phosphacan

Neuroscience Vol. 67, No. I, pp. 23 35, 1995

~ Pergamon

0306-4522(94)00069-0

Elsevier ScienceLtd Copyright "C~1995 IBRO Printed in Great Britain. All rights reserved 0306-4522/95 $9.50 + 0.00

PURIFICATION, CHARACTERIZATION AND DEVELOPMENTAL EXPRESSION OF A BRAIN-SPECIFIC CHONDROITIN SULFATE PROTEOGLYCAN, 6B4 PROTEOGLYCAN/PHOSPHACAN N. M A E D A , * t H. H A M A N A K A , * t A. O O H I R A + and M. N O D A * t § *Division of Molecular Neurobiology, National Institute for Basic Biology, Okazaki 444, Japan tDepartment of Molecular Biomechanics, The Graduate University for Advanced Studies, Okazaki 444, Japan :~Department of Perinatology and Neuroglycoscience, Institute for Developmental Research, Kasugai, Aichi 480-03, Japan Abstract--A large brain-specific chondroitin sulfate proteoglycan, identified with monoclonal antibody 6B4 (6B4 proteoglycan/phosphacan), was isolated from rat brain. Soluble proteoglycans in the phosphatebuffered saline extract from 20-day-old rat whole brain were fractionated by anion exchange chromatography and CsCI density gradient centrifugation. 6B4 proteoglycan was further purified by gel filtration and additional ion exchange chromatography. The molecular mass of 6B4 proteoglycan shifted from 800 to 300 × 103mol. wt after chondroitinase ABC digestion. The core protein was substituted with chondroitin sulfate chains with an average molecular weight of 21,000, keratan sulfate and HNK-1 carbohydrates. Glycosidase digestion of 6B4 proteoglycan with O-glycanase, N-glycanase0 endo-/~-galactosidase, or keratanase did not remove the HNK-1 epitopes. The expression of 6B4 proteoglycan was developmentally regulated in the rat cerebral cortex; appearing first at embryonic day 14, peaking at postnatal day 0, and persisting throughout adulthood at a lower level. Immunohistochemical analysis indicated that 6B4 proteoglycan was distributed along the radial glial fibers and on the migrating neurons in the embryonal rat cerebrum. The radial glial fibers were stained intensely all along their length, but the neurons in the cortical plate were not stained in contrast to the moderate staining of the migrating neurons in the intermediate zone and the subplate. From postnatal day 5 to postnatal day 20, 6B4 proteoglycan was present throughout the cortex. After postnatal day 30, staining of the neuropil was weakened, and the expression of 6B4 proteoglycan was restricted around subsets of neurons. The positive neurons were mostly non-pyramidal cells (>95%) and were relatively concentrated in layers IV and VI of the primary somatosensory cortex. Immunohistochemical analysis of the dissociated cortical neurons indicated that 6B4 proteoglycan was distributed on the cell bodies and neurites. 6B4 proteoglycan strikingly promoted neurite extension of cortical neurons from embryonic day-16 rat embryos when coated on coverslips as a substrate. 6B4 proteoglycan is a brain-specific chondroitin sulfate proteoglycan which carries keratan sulfate and HNK-I carbohydrates. The spatiotemporal expression profile and effects on the dissociated cerebral neurons suggest that 6B4 proteoglycan plays important roles in the migration and differentiation of neurons in the immature cortex, and also in the maintenance of subsets of neurons in the mature cortex.

M a n y cell surface and extracellular matrix proteins are involved in the development and maintenance of the mammalian CNS. Recently, proteoglycans have been recognized to play important roles in the developmental processes of the brain. 2~'23 Proteoglycans are complex cell surface or extracellular matrix constituents composed of a core protein and long

sulfated glycosaminoglycan chains, which are covalently attached to the core protein. In the last several years, especially glycosaminoglycan portions of proteoglycans have attracted attention because of their inhibitory effects on neurite extension and the binding activity of growth factors, s'3~'32 F r o m the detection of at least 25 different proteoglycan core proteins in the rat brain, j3'27'29however, it is possible to speculate that different core proteins have their own different functions being modified by the glycosaminoglycan portions, or vice versa. Indeed, several lines of evidence suggest that different proteoglycans are expressed by different brain components according to different time schedules. ~5'~9'36"38 Therefore, it is a prerequisite for evaluating the functional roles of proteoglycans in the brain to isolate and characterize each proteoglycan.

§To whom correspondence should be addressed at: Division of Molecular Neurobiology, National Institute for Basic Biology, 38 Nishigonaka, Myodaiji-cho, Okazaki 444, Japan. Abbreviations: CMF-HBSS, Ca z+- and Mg2+-free Hank's balanced salt solution; E, embryonic day; MAb, monoclonal antibody; NEM, N-ethylmaleimide; P, postnatal day; PBS, phosphate-buffered saline; PMSF, phenylmethylsulphonyl fluoride; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis. 23

N. Maeda et al.

24

In a previous study, z4 we p r e p a r e d a m o n o c l o n a l a n t i b o d y ( M A b ) , designated M A b 6B4, which is reactive with a large c h o n d r o i t i n sulfate proteoglycan. I m m u n o h i s t o c h e m i c a l analysis o f the adult rat h i n d b r a i n has s h o w n t h a t 6B4 proteoglycan is expressed in fairly restricted areas such as pontine nuclei a n d lateral reticular nucleus. 24 D e v e l o p m e n t a l studies o f the rat h i n d b r a i n have indicated t h a t the expression o f 6B4 p r o t e o g l y c a n is highly correlated spatiotemporally with the circuit f o r m a t i o n of the mossy fiber system. Quite recently, we a n d o t h e r groups succeeded in the c D N A cloning of this proteoglycan, a n d it was revealed t h a t 6B4 proteoglyc a n / p h o s p h a c a n is a n extracellular v a r i a n t o f the receptor-like protein tyrosine p h o s p h a t a s e , P T P ( (RPTPfl), 162°,25,26 which is also present in the f o r m of c h o n d r o i t i n sulfate p r o t e o g l y c a n ? '~5 In the present study, we purified 6B4 proteoglycan a n d characterized its c a r b o h y d r a t e moiety, a n d further e x a m i n e d the effects of this p r o t e o g l y c a n o n dissociated cerebral neurons. In addition, we analysed the d e v e l o p m e n t a l expression of this proteoglycan in the rat cerebral cortex. The functional significance o f 6B4 proteoglycan is discussed from the s p a t i o t e m p o r a l expression p a t t e r n during cerebral d e v e l o p m e n t a n d its effects o n cortical n e u r o n s in culture.

proteoglycan was eluted by 30 ml of linear NaC1 gradient (0.4-0.8 M in 20mM Tris-HC1, pH 7.8).

