C2-A expressed in different tissues

Developmental and Comparative Immunology 26 (2002) 533±541 www.elsevier.com/locate/devcompimm

Diversity of complement factor B/C2 in the common carp (Cyprinus carpio): three isotypes of B/C2-A expressed in different tissues q Miki Nakao a,*, Momoe Matsumoto a, Mika Nakazawa a, Kazuhiro Fujiki a,b, Tomoki Yano a a

Laboratory of Marine Biological Chemistry, Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University, Hakozaki, Fukuoka 812 8581, Japan b Department of Biology, University of Waterloo, 200 University Avenue West,Waterloo, Ont., Canada N2L 3G1 Received 16 July 2001; revised 4 September 2001; accepted 10 September 2001

Abstract Complement factor B and C2 are two critical proteases for complement activation. Some bony ®sh have been reported to possess duplicated genes for B/C2, but there is no direct evidence regarding possible functional divergence. Here, we report the isolation of the second and third isotypes of carp B/C2-A, a close relative of other bony ®sh B reported to date, and designated B/C2-A2 and B/C2-A3. B/C2-A1 (previously reported B/C2-A) and B/C2-A2 share 78% amino acid identity and are synthesized mainly in hepatopancreas. On the other hand, B/C2-A3 showed less (,60%) sequence identity with the other two isotypes. It was expressed mainly in kidney and spleen, and was up-regulated after injection of carp with scleroglucan or sodium alginate, known immunostimulants for ®sh. Phylogenetic analysis suggests that B/C2-A3 diverged before separation of carp and zebra®sh. B/C2-A3 represent a novel B/C2-lineage functioning as an acute phase reactant in cyprinid ®sh. q 2002 Published by Elsevier Science Ltd. Keywords: Complement factor B; Complement factor C2; Gene duplication; Diversity; Expression; Cyprinus carpio

1. Introduction

q

The nucleotide sequence data reported in this paper will appear in the DDBJ, EMBL and GenBank nucleotide sequence databases with the following accession numbers: AB021177 (B/C2-A2) and AB047361 (B/C2-A3). * Corresponding author. Tel.: 181-92-642-2896; fax: 181-92642-2894. E-mail address: [email protected] (M. Nakao). Abbreviations: SCR, short consensus repeat; DIG, digoxigenin; SDS, sodium dodecylsulfate; PCR, polymerase chain reaction; RT, reverse transcription; SSC, standard saline citrate (10 mM sodium citrate, 150 mM NaCl, pH 7.0)

The complement system is a major humoral component of immunity. It is composed of about 30 soluble and membrane-associated proteins. Three pathways of complement activation have been recognized: an antibody-dependent classical pathway, an antibody-independent alternative pathway and a lectin pathway initiated by mannose-binding lectin. Activation of each pathway involves formation of a protease complex, termed C3-convertase, which activates the central component C3 [1]. Complement factor B and C2 are serine protease precursors. Upon activation,

0145-305X/02/$ - see front matter q 2002 Published by Elsevier Science Ltd. PII: S 0145-305 X(01)00 083-0

534

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

each becomes a catalytic subunit of the C3-convertase in the alternative or classical activation pathways, respectively. B and C2 show a signi®cant overall sequence similarity, sharing the same domain structure: three short consensus repeat (SCR) modules, a von Willebrand factor domain and a serine protease domain, from N-terminus in this order [2]. Thus they are considered to have been arisen from a common ancestor, which was probably more B-like, though there is no direct evidence available to specify when B and C2 diverged [3]. Bony ®sh possess all the three activation pathways of complement, and B/C2-like cDNA sequences have been isolated from zebra®sh (Brachydanio rerio) [4], medaka ®sh (Oryzias latipes) [5], rainbow trout (Oncorhynchus mykiss) [6] and carp (Cyprinus carpio) [7]. They are equally similar to mammalian B and C2, making it dif®cult to assign them to B or C2, and thus they are termed `B/C2' in this paper. The zebra®sh, medaka ®sh and trout B/C2 (Bf) sequences closely resemble carp B/C2-A, one of the two diverged carp B/C2-like sequences. Another carp sequence, B/C2-B, has no close relative [7]. Interestingly, rainbow trout, carp and zebra®sh have duplicated B/C2 genes, presumably because of the tetraploid nature of their genomes [6±8]. However, the biological signi®cance of B/C2-diversity is unknown. In the present study, we isolated two additional carp B/C2-like cDNA clones, one from a hepatopancreas cDNA library and another from a peritoneal exudate leukocyte cDNA library. Phylogenetic analysis suggests that, the latter represents a novel B/C2-lineage in cyprinid ®sh. Furthermore, the expression of this novel B/C2-isoform was fairly restricted to leukocyte-rich organs such as kidney and spleen and behaved like an acute phase protein.

