Anti-idiotypic antibodies function as a surrogate surface epitope of Brugia malayi infective larvae

Acta Tropica, 47(1990)391-397 Elsevier

391

ACTROP 00098

Anti-idiotypic antibodies function as a surrogate surface epitope of Brugia malayi infective larvae Clotilde K.S. Carlow, Patricia Busto*, Neil Storey and

Mario Philipp Molecular Parasitology Group, New England Biolabs, Beverly. MA, U,S.A.

Anti-idiotypic (AB2) antibodies were generated in rabbits following immunization with a murine IgM monoclonal antibody (ABI) recognizing a surface determinant of Brugia malayi infective stage larvae. AB2 specifically inhibited the binding of ABI to B. malayi larvae. Furthermore, AB2 had the ability to mimic the original antigen since mice immunized with AB2 possessed serum antibodies (AB3) specific for the B. malayi surface determinant. The presence of anti-surface antibodies (AB3 and AB1) induced either by AB2 immunization or by administration of ABI, did not alter the outcome of an intraperitoneal infection of B. malayi larvae in BABL/c mice when compared to untreated animals. AB3 antibodies like AB1, were IgM, thus indicating an isotype restricted response to the B. malayi epitope. There were no detectable cell mediated responses to the surface determinant in mice immunized with AB2, assessed by lymphocyte blastogenesis or IL3 production in vitro in response to the idiotope as presented by living larvae. The lack of cellular responses and/or the previously demonstrated rapid shedding of the epitope may explain the inability of AB1 or AB2 to protect mice against larval challenge in this study. Key words: Brugia; Larvae; Anti-idiotype; Surface-antigen; Vaccine

I m m u n e - m e d i a t e d resistance to larval Brugia malayi infection in rodent model systems has been induced by various parasite preparations including living B. malayi infective stage larvae (Denham, 1980; Hayashi et al., 1984; Yates and Higashi, 1985; Carlow and Philipp, 1987; A b r a h a m et al., 1989) and saline extracts prepared from this stage (Carlow and Philipp, 1987). The antigens responsible and the immune effector mechanisms involved are yet to be characterized. By analogy with other nonfilarial helminth infections (Grencis et al., 1986; C a p r o n et al., 1987; H a r n 1987) surface antigens m a y be the targets o f lethal immune attack. The surface o f B. malayi infective stage larvae contains an antigenic determinant expressed in a developmentally regulated m a n n e r (Carlow et al., 1987b). Using an I g M m o n o c l o n a l a n t i b o d y (D1E5) we f o u n d that this epitope is confined exclusively to late second and third stage larvae. Interestingly, within 2 - 3 days post infection o f the m a m m a l i a n host the epitope is no longer present on the larval surface (Carlow et al., 1987c). Due to its surface localization and brief expression which coincides with the transmission period, the antigen bearing this epitope m a y have a role in parasite transmission and/ or be a target o f a protective i m m u n e response. We were therefore interested in Correspondence address." C.K.S. Carlow, Molecular Parasitology Group, New England Biolabs, Beverly, MA 01915, U.S.A. [Tel: 508-927-5054; Fax: 508-921-1350] *Present address." Pathology Department, Tufts University School of Medicine, Boston, MA 02111, U.S.A.

