Channel catfish cytotoxic cells: a mini-review

Channel catfish cytotoxic cells: a mini-review

Developmental and Comparative Immunology 26 (2002) 141±149 Channel cat®sh cytotoxic cells: a mini-review Linling S...

146KB Sizes 2 Downloads 55 Views

Developmental and Comparative Immunology 26 (2002) 141±149

Channel cat®sh cytotoxic cells: a mini-review Linling Shen a, Tor B. Stuge b, He Zhou a, Morad Khayat a, Katherine S. Barker a, Sylvie M.A. Quiniou a, Melanie Wilson a, Eva BengteÂn a, V. Gregory Chinchar a, L. William Clem a, Norman W. Miller a,* a

Department of Microbiology, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216, USA b Division of Hematology, Stanford University School of Medicine, Stanford, CA 94305-5156, USA

Abstract The use of allogeneic and autologous lymphoid cell lines has facilitated studies of cytotoxic T lymphocytes (CTL) and natural killer (NK)-like cells in channel cat®sh. NaõÈve cat®sh leukocytes were shown to spontaneously kill allogeneic cells and virally-infected autologous cells without the need for prior sensitization, and allogeneic cytotoxic responses were greatly enhanced by in vitro alloantigen stimulation. Both cat®sh CTL and NK-like cells have been successfully cloned from these alloantigen-stimulated cultures, and represent the ®rst cytotoxic cell lines derived from any ectothermic vertebrate. These cloned cytotoxic cells contain granules and likely induce apoptosis in sensititive targets via a putative perforin/granzyme mechanism. In addition, some cat®sh CTL clones may also kill targets by an additional mechanism, possibly by Fas/FasL-like interactions. Importantly, these cytotoxic cells do not express the marker for cat®sh nonspeci®c cytotoxic cells (NCCs), and thus represent cell types distinct from NCCs. The use of monoclonal antibodies against the cat®sh F and G immunoglobulin light chain isotypes revealed the presence of a putative Fc receptor for IgM (FcmR) on some cat®sh NK-like cells that appears to `arm' these cells with surface IgM. In addition, a potentially important monoclonal antibody (CC41) developed against cat®sh NK-like cells was found to recognize an ~150 kDa molecule on the surface of cat®sh cytotoxic cells. These studies clearly demonstrate that cat®sh possess an array of different cytotoxic cells. The availability of various cloned cytotoxic cell lines should enable unambiguous functional studies to be performed in ways not currently possible with any other ®sh species. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Teleost; NK-like cells; CTL; FcmR; ADCC; Alloantigen

1. Introduction Mammalian natural killer (NK) cells and cytotoxic T-lymphocytes (CTL) play important roles in the * Corresponding author. Tel.: 11-601-984-1719; fax: 11-601984-1708. E-mail address: [email protected] (N.W. Miller). Abbreviations: PBL, peripheral blood leukocytes; NCCs, nonspeci®c cytotoxic cells; CTL, cytotoxic T lymphocytes; NK, natural killer; ADCC, antibody-dependent cellular cytotoxicity; MLC, mixed leukocyte culture; LPS, lipopolysaccharide; ConA, concanavalin A; FcmR, Fc receptor for IgM.

surveillance and destruction of foreign or infected tissues and cells. NK cells in mammals were originally described by their ability to kill certain tumor cells in an apparently non-MHC restricted manner without need for prior sensitization [1,2]. It is now known that these cells represent a distinct lymphocyte subset that arises from a common NK/T cell progenitor [3]. Moreover, both NK cells and CTL are granular lymphocytes that express several cell surface markers in common, and have similar effector functions, i.e. cytotoxicity and cytokine secretion [4,5]. Mammalian NK cells are distinguished from CTL by expression of

0145-305X/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 0145-305 X(01)00 056-8