2. Materials and methods 2.1. Materials Reagents were obtained from the following sources: restriction enzymes and ISOGEN reagent, Nippon Gene (Osaka, Japan); molony murine leukemia virus (MMLV)±reverse transcriptase and oligo(dT) primer, Lifetech Oriental (Tokyo, Japan); PCR±DIG Probe Synthesis Kit, DIG Blocking Reagent, and

alkaline phosphatase-conjugated anti-DIG antibody (Fab fragment), Boehringer Mannheim Japan (Tokyo, Japan); scleroglucan (fungus-derived b-1,3-glucan) [9], Taito (Tokyo, Japan); sodium alginate (low viscosity) [10], Sigma (St. Louis, MS). Synthetic oligonucleotides were obtained from Hokkaido System Science (Sapporo, Japan). 2.2. Library screening A cDNA library from carp hepatopancreas was constructed using lZAP II vector [7] and a leukocyte cDNA library was prepared from pooled mRNA of the head kidney and peritoneal exudate cells of carp after in vivo stimulation with sodium alginate and scleroglucan using ZAP expression vector [11]. Both libraries were screened by plaque hybridization with DIG-labeled probes, essentially as described previously [7]. Positive clones were plaque-puri®ed and converted to plasmids by in vivo excision according to Stratagene's instructions. 2.3. Obtaining nucleotide sequences The nucleotide sequences were determined for both strands by the dideoxy chain-termination method [12] using a model 377 sequencer (Applied Biosystems, Foster City, CA). 2.4. Southern hybridization Genomic DNA (5 mg) isolated from carp erythrocytes as described elsewhere [13] was digested to completion at 37 8C for 16 h with 50 units of PstI. The digests were electrophorezed on a 1% agarose gel and transferred to a Hybond-N 1 membrane (Amersham Pharmacia Biotech Japan, Tokyo, Japan), followed by UV-crosslinking. After prehybridization for 2 h at 42 8C in a solution containing 50% formamide, 5 £ SSC, 7% SDS, 0.1% sodium N-lauroylsarcosinate, 50 mM sodium phosphate buffer (pH 7.0), 2% DIG Blocking Reagent, and 50 mg/ml salmon sperm DNA, the membranes were hybridized at 42 8C for 16 h with DIG-labeled DNA probes prepared using the PCR±DIG Probe Synthesis Kit. The membranes were then washed twice at 68 8C for 30 min in 0.1 £ SSC containing 0.1% SDS, and subjected to chemiluminescent detection with alkaline phosphataseconjugated anti-DIG Fab (Boeringer-Manheim,

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

535

Table 1 PCR primers used in this study Name

Sequence

Target sequence (nucleotide position a)

P1 b P2 c P3 b P4 c P5 b P6 c P7 b P8 c P9 b P10 c P11 b P12 c P13 b P14 c

GATCCTGATGTTTGCTAC CTCTGTGACACTAACGG TACGAAGTCTCCCCAAAC TGTCTCCTGCAAATCCTC ATCGAGATGCTTGTGTTGAA GTTTTCTCAGCCCGCACT ATCGGGATGCTTGTGTTGAA AGTCTGCTAAAGCATTCACA GTGTGACTCGCAACAATG TTTCACGGTGTCTGCAC GCCTTGTACCAGTTCGTC ATCTGTCTGCGAGTCACA TACAGAACGAGAGGGCTTATC TTGGTGACCTTCAGGACATTG

A1 (888±905), A2 (900±917) A1 (1273±1289), A2 (1294±1310) A3 (850±867) A3 (1321±1338) A1 (1907±1926) A1 (2278±2295) A2 (1937±1956) A2 (2286±2305) A3 (1454±1471) A3 (1829±1845) B (2007±2024) B (2380±2397) S11 d (11±31) S11 d (423±443)

a b c d

Nucleotide number starting from the putative initiation codon. Sense strand primer. Antisense strand primer. S11 subunit of 40S ribosomal protein.

