(Salmo salar) and rainbow trout - Springer Link

2 downloads 0 Views 376KB Size Report
BRIEF COMMUNICATION. Kazuhiro Fujiki · Julie Gauley · Niels Bols .... shown to be enhanced in some renal tumours (Kuroda et al. 2001), while expression of ...
Immunogenetics (2002) 54:604–609 DOI 10.1007/s00251-002-0506-0

B R I E F C O M M U N I C AT I O N

Kazuhiro Fujiki · Julie Gauley · Niels Bols Brian Dixon

Cloning and characterization of cDNA clones encoding CD9 from Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss) Received: 18 February 2002 / Revised: 2 September 2002 / Published online: 11 October 2002 © Springer-Verlag 2002

Abstract In comparison to mammals, relatively few of the molecules involved in teleost immune responses have been isolated and characterized. A rapid method of isolating molecules important for immune function is subtractive hybridization. One such experiment using infectious hematopoietic necrosis virus-infected Atlantic salmon produced several cDNA clones with similarity to mammalian immune-specific genes, including granzyme M (accession no. AF434669) and CD9. After cloning the rainbow trout version of CD9, sequence analysis showed that both salmonid sequences contained many features of the tetraspanin receptor family to which CD9 belongs. Phylogenetic analysis revealed a close association of the trout and salmon sequences to known CD9 and CD81 receptors. Southern blotting demonstrated that the rainbow trout gene is single copy. Reverse transcriptase PCR showed strong expression of this clone in many tissues, but liver expression was very low – an observation consistent with the clone being a CD9, not a CD81, equivalent. The evidence suggests that the sequences reported here are bona fide teleost CD9 homologues and we are currently producing recombinant proteins and polyclonal antisera for use in functional studies. Keywords Cluster of differentiation · Teleost · Tetraspanin · Subtractive hybridization In comparison to the vast array of molecules and reagents available for studying immunology in mammals, very few are available to assist in the understanding of teleost immunology. Suppression subtractive hybridization experiments have allowed us to clone and characterThe nucleotide sequences reported in this paper have been submitted to the GenBank database under the following accession numbers: AF434669, AF427519, AF425839 K. Fujiki · J. Gauley · N. Bols · B. Dixon (✉) Department of Biology, University of Waterloo, 200 University Ave. West, Waterloo, Ontario, Canada N2L 3G1 e-mail: [email protected] Tel.: +1-519-888-4567-x2665, Fax: +1-519-746-0614

ize cytokines, such as the chemokine CK2 (Lei et al. 2002), as well as molecules important for antigen presentation such as class II associated invariant chain (K. Fujiki, L. Lei, R. Sundick and B. Dixon, unpublished data). One particularly useful group of molecules that is not represented by many teleost equivalents is the cluster of differentiation (CD) markers. Antibodies to these cell surface molecules are extremely useful for identifying and isolating sub-populations of lymphocytes that cannot be differentiated by morphology alone. In addition, these molecules often play integral roles in cell adhesion, signalling and activation. To date, seven other teleost CD molecules have been reported: catfish CD18 (Qian et al. 1999), rainbow trout CD8α (Hansen and Strassburger 2000), Japanese flounder CD3 (Park et al. 2001), CD45 from carp (Fujiki et al. 2000), pufferfish (Diaz del Pozo et al. 2000) and lamprey (Nagata et al. 2002) and, of particular importance to this study, zebrafish CD81 (Yoder and Litman 2000) – a tetraspanin family member. The tetraspanin receptor superfamily members have been described as “molecular facilitators” (Maecker et al. 1997), since they form a wide range of associations with many other cell surface molecules, particularly other tetraspanins and integrins, to increase the formation and stability of signalling complexes. For example, in macrophages, CD9 associates with Fcγ receptors to assist in cellular activation (Kaji et al. 2001). The signals these molecules produce initiate numerous other cellular functions as well, such as proliferation, differentiation, migration and adhesion. Tetraspanins cross the cell membrane four times, form two extracellular loops and are found in a wide range of organisms from nematodes to vertebrates. CD9 is a tetraspanin family member with a wide tissue distribution that seems to have a specific role in leukocytes, particularly granulocytes, macrophages, pre-B cells and T cells. One tissue it is not predominantly expressed in is liver, where most of its expression is restricted to Kupffer cells, a monocyte subtype (Warskulat et al. 1999), although other tetraspanin family members such as CD81 are expressed strongly in liver (Inoue et al. 2001; Pileri et al. 1998; Tseng and

