Graft-Versus-Host Reaction (GVHR) in Clonal Amago Salmon, Oncorhynchus Rhodurus Q.W

______http:\\www.paper.edu.cn

Veterinary Immunology and Immunopathology 89 (2002) 83–89

Graft-versus-host reaction (GVHR) in clonal amago salmon, Oncorhynchus rhodurus Q.W. Qina,*, M. Ototakeb, H. Nagoyab, T. Nakanishic aTropical Marine Science Institute & Department of Biological Sciences, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Singapore bNational Research Institute of Aquaculture, Tamaki, Mie 519-0423, Japan cDepartment of Veterinary Medicine, Nihon University, 1866 Fujisawa, Kanagawa 252-8510, Japan Received 5 November 2001; received in revised form 7 June 2002; accepted 7 June 2002

Abstract

The graft-versus-host reaction (GVHR) was demonstrated in a salmonid model system of clonal diploid and triploid amago salmon. Triploid operculum grafts on clonal diploid evoked an acute rejection within 12 days. Grafts exchanged among triploid amago salmon exhibited prolonged survival for 18 days. In contrast, diploid grafts on triploid, and allografts among clonal diploid amago salmon were accepted. A typical GVHR was induced in triploid recipients by intraperitonal injection of head kidney cells from sensitised diploid donors. The clinical signs of graft-versus-host disease (GVHD) were observed in the recipients after 1 week of cell injection as a loss of appetite and appearance of solid faeces, followed by haemorrhage, local swelling of ventral skin and an enlarged spleen. Three of six fish died within 1 month. Water temperature and frequency of sensitisation are critical to induce GVHR. Diploid donors had to be sensitised three times at 20 8C to induce the typical GVHR. GVHR was most effectively induced by head kidney cells, followed by peripheral blood leucocytes (PBL) and spleen cells. Ploidy analysis by flow cytometry revealed that the donor head kidney cells greatly increased in the recipient liver, head kidney and spleen, and reached the peak after 9 days of donor cell injection. The results in the present study are quite similar to the findings in ginbuna and ginbuna-gold fish hybrid system, suggesting the presence of T cells in salmonid as well as cyprinid fish. # 2002 Elsevier Science B.V. All rights reserved.

Keywords: Graft-versus-host reaction (GVHR); Clonal amago salmon; Oncorhynchus rhodurus; Graft-versus-host disease (GVHD); Cell-mediated immunity

1. Introduction demonstrated in mammals, birds (Clark, 1991), amphibians (Clark and Newth, 1972; Nakamura, Graft-versus-host reactions (GVHR) is a represen- 1985) and reptiles (Saad and Ridi, 1984). GVHR tative phenomenon of cell-mediated immunity, in can be induced when grafted donor contain immuno- which CD4þ and CD8þ T lymphocytes play a major logical competent cells and the host is unable to role (Jadus and Wepsic, 1992). GVHR has been recognise the donor cells. The grafted donor cells can then react against host which is essentially the * Corresponding author. Tel.: þ65-68743048; reverse of grafting reaction leading to the graft-versus- fax: þ65-67749654. host disease (GVHD) such as tissue damage and death. E-mail address: [email protected] (Q.W. Qin). In higher vertebrates, GVHR is due to the allelic

