temperature and then immunostained with anti-microtubule-associated protein 2 (Leico Technologies Inc., St. Louis, MO) and anti-neurofilament (Sigma) at 1:100 and 1:250 dilutions, respectively, in PBS containing 3% (w/v) bovine serum albumin. The antibodies were then detected using a Vectastain ABC kit (Vector Laboratories Inc., Burlingame, CA) with 3,3'-diaminobenzidine as a chromogen. At least three independent experiments in duplicate were carried out for each culture condition.

The stained hippocampal neurons were analyzed with a morphometric station equipped with an optical microscope (BH-2, Olympus, Tokyo, Japan), a digital camera (HC-300Z/OL, Olympus), and software (Mac Scope Mitani Corp., Tokyo, Japan). Clearly isolated cells with at least one neurite longer than the diameter of the cell body were chosen at random. The length of the neurites and the number of primary neurites (longer than the cell body diameter) were determined by drawing and counting, respectively. The data are expressed as the mean ± S.E., and the significance of difference between the means was evaluated using Mann-Whitney's U test.

Growth Factor Binding Assays Using a BIAcore System—Binding reactions were carried out at 25 °C on a BIAcoreTM J Biosensor (BIAcore AB, Uppsala, Sweden) using streptavidin-derivatized sensor chips. Es-CS/DS and As-CS/DS were biotinylated as described (39), and the excess biotinylation reagent was removed using an ultrafree centrifugal filter tube (molecular mass cut-off, 10 kDa; Millipore, Bedford, MA). The biotinylated Es-CS/DS and As-CS/DS chains were then immobilized on the surface of the streptavidin-derivatized sensor chip. The amount of Es-CS/DS or As-CS/DS, immobilized by repeated injections, was controlled at 320 ± 8 resonance units, which corresponds to 0.37 ng of sugar chain. Growth factors (200 ng each) including FGF1, FGF2, FGF10, FGF18, PTN, MK, and HB-EGF in the running buffer, pH 7.4 (HBS-EP), containing 10 mM HEPES, 0.15 M NaCl, 3 mM EDTA, and 0.005% (w/v) Tween 20 were individually injected onto the surface of the Es-CS/DS- or As-CS/DS-immobilized sensor. To correct for the bulk effects and the nonspecific binding of samples, an untreated sensor channel was used as a control, and the response of the control channel was subtracted from the data obtained for the Es-CS/DS- or As-CS/DS-immobilized sensor.

To investigate the structural characteristics of the functional epitopes in Es-CS/DS responsible for its binding to various growth factors, one aliquot (5 µg) of biotinylated Es-CS/DS was digested with 10 mIU of chondroitinase ABC and another with 4 mIU of chondroitinase B as described above, and the resultant digest was immobilized on the sensor chip surface. The amount of immobilized sugar chains was controlled at 0.37 ng by repeated injections. The binding responses were recorded using 200 ng each of the growth factors tested above.

For kinetic analysis various concentrations of each growth factor in the running buffer were injected onto the surface of the Es-CS/DS-immobilized sensor chip. The association reaction continued for 240 s, and then the sensor surface was washed with the running buffer for 300 s as the dissociation phase. Between each injection, the sensor surface was regenerated by injecting 70 µl of 1.0 M NaCl. The kinetic parameters were evaluated with BIAevaluation software 3.1 (BIAcore AB, Uppsala, Sweden) using a 1:1 binding model with mass transfer.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Preparation of the CS/DS Fractions from Embryonic and Adult Pig Brains—In the brain CS-PGs are present predominantly in extracellular matrices and at cell surfaces. Most of CS-PGs in extracellular matrices can be extracted in soluble fractions by physiological buffers without detergent (40), whereas cell surface CS-PGs are retained in the membrane fraction. Hence, CS-PGs in embryonic and adult pig brains were fractionated into PBS-soluble and PBS-insoluble fractions (see "Experimental Procedures"), which generally represent extracellular CS-PGs and cell surface CS-PGs, respectively. The GAG chains in the PBS-soluble and PBS-insoluble fractions, which contained ~60 and 40% of the chondroitinase-sensitive materials in the brain (data not shown), respectively, were released from the protein cores by actinase E digestion, recovered by ethanol precipitation, and then purified by anion-exchange chromatography as described under "Experimental Procedures." In the case of the GAG chains in the PBS-soluble fraction, the 1.0 M NaCl-eluted fraction, which contained more than 90% of the CS/DS chains of the embryonic or adult pig brain-derived sample, was used for further purification. HA was removed from these fractions by digestion with Streptomyces hyaluronidase (30). The digests were subjected to gel filtration on Superdex 75 (Fig. 1, A and B). Approximately 60 and 15% (as uronic acid) of the GAG chains derived from embryonic and adult pig brains were degraded, respectively. The hyaluronidase-resistant fractions were subjected to digestion with a mixture of heparinase and heparitinase to remove heparan sulfate and subjected to gel filtration on Superdex 75 (Fig. 1, C and D). Approximately 5–10% of the GAG chains in all embryo and adult-derived preparations were degraded by a mixture of the enzymes. The pooled flow-through fractions (marked by bars in Fig. 1, C and D) were quantitatively degraded with chondroitinase ABC into unsaturated disaccharides (Fig. 1, E and F), suggesting that they are CS/DS chains. The purified fractions were totally resistant to Streptomyces hyaluronidase and heparitinase/heparinase (data not shown). The CS/DS-containing fractions were passed through a C-18 cartridge to remove trace amounts of peptides, which are not covalently bound to the CS/DS chains. Likewise, the CS/DS chains in the PBS-insoluble fractions were purified using a similar procedure as described under "Experimental Procedures" (data not shown).



