Characterization of antisera raised against stickleback (Gasterosteus aculeatus) MHC class I and class II molecules

Fish & Shellfish Immunology 23 (2007) 991e1002 www.elsevier.com/locate/fsi

Characterization of antisera raised against stickleback (Gasterosteus aculeatus) MHC class I and class II molecules J.P. Scharsack*, M. Kalbe, H. Schaschl 1 Department of Evolutionary Ecology, Max Planck Institute for Limnology, August-Thienemann-Str. 2, 24306 Ploen, Germany Received 31 January 2007; revised 6 March 2007; accepted 16 March 2007 Available online 24 March 2007

Abstract The three-spined stickleback (Gasterosteus aculeatus) is an important model organism for investigations on the maintenance of polymorphism of the major histocompatibility complex (MHC) of vertebrates. Analysis of functional aspects of MHC diversity in stickleback would benefit from the availability of MHC specific reagents. Here we characterize antisera raised against recombinant fusion proteins of stickleback MHC class I alpha and class II alpha and beta. Western blot analysis using recombinant proteins confirmed the specificity of the antisera. In brain and muscle preparations, neither of the MHC types was detectable. High levels of each MHC receptor type were observed in gills and spleen and lower levels in head kidneys. In histological sections of gills, epithelial cells of primary and secondary lamellae stained positive with MHC class I antiserum, while single, scattered cells stained positive for MHC class II. In sections of spleen and head kidney, considerable numbers of cells positive for either MHC type were detected. Molecular weight shift in SDS-PAGE after deglycosylation of MHC class I alpha and class II beta confirmed the predicted glyco-protein character of the molecules. The majority of MHC II alpha was not glycosylated; only a small fraction of MHC II alpha was susceptible to deglycosylation. This suggests differential expression of the two stickleback MHC II alpha genes (Gaac-DAA, Gaac-DBA) only one of which (Gaac-DBA) has a site for N-linked glycosylation. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Gasterosteus aculeatus; MHC class I; MHC class II; ECL; N-glycosylation; Western blot; Immunohistology

1. Introduction Like all jawed vertebrates, bony fish are endowed with the major histocompatibility complex (MHC) receptor family, subdivided into MHC class I and MHC class II molecules [1,2]. In vertebrates both types of receptors are characterized by a highly polymorphic peptide-binding grove in which self and non-self peptides are presented to T-cells [3,4]. In analogy to mammals, MHC class I receptors in teleosts present peptides from intra-cellular pathogens (e.g. viruses) to cytotoxic CD8þ T-cells, which kill infected cells [5]. Receptors of MHC class II present peptides from extra-cellular pathogens (e.g. bacteria, parasites) to CD4þ T-helper cells that trigger the activation of B-cells resulting * Corresponding author. Tel.: þ49 4522 763 256; fax: þ49 4522 763 310. E-mail addresses: [email protected] (J.P. Scharsack), [email protected] (M. Kalbe), [email protected] (H. Schaschl). 1 Present address: Konrad Lorenz Institute for Ethology, Savoyenstrasse 1a, A-1160 Vienna, Austria. 1050-4648/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2007.03.011

992

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

in an antibody mediated specific immune response and memory [6e8]. Based on the presence of functional T-helper cell activities and MHC class II molecules, comparable mechanisms may exist also in teleosts [9]. In three-spined sticklebacks (Gasterosteus aculeatus), genes of MHC class I and MHC class II [10,11] are highly polymorphic [12e16]. Therefore, sticklebacks have recently become a prime model organism for the investigation of MHC diversity and its consequences on immunocompetence and sexual selection [17e23]. In addition, the genomic organization of both MHC class I and class II have been studied recently [13,16]. In these studies, a bacterial artificial chromosome (BAC) library, which was created from a single stickleback, was screened for clones containing MHC class I and class II genes. In one study the presence of potentially nine MHC class I genes was revealed in the analysed BAC library [16]. In addition, a single BAC clone was sequenced and showed the presence of three MHC class I genes arranged in tandem repeat in the same transcription orientation. The detected MHC class I genes in the BAC clone were very similar to each other and it has been suggested that these MHC class I genes may have been generated by repeated gene duplication events [16]. The sequence analysis of a BAC clone containing MHC class II genes revealed two sets of paralogous class IIa- and b genes in tandem arrangement, probably generated by recent gene duplication events and inter- and intralocus recombination [13,14]. Furthermore, it has been estimated that sticklebacks contain up to six MHC class IIB genes [10]. The question remains, however, whether all three-spined stickleback individuals possess and express the same number of class I and class II genes. The number and the degree of polymorphism of MHC class IIb alleles in stickleback populations, differs among habitats [12,21]. Individual stickleback can have up to 12 alleles of MHC IIb, but for parasite resistance obviously intermediate numbers (i.e. 5e6) of MHC class IIb alleles are advantageous [20]. In infection experiments with three parasite species (Anguillicola crassus, Diplostomum pseudospathacaeum, Camallanus lacustris), sticklebacks with intermediate numbers of MHC IIb alleles were found to have lower parasite loads compared to individuals with more or fewer alleles [20]. In mate choice experiments, female sticklebacks favoured combinations with males, allowing optimal (i.e. intermediate) numbers of MHC IIb alleles for their offspring [17]. The decision of female stickleback was predictably modified towards the optimal number of MHC class IIb alleles using MHC class I ligand peptides [24], hinting at the similarities of peptide binding determinants of MHC class I and II molecules. So far, the investigation of MHC diversity in sticklebacks is based on structural analysis of polymorphisms in genomic and expressed DNA sequences. Studies on their functional consequences would benefit from the availability of specific antibodies against stickleback MHC. Here, we characterize antibodies recognizing stickleback MHC class I and MHC class II molecules. Antibodies were generated by immunization of rabbits with recombinant, bacterially expressed fragments of stickleback MHC class Ia, MHC class IIa and IIb. 2. Materials and methods 2.1. Fish Three spined sticklebacks (Gasterosteus aculeatus) were caught in the lake ‘‘Grosser Plo¨ner See’’ in Northern Germany in winter 2005/2006. They were all offspring from the previous breeding season in spring 2005. Fish were maintained in aquaria with continuous water supply and fed ad libitum three times a week with frozen chironomid larvae. To mimic their natural seasonal cycle, fish were kept first in winter conditions (6  C, 8 h light) then transferred for 2 weeks to spring conditions (12  C, 12 h light) and subsequently to summer conditions (18  C, 16 h light). Only fish from summer conditions were used for Western blot and histology analysis. For preparation of cDNA, fish caught in winter 2003/2004 and maintained as above were used. 2.2. Construction of expression vectors For the production of recombinant fragments of stickleback MHC, sequences were selected that, based on sequence comparisons, are expected to code for conserved, extra-cellular domains. Highly polymorphic areas, likely encoding the peptide binding grooves were excluded (Fig. 1). The MHC fragments were produced in bacteria as N-terminally His-tagged recombinant proteins, using the pQE-30 UA expression vector (Qiagen). Coding sequences were amplified by polymerase chain reaction (PCR) (Table 1) with cDNA templates generated from a stickleback head kidney. PCR products were ligated into linearized expression vector (pQE-30 UA) with the

