Neospora caninum:Role for Immune Cytokines in Host Immunity

Neospora caninum: Role for Immune Cytokines in Host Immunity

IMTIAZ A. KHAN,1 JOSEPHD. SCHWARTZMAN, SUJEEWAFONSEKA, AND LLOYD H. KASPER

Department of Medicine, Department of Pathology, and Department of Microbiology, Dartmouth Medical School,Hanover, New Hampshire 03755, U.S.A.

KHAN, I. A., SCHWARTZMAN, J. D., FONSEKA, S.,AND KASPER, L. H. 1996.Neospora caninum: Rolefor immune cytokines in host immunity.Experimental Parasitology85,24–34.Neospora caninumisa coccidial protozoan parasite that infects a large range of mammals including dogs, cats, mice, andcattle. Morphologically,N. caninumappears indistinguishable fromToxoplasma gondii,althoughthey are genetically distinct. To date there have been no reported cases of this infection in humans,although nonhuman primates may be susceptible to infection. Inbred A/J mice develop no clinical andlittle histologic evidence of infection in spite of a high-dose inoculum ofN. caninum. Splenocytesobtained from infected mice proliferatein vitro in response to bothN. caninumandT. gondii-solubleantigen. A transient state of T cell hyporesponsiveness to parasite antigen and mitogen was observedat Day 7 p.i. This downregulatory response could be partially reversed by the addition of the nitricoxide antagonist LNMMA, but not antibody to IL-10. Mice infected withN. caninumproducesignificant quantities of IL-12 and IFNg, most evident shortly after infection.In vivo, antibody toIL-12 is able to neutralize immune resistance to the parasite. Moreover,in vivo depletion of IFNgwith antibody renders the mice susceptible to infection. These observations suggest thatN. caninuminduces a T cell immune response in the infected host that is at least partially mediated by IL-12 andIFNg. © 1997 Academic Press

INDEX DESCRIPTORS: Neospora caninum; Toxoplasma gondii; Apicomplexa; immunity; IL-12;IL-10; nitric oxide; immune suppression.

INTRODUCTION

Neospora caninumis a recently describedcoccidial parasite that has been reported to in-fect a wide range of mammals. In dogs,N. cani-num causes a severe transplacental acquiredneurologic disease characterized by polyradicu-loneuritis, encephalitis, polymyositis, and as-cending paralysis (Mundayet al. 1990; Bjork-manet al. 1994).Neosporainfection has beenobserved in other domestic animals includingcats (Dubeyet al. 1990) and in livestock(Andersonet al. 1991; Dubeyet al. 1992). Inmice,Neosporais able to induce clinical infec-tion including primary pneumonia, myositis,encephalitis, radiculoneuritis, and pancreatitis(Lindsayet al. 1995). To date there are no re-ported infections by this parasite in humans.Although morphologicallyNeosporaand

Toxoplasmaare quite similar, they are geneti-cally distinct (Marshet al. 1995) and appear tobe divergent (Guo and Johnson 1995). Further,

there is a considerable antigenic difference be-tween these two parasites. For example,Neos-pora lacks both of the major surface proteins,SAG1 (p30) and SAG2 (p22), ofToxoplasma(Brindley et al. 1993). While the host immuneresponse toToxoplasmahas been well studied,little is known regarding the induction of thecellular immune response duringNeosporain-fection. In vitro observations suggest that IFNgmay be involved in host immunity (Inneset al.1995). In the present study, we evaluated themurine immune response to infection withN.caninum. The cellular response to this parasitecan be characterized by the induction of anti-gen-specific T cells and is mediated by the pro-duction of IL-12 and IFNg. Inhibition of IL-12significantly increased mouse susceptibility tothe parasite and caused an increase in IL-10production.

MATERIALS AND METHODS

Mice and parasites.Inbred A/J mice (5- to 6-week-oldfemales) were utilized (Jackson Laboratory, Bar Harbor,ME) for all studies. Mice were infected ip with 1 × 1061 To whom correspondence should be addressed.