EXPERIMENTAL PROCEDURES

Characterization of ehondroitin sulfate Purified 6B4 proteoglycan (150 nmol as hexuronate) was treated with 0.2 M NaOH at room temperature for 24 h. After neutralization with glacial acetic acid, the sample was digested with 5#g/ml pronase at 50°C for 24h in the presence of 0.5 M Tris-HCl, pH7.8. The solution was applied to a Sepharose CL-6B (Pharmacia) column (7 mm × 30 cm), and the chondroitin sulfate was eluted with 0.4 M ammonium acetate. The size of chondroitin sulfate was estimated from the Kay values as described by Wasteson? 5 Proportions of the unsulfated, 4-sulfated and 6-sulfated disaccharide units were determined by the method of Yoshida et al. 37

Purification of 6B4 proteoglyean Twenty-day-old Sprague-Dawley rats (SLC Inc., Japan) were anesthetized, killed by decapitation, and the whole brains were dissected out. Ten grams of brain tissue was homogenized in 50 ml of a solution containing 5 mM EDTA, 5 mM N-ethylmaleimide (NEM), 1 mM phenylmethylsulphonyl fluoride (PMSF), 10 # M leupeptin, 10/~ M pepstatin A, 0.15 M NaC1, and 10 mM sodium phosphate, pH 7.2, with eight strokes in a glass-Teflon Potter homogenizer. The homogenate was centrifuged at 30,000 × g for 30 min at 4°C, and the resultant pellet was homogenized and centrifuged as above. The combined supernatant was adjusted to a final concentration of 7 M urea, 0.05 M NaCl and 0.05 M Tris-HCl, pH 7.5, with 10 M, 3 M and 1 M stock solutions, respectively, and applied to a DEAE-Toyopearl (Tosoh Corporation, Japan) column (1 × 10cm) equilibrated with 7 M urea, 0.1 M NaC1, 1 mM EDTA, 1 mM NEM, 1 mM PMSF, and 50mM Tris-HC1, pH 7.5. The column was washed with 50 ml of a solution containing 0.25 M NaC1, 7 M urea, 1 mM EDTA, 1 mM NEM, 1 mM PMSF and 50mM Tris-HC1, pH7.5, and proteoglycans were eluted with a solution containing 0.4 M NaC1, 7 M urea, 1 mM EDTA, 1 mM NEM, I mM PMSF and 50 mM Tris-HC1, pH 7.5. The eluted sample was dialysed three times against 20 volumes of a solution containing 4 M guanidine-HCl, 5 mM EDTA, 5 mM NEM and 50mM Tris-HC1, pH 8.0. Solid CsCI was added to the dialysed sample to a density of 1.38 g/ml, and the solution was centrifuged at 80,000 rpm for 50 h at 10°C using a Beckman TLA100.4 rotor. The 6B4 proteoglycan-rich fraction was next applied to a TSKgel G6000PW (Tosoh Corporation, Japan) eolumn (21.5 mm × 30 cm) equilibrated with 0.15 M NaC1, 10 mM sodium phosphate, pH 7.4. Finally, the 6B4 proteoglycan fraction was applied to a LiChrospher 4000DMAE column (Cica-MERCK, Germany), and the

Glycosidase digestion of 6B4 proteoglycan Purified proteoglycan equivalent to 20 nmol hexuronate was precipitated with ethanol and dissolved in 30 #1 of a solution containing 30 mM sodium acetate, 1 mM PMSF, 0.1 mM pepstatin A, 10mM EDTA, 10mM NEM, and 0.1 M Tris-HC1, pH 7.5. Two milliunits (mU) of proteasefree chondroitinase ABC (Seikagaku Kogyo, Japan) was added to the solution, which was then incubated for 60 min at 37°C. After precipitation with ethanol, the pellet was dissolved in 30/~1 of a solution containing 5 mM EDTA, 5 m M NEM, l m M PMSF, 0.1 mM pepstatin A, and 50mM sodium acetate, pH 5.0. One milliunit of neuraminidase (Nacalai Tesque, Japan) was added to the solution, which was then incubated for 60 min at 37°C. The same volume of a solution containing 5 mM EDTA, 5 mM NEM, 1 mM PMSF, 0.1 mM pepstatin A, 15mM sodium acetate, and 50 mM Tris-HC1, pH 7.4, was added to the sample, which was then incubated for 60 min at 37°C in the presence of 2 mU of keratanase (Seikagaku Kogyo, Japan). The core protein was precipitated with ethanol, and denatured by boiling for 2min in 5#1 of 1% SDS/10mM sodium phosphate, pH 7.2. The sample was then diluted with 55/tl of a solution containing 1% N-octyl-fl-D-glucoside, 5 mM EDTA, 5 mM NEM, 1 mM PMSF, 0.1 mM pepstatin A, and 10 mM sodium phosphate, pH 7.2. O-glycanase (0.5 mU, Genzyme) and/or N-glycanase (0.5 U, Genzyme) were added to aliquots of the sample, and these were then incubated overnight at 37°C. Endo-/~-galactosidase (Seikagaku Kogyo) digestion was performed with 1 mU of the enzyme in 30/~1 of a solution containing 5 mM EDTA, 5 m M NEM, 1 mM PMSF, 0.1 mM pepstatin A, and 50 mM sodium acetate, pH 5.8 after chondroitinase ABC digestion and ethanol precipitation.

lmmunoblotting The tissue homogenates and PBS-extracts for immunoblotting were prepared as described previously? 4'28 Electrophoresis through a 5 or 4% polyacrylamide gel was carried out by the method reported by Laemmli, ~8 and agarose PAGE was performed as described previously) 2 After electrophoresis, the proteins were transferred to an Immobilon-P membrane (Millipore) according to Towbin et al. 33 The membrane was first blocked with 3% gelatin in PBS for 60 min, and then incubated in the antibody solution for 60 min at room temperature followed by processing with a Vectastain ABC kit according to the manufacturer's protocol (Vector Labs). The membrane was then treated with 0.05% 4-chloronaphthol/0.0125% hydrogen peroxide/PBS. The following first antibodies were used; culture supernatants of hybridomas 6B4~4 and IG2, 28 anti-keratan sulfate monoclonal antibody 5-D-4 (Seikagaku Kogyo, Japan), and HNK-1 monoelonal antibody (Serotec). lmmunohistochemistry After ether anesthesia, Sprague-Dawley rats were perfused with phosphate-buffered saline (PBS), followed by a

6B4 Chondroitin sulfate proteoglycan from rat brain solution containing 4% paraformaldehyde and 0.1 M sodium phosphate buffer, pH 7.4, via the left ventricle and washed out from the right atrium. The brains were then dissected out and embedded in paraffin after dehydration through a graded alcohol series. Paraffin-embedded samples were cut into 6-/~m-thick sections, which were then deparaffinized and equilibrated in PBS. The sections were sequentially incubated in the following solutions: (i) 2.5% H2OjPBS for 10min; (ii) I% bovine serum albumin/4% goat serum/PBS for 60min; (iii) culture supernatant containing MAb 6B4 for 24 h at 4°C; (iv) biotinylated anti-mouse IgM solution for 60min; (v) avidin biotin-peroxidase complex (ABC) solution for 30 min; and (vi) 0.1% diaminobenzidine/0.02% hydrogen peroxide/PBS. A Vectastain ABC kit (Vector Labs) was used according to the supplier's protocol.