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 1.
Purification of CS/DS fractions from embryonic and adult pig brains. The CS/DS-containing GAG fractions of embryonic and adult pig brains were subjected to digestion with hyaluronidase followed by a mixture of heparitinase/heparinase to remove HA and heparan sulfate, respectively, as described under "Experimental Procedures." The digests were chromatographed on a column of Superdex 75. The elution profiles of hyaluronidase digests (A and B) as well as heparitinase/heparinase digests (C and D) of the PBS-soluble GAG fractions from embryonic (A and C) and adult (B and D) pig brains are shown. The flow-through fractions were collected and pooled as indicated by horizontal bars. Aliquots of the CS/DS fractions purified from PBS-soluble fractions of embryonic (E) and adult (F) pig brains were digested with chondroitinase ABC, and the resultant mixtures were chromatographed as described above. The numbers 2, 4, and 6 indicate the elution positions of standard CS di-, tetra-, and hexasaccharides, respectively.

 
Amino acid analysis showed that the CS/DS chains from embryonic and adult pig brains carried small peptides composed of Ser, Asp, Glu, and Gly in an approximate molar ratio of 1:1:1:2 (Es-CS/DS and As-CS/DS) or 0.5~1:1:1~2:1 (Ei-CS/DS and Ai-CS/DS). These compositions are in good agreement with the notion that CS/DS chains are linked to the protein core through a Ser- and Gly-rich region containing some hydrophobic amino acid residues in CS/DS-PGs (41). Taken together, the above results indicated that all these CS/DS preparations contained highly purified CS/DS chains with small linkage region peptides.

Analyses of Disaccharide Compositions—The yields and disaccharide compositions of each CS/DS preparation were determined by HPLC analysis after chondroitinase ABC digestion. The identification of the over-sulfated disaccharides was carried out by digestions with chondro-4 and -6 sulfatases. Contrary to our prediction based on previous studies (see the Introduction), Es- and Ei-CS/DS were relatively low-sulfated due to the presence of a significant proportion of nonsulfated disaccharide units as compared with As- and Ai-CS/DS; the number of sulfate groups per mol of disaccharide was 0.99 (As- and Ai-CS/DS) and 0.83~0.84 (Es- and Ei-CS/DS), respectively (Table I). Both Es- and Ei-CS/DS also showed a higher proportion of 6-O-sulfated disaccharide and a lower proportion of 4-O-sulfated disaccharide than As- and Ai-CS/DS. Although significant yet small proportions of over-sulfated disaccharide D and E units were clearly detected, no striking difference was noted among these four CS/DS fractions (the possible functional importance of these units is discussed below). Es-CS/DS and Ei-CS/DS exhibited similar disaccharide compositions, and As-CS/DS and Ai-CS/DS also showed comparable disaccharide compositions (Table I), suggesting that the CS/DS chains of the PBS-soluble and -insoluble PGs in the brain of the same developmental stage may share similar structural characteristics. In addition, the combined yield of the CS/DS chains of the PBS-soluble and -insoluble PGs from embryonic pig brains was 12% higher than that of the corresponding chains from adult pig brains (Table I), and the amounts of the CS/DS chains of the PBS-soluble PGs accounted for 65 and 55% of all the CS/DS in the embryonic and adult pig brains (Table I), respectively. These results suggest that the expression of CS/DS chains in the developing pig brain changes considerably and that the change in disaccharide composition is consistent with previous results for mouse and embryonic chick brain in that 6-O-sulfation decreases and 4-O-sulfation increases during development (8, 11).


View this table:
[in this window]
[in a new window]
  		TABLE I
Disaccharide compositions of the CS/DS chains purified from embryonic and adult pig brains

The CS/DS chains purified from the embryonic (Es-CS/DS and Ei-CS/DS) and adult (As-CS/DS and Ai-CS/DS) pig brains were extensively digested with chondroitinases ABC (CSase ABC) or AC-I (CSase AC-I), and the unsaturated disaccharides were identified and quantified by anion-exchange HPLC as described under "Experimental Procedures."

 
Determination of Molecular Sizes of the CS/DS Chains—The molecular sizes of the Es-, Ei-, As-, and Ai-CS/DS chains were analyzed by gel filtration chromatography on a column of Superdex 200. Each separated fraction was lyophilized and digested with chondroitinase ABC, and then the resultant digest was analyzed by anion-exchange HPLC. The gel filtration pattern exhibited by unsaturated CS disaccharides in individual fractions was used to determine the size distribution of the parent CS/DS polymers. From the calibration plot generated using the data obtained with polysaccharides of known sizes (Fig. 2, inset), the average molecular size of both Es-CS/DS and As-CS/DS was estimated to be 40 kDa (Fig. 2). Ei-CS/DS and Ai-CS/DS also gave a similar average molecular size showing an indistinguishable gel filtration pattern (data not shown).



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 2.
Determination of the molecular sizes of the CS/DS chains from embryonic and adult pig brains. Ten micrograms of Es-, Ei-, As-, or Ai-CS/DS was fractionated on a column (10 x 300 mm) of Superdex 200 using 25 mM NH4HCO3, pH 7.5, as an effluent with a flow rate of 0.3 ml/min. Fractions were collected at 3-min intervals, and each fraction was subjected to digestion with chondroitinase ABC followed by disaccharide analysis to monitor the elution profile of the CS/DS chains as described under "Experimental Procedures." Vo and Vt were determined using Dextran 2000 (170~200 kDa) and NaCl, respectively. The triangles and squares indicate the elution profiles of the resulting disaccharides generated from Es-CS/DS and As-CS/DS, respectively. Ei-CS/DS and Ai-CS/DS gave a similar average molecular size showing an indistinguishable gel filtration pattern (data not shown). The inset shows the calibration curve, which was generated by commercial polysaccharides of known molecular sizes to verify a linear relation between the log Mr and the elution volumes (32).