605

Fig. 1 Alignment of the nucleotide sequences of Atlantic salmon and rainbow trout CD9 clones. Nucleotides of rainbow trout CD9 identical to those of Atlantic salmon CD9 are shown by dashes. Stars indicate gaps introduced for optimal alignment. The numbers on the right indicate nucleotide position along the Atlantic salmon sequence. Amino acid sequences appear above (salmon) and below (trout) the nucleotide sequences, with the beginning of each protein domain noted above the salmon sequence

Klimpel 2002). CD9 has been shown to associate with several co-stimulatory molecules, such as CD3, CD4 and CD5 (Toyo-oka et al. 1999). In T cells, CD9 has recently been shown to not only co-ordinate with CD28 in transducing the second signal required for activation (Toyo-oka et al. 1997; Yashiro-Ohtani et al. 2000) but in CD28-deficient mice it can replace CD28, providing a signal in combination with T-cell receptor engagement

606

Fig. 2 Alignment of the deduced amino acid sequences of Atlantic salmon, rainbow trout and other vertebrate CD9 proteins. The amino acid number of Atlantic salmon CD9 is presented on the right side. Amino acid residues identical to the Atlantic salmon CD9 sequence are shown by dashes. Stars indicate gaps introduced for the optimal alignment. N-Glycosylation sites are shown by bold italics. The percentages at the end of each sequence represent amino acid identities to Atlantic salmon CD9

that is sufficient for T-cell activation (Tai et al. 1996). It has long been known that CD9 is up-regulated upon T-cell activation (Warskulat et al. 1999). We report here a cDNA clone with a high degree of sequence similarity to tetrapod CD9 genes, which was isolated during a suppression subtractive hybridization experiment (methods described in Fujiki et al. 2000) using head kidney mRNA of Atlantic salmon (Salmo salar) exposed to infectious hematopoietic necrosis (IHN) virus at 1×104 pfu/fish for 5 days. The full-length sequence of clone M-AT18 has a 2,229 bp nucleotide sequence (Fig. 1) encoding a putative 230 amino acid (aa) sequence that should produce a Mr 24,400 protein (Fig. 2), with 105 bp of 5′- and 1,434 bp of 3′-untranslated regions. A BLASTP search with this amino acid sequence produced significant matches with mammalian CD9 sequences with E-values of 2 × 10–25 to 1 × 10–28. In order to further characterize the potential CD9 gene, we cloned the equivalent from a cDNA library constructed in pcDNA3.1 vector (Invitrogen) with phytohemagglutininstimulated head kidney non-adherent cells of rainbow trout (Oncorhynchus mykiss), a species we could keep in the laboratory. The library was subjected to PCR using Atlantic salmon CD9 primers (5′-AAGTGTCTTCCTGCCTGCTG-3′, bp 368–387; 5′-CTCTTGATCTGTTTCACGAG-3′, bp 770–751). Based on the internal sequence of this fragment, gene-specific primers were selected and anchored PCR was performed to extend the sequence of the 5′ and 3′ regions using the pcDNA3.1 forward primer 5′-acgactcactatagggagac-3′ (bp 867–886) and the bovine growth hormone reverse primer (bp

