Identification and distribution of heparan sulfate proteoglycans in the white muscle of Atlantic cod (Gadus morhua) and spotted wolffish (Anarhichas minor)

Comparative Biochemistry and Physiology, Part B 143 (2006) 441 – 452 www.elsevier.com/locate/cbpb

Identification and distribution of heparan sulfate proteoglycans in the white muscle of Atlantic cod (Gadus morhua) and spotted wolffish (Anarhichas minor) Monica G. Tingbø a,⁎, Svein O. Kolset b , Ragni Ofstad a , Grethe Enersen a , Kirsten O. Hannesson a a

Norwegian Food Research Institute, Osloveien 1, 1430 ÅS, Norway b Department of Nutrition, University of Oslo, Norway

Received 1 July 2005; received in revised form 17 December 2005; accepted 21 December 2005 Available online 3 February 2006

Abstract Heparan sulfate proteoglycans (HSPGs) were identified in pre-rigor muscle of two species of cold water fish, Atlantic cod (Gadus morhua) and spotted wolffish (Anarhichas minor) by biochemical and immunological methods. The distribution was described by immunohistology. Special emphasis was directed to the extracellular matrix (ECM) HSPGs perlecan and agrin. In vivo 35S-sulfate labeling combined with ultracentrifugation in CsCl2, DEAE chromatography and scintillation counting of the eluates, revealed that the content of 35S-labeled PGs was much higher in wolffish than in cod. A considerable proportion of the 35S-sulfated PGs in both species was HSPG, as judged by nitrous acid degradation. HSPG represented, however, a higher proportion of the 35S-sulfated PGs in cod compared to wolffish. Dot blot and electrophoresis/western blot using two different HS-mAbs, 10E4 and HepSS-1 indicated structural differences in the HS-chains of the PGs present. This observation was strengthened by immunohistochemistry, showing that both mAbs detected epitopes in the pericellular area, but the staining patterns were not superimposable. Two different agrin isoforms were identified in both species. Furthermore, in the white muscle of both cod and wolffish, perlecan mAb (A7L6) showed positive staining restricted to the transition between myocommata and myofibers. © 2006 Elsevier Inc. All rights reserved. Keywords: Heparan sulfate; Fish; Extracellular matrix; Adhesion; Muscle; Perlecan; Agrin

1. Introduction Heparan sulfate proteoglycans (HSPGs) are members of the proteoglycan family characterized by carrying heparan sulfate (HS) side chains. HS-chains consist of alternating hexuronic acid and glucosamine residues and are known to be structurally heterogenous due to variability in sulfation and epimerization along the carbohydrate backbone. The specific structural properties of the HS-chains determine interactions with a variety of extracellular ligands, growth factors and enzymes (for review see Stringer and Gallagher, 1997; Belting, 2003).

⁎ Corresponding author. Tel.: +47 64 97 01 00; fax: +47 64 97 03 33. E-mail address: [email protected] (M.G. Tingbø). 1096-4959/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpb.2005.12.022

The HSPGs are found in all mammalian organs and tissues that invariably produce more than one HSPG species (Kjellén and Lindahl, 1991). They are located abundantly on the cell surface as syndecans and glypicans, and in basement membranes as perlecan, agrin and type XVIII collagen. HSPGs are involved in various biological processes. Examples are cell adhesion, migration and proliferation, tissue differentiation and organization of extracellular matrix (ECM) (for review see Iozzo, 1998; Dunlevy and Hassel, 2000; Tumova et al., 2000; Nakato and Kimata, 2002; Belting, 2003). The ability of cells to degrade HS is a key factor for the basal membrane break down processes involved in cell motility and metastasis (Kure et al., 1987; Nakajima et al., 1988). Despite the progress concerning the important role of HSPGs for muscle structure, growth and differentiation in mammalia (for review see Erickson and Couchman, 2000; Iozzo, 2001;

442

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

Velleman, 2002), very little information is available on HSPGs in fish muscle. The muscle fibers of fish are, in contrast to mammalian muscle fibers, arranged as myotomes in concentric circles, and joined head to tail by connective tissue (myocommata). Each muscle fiber is surrounded by endomysium, as in mammalia. Lack of adhesion and splitting of the ECM (gaping) is a major quality problem for the fish industry, as it leads to poor presentation of the fish fillets due to gaps in the muscle, and consequently lower prices. The HSPGs perlecan (Brandan, 1994; Eggen et al., 1997) and agrin (Ruegg et al., 1992) have been found in the ECM of mammalian skeletal muscle. Perlecan is reported to play a role in cell adhesion and provides strength to the ECM by interactions with numerous extracellular macromolecules including collagen, laminin, nidogen and fibronectin (Isemura et al., 1987; Battaglia et al., 1992; Bai et al., 1994; Belting, 2003; Iozzo, 1994). In the nematode Caenorhabditis elegans, mutations in the unc-52 gene, the orthologue of the mammalian perlecan gene, lead to disruption of sarcomeres and caused detachment of body wall muscle (Rogalski et al., 1993). Furthermore, the core protein of perlecan in the basement membrane is shown to interact with endothelial cells through β1 and β3 integrins, an interaction modulated by HS (Hayashi et al., 1992). Structure function studies of agrin have shown that this HSPG binds to several proteins of the cell membrane and the ECM, such as laminins (Denzer et al., 1997) and αdystroglycan (Gesemann et al., 1998). The amount of agrin in non-synaptic regions changed in models for animal dystrophies (Eusebio et al., 2003). Pfeiler (1991, 1998) and Pfeiler et al. (1991) reported after series of cellulose acetate electrophoresis that different fish larvae (leptocephali) contained HS in addition to variants of chondroitin sulfate, keratan sulfate and hyaluronan (Pfeiler et al., 2002). A 300-kDa HSPG has been shown to be present in goldfish anterior optic tectum (Su and Elam, 2003). Furthermore, we have recently detected HS-chains in cod and wolffish muscle by biochemical methods after digestion of the protein core with papain (Tingbo et al., 2005). Hence, to our knowledge, little has been done to study HSPGs in skeletal muscle ECM of post-metamorphic teleost fish. In the present study, the composition and distribution of HSPGs in pre-rigor muscle of two species of cold water fish, Atlantic cod and spotted wolffish, were investigated by biochemical and histological methods. Atlantic cod and spotted wolffish were selected due to known differences in the propensity to gap (Ofstad et al., in press). A special emphasis was directed on the ECM HSPGs perlecan and agrin.