 
Es-CS/DS and Ei-CS/DS Promote Neurite Outgrowth in Embryonic Mouse Hippocampal Neurons—The CS/DS chains purified from embryonic and adult pig brains were assessed for activity to promote neurite outgrowth in E16 mouse hippocampal neurons in vitro using an established assay system (7, 17). Neurons were cultured at low density on plastic coverslips coated with P-ORN and then CS/DS chains. A control coverslip was coated with P-ORN alone. After 24 h of culture neurons were immunostained with anti-microtubule-associated protein 2 and anti-neurofilament antibodies. To evaluate the neurite outgrowth-promoting activity of the CS/DS preparations the length of clearly detectable neurites and the number of primary neurites in 100 randomly selected cells were measured.

As shown in Fig. 3 and Fig. 4, both Es-CS/DS and Ei-CS/DS exhibited potent promoting effects on neurite outgrowth in hippocampal neurons at a dose of 2 µg, whereas As-CS/DS and Ai-CS/DS showed no significant activity at the same dose. Moreover, the promoting effects of Es-CS/DS and Ei-CS/DS (data not shown) appeared to increase in a dose-dependent manner in the range of 2~8 µg (Fig. 4). In contrast, As-CS/DS and Ai-CS/DS showed no significant activity even at a dose of 8 µg (data not shown). Because Es-CS/DS and Ei-CS/DS exhibited similar effects on neurite outgrowth with a very similar neuronal morphology, only Es-CS/DS was used for inhibition assays and enzyme digestion experiments. Recently, it was reported that the in vivo injection of chondroitinase ABC permitted axonal regeneration after injury to the central nervous system in adult mice (14, 42), suggesting that the CS chains in the adult central nervous system have inhibitory effects on axonal growth. Hence, to investigate whether the CS/DS chains from adult pig brains have inhibitory or neutralizing effects on the neuritogenic activity of embryonic pig brain-derived CS/DS chains, equal amounts (2 µg) of As-CS/DS and Es-CS/DS were mixed and used to coat a coverslip for cell culture analysis. The results showed that As-CS/DS did not significantly influence the promoting activity of Es-CS/DS (Fig. 4F), suggesting that the CS/DS chains from adult pig brains did not contain an inhibitory or neutralizing component affecting the neuritogenic activity of Es-CS/DS in our assay.



View larger version (75K):
[in this window]
[in a new window]
  		FIG. 3.
Differential neurite outgrowth-promoting activity of the CS/DS chains from embryonic and adult pig brains toward hippocampal neurons. E16 mouse hippocampal neurons were cultured on plastic coverslips coated with P-ORN and then Es-CS/DS (A), or As-CS/DS (B) for 24 h and fixed and stained for microtubule-associated protein 2 and neurofilament as described under "Experimental Procedures." In the control experiments, neurons were cultured on coverslips coated with P-ORN alone (C). Note the increased number of neurites for the neurons cultured on the substrate prepared with Es-CS/DS, whereas no such activity was observed for the substrate prepared with As-CS/DS. Ei-CS/DS showed a similar promoting activity to Es-CS/DS, but Ai-CS/DS exhibited no significant activity (data not shown). Scale bar, 20 µm.

 



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 4.
Dose dependence of the neurite outgrowth promotion by embryonic pig brain-derived CS/DS chains in hippocampal neurons. E16 mouse hippocampal neurons were cultured on different substrates at various doses for 24 h, and then the neurites produced were morphometrically analyzed as described under "Experimental Procedures." A, P-ORN alone; B, P-ORN + Ai-CS/DS (2 µg); C, P-ORN + As-CS/DS (2 µg); D, P-ORN + Ei-CS/DS (2 µg); E, P-ORN + Es-CS/DS (2 µg); F, P-ORN + a mixture of As-CS/DS (2 µg) and Es-CS/DS (2 µg); G, P-ORN + Es-CS/DS (4 µg); H, P-ORN + Es-CS/DS (8 µg). One hundred cells per coverslip were randomly chosen, and the length of the neurites was measured. The total length of neurites per cell is shown. The values represent the mean ± S.E. in duplicate from three independent experiments. **, 0.001 < p < 0.01, and ***, p < 0.001, significant difference from the control values obtained in the experiments using P-ORN-coated coverslips.

