Journal of Immunological Methods, 153 (1992) 161-165
161
© 1992 Elsevier SciencePublishers B.V. All rights reserved 0022-1759/92/$05.00
JIM 06399
Comparison of selected protocols for the generation of IgA isotype monoclona! antibodies against the gut parasite, Eimeria tenella J.A. H o a r e and A.J. T r e e s Department of Veterinary Parasitology, Licerpt~l School of Tropical Medicine, Pembroke Place, Licerpool L3 5QA, UK
(Received 18 October 1991, revisedreceived ! I February 1992,accepted 3 April 1992)
Investigations into the role of mucosal lgA in the protective immune response to alimentary tract parasites could be facilitated by the production of specific lgA monoclonal antibodies. To this end, we sought to generate IgA hybridomas against sporozoites of the protozoan, Eimeria tenella using different protocols. Of the methods investigated, intra-enteric priming followed by intravenous boosting of germ-free mice, using splenic iymphocytes in fusions, optimised the yield of lgA hybridomas. Using this protocol, 7 / 2 7 specific anti-Eimeria hybridoma antibodies isolated were of IgA isotype. When lymphocytes from mesenteric lymph nodes (MLN) were used in fusions more non-specific I r A secretors were produced than when splenic lymphocytes were used, but the yield of specific ant;,-Eimeria IgA secreting hybridomas was not improved. By all protocols, a total of nine lgA secreting hybridomas were identified of which eight have been cloned for further studies. Key words: IgA; Monoclonal antibody; Parasite; Eimeria tenella
Introduction Eimeria tenella is an important protozoal parasite of the alimentary tract of the chicken. Since secretory IgA is the major immunoglobulin of mucosal secretions, a variety of studies has reported on the elaboration and possible role of
Correspondence to: A.J. Trees, Department ot Veterinary Parasitology, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool1.3 5QA, UK. Tel.: 051-708-9393; Fax: 051-708-8733. Abbreciations: DMEM, Dulbecco's modified Eagles medium; GALT, gut associated lymphoid tissue; GF, germfree; i.e., iatraenteric; IFAT, immunofluorescenceantibody test; i.p., intraperitoneal; Mab, monoclonal antibody; MLN, mesenteric lymph node; PP, Peyer's patch.
endogenous lgA in E. tenella infectioc-s (Orlans and Rose, 1972; Davis, Parry and Porter, lCf/8; Davis and Porter, 1979; Mockett and Rose, 1986; Lillehoj and Ruft, 1987; Rose and Hesketh, 1987; Trees et al., 1989, and others). To facilitate the further investigation of the role of I r A in protection at the mucosai surface and the identification of antigens relevant to protection at that site, we sought to generate IgA monoclonal antibodies (Mabs) to the infective stage (sporozoite) of E. tenella. We have produced murine IgA Mabs specific for the avian parasite for use in passive transfer experiments, on the basis that the binding of polymeric lgA to heterologous secretory component has been described for several species combinations, including mammalian to avian species (Rose et al., 1981; Peppard et al., 1983).
162 Compared to Mabs of IgG or IgM isotypes, there are relatively few reports of IgA Mabs and their frequency of occurrence following conventional immunisation and fusion plotocols is low (see Komisar et al., 1982). Murine lgA Mabs to various antigens have been produced based on more rational approaches to preferentially generate IgA secreting hybridomas by using either the alimentary route of immunisation or iymphocytes from gut associated lymphoid tissue (GALT) and mesenteric lymph nodes (MLN) in fusions (Colwell et al., 1982, 1983, 1986; Komisar et al., 1982; Mazanec et al., 1987; Weltzin et ai., 1989). These lgA secretors have been derived using different protocols involving different variables including antigens, immunisation/infection procedures, strains of mice, germ-free (GF) or conventional mice, organ sources of lymphocytes, myeloma fusion partners and fusion protocols and it is difficult to compare the efficiency of different protocols in generating specific IgA Mabs. Some of these IgA Mabs have been directed against mucosal bacteria (Komisar et ai., 1982; Colwell et al., 1983) or viruses (Mazanec et al., 1987; Weitzin et al., 1989). In this paper, we report the rational derivation of IgA Mabs against a mucosal parasite and compare the proportion of both nonspecific and specific IgA secreting hybridomas produced by selected different protocols.