L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149

CD16 (FcgRIII, a low af®nity receptor for IgG) and CD56 as well as by the lack of a rearranged TCR [6±10]. In CTL, rearrangement of the TCR genes enables recognition of foreign peptides presented by `self' MHC molecules. This recognition, in turn, activates the CTL to kill foreign peptide-presenting cells. In contrast, NK cells do not recognize presented antigens, but rather recognize common determinants associated with `self' MHC class I molecules, and this recognition inhibits NK cytotoxic responses. For example, if one of the `self' MHC class I allotypes is absent from a cell, NK cells will spontaneously kill it due to `missing self'. However, if an NK cell detects the appropriate MHC class I molecules, the potential target cell is spared [11]. The discovery and characterization of murine and human NK inhibitory and activating receptors that bind various MHC class I allotypes have given strong support for the `missing self' model of NK cell recognition and activation [12±20]. Evidence that ®sh possess cytotoxic cells has come from in vivo and in vitro studies. For example, allograft rejections and graft versus host reaction have suggested that ®sh are capable of mounting antigen speci®c cytotoxic responses [21,22]. Recent in vitro studies have shown that leukocytes obtained from previously immunized ®sh can speci®cally kill a variety of target cells that include hapten-modi®ed autologous cells [23], allogeneic erythrocytes [24] and allogeneic cell lines [25]. In addition, it has been shown that ®sh leukocytes can spontaneously kill a variety of xenogeneic cells without the need for prior immunization [26,27]. These in vivo and in vitro studies suggest by analogy to mammalian immune responses that ®sh possess cytotoxic T lymphocytes and NK-like cells. However, in contrast to mammalian cytotoxic cells, little is known about the identity or function of cytotoxic cells in teleosts. This has been due, in part, to a lack of immunological reagents that de®ne ®sh lymphocyte subpopulations, as well as to a dearth of functional in vitro culture systems for most ®sh species. The availability of an immunologically relevant in vitro culture system for channel cat®sh leukocytes has enabled a variety of cytotoxic cell types to be identi®ed and partially characterized. This review summarizes recent published and unpublished evidence demonstrating that cat®sh have an array of different types of

cytotoxic cells that function similarly to mammalian NK cells and CTL. 2. Nonspeci®c cytotoxic cells (NCCs) The most extensively studied cytotoxic cells in teleosts are the nonspeci®c cytotoxic cells (NCCs). Originally described in channel cat®sh, these cells are able to spontaneously kill a variety of xenogeneic targets, including certain ®sh parasites and traditional mammalian NK cell targets [26,27]. Unlike mammalian NK cells, cat®sh NCCs are small agranular lymphocytes that are commonly found in lymphoid tissues (i.e. pronephros and spleen), but rarely in the blood. These cells have been de®ned by reactivity with a monoclonal antibody (5C6), which reacts with a 32±34 kD cell surface protein termed NCC receptor protein 1(NCCRP-1) [28,29]. This receptor is believed to recognize a single, highly conserved target antigen (NKTag) found on potential targets ranging from protozoan parasites to human tumor cells. Furthermore, ligation of NCCRP-1 by either mAb 5C6 or NKTag results in NCC activation [30]. In addition to cat®sh, NCC-like activity has been shown in other ®sh species, including rainbow trout [31,32], carp [33,34], damsel®sh [35], and tilapia [36]. Recently, the gene for channel cat®sh NCCRP-1 was sequenced, and found to be a novel type III membrane protein that has no sequence homology to any known mammalian leukocyte receptor [29]. Based on the cross reactivity of mAb 5C6 with mammalian NK cells and the ability of ®sh NCCs to kill traditional mammalian NK targets, NCCs have been postulated to be the evolutionary precursor of mammalian NK cells [26,28,29]. However, formal proof of this notion will require demonstrating that NCCRP-1 gene homologue is expressed in mammalian NK cells. As presented below, there is now considerable evidence to indicate that NCCs are not the only type of ®sh cytotoxic cells capable of spontaneously killing susceptible targets. 3. Spontaneous allogeneic cytotoxic responses Cytotoxic assays employing various channel cat®sh leukocyte lines as targets revealed that PBL from naõÈve cat®sh contain effector cells which