 
Morphometric Analysis of Hippocampal Neurons Cultured on CS/DS-coated Substrates—Morphological analysis of the hippocampal neurons showed that there was no significant difference in the length of the longest neurite per cell among the cells cultured on the Es-, Ei-, As-, and Ai-CS/DS-coated substrates (Fig. 5A), whereas the mean number per cells of primary neurites, which were longer than the cell body, was much larger for the neurons cultured on the Es- or Ei-CS/DS-coated substrate than on As- or Ai-CS/DS or on P-ORN alone (Fig. 5B). In the mixing experiments, As-CS/DS did not neutralize the effects of Es-CS/DS on the mean length of the longest neurites per cell (Fig. 5A) or the mean number of primary neurites per cell (Fig. 5B). These findings suggest that the CS/DS chains from both PBS-soluble and -insoluble fractions of embryonic pig brains have the capacity to induce multiple neurites, giving a dendrite-like morphology. It is unlikely that the neurite-stimulating effects of Es-CS/DS and Ei-CS/DS were attributable to the small peptide moiety attached to the CS/DS chains, because not only Es- and Ei-CS/DS but also As- and Ai-CS/DS carried small linkage region peptides with similar compositions of amino acids but showed distinct effects on neurite outgrowth.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 5.
Promotion of dendrite-like neurite outgrowth by the embryonic pig brain-derived CS/DS chains in hippocampal neurons. E16 mouse hippocampal neurons were cultured on substrate coated with P-ORN then with Es-, Ei-, As-, or Ai-CS/DS (2 µg) for 24 h, and then the neurites generated were morphometrically analyzed as described in the legend to Fig. 4. The mean length of the longest neurite per cell (A) and the mean number of primary neurites per cell (B) is shown. 1, P-ORN + Ei-CS/DS; 2, P-ORN + Es-CS/DS; 3, P-ORN + a mixture of Es-CS/DS and As-CS/DS; 4, P-ORN + Ai-CS/DS; 5, P-ORN + As-CS/DS; 6, P-ORN alone. The values represent the mean ± S.E. from three independent experiments performed in duplicate. **, 0.001 < p < 0.01, significant difference from the control values obtained in the experiments using P-ORN-coated coverslips.

 
Involvement of IdoUA in the Neuritogenic Activity of Es-CS/DS—To confirm the contribution of the CS/DS moiety of Es-CS/DS to the promotion of neurite outgrowth and to characterize the CS/DS motifs responsible for this activity, Es-CS/DS was digested with chondroitinases ABC, AC-I, or B, and the digests were individually coated on coverslips precoated with P-ORN for cell culture analysis (Fig. 6). Chondroitinase ABC catalyzes the eliminative cleavage of nearly all the galactosaminidic linkages in CS/DS chains (3, 43). Chondroitinase AC-I cleaves the galactosaminidic linkages bound to GlcUA in the CS and CS/DS hybrid chains unless GlcUA is 2-O-sulfated (43). In contrast, chondroitinase B cleaves the galactosaminidic linkages bound to IdoUA in DS or CS/DS hybrid chains (43). As expected, chondroitinase ABC digestion largely abolished the promoting effects of Es-CS/DS. Likewise, chondroitinase AC-I treatment resulted in the loss of most activity. Intriguingly, the treatment with chondroitinase B also eliminated the neurite outgrowth-promoting activity of Es-CS/DS, suggesting that the IdoUA-containing domain(s) was deeply involved in the promotion. The neurite extension activity was not affected by treatment with heat-inactivated chondroitin lyases or their optimal buffers (data not shown). These results together suggest that the CS/DS moiety was responsible for the neurite outgrowth-promoting activity of Es-CS/DS, and its IdoUA-containing domain(s) was the essential structural element(s) for this activity.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 6.
Effects of chondroitinase digestion on the neurite outgrowth-promoting activity of Es-CS/DS. Es-CS/DS chains or their enzymatic digests were coated on the P-ORN surface and used as substrates for the neurite outgrowth promotion assays as described in the legend to Fig. 3. Quantitative analyses of the obtained data were carried out as described in the legend to Fig. 4. A, P-ORN + Es-CS/DS (2 µg); B, P-ORN + chondroitinase ABC-treated Es-CS/DS (2 µg); C, P-ORN + chondroitinase AC-I-treated Es-CS/DS (2 µg); D, P-ORN + chondroitinase B-treated Es-CS/DS (2 µg); E, P-ORN alone. The values represent the mean ± S.E. from three independent experiments performed in duplicate. **, 0.001 < p < 0.01, significant difference from the control values obtained in the experiments using P-ORN-coated coverslips.

 
Embryonic CS/DS Chains Contain a Significant Proportion of IdoUA Units as IdoUA-GalNAc(4-O-sulfate)—As described above chondroitinase B digestion resulted in a complete loss of activity to promote neurite outgrowth in Es-CS/DS, suggesting that embryonic pig brain-derived CS/DS chains have some IdoUA-containing structures. Therefore, all the four CS/DS preparations derived from embryonic and adult pig brains were extensively digested with chondroitinases ABC, AC-I, or B, and the resultant unsaturated disaccharides were individually analyzed by anion-exchange HPLC for identification and quantification. The linkages susceptible to chondroitinases ABC and AC-I partly overlap and should correspond to the amounts of GlcUA and IdoUA, or GlcUA alone, respectively. In practice, values tend to be slightly underestimated, reflecting the occasional presence of rare linkages resistant to these enzymes (3, 43). The data, obtained from digestion experiments with chondroitinases ABC and AC-I, are summarized in Table I and illustrated in Fig. 7. Remarkably, the chondroitinase AC-I digests of Es-CS/DS and Ei-CS/DS had lower proportions (39.1% for Es-CS/DS and 37.4% for Ei-CS/DS) of {Delta}Di-4S than the chondroitinase ABC digests of the respective fractions (45.3% for Es-CS/DS and 43.7% for Ei-CS/DS), and the total amounts of disaccharides obtained by the chondroitinase AC-I digestion accounted for only 88.2% (Es-CS/DS) and 86.0% (Ei-CS/DS) of those obtained by the chondroitinase ABC digestion. By contrast, digestions of As- and Ai-CS/DS with chondroitinases ABC and AC-I showed no obvious difference in the proportion of {Delta}Di-4S. The total obtained by digestion with the latter enzyme accounted for 96.2% (As-CS/DS) and 95.4% (Ai-CS/DS) of that obtained by digestion with the former. In view of the substrate specificities of these two enzymes, the difference in the recovery of total disaccharides from the chondroitinases ABC and AC-I digests of Es-CS/DS or Ei-CS/DS should largely correspond to the IdoUA-containing disaccharides. In contrast, the slightly lower recovery of disaccharides in the chondroitinase AC-I digest of As-CS/DS or Ai-CS/DS compared with the chondroitinase ABC digest was likely due to chondroitinase AC-I-resistant structures such as the D unit (43) since chondroitinase B digestion of As-CS/DS and Ai-CS/DS gave rise to no appreciable amount of di- or oligosaccharides (data not shown). The presence of IdoUA units in Es-CS/DS and Ei-CS/DS was supported by the following observations. 1) The chondroitinase AC-I digest gave a small yet significant proportion of resistant oligosaccharides (the degree of polymerization was 4–8) in addition to the predominant disaccharides on gel filtration (Fig. 8A) and anion-exchange HPLC (data not shown). 2) The chondroitinase B digest also gave some minor peaks corresponding to cleavable oligosaccharides in addition to a small proportion of {Delta}Di-4S as revealed by anion-exchange HPLC and gel filtration chromatography (Fig. 8B). Taken together, these results indicated that the Es-CS/DS and Ei-CS/DS chains contained a significant proportion of IdoUA-bearing disaccharides (8 ~ 9%), whereas such disaccharides accounted for less than 1% of the As-CS/DS and Ai-CS/DS chains. Most if not all of the IdoUA residues in Es-CS/DS and Ei-CS/DS appear to occur as the iA unit, IdoUA{alpha}1–3GalNAc(4-O-sulfate) (the "i" in "iA" stands for IdoUA) (44).



