INFECTION AND IMMUNITY,0019-9567/99/$04.0010

Sept. 1999, p. 4661–4667 Vol. 67, No. 9

Copyright © 1999, American Society for Microbiology. All Rights Reserved.

Dominance of Immunoglobulin G2c in the AntiphosphorylcholineResponse of Rats Infected with Trichinella spiralis

PHILIP J. PETERS,1† LUCILLE F. GAGLIARDO,1 ELIZABETH A. SABIN,2‡ ARIN B. BETCHEN,1§KAYA GHOSH,1 JEB B. OBLAK,1 AND JUDITH A. APPLETON1,2*

James A. Baker Institute for Animal Health1 and Department of Microbiology and Immunology,2

College of Veterinary Medicine, Cornell University, Ithaca, New York 14853

Received 12 March 1999/Returned for modification 11 May 1999/Accepted 3 June 1999

The antibody response to the L1 stage of Trichinella spiralis has been described as biphasic. Worms residentin the intestine during the first week of infection stimulate an antibody response against a subset of larvalproteins. L1 larvae in the muscle at the end stage of infection stimulate a second antibody response againsttyvelose-bearing glycoproteins. Antityvelose antibodies protect rats against challenge infection with larvae. Theaim of this study was to characterize the rat B-cell response against larval antigens during the intestinal phaseof T. spiralis infection and to test the antiparasitic effects of such antibodies. Strain PVG rats were infectedorally with 500 larvae. Antibodies specific for phosphorylcholine-bearing proteins of L1 larvae first appearedin serum 9 days postinfection. Absorption experiments showed that the majority of antilarval antibodiesproduced in rats 16 days after infection with T. spiralis were specific for phosphorylcholine-bearing proteins.A fraction of these antibodies bound to free phosphorylcholine. Immunoglobulin G2c (IgG2c) producing cellsin the mesenteric lymph node dominated this early antibody response. IgG2c is associated with T-independentimmune responses in the rat; however, a comparison of athymic rats with euthymic controls suggested that onlya small fraction of the phosphorylcholine-related antibody response against T. spiralis was T independent.Phosphorylcholine is a common epitope in antigens of bacteria and nematode parasites and has been shownto be a target of protective immunity in certain bacteria. A monoclonal IgG2c antibody was prepared frominfected rats and shown to be specific for phosphorylcholine. Monoclonal phosphorylcholine-specific IgG2cfailed to protect rats against intestinal infection with T. spiralis. Therefore, our findings do not support a rolefor phosphorylcholine-bearing antigens in immune defense against T. spiralis; however, the potency of theimmune response induced suggests an immunomodulatory role for the lymphocytes involved.

Trichinella spiralis parasitizes many mammalian species andhas served as a model for the study of parasitic nematodes andmucosal immunity. The rat is a natural host for T. spiralis.Infective larvae (L1 stage) are freed in the stomach when a ratconsumes an infected carcass. Larvae enter the small intestine,where they undergo a series of four molts in a period of 30 h.Worms reside in an epithelial habitat (41). Adult worms re-produce, and newborn larvae migrate from the gut to striatedmuscle, where they invade cells and mature to the infectious L1stage (reviewed by Despommier [10]).

Different antigens are expressed during each life stage of T.spiralis. Philipp et al. (28) showed that the proteins displayedon the epicuticle change with each molt. Jungery and Ogilvie(16) reported that the host antibody response mirrors thechanges in antigen expression. In addition to the antigenicvariation observed in different life stages, Denkers et al. re-ported that the antibody response directed against L1 larvae isbiphasic (8). Of the two responses, the one that has beenstudied most intensively is induced late in infection by muscle-stage L1 larvae. Antibodies are induced by glycans attached to

a number of larval proteins found in the secretory cells, orstichocytes, as well as on the body surface (4, 12). These gly-cans are tri- and tetra-antennary structures that are cappedwith 3,6-dideoxy-D-arabino-hexose (tyvelose) (11, 31, 40). Ty-velose forms the immunodominant epitope. This response isimportant to the survival of host and parasite, as shown bypassive immunization of rats with antityvelose MAbs. Antibod-ies eliminate up to 99% of an oral challenge dose of infectiveT. spiralis larvae within a few hours. IgG1 and IgG2c appear tobe the most protective antibody subclasses (1, 3, 5).

The other phase of the antilarval response is less well char-acterized, but indirect evidence suggests that it is specific forphosphorylcholine-bearing glycoproteins (35). Phosphorylcho-line is widely distributed in tissues, including the cuticle, epi-dermis, hypodermis, hemolymph, and intestinal gland (15).Ubeira et al. (36) investigated the splenic B-cell response in-duced by phosphorylcholine antigens in mice infected withTrichinella; however, the more relevant, intestinal B-cell re-sponse has not been studied. Phosphorylcholine immunogenshave been found in bacteria and nematodes and serve as tar-gets of protective immunity against certain bacteria (19). Inthis report, we describe the rat B-cell response against phos-phorylcholine-bearing glycoproteins during the intestinalphase of T. spiralis infection as well as results of experiments totest the antiparasitic effects of such antibodies.

(Some of these data were presented at the 75th AnnualMeeting of the Conference of Research Workers in AnimalDiseases, Chicago, Ill. [abstract no. 142], and at the 9th Inter-national Congress on Immunology, San Francisco, Calif. [pro-gram no. 931].)