1,039–1,022) were used as a 5′ anchored primer and a 3′ anchored primer, respectively. Because the two fragments overlapped within the internal primers, a third fragment, which overlapped the two previous fragments, was amplified to confirm the sequence upstream of the primers. The consensus nucleotide sequence of rainbow trout CD9 is 1,142 bp with 165 bp and 287 bp of 5′ and 3′ untranslated regions (Fig. 1), respectively, and an open reading frame encoding a 230 aa sequence that is predicted to produce a Mr 24,900 protein (Fig. 2). Three clones for each fragment were sequenced to check for Taq polymerase errors, and consensus full-length sequence was obtained by comparing these three contiguous and overlapping sequences. The coding sequences of the two salmonid clones are 91% identical at the nucleotide level and 80% identical at the amino acid level, whereas amino acid identities between the Atlantic salmon and the tetrapod CD9 sequences are between 30.1% and 33.6% (Fig. 2). Analysis of the putative salmonid CD9 sequences using the Prosite (Falquet et al. 2002) sequence analysis program indicated that the putative salmon CD9 has only three amino acids, D77, T85 and A86, that do not match the 23 aa tetraspanin motif (PS00421) present in tetrapod CD9 and other tetraspanin family members, while the trout CD9 sequence has only one mismatch to the motif – D77. In similar analysis using the program Blocks (Henikoff and Henikoff 1994) the three highest scoring matches were to tetraspanin family motifs: Blocks accession numbers IPB000301B, IPB000301A and IPB000301C, respectively. Additional analysis for transmembrane domains using the DAS program (Cserzo et al. 1997) at the most strict cut-off level indicated that both salmonid sequences contained four such regions at positions 12–32, 55–74, 89–103 and 195–214. Several other important amino acid residues conserved in the tetraspanin family are also present in the salmonid sequences, including K11 in the amino terminal cytoplasmic domain; L14, N18, L23 G25 and G30 in the first transmembrane domain; L62, G65 and G75 in the

607

second transmembrane domain; E82 in the second cytoplasmic domain; L94, E101 (only in the trout), and I102 in the third transmembrane domain; Q118, the CCG at aa 151–153, the CP at aa 166–167, and C180 in the EC2 domain; and finally, L216 in the C-terminal cytoplasmic domain (Fig. 2). The salmon sequence has one potential N-linked glycosylation site at aa positions 83–85, while the rainbow trout sequence does not contain any of these sites (Fig. 2). These features indicate that two salmonid cDNA clones are tetraspanin family members and, based on the fact that they were both isolated from libraries enriched in activated lymphocytes and they show the highest sequence similarity to mammalian CD9, they may represent bona fide teleost homologues of CD9. This would be the eighth teleost equivalent of a mammalian CD molecule to be reported, and only the second tetraspanin family member after zebrafish CD81 (Yoder and Litman 2000). While both salmonid sequences possess many of the features of the tetraspanin family, the trout sequence has more of the conserved family-specific residues (Fig. 2). Although neither salmonid sequence has an N-linked glycosylation site in the EC1 domain, there is one such potential site in the second cytoplasmic domain of the Atlantic salmon sequence, a position that suggests that it is not glycosylated in vivo. While the glycosylation site shown at amino acid positions 50–53 in Fig. 2 is present in most known CD9 sequences, it must not be essential for CD9 function, since the cat sequence also contains no glycosylation sites (Willett and Neil 1995). This suggests that the lack of glycosylation sites in the salmonid sequences does not disqualify them as the teleost CD9 equivalents. A phylogenetic tree was constructed using our cDNA sequences, known tetrapod CD9 sequences, other human tetraspanin family member sequences and the zebrafish CD81 (Yoder and Litman 2000) sequence (Fig. 3). The salmon and trout sequences group with each other 100% of the time, and are within a strongly supported (94%) cluster containing the tetrapod CD9 sequences plus a strong cluster containing the human and zebrafish CD81 sequences, which is one of the tetraspanin family members most closely related to CD9. The human CD53, CD63 and il-TMP sequences, also members of the tetraspanin family, cluster outside the CD9/CD81 group. This suggests that the two salmonid sequences are closely related to the tetraspanin family of receptors, specifically to CD9 and CD81. Within the CD9/CD81 cluster, the mammalian CD9 sequences cluster together 100% of the time and they group with the chicken CD9 sequence very strongly (96%). The salmonid sequences, however, group in a different cluster with the human/zebrafish CD81 pair, although the clustering of these two groups is not strongly supported (49%). The grouping of the salmonid CD9 sequences in a very strongly supported cluster with the other known CD9 sequences supports the hypothesis that these sequences represent bona fide teleost equivalents. Southern blotting revealed that rainbow trout CD9 is a single copy gene, since only one to two bands are ob-