EXPERIMENTAL PARASITOLOGY85, 24–34 (1997)ARTICLE NO. PR964110

240014-4894/97 $25.00Copyright © 1997 by Academic PressAll rights of reproduction in any form reserved.

tachyzoites of the NC-1 strain ofN. caninum(kindly pro-vided by Dr. David Lindsey, Auburn University). The PLKstrains ofToxoplasma gondiiwere used in these studies.Both parasites were maintained by serial passage in humanforeskin fibroblasts.N. caninum(NLA) andT. gondii(TLA)antigens were prepared from parasite-infected human fibro-blasts. Parasites were released by forced extrusion througha 27-gauge needle and separated from host cell debris bypassage over a glass wool column. The protein concentra-tions of the TLA and NLA preparations were determined bya commercial assay (Bio-Rad Labs, Cambridge, MA).In vitro T cell proliferation.The splenocytes of infected

mice were isolated and removed from contaminated RBC inlysing buffer. The cells were culturedin vitro at a concen-tration of 2 × 105/well with 5 mg/ml of ConA (Sigma) oreither TLA or NLA (10 mg/ml). After 72 hr, the cultureswere pulsed with [3H]thymidine (0.5mCi/well; Amersham;specific activity, 40–60 mCi/mmole) for an additional 8 hr.Cells were harvested and DNA incorporation was deter-mined by liquid scintillation count.Cytokine analysis.For cytokine mRNA expression, sple-

nocytes were obtained from mice 7 days postinfection un-less stated otherwise. The splenocytes were collected andcytokine mRNA was determined as previously described(Kasperet al.1995). Briefly, total RNA was extracted usingTRIzol (BRL) and reverse transcription performed with ran-dom hexamer primers (Promega, Madison, WI). PCR de-termination of cytokine production was determined utilizingour previously described technique (Kasperet al. 1995).Aliquots of cDNA were assayed for IL-2, IL-10, and IFNg

by examining the ratio of competitor-to-wild-type-band in-tensity following amplification of each primer set. Separa-tion of the PCR products was done by electrophoresis on3% agarose gel.For cytokine protein determination, splenocytes from in-

fected mice were cultured at a concentration of 1 × 106

cells/ml with ConA (5mg/ml), Neosporaor Toxoplasmaantigen (10mg/ml), or medium alone. After 48 hr incuba-tion the culture supernatants were collected by centrifuga-tion and IFNg (Endogen, Cambridge, MA) and IL-10 (Gen-zyme, Cambridge, MA) in the culture supernatants weredetermined by ELISA.Cytokine depletion assays.Rat anti-mouse IL-10 (Endogen,

Cambridge, MA) was used at a concentration of 40mg/ml.Control antibody was isotype-specific rat IgG (Sigma). Forinvivo IL-12 depletion, mice were treated beginning 1 day priorto infection with goat anti-mouse IL-12 (0.5 mg) (kindly pro-vided by Dr. Maurice Gately, Hoffman LaRoche, Nutley, NJ)and twice weekly thereafter. Control mice were treated with anequal quantity of goat IgG (Sigma, St. Louis). IFNg was neu-tralized by using 3 mg/mouse of rat XMG6 (IgG1) mAb ad-ministered by ip inoculation 1 day prior to parasite infectionand weekly thereafter if necessary. This antibody had beentitrated in a viral neutralization assay in which 1.23 ng ofantibody was required for 50% inhibition of activity of 1 unitIFNg. Control mice were treated with an equal quantity of ratIgG (Sigma, St. Louis).

Microscopy.Brain, heart, lung, liver, spleen, and pan-creas of infected, IL-12 depleted, and of sham treated con-trol mice were fixed in 10% buffered formalin for histologyon Day 7 postinfection. Tissues were embedded in paraffinand 5-mm-thick sections were cut and stained with hema-toxylin and eosin. Sections were examined and photo-graphed with Kodak T Max 100 ASA film using an Olym-pus Van Ox microscope.Statistical analysis.Statistical analysis was performed by

using one-way analysis of variance.