Dissociated cerebral cell culture Cerebra were dissected from embryonic day-16 (El6) Sprague Dawley rats, and the meninges were removed. The tissues were incubated in Ca 2+- and Mg:+-free Hanks' balanced salt solution (CMF-HBSS) containing 0.1% trypsin for 15min at 37°C. After three washings with CMF-HBSS, the tissues were triturated with Pasteur pipettes in CMF-HBSS containing 0.025% DNAase I, 0.4 mg/ml soy bean trypsin inhibitor, 3 mg/ml bovine serum albumin and 12mM MgSO 4. The cell suspension was centrifuged at 160 x g for 5 min at 4°C, and the pelleted cells were washed once with CMF-HBSS. The cells were resuspended at a concentration of 1.0 × 105 cells/ml in culture medium consisting of a 1:1 mixture of Dulbecco's modified Eagle's medium and F12 medium containing 2% B27 supplement (GibcoBRL). Twenty microliters of the cell suspension was plated on a glass coverslip (9 mm in diameter) coated with 0.002% poly-L-tysine (Mr > 300,000) in a 48-well tissue culture plate. After 2 h incubation in a CO2 incubator, 200pl of culture medium was added per well. Under these culture conditions, more than 98% of the cells were identified as neurons by immunohistochemical staining with anti-MAP2 and anti-neurofilament antibodies. For the substrate assay, poly-L-lysine- and fibronectin-coated coverslips were treated overnight with the purified 6B4 proteoglycan (0 ~ 50 #g/ml

1

2

3

(A) 4 5

6

7

25

as protein) or rat chondrosarcoma proteoglycan monomer (0 ~ 10 #g/ml as protein, ICN Biomedicals Inc,) solutions at 37°C. The coverslips were washed with CMF-HBSS and used for plating.

Other methods Protein concentration was determined using a Micro BCA kit (Pierce) employing bovine serum albumin as a standard. Hexuronate concentration was determined as described by Bitter and Muir) RESULTS

Tissue distribution o f 6B4 proteoglycan W e have previously reported t h a t 6B4 proteoglycan shows m a r k e d changes in expression during cerebellar development. 24 The tissue distribution o f this proteoglycan was e x a m i n e d further in this study by i m m u n o b l o t t i n g (Fig. 1). 6B4 proteoglycan was detected in the cerebrum, cerebellum, b r a i n stem a n d spinal cord, b u t not in n o n - n e r v o u s tissues. In the nervous tissue, this proteoglycan was p r e d o m i n a n t l y enriched in the cerebrum. Agarose P A G E analysis o f the tissue h o m o g e n a t e s indicated t h a t the molecular mass o f the intact form o f 6B4 proteoglycan was 500-1000 x 103 mol. wt (Fig. 1A). After chondroitinase A B C digestion of the tissue h o m o g e n a t e s , the molecular mass of the antigen was reduced to 300 × 103tool. wt (Fig. 1B).

Isolation o f 6B4 proteoglycan W h e n 20-day-old rat brain was homogenized in PBS, a b o u t two thirds o f 6B4 proteoglycan was solubilized a n d recovered in the supernatant. The sample was next fractionated t h r o u g h D E A E - T o y opearl c o l u m n in the presence of 7 M urea. The eluted

8

]

2

3

(B) 4 5

6

200='716-358,-'200-

116"--

116,--

97--

Fig. 1. Tissue distribution of 6B4 proteoglycan. Tissue homogenates (50 #g of total protein) from adult rats were applied to the agarose PAGE system and analysed by immunoblotting using MAb 6B4 (A). Tissue homogenates (20 #g of total protein) from adult rats were digested with chondroitinase ABC and analysed by 5% SDS-PAGE and immunoblotting using MAb 6B4 (B). Samples were from cerebrum (I), cerebellum (2), brain stem (3), spinal cord (4), liver (5), spleen (6), kidney (7), and lung (8). The positions of molecular weight markers (given as Mr x 10 -3) are shown at left. Markers used were phosphorylase b, fl-galactosidase, myosin, and dimer and tetramer of ~:-macroglobulin from bottom to top, respectively.

7

26

N. Maeda et al.

materials were further fractionated by CsCI equilibrium density gradient centrifugation in the presence of 4 M guanidine-HC1 (Fig. 2A). 6B4

III M

(A)

1 2

proteoglycan sedimented at a peak density of 1.47 g/ml as shown by immunoblotting with M A b 6B4 (Fig. 2A, inset). The fractions indicated in

++

. . . . ++

3 4

5 6 7 8

9

5OO I

I1

. 400

~(..

O0

•

1.5-E

300 1.4

200 i50

.~"

10f] 1.3

0

0

I

I

1

2

I

I

I

I

I

l

I

3

4

5

6

7

8

9

Fraction

Number

(B) 1

5 7 8

9

3 4

5

6

7 8 9

I

I

I

I

2 3 4

0.04

~'002 QO " II) C= II O 14

O-- I

1 I

I

I

20

I

30 Time

I

40

I

I

I ,

50

min

Fig. 2. CsC1 density gradient centrifugation and TSKgel G6000PW column chromatography of 6B4 proteoglycan. The 6B4 proteoglycan fraction eluted from a DEAE-Toyopearl column was applied to CsCI density gradient centrifugation (A), and aliquots of each resultant fraction were analysed by immunoblotring using MAb 6B4 after chondroitinase ABC digestion (A, inset). Protein (Q) and hexuronate (O) concentrations, and the density ( × ) of each fraction were measured. The fractions indicated by the horizontal bar in (A) were pooled and next applied to a TSKgel G6000PW column (B). Aliquots of each eluted fraction were then analysed by immunoblotting using MAb 6B4 after chondroitinase ABC digestion (B, inset).