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

993

A. Gaac MHC class I alpha ( ) UAA (MW 37.9 kD) Leader peptide 1 23 MRLVGAEISVLSLLMMSLHGAAA Alpha 1-domain 24 . . . . . . .. . . . . . . . . . . . . . . . . 112 LTHSLKYFLTASSGLPNFPEFVIVGLLDEVELFHYDGDTRRAEVRQDWMIRVRGDDPRYLKRGTEVLMDAQQVFKVNIEIAKQR FNQTG Alpha 2-domain 113 . . . . . . . . . . ... . . . . . . . . . .. . 205 GVHIFQRMVGCEWDNETNEFKGYDQFGYDGEDFISYDLQTEQCIAAKQQAVITKQKWDQDRALKAHKKNSLTHVCPESLKTLLNYGR SSLMRT Alpha 3-domain 206 ======== . . . 316 ERPSVSLLQKTPSSPVSCHATGFYPDRADLFWRKDGEELHEDVDLGEILPNHDGTFQM RVDLNLSSVPAEDWRRYDCVFQLSGVDEDIVTKLDKTRTNREK CP/TM/CY domain 317 . . . . . . . . . . . . . . . . . . . .366 PAGSTFIIIIIIAVAVLVVIIVAVVGFKVYRERNAKHSSSSSSSAVGSEL B. Gaac MHC class II alpha ( ) DAA (MW 22.9 kD) MW Leader peptide 1 23 MKTKTMMKMMVVLVLSGVFCVSA Alpha 1-domain 24 . . . .. . . . . .. . . . 105 DGEDIAITGCSDSDGEDMYGLDGEELWYADFKHGKGVKPQPSFVDPIEFQEGTYELAVGNQQICRINLKNRLKGLKDVPLEK Alpha 2-domain 106 . .. 198 DPPSSHMIYPKDGVELGEKNSLICHVTGFYPAPVTFSWTKNQENVNEGSSRNVPFPNNDGTFNQFSTLEFTPKLGDIYSCMVEHLALDHPLVS CP/TM/CY domain 199 233 QPSVGPAVFCGVGLTVGLLGVAAGTFFLIKGNECS C. Gaac MHC class II beta ( ) DAB (MW 26.0 kD) Leader peptide 1 17 MAPSFISVSLLFIGLHA Beta 1 domain 18 . . . . . . .. . . . .. . . . . . . . . 108 ADGFMMFVTDECVFNSTELKDIEFIRSSYFNKKEDTRFSSSVGKFVGFTEQGVKIAANWNKDASFLSAMKAQKEVYCLNHVPVYYTAALTK Beta 2 domain 109 203 SAEPYVRLHSETPPGGGPLSMLVCSVYDFYPKKIIVRWTRDGRPETTGVTSTDELADGDWYYQTHSHLEYTPRSGEKISCVVEHISLSKPLVTDW CP/TM/CY domain 204 249 NPSMPESERNKVAIGASGLILGLTLSLAGFIYYKRKARGRILVPSH

Fig. 1. Amino acid sequences of stickleback MHC class Ia (A), MHC class IIa (B) and MHC class IIb (C). Areas of sequences chosen for production of recombinant fragments for antibody production are indicated in bold type. N-terminal variable regions (amino acids that vary between paralogous genes are indicated by dots, for alignments see [13,16]), connecting peptide (CP), trans membrane (TM) and cytoplasmatic domains were not included in fragments for recombinant proteins. Sites of possible N-linked glycosylation are boxed. Double dash indicates putative site of interaction with the CD8 receptor. Molecular mass in kilodaltons (kDa) is calculated based on the amino acid sequence of the whole receptor molecule without leader peptide.

overhanging adenine according to the manufacturer’s instructions (Qiagen) for the expression system. Plasmids were extracted from bacteria (E. coli strain M15) and the constructs verified by DNA sequencing. Frozen stocks of transformed E. coli cells were prepared as described in the Qiagen handbook. 2.3. Expression and purification of recombinant MHC fragments For expression, a single colony was grown overnight at 37  C in LB medium (10 g l1 tryptone, 5 g l1 yeast extract, 10 g l1 sodium chloride, kanamycin 25 mg l1, ampicilin 100 mg l1) with shaking. From the overnight Table 1 Sequences and primers used for stickleback MHC expression constructsa

MHC I a MHC II a MHC II b a

Gene Gaac:

Accession: nucleotide/protein

Fw primer 50 e30

Rev primer 50 e30

Position in coding sequence

UAA DAA DAB

EF375485/ABN14358.1 AY713945/AAU01917.1 AY713945/AAU01918.1

tctctgatgagaaccgagcgt ttggagaaagatcctccttcca ctgagtgctatgaaggctcaga

catctggaaggtcccgtcgtg atggtccagggccagatgttc aaccagaggtttactcagactg

601e792 307e582 247e600

Schaschl and Wegner [16].

