Protective Immunity against Congenital Toxoplasmosis with ...

2 downloads 0 Views 86KB Size Report
ALEX BOLLEN,1 RALPH BIEMANS,2. AND ALAIN ... B-6041 Gosselies,1 and SmithKline Beecham Biologicals, B-1330 Rixensart,2 Belgium ..... Guy, E. C., H. Pelloux, H. Lappalainen, H. Aspock, A. Hassl, K. K. Melby, .... Editor: W. A. Petri, Jr.
INFECTION AND IMMUNITY, Sept. 2000, p. 4948–4953 0019-9567/00/$04.00⫹0 Copyright © 2000, American Society for Microbiology. All Rights Reserved.

Vol. 68, No. 9

Protective Immunity against Congenital Toxoplasmosis with Recombinant SAG1 Protein in a Guinea Pig Model ` LE HAUMONT,1 LOUIS DELHAYE,1 LIDA GARCIA,1 MARGARITA JURADO,1 MICHE ´ RONIQUE DAMINET,1 VINCENT VERLANT,2 PASQUALINA MAZZU,1 VE 1 ALEX BOLLEN, RALPH BIEMANS,2 AND ALAIN JACQUET1* Department of Applied Genetics, Institut de Biologie et de Me´decine Mole´culaires, Universite´ Libre de Bruxelles, B-6041 Gosselies,1 and SmithKline Beecham Biologicals, B-1330 Rixensart,2 Belgium Received 13 April 2000/Returned for modification 30 May 2000/Accepted 7 June 2000

Primary infection with Toxoplasma gondii during pregnancy can induce fetal pathology and abortion in both humans and animals. The present study describes the development of an experimental model of congenital toxoplasmosis in the guinea pig. In this animal model, we evaluated the protective effect of vaccination with a recombinant form of SAG1 against maternofetal transmission of tachyzoites. The presence of parasites in fetuses was determined by nested PCRs and by an in vivo readout after fetal brain homogenate injections in mice. The absence of parasites was demonstrated in 66 to 86% of fetuses derived from adult guinea pigs immunized with SAG1 and challenged with the mildly virulent T. gondii strain C56. In contrast, more than 80% of fetuses from mock-immunized guinea pigs were infected. The protection was not correlated with titers of antibody to SAG1. Our results indicated that this experimental model constitutes a relevant model for evaluation of vaccine candidates against congenital toxoplasmosis and that SAG1 elicits significant protection against maternofetal transmission. Escherichia coli (27) or Pichia pastoris (2), or with SAG1derived peptides (7, 33) demonstrated the development of significant protection in animal models against lethal challenge. Moreover, a recent study showed that nucleic acid vaccination with plasmids encoding SAG1 induced protection against T. gondii infection in mice (1, 26). The development of a suitable laboratory model is essential for evaluation of the efficacy of the different recombinant subunit vaccine candidates against congenital toxoplasmosis. Different animal models of congenital toxoplasmosis have previously been developed in mice, rats, and sheep (5, 11, 30, 37, 38). The present report describes the development of an experimental model of congenital toxoplasmosis in guinea pigs, for which maternofetal transmission is very similar to that observed in humans (36). To validate this animal model, we evaluated protection against maternofetal transmission by vaccination before pregnancy with a recombinant SAG1 expressed in P. pastoris. We previously showed that adjuvanted anchorless SAG1 was able to induce lymphoproliferation and to prolong the survival of mice infected with live tachyzoites (2). Our data showed that guinea pig infection constitutes a relevant model for congenital toxoplasmosis and that SAG1 elicits significant protection against maternofetal transmission.

The intracellular protozoan Toxoplasma gondii is a widespread opportunistic parasite of humans and animals (10). Transmission from animals to humans occurs mainly through oocysts excreted in the feces of infected cats and meat products from farm animals contaminated with viable tissue cysts. Although infection in immunocompetent humans is usually asymptomatic, toxoplasmosis may cause severe complications in immunocompromised individuals (16). In AIDS patients, in particular, the recrudescence of latent T. gondii infection has induced encephalitis (24). Primary infection during pregnancy can result in neonatal death or in severe congenital defects like hydrocephalus, mental retardation, and retinochoroiditis, which may occur at birth or during development (35). At the veterinary level, toxoplasmosis is also one of the main causes of infectious reproductive wastage in many countries, causing fetal resorption, abortion, stillbirth, and neonatal mortality in sheep, pigs, and goats (10). An effective vaccine should protect against both acute and chronic infection. In humans, this vaccine could be valuable for preventing fetal infection as well as reactivation in immunocompromised individuals. In farm animals, it could prevent spontaneous abortion, thus decreasing economic losses, as well as reducing a major epidemiologic vector for human infection. A live attenuated vaccine, lacking the ability to produce tissue cysts, has been available for sheep (4). However, it was shown that this vaccine induced side effects, and protection lasted no more than 3 years. As this vaccine might revert to a pathogenic strain, it constitutes a poor vaccine candidate for humans. Development of a subunit vaccine against T. gondii has focused mainly on SAG1, the major immunodominant surface antigen of invasive tachyzoites (22). Vaccination with purified natural SAG1 (3, 8, 21), with recombinant SAG1 produced by

