JBC Transcription and Nuclear Factor Monoclonals

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


     

Originally published In Press as doi:10.1074/jbc.M207105200 on September 6, 2002

J. Biol. Chem., Vol. 277, Issue 46, 43707-43716, November 15, 2002
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
277/46/43707    most recent
M207105200v1
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 Deepa, S. S.
Right arrow Articles by Sugahara, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Deepa, S. S.
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?

Specific Molecular Interactions of Oversulfated Chondroitin Sulfate E with Various Heparin-binding Growth Factors

IMPLICATIONS AS A PHYSIOLOGICAL BINDING PARTNER IN THE BRAIN AND OTHER TISSUES*

Sarama Sathyaseelan DeepaDagger §, Yuko UmeharaDagger §, Shigeki Higashiyama, Nobuyuki Itoh||, and Kazuyuki SugaharaDagger **

From the Dagger  Department of Biochemistry, Kobe Pharmaceutical University, Higashinada-ku, Kobe 658-8558, the  Department of Medical Biochemistry, Ehime University School of Medicine, Shitsukawa, Shigenobu-cho, Onsen-gun, Ehime 791 0295, and the ||  Department of Genetic Biochemistry, Kyoto University Graduate School of Pharmaceutical Sciences, Kyoto 606-8501, Japan

Received for publication, July 16, 2002, and in revised form, September 3, 2002

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

We previously observed that the cortical neuronal cell adhesion mediated by midkine (MK), a heparin (Hep)-binding growth factor, is specifically inhibited by oversulfated chondroitin sulfate-E (CS-E) (Ueoka, C., Kaneda, N., Okazaki, I., Nadanaka, S., Muramatsu, T., and Sugahara, K. (2000) J. Biol. Chem. 275, 37407-37413) and that CS-E exhibits neurite outgrowth promoting activities toward embryonic rat hippocampal neurons. We have also shown oversulfated CS chains in embryonic chick and rat brains and demonstrated that the CS disaccharide composition changes during brain development. In view of these findings, here we tested the possibility of CS-E interacting with Hep-binding growth factors during development, using squid cartilage CS-E. The binding ability of Hep-binding growth factors (MK, pleiotrophin (PTN), fibroblast growth factor-1 (FGF-1), FGF-2, Hep-binding epidermal growth factor-like growth factor (HB-EGF), FGF-10, FGF-16, and FGF-18) toward [3H]CS-E was first tested by a filter binding assay, which demonstrated direct binding of all growth factors, except FGF-1, to CS-E. The bindings were characterized further in an Interaction Analysis system, where all of the growth factors, except FGF-1, gave concentration-dependent and specific bindings. The kinetic constants ka, kd, and Kd suggested that MK, PTN, FGF-16, FGF-18, and HB-EGF bound strongly to CS-E, in comparable degrees to the binding to Hep, whereas the intensity of binding of FGF-2 and FGF-10 toward CS-E was lower than that for Hep. These findings suggest the possibility of CS-E being a binding partner, a coreceptor, or a genuine receptor for various Hep-binding growth factors in the brain and possibly also in other tissues.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Proteoglycans (PGs)1 that bear heparan sulfate (HS) glycosaminoglycan side chains have attracted much attention because of the demonstration of the involvement of HS in the developmental processes and specific signaling pathways (for review, see Refs. 1-3). Although the chondroitin sulfate (CS) glycosaminoglycan side chains of CS-PGs have attracted less attention until recently, accumulating evidence suggests the importance of these molecules in various biological functions (2, 4). CS-PGs are also ubiquitous components of the extracellular matrix of connective tissues and are also found at the surfaces of many cell types and in intracellular secretory granules (for review, see Refs. 5 and 6). Developmentally regulated expression of CS epitopes in the rodent fetus has been demonstrated by immunological studies (7), especially during central nervous system development (8-10). CS is a rich component in the extracellular matrix of the brain (11). Developmentally regulated expression and tissue-specific distribution of CS variants suggest that CS chains differing in the degree and profile of sulfation perform distinct functions during development (12).

Among variant forms of CS chains, oversulfated CS chains, such as CS-E and CS-D, are of special interest. The presence of E (GlcUAbeta 1-3GalNAc(4S,6S)) and D (GlcUA(2S)beta 1-3GalNAc(6S)) units in bovine brain (13) and E18 rat brain (14) has been reported (GlcUA stands for D-glucuronic acid whereas 2S, 4S, and 6S represent 2-O-, 4-O-, and 6-O-sulfate, respectively). These units are present in appreciable proportions in chick brains, and their expression is regulated developmentally (15). Faissner et al. (16) demonstrated that the neurite outgrowth-stimulating capacity of DSD-1-PG, derived from neonatal mouse brains, is strongly reduced by the monoclonal antibody 473HD, which recognizes a CS epitope, and the interaction is inhibited by shark cartilage CS-C. We have shown that CS-D, an oversulfated CS also derived from shark cartilage, inhibits the interactions between the monoclonal antibody 473HD and DSD-1-PG and also promotes neurite outgrowth of E18 hippocampal neurons (17, 18) and that the DSD-1-PG contains the D disaccharide unit, which is a rich and poor component in CS-D and CS-C, respectively. We have further observed that another oversulfated variant CS-E, derived from squid cartilage, also exhibits neurite outgrowth promoting activity, which is not inhibited by 473HD, suggesting a different mechanism for CS-E-induced neurite outgrowth (17, 19).

Recently, we demonstrated that the neuronal cell adhesion, mediated by the heparin (Hep)-binding growth factor midkine (MK), was specifically inhibited not only by Hep but also by squid cartilage oversulfated CS-E (14), although it was not inhibited by bovine kidney HS or other CS isoforms. Maeda et al. (20) identified pleiotrophin (PTN), a Hep-binding growth-associated molecule, as a 6B4-PG/phosphacan-binding protein in postnatal day 16 rat brain, and this binding was inhibited by CS-C. 6B4-PG/phosphacan is a major PG in the brain and corresponds to the extracellular region of a receptor-like protein-tyrosine phosphatase, PTPzeta /RPTPbeta (20, 21) and recently turned out to be identical with DSD-1-PG (22). MK and PTN constitute a unique family of Hep-binding proteins and share 45% sequence homology at the amino acid level (23, 24) in addition to many neuroregulatory activities including promotion of neurite outgrowth (25-27). PTN binds to the CS moiety of not only phosphacan, but also neurocan, another major CS-PG, in the brain (28). The CS chain of PTPzeta shows high affinity binding to MK, which is inhibited by CS-C, CS-D, and CS-E (29). Zou et al. (30) also showed that MK bound to oversulfated CS chains with a dermatan sulfate-like domain in PG-M/versican, another CS-PG expressed in midgestation mouse embryos. Despite a large number of CS-PGs identified in the mammalian brain, CS chains attached to individual core proteins have not been structurally characterized, and the specificity of the molecular interactions of the brain CS chains with these growth factors has been largely unknown. However, we found recently that appican, the PG form of amyloid precursor protein, which is a neurotrophic factor (31), contained a significant proportion (14.3%) of E unit (32), demonstrating for the first time the presence of E unit in a particular brain CS-PG.