Materials and methods
Mice Conventionally reared (from the School of Tropical Medicine, Liverpool) or GF (from Harlan Olac, Bicester, England) BALB/c mice were used. GF mice were maintained in a plastic film isolator and provided with sterile dry food (Harlan Plat) and water ad libitum. lmmunisation Beginning at 6-8 weeks of age, mice were immunised by inoculation of live sporozoites of the Houghton strain of Eimeria tenella (kind gift ot" Dr. M.E. Rose), prepared by standard techniques. Live parasites were used in order to optimise the yield of antibodies directed against native surface antigens. Mice were primed either by
the intraperitoneal (i.p.) injection of 250,000 sporozoites on two occasions, 4-6 weeks apart, or by a single intraenteric (i.e.) injection (into the lumen of the jejunum) of 2.5 x 106 sporozoites under general anaesthesia (by halothane inhalation or parenteral sodium pentobarbitone injection, Nembutal, Ceva, UK) via a small laparotomy incision at the umbilicus. All mice received a booster injection of 1 X 106 sporozoites intravenously 14-16 days post priming and 3 days prior to fusions. The serum antibody response of immunised mice was assayed by a three layer immunofluorescence antibody test (IFAT), see below. Fusion method Spleen or mesenteric lymph nodes (MLN) were used as the lymphocyte source. Between 1-3 × 108 lymphocytes were obtained per spleen compared to 3-8 x 107 per MLN. Cell fusion was according to established methods (Galfr~ and Milstein, 1981) with mi~aor modifications. Briefly, myeloma cells (aminopterin-sensitive, NS-1-Ag41, kind gift of Professor M. Hommel) and iymphocytes were mixed in the ratio 1"10 and 50% polyethylene glycol (MW 1500, BDH, England) was used as fusogen in serum free medium. The final cell suspension was made up in HAT selective medium and plated out into four to six 96-well plates per fusion. MLN fusions were plated onto 24 h old feeder layers in four 96-well plates per fusion. All media were based on Dulbecco's modified Eagle's medium (DMEM) (Gibco, Paisley), for further details see Hoare (1989). Hybridoma growth was scored from 7 days post fusion. Results are included only from fusions which produced more than 100 hybridomas from splenic fusions and more than 20 hybridomas from fusions with MLN lymphocytes. Screening, cloning and cryopreservation of hybridomas At 12 days post fusion, hybridoma supernatants were first assayed for specific antibody by a two step IFAT with whole sporozoites as antigen, after the method of Trees et al. (1985) but using an F1TC-conjugated anti-mouse lg antibody (Nordic Immunochemicals, Netherlands). Isotypes of antigen-specific antibodies were deter-
163 mined by Ouchterlony double diffusion in agar employing commercial, affinity purified Fcspecific antisera (Nordic Immunochemicals, Netherlands) and by a three layer IFAT including an isotype specific second antibody stage (see Trees et al., 1985). The specificity and isotype of each hybridoma was assayed on at least two occasions by the three layer IFAT. lsotyping of hybridoma antibodies of undetermined specificity was performed for one 96 well plate per fusion, by Ouchterlony double diffusion, on uncloned material. Cells from positive wells were cloned once or twice by limiting dilution, re-assayed for specific antibody reactivity and isotype and then cryopreserved.
Results The results of 14 fusions involving six different protocols are presented in Table i. The ratio of specific anti-Eimeria hybridoma antibodies per 107 cells fused was similar for both MLN and splenic lymphocytes for comparable fusions but, since more cells were available from the spleen, more specific bybridomas were produced from that source. In comparison with conventionally reared mice, more specific hybridomas of all isotypes were produced from fusions of lymphocytes from G F mice and this is reflected in the im-
proved hybridoma ratio of the latter. Comparing the source of cells for fusion, with respect to non-specific antibody production by bybridomas, IgA secretors were detected from all MLN fusions, but not from splenic fusions. This was not the case for specific lgA hybridoma antibodies, which were derived predominantly from spleen cell fusions. The combination of splenic iymphocytes used in fusions from G F mice after i.e. priming gave the most specific IgA hybridomas; both fusions involving this protocol yielded specific IgA hybridomas (two from one spleen, 16 x 107 lymphocyte fused, and five from the second, 22 × 107 lymphocytes fused). Although these mice had specific serum IgG antibody titres ( > 1/40 dilution) at fusion, neither had specific serum IgA titres detectable by TLIFA. In summary, of nine anti-Eimeria IgA hybridoma antibodies identified, seven resulted from fusions from splenic lymphocytes derived from G F mice after i.e. priming. Of the nine IgA hybridomas, eight were subsequently cloned and cryopreserved.
Discussion On the basis that GALT is an enriched source of IgA B cell precursors (Craig and Cebra, 1971), Komisar and co-workers (1982) used pooled MLN
TABLE ! COMPARATIVE vI'IELDSAND IgA ISOTYPE FREQUENCY OF NON-SPECIFIC AND OF SPECIFIC ANTI-EIMERIA SECRETING HYBRIDOMASFROM DIFFERENT FUSION PROTOCOLS Mouse type
Conventional Conventional Conventional Conventional Germ-free Germ-free
Roote of immunisation
Lymphocyte Totalcells source fused ( )< 107)
Hybridoma:' ratio
Parenteral Parenteral i.e. i.e. i.e. i.e.