RESULTS

Resistance toN. caninuminfection in micewas determined. Three inbred strains of mice(C57BL/6, A/J, Balb/c) and one outbred strain(CD-1) were infected with increasing numbersof N. caninumtachyzoites. No evidence of clini-cal infection (ruffled fur, weight loss, huddling)was evident in any of the infected groups (up to1 × 106 tachyzoites). Although there was noclinical evidence of infection, review of the his-tological sections of brain revealed rare inflam-matory nodules with a few single zoites identi-fiable (Fig. 6F). All other organs were entirelynormal with no evidence of infection or inflam-mation.Splenocytes fromNeospora-infected mice

were assayed for proliferation in response toparasite antigen. Anin vitro proliferation assaywas performed to determine whether mice in-fected with Neosporadevelop cell-mediatedimmunity to the parasite. For this study, micewere infected withNeosporaand their spleno-cytes stimulatedin vitro with eitherToxoplasmaor Neospora lysate at different times post-infection. As shown in Fig. 1, splenocytes fromNeospora-infected mice proliferate when stimu-lated with eitherNeosporaor Toxoplasmaly-sate (Day 10 p.i.). This lymphoproliferative re-sponse was apparent through at least Day 14,when the experiment was terminated. On Day 7postinfection the mitogenic response to ConAwas dramatically reduced when compared tothat of either uninfected mice (P < 0.001) ormice that had been infected and evaluated atDay 3, 10, or 14 postinfection. These observa-tions demonstrate that infected mice developcell-mediated immunity toNeospora. However,this cellular response does not appear to be

N. caninum:CELLULAR IMMUNITY IN MICE 25

parasite specific since the splenocytes fromthese mice responded to bothNeosporaandToxoplasmaparasite preparations.The cytokine-mediated response toNeospora

infection in mice was evaluated. A quantitativePCR was performed to determine the level ofcytokine message expression in mice followingNeosporainfection. Splenocytes from infectedmice were collected either at 6-hr intervals (Fig.2A) or at Days 3, 7, 10, 14, and 18 days postin-fection (Fig. 2B). The mRNA expression forIFNg, IL-2, IL-10, and IL-12 was determinedand the quantitative response determined. Asshown in Fig. 2A, message for IL-12 increaseddramatically during the first few hours postin-fection. By 6 hr p.i., message for this cytokinewas almost 200-fold over that of the control.This message rapidly declined by 24 hr p.i. andremained stable thereafter (Fig. 2B). Also notedwas a dramatic rise in the production of IFNgmessage during the first 6 hr p.i. The level ofmessage for IFNg declined by 24 hr p.i. After24 hr, message for IFNg increased through Day10 (20-fold over that of the control) (Fig. 2B). Agradual decline in message for IFNg was ob-served after Day 10. The IL-2 message re-mained stable throughout the study. There was a

5-fold increase in message for IL-10 by Day 10p.i. Thereafter message for this cytokineshowed a nominal decline, but remained abovebaseline for the remainder of the time pointsinvestigated.Splenocytes from mice infected withNeo-

sporawere assayed for IFNg and IL-10 proteinlevels. In this experiment, splenocytes from in-fected mice were obtained at various time pointsp.i., culturedin vitro for 72 hr in the presence ofeither mitogen or parasite antigen, and the su-pernatant was collected for cytokine determina-tion. As shown in Fig. 3A, murine infectionwith N. caninumstimulates the production ofIFNg in the absence of either mitogen or para-site antigen on all days assayed p.i.In vitroexposure of splenocytes from infected mice toeither mitogen or parasite antigen further en-hances this response. Production of IL-10 re-mained near baseline for the first week p.i. ByDay 14 p.i. there was a demonstrable increase inthe production of IL-10 to both mitogen andparasite antigen stimulation (Fig. 3B).The role of nitric oxide (NO) in regulating

immune suppression was determined. To iden-tify the possible factor(s) responsible for theimmunosuppression observed at Day 7 p.i. (Fig.

FIG. 1. Proliferative response of splenocytes fromN. caninum-infected mice to mitogen and parasite antigens.Spleen cells (2 × 105) from uninfected or Day 3, 7, 10, and 14 infected mice were cultured in the presence ofeither 5mg/ml of ConA or 10mg of NLA or TLA. The proliferation was measured by [3H]thymidine incor-poration. The assay was performed on pooled cells from two mice and is represented as the mean of quadru-plicate wells ± SD. The data are representative of two separate experiments.