Tokyo, Japan) and Lumiphos Plus (Wako Pure Chemical, Osaka, Japan). To check the cross-hybridization of the probes, 10 pg of linearized plasmids containing B/C2-A1 (in pBluescript SK), B/C2-A2 (in pBluescript SK) and B/C2-A3 (in pBK-CMV) were used as internal standards, where SmaI-digested B/C2-A1, B/C2-A2 and B/C2-A3 should give bands of 5.4, 5.1 and 6.8 kbp, respectively. 2.5. RNA preparation Three experimental ®sh groups, each consisting of three carp (about 30 g body weight), were acclimated for more than two weeks in 60 l aquaria kept at 20 8C. Fishes in each group were intraperitoneally injected with scleroglucan or sodium alginate at a dose of 2 mg/100 g-body weight, using 20 mg/ml solutions in 0.9% NaCl (saline) for both stimulants. Control ®shes were given a saline (100 ml) injection. Fortyeight hours later, the ®sh were anesthetized with 50 ppm quinaldine and sacri®ced for RNA preparation from hepatopancreas, head kidney, renal (body) kidney and spleen. Total RNA was isolated by the acid±guanidium±phenol±chroloform method [14] using ISOGEN reagent and stored at 280 8C.

2.6. RT±PCR analysis of carp B/C2 isotypes Oligo (dT)-primed ®rst strand cDNA was synthesized with MMLV±reverse transcriptase (Gibco BRL, MD), from 2 mg of the total RNA and subjected to PCR ampli®cations of the serine protease domain sequences of four B/C2 isotypes and a sequence segment of carp 40S ribosomal protein S11 subunit (S11, database accession No. AB012087) as a control, using primers listed in Table 1. 20, 25 and 30 cycles of ampli®cation were performed on T3-thermocycler (Biometra, Goettingen, Germany) as follows: 94 8C for 0.5 min, 54 8C for 0.5 min, and 72 8C for 1 min. The ampli®ed products were electrophorezed on 2% agarose gels and quantitated using Kodak Digital Science EDAS120 System (Lifetech Oriental, Tokyo, Japan). The levels of B/C2-expression were normalized with respect to the expression level of S11. 3. Results 3.1. Isolation of two distinct cDNA clones encoding carp B/C2 A cDNA fragment (clone m33, database accession

536

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

Fig. 1. Comparison of the deduced amino acid sequence of clone m33 (Accession number, C88395) with those of two previously reported carp B/C2-isotypes, B/C2-A (AB007004) and B/C2-B (AB007005). Dots show residues identical to those of m33, and dashes are gaps introduced for maximum matching. Cysteine residues characteristic of the SCR module are marked by asterisks.

No. C88395) (Fig. 1) found in the suppression subtractive hybridization [11] encoded an amino acid sequence similar to but distinct from B/C2-A and B/C2-B, which were previously isolated from a carp hepatopancreas cDNA library [7], suggesting that the new clone represents a transcript from an additional copy of a carp B/C2-like gene. Thus, the clone m33 was labeled with DIG and used to screen the cDNA libraries prepared from the hepatopancreas and peritoneal exudate leukocytes of carp. Screening of the hepatopancreas cDNA library containing 5 £ 10 5 clones identi®ed 30 weakly hybridized clones, and they were plaque-puri®ed and analyzed for their insert size. A clone carrying the longest insert of 2.4 kbp was sequenced for its entire length. The nucleotide sequence predicted 750 amino acids, in which a signal peptide, three SCR modules, a von Willebrand factor domain and a serine protease domain were identi®ed from the N-terminus as in B/C2-A from carp and B and C2 from other vertebrates. Thus the clone was designated B/C2A2, while the previously isolated carp B/C2-A is renamed here as B/C2-A1. Screening of the leukocyte cDNA library containing 4 £ 10 5 clones yielded 26 positive clones. Among them, a clone with the longest insert of 2.5 kbp was isolated. Its entire nucleotide sequence had an open reading frame encoding 754 amino acids. The deduced sequence is distinct from that of B/C2-A1 and B/C2-A2 but speci®es the same domain structure as them; it is thus designated B/C2-A3.