S MLN S MLN S MLN
0.25 0.26 0.14 0.27 0.71 2.0
75 3.8 44 1I. 1 38 4.5
Number of specificanti-Eimeria hybridomas/107cells fused. b From one 96 well plate only, per fusion, see text. c ND, not done.
Secretingh~bridomas Non-specific Specific IgA/total h anti-Eimeria IgA/total ND " 1/19 ,;/22 0/1 0/24 0/6 2/28 0/3 0/76 7/27 2/17 1/9
no. fusions 4 2 2 2 2 2
164 and Peyer's patch (PP) as lymphocyte sources to produce lgA Mabs. However, various studies have shown that i.e. antigen presentation not only resuits in the induction of precursor IgA B cells in GALT but also IgA secreting and memory cells in the spleen (Michalek et al., 1980; Andrew and Hall, 1982; Jeurissen et al., 1985 and others). Moreover, Jeurissen et al. (1985) proposed that mucosal immunisation activates local IgA precursor cells which leave PP and migrate via MLN, lymph and blood to distant sites, including mucosae and spleen. These cells can then be reactivated by subsequent specific antigenic stimulation to differentiation and lgA production. This evidence indicated that i.e. priming might be efficient in promoting IgA memory cells in the spleen which could be reactivated by parenteral immunisation and used as fusion partners. Thus, although MLN has been proposed as a rational source of lymphocytes for fusions to generate IgA Mabs, our results do not demonstrate any advantage in the use of MLN compared to spleen. MLN was a poorer lymphocyte source, tending to give fewer lymphocytes and giving fewer hybridomas (even when the hybridoma ratio was elevated using GF mice). This disadvantage could have been offset by pooling MLN from 5-10 mice but this has practical and ethical objections. Although the theoretical possibility of obtaining large numbers of specific IgA secretors from MLN was not born out in this study, a higher frequency of lgA isotypes of undetermined specificity was observed from MLN fusions compared with those from splenic lymphocytes. These lgA secreting hybridomas were presumably derived from B cells precommitted to antigens derived from other enteric organisms and food antigens. Whilst there was a marked improvement of yield of antiEimeria bybridomas generated from GF compared to conventionally reared mice, the unchanged isotype ratio in hybridomas of undetermined specificity in fusions generating specific lgA hybridomas suggests that the induction of IgA secreting cells was andgen specific. However, the results do not indicate any preferential induction of specific IgA precursor cells in MLN compared to spleen in GF mice. Other GALT derived cells have been successfully used for fusions by other workers and Weltzin et al. (1989) found
PP generated a considerably higher proportion of specific IgA hybridomas to viral antigens than spleen, but there are considerable technical difficulties in the sterile isolation of PP lymphocytes as well as limitations due to the low numbers of lymphoeytes generated. Although conventional mice are usually used in attempts to derive IgA Mabs, the present results indicate the advantage of GF mice, first used by Colwell and co-workers (Colwell et al., 1982, 1983). The use of GF mice is based on evidence that bacterial lipopolysaccharides negatively influence splenic lgA responses to orally presented antigen (Babb and McGhee, 1980; Michalek et al., 1980; Michalek et ai., 1983). Although GF mice require more sophisticated handling, their merit in optimising the advantages of i.e. priming are likely to reduce the number of fusions, mice, effort and cost required to isolate IgA Mabs of the desired specificity. It is interesting that circulating specific IgA antibodies were not detected at fusion in those mice which yielded the most specific IgA secreting hybridomas. This may have been due to competitive binding of target antigen by high titre IgG isotype antibody. Whilst serum titres have relevance to the likelihood of specific hybridoma production, they may not correlate directly with the number of secreting hybridomas produced after fusion (Campbell, 1984). Our results suggest that lack of detectable specific circulating IgA antibody in mice immunised intraenterically should not discourage fusion and attempts to isolate IgA secreting hybridomas. In conclusion, a rational approach to the production of specific IgA secreting hybridomas i n ferred from these results would employ i.e. priming and intravenous boosting of young GF mice, and utilise splenic lymphocytes in an established fusion protocol. This technically straightforward protocol resulted in about 25% of specific antibody secreting hybridomas being of IgA isotype.
Acknowledgements We are very grateful to M~ J. Gale for technical advice and Miss. S.M. McKellar and Miss. S. Crozier for occasional laboratory assistance. This
165 w o r k w a s s u p p o r t e d by a n A g r i c u l t u r a l a n d F o o d Research Council Veterinary School's Fellowship to J.A.H.