KHAN ET AL.26

1), splenocytes were stimulated with ConA andcultured in the presence of either anti-IL-10 orthe known nitr ic oxide antagonist,N-monomethylL-arginine (LNMMA). As shownin Fig. 4, addition of LNMMA to the culturesupernatant reversed the downregulation of

both the mitogen- and the parasite antigen-mediated proliferative response by 50% (P <0.01). Culture supernatants obtained from sple-nocytes of infected mice were assayed at vari-ous time points p.i. for the production of ni-trites. As shown in Fig. 3C, nitrite production

FIG. 2. Cytokine mRNA expression in mice infected withN. caninum. Mice were infected with 1 × 106

tachyzoites ofN. caninum. The splenocytes were harvested at various time points starting at 3 hr postinfectionuntil 72 hr p.i. (A) or 4 days p.i. to Day 7 p.i. (B). The spleen cells from three mice were pooled for each timepoint and mRNA expression for IL-12, IFNg, and IL-10 was assayed using RT–PCR. The differences in thetranscriptional level for all the genes are expressed relative to the uninfected mice (assigned as 1). The cDNAconcentration examined at each time point was standardized to the HPRT mRNA level (not shown).


was greatest in the culture supernatant at 7 daysp.i. The splenocyte supernatants were alsoevaluated for the effect of IL-10 at Day 7 p.i. Asanticipated, there was only borderline reversalobserved with antibody to IL-10 in both themitogen- and the parasite antigen-stimulatedconditions. Increasing the anti-IL-10 antibody

concentration had no effect (not shown). Anti-body reversal for IL-10 was not performed witheither Day 10 or Day 14 p.i. culture supernatantsince there appeared to be no inhibition of lym-phoproliferation on those days (Fig. 1).The role of IL-12 in host immunity toNeo-

spora was evaluated. The marked increase in

FIG. 3. Level of IFNg, IL-10 protein, and nitrite production by splenocytes from N. caninum-infected mice.Mouse splenocytes (1 × 106) were isolated from Day 3, 7, and 14 p.i. mice and stimulated with ConA, NLA, orTLA. After a 72-hr incubation the supernatants were collected, centrifuged, and stored at −70°C. The IFNg (A)and IL-10 (B) levels in the culture supernatants were determined by ELISA, and the level of nitrite (C) wasdetermined by the Griess reaction.

KHAN ET AL.28

IL-12 at the onset of infection suggested thatIL-12 may be an essential cytokine in the hostimmune response. To determine the importanceof this cytokine, mice were infected withNeo-sporaand treated daily with antibody to IL-12twice a week starting on the day of infection. Asshown in Fig. 5A, IL-12 depletion severely re-versed natural immunity to this parasite. Allmice treated with antibody died by Day 8postinfection whereas all of the mice treatedwith control IgG survived beyond 3 months p.i.The importance of IFNg in protective immunityagainstN. caninumwas evaluatedin vivo.An-tibody to IFNg was administered to mice 1 dayprior to infection. As shown in Fig. 5B, micetreated with anti-IFNg were rendered suscep-tible to infection whereas mice treated with theisotype control antibody remained resistant.These observations indicate that both IL-12 andIFNg are important in the innate response to thisparasite.Histologic analysis of tissues from these mice

demonstrate evidence of overwhelming diffuse

infection in several major organs. As shown inFig. 6, there is a marked proliferation of zoites,both singly and in clusters, throughout the liverand spleen. Within the liver extensive areas ofnecrosis were seen, with little evidence of in-

FIG. 4. Reversal ofN. caninum-mediated immunosup-pression by a NO synthetase antagonist.Pooled splenocytes(2 × 105) from Day 7 p.i. mice (n4 2) were cultured in thepresence of ConA (5mg/ml), NLA (10mg/ml), or TLA (10mg/ml). The cultures were treated with 40mg/ml of ratanti-murine IL-10 antibody, control rat IgG, or 0.5 mMLNMMA. The lymphoproliferation was assayed by[3H]thymidine incorporation. The data are represented asthe mean ± SD of quadruplicate wells and are representativeof one of two experiments.

FIG. 5. Neutralization of IL-12 and IFNg in N. caninum-infected mice. (A) A/J mice (n 4 12) were infected intra-peritoneally with 1 × 106 tachyzoites ofN. caninumandtreated with either 0.5 mg of goat anti mouse IL-12 twiceweekly beginning at the time of infection or an equal vol-ume of goat IgG. The mice were observed daily until Day180 p.i., when the experiment was terminated. (B) A/J mice(n 4 12) were treated with 3 mg/mouse antibody to IFNg

1 day prior to infection with 1 × 106 tachyzoites ofN.caninum. Control mice received an equal amount of rat IgG.The data are representative of one of two experiments.