84 Q.W. Qin et al. / Veterinary Immunology and Immunopathology 89 (2002) 83–89

polymorphism of the major histocompatibility com- were used and maintained in outside ponds with plex (MHC). Acute allograft rejection is mediated via running fresh water at 15 8C, and fed with commercial MHC class I-restricted cytotoxicity. Reactive donor T pelleted trout food at the National Research Institute cells play an important role in GVHR through the of Aquaculture, Mie, Japan. Fish were acclimatised to secretion of cytokines by CD4þ cells, and through the aquarium for at least 2 weeks prior to use. CD8þ-mediated cytotoxity (Jadus and Wepsic, 1992). Fish are the lowest vertebrates with a well devel- 2.2. Gill cover transplantation oped immune system possessing lymphocyte subsets (Miller et al., 1985). T lymphocyte subpopulations Gill cover transplantation was carried out to exam- involved in allograft rejection in a teleost fish, Dicen- ine the allograft responses. Fish were anaesthetised trarchus labrax (L.) have been detected and recog- with 100 mg lÀ1 of ethyl p-amino benzoate. Pieces of nised using monoclonal antibodies (Abelli et al., donor gill cover (4 mm Â 6 mm) were grafted onto the 1999). The presence of GVHR in teleost fish was first recipient’s dorsal trunk region by slipping through the demonstrated employing a model system of clonal incision under the recipient skin. Three days later, a triploid ginbuna, Carassius auratus langsdorfii and small window was opened at the grafting site to tetraploid ginbuna-goldfish, Carassius auratus remove the overlaying recipient skin. The second hybrids (Nakanishi and Ototake, 1999). However, set graft of donor gill cover was transplanted to the crucian carp and goldfish are different at the subspe- same recipients on day 3 after the complete rejection cies level. Triploid donor cells may not recognise of the first set grafts. The survival time of grafted gill antigens originating from goldfish in tetrapolid reci- cover was estimated as previously described by Naka- pients as allogenetically different targets, and there- nishi (1987). Grafting was performed among different fore the killing mechanism may be different from fish groups. (1) Triploid (3N) graft onto diploid (2N); those reported for GVHR in mammals and birds. (2) 2N graft onto 3N; (3) 2N graft onto 2N; (4) 3N Fortunately, homozygous clonal diploid and triploid graft onto 3N. amago salmon, Oncorhynchus rhodurus, have been produced by artificial gynogenesis (Kobayashi et al., 2.3. Sensitisation of donor fish 1994). The principal of the model system in amago salmon is that clonal diploid and triploid fish belong to Diploid fish were sensitised by triploid fish using the same species, and thus triploid fish include all gill cover grafting either at 15 8C(first series of elements of the diploid genotype, but the reverse is not experiments) or at 20 8C (second series of experi- the case. The aim of present study is to induce GVHR ments) of water temperature. Diploid fish were sensi- using the model system of clonal diploid and triploid tised by one, two or three times of grafting. The amago salmon, and analyse the experimental results in interval between two grafts was 2 weeks. comparison with those found in the ginbuna system. 2.4. Preparation and transfer of diploid donor cells for induction of GVHR 2. Materials and methods All equipment used for the cell preparation was 2.1. Fish siliconised. Cells were maintained in RPMI 1640 tissue culture medium (Life Technologies, NY, Homozygous clonal diploid amago salmon, Oncor- USA) containing Penicillin (100 IU mlÀ1) and Strep- hynchus rhodurus, were produced by suppression of tomycin (100 mgmlÀ1). the first mitosis in the first generation followed by Diploid donor fish were bled and killed 7 days after suppression of Meiosis II in the second generation. the last sensitisation. The head kidney and spleen were Triploid amago salmon were produced by the insemi- dissected out sterilely from individual fish and placed nation of intact amago sperm to the eggs of a homo- in RPMI medium containing 10 IU mlÀ1 of heparin on zygous clonal diploid fish followed by suppression of ice. The tissue of each organ was put on a steel-mesh Meiosis II. Fish, weighing 67 g (mean body weight), filter (100 mm) with a small amount of medium and ______中国科技论文在线 http:\\www.paper.edu.cn