fish skeletal muscle. 35S-sulfate was injected intraperitoneally (1.5 mCi per kg fish) into 3 fish of each species. The fish were killed after 72 h by a sharp blow to the head, first being anaesthetized with 0.05% benzocaine. The radiolabeling trial was repeated with 3 independent individuals of each species. 2.2. Extraction and fractionation of HSPG Equal amounts (wet weight) of muscle samples from the white muscle beneath the dorsal fin of three individuals of each species were powdered in liquid nitrogen, pooled and incubated over night at 4 °C in 4 M guanidine-HCl in 0.05 M Na-acetate buffer, pH 6. The solid-to-liquid ratio was 10 g tissue to 150 mL extraction buffer. The protease inhibitors 100 mM 6-aminohexanoic acid, 10 mM Na2EDTA, 10 mM N-ethylmalemide and 1 mM phenylmethylsulfonylfluoride were added to the guanidine extraction buffer. The extracts were clarified by centrifugation for 30 min (Beckman centrifuge, JA-14 fixedangle rotor head, 14 000 rpm, 4 °C). The extracts were subjected to density gradient ultracentrifugation after addition of CsCl2 to a starting density of 1.37 g/mL. The centrifugation was carried out at 140 000 g for 48 h at 15 °C in a Beckman centrifuge (Optima L-80, fixedangle VTi50 rotor head) using polyallomer quick seal centrifuge tubes (Opti-seal 25 × 86 mm, Beckman, Fullerton, CA). After ultracentrifugation, the tubes were punctuated in the bottom. 3 fractions, each of 10 mL, were collected from each tube and marked D1, D2 and D3. D1 represented the bottom fraction of the tube with the highest density; D2 was the middle fraction, whereas D3 was the top fraction with the lowest density. The sticky surface layer was discarded. The density of the fractions was determined using a 300 μL pipette as a pycnometer. The different fractions were dialyzed against distilled water and lyophilized. 2.3. Characterization of HSPG

2.1. Radiolabeling and sampling

2.3.1. Radioactivity To follow the recovery of 35S-sulfated HSPG during fractionation, the content of 35S-sulfate was measured after extraction in 4 M guanidine-HCl, and in the D1, D2 and D3 fractions after ultracentrifugation (results not shown). Furthermore, radioactivity was measured in the eluates after DEAE ion-exchange chromatography of D1, D2 and D3 samples, treated or non-treated with nitrous acid (see below). The radioactivity (counts per minute) was measured by scintillation counting in a scintillation counter (WinSpectral 1414 Liquid Scintillation Counter, Wallac, Turku, Finland) after addition of scintillation fluid (Insta-Gel II, Packard Instruments B.V. Chemical Operations, Groningen, Netherlands) in sample-to-fluid ratio of 1 : 2.

Wild caught Atlantic cod (Gadus morhua L.), kept in cages while fed, and farmed spotted wolffish (Anarhichas minor O.) of average mass 1.8 and 1.7 kg, respectively, were used to study the composition of sulfated proteoglycans in

2.3.2. Chemical analyses Before further analyses were performed, the content of protein in the D1, D2 and D3 fractions obtained after ultracentrifugation, was estimated with a Bio-Rad assay (Bio-Rad

2. Materials and methods

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

Laboratories, Hercules, CA) based on the method of Bradford (1976). (Results not shown) 2.3.3. Nitrous acid digestion For identification of HSPG, aliquots of 3 mL from D1, D2 and D3 fractions of cod and wolffish were treated with a mixture of 0.5 M barium nitrite and 0.5 M sulfuric acid (the nitrous acid reagent), according to the method of Shively and Conrad (1976). This method is described to remove HS-chains. The precipitate of the reagent was discarded after a short centrifugation, and the supernatant was immediately added to the sample (1 : 1 v/v). After incubation in room temperature for 10 min, addition of 2 M Na2CO4 to an ultimate pH of 8.5 stopped the reaction. The nitrous acid treated samples for electrophoresis and Western blotting were dialyzed against distilled water and concentrated in a speedvac system (Thermo Savant ISS110) before being applied to the gels. 2.3.4. Ion-exchange chromatography The lyophilized D1, D2 and D3 fractions were resuspended in 6 M urea. The resuspended samples (aliquots of 3 mL) from D1, D2 and D3 after ultracentrifugation, non-treated and treated with nitrous acid, were then analyzed by ion-exchange chromatography. 10 mL PolyPrep columns (Bio-Rad Laboratories, Hercules, CA, USA) were packed with 1 mL DEAE Sephacel ion-exchange medium (Amersham Pharmacia Biotech, Uppsala, Sweden). Before applying the samples, the columns were equilibrated using 10 mL 4 M urea in 0.05 M Tris–HCl, pH 8. The columns were then washed using 10 mL 0.05 M Tris–HCl, pH 8, containing 0.15 M NaCl. The PGs were eluted using a salt gradient ranging from 0.15 to 1.5 M NaCl in 0.05 M Tris–HCl, pH 8. This concentration of NaCl described to favor fractionation of HSPGs in mammalia (Shworak, 2001). The eluate was collected in 2 mL fractions and the content of 35 S-sulfate in each fraction was measured by scintillation counting. The proportion of HSPG was estimated by calculating the difference in radioactive anionic material before and after nitrous acid treatment. 2.3.5. SDS-PAGE Novex Xcell II apparatus and 4–20% Novex Tris–glycine gels were used (Invitrogen Life Technologies, Paisley, UK). Lyophilized samples of the D1 and D2 fractions were resuspended in 0.05 M Tris–HCl, pH 8 and divided in equal aliquots (1 mg sample to 50 μL fluid); one that remained untreated and one that was subjected to nitrous acid digestion. The nitrous acid treated samples were dialyzed and concentrated in a speedvac system. Both untreated and digested samples were added (1 : 3 v / v) sample buffer containing 0.05 M Tris, 10% glycerol, 5% 2-mercaptoethanol, 2% SDS and 0.001% bromphenol blue, pH 8. Samples were boiled for 10 min and 20 μL per well was applied to the gels. Electrophoresis was run in 0.05 M Tris buffer with 0.1% SDS, pH 7.4, for approx. 90 min at 150 V. The gels were soaked in transfer buffer containing 0.025 M Tris–glycine and 20% methanol, pH 8.3, for 20 min, followed by Western blot to nitrocellulose membranes.