6B4 Chondroitin sulfate proteoglycan from rat brain

0.04

II

U

27

~,,

1 2 3 4 5 6

789

I

.-0°8', s SSS

==

s

0.02

s SS

|

ss

C

== 00

I

I

=E -0.6

Q

SS

L~

m

Sj

t

,,~

Z

ss S

N

J

•=

1 2 J ,

0

9 ,

I

j

i

0

10

/

20 Time

0.4

, I

30 min

40

Fig. 3. LiChrospher 4000 DMAE column chromatography of 6B4 proteoglycan. The fractions indicated by the horizontal bar in Fig. 2B were applied to a LiChrospher 4000DMAE column and eluted by linear NaC1 gradient (. . . . ). Aliquots of each fraction were analysed by immunoblotting using MAb 6B4 after chondroitinase ABC digestion (inset). Fig. 2A were further fractionated through TSKgel G6000PW gel filtration column (Fig. 2B) and LiChrospher 4000DMAE column (Fig. 3), and the elution profile of 6B4 proteoglycan was traced by immunoblotting (Figs 2, 3, insets). 6B4 proteoglycan was eluted from the LiChrospher 4000DMAE column as a single peak. The PBS extract of 20-day-old rat brain contained several kinds of chondroitin sulfate proteoglycans with discrete core proteins, of which the recently identified neurocan is a representative.3° MAb 1G2 recognizes the 220 and 150 × 103mol. wt core proteins of neurocan. 28 The crude proteoglycan fraction, obtained after CsC1 density gradient centrifugation, contained neurocan. However, after HPLC on LiChrospher 4000DMAE, the 6B4 proteoglycan fraction no longer reacted with MAb 1G2 on immunoblotting (Fig. 4C). When the final fraction was treated with chondroitinase ABC, a single 300 × 103 mol. wt proteoglycan core protein was observed (Fig. 4A). About 100/~g as protein and 300nmol as hexuronate of 6B4 proteoglycan was obtained from 10 g of tissue.

Characterization of 6B4 proteoglycan Immunoblotting of the chondroitinase ABC-digested 6B4 proteoglycan showed a broad dispersed band when compared to the CBB staining (Figs 4B, 5A). Neuraminidase treatment of the chondroitinase ABC-digested 6B4 proteoglycan resulted in an increase in the core protein mobility upon SDS-PAGE

(Fig. 5B). Subsequent keratanase digestion substantially reduced the diffuseness of the core protein band (Figs 5C, 6, lane 2). Additional O-glycanase and N-glycanase digestion reduced the molecular weight

A 4-

B --

+

C --

4-

--

200"--

116,-97-'-

66"-Fig. 4. SDS-PAGE of the purified 6B4 proteoglycan. The purified 6B4 proteoglycan (1 #g of protein), before ( - ) or after (+) chondroitinase ABC digestion, was applied to 5% SDS-PAGE, and stained with Coomassie Brilliant Blue (A), or analysed by immunoblotting using MAb 6B4 (B) and MAb IG2 (C).

N. Maeda et al.

28

A

B

C

D

E

F

200="--

Fig. 5. Glycosidase digestion of 6B4 proteoglycan. Purified 6B4 proteoglycan was digested sequentially with chondroitinase ABC (A), neuraminidase (B), keratanase (C), O-glycanase (D) or N-glycanase (E), and both O-glycanase and N-glycanase (F). The digested samples (0.1ttg of protein) were applied to 4% SDS-PAGE and analysed by immunoblotting using MAb 6B4. The position of a molecular weight marker (given as Mr × 10- 3) is shown at left. The arrows indicate the mobilities of 300 and 280 × 103mol. wt species.

1

2

3

of the core protein only slightly (Fig. 5D, E, F). The molecular size of the core protein decreased from 300 to 280 × 103 mol. wt after the glycosidase treatments. Heparitinase digestion of the core protein did not affect its electrophoretic mobility (data not shown). A monoclonal anti-keratan sulfate antibody, 5-D-4, reacted with the chondroitinase ABC-treated 6B4 proteoglycan (Fig. 6C, lane 1). After keratanase or endo-/~-galactosidase digestion, this reactivity disappeared (Fig. 6C, lanes 2 and 3), indicating that 6B4 proteoglycan was substituted with keratan sulfate. The core protein of 6B4 proteoglycan is highly substituted with HNK-1 epitopes (Fig. 6D), and the HNK-1 epitope was not removed from the core protein by digestion with keratanase, endo-~-galactosidase (Fig. 6D), O-glycanase, or N-glycanase (data not shown). To characterize the chondroitin sulfate moiety of 6B4 proteoglycan, purified 6B4 proteoglycan was treated with alkali and digested with pronase. The sample was then chromatographed on a Sepharose CL-6B column, and the elution pattern of chondroitin sulfate was traced by quantifying hexuronic acid (data not shown). Chondroitin sulfate was eluted as a single peak with Kav = 0.48, and its estimated size was 21 × 103mol. wt (Table 1). Analysis of chondroitinase ABC digestion of the sample indicated that 4-sulfated disaccharide unit accounted for 93% and 6-sulfated unit for 6% (Table 1).

Developmental expression o f 6B4 proteoglycan in the cerebral cortex

A

B

C

The developmental expression pattern of 6B4 proteoglycan was examined in the rat cerebral cortex. 6B4 proteoglycan was first detected in the PBS extract of the El4 rat cerebrum by immunoblotting (Fig. 7). The level of 6B4 proteoglycan expression increased until postnatal day 0 (P0), and then gradually decreased until P20. Relatively small amounts of 6B4 proteoglycan persisted in the adult cortex. Essentially the same results were obtained when tissue homogenates of the cerebrum were examined instead of PBS extracts (data not shown). Immunohistochemistry of the developing rat cerebral cortex indicated dynamic changes in the expression profile of 6B4 proteoglycan. The expression of 6B4 proteoglycan began with the initiation of differentiation of cortical layers. At El4, faint staining was firstly observed at the marginal zone. At El6,

D Table I. Properties of 6B4 proteoglycan Fig. 6. Presence of keratan sulfate and HNK-1 epitopes on 6B4 proteoglycan. Purified 6B4 proteoglycan was digested with chondroitinase ABC (1), and further treated with keratanase (2) or endo-fl-galactosidase (3). The samples were applied to 4% SDS-PAGE, and the proteins were stained with Coomassie Brilliant Blue (A), or analysed by immunoblotting using monoclonal antibodies, 6B4 (B), 5-D-4 (C), or HNK-1 (D).

Average molecular size Core protein size Average size of chondroitin sulfate Composition of chondroitin sulfate GlcA-GalNAc (4-SO4) GIcA-GalNAc (6-SO4) GIcA-GalNAc Keratan sulfate HNK-1 epitope

800 × l 0 3 mol. wt 300 × 103mol. wt 21 × 10 3 mol. wt 92.5% 5.6% 1.9% + +

6B4 Chondroitin sulfate proteoglycan from rat brain

ABC

DEFGH

-" ~

Wg

I J

K

O0 am m ~

29

Positive neurons were exceptionally rare in the agranular insular cortex, the prepiriform cortex, and the entorhinal area (data not shown). In all stages, no significant staining was observed when control IgM (Sigma, MOPC104E) was used instead of MAb 6B4.