Fig. 3 A phylogenetic tree of vertebrate tetraspanin receptor family members. Coding sequence nucleotides from salmon and trout CD9 clones were aligned with corresponding regions of mammalian and chicken CD9 genes, as well as other human representatives of the tetraspanin family using the program Clustal W (Thompson et al. 1994). Aligned sequences were used to create a phylogenetic tree in the program MEGA (Kumar et al. 1994). The tree was constructed by the neighbour joining method (Saitou and Nei 1987) using the Jukes and Cantor correction (Jukes and Cantor 1969), with pairwise deletion and 1000 bootstrap replications. Numbers above the line indicate bootstrap confidence values derived from 1,000 replications. The known vertebrate tetraspanin sequences were obtained from the GenBank database and their accession numbers are as follows: Human CD9 – XM006948, Cat CD9 – D30786, Cow CD9 – M81720, Mouse CD9 – NM007657, Rat CD9 – X76489, Chicken CD9 – AB032767, Human CD81 – BC002978, Zebrafish CD81 – AF295377, Human CD53 – NM000560, Human CD63 – BC002349 and Human il-TMP – U31449

served in each lane (Fig. 4). This is interesting, since several other immune system genes that are single copy in mammals have proven to be present in multiple copies in rainbow trout, notably beta-2-microglobulin (Shum et al. 1996), interleukin 1 (Pleguezuelos et al. 2000), the chemokine CK2 (Lei et al. 2002) and invariant chain (K. Fujiki, L. Lei, R. Sundick and B. Dixon, unpublished data). This is thought to be the result of a tetraploidization event that duplicated the entire salmonid genome about 27 million years ago (McKay et al. 1996). However, modern salmonids are thought to be in the process of diploidizing their genome, a process that is often seen in critical immune system genes that cause problems when their products are overproduced, like major histocompatibility receptor genes in polyploid Xenopus species (Courtet et al. 2001; Shum et al. 1993). This can be seen in some trout immune system genes that are single copy despite the tetraploid nature of their genome, for example the chemokine CK1 (Dixon et al. 1998; Shum et al. 1999). CD9 also appears to be in a region of the trout genome that has diploidized. Tetrapod CD9 genes are members of a large family, but they are generally single copy genes. This again supports the hypothesis that our clones encode the teleost homologue of CD9. Mammalian CD9 is ubiquitously expressed, with high levels of receptor found on platelets, eosinophils, basophils, pre-B cells, activated T cells and neural cells. In RT-PCR analysis, CD9 mRNA in rainbow trout was abundant in peripheral blood leukocytes, head kidney, body kidney, spleen, heart, and gill, was present at lower