References Andrew, E. and Hall, J.G. (1982) IgA antibodies in the bile of rats. I. Some characteristics of the primary response. Immunology 45, 169-175. Babb, J.L. and McGhee, J.R. (1980) Mice refractory to lipopolysaccharide manifest high immunoglobulin A responses to orally administered antigen. Infect. Immun. "~9, 322-328. Campbell, A.M. (1984) Monoclonal Antibody Technology. Elsevier, Amsterdam, p. 88. Colwell, D.E., Gollahon, K.A., McGhee, J.R. and Michalek, S.M. (1982) lgA hybridomas: a method for generation in high numbers. J. Immunol. Methods 54, 259-266. Colwell, D.E., Gollahon, K.A., Morisaki, I., McGbce, J.R. and Michalek, S.M. (1983) IgA hybridomas to cell surface components of pathogenic bacteria. Ann. NY Acad. Sci. 409, 794-796. Colwell, D.E., Michalek, S.M. and McGhee, J.R. (1986) Method for generating a high frequency of hybridomas producing monoclonal lgA antibodies. Methods Enzymol. 121, 42-51. Craig, S.W. and Cebra, J.J. (1971) Peyer's patches: an enriched source of precursors for lgA-producing immunocytes in the rabbit. J. Exp. Med. 134, 188-200. Davis, PJ. and Porter, P. (1979) A mechanism for secretory lgA-mediated inhibition of the cell penetration and intracellular development of Eimeria tenella. Immunology 36, 471-477. Davis, P.J., Parry, S.H. and Porter, P. (1978) The role of secretory IgA in anti-coccidial immunity in the chicken. Immunology 34, 879-888. Galfr~, G. and Milstein, C. (1981) Prcparation of monoclonal antibodies: strategies and procedures. Methods Enzymol. 73, 1-46. Hoare, J.A. (1989) The production and evaluation of IgA monoclonal antibodies directed against Eimeria tenella sporozoites. Ph.D. Thesis, University of Liverpool. Jeurissen, S.H.M., Claassen, E., Van Rooijen, N. and Kraal, G. (1985) Intra-intestinal priming leads to antigen-specific
IgA memory cells in peripheral lymphoid organs. Immunology 56, 417-423. Komisar, J.L., Fuhrman, J.A. and Cebra, JJ. (1982) IgA-producing hybridomas are readily derived from gut-associated lymphoid tissue. J. Immunol. 128, 2376-2378. Lillehoj, H.S. and Ruff, M,D. (1987) Comparison of di~ase susceptibility and subclass-specific antibody response in SC and FP chickens experimentally inoculated with Eimeria tenella, E. acerculina or E. maxima. Avian Dis. 31, !12-119. Mazanec. M.B., Nedrod, J.G. and Lamm, M.E. (1987) lmmunoglobulin A monoclonal antibodies protect against Sendal virus. J. Virol. 61, 2624-2626. Michalek, S.M., Kiyono, H., Babb, J.L and McGhee, J.R. (1980) Inheritance of LPS non-responsiveness and elevated splenic IgA immune response in mice orally immunized with heterologous erythrocytes. J. lmmunol. 125, 222O-2224. Michalek, S.M., McGhee, J.R., Kiyono, H., Colwell, D.E., Eldridge, J.H., Wannemuehler, M.J. and Koopman, WJ. (1983) The IgA response: inductive aspects, regulatory cells and effector functions. Ann. NY Acad. Sci. 409, 48-71. Mocken, A.P.A. and Rose, M.E. (1986) Immune responses to eimeria: quantification of antibody isotypes of Eimeria tenella in chicken serum and bile by means of the ELISA. Parasite Immunol. 8, 481-489. Orlans, E. and Rose, M.E. (1972) An IgA-like immunoglobulin in the fowl. Immunochemistry 9, 833-838. Peppard, J.V., Rose, M.E. and Hesketh, P. (1983) A functional homologue of mammalian secretory component exists in chickens. Eur. J. lmmunol. 13, 566-570. Rose, M.E. and Hesketh, P. (1987) Eimeda tenella: effects of immunity on sporozoites within the lumen of the small intestine. Exp. Parasitol. 63, 337-344. Rose, M.E., Orlans, E., Payne, A.W.R. and Hesketh, P. (1981) The origin of igA in chicken bile: its rapid active transport from blood. Eur. J. Immunol. 11,561-564. Trees, A.J., Crozier. S.J., McKellar, S.B. and Wachira, T.M. (1985) Class-specific circulating antibodies in infections with Eimeria tenella. Vet. Parasitol. 18, 349-357. Trees, A.J.0 Karim, M.J., McKellar, S.B. and Car~er, S.D. (1989) Eimeria tenella: local antibodies and interactions with the sporozoite surface. J. Protozool. 36, 326-333. Weltzin, P.L.-J., Micheni, P., Fields, B.N., Kraehenbohl. J.P. and Neutra, M.R. (1989) Binding and transepithelial transport of immunoglobulins by intestinal M cells: demonstration using monoclonal lgA antibodies against enteric viral proteins. J. Cell Biol. 108, 1673-1685.