In view of the findings that MK and PTN are expressed in the brain during embryonic development and the demonstration of oversulfated CS structures, with binding capacities to these Hep-binding growth factors in embryonic brains, we hypothesized that these oversulfated structures might also interact with other Hep-binding growth factors in the brain during embryonic development. This hypothesis was explored in the present study using oversulfated CS-E in particular and various Hep-binding growth factors. The results suggest that all of the tested growth factors can indeed bind CS-E, and kinetic studies show that some of these bindings are stronger than those toward Hep. The preliminary findings have been reported in abstract form (33).

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Materials-- Recombinant human (rh)-MK expressed in Escherichia coli and rh-fibroblast growth factor-1, (FGF-1 or acidic FGF) expressed in E. coli was from PeproTech EC LTD (London, England). rh-PTN expressed in E. coli from RELIA Tech GmbH was from Braunschweig, Germany. rh-Hep-binding epidermal growth factor-like growth factor (HB-EGF) expressed in Sf 21 insect cells was from R&D systems (Minneapolis). rh-FGF-2 (basic FGF) expressed in E. coli was from Genzyme TECHNE (Minneapolis). rh-FGF-10 expressed in E. coli was provided by Takashi Katsumata (Sumitomo Pharmaceutical Research Center, Osaka, Japan). Recombinant rat (rr)-FGF-16 and recombinant mouse (rm)-FGF-18 were prepared as reported previously (34, 35). CS-E sodium salt (super special grade) from squid cartilage (average molecular mass of 70 kDa) (14) and bovine serum albumin (BSA) were purchased from Seikagaku Corp. (Tokyo, Japan), and porcine intestinal Hep sodium salt (200 IU/mg) (average molecular mass of 15 kDa) (14) was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). EZ-LinkTM biotin-LC-hydrazide, ImmunoPure® avidin, and ImmunoPure® HABA were purchased from Pierce. BD BioCoat® streptavidin assay plates (96-well clear, flat bottom) were purchased from BD Biosciences (Bedford, MA). Goat anti-human MK antibody (IgG type) was purchased from Genzyme-Techne (Minneapolis). Alkaline phosphatase-conjugated AffiniPure rabbit anti-goat IgG (H+L) was purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). p-Nitrophenyl phosphate hexahydrate (disodium salt) was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Phosphate-buffered saline (PBS) tablets were purchased from Dainippon Pharmaceutical Co. Ltd. (Osaka, Japan). [3H]Acetyl-labeled CS-E (1.463 × 105 cpm/µg) was prepared as described previously (36) by N-deacetylation with hydrazine followed by N-reacetylation with [3H](CH3CO)2O. [3H]Glucosamine-labeled Hep sodium salt (2.85 × 105 cpm/µg) was purchased from PerkinElmer Life Sciences.

Filter Binding Assay-- Filter binding assays for various growth factors with CS-E and Hep were performed as described previously (36) with slight modifications. Varying amounts (150-450 ng) of individual growth factors were incubated with [3H]CS-E (82 ng, ~12,000 cpm), or a fixed amount (300 ng) of individual growth factors was incubated with [3H]Hep (80 ng, ~22,800 cpm) in 50 µl of 50 mM Tris-HCl, pH 7.3, containing 130 mM NaCl and 1 mg/ml BSA at room temperature for 3 h. Mixed cellulose ester filters (0.45-µm or 1.0-µm pore size, 25-mm diameter; ADVANTEC, Tokyo) were placed onto a 12-well vacuum-assisted manifold filtration apparatus and washed with 10 ml of the same buffer devoid of BSA. The growth factor along with any bound [3H]CS-E/Hep was recovered by the quick passage of the samples through the filters, and the unbound samples were eluted by washing five times with 2 ml of the same buffer. Protein-bound radioactivity was determined after immersing the filters in 1 ml of 1 M NaCl and 0.05 M diethylamine, pH 11.5, for 30 min and the radioactivity in the eluate was determined in a liquid scintillation counter (LSC-700, Aloka Co., Tokyo) using scintillation fluid containing 1.2% (w/v) 2,5-diphenyloxazole and 33% (w/v) Triton X-100.

Preparation of Biotinylated Hep and Biotinylated CS-E (37)-- Hep or CS-E was dissolved in 100 mM MES, pH 5.5, at a concentration of 2 mg/ml. The solution was mixed with a 50 mM solution of freshly prepared biotin-LC-hydrazide (Pierce) in dimethyl sulfoxide. The weight ratio of Hep or CS-E to biotin-LC-hydrazide was 20:1. EDAC hydrochloride, dissolved in the same buffer, was added to this mixture. The weight ratio of Hep or CS-E to EDAC hydrochloride was 8:1. The labeling reaction occurred overnight at room temperature by gently mixing the solution. The reaction mixture was dialyzed against PBS at room temperature for 24 h. Dialysis was carried out in a Spectra/Por molecular porous membrane tubing, with a molecular mass cutoff of 3,500 Da (Spectrum Medical Industries, Inc., Laguna Hills, CA). Hexuronic acid in each preparation was quantified by the carbazole method, using GlcUA as standard (38). The mol of biotin/mol of CS-E or Hep was determined by the modified protocol of Green (39), which uses HABA dye and avidin, which form a complex. In brief, 60 µl of 10 mM HABA in 10 mM NaOH was added to 1.94 ml of 100 mM sodium phosphate, 150 mM NaCl, pH 7.2, containing 1 mg of avidin. The absorbance of this solution at 500 nm was recorded. A solution (10 µl each) of biotinylated CS-E (1.7 mg/ml) or Hep (2.1 mg/ml) was added to 90 µl of the above reagent. A decrease in absorbance, as biotin replaced HABA, was recorded. About 4.2% of the hexuronic acid groups in CS-E were derivatized with the biotin tag. The extent of biotinylation for Hep cannot be determined because addition of Hep produced a precipitate (40). However, a comparable degree of biotinylation with CS-E is expected for Hep.

Successful biotinylation of Hep or CS-E was confirmed by enzyme-linked immunosorbent assay. In brief, 2 µg of biotinylated Hep or CS-E was immobilized on a 96-well BD BioCoat® streptavidin assay plate. After overnight blocking with 3% BSA in PBS, 1 pmol of MK in PBS was added to each well followed by incubation at room temperature for 1 h. Goat anti-human MK antibody (diluted 200-fold with PBS) was added to each well after washing with PBS containing 0.05% Tween 20 (PBST), followed by incubation for 2 h at room temperature. The wells were washed with Tris-buffered saline (TBS) containing 0.05% Tween 20 (TBST), and alkaline phosphatase-conjugated AffiniPure rabbit anti-goat IgG (diluted 5,000-fold in TBS) was added to each well and left for 1 h at room temperature. After TBST washing, p-nitrophenyl phosphate in bicarbonate buffer (0.1 M Na2CO3/NaHCO3 containing 1 mM MgCl2, pH 9.8) was added to the wells and the development of color was measured after 15 min, at 415 nm in a microplate reader (Bio-Rad model 550).