0001-706X/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

392 investigating its immunological properties but numerous attempts to identify and isolate this antigen failed (Carlow et al., 1987b). In particular, the antibody D1E5 did not react with larval extracts immobilised on solid phases, which may in part explain our inability to identify any positive clones after screening one genomic DNA equivalent of recombinants from a B. malayi genomic D N A library (Arasu et al., 1987) in the bacteriophage lambda g t l l (unpublished data). As an alternative we attempted to produce in rabbits anti-idiotypic (anti-id or AB2) antibodies containing an 'internal image' of the D1E5 idiotope, to function as a surrogate antigen. Our initial approach to determine if the epitope is involved in the murine protective immune response was to attempt passive transfer of resistance by administration of the monoclonal antibody during larval infection. Groups of 10 female BALB/cByJ mice, 6-8 weeks old (Jackson Laboratory, Bar Harbor, Maine) received three intraperitoneal (i.p.) injections of 150 Bg purified D1E5 antibody on days 0, 1 and 2 of infection. Control groups received equivalent amounts of an IgM monoclonal antibody (C6C6) reactive with B. malayi somatic antigen, or remained untreated. Antibodies were purified by precipitation in 7% polyethylene glycol (PEG 6000, Sigma) followed by gel filtration on Sephacryl S-200 (Pharmacia). Infections were initiated by an i.p. injection of 50 B. malayi infective larvae isolated from Aedes aegypti mosquitoes (purchased from TRS Laboratories Inc., Athens, Georgia). Parasites were incubated in a 1:50 dilution of appropriate ascitic fluid, or left untreated prior to challenge. Infections were quantified subsequently as previously described (Carlow and Philipp, 1987). There was no significant difference ( P > 0.05) in parasite recoveries expressed as a percentage of the inoculum from mice given DIE5 or C6C6, or left untreated on days 5, 7 or 9 post infection (Table 1). Assuming that anti-surface humoral responses are involved in protection, the failure of AB1 to passively transfer resistance to larval challenge in mice could be explained if the monoclonal antibody is of low affinity or inappropriate isotype. Indeed there is evidence to suggest that protective humoral responses against filariae are isotype dependent (reviewed by Philipp et al., 1984). However in this system perhaps the most likely explanation is the rapid shedding of the epitope, since more recently we discovered that the epitope in the form of an immune complex has a half-life on the surface of only 2.5 minutes (Carlow et al., 1987c; Philipp et al., 1988). In addition, lymphocyte but not antibody transfer experiments have been successful in reconstituting immune competence in immunodeficient (nude) mice, implicating a role for T cells in murine resistance to Brugia larvae (Vickery et al., 1983). TABLE 1 Recoveries"of Brugia malayi larvae from BALB/cmice followingadministration of monoclonalantibody D1E5

Day 5 post challenge Day 7 post challenge Day 9 post challenge

D 1E5b

C6C6c

Untreateda

21.2+3.9 18+3.1 4± 1

23_+1.9 16+_2.9 3 ± 0.8

22+2.1 15+2.1 5.5 + 1.4

aWonn recoveries(% +S.E.) on days 5, 7 or 9 after challengeinfection (50 L3 i.p.), 10 mice per group. bMice received 150 lag purified D1E5 on days 0, 1 and 2 of infection.

CMicereceived 150 lag purified C6C6 on days 0, 1 and 2 of infection. dUntreated controls.

393

To generate a multifactorial response to the D1E5 epitope, mice were actively immunized with purified 'antigen' in the form of rabbit anti-id antibodies. These antibodies were generated in two rabbits by intramuscular (i.m.) injection in multiple sites with a total of 300 lag purified D1E5 in Freunds' complete adjuvant (FCA), followed 10 days later with an identical injection. After 6 weeks rabbits received a booster injection of 300 lag antibody in Freunds' incomplete adjuvant (FIA) subcutaneously (s.c.) in multiple sites. Sequential bleeds were then assayed for rabbit antimouse immunoglobulin activity using a biotin-streptavidin amplified enzyme-linkedimmunosorbent assay (ELISA) with murine IgM as antigen. A pool of serum was made from each rabbit and an immunoglobulin (Ig) fraction was obtained by precipitation with ammonium sulfate (50% final concentration). The Ig salt cut was then purified by affinity chromatography on a protein A-Sepharose column (Pharmacia). Anti-IgM antibodies not directed to the D1E5 idiotope were absorbed by a series of passages through a Sepharose 4B column coupled with 2 heterologous murine IgM monoclonal antibodies. Unbound Ig was then passed through a Sepharose 4B column coupled with purified D1E5 antibody. Bound anti-id antibodies (AB2) were eluted with 1 M acetic acid, 0.15 M sodium chloride and the low pH neutralized with 2 M Trizma base. To examine the anti-id specificity of AB2, classical tests of competition were performed using D1E5 and a control IgM monoclonal antibody, DIM-72, which reacts with a surface determinant of Dirofilaria immitis infective larvae (Abraham et al., 1988). Each monoclonal antibody was incubated with 2 lag rabbit anti-id for 2 h at 37°C. B. malayi or D. immitis (TRS Labs) larvae were then added to the appropriate mixture and an immunofluorescent test (IFAT) was performed. Controls consisted of larvae incubated in homologous respective antibody alone, or normal mouse serum (NMS), followed by fiuorescein-conjugated goat anti-mouse immunoglobulins antibody (Cooper, Malvern, PA). Inhibition of the binding of D1E5 to the surface of B. malayi larvae by anti-id antibodies was observed and shown to be specific since AB2 had no effect on the immunostaining ofD. immitis infective larvae with antibody DIM-72 (Table 2). This result demonstrates that competition was due to the presence of anti-idiotypic and not anti-isotypic antibodies. However, inhibition of AB 1 reactivity by AB2 cannot be used as a sole criterion to distinguish anti-paratopic and anti-non paratopic responses. In this respect induction of an antigen-specific antibody response by anti-id immunization in an animal which has never encountered parasite antigen is conclusive. Consequently, mice received a total of 150 lag anti-id or TABLE 2 Inhibition of IgM monoclonal antibodies binding to larval surfaces by AB2 antibodies Infective larva