L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149

spontaneously kill allogeneic, but not autologous, target cells. This killing is nonspeci®c and not MHC restricted, i.e. the killing of a particular allogeneic cell can be inhibited by different unlabeled allogeneic cells (`cold target') [37]. Importantly, the failure to inhibit allokilling either by the addition of blocking amounts of mAb 5C6, excess xenogeneic cells (i.e. traditional mammalian NK cell targets), or removal of 5C6 reactive cells, argues strongly that the effector cells responsible for the observed allokilling are not NCCs, but rather a distinct NK-like effector [38]. These ®sh NK-like cells form conjugates with and kill allogeneic targets by an apoptotic mechanism [39]. Mammalian cytotoxic cells induce apoptosis in targets by two different mechanisms: one is a calcium-dependent perforin/granzyme-mediated secretory lytic pathway, and the other is a calciumindependent nonsecretory lytic pathway induced through interaction of cell surface molecules such as Fas/FasL [40]. The complete inhibition of allogeneic killing by calcium chelation using EGTA suggests that cat®sh NK-like cells utilize a secretory (i.e. perforin/granzyme), rather than a ligand-based (i.e. Fas/FasL), mechanism to trigger apoptosis [39]. Furthermore, spontaneous killing of allogeneic cells could be inhibited by mAb 1H5 that reacts with a putative leukocyte-function-associated antigen (LFA)1-like molecule, suggesting a possible role of adhesion molecules in conjugate formation [37]. In addition to lysis of allogeneic targets, PBL from naõÈve cat®sh were shown to contain cytotoxic effector cells that kill virus-infected autologous cells [41]. This killing was not abolished by the presence of excess allogeneic `cold target' or by the physical removal of NK-like effectors conjugated to allotargets. These ®ndings indicate the cytotoxic cells which kill virus-infected targets, may represent either a subpopulation of cat®sh NK-like cells or possibly another distinct population of cytotoxic cells. 4. Clonal cytotoxic cell lines In vitro allogeneic stimulation protocols utilizing lymphocytes or various allogeneic cell lines as stimulators have been shown to enrich for mammalian allospeci®c CTL, and in many cases NK cells [42,43]. In attempting to enrich for cat®sh cytotoxic


cells, similar in vitro stimulation studies were initiated using naõÈve cat®sh PBL as responder/effector cells and various irradiated allogeneic lymphoblastoid cell lines as stimulators[44]. These studies revealed that the cells expanded by alloantigen stimulation displayed greatly enhanced cytotoxic activity, i.e. greater than 80% killing at effector to target ratios of 1:1. However, alloantigen-generated cytotoxic responses were not restricted to the stimulator cells since many different allotargets were also effectively killed. The apparent lack of antigen speci®city was con®rmed in inhibition assays where the killing of any particular allotarget could be `cold target' inhibited by a variety of other allotargets. Although allogeneic targets were readily lysed, autologous targets were not, clearly demonstrating that these alloantigen-stimulated effectors could distinguish `self' from `nonself'. Signi®cantly, high levels of cytotoxicity were only generated by allostimulation protocols, whereas stimulation with mitogens such as LPS, ConA, or a combination of phorbol ester and calcium ionophore did not result in cytotoxic activity [44]. Cat®sh cytotoxic cells continued to proliferate in culture with weekly alloantigen stimulation and the addition of conditioned medium (as a presumed source of growth factors) from clonal autonomous T cell lines. However, the lack of appropriate cell surface markers precluded de®nitive identi®cation of the cytotoxic cell type(s) present in these heterogeneous alloantigen expanded cultures. To circumvent this problem, cytotoxic effectors were cloned from these cultures by limiting dilution in the presence of irradiated stimulator cells from allogeneic lymphoid lines and culture supernatants from T cell lines. Each clone was analyzed for TCR ab and Ig m expression by RT±PCR and for target cell speci®city by 51Cr-release assays [45]. This strategy proved to be very successful and multiple alloantigen-dependent cell lines were obtained using PBL from either nonimmune or alloantigen immunized cat®sh (see Fig. 1). Fourteen cloned cell lines, obtained from cultures initiated with PBL from a cat®sh immunized in vivo with an allogeneic B cell line (designated 3B11), were TCR ab positive. Each of these T cell lines expressed unique Va and Vb gene rearrangements indicating that they were clonal and derived from different precursors. Based upon their in vitro cytotoxic and


L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149

Fig. 1. Cytotoxic activity of ®ve representative clonal cat®sh cell lines developed by alloantigen stimulation. TS32.5 and TS32.32 were cloned from an immunized ®sh (immunized in vivo and stimulated in vitro with 3B11 cells). TS.75.2 was cloned from a nonimmune ®sh (stimulated in vitro with 3B11 cells). 2E12 and 3H9 were cloned from a nonimmune ®sh (stimulated in vitro with 1G8 cells). Cells from each clone were used as effectors at the indicated effector to target ratio in 4 h 51Cr release assays against allogeneic target cells (1G8 or 3B11) or for RT±PCR with cat®sh TCR a, TCR b and Ig m primers. Values represent the mean percent speci®c lysis of triplicate wells. Expression of a gene is indicated by 1.