Immobilization of CS-E and Hep to the Interaction Analysis System (IAsys; Affinity Sensors, Cambridge, UK) Cuvette-- Biotinylated CS-E and biotinylated Hep were immobilized individually on the surface of a biotin cuvette as follows. The sensor surface of a biotin cuvette (Affinity sensors, Cambridge, UK) was equilibrated with PBST. Streptavidin (2 µg) was immobilized on the biotin cuvette for 15 min, and 10 µg of biotinylated CS-E or biotinylated Hep was added to the cuvette after washing with PBST. Biotinylated CS-E gave a response of 42 arc seconds corresponding to 0.3 ng of the sample, whereas biotinylated Hep gave a response of 39 arc seconds corresponding to 0.26 ng of the sample, indicating that ~0.003% of the biotinylated sample was immobilized (600 arc seconds correspond to 1 ng of protein/mm2 of the 4-mm2 cuvette surface). After 10 min the cuvette was again washed with PBST and blocked with 20 µg of BSA. The successful immobilization was tested by applying 200 ng of MK to the cuvette surface. It gave a response of 80 arc seconds for the CS-E cuvette and 125 arc seconds for the Hep cuvette. A control cuvette was prepared in a manner similar to that for the CS-E or Hep cuvette, except that no biotinylated sample was added for immobilization.

Binding Assays of Hep-binding Growth Factors to CS-E/Hep in the IAsys-- A single binding assay consisted of the following steps. To the CS-E- or Hep-immobilized cuvette, which was preequilibrated with PBST, a known concentration of each growth factor (final volume 200 µl) was added to initiate the association phase, and the binding reaction was continued for 10 min. The cuvette was then washed three times with 200 µl of PBST, and the dissociation of the bound ligate into the bulk PBST was followed for 3 min. To remove the residual bound ligate and thus regenerate the immobilized ligand, the cuvette was washed with 200 µl of 1 M NaCl. The stirrer speed was maintained at 80% (a percentage value specifying the amplitude of stirrer oscillation at 140 Hz. The scale was linear with 0 corresponding to no oscillation and 100 corresponding to the maximum level), and the temperature was maintained at 25 °C throughout the experiment. The distribution of the immobilized CS-E or Hep and of the bound growth factor on the surface of the biosensor cuvette was inspected by examination of the resonance scan, which showed that at all times these molecules were distributed on the sensor surface and therefore were not microaggregated.

To determine the binding kinetics, growth factors at varying concentrations were applied to the cuvette in the running buffer, followed by dissociation and regeneration, as described previously. Binding parameters were calculated from the association and dissociation phases of the binding reactions using the FASTfit software (Affinity Sensors). Using FASTfit, a plot of the on-rate constant, kon (obtained from association analysis) versus the ligand concentration was obtained. The slope of the line is the association rate constant, ka, and the intercept value on the y axis is the dissociation rate constant, kd. The equilibrium dissociation constant, Kd, was calculated from the ratio (kd/ka) of the dissociation and association rate constants.

For determination of Kd using the Scatchard plot analysis, the response at equilibrium, Req, was plotted against Req/[L], where L was the ligate concentration. The slope of this plot equals -1/Ka, the association equilibrium constant, and Kd was calculated using the equation Kd = 1/Ka. For determination of Kd using the binding curve analysis, the ligate concentration [L] was plotted against Req, and the Kd was equal to the ligate concentration at Rmax/2.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

In view of our previous findings that the neuronal cell adhesion mediated by MK is specifically inhibited by oversulfated CS-E from squid cartilage (14) and the presence of the CS-disaccharide E and D units in bovine, rat, and chick brains (13-15), we explored the possibility of high affinity binding of various other Hep-binding growth factors expressed in the brain during embryonic development. The representative growth factors of the MK family, EGF family, and FGF family were chosen, and the interactions of CS-E with MK, PTN, HB-EGF, FGF-1, FGF-2, FGF-10, and FGF-18 were analyzed. In addition, FGF-16, which is not expressed in the brain but is expressed in the brown adipocytes of rat embryos (41), was also tested to clarify the generality of a binding capacity of a Hep-binding growth factor to CS-E. Even though FGF-16 and FGF-18 are believed to bind Hep, no report is available demonstrating the direct binding of these growth factors to Hep.

Demonstration of the Binding of CS-E to Hep-binding Growth Factors in an Aqueous Solution by Filter Binding Assay-- In the preliminary experiments, 82 ng of a [3H]CS-E preparation was incubated with a fixed amount (300 ng) of various growth factors separately, and the binding ability was evaluated using the filter binding assay. The same amount of [3H]Hep was treated with the individual growth factors in the same manner as positive controls. The results are summarized in Fig. 1. The findings demonstrated direct binding of CS-E to MK, PTN, HB-EGF, FGF-2, FGF-10, FGF-16, and FGF-18, whereas the binding of FGF-1 to CS-E was very low compared with those of other growth factors (Fig. 1C). The binding efficiency toward CS-E of all growth factors tested, except for FGF-1, was higher than that toward Hep, whereas FGF-2 exhibited the same degree of binding to CS-E and Hep. The degree of binding exhibited by most of the growth factors toward CS-E was in the range of 10-25%, except for FGF-1 (1.2%), whereas FGF-18 showed an exceptionally high degree of binding (~40%) (Fig. 1F). In the absence of any growth factor, the amount of CS-E or Hep bound to the filter was less than 0.2%.


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 1.   Binding of squid cartilage [3H]CS-E to Hep-binding growth factors in an aqueous solution. Squid cartilage [3H]CS-E (82 ng, ~1.2 × 104 cpm) or [3H]Hep (80 ng, 2.3 × 104 cpm) was incubated with 300 ng each of rh-MK (A), rh-PTN (B), rh-FGF-1 (C), rh-FGF-2 (D), rh-FGF-10 (E), rm-FGF-18 (F), rh-HB-EGF (G), and rr-FGF-16 (H). The radioactivity bound to each growth factor was quantified by the filter binding assay as described under "Experimental Procedures." Values were obtained from the average of two separate experiments and are expressed as percentages of the radioactivity added for incubation. Estimated errors were within 5%. Filled bars represent [3H]CS-E, and open bars represent [3H]Hep. GAG, glycosaminoglycan.

To confirm the specificity of these bindings toward CS-E, the binding of varying concentrations of the growth factors (150, 300, and 450 ng) to a fixed concentration of CS-E (82 ng) was investigated by filter binding assay. All of the growth factors bound to CS-E in a dose-dependent manner. Even though FGF-1 also bound to CS-E in a dose-dependent manner, the extent of binding was far less compared with that of other growth factors (2% for FGF-1 compared with 15-50% for other growth factors). In the case of PTN, the saturation level was attained with 300 ng of the protein. However, saturation levels were not reached for all others even with 450 ng of the corresponding protein (data not shown), presumably because multiple binding sites for each growth factor are embedded along the CS-E chain. No higher concentrations were tested because of the limited availability of the proteins. As expected, [3H]CS-E also exhibited a dose dependency for its binding toward these growth factors (Fig. 2). For a fixed amount (200 ng) of the individual growth factors, a saturation curve for [3H]CS-E was observed with less than 1 µg each of all of the growth factors tested. FGF-1 also exhibited a similar trend, but only 1.2% of [3H]CS-E was bound at the saturation level (data not shown). Together, these results suggest that the bindings of these growth factors, except for FGF-1, to CS-E were not caused merely by electrostatic interactions, but rather were highly specific.