608

Fig. 4 Southern blotting analysis of rainbow trout CD9 gene. Ten micrograms of genomic DNA was digested with BamHI, EcoRI, EcoRV, HindIII, or PstI and hybridized with a DIG-labelled genomic probe (819 bp), which corresponds to cDNA sequence of bp 761–839. The membrane was washed twice at 65°C for 30 min in 0.1× SSC containing 0.1% SDS. An alkaline phosphataseconjugated anti-DIG antibody and CSPD were used to detect the hybridized DNA fragments according to the protocol of the manufacturer (Roche). DNA fragments bound by the antibody were visualized using a FluorChem 8000 Imager (Alpha Innotech Corporation, San Leandro, Calif.). The positions of the origin and the size marker (λ/HindIII digest) are shown in the left margin

levels in muscle and was present at very low levels in liver (Fig. 5). This is probably a reflection of the expression of CD9 on immune system cells within these tissues. Rainbow trout CD9 was also expressed at high levels in heart and posterior kidney. Expression of human CD9 in human kidney has been reported recently and has been shown to be enhanced in some renal tumours (Kuroda et al. 2001), while expression of CD9 on endomysium and perimysium, but not on muscle cells themselves, has been shown for both cardiac and skeletal muscle in humans (Nakamura et al. 2001). The extremely low expression of trout CD9 in liver is of particular importance in deciding whether it represents the equivalent of mammalian CD9 or CD81. While in mammals CD9 is expressed in liver, it is expressed at very low levels in parenchymal cells and is strongly expressed only on Kuppfer cells (Warskulat et al. 1999), which means an overall low expression in this organ. CD81, on the other hand is strongly expressed in liver cells, so much so that down-regulation of this receptor is associated with diseases like cancer (Inoue et al. 2001) and it is the target of one of the immune evasion strategies of hepatitis C virus (Pileri et al. 1998; Tseng and Klimpel 2002). The low expression of the trout version strongly suggests a functional relationship with mammalian CD9, not CD81. The structural and expression pattern evidence presented above suggests that the sequences reported here do represent teleost equivalents of CD9, or a sequence that is very closely related – as much as human CD81 is

Fig. 5 RT-PCR analysis showing tissue distribution of rainbow trout CD9 mRNA expression. PCR was performed in a PTC-100 HB (MJ Research, Watertown, Mass.), using the following parameters: 95°C, 5 min; 25–33 cycles at (95°C for 30 s, 55°C for 30 s, 72°C for 1 min) and a final extension at 72°C for 5 min. Oligonucleotides used as PCR primers are as follows: rainbow trout elongation factor EfTu-1 (200 bp) sense primer 5′-GAGTGAGCGCACAGTAACAC-3′ (bp 19–38), antisense primer 5′-AAAGAGCCCTTGCCCATCTC-3′ (bp 199–198); rainbow trout CD9 (452 bp) sense primer 5′-AGCTGTGCAAGTGTTTCCTC-3′ (bp 188–207), antisense primer 5′-CAAGGCACCAATGAGTCCAC-3′ (bp 620–639). (The numbers represent the nucleotide numbers of each cDNA sequence in GenBank.) EfTu-1 is the housekeeping protein elongation factor Tu-1, used here as an internal control (Hansen 1997; Hansen and Strassburger 2000). Each experiment was repeated twice and changes in the amount of each specific mRNA at each time point was measured by comparing the number of pixels in each experimental band with the number of pixels in the EfTu-1 control band using the computer program NIH Image. The ratio of these values was compared between control and treated samples to determine if a change in mRNA expression had occurred. RNA extracted from various tissues was used as a template for CD9 amplification, along with the housekeeping gene EfTu-1 (elongation factor Tu-1) used to confirm equal total RNA loading

related to human CD9 or more. The receptors encoded by the sequences reported here are probably involved in activation and signal transduction in trout leukocytes and we are currently producing polyclonal antisera that will be used to elucidate these functions. Acknowledgements The authors would like to thank Dr Garth Traxler for assistance in infecting the Atlantic salmon. This research was funded by Natural Sciences and Engineering Research Council Individual Grant number 217529–1999 to B.D.

