Cloning and phylogenetic analysis of the alpha subunit of the eighth complement component (C8) in rainbow trout

Molecular Immunology 43 (2006) 2188–2194

Cloning and phylogenetic analysis of the alpha subunit of the eighth complement component (C8) in rainbow trout Anastasios D. Papanastasiou, Ioannis K. Zarkadis ∗ Department of Biology, School of Medicine, University of Patras, Rion Patras 26500, Greece Received 2 December 2005; received in revised form 4 January 2006; accepted 6 January 2006 Available online 17 February 2006

Abstract The alpha subunit of the eighth complement component (C8) is a single-chain plasma glycoprotein which functions in the cytolytic process mediated by the complement system through a sequence of polymerization reactions with other terminal components. We have previously isolated and characterized the C8␤ and C8␥ subunits of the eighth complement component in rainbow trout (Oncorhynchus mykiss). Here, we report the primary sequence, the tissue expression profile, the domain architecture and the phylogenetic analysis of the trout C8␣ gene. The deduced amino acid sequence of the trout C8␣ gene exhibits 44 and 43% identity with human and frog orthologs, respectively. The domain architecture of the trout C8␣ resembles that of mammalian orthologs, and the cysteine backbone shows a high degree of conservation. The trout C8␣ shows a similar expression profile with that of trout C8␤ and C8␥, pointing to the liver as the main source of the C8 genes expression. Although the presence of a fully developed lytic pathway of complement system is expected in teleost, this is the first report of the C8␣ gene in an organism other than mammalian. © 2006 Elsevier Ltd. All rights reserved. Keywords: Comparative immunology; Evolution; Phylogeny; Complement lytic pathway; Rainbow trout; C8␣

1. Introduction C8 is one of the five components (C5b, C6, C7, C8 and C9) that interact to form the cytolytic membrane attack complex (MAC) of the complement system (Muller-Eberhard, 1988; Esser, 1994). In human it is composed of an ␣ (64 kDa), ␤ (64 kDa), and ␥ (22 kDa) subunit, each of which is encoded by a separate gene (Steckel et al., 1980; Ng et al., 1987). Within C8, the subunits are arranged as a disulfide-linked C8␣–␥ heterodimer that is noncovalently associated with the C8␤. C8␣ and C8␤ are homologous and together with C6, C7, and C9 belong to the MAC-perforin family of proteins (Hobart et al., 1995; Volanakis, 1998). Both C8␣ and C8␤ contain a pair of tandemly arranged N-terminal modules (TSP1–LDLa) and a pair of C-terminal modules (EGF–TSP1). The intervening segment is referred to as the MACPF domain because of its sequence similarity to other family members and the corresponding region of the pore-forming protein perforin. The C8␥ subunit is unrelated

∗

Corresponding author. Tel.: +30 2610 997621; fax: +30 2610 991769. E-mail address: [email protected] (I.K. Zarkadis).

0161-5890/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2006.01.004

and is a member of the lipocalin family of widely distributed proteins that bind and transport small, hydrophobic ligands (Flower et al., 2000; Schreck et al., 2000; Ortlund et al., 2002). Furthermore, C8␣ and C8␤ have correspondingly similar roles in MAC-mediated lysis of erythrocytes while C8␥ seems not to be required for complement-mediated killing of Gram-negative bacteria (Parker and Sodetz, 2002). The terminal complement components C6, C7, C8␣, C8␤, and C9 (TCCs) belong to the same gene family as the perforins, the lytic proteins of natural killer cells and cytotoxic lymphocytes (DiScipio, 1992; Podack et al., 1988), and they may have emerged through a series of duplications of an ancestral gene. Consequently, the mammalian TCCs share many common structural motifs, such as a thrombospondin type I domain (TSP1), low-density lipoprotein receptor class A (LDLa), epidermal growth factor precursor (EGF) and the membrane attack complex/perforin segment (MACPF). These domains have been conserved and they are also present in their teleost counterparts (Katagiri et al., 1999). Components of the complement lytic pathway have been identified only in vertebrates, while the cloning of a C6-like molecule in the amphioxus, Branchiostoma belcheri (Suzuki et al., 2000) and the presence of