View larger version (61K):
[in this window]
[in a new window]
  		FIG. 7.
Comparison of the yields and disaccharide compositions obtained by digestion with chondroitinases ABC and AC-I of the CS/DS chains from embryonic and adult pig brains. Es-, Ei-, As-, and Ai-CS/DS were exhaustively digested with chondroitinases ABC or AC-I, and the resultant unsaturated disaccharides were identified by anion exchange HPLC as described under "Experimental Procedures." Percent recoveries of disaccharides obtained by digestion were calculated based on the peak area in the HPLC chromatograms and are expressed in molar proportions. Es- or Ei-CS/DS:ABC and Es- or Ei-CS/DS:AC-I represent Es- or Ei-CS/DS digested with chondroitinases ABC or AC-I, whereas As- or Ai-CS/DS:ABC and As- or Ai-CS/DS:AC-I represent As- or Ai-CS/DS digested with chondroitinases ABC or AC-I, respectively.

 



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 8.
Analysis of the distribution of IdoUA-containing disaccharides in the embryonic pig brain-derived CS/DS chains. Es-CS/DS and Ei-CS/DS were digested with either chondroitinase AC-I or chondroitinase B, and the resultant oligosaccharides were labeled with a fluorophore 2AB. The size distribution of the 2AB-labeled oligosaccharides was then analyzed on a column (1.0 x 300 mm) of Superdex Peptide. The elution profiles of the chondroitinase AC-I (A) and chondroitinase B (B) digests of Es-CS/DS are shown. The sizes of the resolved oligosaccharide peaks are indicated by the number of constituent monosaccharides: 2–10, di- to decasaccharides. Note that two putative disaccharide peaks (2a and 2b) in panel A and two putative tetrasaccharide peaks (4a and 4b) in panel B, which differed in sulfate content, were dissolved. The chondroitinase AC-I and chondroitinase B digests of Ei-CS/DS exhibited a similar gel filtration pattern to those of Es-CS/DS (data not shown). The inset is an expanded (8-fold) scale version to show the size distribution of small oligosaccharides at low levels.

 
Distribution of IdoUA Residues along Embryonic CS/DS Hybrid Chains—The distribution of IdoUA along the Es-CS/DS and Ei-CS/DS chains was investigated by enzymatic digestion followed by fluorescence labeling, gel filtration, and MALDETOF MS analysis. As described above, the linkages susceptible to chondroitinases AC-I and B are mutually exclusive. Chondroitinase AC-I-resistant oligosaccharides correspond to the contiguous IdoUA-containing residues in addition to oligosaccharides containing a rare D unit. The unsaturated uronic acid residues, which are located at the non-reducing ends of the chondroitinase B-produced oligosaccharides, must be derived from IdoUA (43). The chondroitinase AC-I digest of Es-CS/DS yielded predominantly disaccharides with a smaller proportion (8% in terms of disaccharide units contained) of tetra- to octasaccharides, as shown by gel filtration on a Superdex Peptide column (Fig. 8A), which corresponded most likely to contiguous sequences of IdoUA-containing disaccharides. The chondroitinase AC-I digest of Ei-CS/DS exhibited a similar elution pattern to that of Es-CS/DS on the Superdex Peptide column (data not shown). Among these oligosaccharides the tetrasaccharides were the major components, accounting for ~70% (as disaccharide units) of the chondroitinase AC-I-resistant oligosaccharides. Chondroitinase B treatment of these oligosaccharides gave predominantly mono-sulfated disaccharides (data not shown), confirming their IdoUA-containing nature. MALDITOF MS analysis of the tetrasaccharide fraction revealed that it was a mixture of non-, mono-, and di-sulfated tetrasaccharides (data not shown). Likewise, the hexasaccharide fraction was found to contain mono-, di-, and tri-sulfated hexasaccharides. Analysis of the octasaccharide fraction was not possible due to the limited amount available. The higher proportion of tetra- than hexa- and octasaccharides in the chondroitinase AC-I-resistant sequences suggested that a majority of the IdoUA-containing disaccharides are flanked by a GlcUA-containing unit in the Es-CS/DS and Ei-CS/DS polymer chains, whereas hexa- and octasaccharides contain contiguous IdoUA-containing units.