Antibodies

Binding

Brugia malayi

D 1E5 D1E5+AB2 NMS

+

Dirofilaria immitis

DIM-72 DIM-72 + AB2 NMS DIE5

+ +

394

protein A purified normal rabbit IgG antibody (Cappel, Malvern, PA) as 3 divided doses emulsified in FCA, in FIA, and in saline, at 2-3 week intervals. AB3 (anti anti-id antibodies) screening was based on the detection of antibodies reactive with the surface of B. malayi infective stage larvae in IFAT. Anti-id immunization generated an antibody response against the parasite surface (Fig. 1A) with the species specificity (as determined by IFAT using B. pahangi as a control) characteristic of DI E5 (Carlow et al., 1987b). Surface reactivity was not observed following immunization of mice with normal rabbit immunoglobulins. Groups consisting of 10 mice were immunized with anti-id antibodies or normal rabbit immunoglobulins as above. One week after completion of immunization, all animals including a control group were challenged with 100 living B. malayi infective larvae i.p. Parasites were recovered 6 days later. In the 2 experiments performed there was no significant difference (P>0.05) in parasite recoveries from AB2-, or normal rabbit immunoglobulin-treated or control mice (Table 3). We investigated whether immunization of mice with AB2 induced a cellular anti-

.A

100o0ol!1:2000 B

125000 ]

E

o

•

T

[]

anti id

75000

E

1500

50000

~

1000

25000

8

0

~ Medium

Con A

C

2500

control

5oo 0

LPS

Larvae

Medium

Con A

LPS

Larvae

Fig. l. Humoral and cellular responses generated in mice by anti-id (AB2) immunization. Specific antibody response to surface of Brugia malayi infective stage larvae (A). Spleen cell responses (B) and IL3 produced (C) following stimulation with Brugia malayi infective larvae in vitro.

395 TABLE 3 Recoveriesa of bodies

Experiment 1 Experiment 2

Brugia malayi larvae from BALB/c mice immunized with anti-DIE5 anti-idiotypic antiAnti-idiotype b

Normal Ig ¢

Control a

53.6_+ 5.2 44.6 _+5.6

41.9_+ 5.6 42.4 _+9.4

42.7_+ 2.7 50.7 _+9.4

"Worm recoveries (% +standard error) on day 6 after challenge infection (100 L3 i.p.), 10 mice per group. b0.15 mg of rabbit antbD 1E5 anti-idiotypic antibodies, given s.c. in three 50 p.g doses at 2-3 weeks interval (in FCA, FIA, Saline), Challenged 1 (Expt. 1) or 2 (Expt. 2) weeks after last injection. cSame as in (b) but using normal rabbit immunoglobulin. aNo treatment.

anti-id response by measuring spleen cell proliferative responses to living larvae in vitro. In addition, T cell responses were quantified more precisely by assaying for interleukin 3 (IL3) production by parasite stimulated spleen cells. Spleens were recovered both from naive and AB2 immunized mice. Single cell suspensions were prepared containing 10v viable cells/ml culture medium (RPMI 1640 supplemented with 1 mM glutamine, 100 i.u./ml penicillin, 100 p.g/ml streptomycin, 100 ~tg/ml Amphotericin-B, 20 mM HEPES, 0.06% sodium bicarbonate (all from Gibco, Grand Island, New York)) and 10% fetal calf serum (Hyclone, Logan, UT). Aliquots of 50 ~tl were dispensed into wells of 96 well flat bottomed microtitre plates (Costar, Cambridge, MA). A 50 ~tl volume containing 5-10 living larvae or iipopolysaccharide (LPS, 25 ~tg/ml, Sigma) or Concanavalin A (ConA, 2.5 p.g/ml, Sigma) and 100 ~tl culture medium were then added. Plates were incubated for 48 h at 37°C, 5% CO2. Supernatants (75 ~tl) were then removed for the IL3 assay, volumes were readjusted to 200 ~tl and 50 p.1of medium containing 0.5 ~tCi tritiated methyl-thymidine ([3H]TdR) (Amersham) were added to each well. After a further 16 h incubation, the cells were harvested onto glass-fibre filters and processed for liquid scintillation counting. Each test was performed in quadruplicate. IL3 production was measured using the method described by Dotsika and Sanderson (1987) employing the IL3-dependent 32D basophil/mast cell line. Briefly, 10 !al of test supernatant collected as above was incubated w i t h 10 4 32D cells in 90 ~tl tissue culture medium for 24 h at 37°C. Four h prior to harvesting 0.5 ~tCi [3H]TdR was added to each well. Tests were performed in quadruplicate. Cells were harvested and counted as above. Proliferative responses (Fig. 1B) and/or IL3 production (Fig. 1C) were only detected in wells containing mitogens. Therefore stimulation o f T cells by anti-id immunization remains questionable since we were unable to detect T cell proliferation or IL3 production, secreted by both TH1 and TH2 lymphocytes (Cherwinsky et al., 1987), in response to the epitope presented by living larvae. In addition, no delayed type hypersensitivity responses were elicited in mice immunized with AB2 following footpad injections of dead B. malayi larvae, or in mice immunized i.p. with living B. malayi infective stage larvae and challenged in the foot pad with AB2. However, a response was observed in AB2 immunized mice challenged in the footpad with rabbit Ig (data not shown). Our results contrast the numerous reports of effective cell mediated immune responses induced by AB2 antibodies (Reviewed by Sacks et al., 1982). However they are consistent with an IgM isotype restricted response to the epitope. Analysis of the AB3 response using anti-heavy chain specific antibodies in immunofluorescence re-