A.D. Papanastasiou, I.K. Zarkadis / Molecular Immunology 43 (2006) 2188–2194

C6-like sequences in the genome of the ascidian Ciona intestinalis (Nonaka and Yoshizaki, 2004), suggest that this pathway may have been established prior to the origin of the vertebrate subphylum. In teleost, the MAC components have been microscopically observed as small pores in the cell surface (Jenkins et al., 1991). We have previously reported the isolation of the C5b, C6, C7, C8␤, C8␥ and C9 complement components in rainbow trout (Oncorhynchus mykiss) (Franchini et al., 2001; Chondrou et al., 2006; Zarkadis et al., 2005; Papanastasiou and Zarkadis, 2005a, 2005b; Kazantzi et al., 2003). In order to fully characterize the components of the complement lytic pathway, we describe here the cloning and characterization of the alpha subunit of the eighth complement component (C8␣) in trout. In addition, we perform an expression analysis of the C8␣ gene in various tissues and we present a phylogenetic tree to better determine the evolutionary position of the trout C8␣ gene. 2. Materials and methods 2.1. Cloning of trout C8α 2.1.1. RNA isolation and cDNA library construction Liver cDNA library was prepared from total RNA extracted from a single liver as previously described (Zarkadis et al., 2001). 2.1.2. Trout C8α probe isolation Degenerated oligonucleotides were designed based on conserved regions of deduced amino acid sequences of lytic complement components from various species: sense 21-mer: 5 -ACNGTNTA(T/C)AA(C/T)GGN GA(A/G)TGG-3 based on TVYNGEW amino acids and antisense 21-mer: 5 (A/G)AA(A/G)TGNGCNGT(T/C)TGNA C(T/C)TT-3 based on KVQTAHF amino acids at positions 179 and 302 amino acid of human C8␣, respectively (the mixtures of nucleotides are represented by N = A, G, C and T). These primers were subsequently applied in a RT-PCR reaction (Qiagen), using as template total trout liver RNA. One cycle was conducted at 48 ◦ C for 30 min. Thirty cycles were conducted, using a PCR thermocycler under the following program: 95 ◦ C for 1 min, 46 ◦ C for 1 min and 72 ◦ C for 45 s, followed by a final extension at 72 ◦ C for 10 min. The PCR product of the expected size (351 bp) was gel-purified (QIAquick, Qiagen), ligated into the T/A cloning vector pGEM-T easy (Promega) at 4 ◦ C overnight and transformed into E. coli DH5a subcloning competent cells (Invitrogen). Positive clones were selected and plasmid DNA was extracted (mini-prep kit, Roche). 2.1.3. Screening of a trout liver cDNA library 1.5 × 105 ␭gt11 recombinant phages of a trout liver cDNA library were screened under high stringency conditions (65 ◦ C) using an ␣-32 P-labelled cDNA probe corresponding to the DNA product, 351 bp in size, described above. The probe was labelled using the random primed DNA labelling kit (Boehringer Mannheim). Positive plaques were cultured, the recombinant phage DNA corresponding to the longest clone in size was iso-