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2.   Saturation curves for the binding of squid cartilage [3H]CS-E to Hep-binding growth factors in an aqueous solution. Varying concentrations of squid cartilage [3H]CS-E (0.05, 0.1, 0.2, 0.5, 1.0, and 2.0 µg) were incubated with 200 ng each of the Hep-binding growth factors, and the radioactivity bound to each growth factor was quantified by the filter binding assay as described under "Experimental Procedures." An equal amount of [3H]CS-E without a growth factor was used as the negative control. Values were obtained by subtracting the radioactivity of the negative control from that of each sample. A, rh-PTN; B, rh-FGF-2; C, rh-FGF-10; D, rm-FGF-18; E, rh-HB-EGF; F, rr-FGF-16.

Characterization of the Binding of the Various Hep-binding Growth Factors to Immobilized CS-E in the IAsys-- The binding of growth factors to CS-E was characterized further using the IAsys, where CS-E was immobilized, and soluble growth factors were added as analytes to mimic physiological conditions. Biotinylated CS-E was immobilized onto the biotin-coated sensor surface of the IAsys cuvette via streptavidin as described under "Experimental Procedures." A Hep-immobilized cuvette was used to determine the positive control values. In both cases, 0.26~0.30 ng of the biotinylated sample was immobilized (see "Experimental Procedures"). The growth factors (200 ng each) were then added to the CS-E- or Hep-immobilized cuvette, individually, which had been equilibrated with PBST. Sensorgrams of the MK binding to the CS-E and Hep cuvettes are shown as representatives in Fig. 3. The binding of the growth factor to immobilized CS-E or Hep was indicated by a steady increase in the response with time up to 10 min, representing the association phase. This was followed by a dissociation phase generated by washing with PBST, and then the sensor surface was regenerated by washing with a 1 M NaCl solution. MK gave a response of 80 arc seconds for the CS-E cuvette (Fig. 3A) and 125 arc seconds for the Hep cuvette (Fig. 3B), indicating that 0.5 ng and 0.8 ng of MK bound to the CS-E cuvette and Hep cuvette, respectively. The response factors generated by each growth factor differed from one another, when compared using a fixed concentration (200 ng) of an analyte, where all of the growth factors gave a higher response with the Hep cuvette than the CS-E cuvette (Fig. 4), except for FGF-18, which gave comparable responses for CS-E and Hep cuvettes (Fig. 4F). Even though FGF-1 bound to the CS-E cuvette appreciably (Fig. 4C), the degree of the binding was very low compared with that of other growth factors, consistent with the results of filter binding assays, where FGF-1 showed a lesser degree of binding to [3H]CS-E (Fig. 1C). The binding of all growth factors to CS-A-, CS-C-, and CS-D-immobilized cuvettes was very low, comparable with their binding to the control cuvette without immobilized CS-E or Hep (data not shown), indicating that among the CS variants, CS-E exhibited high specificity for these growth factors like Hep. The filter binding assay using [3H]CS-D also showed that its binding to MK was less than 2% compared with 12% obtained using [3H]CS-E.


View larger version (11K):
[in this window]
[in a new window]
  		Fig. 3.   Sensorgrams for the binding of MK to CS-E- and Hep-immobilized cuvette. Biotinylated CS-E or biotinylated Hep (10 µg) was immobilized on the streptavidin-coated surface of an IAsys cuvette, and 200 ng of MK was injected into CS-E-immobilized surface (A) and Hep-immobilized surface (B), individually, as described under "Experimental Procedures." Long arrows indicate the beginning of the association phase initiated by the injection of MK, and short arrows indicate the beginning of the dissociation phase initiated by washing with the running buffer.


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 4.   Analysis of the binding of Hep-binding growth factors to immobilized CS-E in an IAsys. Biotinylated CS-E or biotinylated Hep (10 µg) was immobilized on the streptavidin-coated surface of an IAsys cuvette, and the Hep-binding growth factors (200 ng each) were injected individually into the CS-E-immobilized surface as described under "Experimental Procedures." Values on the vertical axis represent the change in resonance angle (response) expressed in arc seconds. Six hundred arc seconds correspond to 1 ng of protein/mm2 of the cuvette surface. A, rh-MK; B, rh-PTN; C, rh-FGF-1; D, rh-FGF-2; E, rh-FGF-10; F, rm-FGF-18; G, rh-HB-EGF; H, rr-FGF-16. Filled bars represent CS-E, and open bars represent Hep.

Kinetics of Growth Factor Binding to Immobilized CS-E-- The kinetic parameters for the interaction of the growth factors with CS-E or Hep were determined by varying the concentrations of the individual growth factors and studying their interactions with CS-E- or Hep-immobilized cuvettes using the IAsys. Fig. 5, A and C, shows the overlay of sensorgrams obtained by applying varying concentrations of MK and PTN, respectively, to the CS-E-immobilized cuvette. Varying concentrations of MK or PTN were added to the CS-E-immobilized cuvette, which had been equilibrated with PBST. Fig. 5, B and D, shows the overlay of sensorgrams for varying concentrations of MK and PTN toward Hep. At higher concentrations of analytes, a decrease in response, after reaching equilibrium in the association phase, was often observed for some sensorgrams, for example in Fig. 5A. This may be caused by either the physisorption, rather than chemisorption, of the growth factor to the sensor surface or the oligomerization of the growth factors at high concentrations. A sharp peak generated at the beginning of the dissociation phase in some of the sensorgrams was presumably caused by a shock by a sudden change in the refractive index or dielectric property created, when the cuvette was emptied and refilled with a fresh buffer. Both MK and PTN exhibited a dose-dependent binding toward CS-E and Hep. These growth factors gave a dose-dependent increase in response until it reached the saturation level. Their binding toward CS-E and Hep was monophasic, and there was no evidence for secondary binding sites.


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 5.   Overlaid sensorgrams for the CS-E binding and Hep binding kinetics of MK and PTN. The binding assays were carried out as described in the legend to Fig. 4, except that the indicated concentrations of rh-MK (A and B) and rh-PTN (C and D) were applied to CS-E-immobilized (A and C) and Hep-immobilized (B and D) cuvettes. Long arrows indicate the beginning of the association phase initiated by the injection of varying concentrations of the growth factors, and short arrows indicate the beginning of the dissociation phase initiated by the running buffer.

The overlay of sensorgrams for FGF-2, FGF-10, and FGF-18, obtained by applying varying concentrations of these growth factors to the CS-E- and Hep-immobilized cuvette, is given in Fig. 6. The ka, kd, and Kd values for each growth factor were calculated using the FASTfit software, and the results are summarized in Table I. For each growth factor, only four sensorgrams of lower concentrations were used for calculating the parameters in most of the cases because excessive binding was observed at the highest concentration (for example, see Fig. 6, C and E) and may partially represent the physisorption or the oligomerization of growth factors as discussed above.