C P,'.~.,~

Fig. 7. Developmental expression of 6B4 proteoglycan in the rat cerebrum. PBS extracts (20yg of total protein) were prepared from the cerebra of rats at various ages, digested with chondroitinase ABC, and then analysed by immunoblotting using MAb 6B4. The samples were from El2 (A), El4 (B), El6 (C), El8 (D), E20 (E), P0 (F), P3 (G), P5 (H), P7 (I), P10 (J), and P20 (K) animals•

VZ

MZ6B4 proteoglycan was detected in all layers of the developing cortex, although the staining of the ventricular zone was less intense (Fig. 8A). At E20, the marginal zone was stained densely (Fig. 8B). From the intermediate zone to the cortical plate, the radial glial fibers were stained intensely (Fig. 9A, B). In addition to the radial glial fibers, the migrating neurons were also stained in the intermediate zone and in the subplate (Fig. 9B). When cerebral cells from El6 rats were cultured and stained immunohistochemically, the neurites and the cell bodies of the neurons were strongly stained with MAb 6B4 (Fig. 9C). In contrast, neurons in the cortical plate were not stained (Fig. 9A). At P0, when the migration of neuroblasts in the intermediate zone almost ceases, the staining in the intermediate zone disappeared, while the staining of the cortical plate and the subplate remained (Fig. 8C). In the early postnatal rat brain, the cortex was uniformly stained. At P7, the white matter began to show positive staining, and until P20, a similar staining pattern was observed (Fig. 10A). After P30, the staining of the neuropil in the cortex and the white matter became weak, and the staining of the surroundings of subsets of neurons in the cortex became evident (Fig. 10B). The stained neurons were mostly multipolar non-pyramidal cells ( > 95%), and their cell bodies and dendrites were outlined with immunopositive reaction products (Fig. 10C). There was a regional difference in the density of the MAb 6B4-positive neurons. Immunopositive neurons were distributed most densely in layers IV and VI of the primary somatosensory cortex (Fig. 10B). In the primary visual cortex, the MAb 6B4-positive neurons were present mainly in layers II IV (data not shown). NSC 67,'1--B

CP

SP IZ

MZ CP

CX SP

IZ~ VZ

C

Fig. 8. lmmunohistochemical analysis of the developing rat cerebrum. Sagittal sections from cerebra of rats at various ages were stained with hematoxylin/eosin (left) or MAb 6B4 (right). The samples were from El6 (A), E20 (B), and P0 (C) animals• MZ, marginal zone; CP, cortical plate; CX, cortex; SP, subplate; IZ intermediate zone; VZ, ventricular zone. Scale bars=A, 40,urn; B, 100pm; C, 200ym.

N. Maeda et al.

30

L'

i

! / y

Fig. 9. Immunohistochemical analysis of the cortex and cortical neurons. Higher magnification figures of the cortical plate (A) and the intermediate zone (B) from E20 rat cerebrum, and dissociated ceils (C) from El6 rat cerebrum are shown. Only the radial glial fibers (arrows) were stained in (A) but the migrating neurons (arrows) were also stained in (B) with MAb 6B4. In (C), dissociated cells from El6 rat cerebrum were cultured for 45 h, and then stained with MAb 6B4. The arrowhead indicates the cell body 0fa neuron• Scale b a r s = A , 30/~m; B, 15/~m; C, 10l~m.

Effects o f 6B4 proteoglycan on dissociated cortical neurons Strong stimulation of neurite o u t g r o w t h was observed w h e n n e u r o n s were cultured on poly-L-lysinecoated coverslips subsequently treated with 6B4 p r o t e o g l y c a n (Fig. 11). The effect was a p p a r e n t even after 1 7 h culture. In the control culture, the cell

B

bodies were well-spread with m a n y s h o r t processes (Fig. 1 IA). In contrast, the n e u r o n s extended one or a few long processes o n the 6B4 proteoglycan-coated coverslips (Fig. 11C). The same effects were observed when c h o n d r o i t i n a s e ABC-digested 6B4 proteoglycan was coated on the coverslips (Fig. l i D ) . The stimulative effects o n neurite extension were a p p a r e n t over

1!!¸¸¸

R

i L

~:5 : ~ ~¸

Fig. 10. Immunohistochemical analysis of postnatal rat cerebrum. Frontal sections from P7 (A) and adult (B and C) rat cerebra were stained immunohistochemically with MAb 6B4. In the P7 rat cerebra, the neuropil in the cortex (CX) and the white matter (WM) were stained with MAb 6B4. In the adult cerebra, surroundings of subsets of neurons were stained. In the primary somatosensory cortex, stained neurons were distributed mainly in layers IV and VI (B). The numbers in (B) indicate the positions of each layer. Scale bars = A, 250 #m; B, 200 #m; C, 20 ~m.

6B4 Chondroitin sulfate proteoglycan from rat brain

31

Fig. 11. Effects of 6B4 proteoglycan on dissociated cortical neurons. Dissociated cells from El6 rat cerebrum were cultured on poly-L-lysine-coatedcoverslips (A) or on those treated with 10/~g/ml rat chondrosarcoma proteoglycan (B), 10/~g/ml6B4 proteoglycan (C), and 10/~g/mlchondroitinase ABC-digested 6B4 proteoglycan (D). Phase-contrast micrographs were taken after 17 h of incubation. The average length of the longest neurites from 100 neurons in total under each condition was 20 + 2/zm (A), 11 + 1/~m (B), 58 _+3/tm (C) and 62 + 10 #m (D) (the mean value + S.E.M. of triplicate determinations). Scale bar = 100pm.

a wide range of concentrations of 6B4 proteoglycan (0.4-50/tg/ml), and saturated at a concentration of l0/lg/ml (ECs0= ~ 2/~ g/ml). We compared the average length of the longest neurites from 100 neurons. The value for the coverslips coated with 6B4 proteoglycan was three-fold higher than that of the control ( ~ 6 0 # m versus 20~m; see Fig. 11 'legend). As another control, rat chondrosarcoma proteoglycan was used to coat the poly-t-lysine-coated coverslips. In this case, neurons extended only short processes (Fig. 11B). After 45 h culture, neurites of the neurons on the control coverslips (poly-L-lysine only) extended subsequently upto 6 2 + 12/~m, while the average length of the neurites on the 6B4 proteoglycan-coated coverslips (100 + 1/~m at 10 #g/ml) was still ~40 ~tm longer than the control. Essentially the same results were obtained after 45h of incubation, when 6B4 proteoglycan (0.4-50#g/ml) was coated on the coverslips pre-treated with fibronectin (20/~g/ml) instead of

poly-L-lysine. The average lengths of the neurites on the 6B4 proteoglycan (10 #g/ml) coated and control (fibronectin only) coverslips were 78 p m and 42 #m, respectively. The proteoglycans did not affect cell viability in the concentration range tested. Detailed analysis of the effects of 6B4 proteoglycan on the cortical neurons will be reported elsewhere. DISCUSSION