2189

lated, and the insert cDNA was subcloned into the pGEM-T easy vector and sequenced. 2.1.4. Sequencing of trout C8α Definition of primary structure of trout C8␣ was performed by dideoxy-chain termination method using the DNA Sequencer Long Read IR 4200 (Li-Cor). All sequences were determined at least twice for both strands. 2.2. Database search/multiple sequence alignment/phylogenetic analysis Analysis and assembly of data derived from DNA sequencing was performed with the Gene Tool Lite software. Basic Local Alignment Search Tool (BLAST, http://www.ncbi.nlm.nih.gov/blast) (Altschul et al., 1990) and S.M.A.R.T (http://smart.embl-heidelberg.de/) (Letunic et al., 2004) were employed for GenBank search, identity/similarity assessment and protein domain determination and characterization. Deduced amino acid sequences were obtained from EMBL and GenBank databases. Amino acid multiple alignments were generated using the Clustal W program (Thomson et al., 1994) within MEGA Version 3 (Kumar et al., 2004). Phylogenetic trees were constructed based on the deduced amino acid sequences of full-length TCCs using the neighbor-joining (NJ) algorithm (Saitou and Nei, 1987) within MEGA Version 3 (Kumar et al., 2004). The phylogenetic tree was constructed using the Poisson correction and branch points were validated by 1000 bootstrap replications. All other conditions were set as “default”. Image analysis was carried out with Kodak Digital Science (Electrophoresis Documentation and Analysis System 120). 2.3. Southern blot analysis Genomic DNA was extracted from trout liver and 12 ␮g were digested overnight at 37 ◦ C with the restriction enzymes BamHI, EcoRI or HindIII. Restricted DNA was electrophoresed on a 0.8% agarose gel and transferred onto a nylon membrane (Zeta Probe Biorad). Transferred DNA was prehybridized at 65 ◦ C for 30 min and then hybridized with a 305 bp (␣-32 P)-radiolabelled DNA probe, at 65 ◦ C for 16 h. The probe was generated by PCR, using the primers TC8aF1: 5 -TGACCCACAATACTATGGAGG-3 and TC8aR1: 5 -TAACGCAGACTCCACATTCCC-3 which span from 611 to 914 nucleotides of the trout C8␣ cDNA sequence, and labelled using the random primed DNA labelling kit (Boehringer Mannheim). Following 16 h hybridization, blots were washed twice with 40 mM sodium phosphate buffer (pH 7.2) and 1 mM EDTA in 5% SDS at room temperature for 15 min. The X-ray film was developed after 2 days of exposure and the hybridized bands visualized by autoradiography. 2.4. RT-PCR analysis RNA was extracted from different trout tissues using the SV Total RNA Isolation system (Promega). Fifty nanograms

2190

A.D. Papanastasiou, I.K. Zarkadis / Molecular Immunology 43 (2006) 2188–2194

Fig. 1. Multiple alignment of trout, frog and mammalian C8␣ deduced amino acid sequences. Amino acid sequence alignment of TC8α (trout C8␣, AJ851931), HC8α (human C8␣, AAA82124), RabC8α (rabbit C8␣, P98136) and FrC8α (frog C8␣, AAH74554) was performed using the Clustal W program. ‘*’ indicates fully conserved residues, ‘:’ indicates conservation of strong groups, and ‘.’ indicates conservation of weak groups. Amino acid numbering is indicated on the right. Signal peptide of trout C8␣ is underlined. Conserved Cys residues are indicated with arrows. Segment sequence (179–198 aa) is boxed. Conserved sequence (165–179 aa) is in dashed box.

A.D. Papanastasiou, I.K. Zarkadis / Molecular Immunology 43 (2006) 2188–2194

of trout total RNA from the brain, heart, intestine, kidney, liver, or spleen, were reverse transcribed using the one-step RT-PCR kit (Qiagen) with the primers TC8␣F2: 5 -GCTCCAGTTCAACTCCTGTCG-3 and TC8␣R2: 5 GTTGTCACATGACCGTCGGC-3 . The program was as follows: 55 ◦ C for 30 min, 95 ◦ C for 15 min, 30 cycles of 95 ◦ C for 30 s, 55 ◦ C for 30 s, 72 ◦ C for 30 s, followed by a final extension at 72 ◦ C for 10 min. Trout ␤-actin mRNA was amplified using the following primers: ␤actinF: 5 -CACCTTCTACAATGAGCTGC-3 , and ␤-actinR: 5 -AGGCAGCTCGTAGCTCTTCT-3 . Amplification products

2191

were analyzed and visualized by electrophoresis on a 2.5% agarose gel. 3. Results 3.1. Cloning and sequence analysis of a full-length cDNA clone encoding the alpha subunit of trout C8 To clone the gene encoding C8␣ in trout, we generated a cDNA probe by RT-PCR using trout liver mRNA and degenerated primers. A single 351 bp product was obtained, after

Fig. 2. Alignment of partial amino acid sequence of trout C6 (TC6), C7-1 (TC7-1), C7-2 (TC7-2), C8␣ (TC8α), C8␤ (TC8β), and C9 (TC9), indicating the indel sequence within trout C8␣. Numbers identify the indel in trout C8␣. Cys187 is indicated by an asterisk. The indel in the C8␣ sequence (179–198 aa) is underlined.