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 6.   Overlaid sensorgrams for the CS-E binding and Hep binding kinetics of FGF-2, FGF-10, and FGF-18. The binding assays were carried out as described in the legend to Fig. 4, except that the indicated concentrations of rh-FGF-2 (A and B), rh-FGF-10 (C and D), and rm-FGF-18 (E and F) were applied to CS-E-immobilized (A, C, and E) and Hep-immobilized (B, D, and F) cuvettes. Long arrows indicate the beginning of the association phase initiated by the injection of varying concentrations of the growth factors, and short arrows indicate the beginning of the dissociation phase initiated by the running buffer.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Kinetic parameters for the interaction of growth factors with immobilized CS-E and Hep
The apparent ka, kd, and Kd values for the interaction of Hep-binding growth factors with immobilized CS-E or Hep were determined using the IAsys as described under "Experimental Procedures." The S.E. is derived from the deviation of the data from a one-site binding model and was calculated by matrix inversion using the FASTfit software provided with the instrument. For each set of values of kon, the resulting values for ka and their associated S.E. were combined.

The ka values for the binding of each growth factor to CS-E or Hep were different from one another. The ka values for the binding of a given growth factor to CS-E and Hep were also different from each other. The ka values for the binding of MK, PTN, and FGF-18 to the CS-E cuvette were approximately 1 order of magnitude larger than those for the binding of FGF-2 and FGF-10 to the CS-E cuvette. They were 1.1 × 105, 4.5 × 105, and 5.7 × 105 M-1 s-1, respectively, for the CS-E cuvette and 3.5 × 104, 2.6 × 105, and 6.3 × 105 M-1 s-1, respectively, for the Hep cuvette, suggesting that the binding rate of MK, PTN, and FGF-18 toward CS-E was higher or comparable with that toward Hep, under the conditions employed. In contrast, the ka values of FGF-2 and FGF-10 for the CS-E cuvette were 3.3 × 104 and 6.5 × 104 M-1 s-1, respectively, and 2.4 × 105 and 1.4 × 105 M-1 s-1, respectively, for the Hep cuvette, suggesting a slower binding of these growth factors toward CS-E compared with the binding to Hep. The ka values for FGF-2 and FGF-10 to Hep were ~7-fold and ~2-fold higher than those for the bindings to CS-E. On the other hand, the kd values representing the dissociation rates of MK, PTN, FGF-2, FGF-10, and FGF-18 from the CS-E cuvette were comparable with those from the Hep cuvette. Consequently, the Kd values for the binding of MK, PTN, and FGF-18 with the CS-E cuvette were 61.6, 11.4, and 8.9 nM, respectively, and were lower or comparable with those for the Hep cuvette (204, 16.1, and 10.8 nM, respectively), whereas those for FGF-2 and FGF-10 for the CS-E cuvette were ~14-fold and ~5-fold higher, respectively, than those for the Hep cuvette. Thus, the difference in the Kd values appears to account for the differences in the ka values.

The overlay of sensorgrams obtained by applying varying concentrations of HB-EGF, FGF-16, and FGF-1 to the CS-E- and Hep-immobilized cuvette is shown in Fig. 7. The sensorgram of HB-EGF for the CS-E cuvette (Fig. 7A) differed from the others. It showed an exceptionally sharp increase in response, reached a plateau within seconds of injecting the sample, and then showed a decrease in response with time (for low concentrations of the growth factor) even during the association phase. The dissociation phase of HB-EGF was also different from others in the fact that the dissociation rate was extremely high and reached a level close to the base line even before regeneration. The ka values for the binding of HB-EGF to the CS-E and Hep cuvettes were 2.8 × 106 and 6.6 × 105, and the kd values were 4.2 × 10-2 and 0.3 × 10-2, respectively, for CS-E and Hep cuvettes. The ka and kd values of HB-EGF for the CS-E cuvette were 4-fold and 14-fold higher than those for the Hep cuvette. The Kd values for the binding of HB-EGF with the CS-E and Hep cuvette were 16 and 4.7 nM, respectively. Even though the Kd values for the CS-E and Hep cuvettes appear to be in the same range, the low Kd value for the CS-E cuvette was a result of the exceptionally high ka and kd values of HB-EGF toward the CS-E cuvette.


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 7.   Overlaid sensorgrams for the CS-E binding and Hep binding kinetics of HB-EGF, FGF-16, and FGF-1. The binding assays were carried out as described in the legend to Fig. 4, except that the indicated concentrations of rh-HB-EGF (A and B) rr-FGF-16 (C and D), and rh-FGF-1 (E and F) were applied to CS-E-immobilized (A, C, and E) and Hep-immobilized (B, D, and F) cuvettes, respectively. Long arrows indicate the beginning of the association phase initiated by the injection of varying concentrations of the growth factors, and short arrows indicate the beginning of the dissociation phase initiated by the running buffer.

The overlay of sensorgrams for FGF-16 for the CS-E cuvette (Fig. 7C) exhibited a peculiar pattern. Upon injection of the sample, there was a steady increase in response with time, and after reaching the plateau, a decline in response with time was observed during the association phase itself. As the concentration of the growth factor was increased, this reduction in response became faster. The physiological significance of these binding patterns is not clear at this stage (and will be discussed later). The overlay of sensorgrams for FGF-1 for the CS-E cuvette (Fig. 7E) showed that the rate of its binding to CS-E is much slower compared with the quick rate of the binding to the Hep cuvette (Fig. 7F). Because the FASTfit software analysis failed to give a fit for FGF-16, Kd values were calculated using two other methods: Scatchard analysis and binding curve analysis, as described under "Experimental Procedures." The results are summarized in Table II. The Kd values obtained by the two different analyses were comparable, and the results showed that FGF-16 had almost comparable affinity for Hep and CS-E. This is the first report that demonstrates the direct binding of FGF-16 and FGF-18 to Hep. For FGF-1, a saturation level was not attained even with 2 µg of the growth factor, hence the binding curve analysis was not used for the determination of Kd value. The Kd value for FGF-1, obtained by Scatchard analysis (Table II), indicated that the affinity of FGF-1 to CS-E was very low and may explain the low degree of binding of FGF-1 to CS-E, although the Kd value for the Hep binding was comparable with the reported value (91 nM), which was determined by affinity coelectrophoresis (42).

                              
View this table:
[in this window]
[in a new window]
 
Table II
Kinetic parameters for the interaction of FGF-16 and FGF-1 with immobilized CS-E and Hep
The Kd values for the interaction of FGF-16 for CS-E and Hep were calculated using Scatchard plot and binding curve analyses, and for FGF-1, the Kd value was calculated using Scatchard plot analysis, as described under "Experimental Procedures."

From the IAsys experiments, it was possible to calculate average mol of each growth factor bound/mol of CS-E or Hep, taking the response of each growth factor at the saturation level. One mol of CS-E (70 kDa) or Hep (15 kDa) appears to have a capacity to bind a maximum of 13 or 4 mol of MK, 17 or 4 mol of PTN, 8 or 5 mol of HB-EGF, 14 or 9 mol of FGF-2, 11 or 5 mol of FGF-10, 17 or 7 mol of FGF-16, and 14 or 5 mol of FGF-18, respectively. These values indicate that a CS-E or Hep chain accommodates multiple yet different numbers of various growth factors, probably reflecting distinct yet overlapping growth factor binding sequences embedded in each sugar chain.