161
© 1992 Elsevier SciencePublishers B.V. All rights reserved 0022-1759/92/$05.00
JIM 06399
Comparison of selected protocols for the generation of IgA isotype monoclona! antibodies against the gut parasite, Eimeria tenella J.A. H o a r e and A.J. T r e e s Department of Veterinary Parasitology, Licerpt~l School of Tropical Medicine, Pembroke Place, Licerpool L3 5QA, UK
(Received 18 October 1991, revisedreceived ! I February 1992,accepted 3 April 1992)
Investigations into the role of mucosal lgA in the protective immune response to alimentary tract parasites could be facilitated by the production of specific lgA monoclonal antibodies. To this end, we sought to generate IgA hybridomas against sporozoites of the protozoan, Eimeria tenella using different protocols. Of the methods investigated, intra-enteric priming followed by intravenous boosting of germ-free mice, using splenic iymphocytes in fusions, optimised the yield of lgA hybridomas. Using this protocol, 7 / 2 7 specific anti-Eimeria hybridoma antibodies isolated were of IgA isotype. When lymphocytes from mesenteric lymph nodes (MLN) were used in fusions more non-specific I r A secretors were produced than when splenic lymphocytes were used, but the yield of specific ant;,-Eimeria IgA secreting hybridomas was not improved. By all protocols, a total of nine lgA secreting hybridomas were identified of which eight have been cloned for further studies. Key words: IgA; Monoclonal antibody; Parasite; Eimeria tenella
Introduction Eimeria tenella is an important protozoal parasite of the alimentary tract of the chicken. Since secretory IgA is the major immunoglobulin of mucosal secretions, a variety of studies has reported on the elaboration and possible role of
Correspondence to: A.J. Trees, Department ot Veterinary Parasitology, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool1.3 5QA, UK. Tel.: 051-708-9393; Fax: 051-708-8733. Abbreciations: DMEM, Dulbecco's modified Eagles medium; GALT, gut associated lymphoid tissue; GF, germfree; i.e., iatraenteric; IFAT, immunofluorescenceantibody test; i.p., intraperitoneal; Mab, monoclonal antibody; MLN, mesenteric lymph node; PP, Peyer's patch.
endogenous lgA in E. tenella infectioc-s (Orlans and Rose, 1972; Davis, Parry and Porter, lCf/8; Davis and Porter, 1979; Mockett and Rose, 1986; Lillehoj and Ruft, 1987; Rose and Hesketh, 1987; Trees et al., 1989, and others). To facilitate the further investigation of the role of I r A in protection at the mucosai surface and the identification of antigens relevant to protection at that site, we sought to generate IgA monoclonal antibodies (Mabs) to the infective stage (sporozoite) of E. tenella. We have produced murine IgA Mabs specific for the avian parasite for use in passive transfer experiments, on the basis that the binding of polymeric lgA to heterologous secretory component has been described for several species combinations, including mammalian to avian species (Rose et al., 1981; Peppard et al., 1983).
162 Compared to Mabs of IgG or IgM isotypes, there are relatively few reports of IgA Mabs and their frequency of occurrence following conventional immunisation and fusion plotocols is low (see Komisar et al., 1982). Murine lgA Mabs to various antigens have been produced based on more rational approaches to preferentially generate IgA secreting hybridomas by using either the alimentary route of immunisation or iymphocytes from gut associated lymphoid tissue (GALT) and mesenteric lymph nodes (MLN) in fusions (Colwell et al., 1982, 1983, 1986; Komisar et al., 1982; Mazanec et al., 1987; Weltzin et ai., 1989). These lgA secretors have been derived using different protocols involving different variables including antigens, immunisation/infection procedures, strains of mice, germ-free (GF) or conventional mice, organ sources of lymphocytes, myeloma fusion partners and fusion protocols and it is difficult to compare the efficiency of different protocols in generating specific IgA Mabs. Some of these IgA Mabs have been directed against mucosal bacteria (Komisar et ai., 1982; Colwell et al., 1983) or viruses (Mazanec et al., 1987; Weitzin et al., 1989). In this paper, we report the rational derivation of IgA Mabs against a mucosal parasite and compare the proportion of both nonspecific and specific IgA secreting hybridomas produced by selected different protocols.