Q.W. Qin et al. / Veterinary Immunology and Immunopathology 89 (2002) 83–89 85

squeezed to give a single cell suspension. The cells 2.6. Determination of spleen enlargement were washed twice with medium followed by centri- fuging at 360 Â g for 5 min at 4 8C. Peripheral blood The spleen weights of experimental fish were mea- leucocytes (PBL) were separated by a density gradient sured, and the spleen enlargement was determined centrifugation method. Two millilitre of blood was using relative spleen index (rSI) according to the diluted with 4 ml of RPMI medium, and then carefully method described by Simonsen (1962).SI¼ spleen layered onto 3 ml of Ficoll-Paque (Pharmacia, Swe- weight/body weight, SI0: SI of triploid fish received den) solution (1.077 g mlÀ1) in a 15 ml centrifuge donor cells, SIo: SI of non-treated triploid fish (mean tube. After centrifugation at 1670 Â g for 15 min at of 0.075%). rSI ¼ SI0/SIo. 4 8C, the leucocyte layer at the interface was collected and washed twice. The cell density and viability were 2.7. Statistics determined using the trypan blue dye staining method. The cell viability was approximate 95%. Triploid fish Data were analysed using student’s t-test (P < 0:05 were intraperitonally injected with 0.2 ml of cell was considered statistically significant). suspension containing 5 Â 106 cells to induce GVHR. The sensitised donor cells were also injected into third party triploid recipients which were allogenetically 3. Results different from each other. Control animals received nonsensitised head kidney leucocytes. Fish were kept 3.1. Allograft responses among the groups at 15 or 20 8C. of amago salmon

2.5. Flow cytometric analysis The alloimmune responses among grafted diploid and triploid fish were summarised in Table 1.Infirst set The fate of diploid donor cells in the triploid host grafting, triploid tissue grafted onto diploid fish showed was analysed by flow cytometry. Triploid recipient an acute type of rejection in a period of 12–17 days after fish were bled and killed on days 3, 7, 9, 14, 21 and 28 grafting (13.4 days of mean survival time of gill grafts), after cell injection. The liver, head kidney and spleen whereas in the reverse situation (2N ! 3N) grafts were were taken to estimate the percentage of donor cells in accepted. The reciprocal grafts between triploid fish the recipients. Head kidney and spleen cells were also showed subacute rejection (18 days mean surviving prepared as described above. Before passing through the steel mesh filter, liver tissue was dissociated by incubating with collagenase (0.5 mg mlÀ1) for 2 h at Table 1 Allograft responses among the group of amago salmon by skin 25 8C. Ploidy was measured from the DNA content of graft the cell nucleus by flow cytometry (EPICS Profile II, Coulter, USA) according to the method of Nakanishi Donor Recipient No. of fish Survival of grafted gill cover (mean of days) and Ototake (1999). In brief, 200 mlof5Â 106 cells mlÀ1 cell suspension was fixed with 3 ml of cold 70% First seta Second setb ethanol for 60 min on ice, then the cells were washed 3Nc 2Nd 15 13 8 twice in RPMI medium with heparin and stained with 3N 3N 10 18 15 10 mgmlÀ1 of propidium iodide (PI) solution with 2N 2N 10 >120 >120 0.2 mg mlÀ1 of RNAse, and passed through the flow 2N 3N 10 >120 >120 cytometer. Three fish were used in each experiment for a The first set gill cover transplantation, a piece of donor fish the ploidy determination. All experimental fish were gill cover was grafted onto the recipient’s dorsal trunk region by anaesthetised and killed to avoid prolonged suffering slipping through the incision under the recipient skin. b after the experiments finished. The experiments were The second set transplantation, 3 days after the complete rejection of the first set grafts, the donor gill cover was transplanted in accord with the ethical recommendations and to the same recipient. approved by the National Research Institute of Aqua- c 3N, clonal triploid amago salmon. culture, Japan. d 2N, clonal diploid amago salmon. ______中国科技论文在线 http:\\www.paper.edu.cn