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

In the present study, we demonstrated the interaction of various Hep-binding growth factors with CS-E in two ways: the filter binding assay and analysis in the IAsys. In both situations, all of the tested growth factors, except for FGF-1, bound to CS-E in a dose-dependent manner, suggesting that these bindings are highly specific. Although these growth factors, except FGF-1, bound more strongly to CS-E than Hep in the filter binding assay, the reverse effects were observed in the IAsys, which may reflect the environment in which the two molecules are interacting. The IAsys more closely mimics the physiological conditions, where the soluble growth factors interact with CS chains in an immobilized form at the cell surface or in the extracellular matrices. On the other hand, complexes measured in the filter binding assay, where both interacting molecules are in solution, would mimic growth factor-CS complexes released enzymatically from cell surfaces or extracellular matrices.

MK and PTN specifically bound to CS-E derived from squid cartilage with high affinity (Kd = 61.6 and 11.4 nM, respectively), and these bindings were higher than or comparable with those for Hep (Kd = 204 and 16.1 nM, respectively), which is in reasonable agreement with our previous results obtained in a BiaCore system where the binding of soluble CS-E to immobilized MK was analyzed (14). The high affinity binding (Kd = 0.58 nM) of MK to PTPzeta was reported, which was inhibited not only by CS-E but also by CS-B, CS-C, and CS-D (29). It was also reported that PG-M/versican isolated from E13 mouse embryos binds MK and PTN strongly (Kd = 1.0 nM), and the binding to MK is abolished by chondroitinase ABC digestion (30). The migration of osteoblast-like cells UMR 106 (43) and that of macrophages (44) promoted by MK are abolished by chondroitinases ABC or B digestion and also inhibited by exogenous CS-E in addition to CS-B. Maeda et al. (20) isolated PTN as a 6B4-PG/phosphacan-binding protein in the rat brain and also demonstrated that PTN binds to the CS moiety of phosphacan with high affinity (Kd = 0.25-3.0 nM), which was decreased by chondroitinase ABC digestion and was inhibited by CS-C. It was also reported that the PTN-stimulated migration of embryonic rat cortical neurons is inhibited by CS-C (21). However, in most studies, detailed structural information about the functional domain of the CS chains was lacking. In this context, our recent demonstration of the significant proportion of CS-E (14.3%) in the CS chain of appican PG (32) is highly significant and is the first demonstration of a specific brain CS-PG that contains the E unit. Previous studies have demonstrated that the CS chain in appican is responsible for the adhesion of neural cells and for promoting neurite outgrowth of primary rat hippocampal cultures, because the core of the Alzheimer's amyloid precursor protein is less potent in promoting adhesion and neurite outgrowth than appican PG (31, 46). The presence of E unit in appican PG may explain the neurotrophic activities of appican. Syndecan-1 from mouse mammary gland epithelial cells was also demonstrated to bear CS-E (48), although it is unknown whether syndecan-1 expressed in the rat central nervous system, which binds MK and PTN (49), contains CS-E.

CS-E also bound FGF family members. The FGF family comprises at least 23 proteins that play important roles in development, tissue maintenance, and tissue repair, and all family members have a Hep-binding property. The tested family members, except for FGF-16, are expressed in the brain. FGF-2 and FGF-10 bound to CS-E with lower affinity compared with Hep (based on Kd values). FGF-18 showed a strong binding, comparable with that toward Hep. In contrast, FGF-1 showed almost no binding. Each growth factor exhibited a peculiar association and dissociation pattern toward CS-E and Hep. The results appear to suggest that even lower affinity binding is physiologically significant as discussed below.

FGF-1 bound only weakly to CS-E, as indicated by a very high Kd value. The reason for this low affinity binding is unclear. Comparison of the Hep-binding domain of the tested FGF family members showed similar distribution of the basic amino acid residues. The only striking feature different from other FGF family members is the presence of the Cys137 residue within the Hep-binding domain (50). Structural characterization of the HS domain involved in the FGF-1 binding showed that the binding depends on a rarely occurring structure hallmarked by L-iduronate(2-O-sulfate)-D-glucosamine(2-N-,6-O-disulfate) units (51). The human aorta HS preparation from an old individual, enriched in such units, showed drastically enhanced binding to FGF-1 (52) compared with HS from the young subject. The low degree of binding exhibited by FGF-1 to CS-E may be correlated with any of these observations.

FGF-2 binds N-syndecan, isolated from neonatal rat brain, with high affinity (Kd = 0.5 nM) through the HS chains, although FGF-1 fails to give such a binding toward N-syndecan (53). Based on the observation that N-syndecan mRNA levels in the neonatal rat brain are significantly higher than those in the adult rat brain, an important role of N-syndecan in nerve tissue differentiation has been suggested (54). Milev et al. (28) demonstrated that the core protein of phosphacan CS-PG showed high affinity binding toward FGF-2 and potentiated its mitogenic effect. Even if the binding of phosphacan to FGF-2 was reduced to 35% after chondroitinase treatment, the mitogenic effect generated by intact phosphacan was comparable with that exhibited by the core protein alone. This was the first demonstration that suggests that CS-PGs may also regulate the access of FGF-2 to cell surface signaling receptors in nervous tissues. The present finding that the FGF-2 interacts with CS-E suggests a possibility that E units in CS side chains of PGs, along with the core protein, may also be involved in the interaction with FGF-2, if it is expressed on certain CS-PGs at specific developmental stages.

Although FGF-10 mRNA is expressed predominantly in the lung, it is also expressed in the brain at low levels, being spatially restricted in some regions, including the hippocampus, thalamus, midbrain, and brainstem, and preferentially in neurons, but not in glial cells (55). Studies using FGF-10 knockout mice have demonstrated that this molecule is essential for limb and lung formation (56). Apart from this, the requirement of FGF-10 in the development of white adipose tissue (57) and its role as mitogen for urothelial cells (58) and in the maintenance of the stem cell compartment in developing mouse incisors (59) were reported recently. The possible involvement of CS-E in these biological events remains to be clarified.

FGF-18 is a unique secreted signaling molecule in the adult lung and developing tissues (60). However, the expression of FGF-18 in lower levels in the mouse brain has also been reported (35, 61). Ohuchi et al. (62) reported the involvement of FGF18-FGF8 signaling for specification of left-right symmetry and development of the chick embryo brain, especially the midbrain, and limbs. Novel roles for FGF-18 in calvarial and limb development (63) as well as in the proximal programming during lung morphogenesis have been reported (64). It is of particular interest to clarify whether E units in CS chains of certain CS-PGs are involved in the FGF-18 signaling in the above biological events, especially because FGF-18 binding to CS-E was exceptionally strong.