994

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

culture, 1 ml was transferred to 100 ml of pre-warmed TB medium (24 g l1 tryptone, 12 g l1 yeast extract, 10 g l1 sodium chloride, 4 g l1 glycerol, 25 mg l1 kanamycin, 100 mg l1 ampicilin). Cells were grown at 37  C until OD at 600 nm had reached 0.5. Expression was induced with IPTG (1 mmol l1). After 3 h at 37  C with shaking, bacteria were harvested by centrifugation (25 min, 4000  g, 4  C) and resuspended with lysis buffer (8 mol l1 urea, 100 mmol l1 NaH2PO4, 10 mmol l1 Tris). Following sonication on ice (3 min, duty cycle 10%, output level 5, Branson Ultrasonic Disrupter Sonifier W-250) lysates were centrifuged at 10 000  g for 20 min at 4  C to remove undissolved material. Recombinant His-tagged proteins in the supernatants were bound on ice to nickel resin (Quiagen) for 20 min. Nickel resin with bound His-tag proteins was loaded on centrifugation devices and washed with five columnvolumes of 8 M urea pH 8 and pH 6.3 to remove non-specific bacterial components. For subsequent immunization and coupling of recombinant proteins to NHS activated Sepharose, MHC fragments were first renatured while still bound to the nickel resin. To this end, centrifugation devices were washed twice with native buffer (2 PBS, 5 mmol l1 MgCl2, 1% v/v glycerol and 0.01% v/v TWEEN 20, 20 mmol l1 imidazol, pH 8.0), refilled with native buffer and incubated for 2 h at 4  C. After three additional washes with native buffer, recombinant proteins were eluted with 250 mmol l1 imidazol, dialyzed against native buffer (as above, without imidazol), adjusted to 200 mg l1 and stored at 80  C. In conventional dialysis against PBS, precipitation of proteins occurred below 2 mol l1 urea. Therefore, recombinant MHC fragments were renatured during the affinity purification step. On the nickel resin, recombinant molecules are presumably spatially separated enough since that rehydration was possible without association and subsequent precipitation of the recombinant proteins. During subsequent dialysis against native buffer, to remove imidazol from the eluted protein samples, precipitation did not occur. 2.4. Immunization of rabbits For each MHC molecule, two rabbits were immunized four times with each 100 mg of antigen. Immunization of rabbits, collection of pre-immune and antisera was performed by Eurogentec (Liege Science Park, Belgium) following a standard protocol (for details see: http://www.eurogentec.de/code/en/page_03_326.htm). 2.5. Affinity purification of polyclonal antisera Bacterial proteins as well as recombinant MHC fragments were coupled to NHS activated Sepharose (Amersham) at a concentration of 1 mg ml1 according to the manufacturer’s specifications. Antiserum was diluted 1:1 with double concentrated PBS, loaded on the column with bacterial proteins (E. coli M15) and incubated for 1 h at 4  C. The flow through was transferred to the column with respective recombinant MHC protein and incubated for 1 h at 4  C. After washing with 10 column-volumes of PBST (PBS 0.01% v/v Tween 20), antibodies were eluted with 1.5 columnvolumes of 100 mol l1 citric acid, pH 2.5 and neutralized immediately with 2 mol l1 Tris pH 8.5 to pH 7. Eluates were dialyzed against PBS and stored at 4  C with 0.02% w/v sodium azide until use. Pre-immune and immune sera were subjected to identical affinity purification protocols. 2.6. Preparation of lysates from stickleback organs Fish were killed by an overdose of MS 222 (Sigma). The body cavity was opened from the ventral side and head kidney, spleen, gills, brain and muscle were sampled. Tissues were transferred immediately to ice cold modified RIPA buffer (150 mmol l1 sodium chloride, 50 mmol l1 TriseHCl, pH 7.4, 1 mmol l1 EDTA, 1% v/v Triton X-100, 1% w/v sodium deoxycholic acid, 0.1% w/v SDS) supplemented with protease inhibitors (2 mg l1 aprotinin, 2 mg l1 leupeptin, 1 mmol l1 PMSF). Tissue samples were sonicated (1 min, duty cycle 10%, output level 5, Branson Ultrasonic Disrupter Sonifier W-250) on ice and centrifuged for 10 min at 4  C with 20 000  g to remove undissolved material. Aliquots of supernatants were stored at 80  C until further processing. 2.7. Glycosylation of stickleback MHC molecules Amino acid sequences of stickleback MHC were analysed for the presence of potential glycosylation sites (for N-linked glycosylation with the NetNGlyc 1.0 Server http://www.cbs.dtu.dk/services/NetNGlyc/, for O-linked