* Corresponding author. Mailing address: James A. Baker Institutefor Animal Health, College of Veterinary Medicine, Cornell Univer-sity, Ithaca, NY 14853. Phone: (607) 256-5600. Fax: (607) 256-5608.E-mail: jaa2@cornell.edu.

† Present address: College of Medicine, Cornell University, NewYork, NY 10021.

‡ Present address: American Veterinary Medical Association,Schaumburg, IL 60173-4360.

§ Present address: Department of Biomedical Sciences, San Fran-cisco, CA 94143.

4661

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

MATERIALS AND METHODS

Abbreviations. ASC, antibody-secreting cells; BSA, bovine serum albumin;CLN, cervical lymph node; DPBS, Dulbecco’s phosphate-buffered saline; ECL,enhanced chemiluminescence; ELISA, enzyme-linked immunosorbent assay;ELISPOT, enzyme-linked immunospot; HRP, horseradish peroxidase; IFN-g,gamma interferon; Ig, immunoglobulin; IL, interleukin; MAb, monoclonal anti-body; MLN, mesenteric lymph node; TI-2, thymus-independent type 2.

Rats. Strain AO and strain PVG rats were produced and maintained underspecific-pathogen-free conditions. Outbred, male NIH-RNU (athymic) andNIH-RNU/1 (euthymic) rats were obtained from the Frederick Cancer Re-search and Development Center and were barrier maintained. All rats werehoused in the James A. Baker Institute vivarium according to American Asso-ciation for Accreditation of Laboratory Animal Care guidelines.

Parasite. T. spiralis (pig strain) was maintained in irradiated adult rats. Infec-tious larvae were harvested from muscle by digestion in a solution of 1% pep-sin–1% HCl at 37°C. Experimental rats were infected orally with larvae sus-pended in a volume of 0.2 to 0.5 ml of 0.6% nutrient broth–2% gelatin (2).

Antigen. Crude larval antigen and excretory/secretory antigens of muscle lar-vae were prepared as previously described (4). Phosphorylcholine-bearing anti-gens were isolated from crude antigen by affinity chromatography with MAb 6G3(see below).

Antibodies. Normal rat serum was collected by cardiac puncture from adultrats anesthetized with ether. Sera were collected at the times specified (see

Results) from rats infected with 500 or 1,000 T. spiralis larvae. Monoclonal ratantibody 9E6 (IgG2c) is specific for tyvelose-capped glycans of T. spiralis (3, 11).

Mouse MAbs specific for rat IgA, IgM, and HRP-conjugated polyclonalanti-rat IgG2c or IgE were obtained from Serotec (Harlan, Indianapolis,Ind.). Mouse MAbs specific for rat IgG2a, IgG2b, and IgG1 were a gift fromT. Springer (Harvard University, Boston, Mass.) (34).

Monoclonal rat antibody 6G3 was produced by somatic cell hybridization (17).MLN cells were collected from a PVG rat which had been infected orally with2,500 larvae 10 days previously. The MLN cells were processed in the mannerdescribed below (“Preparation of Cells”). Rat lymphocytes and mouse myelomacells (SP2/0) were fused at a ratio of 2:1. Fused cells were plated in 96-well plateson rat peritoneal exudate cells. Hybridomas were screened by ELISA in order todetect IgG2c specific for crude T. spiralis larval antigen. Positive hybridomaswere cloned by limiting dilution on rat peritoneal exudate cells. Ascites fluid wasproduced in nude mice by Harlan Sprague-Dawley (Indianapolis, Ind.).

ELISA. Crude larval antigen (5 mg/ml, 50 ml/well) was incubated in polyvinylmicrotiter plates overnight at 4°C. Wells were blocked with DPBS containing0.2% gelatin for 15 min. The remainder of the procedure was performed asdescribed previously (3), except that for hybridoma screening, antibody bindingwas detected with peroxidase-conjugated rabbit antibody specific for rat IgG2c(Serotec).

In one experiment, pooled serum (collected from rats 16 days postinfection) orMAb 6G3 was absorbed with MAb 6G3 affinity-purified antigens or with BSA by

FIG. 1. Kinetics of the T. spiralis-specific ASC response in three lymphoid tissues: MLN (A), CLN (B), and spleen (C). PVG rats were infected with an oral doseof 867 T. spiralis muscle larvae. The numbers of specific ASC were determined by ELISPOT on days 8, 16, 30, and 50 postinfection. Values expressed are the means 1standard deviations (for three rats or, on day 50, two rats). p, IgE was not assayed in this experiment.

4662 PETERS ET AL. INFECT. IMMUN.

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

incubating antibody solutions on antigen-coated (5 mg of protein/ml) petri dishesfor 2 h at 4°C. Absorbed antibody preparations were tested by ELISA for bindingto crude larval antigen. Antibody binding was detected by using peroxidase-con-jugated goat antibody specific for rat IgG (heavy- and light-chain reactive) (3).