Fig. 3. Structural organization of the trout C6 (TC6), C7-2 (TC7-2), C7-1 (TC7-1), C8␣ (TC8α), C8␤ (TC8β), and C9 (TC9) proteins. Shown maps are based on sequences of the proteins and the module boundaries listed in the SWISS-PROT Protein Sequence Data Bank. Abbreviations correspond to thrombospondin type I (TSP1), low-density lipoprotein receptor class A (LDLa), membrane attack complex/perforin (MACPF), epidermal growth factor (EGF), complement control protein (CCP) and factor I MAC module (FIMAC). Numbers correspond to the first residue in each module. Hexagonal symbols designate Asn residues that are potential N-glycosylation sites.

2192

A.D. Papanastasiou, I.K. Zarkadis / Molecular Immunology 43 (2006) 2188–2194

RT-PCR amplification, which was aligned to human C8␣ nucleotide sequence at positions 671–1040 bp. Ten positive clones were isolated after screening of a trout liver ␭gt11 cDNA library, using as probe the 351 bp fragment, and the longest obtained insert was 2037 bp, involving a polyadenylation signal within the 3 -UTR region and a polyA tail (data not shown). The predicted open reading frame of 615 amino acids starts with a putative signal peptide (1–39 aa) (Fig. 1). Its deduced amino acid sequence presented 44 and 43% identity with human and frog (Xenopus tropicalis) C8␣, respectively.

of the mature protein (Fig. 2). This Cys residue is the equivalent of Cys164 of the human ortholog and probably is involved in the formation of the disulfide bond between C8␣ and C8␥. The domain architecture of the trout C8␣ is similar to other mammalian C8␣ proteins, consisting of two thrombospondin domains type 1 (TSP1), a low-density lipoprotein receptor class A (LDLa), an epidermal growth factor domain (EGF), and a MAC/perforin domain (MACPF) (Fig. 3).

3.2. Alignment of trout C8α deduced amino acid sequence

To better determine the evolutionary position of trout C8␣, a phylogenetic tree was constructed using the NJ method. The NJ method was chosen because it is more reliable than other methods in cases where branches of the tree differ with respect to their rates of evolution. The phylogenetic tree constructed using full-length TCC amino acid sequences indicates that known fish C8 sequences are clustered together and form a larger group with mammalian C8. In addition, the group of C8␣ sequences seems to comprise a well-defined cluster within the evolutionary

The deduced amino acid sequence of rainbow trout C8␣ was aligned with those of human, rabbit and frog C8␣ (Fig. 1), as well as with TCCs of representative species (data not shown). All 28 cystein residues of the mature amino acid sequence of trout C8␣ which are involved in disulfide bond formation, are fully conserved when aligned with human C8␣ protein. In addition, trout C8␣ contains an indel, which includes the conserved Cys187

3.3. Phylogenetic analyses

Fig. 4. Phylogenetic tree of the lytic complement components C6, C7, C8a, C8b, and C9 of various species. Trout C8a (CAH65481), trout C7-2 (CAF22025), trout C7-1 (CAD92841), human C7 (NP 000578), porcine C7 (NP 999447), Japanese flounder C7 (BAA88899), amphioxus C6 (Branchiostoma belcheri, BAB47147), human C6 (AAH35723), mouse C6 (NP 057913), human C8a (NP 000553), rabbit C8a (P98136), African frog C8a (Xenopus tropicalis C8␣) (AAH74554), human C8b (NP 000057), rabbit C8b (P98137), trout C8b (AAL16647), Japanese flounder C8b (BAA86877), pufferfish C9 (AAC60288), trout C9 (CAA29037), human C9 (NP 001728), horse C9 (AAB16820), rabbit C9 (AAC48459), mouse C9 (NP 038513). Human perforin (NP 005032) and Japanese flounder perforin (BAC76420) were used as outgroups. Branch points were validated by 1000 bootstrap replications. A scale bar is also shown.