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

995

glycosylation with the NetOglyc 3.1 server http://www.cbs.dtu.dk/services/NetOGlyc/). To examine the presence of carbohydrates on stickleback MHC proteins in vivo, lysates of spleen tissues were subjected to enzymatic deglycosylation using the E-DEGLY kit (Sigma). Samples were processed under denaturing conditions according to the manufacturer’s protocol with slight modifications. Samples were digested overnight at 37  C, instead of 3 h. Enzymes (PNGase F, O-glycosidase, alpha-2(3,6,8,9) neuraminidase, beta(1e4)galactosidase, beta-N-acetylglucosaminidase) were added in combination and singly to differentiate the type of glycosylation. Control samples were subjected to mock treatment without enzymes. Glycosylation was evaluated in Western blot analysis by comparison of molecular weights of antibody labelled MHC molecules in deglycosylated and control samples. 2.8. SDS page and Western blotting of tissue samples Samples were thawed on ice and total protein content was determined colorimetrically by the Bradford method. Proteins were separated by sodium dodecyl sulphate (SDS) page gel electrophoresis [25]. Samples were boiled for 5 min in SDS sample buffer containing mercaptoethanol. From each tissue sample, 10e20 mg of total protein was loaded per slot of a pre cast 4e20% polyacrylamide continuous gradient gel (Cambrex, Rockland, USA). For determination of protein mass, biotinylated recombinant proteins (Strep-tagÒ, BioRad) were used as molecular weight standards. Following electrophoresis proteins were transferred to PVDF membranes. Membranes were blocked with BSA, 3% w/v in TBST buffer (40 mmol l1 TriseHCl, 10 mmol l1 Tris base, 150 mmol l1 sodium chloride, 0.1% w/v Tween 20) and washed with TBST. Then membranes were incubated with affinity purified rabbit anti stickleback MHC antibodies (1:20 to 1:100) in TBST/BSA for 30 min at room temperature. After washing with TBST, membranes were incubated with horseradish peroxidase (HRP) conjugated, polyclonal goat anti rabbit antibodies (Sigma). The part of the membrane containing molecular weight markers was incubated separately with a streptavidin HRP conjugate (Strep-Tactin-HRP, BioRad). Appropriate dilutions for primary and secondary antibodies were determined empirically. Binding of antibodies and Strep-Tactin-HRP was visualized by enhanced chemiluminescence (ECL) with Chemi-glow (Alpha Innotech) as HRP substrate. Membranes were developed up to 15 min with a digital luminescence camera system. Blots were evaluated using the FluorChem (Alpha Innotech) software. 2.9. Immunohistology Fish were killed by an overdose of MS 222 (Sigma). Tissue samples were fixed in 4% w/v paraformaldehyde in phosphate buffered saline (PBS) for 15 min at room temperature and stored for 24 h at 4  C. Fixed tissues were embedded in paraffin using standard procedures. Sections of 4 mm were cut and mounted on glass slides coated with protein glycerol (ROTH, Germany), dried overnight at 37  C and stored at room temperature. Tissue sections were de-waxed with Roticlear (ROTH, Germany) and rehydrated in graded ethanol baths. For antigen retrieval sections were placed in 0.01 mol l1 citrate buffer pH 6.0, placed for 30 min in a 120  C oven, cooled to room temperature and transferred to PBS. To prevent non-specific binding of secondary antibody, sections were incubated for 40 min at 37  C with normal goat serum (1:50 with 3% w/v BSA in PBS). Sections were washed with PBS and incubated for 60 min at room temperature with appropriate dilutions of affinity purified pre-immune or post-immune sera from the respective rabbit in 3% w/v BSA in PBS. After washing with PBS, sections were incubated for 60 min at room temperature with green fluorescent polyclonal goat-anti-rabbit antibodies (Alexa 488, Molecular Probes) 1:200 in 3% w/v BSA in PBS. Sections were washed with PBS and counterstained with propidium iodide (1 mg l1) in anti-fading mounting medium (Fluka, Germany) for fluorescence microscopic analysis. 3. Results Fragments of stickleback MHC molecules were produced as N-terminal histidin tagged proteins with the pQE-30 expression vector (Quiagen) in bacteria. Fragments of the MHC class I alpha chains (MHC Ia), the MHC class II alpha (MHC IIa) and beta (MHC IIb) were produced from conserved extra-cellular parts of the molecules (Fig. 1 in bold type). Selected parts of the molecules show high sequence identity of paralogous genes (e.g. DAA-DBA of MHC IIa). Antibodies raised against these conserved regions of the proteins are expected to detect products of paralogous genes with comparable intensity (for alignments see [13,16]). Polymorphic areas coding for the peptide binding groove and

996

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

sites of interaction with the T-cell receptor were omitted, to avoid allele specific binding effects of antisera and individual differences between fish with different MHC genotypes. Recombinant MHC fragments were isolated from the bacteria under denaturing conditions with urea (8 mol l1), as the majority of the recombinant proteins were found in inclusion bodies and then renatured for further use. A two step affinity purification protocol was applied to optimize the purity of the antibody preparations. The resulting antibody preparations did not show background labelling in Western blot analysis of recombinant MHC preparations (Fig. 2). In Western blot analysis of stickleback organs, MHC molecules were detected as single bands in gill, spleen and head kidney, faintly in muscle, but not in brain (Fig. 3AeC). Protein bands in muscle samples at about 40 kDa and low molecular weight bands in gills and spleen, are not specific, as they were also observed when only the secondary antibody was used (Fig. 3D). When pre-immunization sera were used instead of primary antibodies, faint bands identical to those seen with the secondary antibody alone were observed, but neither recombinant MHC fragments (Fig. 2) nor MHC bands detected in tissue lysates were stained (data not shown). The calculated mass of stickleback MHC class II alpha (MHC IIa) (Gaac-DAA) without leader peptide (22.86 kDa) corresponded to the measured mass of the protein band (23 kDa) detected with antiserum against MHC IIa in Western blot analysis (Fig. 3B). Protein bands of MHC class I alpha (MHC Ia, 44 kDa, Fig. 3A) and MHC class II beta (MHC IIb, 31 kDa, Fig. 3C) showed higher molecular mass in Western blot analysis as compared to calculated masses from the amino acid sequence (MHC Ia: 37.92, MHC IIb: 26.03 kDa, compare Fig. 1), presumably owing to their glycosylation. After deglycosylation of N-linked sugar residues with PNGase F in spleen lysates, bands stained with MHC Ia and MHC IIb antisera showed reduced molecular masses in SDS-PAGE electrophoresis (Fig. 4). Parallel samples developed with MHC IIa antiserum did not show a shift in migration for the strongly stained band. This is in line with absence of glycosylation sites in the MHC IIa sequence (Gaac-DAA, Fig. 1). However, in individual fish a cross-reactive band is apparently reduced in size (arrow in Fig. 4). This likely corresponds to the Gaac-DBA sequence, as this contains a predicted N-glycosylation site at position 149. This observation was not consistent for all stickleback tested here (not shown), suggesting fish to fish variation in Gaac-DBA expression. Furthermore, the calculated molecular mass of Gaac-DBA without leader peptide is 23.9 kDa. This could explain the faint band above the MHC IIa main band, but not the band below the main band after deglycosylation (arrow Fig. 4). O-linked glycosylation was not detected on any of the MHC molecules. Analysis of amino acid sequences predicted possible N-linked glycosylation of MHC Ia (Gaac-UAA) at positions 109, 127 and 268 and of MHC IIb, Gaac-DAB and Gaac-DBB (not shown), at positions 32 and 204 (Fig. 1). Sections of formalin fixed, paraffin embedded tissues from different organs were labelled with affinity purified antisera against MHC Ia, MHC IIa and MHC IIb. Antibody binding was visualized by a green fluorescent conjugate in rMHC llα

rMHC lα i.

pi.

i.

rMHC llβ pi.

i.

pi.