3.2. Multiple alignment and phylogenetic tree The complete amino acid sequences of carp B/C2A1, B/C2-A2 and B/C2-A3 were aligned with all available B and C2 sequences from vertebrates, using the Clustal X software. Only a part of the alignment containing the three sequences and B/C2B of carp is shown in Fig. 2. While B/C2-B has four SCR modules, B/C2-A1, B/C2-A2 and B/C2-A3 contain three. Other sequence motifs, such as the metal binding site sequence (Asp-Xaa-Ser-Xaa-Ser) in the von Willebrand factor domain and the catalytic triad residues (His, Asp and Ser) of the serine protease are conserved in all the four carp B/C2 sequences. Also conserved is an Asp residue in the serine protease domain corresponding to the substrate speci®city-determining residue unique to B and C2 of jawed vertebrates [15]. Table 2 shows amino acid sequence identity values of carp B/C2-A2 and B/C2A3 with B/C2-A1 and B/C2-B of carp and B and C2 of other animals. Whereas carp B/C2-A2 shows close similarity to carp B/C2-A1 sharing 78% amino acid identity, B/C2-A3 shows less similarity with either B/C2-A1 or B/C2-A2, suggesting that B/C2-A3 represents a novel lineage. This was examined by a neighbor-joining phylogenetic tree [16] of B and C2 from vertebrates, using sea urchin B as an outgroup. As shown in Fig. 3, the carp B/C2-A isotypes and B/C2 from other teleosts form a cluster, in which, after separation of medaka and trout B/C2, carp B/C2-A3 diverges ®rst, followed by branching of

Fig. 2. Alignments of the deduced amino acid sequences of B/C2-isotypes of carp. Dots show residues identical to those of B/C2-A1, and dashes are gaps introduced for maximum matching. Boundaries between the following domains are shown by slashes: SCR, VWF, von Willebrand factor and SP, serine protease. Potential N-glycosylation sites are shaded. The metal-binding motif in the VWF domain is shown in bold, the active triad residues of the serine protease are denoted by `#', and the Asp residue which primarily determines the S1-speci®city is shown by `$'.

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

537

538

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

Table 2 Amino acid sequence identity (%) of carp B/C2-isotypes with Bf and C2 from other animals (sequence identity was calculated on the basis of relevant pairwise alignment generated by Clustal X software) Carp B/C2-isotypes

Carp B/C2-A2 Carp B/C2-A3 Carp B/C2-B Zebra®sh Bf Zebra®sh Bf-2 a Trout Bf1 a Trout Bf2 a Medaka ®sh Bf Xenopus Bf A Xenopus Bf B Human Bf Human C2 Mouse Bf Mouse C2 Lamprey Bf Sea urchin Bf

A1

A2

A3

B

78.2 59.9 28.9 61.5 64.2 43.7 44.5 41.9 34.3 33.6 37.4 34.1 36.3 33.9 28.1 18.4

± 61.5 30.3 64.2 63.5 43.1 45.2 41.0 33.4 34.7 36.6 32.9 38.1 32.2 28.1 19.4

± ± 29.5 55.2 59.2 47.6 46.5 42.3 34.9 34.1 38.3 31.7 38.0 30.4 30.0 18.9

± ± ± 29.5 31.9 31.2 31.4 30.0 33.0 32.7 35.0 36.0 35.3 34.0 28.0 19.7

Fig. 3. A phylogenetic tree of the complement components factor B and C2. The tree was drawn by the neighbor-joining method using the sea urchin B as an outgroup. Bootstrap percentages that support each partitioning are given. The bar beneath the tree is a scale to show the genetic distance of 0.05.

a Partial sequence. The identity was calculated on the overlapping region.