A.D. Papanastasiou, I.K. Zarkadis / Molecular Immunology 43 (2006) 2188–2194

2193

4. Discussion

Fig. 5. Southern blot analysis of trout genomic DNA using as probe a cDNA fragment (305 bp) of trout C8␣. The positions of the DNA standards are shown on the left.

scenery of TCCs supported by high bootstrap values (Fig. 4). In support of this view, phylogenetic analysis was carried out using amino acid sequences of the mature C8␣ proteins, producing a tree which showed the same topology (data not shown). 3.4. Presence of C8α gene in trout genome and tissue distribution of trout C8α mRNA In order to estimate the copy number of the C8␣ gene in the trout genome, we performed Southern blot analysis, using as probe the 305 bp fragment. A single and two hybridizing bands were detected after digestion of trout genomic DNA with restriction enzymes BamHI, HindIII, and EcoRI, respectively (Fig. 5), suggesting that the trout C8␣ gene is present as a single copy in the trout genome. In addition, we examined the tissue expression profile of trout C8␣ mRNA by RT-PCR. As shown in Fig. 6, C8␣ mRNA expression is detected mainly in the liver (lane 7) and kidney (lane 6). Expression is also observed at a low level in the brain (lane 3), heart (lane 4), intestine (lane 5), and spleen (lane 8).

Fig. 6. Analysis (RT-PCR) of trout C8␣ mRNA expression in various trout tissues. 1, trout genomic DNA as template; 2, negative control; 3, brain; 4, heart; 5, intestine; 6, kidney; 7, liver; 8, spleen. Beta-actin amplification was included as an internal control and as a measure of RNA integrity.

C8␣ is a subunit of complement component C8, which is one of the five components (C5b, C6, C7, C8, and C9) that interact to form the cytolytic membrane attack complex of complement. Teleost fish express all lytic complement components and are able to kill microorganisms by cytolysis. In order to further elucidate the evolutionary history of the lytic complement pathway, we now report the nucleotide and deduced amino acid sequence of a full-length cDNA clone of the alpha subunit of the eighth complement component in rainbow trout. The sequence of trout C8␣ cDNA is 2037 bp in size encoding 615 amino acids and the mature amino acid sequence (585 aa) is comparable in size to that of the human C8␣ (555 aa). The deduced amino acid sequence shows the highest score of identity 62 and 44% with the Tetraodon nigroviridis unnamed protein product (GenBank, CAF97618) and human C8␣, respectively. All the cysteine residues of trout C8␣, the majority of which are located at the amino or carboxyl terminus, show high conservation compared to the mammalian C8␣ cysteine backbone. Trout C8␣ contains the C-mannosylation motif (WXXWXXW) and two potential N-glycosylation sites (Asn-X-Ser/Thr) in accordance with human C8␣. The trout C8␣ amino acid sequence contains a distinct indel that is located within a segment defined by residues 179–198 of the putative mature protein (Fig. 1). This segment includes a unique indel (residues 179–193) that lies outside the MACPF portion of the trout C8␣ (Fig. 2), in contrast with the human ortholog, where this indel located in the MACPF domain (Plumb et al., 1999). This indel seems well conserved (47% identical to human) between species and this includes Cys187 (human Cys164 ) that is normally linked to C8␥ and which is fully conserved among species (Fig. 1) (Plumb and Sodetz, 2000). An interesting observation, regarding to the amino acid residues lying upstream the indel mentioned above (residues 179–193), is the particularly high degree of conservation. Trout amino acid residues from 165 to 179 are 93% identical to the corresponding human amino acid sequence (residues 142–156). We believe that this region is of importance for the protein regarding its folding and the linkage to C8␥. However, functional assays at the protein level are required to elucidate the biological significance of these highly conserved residues. The evolutionary position of trout C8␣ was revealed by the construction a phylogenetic tree of the TCCs using the neighborjoining (NJ) algorithm within MEGA 3 (Kumar et al., 2004), and the human and Japanese flounder perforin as outgroup. Moreover, the NJ method has been used in the past to examine the phylogeny of the MACPF protein family (Mondragon-Palomino et al., 1999). The phylogenetic tree, constructed using full-length TCCs amino acid sequences, indicates that trout C8␣ is clustered with the mammalian and frog C8␣, comprising a well-defined subgroup inside the MACPF protein family (Fig. 4). The question whether the trout C8␣ as well as the C8␤ and C9 genes are products of a C6/C7 ancestor duplication with loss of modules or have evolved directly from the invertebrate C6-like genes, remain unresolved. Data from the ortholog genes of the lower