Structure of 6B4 proteoglycan In this study, we purified 6B4 proteoglycan under dissociative conditions; DEAE-Toyopearl column chromatography in 7 M urea and CsCI equilibrium density gradient centrifugation in 4 M guanidineHCI. This is essential for the purification of proteoglycans, since they bind various extracellular matrix molecules and growth factors. Tenascin, for instance, was reported to be copurified with some brain proteoglycans under associative conditions. ~2 Accordingly,

32

N. Maeda et al.

establishment of the dissociative purification procedure, as described here, is a prerequisite for in vitro functional assays of individual proteoglycans. Purified 6B4 proteoglycan shows an average molecular weight of about 800,000 in intact form. After chondroitinase ABC digestion, however, the molecular weight of the proteoglycan decreased to 300,000. Based on the average molecular weight of chondroitin sulfate (21,000), about 24 such chains are estimated to be attached to the core protein. However, it should be noted that these values were calculated from the data of the purified 6B4 proteoglycan. At the purification steps of CsC1 density gradient centrifugation and TSKgel G6000PW chromatography, 6B4 proteoglycan containing lesser amounts of chondroitin sulfate was lost. Western blot analysi~ of the tissue homogenates of the cerebrum indicated that small amounts of 6B4 proteoglycan showed an electrophoretic mobility equivalent to that of a 300 x 103 mol. wt species without chondroitinase digestion, suggesting that non-proteoglycan-type 6B4 antigen is also present. Phosphacan was originally reported as two independent proteoglycans, 3H1 proteoglycan and 3F8 proteoglycan, which are distinguishable from each other by monoclonal antibodiesfl9 3HI proteoglycan has a core glycoprotein of ~ 360 x 103 mol. wt in P7 rat brains but its size decreases to ~280 × 103 mol. wt in the adult brainfl 9 On the other hand, the molecular size of the core glycoprotein of 3F8 proteoglycan is constantly 400 x 103mol. wt in P7 and adult rat brain. 29 3H1 proteoglycan carries keratan sulfate, but 3F8 proteoglycan does not. From the amino acid sequence analysis data of the two proteoglycans and cDNA cloning, it was concluded that the core proteins of these two molecules are identical and correspond to an extracellular variant of the receptor-like tyrosine phosphatase PTP~ (RPTPfl). 26 Although c D N A cloning indicated that 6B4 proteoglycan is identical with phosphacan] 5 there are several discrepancies between the two proteoglycans. Immunohistochemical analysis of adult rat cerebellum indicated that 6B4 proteoglycan was distributed around Purkinje cells and Golgi cells,2~whereas phosphacan was not localized around specific types of neurons. 12 6B4 proteoglycan was synthesized mainly by neurons in the dissociation culture of the cortex, 2s but phosphacan was reported to be synthesized by glia. 26 The molecular size of the core glycoprotein of 6B4 proteoglycan did not change from El4 to adulthood (Fig. 7). These discrepancies likely reflect differences in the post-translational modification and the multiple molecular forms derived from the alternative splicing of this proteoglycan, and the antibodies used in these studies seem to recognize only a subpopulation. Consistent with this idea is the high complexity of the carbohydrate moiety of this proteoglycan, which is regulated spatiotemporally in the developmental stages. Anyway, further studies will be necessary to come to a conclusion.

Characterization o f carbohydrate moieties

6B4 proteoglycan bears keratan sulfate, making the molecular size of the core glycoprotein highly diverse. The degree of diversity changes during postnatal development of the rat cerebellum. 24 Recent analysis has indicated that the process of addition of keratan sulfate chains to 6B4 proteoglycan is strictly regulated spatiotemporally during the development of the brain (H. Hamanaka, unpublished observation). Keratan sulfate is considered to be involved in the regulation of neurite extension 5'9 and possibly in cell adhesion. 6 Thus, substitution of 6B4 proteoglycan with keratan sulfate may also be an important step for the acquisition of specific functions. 6B4 proteoglycan is also substituted with HNK-1 epitopes, which are considered to be involved in cell adhesion. HNK-1 epitopes on the 6B4 proteoglycan molecule were not removed by treatment with O-glycanase, N-glycanase, endo-~-galactosidase or keratanase. Gowda et al. 1~ reported that a portion of HNK-I epitopes in the soluble chondroitin sulfate proteoglycan fraction of the brain are present on poly(N-acetyllactosaminyl) oligosaccharides and are removed by digestion with endo-~-galactosidase. Our result is consistent with the report by Rauch et al. 29 in that HNK-1 epitopes on the 3H1 proteoglycan were not removed by endo-/~-galactosidase treatment. Several other brain chondroitin sulfate proteoglycans were reported to carry keratan sulfate and HNK-1 epitopes. Astrochondrin carries HNK-1 carbohydrates and is involved in cerebellar granule cell migration. 34 Somataglycan-S carries keratan sulfate and HNK-1 epitopes, and is expressed on the surface of neurons constituting the spinocerebellar system, 36 but Purkinje cells do not express this proteoglycan in contrast to 6B4 proteoglycan. 24 Krueger et all 7 also reported the presence of keratan sulfatecontaining chondroitin sulfate proteoglycan from chick brain. Some of these proteoglycans with keratan sulfate and/or HNK-1 carbohydrates might be members of a family of proteoglycans with similar core proteins. Expression and possible functions o f 6B 4 proteoglycan

Immunohistochemical and western blot analysis of the rat cerebrum with MAb 6B4 indicated that 6B4 proteoglycan expression changes dynamically during development. M A b 6B4 can also detect PTP~, a splicing variant of 6B4 proteoglycan, however, the total amount of PTP~ is less than one tenth of 6B4 proteoglycan in the brain. 25Accordingly, it is possible to say roughly that the immunostaining with MAb 6B4 is derived from 6B4 proteoglycan. The expression pattern of 6B4 proteoglycan seems to be divided into three stages in the process of cerebral development. In the first stage ( E I ~ P 0 ) , positive immunostaining with MAb 6B4 was observed along the radial glial