396 vealed an IgM isotype response exclusively. In addition, an IgM monoclonal antibody reactive with the surface of B. malayi infective stage larvae was generated using spleen cells from mice demonstrating an AB3 response. Therefore, surprisingly, both a single brief exposure to infective larvae (Carlow et al., 1987a) and repeated immunization with AB2 generated antibodies (AB1 and AB3) of IgM isotype. The possibility exists however that multiple immunization of mice with AB2 may stimulate a population of AB3 antibodies with higher avidity than AB1. Since the protective efficacy of IgM monoclonal antibodies may be a function of avidity (Pincus et al., 1988), the AB3 antibodies generated in the present system could have been superior to D1E5 in this respect. Nevertheless AB3 did not engender any protective immunity. In summary, we adopted the strategy of producing anti-id antibodies to a surface determinant of B. malayi infective stage larvae since other approaches had failed to isolate and identify the molecule. This approach has been shown to be particularly useful when dealing with carbohydrate epitopes (Grzych et al., 1985) which are abundant in surface and secreted filarial antigens (Maizels et al., 1987). We successfully produced anti-id antibodies (AB2) which functioned as surrogate antigen in their ability to specifically inhibit the binding of AB 1 to the parasite surface and to generate further antibodies with specificity identical to AB1. The data presented here strongly suggests that in filariasis an isotype restricted response to an epitope can occur. Recently a similar phenomenon was described in experimental schistosomiasis in which preferential IgE responses to a protective larval antigen were observed following AB2 immunization (Velge-Rousell et al., 1989). Considering that an IgM restricted response was observed it is perhaps not surprising that we found no cell mediated responses to the D1E5 epitope following AB2 immunization nor after injections of living infective larvae. The lack of cell mediated responses and/or the rapid shedding of the epitope (Carlow et al., 1987c; Philipp et al., 1988) may be the reason for the lack of protection against larval challenge in the presence of a surface directed antibody response.

Acknowledgements We gratefully acknowledge the support of Dr. Donald G. Comb. We thank Dr Claude V. Maina for help and advice in screening the genomic D N A library, Dr. Carole Prain for useful comments on the manuscript, Drs. S.F. Wolf and S.C. Clark (Genetics Institute) for the gift of the 32D cell line, and Dr. R.B. Grieve from Colorado State University for providing the DIM-72 antibody.

References Abraham, D., Grieve, R.B., Holy, J.M. and Christensen, B.M. (1989) Immunity to larval Brugia malayi in BALB/cmice: protective immunity and inhibition of larval development. Am. J. Trop. Med. Hyg. 40, 598-604. Abraham, D., Grieve, R.B. and Mika-Grieve, M. (1988) Dirofilaria immitis: surface properties of thirdand fourth-stage larvae. Exp. Parasitol. 65, 157-167. Arasu, P., Philipp, M. and Perler, F. (1987) Brugia malayi: recombinant antigens expressed by genomic DNA clones. Exp. Parasitol. 64, 281-291. Capron, A., Dessaint, J.P., Capron, M., Ouma, J.H. and Butterworth, A.E. (1987) Immunity to schistosomes: progress towards a vaccine. Science238, 1065-1072.