15.7kD 12.7kD 9 .6kD

Fig. 2. Western blot analysis of recombinant MHC fragments. Recombinant proteins (1 ng per lane) were subjected to SDS-PAGE electrophoresis and transferred to PVDF membranes. Membranes were labelled with affinity purified immune (i.) and pre-immune (pi.) sera of the respective rabbit. Antibody binding was visualized by enhanced chemiluminescense (ECL).

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

Br

A

G

S

M

997

Hk

MHC Iα

44kD 31kD 23kD

B

MHC llα

44kD 31kD 23kD

C

MHC llβ

44kD 31kD 23kD

D

Control

44kD 31kD 23kD

Fig. 3. Western blot analysis of MHC molecules in stickleback organs. Tissue lysates (Br, brain; G, gill; S, spleen; M, muscle; Hk, head kidney), 20 mg protein per lane, were separated on 4e20% polyacrylamide gels by SDS-PAGE electrophoresis and transferred to PVDF membranes. Membranes were labelled with affinity purified immune sera against MHC class Ia (A), MHC class IIa (B) and MHC class IIb (C) as primary antibody and with secondary antibody alone (D). Antibody binding was visualized by enhanced chemiluminescense (ECL). Note: absence of MHC bands in brain, faint MHC IIa signal in muscle and 35 kDa band in muscle samples and low molecular weight bands (<23 kDa) were visualized with the secondary antibody alone (D).

998

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002 MHC lα +

MHC llβ

MHC llα -

+

-

+

-

44kD 31kD 23kD

Fig. 4. Western blot analysis of MHC molecules in stickleback spleen after deglycosylation. Lysates of spleen tissue were incubated with (þ) and without () PNGase-F for N-linked deglycosylation. Protein samples (15 mg protein per lane) were separated on a 4e20% polyacrylamide gel by SDS-PAGE electrophoresis and transferred to PVDF membranes. Membranes were labelled with affinity purified immune sera against MHC Ia, IIa and IIb. Antibody binding was visualized by enhanced chemiluminescense (ECL). Note: upon deglycosylation (þ), MHC class I alpha and II beta showed reduced molecular weight compared to controls (), while in MHC II alpha only a small fraction of protein (arrow) was susceptible to PNGase-F digestion.

sections counterstained with propidium iodide (red fluorescence) (Fig. 5). Neuronal cells in the brain were not stained by any of the tested antisera (Fig. 5, IeIIIa). In gill sections, epithelial cells of secondary lamellae were stained intensely with anti MHC Ia antibodies (Fig. 5, Ib). With anti-MHC IIa (Fig. 5, IIb) and MHC IIb antiserum (Fig. 5, IIIb), single cells stained positive in gill sections (putative residual macrophages, Fig. 5, IIeIIIb). In spleen tissue, MHC Ia antisera stained the majority of cells (Fig. 5, Ic), but cells positive for MHC IIa and IIb (putative antigen presenting mononuclear cells) were distributed in the spleen tissue (Fig. 5, IIeIIIc). Myocytes (M) were not stained with anti-MHC Ia, anti-MHC IIa and IIb antisera (Fig. 5, IeIIId). In muscle tissue, putative endothelial cells of capillaries in the perimysium stained positive for MHC Ia (arrow in Fig. 5, Id). In sections of head kidneys, considerable numbers of cells showed distinct labelling with anti MHC Ia antiserum (Fig. 5, Ie), but with anti MHC IIa and IIb antisera relatively few cells stained positive (Fig. 5, IIeIIIe). 4. Discussion In the present study, polyclonal antisera produced against MHC class I a3, MHC class II a2 and MHC class II b2 chains of the three spined stickleback (Gasterosteus aculeatus) were characterized. In Western blot analysis of organ lysates, affinity-purified antisera reacted with protein fractions of the appropriate molecular mass expected for the respective MHC molecule (Fig. 3). While MHC IIa matched the molecular mass calculated from the amino acid sequence precisely, MHC Ia and MHC IIb showed elevated molecular mass in Western blot analysis compared to the calculated values, most probably due to N-linked glycosylation. In Western blot analysis, distinct staining of all tested MHC molecules was observed in gill tissue and spleen, less prominent in the head kidney; it was almost absent in brain and muscle (Fig. 3). Levels of MHC expression in individual organs depend on the type and number of MHC expressing cells and their state of activation [26,27]. Frequency and activation of MHC expressing cells in the respective organs appear to be regulated according to the type of invading pathogens [28e32]. In amoebic gill disease of Atlantic salmon (Salmo salar), numerous MHC IIb positive cells were detected within gill lesions, indicative of immune cell trafficking [33]. In laboratory-raised sticklebacks, experimentally infected with three naturally occurring parasite species, expression of MHC IIb coding messenger RNA (mRNA) relative to beta actin, was highest in gills, intermediate in spleen and low in the head kidney [22]. The sticklebacks investigated in the present study were caught in the wild and had been in contact with natural pathogens. These sticklebacks too showed comparably low MHC expression in head kidneys, but high expression in gills and spleen in Western blot analysis. With the present study we did not investigate the extent of intra- and extra-cellular binding of the antibodies. Initially cryostat sections of tissues were used for antibody staining, but only low binding intensities were observed,