Preparation of cells. MLNs, CLNs, spleens, and blood were collected frominfected rats at intervals following infection. Lymph nodes and spleens weredissected to remove the fat, cut into several pieces, and pressed through a sterilesieve. Cell suspensions were filtered through a cell strainer (Becton Dickinson,Lincoln Park, N.J.), diluted to 25 ml in DPBS–0.05% BSA (Sigma, St. Louis,Mo.), and centrifuged at 153 3 g for 10 min at 4°C. This step was repeated twice.The cells were resuspended, counted (Coulter Counter ZM; Coulter Electronics,Hialeah, Fla.), centrifuged, and resuspended at 107 cells/ml in RPMI mediumcontaining 5% fetal calf serum, 50 U of penicillin per ml, 50 mg of streptomycinper ml, and 1 mM sodium pyruvate (all from Gibco, Grand Island, N.Y.), as wellas 2 mM L-glutamine and 50 mM 2-mercaptoethanol (Sigma).

ELISPOT. Tissue culture plates (96 well) with nitrocellulose filters (Millipore,Bedford, Mass.) were incubated for 24 h at 4°C with 50 ml of crude antigen (100mg/ml) in DPBS. The filters were washed with DPBS and then blocked for 15 minin DPBS–1% dry milk. The plates were washed twice with DPBS, and 100 ml ofcomplete RPMI medium was added to each well, together with 100 ml of theappropriate cell suspension. Cells were tested at three dilutions. Each dilutionwas tested in duplicate. The plates were left at room temperature for 5 min andthen incubated for 3 h at 37°C in 5% CO2. The filters were washed three timeswith DPBS and incubated with unconjugated mouse anti-rat isotype antibodiesor with HRP-conjugated anti-IgG2c or anti-IgE for 1 h at 4°C with agitation.Filters were washed three times in DPBS–0.05% BSA. For detection of IgG1,IgG2a, IgG2b, IgM, and IgA, filters were next incubated with agitation for 1 h at4°C with goat anti-mouse IgG conjugated to HRP (Southern Biotech, Birming-ham, Al.) diluted in DPBS–0.05% BSA plus 10% normal rat serum. For detec-tion of IgE, the signal was amplified further by an additional incubation with anHRP-conjugated antiperoxidase antibody. The filters were washed three times inDPBS–0.05% Tween 20 (Sigma) and incubated with 4-chloro-naphthol (Sigma)for 30 min at room temperature with agitation.

Dark purple spots, representing ASC, were counted with a Nikon dissectingmicroscope (63) fitted with a Sony DXC-107 video camera attached to a videodisplay. Wells were counted if they contained between 1 and 200 spots per well.Data were expressed as the mean numbers of ASC per tissue 6 1 standarddeviation. Hybridoma cells that produce anti-T. spiralis MAbs of known isotypes(3) were used as positive controls and to confirm the specificity of the assay.

Polyacrylamide gel electrophoresis and Western blotting. Ten percent poly-acrylamide gels with 3% stacking gels in standard format or miniformat (as

noted) were used. Samples were boiled for 3 to 5 min prior to loading in samplepreparation buffer (0.0625 M Tris–2% sodium dodecyl sulfate–10% glycerol [pH6.8]) containing 5% 2-mercaptoethanol. Prestained molecular weight standards(Diversified Biotech, Newton Center, Mass.) were used. Gel electrophoresis andWestern transfer to nitrocellulose were performed according to the methodsdescribed by Burnette (7). Nitrocellulose was blocked with DPBS–0.2% Tween–7.5% dry milk–2% BSA (Sigma). Antibodies were diluted in DPBS–0.2%Tween–2% dry milk–0.25% BSA. The nitrocellulose was washed in DPBS–0.2%Tween. ECL reagent (Amersham Life Science, Little Chalfont, Buckingham-shire, England) was prepared according to the manufacturer’s instructions. Inone experiment, primary antibodies were prepared in the presence of phospho-rylcholine chloride (Sigma).

Passive immunization of rats. Fourteen-day-old AO rats were injected intra-peritoneally with antibody (2.5 mg/20 g of body weight) 90 min prior to oralinfection with 185 larvae. Twelve rats each were treated with serum immuno-globulin from uninfected rats, antityvelose MAb 9E6 (IgG2c), or MAb 6G3(IgG2c). Worms were harvested from intestines on days 1, 8, and 10 afterinfection according to methods that have been described (2).

Statistical analyses. Student’s t test was used to compare means for treatmentgroups. Analysis of variance and Tukey’s least significant difference test wereused to detect differences among three or more means. Differences were con-sidered significant when P was ,0.05.

RESULTS

IgG2c ASC dominated the early immune response. We usedcrude larval antigen in the ELISPOT. Repeated attempts todetect ASC specific for larval excretory/secretory antigens(which are largely tyvelose-bearing glycoproteins) during theintestinal stage of infection were unsuccessful. Thus, antigensrecognized in crude larval homogenates by ASC were distinctfrom those found in larval excretory/secretory antigens andwere not tyvelose-bearing glycoproteins. The presence oftyvelose-bearing proteins in crude antigen was confirmed byELISPOT with specific hybridomas (not shown).

We assayed parasite-specific ASC in three lymphoid tissueson days 8, 16, 30, and 50 postinfection (Fig. 1). The MLN

FIG. 2. Comparison of the kinetics of the T. spiralis-specific ASC responses in the MLNs of euthymic (haired) and athymic (nude) rats. Rats were infected with anoral dose of 500 T. spiralis muscle larvae. The numbers of specific ASC were determined by ELISPOT on days 8 and 16 postinfection. Values expressed are the means 1standard deviations (n 5 three rats).