proliferative responses, the T cell lines were placed into one of three groups. Group I (10 clones) contains antigen speci®c cytotoxic T cells that speci®cally proliferate in response to, and kill, only 3B11 cells. Group II (three clones) contains T cells that demonstrate broad speci®city since they proliferate to and kill some, but not all, allogeneic cells. Group III (one clone) contains non-cytotoxic T cells that speci®cally proliferate in response to irradiated 3B11 cells. This latter group may represent the cat®sh equivalent of alloantigen-speci®c T helper cells. Either EGTA or concanamycin A completely inhibited killing by the antigen-speci®c (group I) T cells, suggesting exclusive reliance on a secretory (perforin/granzyme) cytotoxic mechanism. However, these reagents only partially inhibited killing by the broadly speci®c (group II) T cells giving rise to the speculation that these cells may use both secretory and ligand-based (Fas/FasL) mechanisms for killing [46]. In contrast to the above situation employing cat®sh immune PBL, cultures initiated in vitro using two different allogeneic B cell lines (3B11 and 1G8) as stimulators of nonimmune PBL yielded exclusively TCR ab-negative cytotoxic cell clones (Fig.1).

Some of these cytotoxic cell clones proliferate in response to and kill a number of different allogeneic cells and appear to be nonspeci®c. Other clones appear to kill in a more speci®c fashion similar to the situation seen with certain cloned mammalian NK cells [47]. The question of whether these TCR ab-negative nonspeci®c cytotoxic clones represent TCR gd cells, NK-like cells, or both is not currently known due to lack of sequence information concerning cat®sh TCR g or d genes. Similar to the T cell lines mentioned above, these TCR ab-negative cytotoxic cells require weekly stimulation with both alloantigen and T cell culture supernatants for maximum viability and continued proliferation. In contrast to cat®sh NCCs, transmission electron microscopy revealed that the cells in each of the examined cytotoxic lines are granular. Depending on the cytotoxic cell line used, the cytotoxic responses of these cells can be partially to completely inhibited by EGTA. This latter ®nding is consistent with cells having a granular morphology and suggests they use a perforin/granzyme-like killing mechanism. Finally, it was noted that none of the TCR ab-negative cytotoxic cells react with the NCC-de®ning mAb 5C6,

L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149

Fig. 2. Flow analyses of cat®sh cell surface antigens recognized by CC41 or 9E1 mAbs on polyclonal or clonal NK-like cells. NK-like clones (2E10, 2E12 and 3H9) were derived from an alloantigenstimulated PBL culture (MLC). These cells were analyzed by ¯ow cytometry using mAbs 9E1 (anti-cat®sh m chain) or CC41, followed by PE-conjugated goat anti-mouse IgG1 (solid area). The control histograms represent negative staining using isotype matched mAb 1.14 (IgG1), speci®c for rainbow trout IgM (open area). All of these NK-like cells are negative for expression of TCR a, TCR b and Ig m as assessed by RT±PCR analyses. Polyclonal NK-like cells were passaged weekly more than 10 times before analysis.

indicating that they likely represent a population of cytotoxic cells distinct from NCCs. 5. Putative FcmR on cat®sh cytotoxic cells Flow cytometric analyses revealed the presence of cat®sh IgM on the surface of individual cells from some of the TCR ab-negative cytotoxic cell lines (Fig. 2). While some clones displayed high levels of surface IgM, others showed intermediate to low


levels, indicating heterogeneity among the clones with respect to surface IgM. Since these cells do not express message for either Ig m or L chains, surface IgM is most likely acquired passively from the cat®sh serum-supplemented culture medium. This ®nding suggests that some cat®sh cytotoxic cells may `arm' themselves with IgM via an FcmR mechanism. In the presence of serum-free medium, cell surface bound IgM was removed from these cells by modulation with an anti-cat®sh IgH chain speci®c mAb (9E1), and replaced with af®nity puri®ed cat®sh antibody to the trinitrophenol (TNP) moiety. Cell surface bound anti-TNP antibodies were detected by ¯ow cytometry using mAb 9E1 and by rosette formation using TNP-conjugated sheep erythrocytes (Fig. 3). To determine if IgM-`armed' cytotoxic cells are present in freshly isolated normal cat®sh PBL, two color ¯ow cytometry was conducted using mAbs speci®c for the two antigenically distinct isotypes (F and G) of channel cat®sh Ig L chains that are encoded by different L chain gene clusters [48±50]. It was found that approximately 4±8% of normal PBL have both L chain isotypes on their surface. The isolation and assay of these double L chain positive cells from normal PBL by ¯uorescent activated cell sorting revealed that they have high levels of nonspeci®c allocytotoxic activity. In addition, 9±12 days after allogeneic stimulation of normal PBL, the percentage of double light chain positive cells is greatly increased (. 40%). However, both the IgM-positive and IgM-negative populations from alloantigen stimulated cultures display high levels allocytotoxic activity, suggesting phenotypic heterogeneity among the cytotoxic cells present in these cultures (Fig. 4). Although ConA stimulated normal PBL also show increases in double light chain bearing cells, they possess very little, if any, spontaneous allocytotoxic activity (data not shown). These ®ndings suggest that some ®sh cytotoxic cells, and possibly some T cells, express an IgM binding molecule on their surface, presumably an FcmR molecule. Preliminary studies have indicated that cloned cytotoxic cells bearing the putative FcmR can kill otherwise resistant targets by an ADCC mechanism. For example, cat®sh cytotoxic cells `armed' with anti-TNP antibody effectively kill TNP-conjugated human B lymphoblastoid cells (IM9) which are refractory to killing by the same cytotoxic cells `armed' with normal cat®sh IgM