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

999

Fig. 5. MHC class I and class II positive reaction shows green. (I) MHC class I alpha, (II) MHC class II alpha, (III) MHC class II beta, (a) brain, (b) gill tissue with primary (PL) and secondary (SL) lamellae and goblet cell (G), (c) spleen, (d) muscle, myocyte (M), (e) head kidney, bar 10 mm; arrows, MHC positive cells.

suggesting that native MHC molecules are insufficiently detected by the antibodies produced here. Permeabilization of cryostat sections to demonstrate potential intra-cellular binding of the antibodies was not done. However, distinct labelling of individual cells was detectable on sections of formaldehyde fixed, paraffin embedded tissues after antigen retrieval with citrate buffer. In immunohistology, MHC positive cells were detected in gills, spleen and head kidneys

1000

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

(Fig. 5). In gill sections, epithelia covering the primary and secondary lamella stained intensively with anti MHC Ia. In rainbow trout (Oncorhynchus mykiss) gill epithelial cells also stained positive for MHC class I [34]. With MHC IIa and IIb antisera, individual cells were stained positive in primary and secondary lamellae in gill sections. Positive cells located at the outer surface of the secondary lamellae might be epithelial cells, as observed in Atlantic salmon [33,35]. Positive cells scattered in the primary lamellae might be residual macrophages and leukocytes from blood. In spleen, the majority of cells stained with MHC Ia antiserum while scattered individual cells stained positive for MHC IIa and IIb. A comparable abundance of positive cells with either MHC antiserum was observed in the head kidney. In brain tissue, MHC molecules of either type were not detectable, either in protein lysates (Fig. 3), or on neuronal cells in histological sections (Fig. 5). In mammals, MHC expression is absent in neuronal tissue of a healthy central nervous system distal from the blood/brain barrier [36,37]. Quantitative PCR indicates that MHC expression is low in the brain tissue of teleosts as well [22,27,38e40]. In immunohistological analyses of sections of brain from rainbow trout, neuronal cells were MHC class I negative, only endothelial cells of blood vessels and macrophages lining the sub epithelial capillaries stained positive [34]. In support of these earlier findings, in the present study MHC IIa and IIb molecules were not detected in neuronal cells in brain sections from stickleback. The observations described above, together with the present data, suggest that in the brain of teleosts, comparable to mammals, expression of MHC class I and class II is absent distal from the brain/blood barrier. In immunohistology, MHC was not detected in myocytes, but in lysates of muscle tissue of stickleback, faint bands of MHC IIa and IIb were observed. Similarly, in protein lysate from muscle tissue of rainbow trout (Oncorhynchus mykiss) a faint MHC IIb band was present and it was suggested that this was due to co-isolation of interstitial macrophages [41]. In lysates from muscle tissue of Atlantic salmon, MHC IIb was not detected [35]. However, the very low or absent MHC staining in muscle and brain samples, respectively, confirms the specificity of the tested antibody preparations as in both organs low or absent MHC expression is expected. Additional indication for the specificity of the produced antisera was provided by analysis of glycosylation of MHC molecules in stickleback. Among vertebrates particular sites of glycosylation of MHC class I and class II molecules are fairly conserved [42]. In several teleost species, including stickleback, conserved N-glycosylation sites are predicted in MHC class I [16,43,44]. In the present study, N-linked glycosylation of stickleback MHC Ia is suggested by deglycosylation experiments. Glycosylation of MHC class II in teleosts is in accordance with other vertebrates for class IIb but deviates in class IIa. In the MHC IIb chain of many teleosts N-glycosylation was predicted [45e47] and confirmed by deglycosylation experiments [35,41,48e50]. In stickleback N-linked glycosylation sites are predicted at positions 32 and 204 in both MHC IIb genes (Gaac-DAB, DBB). Here we confirm N-linked glycosylation of stickleback MHC IIb molecules. For MHC IIa molecules in teleosts both glycosylated and non-glycosylated forms are described. In MHC IIa of gilthead sea bream (Sparus aurata) and rainbow trout (O. mykiss) N-linked glycosylation was identified [27,50]. In contrast, channel catfish (Ictalurus punctatus) MHC IIa molecules lack sites of glycosylation [46,48]. In stickleback two genes (Gaac-DAA and -DBA) coding for MHC IIa molecules are described [13], only one of which (Gaac-DBA) contains a predicted site for N-linked glycosylation. In the present study, the majority of stickleback MHC IIa was not glycosylated; only a small fraction of the protein was susceptible to digestion with N-glycosidase (Fig. 4). This finding suggests that in stickleback spleen mainly Gaac-DAA is expressed and Gaac-DBA to a minor degree only. A glycosylated fraction of MHC IIa (putative Gaac-DBA) was not detectable in spleen samples from every stickleback tested here suggesting individual differences in Gaac-DBA expression. However, the deduced molecular mass of nonglycosylated Gaac-DBA (23.9 kDa) is higher than that of Gaac-DAA (22.9 kDa). This could explain the faint band (putative Gaac-DBA) above the main band (Gaac-DAA) which is susceptible to deglycosylation, but is contradictory to the assumption that the faint MHC IIa fraction occurring below 23 kDa after deglycosylation is Gaac-DBA. At the moment it is not possible to explain why in individual sticklebacks fractions of MHC IIa proteins are glycosylated. Current available sequence information for MHC IIa genes in stickleback is based on one individual only [13]. It is suggested that copy number variation of MHC genes in stickleback could be causative for the observed individual differences. In channel catfish, lack of MHC IIa glycosylation is suggested to interfere with association of the molecule with the chaperone calnexin (CNX), and assembly of the molecule thus might proceed via different pathways compared to mammals [48]. Sticklebacks appear to have both glycosylated and non-glycosylated MHC IIa molecules. Further investigation of differential gene expression of the two types of MHC IIa in stickleback may contribute to understanding of the function of non-glycosylated MHC IIa molecules in teleosts.