3.4. Expression analysis of the four carp B/C2isotypes

zebra®sh B. The divergence of carp B/C2-A1 and B/C2-A2 seems to be the most recent event. This branching sequence suggests that zebra®sh also possess an ortholog of carp B/C2-A3, though this is yet to be found.

The tissue distribution of mRNAs of the four B/C2 isotypes was analyzed by RT±PCR, using the isotypespeci®c primers shown in Table 1. Fig. 5(A) shows the results of 25 cycles of ampli®cation from the tissues of a representative ®sh. In normal (unstimulated) ®sh, B/C2-A3 is expressed mainly in head kidney and renal kidney rather than in hepatopancreas, which is the major site of expression of B/C2-A1, B/C2-A2 and B/C2-B. In the spleen of unstimulated ®sh, only a weaker signal of B/C2-A3 was detected. Since the clone m33, which was used as the probe for B/C2-A3, was found in a cDNA subtraction between ®sh stimulated with sodium alginate and scleroglucan and normal ®sh [11], we examined if the in vivo stimulation actually up-regulates the expression of B/C2-A3 as well as the other three B/ C2-isotypes in various tissues 48 h after intraperitoneal injection of saline, scleroglucan or sodium alginate. Fig. 5(B) shows the result of semi-quantitative RT±PCR after 20 cycles of ampli®cation, which gave a dose-response linearity better than that after 25 and 30 cycles. In the hepatopancreas, expression was not increased signi®cantly for any of the isotypes

3.3. Southern hybridization To determine whether B/C2-A1, B/C2-A2 and B/C2-A3 are derived from distinct genes or whether some of them are allotypes of a single gene, southern hybridization was performed using PstI-digests of genomic DNA isolated from seven representative carp and probes speci®c to the von Willebrand factor domain of each of the three B/C2-isoforms. No crosshybridization of the probes was observed as shown in lanes C of Fig. 4. Each probe detected one or two polymorphic bands from each carp, each of which gave a distinct band pattern with the three different probes. These results suggest that B/C2-A1, B/C2-A2 and B/C2-A3 are distinct isotypes each encoded by a single gene in the carp genome.

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

539

Fig. 4. Southern blotting analysis of carp B/C2-A1, B/C2-A2 and B/C2-A3. 5 mg of PstI-digests of genomic DNA from seven representative carp (Lanes 1±7) and mixtures of 10 pg each of linearized plasmids encoding B/C2-A1, B/C2-A2 and B/C2-A3 (Lane C) were electrophoresed in a 1% agarose gel, transferred to a nylon membrane, and hybridized under stringent conditions with DIG-labeled probes speci®c for each B/C2-isoform. An arrowhead shows the origin of electrophoresis, and positions and sizes (kbp) of the l/HindIII-digest markers are shown on the left and between the panels.

following either stimulation; the apparent increase of B/C2-A1 was not signi®cant due to substantial variation among individuals. On the other hand, in spleen and the two parts of the kidney, the expression of B/ C2-A3 was remarkably enhanced by each stimulation, while that of the other isotypes remained low or undetectable after the 20 cycle-ampli®cation.

4. Discussion We have previously isolated two diverged B/C2like clones, B/C2-A and B/C2-B, sharing less than 30% amino acid identity, from carp and predicted the presence of multiple B/C2-A genes and a single B/C2-B gene, based on the results of southern

Fig. 5. Expression analysis of carp B/C2-A1, B/C2-A2, B/C2-A3 and B/C2-B by RT±PCR. (A) RT±PCR products by 25 cycles of ampli®cation from hepatopancreas (HP), head kidney (HK), spleen (SP) and renal kidney (RK) of an unstimulated ®sh were electrophorezed on a 2% agarose gel. Lanes A1, A2, A3 and B denote the four B/C2-isotypes, and lane C shows the S11 subunit of 40S ribosomal protein used as a positive control. The sizes of the ampli®ed products are shown below the lane labels. (B) Semi-quantitative RT±PCR of the four carp B/C2isotypes from four organs of ®sh that had been injected with saline (control), scleroglucan or sodium alginate 48 h before. Labels are the same as in (A). Expressions level relative to those of S11 in each organ were determined from the band intensities after 20 cycles of RT±PCR and are shown as the average ^ SD (n ˆ 3).