2194

A.D. Papanastasiou, I.K. Zarkadis / Molecular Immunology 43 (2006) 2188–2194

vertebrates is required for the better understanding of the evolutionary history of the complement lytic pathway. Although a variety of complement genes in rainbow trout have been found to exist in multiple forms (Zarkadis et al., 2001), Southern blot analysis indicates that C8␣ gene is probably found as a single copy in the trout genome, in accordance with data from mammalian species (Fig. 5). In addition, C8␣, C8␤ and C8␥ proteins in trout are products of three different genes, as it occurs in higher vertebrates (Ng et al., 1987). Furthermore, the comparison of trout TCCs with those of the T. nigroviridis and D. rerio, implies the duplication of only the C7 gene among the TCCs in the teleost fish genomes (Papanastasiou and Zarkadis, 2005a). mRNA expression studies by RT-PCR analysis revealed that the trout C8␣ mRNA is expressed mainly in the kidney and liver (Fig. 6). These results seem to be in partial accordance with human C8␣ expression profile that points to the liver as the main source of C8␣ expression. In addition, trout C8␤ and C8␥ mRNA expression patterns (Papanastasiou and Zarkadis, 2005b; Kazantzi et al., 2003) exhibit partial similarity with that of the trout C8␣, supporting that liver and kidney are the main sources of C8␣,␤,␥ mRNA production in the rainbow trout. Finally, the functional role of the trout C8␣ remains to be elucidated. Acknowledgements We thank European Social Fund (ESF), Operational Program for Educational and Vocational Training II (EPEAEK II), and particularly, the Program PYTHAGORAS for funding the above work. The sequence described in this paper has been deposited in the EMBL database under accession number AJ851931 (Trout C8␣). References Altschul, S.F., Gish, W., Myers, E., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403–410. Chondrou, M.P., Mastellos, D., Zarkadis, I.K., 2006. cDNA cloning and phylogenetic analysis of the sixth complement component in rainbow trout. Mol. Immunol. 43, 1080–1087. DiScipio, R.G., 1992. Formation and structure of the C5b-7 complex of the lytic pathway of complement. J. Biol. Chem. 267, 17087–17094. Esser, A.F., 1994. The membrane attack complex of complement: assembly, structure and cytotoxic activity. Toxicology 87, 229–247. Flower, D.R., North, C.T., Sansom, C.E., 2000. The lipocalin protein family: structural and sequence overview. Biochim. Biophys. Acta 1482, 9–24. Franchini, S., Zarkadis, I.K., Sfyroera, G., Sahu, A., Moore, W.T., Mastellos, D., LaPatra, S.E., Lambris, J.D., 2001. Cloning and purification of the rainbow trout fifth component of complement (C5). Dev. Comp. Immunol. 25, 419–430. Hobart, M.J., Fernie, B.A., DiScipio, R.G., 1995. Structure of the human C7 gene and comparison with the C6, C8a, C8b, and C9 genes. J. Immunol. 154, 5188–5194. Jenkins, J.A., Rosell, R., Ourth, D., Coons, L.B., 1991. Electron microscopy of bactericidal effects produced by the alternative complement pathway of channel catfish. J. Aquat. Anim. Health 3, 16–22. Katagiri, T., Hirono, I., Aoki, T., 1999. Molecular analysis of complement component C8 beta and C9 cDNAs of Japanese flounder Paralichthys olivaceus. Immunogenetics 50, 43–48.