86 Q.W. Qin et al. / Veterinary Immunology and Immunopathology 89 (2002) 83–89

time). Rejection was not observed in the reciprocal graft Table 2 between clonal diploid fish. In comparison with the first Comparison of GVHR in triploid (3N) host fish injected with 5 Â 106 donor cells from various lymphoid tissues set grafts, the allorejection time was accelerated by exploring the same donor to a second set graft. Tissue Time of appearance of GVHD (day) Head kidney 8, 10, 10, 12, 12, 14 (three fish 3.2. GVHR in clonal amago salmon died within 1 month) Third party fisha 8, 9, 10, 12, 14, 14 When the diploid donor fish were sensitised three PBL 10, 12, 12, 14, 14, 14 times and kept at 20 8C water temperature, injection of Third party fish 10, 12, 14, 14, 14, 14 donor head kidney cells generated acute GVHD in Spleen 11, 11, 13, 14, 14, 14 triploid recipients at 20 8C of water temperature. The Third party fish 11, 12, 12, 13, 14, 14 clinical symptoms of GVHD were observed within 14 days as loss of appetite and appearance of solid faeces, At 20 8C water temperature. a Third party fish, triploid recipients which were allogenetically eventually leading to body weight loss, scale protru- different from each other. sion, haemorrhage of ventral skin, swelling and ulcera- tion of tail and anus. During succeeding stages, a green colouration of the liver and an enlargement of spleen tion of 5 Â 106 sensitised donor head kidney, spleen have also been observed. GVHR was very weak when cells or PBL were able to generate an acute GVHD recipients were kept at 15 8C water temperature exhi- within 2 weeks in the recipients of triploid fish at 20 8C. biting some symptoms of GVHD, such as appetite loss Head kidney cells were most effective in the induction and appearance of solid faeces 1 week after injection of GVHR, and three of six recipients died within 1 but recovered after 1 month. There were no clinical month. No difference was observed in the appearance signs observed in control recipients which received of GVHD in the recipients of third party fish. non-sensitised donor cells at 20 8C. Diploid donor cells To examine whether the observed GVHR corre- sensitised three times at 15 8C were not able to induce sponded with the enlargement of the spleen, fish were GVHD in triploid recipients, even when the triploid killed on days 3, 7, 9, 14, 21, 28 and 35 after injection fish were reared at 20 8C. As shown in Table 2, injec- of donor cells, and their spleens were weighed to

Fig. 1. Relative spleen indices of triploid amago salmon. The triploid ﬁsh were intraperitonally injected with head kidney cells from in vivo pre-sensitised diploid donors at various periods. Note the splenic enlargement corresponds with the observed GVHR in the triploid recipients. Each symbol represents one ﬁsh. ______中国科技论文在线 http:\\www.paper.edu.cn

Q.W. Qin et al. / Veterinary Immunology and Immunopathology 89 (2002) 83–89 87

Fig. 2. Percentage of injected diploid donor cells in lymphoid organs of triploid host. Data are means S:D: of three ﬁsh. P < 0:05, P < 0:01 compared to the donor cells in host ﬁsh organs after 3 days injection.

determine the relative spleen indices. As shown in et al., 1991; Abelli et al., 1999). The specific cell- Fig. 1, the spleens of fish did not enlarge for the first 3 mediated immunity of teleost fish has been studied in days, but significantly increased from 7 to 9 days in vitro as specific cell-mediated cytotoxicity (CMC) weight in comparison to the non-treated fish against allogeneic targets (Fischer et al., 1998; Hase- (P < 0:05). During the succeeding stage, the rSI of gawa et al., 1998) and virus infected target cells surviving fish decreased from 14 days and returned to (Somamoto et al., 2000). In contrast, most information 1.0 after 28 days. Spleen weights of the recipients on in vivo cell-mediated immunity of teleost fish has which received donor cells for more than 28 days were been obtained by transplantation experiments, parti- significantly lower than those at 9 days (P < 0:05). cularly by skin allograft rejection and the graft-versus- host reaction (reviewed by Nakanishi et al., 1999). 3.3. Fate of the transplanted cells The first evidence of GVHR in teleosts was reported by Nakanishi and Ototake (1999). A model system of Ploidy analysis of the donor cells in different organs clonal triploid ginbuna and tetraploid ginbuna-gold- of recipients is shown in Fig. 2. In comparison to the fish, Carassius aruatus hybrids was used in their donor cells in host liver, spleen and head kidney after 3 studies. When triploid cells, sensitised by scale grafts days injection, the number of injected donor cells from tetraploid donors, were injected into tetraploid significantly increased in all organs examined in the recipients, a typical GVHR was induced as evidenced host at 14 days (P < 0:05). The percentage of donor by the death of recipients within 1 month. A wide cells increased up to 46.50% of the lymphoid popula- variety of organs and tissues showed pathological tion in the liver, 13.25% in the head kidney at 9 days changes during the course of the clinically apparent and 11.43% in the spleen at 14 days, respectively, after GVHD including enlargement of the spleen, infiltra- cell injection. By 28 days after injection, the percen- tion of mononuclear cells and focal necrosis particu- tage of donor cells decreased and remained around larly in the skin, liver and lymphoid tissues. Ploidy 10% in the liver and spleen, and 5% in the head kidney. analyses revealed that donor cells greatly increased in the host liver and spleen. However, the ginbuna and ginbuna-goldfish model system has limitations for 4. Discussion induction of GVHR. The main problem is that ginbuna and goldfish are different at the subspecies level. There- It is well established that teleost fish possess the fore, strictly, it is not possible to tell if the MHC is functional equivalents of T- and B-lymphocytes (Clem involved in the GVHR in ginbuna and ginbuna-gold ______中国科技论文在线 http:\\www.paper.edu.cn