FIG. 6. Histologic effect of IL-12 depletion duringN. caninuminfection in mice. Photomicrographs of tissuesfrom IL-12-depleted mice on Day 7 after infection withN. caninum(A–E) and untreated control mouse brain(F). (A) A large cyst-like cluster of zoites in liver with marked fatty change of hepatocytes. (B) Small clustersof zoites (arrowheads) at the edge of an extensive area of coagulate necrosis (arrows). (C) A large cluster ofzoites at the junction of red and white pulp of spleen (arrowheads). There are small foci of inflammatory infiltratein the white pulp (arrows). (D) A cluster of zoites at the edge of a pancreatic acinus (arrow). (E) Clusters andsingle zoites (arrows) in the cerebral cortex. (F) An inflammatory nodule in the cerebral cortex of a controlmouse; single zoites are seen (arrows). Bars represent 100mm.

KHAN ET AL.30

flammation (Figs. 6A and 6B). The spleen con-tained numerous proliferating zoites, with onlysmall foci of inflammatory cells (Fig. 6C). Thepancreas showed occasional clusters of zoites,but little evidence of inflammation (Fig. 6D).Within the brain, numerous foci of proliferatingzoites, some with areas of necrosis but no in-flammatory infiltrate, were scattered throughoutthe gray matter (Fig. 6E), in contrast to the un-treated control, where marked inflammatory in-filtration was seen in the rare lesions (Fig. 6F).Zoites could not be demonstrated in the heartand lungs, and these organs were histologicallynormal in IL-12-depleted mice.An analysis of the IL-10 response following

IL-12 depletion was then performed. In thisstudy, splenocytes from infected mice depletedof IL-12 were isolated at Day 7 p.i. These miceshowed clinical evidence of infection (ruffledfur, huddling) at the time of splenocyte harvest.The isolated splenocytes were analyzed for pro-duction of IL-10 by quantitative PCR. Asshown in Fig. 7, there was a 200-fold increase inmessage for IL-10 in the splenocytes of 7-day-p.i. mice.

DISCUSSION

We demonstrate thatN. caninum, a coccidianparasite that is morphologically similar toT.gondii, is able to stimulate a T cell response inan experimental murine model. This immuneresponse is characterized by a proliferative Tcell response and the production of immune cy-tokines postinfection. In particular, IL-12 ap-pears to play an essential role in host immunityto this organism.N. caninumwas first isolated from infected

dogs (Mundayet al. 1990). More recently thiscoccidian parasite has been shown to infect awide range of mammals including cats, al-though cats do not appear to be an essentialcomponent of its life cycle. In mice,N. caninuminfection has been associated with acute pri-mary pneumonia, myositis, encephalitis, gan-glioradiculoneuritis, and pancreatitis (Lindsayand Dubey 1990). The development of infectionin mice appears to be genetically restricted inthat some mouse strains are more susceptible to

infection of the CNS (Lindsayet al. 1995). Inour study, we utilized inbred A/J mice. We ob-served nominal infection in these mice when theparasite was administered by intraperitoneal in-oculation. Infection was limited to a few scat-tered foci of cysts containing bradyzoite inliver, pancreas, and brain. Mice infected withN.caninumwere resistant to very high doses ofparasite (up to 1 × 106). This inoculum is manytimes greater than that observed for certainstrains ofT. gondii(Type I clones), such as RH,

FIG. 7. IL-10 mRNA expression inN. caninum-infectedmice depleted of IL-12. Mice were infected with 1 × 106

tachyzoites ofN. caninumand depleted of IL-12 by anti-body treatment as mentioned above. Splenocytes from in-fected mice treated with anti-IL-12 antibody, goat IgG, anduninfected controls were harvested on Day 7 p.i. mRNAexpression for IL-10 was measured by RT–PCR. The dif-ferences in the transcriptional levels are relative to those ofthe spleen cells from uninfected animals. The cDNA con-centration was standardized to the HPRT mRNA level.Pooled splenocytes from three mice were used for eachcondition.