Consistent with the extensive breakdown of Es-CS/DS by chondroitinase AC-I (Fig. 8A), chondroitinase B cleaved only 7% of all the galactosaminidic bonds in Es-CS/DS (Fig. 8B), suggesting that the Es-CS/DS chains have hybrid structures containing both GlcUA and IdoUA in the individual polymer chains. Among the oligosaccharides released by chondroitinase B, disaccharides accounted for only a small proportion (12% (mol/mol)) of the resolved total, and several intermediate-sized oligosaccharide peaks ranging from tetra- to at least decasaccharides were also observed (Fig. 8B), suggesting that only a small portion of IdoUA-containing disaccharides formed contiguous sequences in the parent polymers. These results are consistent with the above finding that more than half of the chondroitinase AC-I-resistant structures were isolated as tetrasaccharides. On the other hand chondroitinase B digestion produced, in addition to disaccharides, some tetrasaccharides, which accounted for 33% (mol/mol) of the resultant oligosaccharides. These disaccharides should be derived from tetrasaccharide sequences formed by two consecutive IdoUA-containing disaccharide units, whereas the tetrasaccharides are probably derived from hexasaccharide sequences containing an internal GlcUA-containing disaccharide unit sandwiched by two IdoUA-containing disaccharide units. Because disaccharides and tetrasaccharides accounted for 45% (mol/mol) of the total amount of resolved oligosaccharides obtained by chondroitinase B digestion, approximately half the IdoUA-containing disaccharides are clustered to form relatively short IdoUA-containing domains scattered along the Es-CS/DS chains. In the case of Ei-CS/DS, chondroitinase B cleaved ~8% of all the galactosaminidic bonds, and the digest exhibited a very similar elution pattern (data not shown) to that of Es-CS/DS described above, suggesting that the structures of Es-CS/DS and Ei-CS/DS are similar in the disaccharide composition, hydrodynamic size, proportions of IdoUA to GlcUA residues, and their distribution along the parent polymers.

The Es-CS/DS Chains Have Greater Capacity to Bind Various Heparin Binding Growth Factors Than the As-CS/DS Chains—Accumulating evidence indicates that the CS/DS chains can act as co-receptors for various growth factors, and such interactions have been demonstrated to be involved in the stimulation of neuronal migration as well as the promotion of neurite outgrowth by heparin binding growth factors such as PTN and MK (45–48). Provided that the signaling of various growth factors is involved in the neuritogenic activities of Es-CS/DS, Es-CS/DS may bind certain growth factors. Therefore, we compared the binding activities of Es-CS/DS and As-CS/DS using a BIAcore system toward the heparin binding growth factors FGF1, FGF2, FGF10, FGF18, PTN, MK, and HB-EGF, all of which are expressed in the developing brain and have neuroregulatory effects (reviewed in "Discussion" in Ref. 39). An aliquot (200 ng) of each growth factor was passed over the sensor chip on which Es-CS/DS or As-CS/DS (0.37 ng each) was immobilized, and the changes in response were recorded. As shown in Fig. 9, all the growth factors tested except for FGF1 and HB-EGF bound to both Es-CS/DS and As-CS/DS. However, Es-CS/DS had greater capacity to bind five of the growth factors (Fig. 9), suggesting that it possesses more binding sites for these growth factors than As-CS/DS.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 9.
Comparison of the binding of various heparin binding growth factors to immobilized As-CS/DS, Es-CS/DS, or chondroitinase B-treated Es-CS/DS. Various growth factors (200 ng each) including PTN, MK, FGF1, FGF2, FGF10, FGF18, and HB-EGF were injected onto a sensor chip on which As-CS/DS, Es-CS/DS, or chondroitinase B-treated Es-CS/DS was immobilized. The maximum responses in the association phase are recorded. The values represent the mean ± S.D. from three injections. RU, resonance units.

 
IdoUA Residues Are Essential for the Binding of Growth Factors to Es-CS/DS—As described above IdoUA residues are indispensable for the neurite outgrowth-promoting activity of Es-CS/DS. Hence, to investigate whether the IdoUA-containing epitopes in Es-CS/DS are required also for its efficient binding to the growth factors, the biotinylated Es-CS/DS was digested with chondroitinase B, and the resultant digest was immobilized on the sensor chip. The immobilized amount of sugar chains was controlled at 0.37 ng, which was the same as that of the undigested Es-CS/DS or As-CS/DS immobilized. Various growth factors (200 ng) were then individually applied to this sensor chip, and the responses were compared with those obtained using the sensor chip coated with Es-CS/DS- or As-CS/DS. Fig. 9 shows that chondroitinase B digestion markedly reduced the binding activity of Es-CS/DS toward all the growth factors tested at similar levels to those obtained with As-CS/DS, suggesting that the IdoUA-containing structures in Es-CS/DS are essential for its binding to these growth factors. Chondroitinase ABC in addition to chondroitinase B also displayed similar effects on the binding of Es-CS/DS to various growth factors (data not shown).

Kinetic Analysis of the Binding of Growth Factors to Es-CS/DS—Kinetic analysis of the binding of various growth factors to immobilized Es-CS/DS was carried out. The sensorgrams are shown in Fig. 10, and the obtained kinetic parameters are summarized in Table II. PTN, MK, FGF2, FGF10, and FGF18 showed high affinity binding, and the Kd values were roughly comparable with those reported for CS-E and heparin (39). Among the growth factors tested, FGF18 showed the strongest binding to E-CS/DS at a Kd value of 6.5 nM, and MK bound to Es-CS/DS with a relatively low affinity (Kd = 55.9 nM). Kinetic analysis of the binding of various growth factors to As-CS/DS or to a chondroitinase B digest of Es-CS/DS was not possible due to the slow and non-significant binding. These results together suggest that the activities of the brain CS/DS chains to bind various heparin binding growth factors change during development and that the IdoUA residues are involved in the binding.