References Courtet M, Flajnik M, Du Pasquier L (2001) Major histocompatibility complex and immunoglobulin loci visualized by in situ hybridization on Xenopus chromosomes. Dev Comp Immunol 25:149–157 Cserzo M, Wallin E, Simon I, von Heijne G, Elofsson A (1997) Prediction of transmembrane alpha-helices in prokaryotic membrane proteins: the dense alignment surface method. Protein Eng 10:673–676

609 Diaz del Pozo E, Beverley PC, Timon M (2000) Genomic structure and sequence of the leukocyte common antigen (CD45) from the pufferfish Fugu rubripes and comparison with its mammalian homologue. Immunogenetics 51:838–846 Dixon B, Shum B S, Adams E J, Magor KE, Hedrick RP, Muir DG, Parham P (1998)CK-1, a putative chemokine of rainbow trout (Oncorhynchus mykiss). Immunol Rev 166:345–352 Falquet L, Pagni M, Bucher P, Hulo N, Sigrist CJ, Hofmann K, Bairoch A (2002) The PROSITE database, its status in 2002. Nucleic Acids Res 30:235–238 Fujiki K, Shin DH, Nakao M, Yano T (2000) Molecular cloning of carp (Cyprinus carpio) leucocyte cell-derived chemotaxin 2, glia maturation factor beta, CD45 and lysozyme C by use of suppression subtractive hybridisation. Fish Shellfish Immunol 10:643–650 Hansen JD (1997) Characterization of rainbow trout terminal deoxynucleotidyl transferase structure and expression. TdT and RAG1 co-expression define the trout primary lymphoid tissues. Immunogenetics 46:367–375 Hansen JD, Strassburger P (2000) Description of an ectothermic TCR coreceptor, CD8 alpha, in rainbow trout. J Immunol 164:3132–3139 Henikoff S, Henikoff JG (1994) Protein family classification based on searching a database of blocks. Genomics 19:97–107 Inoue G, Horiike N, Onji M (2001) The CD81 expression in liver in hepatocellular carcinoma. Int J Mol Med 7:67–71 Jukes TH, Cantor CR (1969) Evolution of protein molecules. In: Munro HN (ed) Mammalian protein metabolism.Academic Press, New York, pp 21–132 Kaji K, Takeshita S, Miyake K, Takai T, Kudo A (2001) Functional association of CD9 with the Fc gamma receptors in macrophages. J Immunol 166:3256–3265 Kumar S, Tamura K, Nei M (1994) MEGA: Molecular Evolutionary Genetics Analysis software for microcomputers. Comput Appl Biosci 10:189–191 Kuroda N, Inoue K, Guo L, Miyazaki E, Hayashi Y, Naruse K, Toi M, Hiroi M, Shuin T, Enzan H (2001) Expression of CD9/motility-related protein 1 (MRP-1) in renal parenchymal neoplasms: consistent expression in papillary and chromophobe renal cell carcinomas. Hum Pathol 32:1071–1077 Lei L, Fujiki K, Dixon B, and Sundick R (2002) Cloning of CK2, a CC chemokine with a CXXXC type structure from rainbow trout (Oncorhynchus mykiss). Cytokine 17:71–81 Maecker HT, Todd SC, Levy S (1997) The tetraspanin superfamily: molecular facilitators. FASEB J 11:428–442 McKay S.J, Devlin RH, Smith MJ (1996) Phylogeny of Pacific salmon and trout based on growth hormone type-2 and mitochondrial NADH dehydrogenase subunit 3 DNA sequences. Can J Fish Aquat Sci 53:1165–1176 Nagata T, Suzuki T, Ohta Y, Flajnik MF, Kasahara M (2002) The leukocyte common antigen (CD45) of the Pacific hagfish, Eptatretus stoutii: implications for the primordial function of CD45. Immunogenetics 54:286–291 Nakamura Y, Handa K, Iwamoto R, Tsukamoto T, Takahasi M, Mekada E (2001) Immunohistochemical distribution of CD9, heparin binding epidermal growth factor-like growth factor, and integrin alpha3beta1 in normal human tissues. J Histochem Cytochem 49:439–444 Park CI, Hirono I, Enomoto J, Nam BH, Aoki T (2001) Cloning of Japanese flounder Paralichthys olivaceus CD3 cDNA and gene, and analysis of its expression. Immunogenetics 53:130–135