which are highly lethal at very small inoculumdoses (Howe and Sibley 1995).In spite of this large inoculum of parasites

mice remained clinically resistant to infectionwith N. caninum. Splenocytes from these miceproliferated in response to a parasite lysate. Aresponse of similar magnitude was also ob-served by these splenocytes to stimulation witha lysate ofT. gondii. This proliferative responseto both parasite preparations would suggest thatthere are crossreactive immune determinants onthese two parasites. Earlier studies have shownthat these parasites are antigenically distinct(Brindley et al. 1993). Evaluation of theN.caninumin our laboratory using IFA confirmedthe absence of the majorT. gondii surface an-tigen, P30 (SAG1), on their surface (data notshown). However, evaluation by Western blotanalysis of the surface antigens of these twoparasites does identify antigens of similar mo-lecular weights (manuscript in preparation).In addition to the proliferative response of the

splenocytes, we observed transient lymphocytehyporesponsiveness at Day 7 postinfection.This transient period is characterized by a lowproliferative index to both mitogen and parasiteantigen. Lymphocyte hyporesponsiveness atDay 7 postinfection has been observed duringacuteToxoplasmainfection in mice (Haqueetal. 1994, 1995; Khanet al. 1995). Duringmouse infection withT. gondii the period oflymphocyte hyporesponsiveness persiststhrough Day 14 postinfection for several differ-ent strains of parasite including Type I (viru-lent) and Type II (avirulent) strains (Howe andSibley 1995). In contrast, the lymphocyte pro-liferative response toN. caninumis restored byDay 10 p.i. This would suggest that althoughNeosporainfection is able to elicit a downregu-latory response on host immunity the effect ismore restricted, at least for the strain ofN. cani-numused in these studies.The principal mechanism for this downregu-

latory response appears to be the induction ofnitric oxide. Nitric oxide has been demonstratedto be an important immunoregulatory moleculein various infections includingPlasmodium(Rockettet al. 1994), Listeria monocytogenes

(Gregoryet al.1993), andTrypanosome bruceii(Sternberg and McGuigan 1992). The effect ofthis reactive intermediate during acuteToxo-plasmainfection is well recognized (Candolfietal. 1995; Haqueet al. 1994; Khanet al. 1995;Gazzinelliet al.1994a). The treatment of spleencell cultures with LNMMA, an inhibitor of ni-tric oxide synthetase, restores 50% of the sple-nocyte proliferative response from Day-7-p.i.mice. This reversal of the transient lymphocytehyporesponsive state was observed under bothmitogen- and antigen-stimulated conditions.IL-10 does not appear to be a required modu-

lator of the hyporesponsive state observed atDay 7 p.i. Our data demonstrate that messagefor this cytokine is near above baseline at thistime corresponding to the low production of cy-tokine product in response to both mitogen andparasite antigen stimulation. This is further con-firmed by inhibition with antibody to IL-10,which failed to restore the proliferative re-sponse. This observation differs from thatwhich has been reported during acuteToxo-plasmainfection (Khanet al. 1995; Gazzinelliet al.1994b). In those studies, antibody to IL-10was able to reverse the downregulatory re-sponse by almost 50%. In combination with aNO inhibitor (LNMMA), the downregulationby T. gondiiwas reversed by over 80%.The principal mechanism for murine protec-

tion against this parasite appears to be via in-duction of IL-12 and IFNg. These two cyto-kines have been demonstrated to be essential forhost immunity to a number of infectious agents,in particular,T. gondii (Gazzinelliet al. 1993;Khanet al. 1994; Hunteret al. 1995). The ob-servations reported herein suggest that in miceendogenous IL-12 is important for host protec-tion against this organism. Administration ofantibody to IL-12 reversed natural protectionsuch that all mice succumbed to challenge by 7days p.i. These mice were found to have foci ofcysts containing bradyzoites in all organs ana-lyzed. Interestingly, there was minimal inflam-mation evident in these infected foci, suggestingthat the antibody to IL-12 inhibited the hostinflammatory response to the parasite. Mice ad-ministered a control antibody were essentially

KHAN ET AL.32

disease free. Mice that were treated with anti-IL-12 antibody demonstrated a significant risein message expression of the downregulatorycytokine IL-10. Since all mice treated with an-tibody to IL-12 succumbed to infection by Day7, it is possible that this increase in IL-10 ex-pression was responsible for the observed lackof inflammation in the infected foci of thesemice. Thein vivo IFNg neutralization studiesindicate that this cytokine is an essential com-ponent of the host response to this parasite. Thisalso is consistent with the observations made bya number of laboratories that it is a principalmediator of host protection againstToxoplasma.Mice that receive a neutralizing dose of anti-body to this cytokine are rendered susceptible toinfection. It is presumed, as withToxoplasma,that IL-12 is the principal cytokine that stimu-lates the production of IFNg in Neospora-infected mice. Neutralization of either of thesecytokines results in the loss of innate resistanceto the parasite.The mechanism of host protection againstN.