6B4 Chondroitin sulfate proteoglycan from rat brain

33

fibers and on migrating neurons. At E20, the radial sulfate. 2'4'7 6B4 proteoglycan may also be a comglial fibers were intensely stained all along their ponent of the perineural nets and involved in the length, and migrating neurons in the intermediate maintenance of specific types of neurons. zone and the subplate neurons were also stained with MAb 6B4. However, the neurons in the cortical plate Possible cellular sources o f 6B4 proteoglycan were not stained. During the first stage, neuroblasts Previously, we demonstrated that 6B4 proteoglyare actively generated at the ventricular zone, migrate can was secreted more actively by cultured cortical along the radial glial fibers, and settle at the cortical neurons than by astrocytes. 28 In the present study, plate in an inside-out order. The expression profile of immunohistochemical analysis of the cerebral cortex 6B4 proteoglycan strongly suggests that this molecule of rat embryos indicated that 6B4 proteoglycan was plays important roles in neuronal migration and in present on the radial glial fibers and migrating neurcortical layer formation. This is consistent with the ons. In dissociation culture of El6 rat cerebrum, the observation that staining of the radial glial fibers in cell bodies and neurites of neurons were strongly the intermediate zone disappeared at P0, when the stained with MAb 6B4. It is also conceivable that 6B4 migration of neurons ceases in this zone. proteoglycans are secreted by migrating neurons and In the second stage (P5-20), positive immunoreac- associated with the surroundings of the radial glial tivity with M A b 6B4 was observed throughout the fibers. However, these observations do not exclude cortex, except in layer IV of the primary somatosen- the possibility that 6B4 proteoglycan is also synsory cortex. From P5 to Pl6, staining in this region thesized by radial glia. was relatively weak compared to that in other regions. Several extracellular matrix proteins, such as CONCLUSIONS neurocan and cytotactin/tenascin, have been reported to be distributed in a barrel pattern in layer IV of the In the hindbrain, 6B4 proteoglycan is highly exprimar~ somatosensory cortex. ~°'28 The distribution pressed on the cerebellar Purkinje cells and Golgi of 6B4 proteoglycan, however, does not seem to be in cells, and at particular nuclei including the pontine a barrel pattern. nuclei and lateral reticular nucleus, which are related During the second stage, growth of neurites and to the mossy fiber system. 24 These expression stages synapse formation actively proceed in the cortex, and correspond with the onset of their synapse formation, 6B4 proteoglycan may be involved in this process. In which suggests that 6B4 proteoglycan is closely inthis context, it should be noted that 6B4 proteoglycan volved in the development of the cerebellar mossy promoted neurite extension of cortical neurons (Fig. fiber system. In the forebrain, the spatiotemporal 11). Iijima et al. ~4 indicated that soluble chondroitin expression pattern of 6B4 proteoglycan suggests that sulfate proteoglycan mixture from 10-day-old rat it plays important roles in the neuronal migration and whole brain promoted neurite extension. 6B4 proteo- differentiation in the developing cortex and the mainglycan seems to be partly responsible for this effect. tenance of subsets of neurons in the mature cortex. It The growth-promoting effects of 6B4 proteoglycan is known that proteoglycans can interact with various are in contrast to the effect of rat chondrosarcoma molecules including extracellular matrix components, proteoglycan. Rat chondrosarcoma proteoglycan growth factors and cell adhesion molecules. In view rather depressed the neurite extension of cortical of the highly specific expression pattern of 6B4 neurons (Fig. l 1). This suggests that core protein proteoglycan/phosphacan and the presence of the structures are important for these functions of promultiple molecular forms in the CNS, it is conceivteoglycans in brain development. able that this proteoglycan bears multiple functions, After P30 (the third stage), staining of the neuropil which are determined by the combination of the became weak, and strong positive reactions remained interacting molecules extra- and intracellularly~ in the around a subset of neurons. Some neurons in the individual developmental stages. mature cerebrum are wrapped with a reticular wrapping called 'perineural nets', which has been Acknowledgements--We thank Miss Akiko Kodama and suggested to be involved in neuron-glia interactions. 7 Mrs Naoko Oohashi for secretarial assistance. This work This structure is rich in extracellular matrix com- was partly supported by grants from the Ministry of Eduponents such as cytotactin/tenascin and chondroitin cation, Science and Culture of Japan.

REFERENCES

1. Barnea G., Grumet M., Milev, P., Silvennoinen O., Levy J. B., Sap J. and Schlessinger J. (1994) Receptor tyrosine phosphatase/~ is expressed in the form of proteoglycan and binds to the extracellular matrix protein tenascin. J. biol. Chem. 269, 14349-14352. 2. Bertolotto A., Rocca G. and Schiffer D. (1990) Chondroitin 4-sulfate proteoglycan forms an extracellular network in human and rat nervous system. J. Neurol. Sci. 100, 113 123. 3. Bitter T. and Muir H. E. (1962) Modified uronic acid carbazole reaction. Analyt. Biochem. 4, 330-334.