397 Carlow, C.K.S., Edwards, M.K., James, E.R. and Philipp, M. (1987a) Monoclonal antibodies to parasite antigens: a rapid immunization protocol requiring small numbers of parasites. J. Parasitol. 73, 10541057. Carlow, C.K.S., Franke, E.D., Lowrie, R.C., Partono, F. and Philipp, M. (1987b) Monoclonal antibody to a unique surface epitope of the human filaria Brugia malayi identifies infective larvae in mosquito vectors. Proc. Natl. Acad. Sci. U.S.A. 84, 6914-6918. Carlow, C.K.S., Perrone, J., Spielman, A. and Philipp, M. (1987c). A developmentally regulated surface epitope expressed by the infective stage larvae of Brugia malayi which is rapidly lost after infection. UCLA Symp. Mol. Cell. Biol. 60, 301-310. Carlow, C.K.S. and Philipp, M. (1987) Protective immunity to Brugia malayi larvae in BALB/c mice: potential of this model for the identification of protective antigens. Am. J. Trop. Med. Hyg. 37, 597-604. Cherwinsky, H.M., Schumacher, J.H., Brown, K.D. and Mossman, T.R. (1987). Two types of mouse helper T cell clone. III. Further differences in lymphokine synthesis between TH1 and TH2 clones revealed by RNA hybridization, functionally monospecific bioassays, and monoclonal antibodies. J. Exp. Med. 166, 1229-1244. Denham, D.A. (1980) Vaccination against filarial worms using radiation-attenuated vaccines. J. Nuc. Med. Biol. 7, 1105-1111. Dotsika, E.N. and Sanderson, C.J. (1987) Interleukin 3 production as a sensitive measure of T lymphocyte activation in the mouse. Immunology 62, 665-668. Grencis, R.K., Crawford, C., Pritchard, D.I., Behnke, J.M. and Wakelin, D. (1986) Immunization of mice with surface antigens from muscle larvae of Trichinella spiralis. Parasite Immunol. 8, 587-596. Grzych, J.M., Capron, M., Lambert, P.H., Dissous, C., Torres, S. and Capron, A. (1985) An anti-idiotype vaccine against experimental schistosomiasis. Nature 316, 74-76. Harn, D.A. (1987) Immunization with schistosome membrane antigens. Acta Trop. 44 (suppl. 12), 46-49. Hayashi, Y., Noda, K., Sirasaka, A., Nogami, S. and Nogami, S. (1984) Vaccination of BALB/c mice against Brugia malayi and Brugia pahangi with larvae attenuated by gamma irradiation. Jap. J. Exp. Med. 54, 177-181. Maizels, R.M., Bianco, A.E., Flint, J.E., Gregory, W.F., Kennedy, M.W., Lim, G.E., Robertson, B.D. and Selkirk, M.E. (1987) Glycoconjugate antigens from parasitic nematodes. UCLA Symp. Mol. Cell. Biol 60, 267-279. Philipp, M., Davis, T.B., Storey, N. and Carlow, C.K.S. (1988) Immunity in filariasis: perspectives for vaccine development. Annu. Rev. Microbiol. 42, 685-716. Philipp, M., Worms, M.J., Maizels, R. and Ogilvie, B.M. (1984) Rodent models of filariasis. Contemporary Topics in Immunobiology 12, 275-321. Pincus, S.H., Shigeoka, A.O., Moe, A.A., Ewing, L.P. and Hill, H.R. (1988) Protective efficacy of IgM monoclonal antibodies in experimental Group B streptococcal infection is a function of antibody avidity. J. Immunol. 140, 2779-2785. Sacks, D.K., Esser, K.M. and Sher, A. (1982) Immunization of mice against African trypanosomiasis using anti-idiotypic antibodies. J. Exp. Med. 155, 1108-1119. Velge-Roussel, F., Verwaerde, C., Grzych, J.M., Auriault, A. and Capron, A. (1989) Protective effects of anti-antiidiotypic IgE antibodies obtained from an IgE monoclonal antibody specific for a 26-kilodalton Schistosoma mansoni antigen. J. Immunol. 142, 2527-2532. Vickery, A.C., Vincent, A.L. and Sodeman, W. (1983) Effect of immune reconstitution on resistance to Brugia pahangi in congenitally athymic nude mice. J. Parasitol. 69, 478-485. Yates, J.A. and Higashi, G.I. (1987) Brugia malayi: vaccination ofjirds with 6°Cobalt-attenuated infective stage larvae protects against homologous challenge. Am. J. Trop. Med. Hyg. 34, 1132-1137.