VOL. 67, 1999 ANTIPHOSPHORYLCHOLINE RESPONSE TO T. SPIRALIS INFECTION 4663

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

dominated the antibody response, generating 50 times theASC found in spleen or CLN. Over 105 T. spiralis-specific ASCwere detected in MLN 8 and 16 days after infection. The peakactivity in the spleen and CLN did not occur until day 16. ASCnumbers in all tissues decreased more than 10-fold betweendays 16 and 30.

IgG2c was the dominant isotype in the MLN, comprising55% of the total ASC on day 8 and 89% of the total ASC onday 16 (Fig. 1A). IgG2b made a large contribution on day 8(41% of the total ASC) but was more variable and declinedquickly. IgA ASC contributed 4.3% of the total ASC on day 16.IgM, IgG1, and IgG2a ASC together comprised less than 4%of the total ASC at any given time. Parasite-specific IgE ASCwere not assayed in this experiment.

The ASC response in the CLN was similar in isotype distri-bution but muted in comparison to that of the MLN (Fig. 1C).IgG2c accounted for 60% of the total ASC on day 16. No ASCwere detected on day 30. On day 50, specific ASC again weredetected, coincident with the onset of the antityvelose responsedescribed previously (1).

The isotype profile of ASC in the spleen was distinct fromthose of the MLN and CLN (Fig. 1B). IgA ASC accounted for64% of the total on day 16. IgA, IgM, and IgG2c ASC declinedfrom day 16 to day 30 and then rebounded on day 50.

The kinetics of the early immune response were confirmedin three additional experiments. In one experiment, parasite-specific IgE ASC were rare but detectable in the MLN (lessthan 0.1% of the total ASC) on day 16. In a third experiment,we detected low numbers of IgE ASC in the CLN 47 dayspostinfection (data not shown).

The ASC response was largely T dependent. Euthymic NIHrats produced an antibody response with an isotype distribu-tion similar to that of the PVG rats, although it was not ofequal magnitude (Fig. 2). Nude rats mounted a weak (less than10% of the haired-rat response) but constant ASC response inthe MLN, with relatively constant numbers of ASC at 8, 15,and 22 days postinfection. The isotypes produced by nude ratswere the same as those produced by euthymic rats: IgG2b,IgG2c, and IgA. The overall level of activity in the spleens ofrats in this experiment was too small to allow any conclusionsto be drawn (data not shown). When euthymic rats were in-fected orally with larvae, they eliminated nearly all of theworms by 15 days postinfection. As expected based on previ-ously reported studies (27, 37), athymic rats failed to eliminatesignificant numbers of adult worms by 22 days postinfection(data not shown), confirming the T-cell dependence of adultworm rejection and showing that the weak B-cell response didnot promote worm rejection.

Isotype profiles of specific antibodies in sera from T. spira-lis-infected PVG rats correlated with ASC numbers. Westernblots of crude larval antigens were probed with pooled serumantibodies collected from uninfected rats, rats infected with1,000 larvae (56 days postinfection), or rats infected with 500larvae (16 days postinfection) (Fig. 3). Antibodies in the seracollected 16 days postinfection bound to at least three proteins(reduced) of 40 to 60 kDa. The difference in binding specific-ities between antibodies collected on days 16 and 56 of infec-tion illustrates the switch from phosphorylcholine antigens totyvelose antigens that takes place in the biphasic antibodyresponse (8). The predominance of IgG1, IgG2a, and IgG2c

FIG. 3. Isotype profile of specific antibodies in sera of PVG rats on days 16 and 56 postinfection (PI). Western-blotted T. spiralis crude antigen was probed withpooled sera from normal rats (NRS), rats infected with 1,000 larvae 56 days prior to bleeding, and rats infected with 500 larvae 16 days prior to bleeding. Antibodybinding was detected with HRP-conjugated antibodies developed in an ECL system, as described in Materials and Methods. Molecular mass standards are indicatedin kilodaltons. Autoradiograph exposure time for each isotype is shown at the bottom of the figure. The small arrow marks migration of the 49-kDa phosphorylcholine-bearing protein. The large arrowhead marks migration of the smallest tyvelose-bearing glycoprotein.

4664 PETERS ET AL. INFECT. IMMUN.

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

among antityvelose antibodies detected on day 56 postinfec-tion corresponds with ELISA data reported previously (3).

The isotype profile of pooled antibodies in sera collected 16days postinfection correlated with ELISPOT data obtainedfrom the same rats. IgG2c was the predominant isotype. IgG2bhad a weak signal, correlating with a variable and volatileIgG2b ASC response. The IgE signal was extremely weak,requiring a 2-h exposure to develop, compared to the 3- to 12-sexposures for the other isotypes. The specificities of the iso-types were similar, but the intensity of binding to differentproteins varied. IgE was distinctive in that it bound to a singleprotein species.