Materials and methods
Mice Conventionally reared (from the School of Tropical Medicine, Liverpool) or GF (from Harlan Olac, Bicester, England) BALB/c mice were used. GF mice were maintained in a plastic film isolator and provided with sterile dry food (Harlan Plat) and water ad libitum. lmmunisation Beginning at 6-8 weeks of age, mice were immunised by inoculation of live sporozoites of the Houghton strain of Eimeria tenella (kind gift ot" Dr. M.E. Rose), prepared by standard techniques. Live parasites were used in order to optimise the yield of antibodies directed against native surface antigens. Mice were primed either by
the intraperitoneal (i.p.) injection of 250,000 sporozoites on two occasions, 4-6 weeks apart, or by a single intraenteric (i.e.) injection (into the lumen of the jejunum) of 2.5 x 106 sporozoites under general anaesthesia (by halothane inhalation or parenteral sodium pentobarbitone injection, Nembutal, Ceva, UK) via a small laparotomy incision at the umbilicus. All mice received a booster injection of 1 X 106 sporozoites intravenously 14-16 days post priming and 3 days prior to fusions. The serum antibody response of immunised mice was assayed by a three layer immunofluorescence antibody test (IFAT), see below. Fusion method Spleen or mesenteric lymph nodes (MLN) were used as the lymphocyte source. Between 1-3 × 108 lymphocytes were obtained per spleen compared to 3-8 x 107 per MLN. Cell fusion was according to established methods (Galfr~ and Milstein, 1981) with mi~aor modifications. Briefly, myeloma cells (aminopterin-sensitive, NS-1-Ag41, kind gift of Professor M. Hommel) and iymphocytes were mixed in the ratio 1"10 and 50% polyethylene glycol (MW 1500, BDH, England) was used as fusogen in serum free medium. The final cell suspension was made up in HAT selective medium and plated out into four to six 96-well plates per fusion. MLN fusions were plated onto 24 h old feeder layers in four 96-well plates per fusion. All media were based on Dulbecco's modified Eagle's medium (DMEM) (Gibco, Paisley), for further details see Hoare (1989). Hybridoma growth was scored from 7 days post fusion. Results are included only from fusions which produced more than 100 hybridomas from splenic fusions and more than 20 hybridomas from fusions with MLN lymphocytes. Screening, cloning and cryopreservation of hybridomas At 12 days post fusion, hybridoma supernatants were first assayed for specific antibody by a two step IFAT with whole sporozoites as antigen, after the method of Trees et al. (1985) but using an F1TC-conjugated anti-mouse lg antibody (Nordic Immunochemicals, Netherlands). Isotypes of antigen-specific antibodies were deter-
163 mined by Ouchterlony double diffusion in agar employing commercial, affinity purified Fcspecific antisera (Nordic Immunochemicals, Netherlands) and by a three layer IFAT including an isotype specific second antibody stage (see Trees et al., 1985). The specificity and isotype of each hybridoma was assayed on at least two occasions by the three layer IFAT. lsotyping of hybridoma antibodies of undetermined specificity was performed for one 96 well plate per fusion, by Ouchterlony double diffusion, on uncloned material. Cells from positive wells were cloned once or twice by limiting dilution, re-assayed for specific antibody reactivity and isotype and then cryopreserved.
Results The results of 14 fusions involving six different protocols are presented in Table i. The ratio of specific anti-Eimeria hybridoma antibodies per 107 cells fused was similar for both MLN and splenic lymphocytes for comparable fusions but, since more cells were available from the spleen, more specific bybridomas were produced from that source. In comparison with conventionally reared mice, more specific hybridomas of all isotypes were produced from fusions of lymphocytes from G F mice and this is reflected in the im-
proved hybridoma ratio of the latter. Comparing the source of cells for fusion, with respect to non-specific antibody production by bybridomas, IgA secretors were detected from all MLN fusions, but not from splenic fusions. This was not the case for specific lgA hybridoma antibodies, which were derived predominantly from spleen cell fusions. The combination of splenic iymphocytes used in fusions from G F mice after i.e. priming gave the most specific IgA hybridomas; both fusions involving this protocol yielded specific IgA hybridomas (two from one spleen, 16 x 107 lymphocyte fused, and five from the second, 22 × 107 lymphocytes fused). Although these mice had specific serum IgG antibody titres ( > 1/40 dilution) at fusion, neither had specific serum IgA titres detectable by TLIFA. In summary, of nine anti-Eimeria IgA hybridoma antibodies identified, seven resulted from fusions from splenic lymphocytes derived from G F mice after i.e. priming. Of the nine IgA hybridomas, eight were subsequently cloned and cryopreserved.
Discussion On the basis that GALT is an enriched source of IgA B cell precursors (Craig and Cebra, 1971), Komisar and co-workers (1982) used pooled MLN
TABLE ! COMPARATIVE vI'IELDSAND IgA ISOTYPE FREQUENCY OF NON-SPECIFIC AND OF SPECIFIC ANTI-EIMERIA SECRETING HYBRIDOMASFROM DIFFERENT FUSION PROTOCOLS Mouse type
Conventional Conventional Conventional Conventional Germ-free Germ-free
Roote of immunisation
Lymphocyte Totalcells source fused ( )< 107)
Hybridoma:' ratio
Parenteral Parenteral i.e. i.e. i.e. i.e.