34

N. Maeda et al.

4. Bruckner G., Brauer K., Hartig W., Wolff J., Rickmann M. J., Derouiche A., Delpech B,, Girard N., Oertel W. H. and Reichenbach A. (1993) Perineural nets provide a polyanionic, glia-associated form of microenvironment around certain neurons in many parts of the rat brain. Glia 8, 183-200. 5. Burg M. A. and Cole G. J. (1994) Claustrin, an antiadhesive neural keratan sulfate proteoglycan, is structurally related to MAP1B. J. Neurobiol. 25, 1 22. 6. Carson D. D., Tang J.-P., Julian J. and Glasser S. R. (1988) Vectorial secretion of proteoglycans by polarized rat uterine epithelial cells. J. Cell Biol. 107, 2425~435. 7. Celio M. R. and Chiquet-Ehrismann R. (1993) 'Perineural nets' around cortical interneurons expressing parvalbumin are rich in tenascin. Neurosci. Lett. 162, 137-140. 8. Chernousov M. A. and Carey D. J. (1993) N-syndecan (syndecan 3) from neonatal rat brain binds basic fibroblast growth factor. J. biol. Chem. 268, 1681(~16814. 9. Cole G. J. and McCabe C. F. (1991) Identification of a deyelopmentally regulated keratan sulfate proteoglycan that inhibits cell adhesion and neurite outgrowth. Neuron 7, 1007-1018. 10. Crossin K. L., Hoffman S., Tan S.-S. and Edelman G. M. (1989) Cytotactiri and its proteoglycan ligand mark structural and functional boundaries in somatosensory cortex of the early postnatal mouse. Devl Biol. 136, 381 392. 11. Gowda D. C., Margolis R. U. and Margolis R. K. (1989) Presence of the HNK-1 epitope on poly(N-acetyllactosaminyl) oligosaccharides and identification of multiple core proteins in the chondroitin sulfate proteoglycans of brain. Biochemistry 28, 4468~4474. 12. Grumet M., Milev P., Sakurai To, Karthikeyan L., Bourdon M., Margolis R. K. and Margolis R. U. (1994) Interactions with tenascin and differential effects on cell adhesion of neurocan and phoSphacan, two major chondroitin sulfate proteoglycans of nervous tissue. J. biol. Chem. 269, 12142 12146. 13. Herndon M. E. and Lander A. D. (1990) A diverse set of developmentally regulated proteoglycans is expressed in the rat central nervous system. Neuron 4, 949-961. 14. Iijima N., Oohira A., Mori T., Kitabatake K. and Kohsaka S. (1991) Core protein of chondroitin sulfate proteoglycan promotes neurite outgrowth from cultured neocortical neurons. J. Neurochem. 56, 706 708. 15. Iwata M., Wight T. N. and Carlson S. S. (1993) A brain extracellular matrix proteoglycans form aggregates with hyaluronan. J. biol. Chem. 268, 15061 15069. 16. Krueger N. X. and Saito H. (1992) A human transmembrane protein-tyrosine-phosphatase, PTP~-, is expressed in brain and has an N-terminal receptor domain homologous to carbonic anhydrases. Proc. natn. Acad. Sci. U.S.A. 89, 7417 7421. 17. Krueger R. C., Hennig A. K. and Schwartz N. B. (1992) Two immunologically and developmentally distinct chondroitin sulfate proteoglycans in embryonic chick brain. J. biol. Chem. 267, 12149 12161. 18. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680~85. 19. Levine J. M. and Card J. P. (1987) Light and electron microscopic localization of a cell surface antigen (NG2) in the rat cerebellum: Association with smooth protoplasmic astrocytes. J. Neurosci. 7, 2711 2720. 20. Levy J. B., Cannol P. D., Silvennoinen O., Barnea G., Morse B., Honegger A. M., Huang J.-T., Cannizzaro L. A., Park S.-H., Druck T., Huebner K., Sap J., Ehrlich M., Musacchio J. M. and Schlessinger J. (1993) The cloning of a receptor-type tyrosine phosphatase expressed in the central nervous system. J. biol. Chem. 268, 10,573-10,581. 21. Margolis R. K. and Margolis R. U. (1993) Nervous tissue proteoglycans. Experientia 49, 429-446. 22. Maeda N., Kawasaki T., Nakade S., Yokota N., Taguchi T., Kasai M. and Mikoshiba K. (1991) Structural and functional characterization of inositol 1,4,5-trisphosphate receptor channel from mouse cerebellum. J. biol. Chem. 266, 1109-1116. 23. Maeda N. and Oohira A. (1991) Proteoglycan and development of the brain: functions of brain chondroitin sulfate proteoglycan. Trends Glycosci. Glycotechnol. 3, 28-34. 24. Maeda N., Matsui F. and Oohira A. (1992) A chondroitin sulfate proteoglycan that is developmentally regulated in the cerebellar mossy fiber system. Devl Biol. 151, 564 574. 25. Maeda N., Hamanaka H., Shintani T., Nishiwaki T. and Noda M. (1994) Multiple receptor-like protein tyrosine phosphatases in the form of chondroitin sulfate proteoglycan. Fedn Eur. biochem. Socs Left. 354, 67 70. 26. Mauel P., Rauch U., Flad M., Margolis R. K. and Margolis R. U. (1994) Phosphacan, a chondroitin sulfate proteoglycan of brain that interact with neurons and neural cell-adhesion molecules, is an extracellular variant of a receptor-type protein tyrosine phosphatase. Proc. natn. Acad. Sci. U.S.A. 91, 2512-2516. 27. Oohira A., Matsui F., Matsuda M., Takida Y. and Kuboki Y. (1988) Occurrence of three distinct molecular species of chondroitin sulfate proteoglycans in the developing rat brain. J. biol. Chem. 263, 10240-10246. 28. Oohira A., Matsui F., Watanabe E., Kushima Y. and Maeda N. (1994) Developmentally regulated expression of a brain specific species of chondroitin sulfate proteoglycan, neurocan, identified with a monoclonal antibody 1G2 in the rat cerebrum. Neuroscience 60, 145 157. 29. Rauch U., Gao P., Janetzko A., Flaccus A., Hilgenberg L., Tekotte H., Margolis R. K. and Margolis R. U. (1991) Isolation and characterization of developmentally regulated chondroitin sulfate and chondroitin/keratan sulfate proteoglycans of brain identified with monoclonal antibodies. J. biol. Chem. 266, 14785 14801. 30. Rauch U., Karthiokeyan L., Maurel P., Margolis R. U. and Margolis R. K. (1992) Cloning and primary structure of neurocan, a developmentally regulated, aggregating chondroitin sulfate proteoglycan of brain. J. biol. Chem. 267, 19536 19547. 31. Snow D. M., Lemmon V., Carrino D. A., Caplan A. I. and Silver J. (1990) Neurite outgrowth on a step gradient of chondroitin sulfate proteoglycan (CS-PG). Expl Neurol. 109, 111 130. 32. Snow D. M. and Letourneau P. C. (1992) Sulfated proteoglycans in astroglial barriers inhibit neurite outgrowth in vitro. J. Neurobiol. 23, 322 336. 33. Towbin H., Stahelin T. and Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gel to nitrocellulose sheets: Procedure and some applications. Proc. hath. Acad. Sci. U.S.A. 76, 4350-4354. 34. Streit A., Nolte C., Rasony T. and Schachner M. (1993) Interaction of astrochondrin with extracellular matrix components and its involvement in astrocyte process formation and cerebellar granule cell migration. J. Cell Biol. 120, 799 814.

6B4 Chondroitin sulfate proteoglycan from rat brain

35

35. Wasteson A. (1971) A method for the determination of the molecular weight and molecular-weight distribution of chondroitin sulfate. J. Chromatogr. 59, 89-97. 36. Williams C., Hinton D. R. and Miller C. A. (1994) Somataglycan-S: a neuronal surface proteoglycan defines the spinocerebellar system. J. Neurochem. 62, 1615-1630. 37. Yoshida K., Miyauchi S., Kikuchi H., Tawada A. and Tokuyasu K. (1989) Analysis of unsaturated disaccharides from glycosaminoglycan by high-performance liquid chromatography. Analyt. Biochem. 177, 327 332. 38. Zaremba S., Guimaraes A., Kalb R. G. and Hockfield S. (1989) Characterization of an activity-dependent, neuronal surface proteoglycan identified with monoclonal antibody Cat-301. Neuron 2, 1207 1219. (Accepted 1 February 1995)