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

1001

The results from glycosylation analysis, molecular mass measurements and organ specific MHC labelling confirm that antibodies developed here are detecting stickleback MHC class I and class II molecules in Western blot analyses and immunohistology. In future experiments, MHC antibodies will be used in combination with quantitative PCR, to understand if tissue specific up-regulation of MHC expression is due to invading MHC positive cells or if single cells express more of these molecules. Furthermore, antibodies may be used to detect functional interactions between MHC, mate choice and parasite resistance. Acknowledgements The authors are grateful to Professor T. Boehm, Max-Planck Institute for Immunobiology, Freiburg, Germany, for thoughtful suggestions during the progress of the project and for the critical review of the manuscript. Furthermore we want to thank Professor M. Milinski, Max Planck Institute for Limnology, Plo¨n, Germany, for the support of the project. For technical assistance we are grateful to V. Bu¨scher, H. Luttman, A. Hasselmeyer, A. Busekow and S. Arndt. References [1] Stet RJ, Kruiswijk CP, Saeij JP, Wiegertjes GF. Major histocompatibility genes in cyprinid fishes: theory and practice. Immunol Rev 1998;166:301e16. [2] Stet RJ, Kruiswijk CP, Dixon B. Major histocompatibility lineages and immune gene function in teleost fishes: the road not taken. Crit Rev Immunol 2003;23:441e71. [3] Kaufman J, Skjoedt K, Salomonsen J. The MHC molecules of nonmammalian vertebrates. Immunol Rev 1990;113:83e117. [4] Hughes AL. Natural selection and the diversification of vertebrate immune effectors. Immunol Rev 2002;190:161e8. [5] Fischer U, Utke K, Somamoto T, Kollner B, Ototake M, Nakanishi T. Cytotoxic activities of fish leucocytes. Fish Shellfish Immunol 2006;20:209e26. [6] Germain RN. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 1994;76:287e99. [7] Cresswell P. Assembly, transport, and function of MHC class II molecules. Annu Rev Immunol 1994;12:259e93. [8] Ahmed R, Gray D. Immunological memory and protective immunity: understanding their relation. Science 1996;272:54e60. [9] Secombes CJ, Bird S, Zou J. Adaptive immunity in teleosts: cellular immunity. Dev Biol (Basel) 2005;121:25e32. [10] Sato A, Figueroa F, O’hUigin C, Steck N, Klein J. Cloning of major histocompatibility complex (Mhc) genes from threespine stickleback, Gasterosteus aculeatus. Mol Mar Biol Biotechnol 1998;7:221e31. [11] Sato A, Figueroa F, Murray BW, Malaga-Trillo E, Zaleska-Rutczynska Z, Sultmann H, et al. Nonlinkage of major histocompatibility complex class I and class II loci in bony fishes. Immunogenetics 2000;51:108e16. [12] Reusch TB, Wegner KM, Kalbe M. Rapid genetic divergence in postglacial populations of threespine stickleback (Gasterosteus aculeatus): the role of habitat type, drainage and geographical proximity. Mol Ecol 2001;10:2435e45. [13] Reusch TB, Schaschl H, Wegner KM. Recent duplication and inter-locus gene conversion in major histocompatibility class II genes in a teleost, the three-spined stickleback. Immunogenetics 2004;56:427e37. [14] Reusch TB, Langefors A. Inter- and intralocus recombination drive MHC class IIB gene diversification in a teleost, the three-spined stickleback Gasterosteus aculeatus. J Mol Evol 2005;61:531e41. [15] Schaschl H, Wegner KM. Polymorphism and signature of selection in the MHC class I genes of the three-spined stickleback Gasterosteus aculeatus. J Fish Biol 2006;69:177e88. [16] Schaschl H, Wegner KM. Contrasting mode of evolution between the MHC class I genomic region and class II region in the three-spined stickleback (Gasterosteus aculeatus L.; Gasterosteidae: Teleostei). Immunogenetics 2007;59:295e304. [17] Reusch TB, Haberli MA, Aeschlimann PB, Milinski M. Female sticklebacks count alleles in a strategy of sexual selection explaining MHC polymorphism. Nature 2001;414:300e2. [18] Kurtz J, Kalbe M, Aeschlimann PB, Haberli MA, Wegner KM, Reusch TB, et al. Major histocompatibility complex diversity influences parasite resistance and innate immunity in sticklebacks. Proc R Soc B 2004;271:197e204. [19] Kurtz J, Wegner KM, Kalbe M, Reusch TB, Schaschl H, Hasselquist D, et al. MHC genes and oxidative stress in sticklebacks: an immunoecological approach. Proc R Soc B 2006;273:1407e14. [20] Wegner KM, Kalbe M, Kurtz J, Reusch TB, Milinski M. Parasite selection for immunogenetic optimality. Science 2003;301:1343. [21] Wegner KM, Reusch TB, Kalbe M. Multiple parasites are driving major histocompatibility complex polymorphism in the wild. J Evol Biol 2003;16:224e32. [22] Wegner KM, Kalbe M, Rauch G, Kurtz J, Schaschl H, Reusch TB. Genetic variation in MHC class II expression and interactions with MHC sequence polymorphism in three-spined sticklebacks. Mol Ecol 2006;15:1153e64. [23] Milinski M. The major histocompatibility complex, sexual selection, and mate choice. Annu Rev Ecol Evol Syst 2006;37:159e86. [24] Milinski M, Griffiths S, Wegner KM, Reusch TB, Haas-Assenbaum A, Boehm T. Mate choice decisions of stickleback females predictably modified by MHC peptide ligands. Proc Natl Acad Sci USA 2005;102:4414e8. [25] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680e5.