L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149

accessory molecule for various Fc receptors including CD16 [53], and NKp46, a recently described activating receptor found on human NK cells [17]. 6. A cell surface marker for cat®sh cytotoxic cells

Fig. 3. Rosette formation detects the presence of anti-TNP antibodies bound to cat®sh NK-like cells. IgM positive NK-like cytotoxic cells derived from an alloantigen stimulated culture were treated with mAb 9E1 in serum free medium to modulate the surface bound cat®sh IgM from the cell surface. The cells were washed and cultured in the presence of af®nity-puri®ed cat®sh anti-TNP antibody (100 mg/ml). After 20 h incubation, the antibody treated cells were mixed with TNP-haptenated sheep RBC (TNP±SRBC). Rosette formation was examined by light microscopy after 10 min. Greater than 95% of treated cells formed rosettes with TNP±SRBC when incubated with cat®sh anti-TNP antibody. No rosettes were observed with similarly treated control cells incubated with normal cat®sh IgM, or with use of unhaptenated SRBC in the rosette assays (data not shown).

(manuscript in preparation). The `arming' of cat®sh cytotoxic cells with antibody is reminiscent of mammalian mast cells that are `armed' with IgE by the high af®nity FceR [51]. In contrast, CD16 bearing mammalian NK cells bind soluble IgG with low af®nity, but bind IgG on target cells with higher af®nity [52]. It is of interest to note that recent RT±PCR and sequencing studies revealed that all channel cat®sh cytotoxic (both T and NK-like) cell lines tested express a signal transducing molecule with homology to the FceR g chain. Signi®cantly, the FceR g chain contains an immunoreceptor tyrosine-based activation motif (ITAM). In mammals, the g chain serves as an

Hybridoma production using a cat®sh TCR ab-negative NK-like cell line (clone 10.1) as an immunogen yielded a mAb (designated CC41) that recognizes an ~150 kDa surface molecule. This surface molecule is found at high densities on cells from TCR ab-negative cytotoxic clones (both IgMpositive and IgM-negative), and at low densities on cells from the cytotoxic T cell clones (Fig. 2). It is also found on approximately 8±14% of normal PBL. Cell separation studies revealed that cat®sh PBL reactive with mAb CC41 have high levels of nonspeci®c allocytotoxic activity. Moreover, the percentage of mAb CC41 reactive cells is greatly increased (.60%) after 7±12 days in allostimulated cultures (manuscript in preparation). This cell surface marker is of potential importance since it appears to identify ®sh cytotoxic cells in a fashion similar to that observed in mammals with CD56 [54]. 7. Conclusions There is now compelling evidence that the channel cat®sh, and probably other ®sh species, possess a number of different types of cytotoxic leukocytes including NCCs, NK-like cells, and CTL. It appears that each type of ®sh cytotoxic cell has a different target cell preference, a situation which may enable ®sh to destroy a diverse array of potential foreign targets through both innate and adaptive immune responses. The cat®sh cloned alloantigen-dependent CTL and NK-like cells represent the ®rst cytotoxic cell lines derived from any ectothermic vertebrate, and should facilitate functional studies in ways not currently possible with any other ®sh species. The future use of these clonal cell lines for both monoclonal antibody production and establishment of expressed sequence tag (EST) cDNA libraries should enable the development of additional molecular probes and immunological reagents that functionally de®ne the various cytotoxic cell types in ®sh. In addition, characterization of the putative FcmR