88 Q.W. Qin et al. / Veterinary Immunology and Immunopathology 89 (2002) 83–89

fish system since the MHC works within a species but while CD8þ cells are part of the efferent arm in not between species. GVHR was successfully induced GVHD (Chao, 1994). Taken together, it is possible in a clonal amago salmon system in the present studies. that low water temperature suppresses both CD8þ Injection of 5 Â 106 sensitised diploid donor cells per cytotoxic T cells during the effect phase and CD4þ T fish was able to induce GVHR in triploid amago helper cells during the recognition phase of GVHR in salmon. Although only three of six fish died within 1 teleost fish. month, acute clinical GVHD symptoms and the patho- In conclusion, the GVHR was successfully demon- logical changes of internal organs were observed in strated in the model system of clonal amago salmon. recipients which are quite similar findings to those Environmental temperature is a critical factor for found in ginbuna (Nakanishi and Ototake, 1999). induction of acute GVHD in amago salmon. The The flow cytometry analysis of ploidy revealed that present results, which are similar to those found in the cytotoxic donor cells could proliferate and cause the clonal ginbuna crucian carp system, suggest the pre- liver necrosis and enlargement of the spleen in host fish. sence of cytotoxic T cells in salmonid as well as The head kidney, PBL and spleen cells are rich sources cyprinid fish. of lymphocytes, highly efficient in inducing GVHR. The head kidney cells were most effective at inducing GVHR in amago salmon. The results reported in the Acknowledgements present studies are quite similar to those found in the ginbuna and ginbuna-goldfish hybrid system, suggest- This study was supported in part of a postdoctoral ing the existence of similar killing mechanisms in the fellowship for Qiwei Qin from the Science and Tech- amago and ginbuna system. nology Agency of Japan and by a grant from the Theimmunesystemofteleostfish is influenced by Ministry of Agriculture, Forestry and Fisheries of environmental temperature. In channel catfish, sup- Japan. pression of humoral immunity at low temperatures was the result of an inhibitory effect on T rather than B cells (Bly and Clem, 1991). Also, the mixed References leucocyte reaction (MLR) was suppressed at low temperature (Miller et al., 1986). Thus, it is not Abelli, L., Baldassini, M.R., Mastrolia, L., Scapigliti, G., 1999. surprising that GVHR in fish can be affected by Immunodetection of lymphocyte subpopulations involved in environmental temperatures. Fischer et al. (1999) allograft rejection in a teleost, Dicentrarchus labrax (L.). Cellular Immunol. 191, 152–160. reported that water temperature affected the effecter Bly, J.E., Clem, L.W., 1991. Temperature-mediated processes in phase of GVHR in the ginbuna system. After injection teleost immunity: in vitro immunosuppression induced by in of sensitised triploid donor cells, the death of tetra- vivo low temperature in channel catfish. Vet. Immunol. ploid recipients occurred earliest when fish were Immunopathol. 28, 365–377. reared at 25 8C; fish kept at 15 8C exhibited a lower Chao, N.J., 1994. Graft-Versus-Host Disease. R.G. Landes Company, Austin. death rate and died later than those kept at high Clark, W.R., 1991. The Experimental Foundations of Modern temperatures. These results suggest that mainly cyto- Immunology, 4th Edition. Wiley, New York. toxic effector cells were suppressed at low tempera- Clark, J.C., Newth, D.R., 1972. Immunological activity of ture (Fischer et al., 1999). The present study showed transplanted spleens in Xenopus laevis. Experientia 28, 951– that the recognition phase (sensitisation of donor 953. Clem, L.W., Miller, N.W., Bly, J.E., 1991. Evolution of lymphocyte cells) of GVHR was highly affected by environmental populations, their interactions, and temperature sensitive. In: temperature in amago salmon. The optimal sensitisa- Warr, G.W., Cohen, N. (Eds.), Phylogenesis of the Immune tion condition was essential to induce GVHD in System. CRC Press, Boca Raton, FL, pp. 191–213. amago. For preparation of effective donor cells, clo- Fischer, U., Ototake, M., Nakanishi, T., 1998. In vitro cell- nal diploid fishhadtobesensitisedbytriploidfish mediated cytotoxicity against allogeneic erythrocytes in ginbuna crucian carp and goldfish using a non-radioactive whilstheldat208C, with a water temperature of assay. Dev. Comp. Immunol. 22, 195–206. þ 15 8C being ineffective. In mammals, CD4 cells are Fischer, U., Ototake, M., Nakanishi, T., 1999. Effect of environ- critical in the afferent arm for the induction of GVHD, mental temperature on in vitro cell-mediated cytotoxicity ______中国科技论文在线 http:\\www.paper.edu.cn