caninumappears to be similar in many respectsto the immune response elicited byT. gondii.Although these two parasites are geneticallydistinct and express different surface antigens,there does appear to be crossreactive immunitybetween the organisms. This crossreactivity canbe most easily explained by antigenic overlapbetween immune epitopes. Interestingly, bothof these parasites stimulate the production ofIL-12 in the murine host. Reversal of this cyto-kine during either infection is uniformly fatal.Acute T. gondii infection, however, is uni-formly fatal to almost all inbred mouse strains,whereasN. caninuminfection in those inbredstrains evaluated to date is nonfatal even whenthe mice have been infected with enormousnumbers of parasites. It is possible that induc-tion of the downregulatory cytokine IL-10 isresponsible in part for the loss of protectionfollowing depletion of IL-12. Further studiesexploring the immunologic overlap betweenthese two divergent parasites are currently un-derway and may provide a more comprehensiveunderstanding of the immune mechanisms acti-vated in response to infection withT. gondii.

ACKNOWLEDGMENTS

The authors thank Dr. Tadashi Matsuura for his helpduring this study. We thank Dr. David Lindsay, AuburnUniversity, Alabama, and Dr. J. P. Dubey, USDA, for help-ful discussions.

REFERENCES

ANDERSON, M. L., BLANCHARD, P. C., BARR, B. C., DUBEY,J. P., HOFFMAN, R. L., AND CONRAD, P. A. 1991. Neo-spora-like protozoan infection as a major cause of abor-tion in California dairy cattle.Journal of the AmericanVeterinary Association198,241–244.

BJORKMAN, C. A., LUNDEN, J., HOLMDAHL , J., BARBER, A. J.,TREES, E. Y., AND UGGLA, A. 1994.Neospora caninumindogs: Detection of antibodies by ELISA using an iscomantigen.Parasite Immunology16, 643–648.

BRINDLEY, P. J., GAZZINELLI , R. T., DENKERS, S. W., DAVIS,J. P., DUBEY, R., BELFORT, M. C., JR., MARTINS, C., SIL-VEIRA, L., JAMRA, A. P., WATERS, ET AL. 1993. Differen-tiation of Toxoplasma gondiifrom closely related coc-cidia by riboprint analysis and a surface antigen genepolymerase chain reaction.American Journal of TropicalMedicine and Hygiene48, 447–456.

CANDOLFI, E., HUNTER, C. A., AND REMINGTON, J. S. 1995.Roles of gamma interferon and other cytokines in sup-pression of the spleen cell proliferative response to con-canavalin A and toxoplasma antigen during acute toxo-plasmosis.Infection and Immunity63, 751–756.

DUBEY, J. P., LINDSAY, D. S., ANDERSON, M. L., DAVIS,S. W., AND SHEN, S. K. 1992. Induced transplacentaltransmission ofNeospora caninumin cattle.Journal ofthe American Veterinary Medical Association201,709–713.

DUBEY, J. P., LINDSAY, D. S., AND LIPSCOMB, T. P. 1990.Neosporosis in cats.Veterinary Pathology27, 335–339.

GAZZINELLI , R. T., HAYASHI, S., WYSOCKA, M., CARRERA,L., KUHN, R., MULLER, W., ROBERGE, F., TRINCHIERI, G.,AND SHER, A. 1994a. Role of IL-12 in the initiation of cellmediated immunity byToxoplasma gondiiand its regu-lation by IL-10 and nitric oxide.Journal of EukaryoticMicrobiology41, 9S.

GAZZINELLI , R. T., HIENY, S., WYNN, T. A., WOLF, S., ANDSHER, A. 1993. Interleukin 12 is required for the T-lymphocyte-independent induction of interferon gammaby an intracellular parasite and induces resistance in T-cell-deficient hosts.Proceedings of the National Acad-emy of Science USA90, 6115–6119.