Pileri P, Uematsu Y, Campagnoli S, Galli G, Falugi F, Petracca R, Weiner AJ, Houghton M, Rosa D, Grandi G, Abrignani S (1998) Binding of hepatitis C virus to CD81. Science 282:938– 941 Pleguezuelos O, Zou J, Cunningham C, Secombes CJ (2000) Cloning, sequencing, and analysis of expression of a second IL-1beta gene in rainbow trout (Oncorhynchus mykiss). Immunogenetics 51:1002–1011 Qian Y, Ainsworth AJ, Noya M (1999) Identification of a beta 2 (CD18) molecule in a teleost species, Ictalurus punctatus Rafinesque. Dev Comp Immunol 23:571–583 Saitou N, Nei M (1987) The neighbour-joining method: a new method for reconstructing phyogenetic trees. Mol Biol Evol 4:406–425 Shum BP, Avila D, Du Pasquier L, Kasahara M, Flajnik MF (1993) Isolation of a classical MHC class I cDNA from an amphibian: evidence for only one class I locus in the Xenopus MHC. J Immunol 151:5376–5386 Shum BP, Azumi K, Zhang S, Kehrer SR, Raison RL, Detrich HW, Parham P (1996) Unexpected beta2-microglobulin sequence diversity in individual rainbow trout. Proc Natl Acad Sci USA 93:2779–2784 Shum BP, Rajalingam R, Magor KE, Azumi K, Carr WH, Dixon B, Stet RJ, Adkison MA, Hedrick RP, Parham P (1999) A divergent non-classical class I gene conserved in salmonids. Immunogenetics 49:479–490 Tai XG, Yashiro Y, Abe R, Toyooka K, Wood CR, Morris J, Long A, Ono S, Kobayashi M, Hamaoka T, Neben S, Fujiwara H (1996) A role for CD9 molecules in T cell activation. J Exp Med 184:753–758 Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680 Toyo-oka K, Tai XG, Yashiro Y, Ahn HJ, Abe R, Hamaoka T, Kobayashi M, Neben S, Fujiwara H (1997) Synergy between CD28 and CD9 costimulation for naive T-cell activation. Immunol Lett 58:19–23 Toyo-oka K, Yashiro-Ohtani Y, Park CS, Tai XG, Miyake K, Hamaoka T, Fujiwara H (1999) Association of a tetraspanin CD9 with CD5 on the T cell surface: role of particular transmembrane domains in the association. Int Immunol 11:2043– 2052 Tseng CT, Klimpel GR (2002) Binding of the hepatitis C virus envelope protein E2 to CD81 inhibits natural killer cell functions. J Exp Med 195:43–49 Warskulat U, Hammermann R, Haussinger D (1999) Lack of osmosensitive CD9 expression in rat liver cells. Cell Physiol Biochem 9:99–105 Willett BJ, Neil JC (1995) cDNA cloning and eukaryotic expression of feline CD9. Mol Immunol 32:417–423 Yashiro-Ohtani Y, Zhou XY, Toyo-Oka K, Tai XG, Park CS, Hamaoka T, Abe R, Miyake K, Fujiwara H (2000) Non-CD28 costimulatory molecules present in T cell rafts induce T cell costimulation by enhancing the association of TCR with rafts. J Immunol 164:1251–1259 Yoder JA, Litman GW (2000) The zebrafish fth1, slc3a2, men1, pc, fgf3 and cycd1 genes define two regions of conserved synteny between linkage group 7 and human chromosome 11q13. Gene 261:235–242