Serum antibodies from infected rats and MAb 6G3 werespecific for phosphorylcholine. Based on indirect evidence,Takahashi has concluded that the antibody response inducedin the first 2 weeks of infection with T. spiralis is specific forphosphorylcholine-bearing proteins (35). In order to confirmthe specificity of the antibodies produced in infected rats, weperformed blocking experiments. Incubation of MAb 6G3 with10 mM phosphorylcholine blocked binding of the antibody toWestern blots of crude larval antigen (Fig. 4), proving thespecificity of the antibody. In contrast, phosphorylcholine com-pletely blocked serum antibody binding to only one of severalproteins, suggesting that other epitopes on these proteins werebeing recognized (Fig. 4). We confirmed the specificity of se-rum antibodies by absorbing rat sera with phosphorylcholine-bearing proteins that were affinity purified from crude antigenby using MAb 6G3. Absorption removed all serum antibodiescapable of binding crude antigens in ELISA (Fig. 5), confirm-ing the phosphorylcholine-bearing antigens to be dominant inthis phase of the antilarval response.

Antiphosphorylcholine MAb 6G3 was not protective. In or-der to test the effects of antiphosphorylcholine antibody onintestinal T. spiralis, we used a passive immunization and chal-lenge protocol in suckling rats. We have shown this system tobe highly sensitive in revealing protective activity of rat anti-bodies against intestinal infection with the parasite (3). Passiveimmunization with MAb 6G3 failed to accelerate expulsion ofintestinal larvae or adult worms from rat pups challenged withT. spiralis (Fig. 6). We used antityvelose antibody 9E6 (IgG2c)as an isotype-matched, positive control in this experiment. As

described previously, passive immunization with 9E6 causedrats to eliminate more than 90% of the challenge larvae within24 h (3). Antibody 9E6 had no effect on the few adult wormsthat developed in passively immunized rats.

DISCUSSION

MLNs are sites of intense B- and T-lymphocyte activity be-tween 8 and 16 days after infection with T. spiralis in rats andmice (13, 22). The increased cellular activity in the MLN iscorrelated with the expulsion of the adult worms. T lympho-cytes from this tissue (13) and from lymph vessels draining the

FIG. 4. Soluble phosphorylcholine (pc) blocked binding of MAbs and poly-clonal antibodies to crude larval antigens. Crude antigens were resolved on 10%acrylamide minigels, blotted, immunostained, and developed with ECL. Phos-phorylcholine was titrated in the presence of the primary antibodies: 6G3 hy-bridoma culture supernatant (diluted 1:40) (20-s exposure) or pooled serumfrom rats infected 16 days prior with 500 larvae (diluted 1:5,000) (1-min expo-sure). Similarly diluted serum from uninfected rats showed no reaction. Molec-ular mass standards are indicated in kilodaltons.

FIG. 5. Serum antibodies in rats infected 16 days with T. spiralis were specificfor phosphorylcholine-bearing antigens. Pooled serum collected from rats 16days postinfection or antibody 6G3 was absorbed with phosphorylcholine-bear-ing antigens or BSA as described elsewhere and then titrated in an ELISA oncrude larval antigens. Absorbances are graphed for pooled serum collected fromrats 16 days postinfection with 500 T. spiralis absorbed with phosphorylcholineantigens (■), absorbed with BSA (F), or unabsorbed (E); for pooled seracollected from uninfected rats (h); and for MAb 6G3 absorbed with phospho-rylcholine antigens (Œ) or BSA (}).

FIG. 6. Comparison of intestinal worm burdens in rat pups that were pas-sively immunized with 6G3 MAb (F), 9E6, tyvelose-specific MAb (E), or serumantibodies from an uninfected rat (h). PVG rats were infected with an oral doseof 185 T. spiralis muscle larvae. The numbers of worms per intestine werecounted on 1, 8, and 10 days postinfection. Values are the means 1 standarddeviations (four or five rats). p, 9E6-treated rats bore significantly fewer wormsthan the two other treatment groups (Tukey’s least significant difference test, P ,0.05).

VOL. 67, 1999 ANTIPHOSPHORYLCHOLINE RESPONSE TO T. SPIRALIS INFECTION 4665

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

intestine (18) are protective in adoptive transfer experiments.We found that the isotype distribution among ASC in lymphnodes draining the intestine correlated well with specific anti-body isotypes in serum, suggesting that in addition to the T-cellresponse, the MLN is playing a major role in the serum anti-body response to infection.

Intestinal nematodes have long been known to induce strongIgE responses in rodents. Although we expected to measurelarge numbers of IgE ASC in our experiments, such cells wererare in the tissues examined from infected PVG rats. We ob-tained similar results from other rat strains, including BrownNorway (data not shown), a strain that is reported to producestrong allergic responses (42). It is possible that IgE ASC mayreside in tissues that we did not evaluate, such as intestinallamina propria. Circulating IgE levels similarly were barelydetectable. Interpretation of our results is aided by the findingsof Ramaswamy and coworkers, which showed that IgE is trans-ported from serum to the gut lumen in T. spiralis-infected rats(29). This transport depletes the serum of circulating IgE (butnot IgG) between 4 and 14 days postinfection, times that cor-relate with our assessments. Thus, we cannot rule out thepossibility that IgE was produced by B cells in tissues we didnot assay or that it was cleared from the circulation so that oursampling methods failed to detect it.