L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149


Fig. 4. Two-color ¯ow analyses and cytotoxic responses of freshly isolated and allostimulated PBL. Freshly isolated PBL or PBL stimulated for 12 days with irradiated 1G8 cells (MLC), were stained with mAbs 3F12 (IgG1) and 11A2 (IgG2b) which speci®cally react with the cat®sh Ig L chain isotypes F and G, respectively. This was followed by goat anti-mouse isotype-speci®c FITC- or PE-conjugated secondary reagents. The dot plots were divided into quadrants (QUAD) representing four different cell populations: (1) lower right (LR), B cells that express the G type L chain; (2) upper left (UL), B cells that express the F type L chain; (3) lower left (LL), cells that express neither L chain type; and (4) upper right (UR), cells that express both L chain types. The percentages (%) of cells in each quadrant are indicated in the bottom panel. Each of four different cell populations (based on L chain expression) was used as effectors in a 4 h 51Cr release assay against allogeneic targets at a 5:1 effector to target ratio. The percent speci®c 51Cr release (CTX) is indicated by 1111 (60±80%), 111 (40±59%), 1 1 (20±39%), 1 (10± 19%), 2 (0±9%).

found on some ®sh cytotoxic cells may provide valuable information concerning not only the evolutionary aspects of FcRs in general, but also the possible role that the poorly understood FcmR plays in mammalian immune responses. Acknowledgements

[4] [5]


This work was supported by grants from NIH (R01AI-19530) and USDA (NRI-CGP-99-35204-7844).




[1] Herberman RB, Nunn ME, Holden HT, Lavrin DH. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic and allogeneic tumors. II. Characterization of effector cells. Int J Cancer 1975;16:230±9. [2] Herberman RB, Ortaldo JR. Natural killer cells: their roles in defenses against disease. Science 1981;214:24±30. [3] Carlyle JR, Michie AM, Furlonger C, Nakano T, Lenardo MJ, Paige CJ, ZuÂnÄiga-P¯uÈcker JC. Identi®cation of a novel


developmental stage marking lineage commitment of progenitor thymocytes. J Exp Med 1997;186:173±82. Doherty PC. Cell-mediated cytotoxicity. Cell 1993;75:607±12. Trinchieri G. Interleukin-12: a proin¯ammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-speci®c adaptive immunity. Annu Rev Immunol 1995;13:251±76. Perussia B, Trinchieri G, Jackson A, Warner NL, Faust J, Rumpold H, Kraft D, Lanier LL. The Fc receptor for IgG on human natural killer cells: phenotypic, functional, and comparative studies with monoclonal antibodies. J Immunol 1984;133:180±9. Grif®n JD, Hercend T, Beveridge R, Schlossman SF. Characterization of an antigen expressed by human natural killer cells. J Immunol 1983;130:2947±51. McGarry RC, Pinto A, Hammersley-Straw DR, Trevenen CL. Expression of markers shared between human natural killer cells and neuroblastoma lines. Cancer Immunol Immunother 1988;27:47±52. Hercend T, Grif®n JD, Bensussan A, Schmidt RE, Edson MA, Brennan A, Murray C, Daley JF, Schlossman SF, Ritz J. Generation of monoclonal antibodies to a human natural killer clone. Characterization of two natural killer-associated antigens, NKH1A and NKH2, expressed on subsets of large granular lymphocytes. J Clin Invest 1985;75:932±43.