From the calculated Kd values, it appears that the affinity of FGF-16 toward CS-E is comparable with that for Hep. However, the pattern of the sensorgrams generated by FGF-16 with CS-E cuvette was markedly different from other growth factors in that they decreased significantly even during the association phase. Because this pattern closely resembles that generated by FGF-1 in a Hep-immobilized cuvette, there is a strong possibility that these bindings are physiologically significant. It will be interesting to clarify whether the binding of FGF-16 to CS-E can stimulate cellular growth or the signal transduction cascade in vivo. FGF-16 is expressed predominantly in the brown adipocytes of rat embryos during embryonic days 17.5-19.5 and appears to be a unique growth factor involved in the proliferation of embryonic brown adipose tissue (65). It also induces hepatocellular proliferation and increases liver weight in vivo (34). However, a truncated rat FGF-16 with 34 amino acids removed from the N terminus induces proliferation of oligodendrocytes, isolated from the rat brain, in vitro (34). It is noteworthy that FGF-16 shares 62% amino acid identity with FGF-20, whose mRNA is expressed preferentially in the brain among the adult rat major tissues and is reported to enhance the survival of midbrain dopaminergic neurons (66).

The high affinity of HB-EGF, an EGF family member (67, 68), toward CS-E observed in the present study, as indicated by the low Kd value, is interesting in that it is a consequence of the high association and dissociation rates (Table I). Thus, the kinetic studies have revealed the quick binding of the growth factor to CS-E in the environment of an increasing concentration of HB-EGF and its fast release with a decrease in the growth factor concentration. The polypeptides of EGF family produced by neurons and glial cells play important roles in the development of the nervous system, stimulating proliferation, migration, and differentiation of neuronal, glial, and Schwann precursor cells (69). Nakagawa et al. (70) reported the neuronal and glial expression of HB-EGF in the central nervous system of prenatal and early postnatal rats and suggested that it may contribute to brain development. HB-EGF is also reported to regulate the survival of midbrain dopaminergic neurons utilizing mitogen-activated protein kinase in addition to the Akt- signaling pathway (71). In addition, a new role of HB-EGF was reported recently by Asakura et al. (72), who found that the cleavage and shedding of the membrane-bound HB-EGF by metalloproteases contribute to the cardiac hypertrophic process and suggested that inhibition of HB-EGF shedding could be a potent therapeutic strategy for cardiac hypertrophy. It is of interest to clarify whether CS-PGs, in addition to HS-PGs, are involved in the regulation of these functions of HB-EGF. The physiological significance of the low affinity bindings caused by the slow binding observed for FGF-2 and FGF-10, compared with high affinity bindings, in addition to the quick binding and release of HB-EGF remains to be clarified. There is a possibility that these low affinity motifs act as a scaffold to capture the growth factors and then hand them over to other high affinity motifs and finally direct the growth factor toward the site of interaction with its receptor target at the cell surface (73). The graded affinities of various HS oligosaccharides for FGF-1 and FGF-2 (74, 75) are consistent with this concept. CS-E with higher affinity for certain growth factors than Hep may in turn receive them from HS chains as was discussed previously for a hybrid PG, syndecan-1 containing CS-E (48).

Although most studies of growth factor interactions with PGs have concerned HS-PGs and more specifically their HS-glycosaminoglycan chains, the present findings strengthen the emerging concept (4, 44) that the interactions of CS-PGs with growth/differentiation factors might also play significant roles in developmental processes of the central nervous system and other systems. In this context, Muramatsu (44) recently reported the isolation of CS that bound strongly to MK by affinity chromatography after labeling neurons from E13 embryos with [14C]glucosamine. The disaccharide composition analysis of the strongly bound fraction showed that 30% was E unit. Kawashima et al. (76) recently demonstrated specific high affinity interactions of L- and P-selectins and chemokines with CS-E and suggested the involvement of the E unit-containing CS/dermatan sulfate in selectin- and/or chemokine-mediated cellular responses (76). The results obtained in the present study, together with accumulating evidence, may suggest the possibility that CS-glycosaminoglycan chains produced at certain developmental stages act as a binding partner, a coreceptor, or a genuine receptor for various Hep-binding growth factors and chemokines/cytokines with developmentally regulated expression. Recently, Muramatsu (44) proposed a model of MK signaling through PTPzeta , where the MK receptor is a molecular complex of PTPzeta and a transmembrane protein, viz. low density lipoprotein receptor-related protein, which was reported to function as signaling receptors upon signal reception of Wnt and reelin. The essential function of PGs in this complex may be the activation of Src family kinases, by an as yet undetermined mechanism, which will in turn activate the phosphoinositide 3-kinase to ERK downstream signaling systems. It was also reported that the phosphoinositide 3-kinase to ERK pathway acts as a downstream signaling system for MK and PTN (77, 78).

Elucidation of the minimal structural units for these interactions will be essential for therapeutic strategies to develop drugs that may selectively interfere with CS-protein interactions. Development of CS-E-based drugs will certainly be a major breakthrough, because they can even substitute for endogenous CS in agonist mode. To achieve this, two major obstacles for such advancement have to be overcome: the lack of information regarding protein-binding epitopes in CS chains and the structural variability of CS molecules from different cells and tissues. The squid cartilage CS-E, used in the present study, has unusual GlcUA(3S)-containing disaccharide units such as GlcUA(3S)-GalNAc(4S), GlcUA(3S)-GalNAc(6S), and GlcUA(3S)-GalNAc(4S,6S), where 3S represents 3-O-sulfate (45, 47) in addition to the conventional E unit. Some of these heavily sulfated structures in CS-E are comparable with the highly sulfated regions in Hep. It is unclear whether these units are involved in the observed binding of the growth factors and whether such unique structures contribute to the binding of CS chains to these growth factors in mammalian systems because the existence of such structures in the mammalian system has not yet been reported.

    FOOTNOTES

* The work at Kobe Pharmaceutical University was supported in part by a science research promotion fund from the Japan Private School Promotion Foundation and Grant-in-aid for Scientific Research (B) 12557214.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ These authors contributed equally to this work.

** To whom correspondence should be addressed: Dept. of Biochemistry, Kobe Pharmaceutical University, 4-19-1 Motoyama-kita-machi, Higashinada-ku, Kobe 658-8558, Japan. (Tel.: Int +81-78-441-7570; Fax: Int +81-78-441-7569); E-mail: k-sugar@kobepharma-u.ac.jp.