443

2.4. Immunological identification of HSPG 2.4.1. Antibodies Monoclonal antibodies against different HS-epitopes (HepSS-1 and 10E4, Seikagaku America, MA, USA) and the HSPG, perlecan (A7L6, Chemicon International LTD, Hofheim, Germany) and agrin (Agr-131 and Agr-86, Calbiochem/ Merck Biosciences, Nottingham, UK) were used. HepSS-1 recognizes an epitope present in the HS-chain involving Nsulfate (Wada et al., 1995; van den Born et al., 2005) and treatment with heparinase or heparitinase resulted in loss of HepSS-1 binding (Kure and Yoshie, 1986; Kure et al., 1987). 10E4 reacts with HS domains containing both N-sulfated and N-acetylated disaccharide units (David et al., 1992; van den Born et al., 2005). The reactivity is completely abolished by heparitinase, but only partially with heparinase (David et al., 1992; Bai et al., 1994). Perlecan clone A7L6 reacts with perlecan core domain IV (Couchman and Ljubimov, 1989). Anti-Agrin-131 recognizes all rat agrin isoforms. The epitope has been mapped near the first EGF-like repeat. Anti-Agrin-86 recognizes epitopes specific for the central nervous system isoforms, and does not recognize agrin lacking these specific amino acid inserts at splicing site Z (Ferns et al., 1993; Hoch et al., 1994). 2.4.2. Western blot Bio-Rad Trans-Blot apparatus and nitrocellulose membranes (Bio-Rad Laboratories) were used for electrophoretic transfer of bands from gel to membrane. The Novex Tris–glycine gels were blotted to the membranes for 1 h at 80 V in chilled 0.025 M Tris–glycine transfer buffer, pH 8.3, containing 20% methanol. Blocking of unspecific binding sites was done using 5% teleost gelatin (Sigma-Aldrich) in 0.1 M Tris–saline with azide, pH 7.4. The monoclonal antibodies HepSS-1 and 10E4 mentioned previously were applied. Alkaline Phosphatase (AP) conjugated secondary antibodies (anti-mouse IgG, (H+L), Promega Corp. MI, USA), combined with BCIP/NBT color development substrate (Promega) were used. Washing the membranes in distilled water stopped the AP/substrate reaction, after which the membranes were air-dried. 2.4.3. Dot blot Aliquots from the D1 and D2 fractions were diluted in 0.1 M Tris–HCl, pH 7.4, and applied to nitrocellulose membranes. Blocking of unspecific binding sites was done using 5% teleost gelatin (Sigma-Aldrich) in 0.1 M Tris–saline with azide, pH 7.4. The monoclonal antibodies HepSS-1, 10E4, A7L6, Agr131 and Agr-86 were used. The same AP-conjugated secondary antibodies and BCIP/NBT color development substrate as in Western blotting were used. Washing the membranes in distilled water stopped the AP/substrate reaction, after which the membranes were air-dried. 2.4.4. Light microscopy For the histology study, samples were collected from the same area of the white muscle beneath the dorsal fin as in the biochemical study. The samples were embedded in O.C.T

444

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

compound (Tissue Tek 4583, Miles Inc. Diagnostic Division, Elkhart, IN, USA) for 30 min and then frozen in liquid nitrogen and stored at − 80 °C. Cross-sections were cut in a cryostat (Leitz 1720 Digital, Leica Instruments GMBH, Heidelberg, Germany) into 5 or 10 μm sections, and mounted on poly-Llysine coated glass slides and kept in − 20 °C. Before further treatment, the sections were dried for 1 h at room temperature. To outline the muscle architecture, hematoxylin (Riedel-de Haen, Seelze, Germany) in combination with erythrosin B (Sigma-Aldrich Chemie, Steinheim, Germany) was used. The dried 10 μm sections were rinsed in hematoxylin for 5–10 min. After washing in running water for 20 min, the sections were fixed in 0.25% hexamine for 3 min and rinsed in running water before staining in erythrosin B for 4 min. Finally, the sections were rinsed in running water, dehydrated and mounted in Eukitt (O. Kindler GmbH and Co., Freiburg, Germany). 2.4.5. Immunohistochemical staining For immunostaining, an immunoperoxidase system Vectastain Universal Elita ABC kit (Vector Laboratories, Inc., Burlingame, CA, USA) was used according to the manufacturer's recommendations. Before immunostaining, the 5 μm cross-sections were fixed in 8% formaldehyde in phosphate buffered saline (PBS, pH 7.4) for 5 min. and washed in PBS (2 × 5 min). To enhance the exposure of the epitopes, the sections were digested for 2 h at 37 °C with chondroitinase ABC lyase (0.5 IU/mL in 0.1 M Tris–HCl, pH 8) from Proteus vulgaris (EC 4.2.2.4, Sigma). The sections were then washed in PBS 2 × 5 min. Unspecific binding sites were blocked using 5% teleost gelatin (Sigma) in PBS added normal serum from horse (Vector Laboratories). The sections were then incubated over night at 4 °C with the following monoclonal HS antibodies;

10E4 (diluted 1 : 200) or HepSS-1 (diluted 1 : 100). For the study of agrin and perlecan, the following mAbs; Agr-131 (diluted 1 : 150), Agr-86 (diluted 1 : 100) and A7L6 (diluted 1 : 100) were used, respectively. The mAbs were diluted in PBS added 5% teleost gelatin and 0.005% Tween-20 (Sigma). After rinsing in PBS with Tween-20 (0.005%), the sections were incubated for 1 h with diluted biotinylated secondary antibodies (horse antimouse, Vectastain Universal Elita ABC kit). After a second rinsing, the sections were incubated in a preformed avidin and biotinylated horseradish peroxidase macromolecular complex (Vectastain Universal Elita ABC kit). Peroxidase activity was revealed by use of a DAB-kit according to the manufacturer's recommendation, based on the method of Graham and Karnovsky (1966). After rinsing, the sections were dehydrated in ascending concentrations of alcohol, cleared in xylol and mounted in Eukitt. A Spot RT Color Camera (Diagnostic Instruments Inc. Burroughs Sterling Heights, Michigan Heights) photographed the sections in a LEICA DMLB microscope (Leica Microsystems Nussloch GmbH, Germany). 2.4.6. Controls Non-specific binding of the primary antibodies was controlled by substitution of the primary antibody with mouse whole serum (X-0910, Dako). Non-specific binding of the secondary antibodies was tested by replacing the latter with dilution buffer (PBS). In addition, some sections were treated with 0.3% H2O2 in dilution buffer for 15 min to block possible endogenous peroxidase activity in the tissue. 3. Results 35

S-labeled PGs were isolated from both species by ultracentrifugation in a CsCl2 gradient and ion-exchange

Fig. 1. Results obtained after DEAE ion-exchange chromatography and scintillation counting of samples from cod and wolffish. Panels (a) and (b) represent 35Ssulfated material from the D1 fraction after ultracentrifugation, of cod and spotted wolffish, respectively. Panels (c) and (d) represent 35S-sulfated material from the D2 fraction after ultracentrifugation, of cod and wolffish. Solid lines represent the untreated samples, whereas dotted lines represent nitrous acid treated fractions. The linear salt gradient is indicated with a thin solid line.

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

Fig. 2. Western blot of samples from wolffish and cod stained using HepSS-1 (left panel) and 10E4 (right panel). Lane 1 in both panels represents nitrous acid treated samples from wolffish, whereas lane 2 in the panels represents untreated samples from wolffish. Lane 3 represents nitrous acid treated samples from cod, whereas lane 4 represents untreated samples from cod. The apparent molecular sizes are indicated to the right.

chromatography. The major proportion of the 35S-labeled PGs (86%) in wolffish and most of the material in cod, were recovered after DEAE ion-exchange chromatography with a NaCl-gradient in the fractions obtained after ultracentrifugation (D1 and D2), which exhibited a buoyant density above 1.36 g/mL (Fig. 1a–d). D3 also contained 35S-labeled material. However, only a small proportion of the 35Slabeled material recovered in D3 from wolffish, and nonmeasurable amounts from cod, showed PG properties by binding to the DEAE column (results not shown). Focus for the present study was therefore on the D1 and the D2 fractions.

The content of 35S-labelled PGs was much higher in spotted wolffish than in cod, in the D1 as well as D2 fraction (Fig. 1a–d). The PG preparation eluted by the NaCl-gradient, showed a broad profile in both species indicating the presence of several 35S-labeled PGs. To identify and estimate the amount of 35S-HSPGs present, equal amounts of samples from D1 and D2 fractions, before and after removal of the HS-chains with HNO2 treatment, were examined by DEAE ion-exchange chromatography. Fig. 1a and b shows the results obtained from 35S-sulfated material in D1, whereas Fig. 1c and d shows the results obtained from 35Ssulfated material in D2. HNO2 treatment changed the elution profiles in both cod and wolffish, but in somewhat different ways. In cod the eluted nitrous acid treated material appeared as one sharp peak that needed a higher salt concentration for elution (Fig. 1a and c). In the wolffish a reduction in 35Ssulfated PGs binding to the column material was observed after HNO2 treatment, but the peak still exhibited a similar profile as before HNO2 treatment (Fig. 1b and d). D1 and D2 material showed similar elution profiles in both species. The results imply that the 35S-PG preparation from the D1 fractions contained 30% HS in cod and 20% HS in wolffish, whereas the D2 fractions contained 25% HS in cod and 8% HS in wolffish, based on reduction of 35S material binding to the columns after nitrous acid treatment. 3.1. Gel-electrophoresis and Western blot To obtain more information on the size of the HSPGs and structure of the HS-chains in the muscles of the two fish species, gel-electrophoresis and Western blot analyses,

HEPSS-1 Cod

445

10E4

Wolffish

Cod

Wolffish

1:1 1:2 1:4 1:8 1:16 1:32 1:64 1:128 1:256

D1

D2

D1

D2

D1

D2

D1

D2

Fig. 3. Results obtained after dot blot studies of the D1 and D2 fractions of cod and wolffish using mAbs against structures in the HS-chains. Left panel shows the membranes after staining using HepSS-1. Right panel shows the results obtained after staining using 10E4. The first and second vertical lines in each panel represent cod (D1 and D2, respectively), the third and fourth vertical lines represent wolffish. Sample dilutions are indicated to the right.

446

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

Agr-131 Cod

Agr-86 Cod

Wolffish

Wolffish 1:1 1:2 1:4 1:8 1:16 1:32 1:64 1:128 1:256

D1

D2

D1

D2

D1

D2

D1

D2

Perlecan Cod

Wolffish

1:1 1:2 1:4 1:8 1:16 1:32 1:64 1:128 1:256

D1

D2

D1

D2

Fig. 4. Results obtained after dot blot studies of the D1 and D2 fractions of cod and wolffish using mAbs against the HSPGs, agrin and perlecan. Upper panels show the results obtained after staining with the mAbs Agr-131 and Agr-86, whereas lower panel shows the results after staining with A7L6 against perlecan. The first and second vertical lines in each panel represent cod (D1 and D2, respectively), the third and fourth vertical lines represent wolffish. The sample dilutions are indicated to the right.

using antibodies directed against structures in the HS part of the PGs, (HepSS-1 and 10E4, see Materials and Methods section), were performed. Variations were observed in the amount of material expressed on the blots from run to run despite similar amounts of material applied. The results illustrated in Fig. 2 were consistent with findings reproduced in several experiments. After immunostaining using HepSS-1 antibodies, the major proportion of the material was apparent in the molecular mass region above 250 kDa (Fig. 2, left panel). HepSS-1 also detected a band with a molecular mass

around 200 kDa. This band was not visible using the other HS-mAb against N-sulfated/N-acetylated disaccharide epitopes. However, in wolffish, the 10E4 mAb detected a band of molecular mass around 100 kDa. Staining using the 10E4 mAb revealed bands of high molecular masses in both the wolffish and cod preparations (Fig. 2, right panel). After HNO2 treatment the bands disappeared (Fig. 2, lanes 1 and 3 in both panels). The Western blots confirmed the presence of several HSPGs with varying, but mostly high molecular masses, in the high-density fractions of both species, and

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

Wolffish

447

Cod

Mc

Mc E

E

a

b

25 µm

25 µm

Fig. 5. HE stained transverse sections from white muscle collected from wolffish (panel a) and cod (panel b). The arrows point to endomysium (E). Myocommata (Mc) is clearly outlined.

indicated structural differences in the HS-chains of the PGs present in the two species. 3.2. Dot blot analysis As some of the HSPG hardly entered the gel as observed by Western blots, the HSPG composition of the D1 and D2 fractions from cod and wolffish were furthermore analyzed by dot blotting. Fig. 3 shows the dot blot membranes of D1 and D2 fractions of both species investigated using the same anti-HS mAbs, HepSS-1 and 10E4, as in Western blot. The HepSS-1 antibody reacted to epitopes present in both D1 and D2 fractions from cod and wolffish, whereas the 10E4 antibody reacted to epitopes mainly present in the D2 fractions, confirming that HepSS-1 and 10E4 epitopes were not evenly distributed among the HSPGs with different buoyant densities separated by ultracentrifugation. In order to identify PGs present carrying the HS-chains, antibodies against the core protein of the basement membrane HSPGs, agrin and perlecan, were furthermore included in the

dot blot experiments. Antibodies directed against two different isoforms of the agrin core protein, Agr-131 and Agr-86, were applied and showed reaction on the blots (Fig. 4). The Agr-131, which reacts with an epitope present in all agrin isoforms, reacted against PGs present mainly in the D2, but a faint reaction could be seen also in the D1 fraction of both species. A similar staining pattern, although weaker, was observed using Agr-86 (neuronal isoforms). Immunostaining with the mAb A7L6 directed against epitopes present in the core protein of perlecan (domain IV), showed a strong staining of molecules in the D2 fractions with a faint stain in the D1 fractions from both cod and wolffish (Fig. 4). A stronger reactivity was evident in the D2 samples from wolffish than cod. 3.3. Tissue distribution To identify the tissue localization of the HS-structures and the basement membrane HSPGs agrin and perlecan, identified by Western and dot blot, analyses using immunohistochemistry and light microscopy were performed.

10E4 Wolffish

Cod

M

Mc

M

a

Mc

50 µm

b

µm 100µ

Fig. 6. Results obtained after staining using the mAb 10E4. Panel (a) illustrates the tissue section from wolffish, whereas panel (b) illustrates the section from cod. The arrows indicate some stained areas in cod and the areas with particularly strong staining in wolffish. Mc = myocommata; M = myofiber.

448

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

HEPSS-1 Wolffish

Cod

Mc Mc

a

50 µm

b

100µm

Fig. 7. Results obtained after staining using the mAb HepSS-1. Panel (a) illustrates the tissue section from wolffish, whereas panel (b) illustrates the section from cod.

In the transverse sections of white muscle from cod and spotted wolffish, stained with hematoxylin–erythrosin B, the general histological architecture of the tissue is demonstrated (Fig. 5a and b). The endomysia (E) that surround the individual myofibers (M) and the myocommata (Mc) are clearly outlined in both wolffish and cod. Immunostaining with the mAb 10E4, against an epitope that includes N-sulfated and N-acetylated glucosamine residues in the HS-chain, showed a strong and continuous reaction in the cell surface/basement membrane area of the endomysia, encircling every individual myofiber (M) in both wolffish and cod (Fig. 6a and b). In the borderline between the myofibers and the myocommata (Mc), the staining was less distinct. Some areas showed a well-defined borderline between myocommata and the myofibers, whereas in other regions the transition was more diffuse. Moreover, the matrix proper of myocommata and endomysia showed a diffuse stain. Using the other mAb against HS-GAG, HepSS-1, a somewhat different staining pattern was observed (Fig. 7a and b). In wolffish, this mAb stained mainly the cell surface area close to the myocommata, and the surface of some specific cells (see arrows Fig. 7a). In cod a more

uniform distribution with similar stain intensity in all the pericellular areas was seen, but with a faint and granular appearance (Fig. 7b). The neuronal isoform of agrin was present in the vessel walls and at the neuromuscular junctions (NMJ) in the myocommata of both cod and wolffish, as judged by the staining pattern after application of the mAb Agr-86 antibody (Fig. 8a and b). Epitopes for the mAb Agr-131, which react with all agrin isoforms, showed a somewhat different distribution. In addition to vessel walls and NMJs, distinct staining was seen on one side of myocommata in wolffish (Fig. 9a). In cod, a weak, granular stain was also observed on the surface of the myofibers (Fig. 9b), but the distinct staining between mycommata and the myofibrillar area was not visible. The results indicated the presence of at least two agrin isoforms in fish muscle, with a somewhat different distribution in wolffish and cod. The results in Fig. 10 showed that also perlecan was a member of the HSPG population in fish muscle, due to the reaction seen using the antibodies A7L6. A strong staining was observed in the transition between myocommata and myofibers, with invaginations into the endomysia of some of the myofibers

Agrin 86 Wolffish

a

Cod

50 µm

b

100µ µm

Fig. 8. Results obtained after staining using the mAb Agr-86 (the neuronal isoform). Panel (a) shows the tissue section from wolffish, whereas panel (b) shows the section from cod. The arrows indicate the strong staining of the vessel walls and NMJs.

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

449

Agrin 131 Wolffish

a

Cod

50 µm

b

100µm

Fig. 9. Results obtained after staining using the mAb Agr-131 (all isoforms). Panel (a) shows the distribution pattern obtained in the tissue section from wolffish, whereas panel b shows the section from cod. The double arrows in panel (a) show the strong staining in the transition to myocommata in wolffish, whereas the single arrows in panel (b) show the endomysial staining in cod.

close to the myocommata. The remaining endomysia showed no stain. Similar localization patterns were observed in both species. 4. Discussion In the intramuscular connective tissue of two species of commercially important cold water teleost fish, Atlantic cod and spotted wolffish, HSPGs carrying HS-chains with different structural epitopes were identified. Furthermore, the presence and distribution of the ECM-HSPGs perlecan and agrin, were demonstrated by immunological methods. Differences were observed between the two species in the composition and structure of the HSPGs. DEAE ion-exchange chromatography showed that a major proportion of the 35S-labeled PGs in cod came from HSPGs (50%) compared to wolffish (33%). Furthermore, the elution profiles of the 35S-sulfated PGs in

cod and wolffish changed in different ways after nitrous acid treatment (Fig. 1a–d). After removal of HS, anionic groups in chondroitin/dermatan (CS/DS) and keratan sulfate (KS) chains of PGs present in the preparation contribute to attachment to the column material. The different elution profiles in cod and wolffish, may reflect that a different population of CS-, DS- and KSPGs is present, with great variations in charge. Another possibility is that HSPGs in wolffish carry CS/DS chains to an extent providing sufficient anionic groups for binding to the DEAE-material, despite removal of the HS-chains. HSPGs, such as perlecan, are described to be hybrid HS/CSPGs in mammalia (Hassell et al., 1993). In a separate study a higher content of 35S-sulfated CS/DS free chains were detected in wolffish than cod muscle after papain digestion (Tingbo et al., 2005). Further studies are at present going on in our laboratory to describe the localization and structure of CS/DS and KS PGs present in the two species.

A7L6 Cod

M c

Wolffish

Mc

a

100µm

b

100µm

Fig. 10. Results obtained after staining using the mAb A7L6 against perlecan core protein domain IV. Panel (a) shows the distribution pattern obtained in the transverse section from wolffish, whereas panel (b) shows the section from cod. The myocommata (Mc) is outlined. The arrows indicate the strong stain in the myocommata borderline. Note the absence of staining pericellularly in the endomysia.

450

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

The differences in staining pattern with the two HS-mAbs 10E4 and HepSS-1, in dot blot and Western blot experiments, showed that the epitopes for these two mAbs were not evenly distributed among the HSPGs present in fish muscle, reflecting structural differences in the HS-chains of the PGs. This observation was strengthened by immunohistochemistry, showing that although both HepSS-1 and 10E4 detected epitopes in the pericellular area, the staining patterns were not superimposable. Using 10E4, a distinct uniform staining was observed lining all the myofibers in a similar manner in both species (Fig. 6). The intensity of the staining reaction obtained using HepSS-1 was weaker and showed marked variations between the two species. In cod the same faint and granular reaction was observed around all the myofibers, whereas in wolffish a very strong regional staining in specific areas near myocommata was evident (see arrows in Fig. 7). The high level of expression of HepSS-1 epitopes at specific sites in wolffish may have consequences for the interactions and thereby adhesion strength in this area, and for growth factor activity and storage (for review see Stringer and Gallagher, 1997). To approach the identity of the proteoglycans carrying the HS-chains, antibodies directed against the basement membrane HSPGs agrin (two isoforms) and perlecan were applied in dot blot analysis and immunohistochemistry. Both the neuronal agrin and alternatively spliced isoforms were present in fish muscle, as judged by the discrepancy in staining patterns observed between the two mAbs Agr-86 and Agr-131. The Agr-131 mAb showed the most widespread distribution in the connective tissue in both species, but with a somewhat different expression pattern. The Agr-86 staining was restricted to NMJs and vessel walls (Figs. 8 and 9). The different expression of Agr-131 in cod and wolffish may be a consequence of a different need for enforcement in the myocommata–myofiber junction, as agrin is reported to bind proteins of the cell membrane and ECM, such as laminins (Denzer et al., 1997) and α-dystroglycan (Gesemann et al., 1998). Dot blot showed that both perlecan and agrin appeared mainly in the D2 fractions, although reactions were visible also in the high-density fraction (D1). The results indicated that the content of perlecan was somewhat higher in wolffish than in cod. The finding of perlecan and agrin epitopes in both D1 and D2 fractions may indicate differences in the glycosylation of the molecules, which result in variations in buoyant densities. Both agrin and perlecan may be highly glycosylated (Tsen et al., 1995). The most surprising observation shown in the present study was in the location of perlecan. In a previous study on bovine skeletal muscle, using the same mAb as in the present study, perlecan epitopes were strongly expressed pericellularly, everywhere in the basement membranes around each individual myofiber (Eggen et al., 1997). In the fish muscle, perlecan staining was restricted to the muscle fiber–myocommata junction, showing a somewhat stronger staining in wolffish. No staining was observed in the basement membrane area surrounding the myofibers in cod nor wolffish