L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149

[10] Lanier LL, Phillips JH, Hackett Jr J, Tutt M, Kumar V. Natural killer cells: de®nition of a cell type rather than a function. J Immunol 1986;137:2735±9. [11] Ljunggren HG, Karre K. In search of the `missing self': MHC molecules and NK cell recognition. Immunol Today 1990; 11:237±44. [12] Moretta L, Ciccone E, Mingari MC, Biassoni R, Moretta A. Human natural killer cells: origin, clonality, speci®city, and receptors. Adv Immunol 1994;55:341±80. [13] Moretta A, Bottino C, Vitale M, Pende D, Biassoni R, Mingari MC, Moretta L. Receptors for HLA class-I molecules in human natural killer cells. Annu Rev Immunol 1996;14: 619±48. [14] Moretta A, Biassoni R, Bottino C, Pende D, Vitale M, Poggi A, Mingari MC, Moretta L. Major histocompatibility complex class I-speci®c receptors on human natural killer and T lymphocytes. Immunol Rev 1997;155:105±17. [15] Lopez-Botet M, Perez-Villar JJ, Carretero M, Rodriguez A, Melero I, Bellon T, Llano M, Navarro F. Structure and function of the CD94 C-type lectin receptor complex involved in recognition of HLA class I molecules. Immunol Rev 1997;155:165±74. [16] Lanier LL, Corliss B, Phillips JH. Arousal and inhibition of human NK cells. Immunol Rev 1997;155:145±54. [17] Sivori S, Vitale M, Morelli L, Sanseverino L, Augugliaro R, Bottino C, Moretta L, Moretta A. p46, a novel natural killer cell-speci®c surface molecule that mediates cell activation. J Exp Med 1997;186:1129±36. [18] Vitale M, Bottino C, Sivori S, Sanseverino L, Castriconi R, Marcenaro E, Augugliaro R, Moretta L, Moretta A. NKp44, a novel triggering surface molecule speci®cally expressed by activated natural killer cell, is involved in non-major histocompatibility complex-restricted tumor cell lysis. J Exp Med 1998;187:2065±72. [19] Pende D, Parolini S, Pessino A, Sivori S, Augugliaro R, Morelli L, Marcenaro E, Accame L, Malaspina A, Biassoni R, Bottino C, Moretta L, Moretta A. Identi®cation and molecular characterization of NKp30, a novel triggering receptor involved in natural cytotoxicity mediated by human natural killer cells. J Exp Med 1999;190:1505±16. [20] Moretta A, Biassoni R, Bottino C, Mingari MC, Moretta L. Natural cytotoxicity receptors that trigger human NK-cellmediated cytolysis. Immunol Today 2000;21:228±34. [21] Manning MJ, Nakanishi T. The speci®c immune system: cellular defenses. In: Iwama GK, Nakanishi T, editors. The ®sh immune system. London: Academic Press, 1996. p. 159±205. [22] Nakanishi T, Aoyagi K, Xia C, Dijkstra JM, Ototake M. Speci®c cell-mediated immunity in ®sh. Vet Immunol Immunopathol 1999;72:101±9. [23] Verlhac V, Sage M, Deschaux P. Cytotoxicity of carp (Cyprinus carpio) leucocytes induced against TNP-modi®ed autologous spleen cells and in¯uence of acclimatization temperature. Dev Comp Immunol 1990;14:475±80. [24] Fischer U, Ototake M, Nakanishi T. In vitro cell-mediated cytotoxicity against allogeneic erythrocytes in ginbuna crucian carp and gold®sh using a non-radioactive assay. Dev Comp Immunol 1998;22:195±206.

[25] Hasegawa S, Nakayasu C, Yoshitomi T, Nakanishi T, Okamoto N. Speci®c cell-mediated cytotoxicity against an allogeneic target cell line in isogeneic ginbuna crucian carp. Fish Shell®sh Immunol 1998;8:303±13. [26] Graves SS, Evans DL, Cobb D, Dawe DL. Nonspeci®c cytotoxic cells in ®sh (Ictalurus punctatus). I. Optimum requirements for target cell lysis. Dev Comp Immunol 1984;8:293±302. [27] Evans DL, Jaso-Friedmann L. Nonspeci®c cytotoxic cells as effectors of immunity in ®sh. Ann Rev Fish Dis 1992;2:109± 21. [28] Evans DL, Jaso-Friedmann L, Smith Jr EE, St John A, Koren HS, Harris DT. Identi®cation of a putative antigen receptor on ®sh nonspeci®c cytotoxic cells with monoclonal antibodies. J Immunol 1988;141:324±32. [29] Jaso-Friedmann L, Leary 3rd JH, Evans DL. NCCRP-1: a novel receptor protein sequenced from teleost nonspeci®c cytotoxic cells. Mol Immunol 1997;34:955±65. [30] Evans DL, Leary 3rd JH, Weisman Z, Warren J, Jaso-Friedmann L. Mapping of the epitope recognized by non-speci®c cytotoxic cells: determination of the ®ne speci®city using synthetic peptides. Scand J Immunol 1996: 43556±65. [31] Greenlee AD, Brown RA, Ristlow SS. Nonspeci®c cytotoxic cells of rainbow trout (Oncorhynchus mykiss) kill Yac-1 targets by both necrotic and apoptotic mechanisms. Dev Comp Immunol 1991;15:153±64. [32] Hayden BJ, Laux DC. Cell-mediated lysis of murine target cells by non-immune salmonoid lymphoid preparations. Dev Comp Immunol 1985;9:627±39. [33] Hinuma S, Abo T, Kumagai K, Hata M. The potent activity of fresh water ®sh kidney cells in cell-killing: I. Characterization and species-distribution of cytotoxicity. Dev Comp Immunol 1980;4:653±66. [34] Suzumura E, Kurata O, Okamoto N, Ikeda Y. Characteristics of natural killer-like cells in carp. Fish Path 1994;29:199±203. [35] McKinney EC, Schmale MC. Damsel®sh with neuro®bromatosis exhibit cytotoxicity toward tumor targets. Dev Comp Immunol 1994;18:305±13. [36] Faisal M, Ahmed II PetersG, Cooper EL. Natural cytotoxicity of talapia leukocytes. Dis Aquatic Org 1989;7:17±22. [37] Yoshida SH, Stuge TB, Miller NW, Clem LW. Phylogeny of lymphocyte heterogeneity: cytotoxic activity of channel cat®sh peripheral blood leukocytes directed against allogeneic targets. Dev Comp Immunol 1995;19:71±77. [38] Stuge T, Miller NW, Clem LW. Channel cat®sh cytotoxic effector cells from peripheral blood and pronephroi are different. Fish Shell®sh Immunol 1995;5:169±71. [39] Hogan RJ, Taylor WR, Cuchens MA, Naftel JP, Clem LW, Miller NW, Chinchar VG. Induction of target cell apoptosis by channel cat®sh cytotoxic cells. Cell Immunol 1999;195:110± 8. [40] Kagi D, Ledermann B, Burki K, Zinkernagel RM, Hengartner H. Molecular mechanisms of lymphocyte-mediated cytotoxicity and their role in immunological protection and pathogenesis in vivo. Annu Rev Immunol. 1996;14:207±32. [41] Hogan RJ, Stuge TB, Clem LW, Miller NW, Chinchar VG.