The dominant isotype in the antiphosphorylcholine antigenresponse against T. spiralis was IgG2c. In the rat, polysaccha-ride and hapten-polysaccharide antigens (TI-2 antigens) stim-ulate IgG2c predominantly (9). Our studies of athymic ratsshowed that less than 10% of the early antibody responseagainst T. spiralis was T-lymphocyte independent. This resultis compatible with findings for nude mice, where IgG3 (themouse homologue of rat IgG2c [6, 9]) responses to TI-2 anti-gens are reduced by more than 90% compared to those ineuthymic controls (32). Biological activities unique to ratIgG2c are not known.

Other investigations have shown that anti-Ig antibodies con-jugated to dextran mimic the B-cell-stimulatory activity of TI-2antigens in mice (reviewed by Mond et al. [23]). The influencesof cytokines on this stimulation have been investigated. Anti-Ig–dextran conjugates trigger B-cell proliferation in vitro, andaddition of IL-5 induces these cells to produce IgG3. Further-more, IgG3 production is enhanced 10-fold by the addition ofIFN-g together with IL-5 (33). These findings can be applied tothe analysis of the immune response against T. spiralis in rats.Ramaswamy et al. (30) showed that IFN-g was released intothe intestinal lymph in a cyclical manner, increasing at 35, 75,and 180 h following infection. IL-5 was detected as early as20 h following infection and subsequently was present togetherwith IFN-g. Thus, IL-5 and IFN-g are present in intestinallymphoid tissue at the time that worms are releasing theirphosphorylcholine-bearing antigens in the intestine. These cy-tokines may drive the production of IgG2c against phospho-rylcholine-bearing antigens of T. spiralis.

Phosphorylcholine epitopes in other nematodes have beendescribed to have immunomodulatory effects. A phosphoryl-choline-bearing glycoprotein of the filarial nematode Acantho-cheilonema viteae has been shown to inhibit B-cell activation(14). In addition, phosphorylcholine has been implicated in theinduction of IL-10 secretion by B-1 cells from mice immunizedwith microfilarial lysates of Brugia malayi (26). Although wehave not investigated B-1-cell proliferation or cytokine produc-tion in T. spiralis-infected rats, the magnitude of the responseto phosphorylcholine-bearing antigens suggests that it is likelyto influence the other immune responses induced by the par-asite.

Bacterial organisms incorporate phosphorylcholine into tei-

choic acids (24), lipopolysaccharide (39), protein, and pili (38).Using mice transgenic for an antiphosphorylcholine antibody,Lim and Choy demonstrated that phosphorylcholine is a targetof protective immunity against Streptococcus pneumoniae (20).In these studies, transgenic adult mice were not resistant to T.spiralis infection. Because it was found that suckling rats aremore sensitive to the effects of protective antilarval antibodiesthan adult rats (3, 5), we tested phosphorylcholine-specificantibody for protection in suckling rats. AntiphosphorylcholineIgG2c did not alter the establishment or development of T.spiralis in the intestine. Previously, it was shown that protec-tive, antityvelose IgG2c antibodies bind to tyvelose-bearingsurface and secreted products of L1 larvae (4, 11, 12) and thatsecreted products are the primary targets in protection (21). Incontrast, phosphorylcholine antigens are sequestered in thebody of the larva and are not secreted by this stage (reference25 and unpublished observations). Sequestration would ex-plain both the failure of these antigens to induce an immuneresponse during the muscle stage of infection and also theirfailure to serve as targets in protective intestinal immunity. Itseems plausible, although not proven, that phosphorylcholineantigens would be released during molting. Although phospho-rylcholine antigens have been detected in adult worms (25), wefound that antiphosphorylcholine IgG2c antibody did not ac-celerate expulsion of adult worms from the gut. Thus, phos-phorylcholine antigens are highly immunogenic but do notappear to serve as targets of antibody-mediated, protectiveimmunity against intestinal T. spiralis.

In summary, we have described an IgG2c-dominated anti-body response in the intestines of rats infected with the para-sitic nematode, T. spiralis. This response is directed againstpreviously described phosphorylcholine-bearing glycoproteinsproduced by the first larval stage and is distinct from thewell-characterized antityvelose response which is induced bymuscle-stage larvae. The antiphosphorylcholine response islargely T dependent and, although it is induced in the intestine,does not appear to directly influence the outcome of intestinalinfection. Unique biological activities of rat IgG2c are notknown. The strong representation of IgG2c in two distinctantibody responses directed against T. spiralis larvae supportsa conclusion that this isotype, as well as the cells that produceit, has a significant and unique role in host responses to infec-tion with this nematode.

ACKNOWLEDGMENTS

This work was supported by USPHS grant AI 14490, a grant fromthe Cooperative State Research, Education and Extension Service,U.S. Department of Agriculture, under project no. 473401, and aFogarty Senior International Fellowship (J.A.A.). P.J.P. and A.B.B.were supported by a grant from the Howard Hughes Medical Institute(Cornell Hughes Scholars Program), and E.A.S. was supported by afellowship from the College of Veterinary Medicine, Cornell Univer-sity.

We thank F. Ubeira for helpful discussion and A. Hesser for assis-tance in preparing the manuscript.

REFERENCES

1. Appleton, J. A., and D. D. McGregor. 1987. Characterization of the immunemediator of rapid expulsion of Trichinella spiralis in suckling rats. Immunol-ogy 62:477–484.

2. Appleton, J. A., and D. D. McGregor. 1985. Life-phase specific induction andexpression of rapid expulsion in rats suckling Trichinella spiralis-infecteddams. Immunology 55:225–232.