GAZZINELLI , R. T., WYSOCKA, M., HAYASHI, S., DENKERS,E. Y., HIENY, S., CASPAR, P., TRINCHIERI, G., AND SHER,A. 1994b. Parasite-induced IL-12 stimulates early IFN-gamma synthesis and resistance during acute infectionwith Toxoplasma gondii. Journal of Immunology153,2533–2543.

GREGORY, S. H., WING, E. J., HOFFMAN, R. A., AND SHER,R. L. 1993. Reactive nitrogen intermediates suppress the


primary immunologic response to Listeria.Journal of Im-munology150,2901–2908.

GUO, Z. G.,AND JOHNSON, A. M. 1995. Genetic comparisonof Neospora caninumwith Toxoplasma and Sarcocystisby random amplified polymorphic DNA-polymerasechain reaction.Parasitology Research81, 365–370.

HAQUE, S., HAQUE, A., AND KASPER, L. H. 1995. AToxo-plasma gondii-derived factor(s) stimulates immunedownregulation: An in vitro model.Infection and Immu-nity 63, 3442–3447.

HAQUE, S., KHAN, I. A., HAQUE, A., AND KASPER, L. H.1994. Impairment of the cellular immune response inacute murine toxoplasmosis: Regulation of interleukin 2production and macrophage-mediated inhibitory effects.Infection and Immunity62, 2908–2916.

HOWE, D. K., AND SIBLEY, L. D. 1995.Toxoplasma gondiicomprises three clonal lineages: Correlation of parasitegenotype with human disease.Journal of Infectious Dis-eases172,1561–1566.

HUNTER, C. A., CANDOLFI, E., SUBAUSTE, C., VAN CLEAVE,V., AND REMINGTON, J. S. 1995. Studies on the role ofinterleukin-12 in acute murine toxoplasmosis.Immunol-ogy84, 16–20.

INNES, E. A., PANTON, W. R., MARKS, J., TREES, A. J.,HOLMDAHL , J.,AND BUXTON, D. 1995. Interferon gammainhibits the intracellular multiplication ofNeospora cani-num, as shown by incorporation of [3H]uracil. Journal ofComparative Pathology113,95–100.

KASPER, L. H., MATSUURA, T., AND KHAN, I. A. 1995. IL-7stimulates protective immunity in mice against the intra-cellular pathogen,Toxoplasma gondii. Journal of Immu-nology155,4798–4804.

KHAN, I. A., MATSUURA, T., AND KASPER, L. H. 1994. In-

terleukin-12 enhances murine survival against acute toxo-plasmosis.Infection and Immunity62, 1639–1642.

KHAN, I. A., MATSUURA, T., AND KASPER, L. H. 1995. IL-10mediates immunosuppression following primary infec-tion with Toxoplasma gondiiin mice.Parasite Immunol-ogy17, 185–195.

LINDSAY, D. S.,AND DUBEY, J. P. 1990. Infections in micewith tachyzoites and bradyzoites of Neospora caninum(Protozoa: Apicomplexa).Journal of Parasitology76,410–413.

LINDSAY, D. S., LENZ, S. D., COLE, R. A., DUBEY, J. P.,ANDBLAGBURN, B. L. 1995. Mouse model for central nervoussystemNeospora caninuminfections.Journal of Parasi-tology81, 313–315.

MARSH, A. E., BARR, B. C., SVERLOW, K., HO, M., DUBEY,J. P., AND CONRAD, P. A. 1995. Sequence analysis andcomparison of ribosomal DNA from bovine Neospora tosimilar coccidial parasites.Journal of Parasitology81,530–535.

MUNDAY, B. L., DUBEY, J. P., AND MASON, R. W. 1990.Neospora caninuminfection in dogs.Austrian VeterinaryJournal67, 76–77.

ROCKETT, K. A., AUBURN, M. M., ROCKETT, A. J., COWDEN,W. B., AND CLARK, I. A. 1994. Possible role of nitricoxide in malarial immunosuppression.Parasite Immunol-ogy16, 243–249.

STERNBERG, J., AND MCGUIGAN, F. 1992. Nitric oxide me-diates suppression of T cell responses in murineTrypana-soma bruceiinfection.European Journal of Immunology22, 2741–2748.

Received 14 June 1996; accepted with revision 13 Septem-ber 1996

KHAN ET AL.34

Documents

Neospora caninum:Role for Immune Cytokines in Host Immunity