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 10.
Sensorgrams of the binding of various growth factors to the immobilized Es-CS/DS chains. Various concentrations of PTN (A), MK (B), FGF2 (C), FGF10 (D), or FGF18 (E) and HB-EGF (F) were injected onto the Es-CS/DS-immobilized sensor chip. The long arrows indicate the beginning of the association phase, and the short arrows indicate the beginning of the dissociation phase. RU, resonance units.

 


View this table:
[in this window]
[in a new window]
  		TABLE II
Kinetic parameters for the interactions of various heparin binding growth factors with immobilized CS/DS chains from embryonic pig brains

The ka, kd, and Kd values were calculated from the sensorgrams using five or more concentrations of each growth factor. The values are expressed as the mean ± S.D.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In this study we demonstrated that the CS/DS chains derived from embryonic pig brains displayed marked neuritogenic activity and strong binding to various heparin binding growth factors. In contrast, the CS/DS chains derived from adult pig brains showed lower growth factor binding activity and no significant neuritogenic activity. Structural analysis of the CS/DS chains showed that the embryonic pig brain-derived CS/DS chains are relatively low-sulfated in contrast to our prediction of larger amounts of the D and/or E units and also unexpectedly contain a significant proportion of IdoUA-containing iA disaccharide units, which were hardly detected at all in adult pig brain-derived CS/DS chains. Here, we analyzed the CS/DS chains of the PBS-soluble and -insoluble PGs in embryonic or adult pig brains, which showed similar neuritogenic activity and were structurally indistinguishable in the disaccharide composition, molecular size, proportions of IdoUA to GlcUA residues and their distribution along the CS/DS chains. The former would at least represent major secreted PGs such as neurocan, phosphacan, brevican, versican, and aggrecan (1), whereas the latter would contain glycosylphosphatidylinositolanchored or transmembrane PGs such as receptor-type protein-tyrosine phosphatase {zeta} (RPTP{beta}/PTP{zeta}), neuroglycan C, appican, and NG2 (1). It remains to be investigated whether the structural and functional changes of the CS/DS side chains of PGs during the brain development are unique to individual PGs by isolating each CS-PG species.

The chondroitinase B-susceptible IdoUA-containing structures in Es-CS/DS were essential for its neuritogenic and growth factor binding activities. The structural importance of IdoUA relates to the flexibility of the sugar ring (49, 50), which allows substantial conformational changes and is important in interactions with proteins (for review, see Ref. 51) including growth factors such as FGF7 (52) and hepatocyte growth factor/scatter factor (53, 54). These and our observations suggest that not only sulfation profiles but also the composition and arrangements of GlcUA/IdoUA in brain CS/DS chains significantly change during development and are involved in the striking biological activities.

The marked neuritogenic activity of Es-CS/DS and Ei-CS/DS afforded E16 mouse hippocampal neurons the dendrite-like morphology, which is consistent with the recent finding that over-sulfated DS preparations from marine organisms stimulated multiple neurite formation (19). Further structural characterization of the IdoUA-containing structures in the Es-CS/DS and Ei-CS/DS chains showed that most of the chondroitinase AC-I-resistant oligosaccharides were tetrasaccharides and that only a small amount of disaccharides was released by chondroitinase B, indicating that most of the IdoUA residues were not connected with each other to form long sequences in the Es-CS/DS and Ei-CS/DS chains. The elimination of the neuritogenic activity of Es-CS/DS by chondroitinase AC-I digestion was probably due to the extensive fragmentation of the CS/DS chains, which might have caused inefficient immobilization of the resultant small oligosaccharides onto the coverslip (19). Thus, it is unlikely that sequences containing consecutive IdoUA residues are required for the promotion of neurite outgrowth, but CS/DS hybrid structures, which contain GlcUA and IdoUA linked to each other, are probably essential for the activity. This concept is consistent with our recent finding that the DSD-1 epitope, essential for the neurite outgrowth of DSD-1-PG, is also a CS/DS hybrid structure as revealed by enzymatic analysis and that the chondroitinase B treatment of DSD-1-PG resulted in a significant reduction in its neuritogenic activity (19). Hence, the biosynthetic expression of such CS/DS hybrid structures in the early developmental stages of the brain appears to be regulated to play important roles in the neurogenesis. In this context, Mikami et al. (55) recently demonstrated the acceptor specificity of three human CS/DS 4-O-sulfotransferases, which transfer sulfate to the C4 position of GalNAc residues in CS/DS hybrid sequences. It is of particular interest to investigate the regulatory mechanisms in the brain of biosynthetic glycosyltransferases and sulfotransferases for CS/DS chains (2), including these enzymes, other sulfotransferases (56), and GlcUA C5-epimerase (57), which converts GlcUA to IdoUA.