3. Appleton, J. A., L. R. Schain, and D. D. McGregor. 1988. Rapid expulsion ofTrichinella spiralis in suckling rats: mediation by monoclonal antibodies.Immunology 65:487–492.

4. Appleton, J. A., and L. Usack. 1993. Identification of potential antigenictargets for rapid expulsion of Trichinella spiralis. Mol. Biochem. Parasitol.58:53–62.

4666 PETERS ET AL. INFECT. IMMUN.

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

5. Bell, R. G., J. A. Appleton, D. A. Negrao-Correa, and L. S. Adams. 1992.Rapid expulsion of Trichinella spiralis in adult rats mediated by monoclonalantibodies of distinct IgG isotypes. Immunology 75:520–527.

6. Bruggemann, M., P. Delmastro-Galfre, H. Waldmann, and F. Calabi. 1988.Sequence of a rat immunoglobulin g2c heavy chain constant region cDNA:extensive homology to mouse g3. Eur. J. Immunol. 18:317–319.

7. Burnette, W. N. 1981. Western blotting: electrophoretic transfer of proteinsfrom sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellu-lose and radiographic detection with antibody and radioiodinated protein A.Anal. Biochem. 112:195–203.

8. Denkers, E. Y., D. L. Wassom, C. J. Krco, and C. E. Hayes. 1990. The mouseantibody response to Trichinella spiralis defines a single, immunodominantepitope shared by multiple antigens. J. Immunol. 144:3152–3159.

9. der Balian, G., J. Slack, B. L. Clevinger, H. Bazin, and J. M. Davie. 1980.Subclass restriction of murine antibodies. III. Antigens that stimulate IgG3in mice stimulate IgG2c in rats. J. Exp. Med. 152:209–218.

10. Despommier, D. D. 1983. Biology, p. 75–80. In W. C. Campbell (ed.),Trichinella and trichinosis. Plenum Press, New York, N.Y.

11. Ellis, L. A., C. S. McVay, M. A. Probert, J. Zhang, D. R. Bundle, and J. A.Appleton. 1997. Terminal b-linked tyvelose creates unique epitopes inTrichinella spiralis glycan antigens. Glycobiology 7:383–390.

12. Ellis, L. A., A. J. Reason, H. R. Morris, A. Dell, R. Iglesias, F. M. Ubeira, andJ. A. Appleton. 1994. Glycans as targets for monoclonal antibodies thatprotect rats against Trichinella spiralis. Glycobiology 4:585–592.

13. Grencis, R. K., and D. Wakelin. 1982. Short lived, dividing cells mediateadoptive transfer of immunity to Trichinella spiralis in mice. I. Availability ofcells in primary and secondary infections in relation to cellular changes in themesenteric lymph node. Immunology 46:443–450.

14. Harnett, W., and M. M. Harnett. 1993. Inhibition of murine B cell prolifer-ation and down-regulation of protein kinase C levels by a phosphorylcholine-containing filarial excretory-secretory product. J. Immunol. 151:4829–4837.

15. Hernandez, S., F. Romarıs, I. Acosta, P. N. Gutierrez, and F. M. Ubeira.1995. Ultrastructural colocalization of phosphorylcholine and a phosphoryl-choline-associated epitope in first-stage larvae of Trichinella spiralis. Parasi-tol. Res. 81:643–650.

16. Jungery, M., and B. M. Ogilvie. 1982. Antibody response to stage-specificTrichinella spiralis surface antigens in strong and weak responder mousestrains. J. Immunol. 129:839–843.

17. Kohler, G., and C. Milstein. 1975. Continuous cultures of fused cells secret-ing antibody of predefined specificity. Nature 256:495–497.

18. Korenaga, M., C. H. Wang, R. G. Bell, D. Zhu, and A. Ahmad. 1989.Intestinal immunity to Trichinella spiralis is a property of OX82OX222T-helper cells that are generated in the intestine. Immunology 66:588–594.

19. Lim, P. L., and W. F. Choy. 1990. A thymus-independent (type 1) phospho-rylcholine antigen isolated from Trichinella spiralis protects mice againstpneumococcal infection. Immunology 69:443–448.

20. Lim, P. L., W. F. Choy, S. T. H. Chan, D. T. Leung, and S. S. Ng. 1994.Transgene-encoded antiphosphorylcholine (T151) antibodies protectCBA/N (xid) mice against infection with Streptococcus pneumoniae but notTrichinella spiralis. Infect. Immun. 62:1658–1661.

21. McVay, C. S., A. Tsung, and J. Appleton. 1998. Participation of parasitesurface glycoproteins in antibody-mediated protection of epithelial cellsagainst Trichinella spiralis. Infect. Immun. 66:1941–1945.

22. Molinari, J. A., J. A. Hess, and R. K. Wesley. 1982. Sequential mesentericlymph node histological response following Trichinella spiralis infection. Int.Arch. Allergy Appl. Immunol. 69:81–85.

23. Mond, J. J., A. Lees, and C. M. Snapper. 1995. T cell-independent antigenstype 2. Annu. Rev. Immunol. 13:655–692.

24. Mosser, J. L., and A. Tomasz. 1970. Choline-containing teichoic acid as astructural component of pneumococcal cell wall and its role in sensitivity tolysis by an enzyme. J. Biol. Chem. 245:287–298.