L. Shen et al. / Developmental and Comparative Immunology 26 (2002) 141±149







Anti-viral cytotoxic cells in the channel cat®sh (Ictalurus punctatus). Dev Comp Immunol 1996;20:115±27. Peck AB, Wigzell H, Janeway Jr C, Andersson LC. Environmental and genetic control of T cell activation in vitro: a study using isolated alloantigen-activated T cell clones. Immunol Rev 1977;35:146±80. London L, Perussia B, Trinchieri G. Induction of proliferation in vitro of resting human natural killer cells: IL 2 induces into cell cycle most peripheral blood NK cells, but only a minor subset of low density T cells. J Immunol 1986;137:3845±54. Stuge TB, Yoshida SH, Chinchar VG, Miller NW, Clem LW. Cytotoxic activity generated from channel cat®sh peripheral blood leukocytes in mixed leukocyte cultures. Cell Immunol 1997;177:154±61. Stuge TB, Wilson MR, Zhou H, Barker KS, Bengten E, Chinchar G, Miller NW, Clem LW. Development and analysis of various clonal alloantigen-dependent cytotoxic cell lines from channel cat®sh. J Immunol 2000;164:2971±7. Zhou H, Stuge TB, Miller NW, Bengten E, Naftel JP, Bernanke JM, Chinchar G, Clem LW, Wilson MR. Heterogeneity of channel cat®sh cytotoxic lymphocytes with respect to target recognition and cytotoxic mechanisms employed. J Immunol 2001;167:1325±32. Moretta A, Biassoni R, Bottino C, Mingari MC, Moretta L.

[48] [49]


[51] [52]




Natural cytotoxicity receptors that trigger human NK-cellmediated cytolysis. Immunol Today 2000;21:228±34. Lobb CJ, Olson MO, Clem LW. Immunoglobulin light chain classes in a teleost ®sh. J Immunol 1984;132:1917±23. Ghaffari SH, Lobb CJ. Structure and genomic organization of immunoglobulin light chain in the channel cat®sh. An unusual genomic organizational pattern of segmental genes. J Immunol 1993;151:6900±12. Ghaffari SH, Lobb CJ. Structure and genomic organization of a second class of immunoglobulin light chain genes in the channel cat®sh. J Immunol 1997;159:250±8. Metzger H. The high af®nity receptor for IgE on mast cells. Clin Exp Allergy 1991;21:269±79. Vance BA, Huizinga TW, Wardwell K, Guyre PM. Binding of monomeric human IgG de®nes an expression polymorphism of Fc gamma RIII on large granular lymphocyte/natural killer cells. J Immunol 1993;151:6429±39. Moingeon P, Lucich JL, Reinherz EL. Characterization of Fc epsilon RI gamma in human natural killer cells. Int Immunol 1992;4:955±8. Grif®n JD, Hercend T, Beveridge R, Schlossman SF. Characterization of an antigen expressed by human natural killer cells. J Immunol 1983;130:2947±51.