Analysis of the interactions between various heparin binding growth factors and Es-CS/DS or As-CS/DS in the BIAcore system revealed that all the growth factors tested except for FGF1 and HB-EGF bound to both Es-CS/DS and As-CS/DS and that the Es-CS/DS chains were better able to bind these growth factors than were the As-CS/DS chains, suggesting that Es-CS/DS has more binding sites for these growth factors despite its similar molecular size to As-CS/DS. Chondroitinase B digestion resulted in a marked reduction in the binding of Es-CS/DS to all these growth factors to levels comparable with those of As-CS/DS, suggesting that the binding of Es-CS/DS to these growth factors is attributable to the IdoUA-containing structures. Kinetic analysis revealed strong binding (Kd values of 6.5~55 nM) between Es-CS/DS and FGF2, FGF10, FGF18, PTN, or MK, all of which are highly expressed in the developing brain (for review, see "Discussion" in Ref. 39) and have neuroregulatory activities, implying that the CS/DS chains at cell surfaces and/or in the extracellular matrices in the embryonic pig brain may play roles in the regulation of biological functions of various heparin binding growth factors.

Although the molecular mechanism of the action of CS/DS chains in neuritogenesis is only poorly understood, emerging data suggest that the neuroregulatory effects of these chains may be attributable at least in part to their binding to various growth factors, thereby regulating the signaling of the growth factors (8, 45, 46). For example, Maeda et al. (8, 46) report that the high affinity binding of PTN to receptor-type protein-tyrosine phosphatase {zeta} (PTP{zeta}/RPTP{beta}), which is a transmembrane CS-PG (or CS/DS-PG), occurs via its CS/DS chains. Moreover, the disruption of PTP{zeta}-PTN signaling by chondroitinase ABC digestion or by the addition of exogenous CS or DS chains leads to an aberrant morphogenesis of Purkinje cells in vitro (47), implying the importance of the CS/DS-mediated PTN signaling system in vivo. Our enzyme digestion experiments revealed that both neuritogenic activity and growth factor binding activities of Es-CS/DS are associated with the IdoUA-containing structures, which supports the notion that the neuritogenic activity of the CS/DS chains with specific structures may involve efficient signaling of various heparin binding growth factors or morphogens. The immobilized CS/DS chains may mimic CS/DS-PGs in the brain extracellular matrix to present growth factors secreted by cultured neurons to the neurons themselves in our assay system. The possibility cannot be excluded that the neurons express a receptor(s) for specific CS/DS structures.

It has been reported that over-sulfated CS/DS variants from various marine organisms possess neuritogenic activities and growth factor binding activities, which may reflect the possible involvement of the over-sulfated disaccharide-containing domains of the brain CS/DS in neurogenesis during the development of the brain (7, 8, 39). In this study we also found small proportions of over-sulfated disaccharides (D and E units) not only in Es-CS/DS or Ei-CS/DS (2.5–3.6%) but also in As-CS/DS and Ai-CS/DS (3.3–3.8%) preparations. However, the contributions of these over-sulfated disaccharides to the neurite outgrowth-promoting activity and/or growth factor binding activities of Es-CS/DS remain to be clarified. Even As-CS/DS, which did not promote neurite outgrowth and exhibited less activity to bind various growth factors than did Es-CS/DS, contained slightly higher proportions of D and E units than Es-CS/DS. Thus, it remains to be solved whether the functional motifs in Es-CS/DS and Ei-CS/DS necessary for its neuritogenic and/or growth factor binding activity include the over-sulfated disaccharides as previously suggested (2, 3) in addition to the IdoUA-containing disaccharide units. However, based on the present and previous findings (19), it is tempting to speculate that combinations of iA units and D/iD and/or E/iE units are likely involved in the described activities. A better understanding of the structural motifs required for neuritogenesis would provide useful information applicable to drug development for regenerating neurons and treating dementia in view of the emerging concept of neurogenesis even in the adult hippocampus (58).


   FOOTNOTES
 
* This work was supported in part by the Scientific Research Promotion Fund of the Japan Private School Promotion Foundation Grant-in aid for Exploratory Research 15659021 (to K. S.) and the National Project on Functional Glycoconjugate Research Aimed at Developing New Industry (to K. S.) from the Ministry of Education, Science, Sports, and Culture of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

§ Recipient of a postdoctoral fellowship from the Japan Society for the Promotion of Science. Back

|| To whom correspondence should be addressed: Dept. of Biochemistry, Kobe Pharmaceutical University, 4-19-1, Motoyamakita-machi, Higashinada-ku, Kobe 658-8558, Japan. Tel.: 81-78-441-7570; Fax: 81-78-441-7569; E-mail: k-sugar{at}kobepharma-u.ac.jp.

1 The abbreviations used are: CS, chondroitin sulfate; PG, proteoglycan; GAG, glycosaminoglycan; DS, dermatan sulfate; HA, hyaluronic acid; E-CS/DS, embryonic pig brain-derived CS/DS; A-CS/DS, adult pig brain-derived CS/DS; GlcUA, D-glucuronic acid; IdoUA, L-iduronic acid; HPLC, high performance liquid chromatography; 2AB, 2-aminobenzamide; MALDI-TOF MS, matrix-assisted laser desorption ionization time-of-flight mass spectrometry; FGF, fibroblast growth factor; MK, midkine; PTN, pleiotrophin; HB-EGF, heparin binding epidermal growth factor-like growth factor; PBS, phosphate-buffered saline; {Delta}HexUA, 4-deoxy-L-threo-hex-4-enepyranosyluronic acid; {Delta}Di-4S, {Delta}HexUA{alpha}1–3GalNAc(4-O-sulfate); rh, recombinant human; mIU, milli-international units; PTP, protein-tyrosine phosphatase; P-ORN, poly-DL-ornithine. Back

2 S. S. Deepa and K. Sugahara, unpublished results. Back


   ACKNOWLEDGMENTS
 
We thank Takashi Kokawa, Kobe Pharmaceutical University, for technical assistance and Prof. Akira Kurosaka, Kyoto Sangyo University, for amino acid analysis.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESUL