Isolation and partial characterisation of immunoglobulin from southern bluefin tuna Thunnus maccoyii Castelnau

Fish & Shellfish Immunology (2001) 11, 491–503 doi:10.1006/fsim.2000.0329 Available online at http://www.idealibrary.com on

Isolation and partial characterisation of immunoglobulin from southern bluefin tuna Thunnus maccoyii Castelnau M. WATTS, B. L. MUNDAY1

AND

C. M. BURKE

Tasmanian Aquaculture and Fisheries Institute, School of Aquaculture, Locked Bag 1–370, Launceston, Tasmania 7250, Australia and CRC for Aquaculture, University of Tasmania, Launceston, Tasmania 7250, Australia and 1University of Tasmania, School of Biomedical Science, Locked Bag 1–320, Launceston, Tasmania, 7250, Australia (Received 15 June 2000, accepted after revision 27 November 2000) Specific and total serum immunoglobulins were extracted by immunoa$nity, mannan-binding protein and Protein A a$nity chromatography from southern bluefin tuna (Thunnus maccoyii Castelnau) immunised with rabbit IgG, and from non-immunised southern bluefin tuna. SDS-PAGE in 10% reducing gels revealed two heavy chains with molecular weights of approximately 74·61·3 kDa and 71·20·9 kDa, and two light chains with molecular weights of approximately 291·2 kDa and 281·0 kDa. Under non-reducing, but denaturing, conditions in 4% and 5% SDS-PAGE gels, a high molecular weight and a low molecular weight fraction were demonstrated. By gel filtration using Sephacryl HR 300 a molecular weight of 845 kDa, consistent with a tetramer, was obtained for the high molecular weight fraction, and a molecular weight of 168 kDa, consistent with a monomer, was obtained for the low molecular weight fraction. The extinction coe$cient at A280 for the purified immunoglobulin (Ig) was determined to be 1·24. Tuna
-rabbit IgG Ig was reactive with all non-reduced mammalian IgG antigens tested, suggesting that common conformational antigenic determinants were recognised.  2001 Academic Press

Key words:

southern bluefin tuna, immunoglobulins, Protein A, mannanbinding protein, a$nity purification.

I. Introduction Two classes of immunoglobulin (Ig) have been described in teleosts; IgM and the recently discovered IgD homologue (Miller et al., 1998; Hordvik et al., 1999). Teleost IgM is similar to mammalian IgM in that both are composed of equimolar amounts of heavy (H) chains (70 to 81 kDa) and light (L) chains (22 to 32 kDa) (reviewed in Wilson & Warr, 1992). Individual H and L chains form pairs which are held together by disulphide bonds and two H:L chain pairs form the basic monomer unit (H2L2). In mammalian IgM the monomer is disulphide-bonded between the two H chains and monomers associate into pentamers, linked by a J chain. In teleost IgM, the J chain is generally absent (Kobayashi et al., 1982; Wilson & Warr, 1992) and the way the polymer is 1050–4648/01/060491+13 $35.00/0

491

 2001 Academic Press

492

M. WATTS ET AL.

associated is strikingly di#erent. In teleost IgM, disulphide bonding between adjacent H chains is not uniform and non-covalent bonding is a frequent feature of the association of sub-units to form a complete tetramer (Lobb, 1985; Kaattari et al., 1998). Although stable under physiological conditions (Wilson & Warr, 1992), under denaturing conditions teleost IgM has been shown to exist both as fully and incompletely cross-linked tetramers, termed redox forms by Kaattari et al. (1998). It has been suggested that possession of di#erent polymerisation forms is a mechanism by which teleosts generate antibody diversity (Kaattari et al., 1998). However, di#erent redox forms do not account for all of the diversity observed (reviewed in Wilson & Warr, 1992). Isotypes of IgM have been described based on antigenic (Sanchez & Dominguez, 1991; Bang et al., 1996), electrophoretic (Lobb & Clem, 1981; Havarstein et al., 1988) and functional (Elcombe et al., 1985) heterogeneity. Di#erent H chains have been demonstrated, also based on antigenic (Lobb & Olson, 1988) and electrophoretic (Lobb & Clem, 1981; Fuda et al., 1992; Suzuki et al., 1994) characteristics. That the observed heterogeneity lies within the constant region of the H chain has not been demonstrated, and so a class basis for these di#erences has not been established. The aims of this work were to begin the characterisation and structural elucidation of southern bluefin tuna (Thunnus maccoyii Castelnau) Ig for comparative purposes. One-step extraction methods involving specific antigen/antibody reactions (Sanchez et al., 1991; Smith, 1992; Palenzuela et al., 1996), lectin-binding (mannanbinding protein) (Nevens et al., 1992; Cobb et al., 1998) and bacterial cell wall proteins (Protein A) (Estevez et al., 1993a,b; Kanlis et al., 1995; Scapigliatti et al., 1996) were chosen because of their ease of use and the purity of the extracted Ig. Specificity and reactivity of serum Ig from southern bluefin tuna immunised with rabbit IgG were assessed in immunoblots using various Ig and serum antigens.

II. Materials and Methods PRODUCTION OF SPECIFIC SOUTHERN BLUEFIN TUNA Ig

Two captive southern bluefin tuna (designated L and S), held in a sea-cage at Pt Lincoln, South Australia (3444 S 13552 E), were inoculated with rabbit IgG (Sigma). A total of 10 mg of rabbit IgG, made up in 6 ml of sterile 0·01 M phosphate, 0·15 M NaCl, pH 7·2 was given by intraperitoneal inoculation into each fish in three injections over a 6-week period. The first inoculation (at 0 weeks) was mixed with 1 ml of Freund’s complete adjuvant; the second (at four weeks) and third (at six weeks) were each mixed with 1 ml of Freund’s incomplete adjuvant. Eight weeks after the first inoculation the fish were killed and bled from the lateral blood vessel in the pectoral fin recess. The blood was allowed to clot at ambient temperature, then placed at 4 C for clot retraction to take place. The serum was separated, frozen and delivered to the School of Aquaculture, University of Tasmania, where it was aliquotted and held at
20 C until use.

IMMUNOGLOBULIN FROM SOUTHERN BLUEFISH TUNA

493

TITRATION OF SPECIFIC SOUTHERN BLUEFIN TUNA Ig

Response to rabbit IgG was determined by dot-blot titration using a dot-blot apparatus (Millipore) fitted with a 0·45 m pore nitro-cellulose membrane (Biorad) pre-soaked in 20 mM Tris-HCl, 0·5 M NaCl, pH 7·5 (TBS) following the procedure described in Microfiltration Blotting (Biorad Bulletin 1721). Three micrograms of rabbit IgG, in 100 l of TBS, was applied to each well and unoccupied membrane sites were blocked with 300 l well
1 of 3% bovine serum albumin in TBS. Each tuna
-rabbit IgG antiserum, was serially diluted from 1/10 to 1/20 480 in 1% bovine serum albumin in TBS (TBSD). Bound antibodies were detected with sheep
-tuna Ig antiserum (Watts, 2000) diluted 1/1000 in TBSD. The complex was detected with 100 l per well of alkaline phosphatase-conjugated donkey anti-sheep IgG antibody (Sigma), diluted to 1/30 000 in TBSD, and developed with 5-bromo-4-chloro-3-indolyl phosphate/ nitro-blue tetrazolium (BCIP/NBT) substrate (Biorad kit). The titre was expressed as the ‘dilution of antibody solution which gives one-half the maximal signal with the detection system being used’ (Bollag et al., 1996). Appropriate negative controls were used.

IMMUNOGLOBULIN PURIFICATION

Extraction of specific southern bluefin tuna Ig by immunoaffinity chromatography Tuna
-rabbit IgG Ig was extracted from immunised southern bluefin tuna serum by a$nity purification on a 2 ml rabbit IgG agarose column (Sigma) (Smith, 1992). Extraction of southern bluefin tuna Ig via mannan-binding protein A 5-ml gel-bed capacity Immunopure Immobilized Mannan-Binding Protein kit (Pierce Co) was used to extract serum Ig from immunised and wild southern bluefin tuna (Nevens et al., 1992). As a preparative step, sera were dialysed against 1 l of 10 mM Tris-HCl, 1·25 M NaCl, 0·02% sodium azide, pH 7·4 for 24 h, to remove interfering potassium ions. Extraction of southern bluefin tuna Ig via Protein A One millilitre of Protein A (Sigma) was packed into a 101 cm borosilicate column (Biorad) and the procedure described in Bollag et al. (1996) was used to extract serum Ig from immunised and wild southern bluefin tuna. All sera were filter sterilised by passing through a 0·22-m low proteinbinding syringe filter (Millipore) before layering onto columns. All extraction procedures were monitored at A280 to determine elution of Ig fractions. Eluents from the same procedures were pooled and filter-concentrated by centrifugation at 2000g with a Millipore Ultrafree-15 Biomax-10 device to approximately 1·4 mg ml
1. During concentration the elution bu#er was changed to 0·02 M Tris-HCl, 0·15 M NaCl, pH 7·4. The protein concentration of concentrated samples was estimated using the Biorad Protein Assay kit (microassay method). The extinction coe$cient of a purified, lyophilised sample was determined at A280.

494

M. WATTS ET AL.

STRUCTURE AND MOLECULAR WEIGHT DETERMINATION

SDS-PAGE reducing and non-reducing conditions The purity, structure and molecular weight of the tuna Ig, purified by each method, were determined by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) in 10% reducing gels and 4% and 5% non-reducing gels (Bollag et al., 1996) in a Mighty Small SE 250 vertical electrophoresis unit (Hoefer). Samples at a concentration of approximately 1·4 mg ml
1 were diluted 1:1 in 0·125 M Tris, 4% SDS, 20% glycerol, 10% mercaptoethanol (sample bu#er) for reducing conditions, or omitting mercaptoethanol for non-reducing conditions. Electrophoresis was carried out for 1 h in 25 mM Tris, 192 mM glycine, 0·1% SDS, pH 8·3. Human IgM and rainbow trout (Oncorhynchus mykiss) IgM at a final concentration of 1 mg ml
1 were included on some gels for comparison. Wide Range Protein Standards (Mark 12 Novex) or High Molecular Weight Standards (Pharmacia LKB) were included for molecular weight determination. The gels were stained with 0·025% Coomassie Brilliant Blue R-250 (CBB) (Bollag et al., 1996) or silver. Gel filtration The molecular weight of purified Ig extracted by immunoa$nity over rabbit IgG agarose was also determined using gel filtration on Sephacryl HR 300. Molecular weight standards were human IgM (950 kDa), thyroglobulin (669 kDa), rabbit IgG (150 kDa) and bovine albumin (66 kDa). Two-dimensional SDS-PAGE To obtain more detailed structural information, two-dimensional SDSPAGE with non-reducing conditions in the first dimension, followed by reducing conditions in the second dimension, was carried out as follows: Stage 1. One hundred microlitres of tuna
-rabbit IgG Ig, diluted 1:1 in sample bu#er, was loaded into a single preparative well, formed in a 3% stacking gel over a 5% resolving gel, and electrophoresed under non-reducing conditions until the dye front was approximately 1 cm from the bottom of the gel. Stage 2. On completion of Stage 1, the gel was removed and a strip equivalent to a whole lane was excised vertically from the gel and placed into a tray with 0·125 M Tris, 4% SDS, 20% glycerol, 0·5% dithiothreitol, and heated at 60 C for 10 min. After reduction the strip was placed horizontally into a single preparative well formed in the top of a 4% stacking gel over a 10% resolving gel. The gap between the gel strip and the well was filled with 4% stacking gel solution, and left to polymerise for 20 min. Electrophoresis was then carried out in the second dimension. On completion, gels were stained with silver. Wide Range Protein Standards (Mark 12 Novex) were used for molecular weight calibration.

IMMUNOGLOBULIN FROM SOUTHERN BLUEFISH TUNA

495

ANTIBODY ACTIVITY

All purified Igs were tested for activity by their reaction with rabbit IgG in immunoblots. Preliminary immunoblots and dot-blots (data not shown) demonstrated that tuna
-rabbit IgG Ig was reactive with the following alkaline-phosphatase conjugated Igs (Sigma): donkey
-sheep IgG, goat
-rabbit IgG, rabbit
-mouse and sheep
-human IgG. To further characterise this activity, specificity and cross-reactivity were determined in immunoblots against various other purified mammalian IgGs (rabbit, bovine and human (Sigma)), and human, rabbit, monotreme (platypus: Ornithorhynchus anatinus), reptile (blue tongued lizard: Tiliqua nigrolutea), amphibian (cane toad: Bufo marinus), bird (native hen: Egernia whitii), and fish (rainbow trout and Atlantic salmon: Salmo salar) sera. Immunoglobulin antigens were used at a final concentration of approximately 1 mg ml
1 and all serum samples were used at a final dilution of 1/20. Electrophoresis was carried out in non-reducing 5% gels and electrophoresed proteins were electro-transferred onto 0·45-m nitro-cellulose membrane (Biorad) on a Hoefer Semi-Phor Semi-dry transfer unit at 0·8 mA cm
2 for 1·5 h in 48 mM Tris, 39 mM glycine, 20% methanol, pH 9·2. Following transfer, unoccupied membrane sites were blocked with 5% skim milk in TBS and immunorecognition was carried out with tuna
-rabbit IgG Ig at a concentration of 75 g in 10 ml TBS, 3% skim milk. Bound antibodies were detected with sheep
-tuna Ig antiserum diluted 1/1000 in the same bu#er, followed by incubation with alkaline phosphatase-conjugated donkey anti-sheep IgG antibody (Sigma) at a dilution of 1/30 000 also in the same bu#er. The complex was detected with BCIP/NBT (Biorad substrate kit). All antibody incubations were carried out at ambient temperature for 1 h. Pooled tuna sera from wild southern bluefin tuna was used as the negative antibody control. III. Results TITRATION OF TUNA
-RABBIT IgG ANTISERUM

Both immunised southern bluefin tuna responded to rabbit IgG. ‘L’ gave a response three times greater than ‘S’ with a titre of 1/5120 compared with 1/1280. IMMUNOGLOBULIN PURIFICATION

All three purification methods successfully extracted proteins consistent with Ig from southern bluefin tuna sera. Protein A extraction gave the greatest yield of approximately 1·5 mg Ig ml
1 serum. Immunoa$nity and mannan-binding protein extraction yielded from 0·16 to 0·3 mg Ig ml
1 serum. The extinction coe$cient of lyophilised, immunoa$nity-purified tuna
-rabbit IgG Ig was 1·24. MOLECULAR WEIGHT DETERMINATION

Reduced immunoglobulin SDS-PAGE in reducing conditions (n=5) demonstrated two H chains with molecular weights 71·21.3 kDa and 74·60.9 kDa and two L chains

496

M. WATTS ET AL.

kD

1

2

3

4

5

6

200 116 97 66 55

37 31

22

Fig. 1. SDS-PAGE analysis of reduced southern bluefin tuna serum Ig in 10% gels stained with Coomassie Brilliant Blue. Lane 1: molecular weight standards; lane 2: human IgM; lane 3: serum Ig extracted from immunised southern bluefin tuna via mannan-binding protein; lane 4: serum Ig extracted from non-immunised southern bluefin tuna via mannan-binding protein; lane 5: serum Ig extracted via Protein A from non-immunised southern bluefin tuna; lane 6: immunoa$nity-purified southern bluefin tuna serum Ig. Molecular weight is given in kilodaltons (kDa). Top arrows indicate the 2H chains and bottom arrows indicate the 2L chains.

with molecular weights of 281 kDa and 291 kDa in Ig purified by all methods, with only minor bands in other locations (Fig. 1). Within the H chain doublet the band at 74·60·9 kDa was always more intensely stained. Unreduced immunoglobulin In non-reducing conditions a high molecular weight component was demonstrated in Southern bluefin tuna Ig extracted by all methods (Fig. 2a and b, top arrows). Immunoa$nity purification and purification over Protein A also revealed a low molecular weight component (Fig. 2a and b, bottom arrows). This fraction was not discernible in total serum Ig extracted over mannanbinding protein from non-immunised southern bluefin tuna (Fig. 2a, lane 4) and only slightly present in serum Ig extracted via mannan-binding protein from immunised tuna (Fig. 2a, lane 3). In 4% gels a molecular weight of 778 kDa (consistent with a tetramer) was obtained for the high molecular weight component and 180 kDa (consistent with a monomer) for the low molecular weight component. Human IgM and rainbow trout IgM included on the 4% gel gave values of 922 kDa and 736 kDa, which is consistent with literature values. Trout IgM dissociated into the characteristic bands associated with numerous redox forms (Fig. 2, lane 1), and human IgM did not dissociate (Fig. 2, lane 6). In 5% non-reducing gels the molecular weight of the low molecular weight fraction of tuna Ig was calculated to be 160 kDa

497

IMMUNOGLOBULIN FROM SOUTHERN BLUEFISH TUNA

(a) 1

(b) 2

3

4

5

6

kD

1

M

kD

669

200 440 200

116 116

97

Fig. 2. SDS-PAGE analysis of non-reduced southern bluefin tuna serum Ig stained with silver. (a) 4% gel. Lane 1: rainbow trout IgM; lane 2: serum Ig extracted from non-immunised southern bluefin tuna via Protein A; lane 3: serum Ig extracted from immunised southern bluefin tuna via mannan-binding protein; lane 4: serum Ig extracted from non-immunised southern bluefin tuna via mannan-binding protein; lane 5: immunoa$nity-purified southern bluefin tuna serum Ig; lane 6: human IgM. (b) 5% gel. Lane 1: serum Ig extracted from immunised southern bluefin tuna via Protein A. M: molecular weight markers. Molecular weight is given in kilodaltons (kDa). Arrows indicate high and low molecular weight components.

(Fig. 2b) (somewhat less than that obtained in 4% gels). By gel filtration, the molecular weight of the high molecular weight fraction was estimated to be 845 kDa and the low molecular weight fraction was 168 kDa. Two-dimensional SDS-PAGE In the first, non-reducing dimension in a 5% resolving gel, immunoa$nitypurified southern bluefin tuna Ig was resolved into both high molecular weight and low molecular weight fractions (Fig. 3) as previously demonstrated (Fig. 2, lane 5). In the second dimension, SDS-PAGE in reducing conditions confirmed that both fractions consisted of H chains of molecular weight 74·6 and 72·1 kDa and L chains of 29 and 28 kDa. ANTIBODY ACTIVITY

Serum Ig extracted from immunised tuna was reactive with non-reduced rabbit IgG. In immunoblots, immunoa$nity-purified tuna
-rabbit IgG Ig also reacted with non-reduced Igs and bands consistent with Ig in all mammalian sera tested (Fig. 4), but not with sera from selected fish, amphibian, reptile or bird species. Also, no reaction was seen with any reduced Ig or serum, even with those from rabbit (data not shown).

498

M. WATTS ET AL.

First dimension kD

66

97 116

Second dimension

LMW

200

HMW

200 116 97 66 55 36 31 21

Fig. 3. Two-dimensional SDS-PAGE analysis of immunoa$nity-purified southern bluefin tuna serum Ig. The first dimension was run in non-reducing conditions in the direction of the long horizontal arrow. A gel strip was then excised, reduced and electrophoresed in the second dimension, in the direction of the vertical arrow. Two H chains and two L chains in the high molecular weight (HMW) and low molecular weight (LMW) components are indicated (small arrows).

IV. Discussion STRUCTURE OF SOUTHERN BLUEFIN TUNA Ig

Molecular weight characterisation All extraction methods were successful for purification of Ig from southern bluefin tuna sera, which, on reduction, dissociated into the same molecular weight H and L chain bands, comparable with those found in the literature for other teleost species (Havarstein et al., 1988; Lobb & Olson, 1988; Pilstrom & Petersson, 1991; Romboult et al., 1993; Palenzuela et al., 1996) and reviewed in Wilson & Warr (1992). Under non-reducing conditions immunoa$nity-extracted Ig (and to a lesser extent Protein A-extracted Ig), revealed two major molecular weight populations; one of high molecular weight and one of low molecular weight. Two-dimensional SDS-PAGE revealed that both of these populations were composed of H and L chains. In 4% gels the high molecular weight population had an apparent molecular weight of 778 kDa and migrated between human IgM and trout IgM. Assuming southern bluefin tuna Ig has a tetrameric structure, addition of 8 H chains and 8 L chains would yield a molecule of 815 kDa. A pentamer with these H and L chains would have a molecular weight of >1000 kDa, and would therefore migrate more slowly than human IgM. Consequently, from the migration pattern and the apparent molecular weight it is reasonable to conclude that the molecule is a tetramer. The weight of 845 kDa obtained by gel filtration was much closer to the predicted value. By SDS-PAGE the low molecular weight fraction had a molecular weight of between 160 and 180 kDa and by gel filtration the molecular weight was 168 kDa. As this

IMMUNOGLOBULIN FROM SOUTHERN BLUEFISH TUNA

kD

1

2

3

4

5

499

6

200

97

66

Fig. 4. Antibody activity of immunoa$nity-purified tuna
-rabbit IgG Ig against mammalian IgGs and sera. Immunoblot analysis from SDS-PAGE non-reducing 8% gel. Antigen lanes: 1, human IgG; 2, human serum; 3, bovine IgG; 4, rabbit serum; 5, rabbit IgG; 6, platypus serum. Appropriate negative controls were used.

fraction was also composed of H and L chains a monomeric structure is suggested. However, by addition of H and L chains, the predicted weight of a monomer should be 210 kDa and, therefore, the weight derived from non-reducing gels and gel filtration is a little too low. This molecular weight discrepancy has also been described for other fish species and was suggested to be due to conformational di#erences between the molecular weight markers and non-reduced Ig (Adkinson et al., 1996). This may also apply to tuna Ig. Presence of isotypes As both tetrameric and monomeric Ig forms were isolated by molecular sieving with gel filtration it is unlikely that the monomer is a (redox) component of a non-covalently bonded tetramer, as has been described for numerous other teleost species (Kobayashi et al., 1982; Lobb, 1986; Sanchez et al., 1991; Navarro et al., 1993; Romboult et al., 1993; Whittington, 1993; VanderHeijden et al., 1995; Kaattari et al., 1998). Distinct monomeric Ig forms are not consistently present in teleosts, but have been demonstrated in: giant grouper (Epinephelus itaira) (Clem, 1971), margate (Haemulon album) (Clem & McLean, 1975), toadfish (Speroides glaber) (Warr, 1982), rainbow trout (Elcombe et al., 1985; Sanchez & Dominguez, 1991), sheepshead (Archosargus probatocephalus) (Lobb & Clem, 1981), European perch (Perca fluviatilis) (Whittington, 1993) and flounder (Paralichthys olivaceus) (Bang et al., 1996) sera.

500

M. WATTS ET AL.

That the monomer is not derived from the tetramer is further supported by the di#erential extraction results. Serum Ig extracted by mannan-binding protein from both immunised and non-immunised southern bluefin tuna consistently demonstrated only the tetramer. Binding of mannan-binding protein involves non-specific interactions of carbohydrate moities on the Fc region with carbohydrate-binding regions on mannan-binding protein (Nevens et al., 1992). As only the tetramer was demonstrated following extraction with mannan-binding protein, it is likely that the tetramer and monomer di#er in amino acid glycosylation in the Fc region. As immunoa$nity extraction binds via the Fab by virtue of the antigen/ antibody reaction between the eliciting antigen and corresponding antibody, then both tetrameric and monomeric Ig populations must have antibody reactivity towards rabbit IgG, as both were extracted. Protein A binding usually involves non-specific interactions via the Fc region, but reactions via the Fab regions are possible (Sasso et al., 1991), which may explain the results obtained here. However, as two-dimensional SDS-PAGE demonstrated that both fractions have the same molecular weight H and L chain components, it is unlikely that they are di#erent Ig classes. SPECIFICITY OF TUNA
-RABBIT IgG Ig

Not only did tuna
-rabbit IgG Ig react with rabbit IgG, the eliciting antigen, but it also reacted with all unreduced, purified mammalian IgG tested. This reaction was initially confusing as cross-reactivity occurred with any conjugated antiserum that was used. However, it became clear that tuna
-rabbit IgG Ig was highly reactive with mammalian IgG (Fig. 4). No reaction was demonstrated with any teleost, amphibian, reptilian or avian sera or purified teleost Ig tested. Neither was there any reaction with reduced IgG, even rabbit IgG. Therefore, tuna
-rabbit IgG Ig is likely to be reacting with common conformational determinant(s) which are lost on reduction of the molecule. Other studies have demonstrated this type of recognition (Neoh et al., 1973; Marchalonis et al., 1992). Generally, antibodies produced by teleosts are as specific as mammalian antibodies (Mochida et al., 1994). Therefore, the demonstrated cross-reactivity probably reflects high homology between mammalian IgG rather than the production of low-specificity antibodies by tuna. In conclusion, the results show that high purity Ig was extracted by each method and extraction with Protein A gave the highest yield. However, di#erent extraction methods produced di#erent results, highlighting the problem of interpretation regarding the presence of isotypes when only a single extraction procedure is employed. With regard to Ig structure it appears that southern bluefin tuna are similar to other teleosts in that they possess the predominant high molecular weight Ig tetramer which dissociates into H chains of 70–80 kDa, and L chains of 22–30 kDa. However, H chains were found as a doublet, which is unusual, and a significant amount of serum Ig was present as a monomer. Both forms had antibody activity, demonstrated by immunoa$nity extraction with rabbit IgG agarose. The cross-reactivity between mammalian IgG demonstrated with tuna
-rabbit IgG Ig is probably a result of the phylogenetic distance between tuna and mammals.

IMMUNOGLOBULIN FROM SOUTHERN BLUEFISH TUNA

501

This work was supported by an Australian Research Council Post Graduate Award (Industry) in partnership with Veterinary Pathology Services. Additional funding and technical assistance was provided by the Tuna Boat Owners Association of Australia. Serum samples from wild southern bluefin tuna were collected by the Observer Programme of the Australian Fisheries Management Authority. Pierce Chemical Company kindly provided a mannan-binding protein a$nity column, and Professor Rob Raison of the Immunobiology Unit, University of Technology, Sydney, made a generous gift of human IgM and trout IgM. The assistance of sta# and students in a number of Schools of the University of Tasmania is gratefully acknowledged.

References Adkinson, M. A., Basurko, B. & Hendrick, R. P. (1996). Humoral immunoglobulins of the white sturgeon, Acipenser transmontanus: partial characterisation of and recognition with monoclonal antibodies. Developmental and Comparative Immunology 20, 285–298. Bang, J. D., Kim, J. W., Lee, S. D., Park, S. I., Chun, S. G., Jeong, C. S. & Park, J. W. (1996). Humoral immune response of flounder to Edwardsiella tarda: the presence of various sizes of immunoglobulins in flounder. Diseases of Aquatic Organisms 26, 197–203. Bollag, D. M., Rozycki, M. D. & Edelstein, S. J. (1996). Protein Methods. New York: Wiley-Liss, John Wiley and Sons Inc. Clem, L. W. (1971). Phylogeny of immunoglobulin structure and function IV. Immunoglobulins of the giant grouper, Epinephelus itaira. Journal of Biological Chemistry 246, 9–15. Clem, L. W. & McLean, W. E. (1975). Phylogeny of immunoglobulin structure and function VII. Monomeric and tetrameric immunoglobulins of the margate, a marine teleost fish. Immunology 29, 791–799. Cobb, C. S., Levy, M. G. & Noga, E. J. (1998). Acquired immunity to amylodiniosis is associated with an antibody response. Diseases of Aquatic Organisms 34, 125–133. Elcombe, B. M., Chang, R. J., Taves, C. J. & Winkelhake, J. L. (1985). Evolution of antibody structure and e#ector functions: comparative activities of monomeric and tetrameric IgM from rainbow trout, Salmo gairdneri. Comparative Biochemistry and Physiology 80B, 697–706. Estevez, J., Leiro, J., Sanmartin, M. L. & Ubiera, F. (1993a). Isolation and partial characterisation of turbot (Scophthalmus maximus) immunoglobulins. Comparative Biochemistry and Physiology 105A, 275–281. Estevez, J., Sanchez, C., Dominguez, J., Leiro, J., Sanmartin, M. L. & Ubiera, F. M. (1993b). Protein-A binding characteristics of rainbow trout (Oncorhynchus mykiss) immunoglobulins. Comparative Biochemistry and Physiology 106B, 173–180. Fuda, H., Hara, A., Yamazaki, F. & Kobayashi, K. (1992). A peculiar immunoglobulin M (IgM) identified in eggs of chum salmon (Oncorhynchus keta). Developmental & Comparative Immunology 16, 415–423. Havarstein, L. S., Aasjord, P. M., Ness, S. & Endresen, C. (1988). Purification and partial characterisation of an IgM-like serum immunoglobulin from Atlantic salmon (Salmo salar). Developmental and Comparative Immunology 12, 773–785. Hordvik, I., Thevarajan, J., Samdal, I., Bastani, N. & Krossoy, B. (1999). Molecular cloning and phylogenetic analysis of the Atlantic salmon immunoglobulin D gene. Scandinavian Journal of Immunology 50, 202–210. Kaattari, S. L., Evans, D. A. & Klemer, J. V. (1998). Varied redox forms of teleost IgM: an alternative to isotypic diversity? Immunological Reviews 166, 133–142. Kanlis, G., Suzuki, Y., Tauchi, M., Numata, T., Shirojo, Y. & Takashima, F. (1995). Immunoglobulin in oocytes, fertilised eggs, and yolk sac larvae of red sea bream. Fisheries Science 61, 787–790.

502

M. WATTS ET AL.

Kobayashi, K., Hara, A., Takano, K. & Hirai, H. (1982). Studies on subunit components of immunoglobulin from a bony fish, the chum salmon (Oncorhynchus keta). Molecular Immunology 19, 95–103. Lobb, C. J. (1985). Covalent structure and a$nity of channel catfish anti-dinitrophenyl antibodies. Molecular Immunology 22, 993–999. Lobb, C. J. (1986). Structural diversity of channel catfish immunoglobulins. Veterinary Immunology and Immunopathology 12, 7–12. Lobb, C. J. & Clem, L. W. (1981). Phylogeny of immunoglobulin structure and function-X. Humoral immunoglobulins of the sheepshead, Archosargus probatocephalus. Developmental and Comparative Immunology 5, 271–282. Lobb, C. J. & Olson, M. O. (1988). Immunoglobulin heavy H chain isotypes in a teleost fish. Journal of Immunology 141, 1236–1245. Marchalonis, J. J., Schluter, S. F., Yang, H. Y., Hohman, V. S., McGee, K. & Yeaton, L. (1992). Antigenic cross-reactions among immunoglobulins of diverse vertebrates (elasmobranchs to man) detected using xenoantisera. Comparative Biochemistry and Physiology 101A, 675–687. Miller, N., Wilson, M., Bengten, E., Stuge, T., Warr, G. & Clem, W. (1998). Functional and molecular characterisation of teleost leucocytes. Immunological Reviews 166, 187–197. Mochida, K., Lou, Y. H., Hara, A. & Yamauchi, K. (1994). Physical biochemical properties of IgM from a teleost fish. Immunology 83, 675–680. Navarro, V., Quesada, J. A., Abad, M. E., Taverne, N. & Romboult, J. H. M. W. (1993). Immuno(cyto)chemical characterisation of monoclonal antibodies to gilthead seabream (Sparus aurata) immunoglobulin. Fish & Shellfish Immunology 3, 167–177. Neoh, S. H., Jahoda, D. M. & Rowe, D. S. (1973). Immunoglobulin classes in mammalian species identified by cross-reactivity with antisera to human immunoglobulin. Immunochemistry 10, 805–813. Nevens, J. R., Mallia, A. K., Wendt, M. W. & Smith, P. K. (1992). A$nity chromatographic purification of immunoglobulin M utilising immobilised mannan binding protein. Journal of Chromatography 597, 247–256. Palenzuela, O., Sitja-Bobadilla, A. & Alvarez-Pellitero, P. (1996). Isolation and partial characterisation of serum immunoglobulins from sea bass (Dicentrarchus labrax L.) and gilthead sea bream (Sparus aurata L.). Fish & Shellfish Immunology 6, 81–95. Pilstrom, L. & Petersson, A. (1991). Isolation and partial characterisation of immunoglobulin from cod (Gadus morhua L). Developmental and Comparative Immunology 15, 143–152. Romboult, J. H. M. W., Taverna, N., VanDeKamp, M. & Taverna-Thiele, A. J. (1993). Di#erences in mucus and serum immunoglobulin of carp (Cyprinus carpio L.). Developmental and Comparative Immunology 17, 309–317. Sanchez, C., Coll, J. & Dominguez, J. (1991). One-step purification of the major rainbow trout immunoglobulin. Veterinary Immunology and Immunopathology 27, 383–392. Sanchez, C. & Dominguez, J. (1991). Trout immunoglobulin populations di#ering in light chains revealed by monoclonal antibodies. Molecular Immunology 28, 1271–1277. Sasso, E. H., Silverman, G. J. & Mannik, M. (1991). Human IgA and IgG F(ab)2 that bind to staphylococcal protein A belong to the VHIII subgroup. Journal of Immunology 147, 1877–1883. Scapigliatti, G., Romano, N., Picchetti, S., Mazzini, M., Mastrolia, L., Scalia, D. & Abelli, L. (1996). Monoclonal antibodies against sea bass Dicentrarchus labrax (L.) immunoglobulins: immunolocalisation of immunoglobulin-bearing cells and applicability in immunoassays. Fish & Shellfish Immunology 6, 383–401. Smith, S. A. (1992). A$nity purification of serum immunoglobulin from fish. In Techniques in Fish Immunology (J. S. Stolen, T. C. Fletcher, D. P. Anderson,

IMMUNOGLOBULIN FROM SOUTHERN BLUEFISH TUNA

503

S. L. Kaattari & A. F. Rowley, eds) pp. 125–129. Fairhaven, NJ: SOS Publications. Suzuki, Y., Orito, M., Furukawa, K. & Aida, K. (1994). Existence of low molecular weight immunoglobulin in carp eggs. Fisheries Science 60, 159–162. VanderHeijden, M. H. T., Rooijakkers, J. B. M. A., Booms, G. H. R., Romboult, J. H. M. W. & Boon, J. J. (1995). Production, characterisation and applicability of monoclonal antibodies to European eel (Anguilla anguilla L., 1758) immunoglobulin. Veterinary Immunology and Immunopathology 45, 151–164. Warr, G. W. (1982). Immunoglobulin of the toadfish, Spheroides glaber. Comparative Biochemistry and Physiology 76, 507–514. Watts, M. (2000). Immunology of Southern Bluefin Tuna. PhD Thesis, University of Tasmania, Launceston, Australia. Whittington, R. (1993). Purification and partial characterisation of serum immunoglobulin of the European perch (Perca fluviatilis L.). Fish & Shellfish Immunology 3, 331–343. Wilson, M. R. & Warr, G. W. (1992). Fish immunoglobulins and the genes that encode them. Annual Review of Fish Diseases 2, 201–221.