540

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541

hybridization [7]. The present study provides the fulllength cDNA sequences of two additional B/C2-A isotypes, named B/C2-A2 and B/C2-A3. It is still unclear whether the B/C2-A isotypes and B/C2-B correspond to mammalian B and C2, respectively, in spite of a recent phylogenetic analysis using B/C2 sequences from various vertebrates including a cyclostome, cartilaginous ®sh, bony ®sh, an amphibian and mammals [15]. Functional evidence at the protein level will be required to identify C2 in bony and cartilaginous ®sh, and this would provide an inference as to the time of divergence of B and C2. The presence of three B/C2-A isotypes indicates that at least two cycles of gene duplication occurred after the divergence of their common ancestor from the B/C2-B lineage. The tight clustering of B/C2-A1 and B/C2-A2 in the phylogenetic tree and the tetraploid nature of carp chromosomes [17] together suggest that these two isotypes have diverged by ancient tetraploidization, which is known to have occurred in carp. It seems also possible that the tetraploidization generated B/C2-A3, followed by tandem duplication to generate B/C2-A1 and B/C2-A2. In either case, it is intriguing that, in the phylogenetic tree, the divergence of B/C2-A3 predated the separation of carp B/C2-A1 and B/C2-A2 from zebra®sh B. This suggests that zebra®sh, and possibly other cyprinid ®sh species, also possess a B/C2-A3-like isotype. Mapping of these genes either by physical means or by linkage analysis will facilitate the elucidation of the evolutionary mechanisms that have generated the carp B/C2-isotypes. The substantial restriction fragment length polymorphism of each isotype (Fig. 4) may be useful for the linkage analysis. The biological signi®cance of the B/C2-diversi®cation described here is of great interest. Although further biochemical analyses at the protein level are required to reveal any functional differentiation of the carp B/C2-isotypes, both the sequences and the pattern of expression lead us to predict functional differences. The amino acid identity (78%) between B/C2-A1 and B/C2-A2 is similar to that (74%) between two trout Bf isotypes (Bf1 and Bf2) [6]. However it should be noted that, in contrast to the two trout sequences, carp B/C2-A1 and B/C2-A2 show substantial sequence diversity within the von Willebrand factor domain. This domain in mammalian B has been

shown to be important for speci®c binding of B to C3b-fragment when they form the C3-convertase complex [18]. In this context, it is interesting to note that carp have also multiple C3 isoforms with distinct binding speci®city to target molecules [19]. It is possible that the B/C2-A isoforms have distinctive preferences for the particular C3 isoforms in the C3-convertase formation. B/C2-A3 is considerably different in primary structure from those of B/C2-A1 and B/C2-A2, though it retains the same domain organization. B/C2-A3 also showed a novel expression pattern in which it was expressed in the kidney and spleen predominantly, and transcription was up-regulated by the administration of immunostimulating polysaccharides that may mimic bacterial infection. Extrahepatic expression of B, which is encoded by a single gene, has also been reported for mammals such as humans and mice, in which monocytes and granulocytes constitutively secrete B, with the level increasing in response to infections [20,21]. In carp, however, only the B/C2A3 shows signi®cant expression in tissues other than hepatopancreas, and responds to in vivo stimulation. Thus, B/C2-A3 may represent a novel B/C2 subtype that plays a role as an acute phase reactant to enhance complement activation. In summary, the present study revealed the presence of three B/C2-A isotypes including a novel one that behaves like an acute phase reactant in extrahepatic organs. Preparation of recombinant protein corresponding to the isotypes, now in progress in our laboratory, will lead to the biochemical evaluation of their possible functional differences. Acknowledgements We are grateful to Dr C.J. Bayne, Oregon State University, for critically reading the manuscript. This study was supported partly by a grant-in-aid for scienti®c research from the Japanese Ministry of Education, Science, Sports and Culture (12660173 to M.N., 11694220 and 11460098 to T.Y.). References [1] Law SKA, Reid KBM. Complement. 2nd ed. Oxford: Oxford University Press, 1995.