Published, JBC Papers in Press, September 6, 2002, DOI 10.1074/jbc.M207105200

    ABBREVIATIONS

The abbreviations used are: PG(s), proteoglycan(s); BSA, bovine serum albumin; CS, chondroitin sulfate; E, embryonic day(s); EDAC, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide; ERK, extracellular signal-regulated kinase; FGF, fibroblast growth factor; HABA, 2-(4-hydroxyazobenzene) benzoic acid; HB-EGF, heparin-binding epidermal growth factor-like growth factor; Hep, heparin; HS, heparan sulfate; IAsys, Interaction analysis system; MES, 2-(N-morpholino)ethanesulfonic acid; MK, midkine; PBS, phosphate-buffered saline; PTN, pleiotrophin; PTP, protein-tyrosine phosphatase; Req, response at equilibrium; RPTP, receptor-like PTP; rh, recombinant human; rm, recombinant mouse; rr, recombinant rat; TBS, Tris-buffered saline.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

1. Sugahara, K., and Kitagawa, H. (2002) IUBMB Life, in press
2. Sugahara, K., and Kitagawa, H. (2000) Curr. Opin. Struct. Biol. 10, 518-527[CrossRef][Medline] [Order article via Infotrieve]
3. Perrimon, N., and Bernfield, M. (2000) Nature 404, 725-728[CrossRef][Medline] [Order article via Infotrieve]
4. Sugahara, K., and Yamada, S. (2000) Trends Glycosci. Glycotechnol. 12, 321-349
5. Rodén, L. (1980) in The Biochemistry of Glycoproteins and Proteoglycans (Lennarz, W. J., ed) , pp. 491-517, Plenum Publishing Corp., New York
6. Hascall, V. C., and Hascall, G. K. (1981) in Cell Biology of Extracellular Matrix (Hay, E. D., ed) , pp. 39-63, Plenum Publishing Corp., New York
7. Mark, M. P., Baker, J. R., Kimata, K., and Ruch, J. V. (1990) Int. J. Dev. Biol. 34, 191-204[Medline] [Order article via Infotrieve]
8. Herndon, M. E., and Lander, A. D. (1990) Neuron 4, 949-961[CrossRef][Medline] [Order article via Infotrieve]
9. Lander, A. D. (1993) Curr. Opin. Neurobiol. 3, 716-723[CrossRef][Medline] [Order article via Infotrieve]
10. Oohira, A., Matsui, F., Tokita, Y., Yamauchi, S., and Aono, S. (2000) Arch. Biochem. Biophys. 374, 24-34[CrossRef][Medline] [Order article via Infotrieve]
11. Ruoslahti, E. (1989) J. Biol. Chem. 264, 13369-13372[Free Full Text]
12. Poole, A. R. (1986) Biochem. J. 236, 1-14[Medline] [Order article via Infotrieve]
13. Saigo, K., and Egami, F. (1970) J. Neurochem. 17, 633-647[CrossRef][Medline] [Order article via Infotrieve]
14. Ueoka, C., Kaneda, N., Okazaki, I., Nadanaka, S., Muramatsu, T., and Sugahara, K. (2000) J. Biol. Chem. 275, 37407-37413[Abstract/Free Full Text]
15. Kitagawa, H., Tsutsumi, K., Tone, Y., and Sugahara, K. (1997) J. Biol. Chem. 272, 31377-31381[Abstract/Free Full Text]
16. Faissner, A., Clement, A., Lochter, A., Streit, A., Mandl, C., and Schachner, M. (1994) J. Cell Biol. 126, 783-799[Abstract/Free Full Text]
17. Nadanaka, S., Clement, A., Masayama, K., Faissner, A., and Sugahara, K. (1998) J. Biol. Chem. 273, 3296-3307[Abstract/Free Full Text]
18. Clement, A. M., Nadanaka, S., Masayama, K., Mandl, C., Sugahara, K., and Faissner, A. (1998) J. Biol. Chem. 273, 28444-28453[Abstract/Free Full Text]
19. Clement, A. M., Sugahara, K., and Faissner, A. (1999) Neurosci. Lett. 269, 125-128[CrossRef][Medline] [Order article via Infotrieve]
20. Maeda, N., Nishiwaki, T., Shintani, T., Hamanaka, H., and Noda, M. (1996) J. Biol. Chem. 271, 21446-21452[Abstract/Free Full Text]
21. Maeda, N., and Noda, M. (1998) J. Cell Biol. 142, 203-216[Abstract/Free Full Text]
22. Garwood, J., Schnadelbach, O., Clement, A., Schutte, K., Bach, A., and Faissner, A. (1999) J. Neurosci. 19, 3888-3899[Abstract/Free Full Text]
23. Kadomatsu, K., Tomomura, M., and Muramatsu, T. (1998) Biochem. Biophys. Res. Commun. 151, 1312-1318
24. Merenmies, J., and Rauvala, H. (1990) J. Biol. Chem. 265, 16721-16724[Abstract/Free Full Text]
25. Rauvala, H. (1989) EMBO J. 8, 2933-2941[Medline] [Order article via Infotrieve]
26. Muramatsu, H., and Muramatsu, T. (1991) Biochem. Cell Biol. 177, 652-658
27. Kaneda, N., Talukder, A. H., Nishiyama, H., Koizumi, S., and Muramatsu, T. (1996) J. Biochem. (Tokyo) 119, 1150-1156[Abstract/Free Full Text]
28. Milev, P., Monnerie, H., Popp, S., Margolis, R. K., and Margolis, R. U. (1998) J. Biol. Chem. 273, 21439-21442[Abstract/Free Full Text]
29. Maeda, N., Ichihara-Tanaka, K., Kimura, T., Kadomatsu, K., Muramatsu, T., and Noda, M. (1999) J. Biol. Chem. 274, 12474-12479[Abstract/Free Full Text]
30. Zou, K., Muramatsu, H., Ikematsu, S., Sakuma, S., Salama, R. H., Shinomura, T., Kimata, K., and Muramatsu, T. (2000) Eur. J. Biochem. 267, 4046-4053[Medline] [Order article via Infotrieve]
31. Salinero, O., Moreno-Flores, M. T., and Wandosell, F. (2000) J. Neurosci. Res. 60, 87-97[CrossRef][Medline] [Order article via Infotrieve]
32. Tsuchida, K., Shioi, J., Yamada, S., Boghosian, G., Wu, A., Cai, H., Sugahara, K., and Robakis, N. K. (2001) J. Biol. Chem. 276, 37155-37160[Abstract/Free Full Text]
33. Deepa, S. S., Umehara, Y., Higashiyama, S., Itoh, N., and Sugahara, K. (2002) in 23rd Japanese Carbohydrate Symposium, Yokohama, Japan, August 22-24, 2002 (Hashimoto, H., ed) , p. 91, Gakushin Publishing Co., Suita, Japan, Abstract PI-37
34. Danilenko, D. M., Montestruque, S., Philo, J. S., Li, T., Hill, D., Speakman, J., Bahru, M., Zhang, M., Konishi, M., Itoh, N., Chirica, M., Delaney, J., Hernday, N., Martin, F., Hara, S., Talvenheimo, J., Narhi, L. O., and Arakawa, T. (1999) Arch. Biochem. Biophys. 361, 34-46[CrossRef][Medline] [Order article via Infotrieve]
35. Hu, M. C., Qiu, W. R., Wang, Y. P., Hill, D., Ring, B. D., Scully, S., Bolon, B., DeRose, M., Luethy, R., Simonet, W. S., Arakawa, T., and Danilenko, D. M. (1998) Mol. Cell. Biol. 10, 6063-6074
36. Yamane, Y., Tohno-oka, R., Yamada, S., Furuya, S., Shiokawa, K., Hirabayashi, Y., Sugino, H., and Sugahara, K. (1998) J. Biol. Chem. 273, 7375-7378[Abstract/Free Full Text]
37. Yang, B., Yang, B. L., and Goetinck, P. F. (1995) Anal. Biochem. 228, 299-306[CrossRef][Medline] [Order article via Infotrieve]
38.