1002

J.P. Scharsack et al. / Fish & Shellfish Immunology 23 (2007) 991e1002

[26] Rodrigues PNS, Hermsen TT, Rombout JHWM, Egberts E, Stet RJM. Detection of MHC class II transcripts in lymphoid tissues of the common carp (Cyprinus carpio L.). Dev Comp Immunol 1995;19:483e96. [27] Cuesta A, Angeles Esteban M, Meseguer J. Cloning, distribution and up-regulation of the teleost fish MHC class II alpha suggests a role for granulocytes as antigen-presenting cells. Mol Immunol 2006;43:1275e85. [28] Sigh J, Lindenstrom T, Buchmann K. The parasitic ciliate Ichthyophthirius multifiliis induces expression of immune relevant genes in rainbow trout, Oncorhynchus mykiss (Walbaum). J Fish Dis 2004;27:409e17. [29] Tafalla C, Coll J, Secombes CJ. Expression of genes related to the early immune response in rainbow trout (Oncorhynchus mykiss) after viral haemorrhagic septicemia virus (VHSV) infection. Dev Comp Immunol 2005;29:615e26. [30] Chen SL, Zhang YX, Xu MY, Ji XS, Yu GC, Dong CF. Molecular polymorphism and expression analysis of MHC class II B gene from red sea bream (Chrysophrys major). Dev Comp Immunol 2006;30:407e18. [31] Zhang YX, Chen SL. Molecular identification, polymorphism, and expression analysis of major histocompatibility complex class IIA and B genes of turbot (Scophthalmus maximus). Mar Biotechnol (NY) 2006;8:611e23. [32] Bridle AR, Morrison RN, Nowak BF. The expression of immune-regulatory genes in rainbow trout, Oncorhynchus mykiss, during amoebic gill disease (AGD). Fish Shellfish Immunol 2006;20:346e64. [33] Morrison RN, Koppang EO, Hordvik I, Nowak BF. MHC class IIþ cells in the gills of Atlantic salmon (Salmo salar L.) affected by amoebic gill disease. Vet Immunol Immunopathol 2006;109:297e303. [34] Dijkstra JM, Kiryu I, Kollner B, Yoshiura Y, Ototake M. MHC class II invariant chain homologues in rainbow trout (Oncorhynchus mykiss). Fish Shellfish Immunol 2003;15:91e105. [35] Koppang EO, Hordvik I, Bjerkas I, Torvund J, Aune L, Thevarajan J, et al. Production of rabbit antisera against recombinant MHC class II beta chain and identification of immunoreactive cells in Atlantic salmon (Salmo salar). Fish Shellfish Immunol 2003;14:115e32. [36] Minagar A, Alexander JS. Blood-brain barrier disruption in multiple sclerosis. Mult Scler 2003;9:540e9. [37] Bailey SL, Carpentier PA, McMahon EJ, Begolka WS, Miller SD. Innate and adaptive immune responses of the central nervous system. Crit Rev Immunol 2006;26:149e88. [38] Hansen JD, Strassburger P, Du Pasquier L. Conservation of an alpha 2 domain within the teleostean world, MHC class I from the rainbow trout Oncorhynchus mykiss. Dev Comp Immunol 1996;20:417e25. [39] Koppang EO, Press CML, Rønningen K, Lie Ø. Expression of Mhc class I mRNA in tissues from vaccinated and non-vaccinated Atlantic salmon (Salmo salar L.). Fish Shellfish Immunol 1998;8:577e87. [40] Koppang EO, Lundin M, Press CML, Rønningen K, Lie Ø. Differing levels of Mhc class II b chain expression in a range of tissues from vaccinated and non-vaccinated Atlantic salmon (Salmo salar L.). Fish Shellfish Immunol 1998;8:183e96. [41] Van Lierop MJC, Knight J, Secombes CJ, Hermsen TT, Groeneveld A, Stet RJM. Production and characterization of an antiserum raised against recombinant rainbow trout (Oncorhynchus mykiss) MHC class II beta-chain (MhcOnmy-Dab). Fish Shellfish Immunol 1998;8:231e43. [42] Kaufman J, Salomonsen J, Flajnik M. Evolutionary conservation of MHC class I and class II molecules e different yet the same. Semin Immunol 1994;6:411e24. [43] Antao AB, Chinchar GV, McConnell TJ, Miller NW, Clem WL, Wilson MR. MHC class I genes of the channel catfish: sequence analysis and expression. Immunogenetics 1999;49:303e11. [44] Fujiki K, Booman M, Chin-Dixon E, Dixon B. Cloning and characterization of cDNA clones encoding membrane-bound and potentially secreted major histocompatibility class I receptors from walleye (Stizostedion vitreum). Immunogenetics 2001;53:760e9. [45] McConnell TJ, Godwin UB, Norton SF, Nairn RS, Kazianis S, Morizot DC. Identification and mapping of two divergent, unlinked major histocompatibility complex class II B genes in Xiphophorus fishes. Genetics 1998;149:1921e34. [46] Godwin UB, Flores M, Quiniou S, Wilson MR, Miller NW, Clem LW, et al. MHC class II A genes in the channel catfish (Ictalurus punctatus). Dev Comp Immunol 2000;24:609e22. [47] Kruiswijk CP, Hermsen T, Fujiki K, Dixon B, Savelkoul HF, Stet RJ. Analysis of genomic and expressed major histocompatibility class Ia and class II genes in a hexaploid Lake Tana African ‘large’ barb individual (Barbus intermedius). Immunogenetics 2004;55:770e81. [48] Fuller JR, Pitzer JE, Godwin U, Albertino M, Machon BD, Kearse KP, et al. Characterization of the molecular chaperone calnexin in the channel catfish, Ictalurus punctatus, and its association with MHC class II molecules. Dev Comp Immunol 2004;28:603e17. [49] Thankappan A, Fuller JR, Godwin UB, Kearse KP, McConnell TJ. Characterization of glycans on major histocompatibility complex class II molecules in channel catfish, Ictalurus punctatus. Dev Comp Immunol 2006;30:772e82. [50] Nath S, Kales S, Fujiki K, Dixon B. Major histocompatibility class II genes in rainbow trout (Oncorhynchus mykiss) exhibit temperature dependent downregulation. Immunogenetics 2006;58:443e53.