S MLN S MLN S MLN
0.25 0.26 0.14 0.27 0.71 2.0
75 3.8 44 1I. 1 38 4.5
Number of specificanti-Eimeria hybridomas/107cells fused. b From one 96 well plate only, per fusion, see text. c ND, not done.
Secretingh~bridomas Non-specific Specific IgA/total h anti-Eimeria IgA/total ND " 1/19 ,;/22 0/1 0/24 0/6 2/28 0/3 0/76 7/27 2/17 1/9
no. fusions 4 2 2 2 2 2
164 and Peyer's patch (PP) as lymphocyte sources to produce lgA Mabs. However, various studies have shown that i.e. antigen presentation not only resuits in the induction of precursor IgA B cells in GALT but also IgA secreting and memory cells in the spleen (Michalek et al., 1980; Andrew and Hall, 1982; Jeurissen et al., 1985 and others). Moreover, Jeurissen et al. (1985) proposed that mucosal immunisation activates local IgA precursor cells which leave PP and migrate via MLN, lymph and blood to distant sites, including mucosae and spleen. These cells can then be reactivated by subsequent specific antigenic stimulation to differentiation and lgA production. This evidence indicated that i.e. priming might be efficient in promoting IgA memory cells in the spleen which could be reactivated by parenteral immunisation and used as fusion partners. Thus, although MLN has been proposed as a rational source of lymphocytes for fusions to generate IgA Mabs, our results do not demonstrate any advantage in the use of MLN compared to spleen. MLN was a poorer lymphocyte source, tending to give fewer lymphocytes and giving fewer hybridomas (even when the hybridoma ratio was elevated using GF mice). This disadvantage could have been offset by pooling MLN from 5-10 mice but this has practical and ethical objections. Although the theoretical possibility of obtaining large numbers of specific IgA secretors from MLN was not born out in this study, a higher frequency of lgA isotypes of undetermined specificity was observed from MLN fusions compared with those from splenic lymphocytes. These lgA secreting hybridomas were presumably derived from B cells precommitted to antigens derived from other enteric organisms and food antigens. Whilst there was a marked improvement of yield of antiEimeria bybridomas generated from GF compared to conventionally reared mice, the unchanged isotype ratio in hybridomas of undetermined specificity in fusions generating specific lgA hybridomas suggests that the induction of IgA secreting cells was andgen specific. However, the results do not indicate any preferential induction of specific IgA precursor cells in MLN compared to spleen in GF mice. Other GALT derived cells have been successfully used for fusions by other workers and Weltzin et al. (1989) found
PP generated a considerably higher proportion of specific IgA hybridomas to viral antigens than spleen, but there are considerable technical difficulties in the sterile isolation of PP lymphocytes as well as limitations due to the low numbers of lymphoeytes generated. Although conventional mice are usually used in attempts to derive IgA Mabs, the present results indicate the advantage of GF mice, first used by Colwell and co-workers (Colwell et al., 1982, 1983). The use of GF mice is based on evidence that bacterial lipopolysaccharides negatively influence splenic lgA responses to orally presented antigen (Babb and McGhee, 1980; Michalek et al., 1980; Michalek et ai., 1983). Although GF mice require more sophisticated handling, their merit in optimising the advantages of i.e. priming are likely to reduce the number of fusions, mice, effort and cost required to isolate IgA Mabs of the desired specificity. It is interesting that circulating specific IgA antibodies were not detected at fusion in those mice which yielded the most specific IgA secreting hybridomas. This may have been due to competitive binding of target antigen by high titre IgG isotype antibody. Whilst serum titres have relevance to the likelihood of specific hybridoma production, they may not correlate directly with the number of secreting hybridomas produced after fusion (Campbell, 1984). Our results suggest that lack of detectable specific circulating IgA antibody in mice immunised intraenterically should not discourage fusion and attempts to isolate IgA secreting hybridomas. In conclusion, a rational approach to the production of specific IgA secreting hybridomas i n ferred from these results would employ i.e. priming and intravenous boosting of young GF mice, and utilise splenic lymphocytes in an established fusion protocol. This technically straightforward protocol resulted in about 25% of specific antibody secreting hybridomas being of IgA isotype.
Acknowledgements We are very grateful to M~ J. Gale for technical advice and Miss. S.M. McKellar and Miss. S. Crozier for occasional laboratory assistance. This
165 w o r k w a s s u p p o r t e d by a n A g r i c u l t u r a l a n d F o o d Research Council Veterinary School's Fellowship to J.A.H.