M. Nakao et al. / Developmental and Comparative Immunology 26 (2002) 533±541 [2] Ishikawa N, Nonaka M, Wetsel RA, Colten HR. Murine complement C2 and factor B genomic and cDNA cloning reveals different mechanisms for multiple transcripts of C2 and B. J Biol Chem 1990;265(31):19,040±6. [3] Nonaka M, Kuroda N, Naruse K, Shima A. Molecular genetics of the complement C3 convertases in lower vertebrates. Immunol Rev 1998;166:59±65. [4] Seeger A, Mayer WE, Klein J. A complement factor B-like cDNA clone from the zebra®sh (Brachydanio rerio). Mol Immunol 1996;33:511±20. [5] Kuroda N, Wada H, Naruse K, Shimada A, Shima A, Sasaki M, Nonaka M. Molecular cloning and linkage analysis of the Japanese medaka ®sh complement Bf/C2-gene. Immunogenetics 1996;44:459±67. [6] Sunyer JO, Zarkadis I, Sarrias MR, Hansen JD, Lambris JD. Cloning, structure and function of two rainbow trout Bf molecules. J Immunol 1998;161(8):4106±14. [7] Nakao M, Fushitani Y, Fujiki K, Nonaka M, Yano T. Two diverged complement factor B/C2-like cDNA sequences from a teleost, the common carp (Cyprinus carpio). J Immunol 1998;161:4811±8. [8] Gongora R, Figueroa F, Klein J. Independent duplications of Bf and C3 complement genes in the zebra®sh. Scand J Immunol 1998;48(6):651±8. [9] Yano T, Mangindaan REP, Matsuyama H. Enhancement of the resistance of carp Cyprinus carpio to experimental Edwardsiella tarda infection, by some b-1,3-glucans. Nippon Suisan Gakkaishi 1989;55(10):1815±9. [10] Fujiki K, Yano T. Effects of sodium alginate on the nonspeci®c defence system of the common carp (Cyprinus carpio L.). Fish Shell®sh Immunol 1997;7:417±27. [11] Fujiki K, Shin DH, Nakao M, Yano T. Molecular cloning of carp (Cyprinus carpio) CC chemokine, CXC chemokine receptors, allograft in¯ammatory factor-1, and natural killer cell enhancing factor by use of suppression subtractive hybridization. Immunogenetics 1999;49:909±14. [12] Sanger F, Nicklen S, Coulson AR. DNA sequencing with

[13]

[14]

[15]

[16]

[17] [18]

[19]

[20]

[21]

541

chain-terminating inhibitors. Proc Natl Acad Sci USA 1977; 74:5463±7. Sambrook J, Fritsch EF, Maniatis T, editors. Molecular Cloning. A Laboratory Manual 2 ed. New York: Cold Spring Harbor Laboratory Press, 1989. Chomczynski P. A reagent for the single-step simultaneous isolation of RNA, DNA and proteins from cell and tissue samples. Biotechniques 1993;15:532±4. Terado T, Smith SL, Nakanishi T, Nonaka MI, Kimura H, Nonaka M. Occurrence of structural specialization of the serine protease domain of complement factor B at the emergence of jawed vertebrates and adaptive immunity. Immunogenetics 2001;53:250±4. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987; 4:406±25. Ohno S. Evolution by Gene Duplication. New York: Springer, 1970. Tuckwell DS, Xu Y, Newham P, Humphries MJ, Volanakis JE. Surface loops adjacent to the cation-binding site of the complement factor B von Willebrand factor type A module determine C3b binding speci®city. Biochemistry 1997;36(22): 6605±13. Nakao M, Mutsuro J, Obo R, Fujiki K, Nonaka M, Yano T. Molecular cloning and protein analysis of divergent forms of the complement component C3 from a bony ®sh, the common carp (Cyprinus carpio): presence of variants lacking the catalytic histidine. Eur J Immunol 2000;30:858±66. Sundsmo JS, Chin JR, Rapin RA, Fair DS, Werb Z. Factor B, the complement alternative pathway serine protease, is a major constitutive protein synthesized and secreted by resident and elicited mouse macrophages. J Exp Med 1985;161: 306±22. Okuda T. Murine polymorphonuclear leukocytes synthesize and secrete the third component and factor B of complement. Int Immunol 1991;3:293±6.