25. Ortega-Pierres, M. G., L. Yepez-Mulia, W. Homan, H. R. Gamble, P. L. Lim,

Y. Takahashi, D. L. Wassom, and J. A. Appleton. 1996. Workshop on adetailed characterization of Trichinella spiralis antigens: a platform for futurestudies on antigens and antibodies to this parasite. Parasite Immunol. 18:273–284.

26. Palanivel, V., C. Posey, A. M. Horauf, W. Solbach, W. F. Piessens, and D. A.Harn. 1996. B-cell outgrowth and ligand-specific production of IL-10 corre-late with Th2 dominance in certain parasitic diseases. Exp. Parasitol. 84:168–177.

27. Perrudet, B. A., A. Y. Boussac, E. J. Ruitenberg, W. Kruizinga, and A.Elgersma. 1983. Protection to challenge with Trichinella spiralis after primaryoral infection in congenitally athymic (nude) mice. J. Parasitol. 69:253–255.

28. Philipp, M., R. M. Parkhouse, and B. M. Ogilvie. 1980. Changing proteins onthe surface of a parasitic nematode. Nature 287:538–540.

29. Ramaswamy, K., J. Hakimi, and R. G. Bell. 1994. Evidence for an interleukin4-inducible immunoglobulin E uptake and transport mechanism in the in-testine. J. Exp. Med. 180:1793–1803.

30. Ramaswamy, K., D. Negrao-Correa, and R. Bell. 1996. Local intestinal im-mune responses to infections with Trichinella spiralis. J. Immunol. 156:4238–4337.

31. Reason, A. J., L. A. Ellis, J. A. Appleton, N. Wisnewski, R. B. Grieve, M.McNeil, D. L. Wassom, H. R. Morris, and A. Dell. 1994. Novel tyvelose-containing tri- and tetra-antennary N-glycans in the immunodominant anti-gens of the intracellular parasite Trichinella spiralis. Glycobiology 4:593–603.

32. Slack, J. H., and J. M. Davie. 1982. Subclass restriction of murine antibodies.V. The IgG plaque-forming cell response to thymus-independent and thy-mus-dependent antigens in athymic and euthymic mice. Cell. Immunol.68:139–145.

33. Snapper, C. M., T. M. McIntyre, R. Mandler, L. M. Pecanha, F. D. Finkel-man, A. Lees, and J. J. Mond. 1992. Induction of IgG3 secretion by inter-feron-gamma: a model for T cell-independent class switching in response toT cell-independent type 2 antigens. J. Exp. Med. 175:1367–1371.

34. Springer, T. A., A. Bhattacharya, J. T. Cardoza, and F. Sanchez-Madrid.1982. Monoclonal antibodies specific for rat IgG1, IgG2a, and IgG2b sub-classes, and kappa chain monotypic and allotypic determinants: reagents foruse with rat monoclonal antibodies. Hybridoma 1:257–273.

35. Takahashi, Y. 1997. Antigens of Trichinella spiralis. Parasitol. Today 13:104–106.

36. Ubeira, F. M., J. Leiro, M. T. Santamarina, T. G. Villa, and M. L. Sanmar-tın-Duran. 1987. Immune response to Trichinella epitopes: the antiphospho-rylcholine plaque-forming cell response during the biological cycle. Parasi-tology 94:543–553.

37. Vos, J. G., E. J. Ruitenberg, N. Van Basten, J. Buys, A. Elgersma, and W.Kruizinga. 1983. The athymic nude rat. IV. Immunocytochemical study todetect T-cells, and immunological and histopathological reactions againstTrichinella spiralis. Parasite Immunol. 5:195–215.

38. Weiser, J. N., J. B. Goldberg, N. Pan, L. Wilson, and M. Virji. 1998. Thephosphorylcholine epitope undergoes phase variation on a 43-kilodaltonprotein in Pseudomonas aeruginosa and on pili of Neisseria meningitidis andNeisseria gonorrhoeae. Infect. Immun. 66:4263–4267.

39. Weiser, J. N., N. Pan, K. L. McGowan, D. Musher, A. Martin, and J.Richards. 1998. Phosphorylcholine on the lipopolysaccharide of Haemophi-lus influenzae contributes to persistence in the respiratory tract and sensitiv-ity to serum killing mediated by C-reactive protein. J. Exp. Med. 187:631–640.

40. Wisnewski, N., M. McNeil, R. B. Grieve, and D. L. Wassom. 1993. Charac-terization of novel fucosyl- and tyvelosyl-containing glycoconjugates fromTrichinella spiralis muscle stage larvae. Mol. Biochem. Parasitol. 61:25–35.

41. Wright, K. D. 1979. Trichinella spiralis: an intracellular parasite in the intes-tinal phase. J. Parasitol. 65:441–445.

42. Yoo, T. J., and C. Y. Kuo. 1980. Cellular and reaginic immune responses toragweed antigen E in inbred rats. Int. Arch. Allergy Appl. Immunol. 61:259–270.

Editor: S. H. E. Kaufmann

VOL. 67, 1999 ANTIPHOSPHORYLCHOLINE RESPONSE TO T. SPIRALIS INFECTION 4667

on March 14, 2020 by guest

http://iai.asm.org/

Dow

nloaded from