References Andrew, E. and Hall, J.G. (1982) IgA antibodies in the bile of rats. I. Some characteristics of the primary response. Immunology 45, 169-175. Babb, J.L. and McGhee, J.R. (1980) Mice refractory to lipopolysaccharide manifest high immunoglobulin A responses to orally administered antigen. Infect. Immun. "~9, 322-328. Campbell, A.M. (1984) Monoclonal Antibody Technology. Elsevier, Amsterdam, p. 88. Colwell, D.E., Gollahon, K.A., McGhee, J.R. and Michalek, S.M. (1982) lgA hybridomas: a method for generation in high numbers. J. Immunol. Methods 54, 259-266. Colwell, D.E., Gollahon, K.A., Morisaki, I., McGbce, J.R. and Michalek, S.M. (1983) IgA hybridomas to cell surface components of pathogenic bacteria. Ann. NY Acad. Sci. 409, 794-796. Colwell, D.E., Michalek, S.M. and McGhee, J.R. (1986) Method for generating a high frequency of hybridomas producing monoclonal lgA antibodies. Methods Enzymol. 121, 42-51. Craig, S.W. and Cebra, J.J. (1971) Peyer's patches: an enriched source of precursors for lgA-producing immunocytes in the rabbit. J. Exp. Med. 134, 188-200. Davis, PJ. and Porter, P. (1979) A mechanism for secretory lgA-mediated inhibition of the cell penetration and intracellular development of Eimeria tenella. Immunology 36, 471-477. Davis, P.J., Parry, S.H. and Porter, P. (1978) The role of secretory IgA in anti-coccidial immunity in the chicken. Immunology 34, 879-888. Galfr~, G. and Milstein, C. (1981) Prcparation of monoclonal antibodies: strategies and procedures. Methods Enzymol. 73, 1-46. Hoare, J.A. (1989) The production and evaluation of IgA monoclonal antibodies directed against Eimeria tenella sporozoites. Ph.D. Thesis, University of Liverpool. Jeurissen, S.H.M., Claassen, E., Van Rooijen, N. and Kraal, G. (1985) Intra-intestinal priming leads to antigen-specific
IgA memory cells in peripheral lymphoid organs. Immunology 56, 417-423. Komisar, J.L., Fuhrman, J.A. and Cebra, JJ. (1982) IgA-producing hybridomas are readily derived from gut-associated lymphoid tissue. J. Immunol. 128, 2376-2378. Lillehoj, H.S. and Ruff, M,D. (1987) Comparison of di~ase susceptibility and subclass-specific antibody response in SC and FP chickens experimentally inoculated with Eimeria tenella, E. acerculina or E. maxima. Avian Dis. 31, !12-119. Mazanec. M.B., Nedrod, J.G. and Lamm, M.E. (1987) lmmunoglobulin A monoclonal antibodies protect against Sendal virus. J. Virol. 61, 2624-2626. Michalek, S.M., Kiyono, H., Babb, J.L and McGhee, J.R. (1980) Inheritance of LPS non-responsiveness and elevated splenic IgA immune response in mice orally immunized with heterologous erythrocytes. J. lmmunol. 125, 222O-2224. Michalek, S.M., McGhee, J.R., Kiyono, H., Colwell, D.E., Eldridge, J.H., Wannemuehler, M.J. and Koopman, WJ. (1983) The IgA response: inductive aspects, regulatory cells and effector functions. Ann. NY Acad. Sci. 409, 48-71. Mocken, A.P.A. and Rose, M.E. (1986) Immune responses to eimeria: quantification of antibody isotypes of Eimeria tenella in chicken serum and bile by means of the ELISA. Parasite Immunol. 8, 481-489. Orlans, E. and Rose, M.E. (1972) An IgA-like immunoglobulin in the fowl. Immunochemistry 9, 833-838. Peppard, J.V., Rose, M.E. and Hesketh, P. (1983) A functional homologue of mammalian secretory component exists in chickens. Eur. J. lmmunol. 13, 566-570. Rose, M.E. and Hesketh, P. (1987) Eimeda tenella: effects of immunity on sporozoites within the lumen of the small intestine. Exp. Parasitol. 63, 337-344. Rose, M.E., Orlans, E., Payne, A.W.R. and Hesketh, P. (1981) The origin of igA in chicken bile: its rapid active transport from blood. Eur. J. Immunol. 11,561-564. Trees, A.J., Crozier. S.J., McKellar, S.B. and Wachira, T.M. (1985) Class-specific circulating antibodies in infections with Eimeria tenella. Vet. Parasitol. 18, 349-357. Trees, A.J.0 Karim, M.J., McKellar, S.B. and Car~er, S.D. (1989) Eimeria tenella: local antibodies and interactions with the sporozoite surface. J. Protozool. 36, 326-333. Weltzin, P.L.-J., Micheni, P., Fields, B.N., Kraehenbohl. J.P. and Neutra, M.R. (1989) Binding and transepithelial transport of immunoglobulins by intestinal M cells: demonstration using monoclonal lgA antibodies against enteric viral proteins. J. Cell Biol. 108, 1673-1685.