Q.W. Qin et al. / Veterinary Immunology and Immunopathology 89 (2002) 83–89 89

(CMC) and graft-versus-host reaction (GVHR) in ginbuna Nakamura, T., 1985. Lethal graft-versus-host reaction induced by crucian carp (Carassius auratus langsdorfii). Fish Shellfish parental cells in the clawed frog, Xenopus laevis. Transplanta- Immunol. 9, 233–236. tion 40, 393–397. Hasegawa, S., Nakayasu, C., Yoshitomi, T., Nakanishi, T., Nakanishi, T., 1987. Transfer ability of immune plasma and Okamoto, N., 1998. Specific cell-mediated cytotoxicity against pronephric cells in isogenic, allogeneic and xenogeneic transfer an allogeneic target cell line in isogeneic ginbuna crucian carp. systems in crucian carp. Dev. Comp. Immunol. 11, 521–528. Fish Shellfish Immunol. 8, 303–313. Nakanishi, T., Ototake, M., 1999. The graft-versus-host reaction Jadus, M.R., Wepsic, H.T., 1992. The role of cytokines in graft- (GVHR) in the ginvuna crucian carp, Carassius auratus versus-host reactions and disease. Bone Marrow Transplant. 10, langsdorfii. Dev. Comp. Immunol. 23, 15–26. 1–14. Nakanishi, T., Aoyagi, K., Xia, C., Dijkstra, J.M., Ototake, M., Kobayashi, T., Ide, A., Hiasa, T., Fushiki, S., Ueno, K., 1994. 1999. Specific cell-mediated immunity in fish. Vet. Immunol. Production of cloned amago salmon Oncorhynchus rhodurus. Immunopathol. 72, 101–109. Fisheries Sci. 66, 275–281. Saad, A.H., Ridi, R.E., 1984. Mixed leukocyte reaction, graft- Miller, N.W., Sizemore, R., Clem, L.W., 1985. Phylogeny of versus-host reaction, and skin allograft rejection in the lizard, lymphocyte heterogeneity: the cellular requirement for in vitro Chalcides ocellatus. Immunobiology 166, 484–493. antibody responses of channel catfish leukocytes. J. Immunol. Simonsen, M., 1962. Graft-versus-host reactions, their natural history 134, 2884–2888. and applicability as tools research. Progr. Allergy 6, 349–367. Miller, N.W., Deuter, A., Clem, L.W., 1986. Phylogeny of Somamoto, T., Nakanishi, T., Okamoto, N., 2000. Specific cell- lymphocyte heterogeneity: the cellular requirement for the mediated cytotoxicity against a virus-infected syngeneic cell mixed leukocyte reaction in channel catfish leukocytes. line in isogeneic ginbuna crucian carp. Dev. Comp. Immunol. Immunology 59, 123–128. 24, 633–640.