(Fig. 10a and b). The importance of perlecan for basement membrane integrity, and fragility, is clearly demonstrated in perlecan knock out mice (Costell et al., 1999). In addition, a defective collagenous network was observed in the cartilage of perlecan null mutants (Arikawa-Hirasawa et al., 1999). The restriction of perlecan staining to the muscle fiber–myocommata junction in fish may reflect the difference in force transmission in fish and mammalian muscles. In both mammalia and fish there is need for good mechanical integration and tensile strength between the muscle fibers and the connective tissue in the myotendinosus/myocommata junction in order to transmit contractile forces. Furthermore, in mammalia, transmission of contractile forces is not solely in the myotendinous junction, but also lateral transmission occurs between neighbouring myofibers and fascicles through the intramuscular connective tissue, that subsequently need tensile strength (Trotter and Purslow, 1992). The different localization of perlecan in fish muscle, observed in the present study, may indicate that force transmission in fish muscle mainly occurs in the myofiber–myocommata junction (for review see Purslow, 2002). It is possible that the absence of perlecan around the separate myofibers, provides a more vulnerable pericellular area, and thus contributes to the reported post-mortem changes in the connective tissue of fish muscle resulting in gaping. The higher expression of perlecan observed in the non-gaping species may support this view. In the present study we have demonstrated the presence of agrin and perlecan in fish muscle. But the distribution of HS-chains visualized by the mAbs against HS epitopes, did only partially coincide with the distribution patterns obtained for perlecan and agrin. The different distribution patterns could be influenced by discrepancies in epitope exposure. The intense staining of HS around the individual myofibers, using the monoclonal anti HS antibodies 10E4 and HepSS-1, most likely reflects the presence of other HSPGs in this area of fish muscle. Syndecans and glypicans are integral membrane HSPGs that are expressed in both vertebrates and invertebrates, and are ascribed a role in cell adhesion (for review Couchman and Woods, 2000; Filmus and Song, 2000). Further studies are needed to elucidate the composition and functional roles of the different HSPGs of fish white muscle and the relationship to quality. Acknowledgements This study was financially supported by Norwegian Research Council, grant no. 140791/130. Professor Jarl Bøgwald and Dr. Roy Dalmo, Norwegian College of Fishery Sciences/University of Tromsø, were of great assistance during radiolabeling and sampling trials. Professor Terje Lømo and his group at Department of Physiology/University of Oslo, kindly provided samples of monoclonal agrin antibodies (Agr-131 and Agr-86). Anne-Birgit Bævre, The Norwegian Food Research Institute, was most helpful in the lab and gave valuable assistance.

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

References Arikawa-Hirasawa, E., Watanabe, H., Takami, H., Hassell, J.R., Yamada, Y., 1999. Perlecan is essential for cartilage and cephalic development. Nat. Genet. 23, 354–358. Bai, X.M., Van der Schueren, B., Cassiman, J.J., Van den Berghe, H., David, G., 1994. Differential expression of multiple cell-surface heparan sulfate proteoglycans during embryonic tooth development. J. Histochem. Cytochem. 42, 1043–1054. Battaglia, C., Mayer, U., Aumailley, M., Timpl, R., 1992. Basement-membrane heparan sulfate proteoglycan binds to laminin by its heparan sulfate chains and to nidogen by sites in the protein core. Eur. J. Biochem. 208, 359–366. Belting, M., 2003. Heparan sulfate proteoglycan as a plasma membrane carrier. Trends Biochem. Sci. 28, 145–151. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Brandan, E., 1994. Proteoglycans in skeletal muscle. Braz. J. Med. Biol. Res. 27, 2109–2116. Costell, M., Gustafsson, E., Aszódi, A., Mörgelin, M., Bloch, W., Hunziker, E., Addicks, K., Timpl, R., Fässler, R., 1999. Perlecan maintains the integrity of cartilage and some basement membranes. J. Cell Biol. 147, 1109–1122. Couchman, J.R., Ljubimov, A.V., 1989. Mammalian tissue distribution of a large heparan sulfate proteoglycan detected by monoclonal antibodies. Matrix 9, 311–321. Couchman, J.R., Woods, A., 2000. Signaling through the syndecan proteoglycans. In: Iozzo, R.V. (Ed.), Proteoglycans: Structure, Biology and Molecular Interactions. Marcel Dekker Inc., New York-Basel, pp. 147–160. David, G., Bai, X.M., Van Der Schueren, B., Cassiman, J.J., Van Den Berghe, H., 1992. Developmental changes in heparan sulfate expression: in situ detection with mAbs. J. Cell Biol. 119, 961–975. Denzer, A.J., Brandenberger, R., Gesemann, M., Chiquet, M., Ruegg, M.A., 1997. Agrin binds to the nerve–muscle basal lamina via laminin. J. Cell Biol. 137, 671–683. Dunlevy, J.R., Hassel, J.R., 2000. Heparan sulfate proteoglycans in basement membranes. In: Iozzo, R.V. (Ed.), Proteoglycans: Structure, Biology and Molecular Interactions. Marcel Dekker Inc., New York-Basel, pp. 275–315. Eggen, K.H., Malmstrøm, A., Sørensen, T., Høst, V., 1997. Identification of proteoglycans in bovine M. semimebranosus by immunohistochemical methods. J. Muscle Foods 8, 121–136. Erickson, A.C., Couchman, J.R., 2000. Still more complexity in mammalian basement membranes. J. Histochem. Cytochem. 48, 1291–1306. Eusebio, A., Oliveri, F., Barzaghi, P., Ruegg, M.A., 2003. Expression of mouse agrin in normal, denervated and dystrophic muscle. Neuromuscul. Disord. 13, 408–415. Ferns, M.J., Campanelli, J.T., Hoch, W., Scheller, R.H., Hall, Z., 1993. The ability of agrin to cluster AChRs depends on alternative splicing and on cell surface proteoglycans. Neuron 11, 491–502. Filmus, J., Song, H.H., 2000. Glypicans. In: Iozzo, R.V. (Ed.), Proteoglycans: Structure, Biology and Molecular Interactions. Marcel Dekker Inc., New York-Basel, pp. 161–176. Gesemann, M., Brancaccio, A., Schumacher, B., Ruegg, M.A., 1998. Agrin is a high-affinity binding protein of dystroglycan in non-muscle tissue. J. Biol. Chem. 273, 6000–6005. Graham Jr., R.C., Karnovsky, M.J., 1966. The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem. 14, 291–302. Hassell, J.R., Yamada, Y., Arikawa-Hirasawa, E., 1993. Role of perlecan in skeletal development and diseases. Glycoconj. J. 19, 263–267. Hayashi, K., Madri, J.A., Yurchenco, P.D., 1992. Endothelial cells interact with the core protein of basement membrane perlecan trough β1 and β3 integrins: an adhesion modulated by glycosaminoglycan. J. Cell Biol. 119, 945–959. Hoch, W., Campanelli, J.T., Harrison, S., Scheller, R.H., 1994. Structural domains of agrin required for clustering of nicotinic acetylcholine receptors. EMBO J. 13, 2814–2821.