Kazantzi, A., Sfyroera, G., Holland, C.M., Lambris, J.D., Zarkadis, I.K., 2003. Molecular cloning of the ␤-subunit of complement component eight of rainbow trout. Dev. Comp. Immunol. 27, 167–174. Kumar, S., Tamura, K., Nei, M., 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings Bioinformatics 5, 150–163. Letunic, I., Copley, R.R., Schmidt, S., Ciccarelli, F.D., Doerks, T., Schultz, J., Ponting, C.P., Bork, P., 2004. SMART 4.0: towards genomic data integration. Nucl. Acids Res. 1 (32(database issue)), D142–D144. Mondragon-Palomino, M., Pinero, D., Nicholson-Weller, A., Laclette, J.P., 1999. Phylogenetic analysis of the homologous proteins of the terminal complement complex supports the emergence of C6 and C7 followed by C8 and C9. J. Mol. E 49, 282–289. Muller-Eberhard, H.J., 1988. Molecular organization and function of the complement system. Annu. Rev. Biochem. 57, 321–347. Ng, S.C., Rao, A.G., Howard, O.M., Sodetz, J.M., 1987. The eighth component of human complement (C8): evidence that it is an oligomeric serum protein assembled from products of three different genes. Biochemistry 26, 5229–5233. Nonaka, M., Yoshizaki, F., 2004. Primitive complement system of invertebrates. Immunol. Rev. 198, 203–215. Ortlund, E., Parker, C.L., Schreck, S.F., Ginell, S., Minor, W., Sodetz, J.M., Lebioda, L., 2002. Crystal structure of human complement protein C8␥ ˚ resolution reveals a lipocalin fold and a distinct ligand binding at 1.2 A site. Biochemistry 41, 7030–7037. Papanastasiou, A.D., Zarkadis, I.K., 2005a. Gene duplication of the seventh component of complement in rainbow trout. Immunogenetics. Papanastasiou, A.D., Zarkadis, I.K., 2005b. The gamma subunit of the eighth complement component (C8) in rainbow trout. Dev. Comp. Immunol. Parker, C.L., Sodetz, J.M., 2002. Role of the human C8 subunits in complement-mediated bacterial killing: evidence that C8␥ is not essential. Mol. Immunol. 39, 453–458. Plumb, M.E., Scibek, J.J., Barber, T.D., Dunlap, R.J., Platteborze, P.L., Sodetz, J.M., 1999. Chimeric and truncated forms of human complement protein C8␣ reveal binding sites for C8␤ and C8␥ within the membrane attack complex/perforin region. Biochemistry 38, 8478–8484. Plumb, M.E., Sodetz, J.M., 2000. An indel within the C8␣ subunit of human complement C8 mediates intracellular binding of C8␥ and formation of C8␣–␥. Biochemistry 39, 13078–13083. Podack, E.R., Olsen, K.J., Lowery, D.M., Lichtemheld, M., 1988. Structure and function of perforin. Curr. Top. Microbiol. Immunol. 140, 11–17. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. E 4, 406–425. Schreck, S.F., Parker, C., Plumb, M.E., Sodetz, J.M., 2000. Human complement protein C8␥. Biochim. Biophys. Acta 1482, 199–208. Steckel, E.W., York, R.G., Monahan, J.B., Sodetz, J.M., 1980. The eighth component of human complement: purification and physiochemical characterization of its unusual subunit structure. J. Biol. Chem. 255, 11997–12005. Suzuki, N.M., Satoh, N., Nonaka, M., 2000. C6-like and C3-like molecules from the cephalochordate, amphioxus, suggest a cytolytic complement system in invertebrates. J. Mol. E 54, 671–679. Thomson, J.D., Higgins, D.G., Gibson, T.J., 1994. Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucl. Acids Res. 22, 4673–4680. Volanakis, J.E., 1998. Overview of the complement system. In: Volanakis, J.E. (Ed.), The Human Complement System in Health and Disease. New York, pp. 9–32. Zarkadis, I.K., Sarrias, M.R., Sfyroera, G., Sunyer, O.J., Lambris, J.D., 2001. Cloning and structure of three rainbow trout C3 molecules: a plausible explanation for their functional diversity. Dev. Comp. Immunol. 25, 11–24. Zarkadis, I.K., Duraj, S., Chondrou, M., 2005. Molecular cloning of the seventh component of complement in rainbow trout. Dev. Comp. Immunol. 29, 95–102.