451

Iozzo, R.V., 1994. Perlecan: a gem of a proteoglycan. Matrix Biol. 14, 203–208. Iozzo, R.V., 1998. Matrix proteoglycans: from molecular design to cellular function. Annu. Rev. Biochem. 67, 609–652. Iozzo, R.V., 2001. Series introduction: heparan sulfate proteoglycans: intricate molecules with intriguing functions. J. Clin. Invest. 108, 165–167. Isemura, M., Sato, N., Yamaguchi, Y., Aikawa, J., Munakata, H., Hayashi, N., Yosizawa, Z., Nakamura, T., Kubota, A., Arakawa, M., Hsu, C.C., 1987. Isolation and characterization of fibronectin-binding proteoglycan carrying both heparan sulfate and dermatan sulfate chains from human placenta. J. Biol. Chem. 262, 8926–8933. Kjellén, L., Lindahl, U., 1991. Proteoglycans: structures and interactions. Annu. Rev. Biochem. 60, 443–475. Kure, S., Yoshie, O., 1986. A syngeneic monoclonal antibody to murine Meth-A sarcoma (HepSS-1) recognizes heparan sulfate glycosaminoglycan (HSGAG): cell density and transformation dependent alteration in cell surface HS-GAG defined by HepSS-1. J. Immunol. 137, 3900–3908. Kure, S., Yoshie, O., Aso, H., 1987. Metastatic potential of murine B16 melanoma correlates with reduced surface heparan sulfate glycosaminoglycan. Jpn. J. Cancer Res. 78, 1238–1245. Nakajima, M., Irimura, T., Nicolson, G.L., 1988. Heparanases and tumor metastasis. J. Cell. Biochem. 36, 157–167. Nakato, H., Kimata, K., 2002. Heparan sulfate fine structure and specificity of proteoglycan functions. Biochim. Biophys. Acta 1573, 312–318. Ofstad, R., Olsen, R.L., Taylor, R., Hannesson, K., in press. Breakdown of intramuscular connective tissue in cod (Gadus morhua) and spotted wolffish (Anarhichas minor O.) related to gaping. LWT. Pfeiler, E., 1991. Glycosaminoglycan composition of anguilliform and elopiform leptocephali. J. Fish Biol. 38, 533–540. Pfeiler, E., 1998. Acidic glycosaminoglycans in marine teleost larvae: evidence for a relationship between composition and negative charge density in elopomorph leptocephali. Comp. Biochem. Physiol. B 119, 137–144. Pfeiler, E., Donnelly, J., Torres, J.J., Crabtree, R.E., 1991. Glycosaminoglycan composition of tarpon (Megalops atlanticus) and ladyfish (Elops saurus) leptocephali. J. Fish Biol. 39, 613–615. Pfeiler, E., Toyoda, H., Williams, M.D., Nieman, R.A., 2002. Identification, structural analysis and function of hyaluronan in developing fish larvae (leptocephali). Comp. Biochem. Physiol. B 132, 443–451. Purslow, P.P., 2002. The structure and functional significance of variations in the connective tissue within muscle. Comp. Biochem. Physiol. A 133, 947–966. Rogalski, T.M., Williams, B.D., Mullen, G.P., Moerman, D.G., 1993. Products of the unc-52 gene in Caenorhabditis elegans are homologous to the core protein of the mammalian basement membrane heparan sulfate proteoglycan. Genes Dev. 7, 1471–1484. Ruegg, M.A., Tsim, K.W., Horton, S.E., Kröger, S., Escher, G., Gensch, E.M., McMahan, U.J., 1992. The agrin gene codes for a family of basal lamina proteins that differ in function and distribution. Neuron 6, 811–823. Shively, J.E., Conrad, H.E., 1976. Nearest neighbor analysis of heparin: identification and quantitation of the products formed by selective depolymerization procedures. Biochemistry 15, 3943–3950. Shworak, N.W., 2001. High-specific-activity 35S-labeled heparan sulfate prepared from cultured cells. Methods Mol. Biol. 171, 79–89. Stringer, S.E., Gallagher, J.T., 1997. Molecules in focus: heparan sulphate. Int. J. Biochem. Cell Biol. 29, 709–714. Su, Y.K., Elam, J.S., 2003. Proteoglycan regulation of goldfish retinal explant growth on optic tectal membranes. Brain Res. Dev. Brain Res.142, 169–175. Tingbo, M.G., Kolset, S.O., Ofstad, R., Enersen, G., Hannesson, K.O., 2005. Sulfated glycosaminoglycans in the extracellular matrix of muscle tissue in Atlantic cod (Gadus morhua) and spotted wolffish (Anarhichas minor). Comp. Biochem. Physiol. B 140, 349–357. Trotter, J.A., Purslow, P.P., 1992. Functional morphology of the endomysium in series fibered muscles. J. Morphol. 212, 109–122. Tsen, G., Halfter, W., Kröger, S., Cole, G.J., 1995. Agrin is a heparan sulfate proteoglycan. J. Biol. Chem. 270, 3392–3399.

452

M.G. Tingbø et al. / Comparative Biochemistry and Physiology, Part B 143 (2006) 441–452

Tumova, S., Woods, A., Couchman, J.R., 2000. Heparan sulfate proteoglycans on the cell surface: versatile coordinators of cellular functions. Int. J. Biochem. Cell Biol. 32, 269–288. van den Born, J., Salmivirta, K., Henttinen, T., ôstman, N., Ishimaru, T., Miyaura, S., Yoshida, K., Salmivirta, M., 2005. Novel heparan sulfate structures revealed by monoclonal antibodies. J. Biol. Chem. 280, 20516–20523.

Velleman, S.G., 2002. Role of the extracellular matrix in muscle growth and development. J. Anim. Sci. 80E, E8–E13. Wada, M., Kato, T., Kurita, K., Shikama, Y., Toyoshima, I., Sasaki, H., 1995. Detection of heparan sulfate in spinal spheroids of β,β′-iminodipropionitrile (IDPN)-treated rats. Neurosci. Lett. 202, 137–140.