MATERIALS AND METHODS Animals. Female and Male Dunkin-Hartley guinea pigs were purchased from Harlan (Zeist, The Netherlands). Guinea pigs were housed at Harlan’s facilities during vaccination and mating. Afterwards, animals were transferred to our animal facilities. Female BALB/c mice (6 weeks old) were obtained from IffaCredo Laboratories (Brussels, Belgium). Parasites. T. gondii C56, a mildly virulent strain (kindly supplied by M. L. Darde, Centre Hospitalier Universitaire [CHU], Limoges, France) maintained in BALB/c mice by intraperitoneal inoculation of brain tissue cysts, was used for experimental infections of guinea pigs. To develop chronic toxoplasmosis in mice, sulfadiazine at 400 mg/ml (Sigma) was added to drinking water for 19 days after intraperitoneal infection with 104 tachyzoites. Chronic Toxoplasma infection appeared 30 days after parasite inoculation. To obtain fresh tachyzoites, a chronically infected mouse was sacrificed and the brain was homogenized with 1 ml of phosphate-buffered saline (PBS) and injected intraperitoneally into one to three mice. After 5 to 7 days, mice were

* Corresponding author. Mailing address: Rue des Professeurs Jeener et Brachet, 12, B-6041 Gosselies, Belgium. Phone: 32 2 650 99 09. Fax: 32 2 650 99 00. E-mail: [email protected]. 4948

VOL. 68, 2000

VACCINE AGAINST CONGENITAL TOXOPLASMOSIS

killed and parasites were recovered by peritoneal lavage with 10 ml of PBS. The peritoneal lavage was forced through a 27-gauge needle and purified by filtration through 3-␮m-pore-size polycarbonate filters (Nuclepore). Tachyzoite preparations were used either for the challenge or for the maintenance of chronically infected mice. The preparation of tachyzoites is schematically depicted in the upper right portion of Fig. 1. Immunization. Guinea pigs (20/group) were immunized subcutaneously three times at 3-week intervals with 10 ␮g (500 ␮l) of recombinant SAG1, expressed in P. pastoris and purified as previously described (2), formulated with the SBAS1 adjuvant (a proprietary formulation of SmithKline Beecham, Rixensart, Belgium). As a negative control, SBAS1 alone was injected. Measurement of antibody response. Animals were bled 2 weeks after the last immunization, and sera were tested for the presence of anti-SAG1 immunoglobulin G (IgG) antibodies in an enzyme-linked immunosorbent assay (ELISA). Briefly, immunoplates were coated with recombinant SAG1 (100 ng/well) for 16 h at 4°C. Plates were then washed five times with Tris-buffered saline (TBS)Tween buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.1% Tween 80) and saturated for 1 h at 37°C with 150 ␮l of the same buffer supplemented with 1% bovine serum albumin (Sigma). Serial dilutions of sera to be tested were then incubated for 1 h at 37°C. Plates were washed five times with TBS-Tween buffer, and the antigen-IgG complexes were detected after incubation with a goat anti-guinea pig IgG antibody coupled to alkaline phosphatase (dilution, 1/10,000 in TBS-Tween buffer; Sigma). Enzymatic activity was measured using the pnitrophenylphosphate substrate (Sigma) dissolved in diethanolamine buffer (pH 9.8). Optical density at 415 nm was measured in a Bio-Rad Novapath ELISA reader. Titers were expressed as the reciprocal of the dilution giving 50% of the maximal optical density. Antibodies to SAG1 were also measured in fetuses by using the same protocol. To determine whether sera contained antibodies recognizing T. gondii SAG1, plates were coated with tachyzoite antigen extract (900 ng/well; Argene, Varhiles, France). Congenital infection model. Before antigen immunization, all guinea pigs were monitored for the absence of seroreactivity to Toxoplasma. Females were mated with males (ratio, 1:4) for breeding 3 weeks after the last immunization for a 5-week period. Female guinea pigs were challenged intradermally with 5 䡠 105 filtered tachyzoı¨tes 3 weeks after mating. The infectious status of pups delivered from guinea pigs were evaluated in a lethal mouse model. Fetuses were sacrificed within 48 h following delivery, and each brain was homogenized with 1 ml of PBS and inoculated intraperitoneally into two female BALB/c mice (0.5 ml each). Mice that did not survive from 21 days onwards after brain homogenate injection were considered infected. A pup was considered infected once one of the two injected mice died. PCR. For the detection of T. gondii DNA in the brain homogenate, a brain suspension (100 ␮l) was used for DNA extraction according to the Easy-DNA kit instruction manual (Invitrogen). After precipitation, DNA was resuspended in 100 ␮l of H2O supplemented with 2 ␮l of RNase (final concentration, 40 ␮g/ml) and incubated 30 min at 37°C. Nested PCR was performed in a final volume of 50 ␮l containing 20 pmol of primers, 200 ␮M deoxynucleoside triphosphates, 5% dimethyl sulfoxide, 2.5 U of Taq DNA polymerase (Roche) in PCR buffer (10 mM Tris HCl, 1.5 mM MgCl2, 50 mM KCl [pH 8.3]), and 1, 0.1, or 0.01 ␮l of the purified DNA. The denaturing, annealing, and extension times were 1 min each at 95, 52, and 72°C, respectively. The final extension step continued for an additional 10 min. In the first round, DNA samples were amplified for 35 cycles using primers 5⬘ GGAACTGCATCCGTTCATGAG 3⬘ and 5⬘ TCTTTAAAG CGTTCGTGGTC 3⬘, derived from the T. gondii B1 gene (34), to produce a 194-bp fragment. In the second round, a portion of the first amplification product was amplified under the same conditions used in the first round but with primers 5⬘ TGCATAGGTTGCAGTCACTG 3⬘ and 5⬘ GGCGACCAATCTGCGAATA CACC 3⬘ (for the amplification of a 97-bp fragment). The products were separated on a 3% agarose gel and detected after staining with ethidium bromide. Statistical analysis. Statistical analyses were performed by Fisher’s exact test and one-way analysis of variance (ANOVA). Kappa and McNemar’s tests were used for comparison between the in vivo readout and PCR results.

RESULTS Establishment of a guinea pig model of congenital toxoplasmosis. To test vaccine candidate antigens, an experimental model of guinea pig congenital toxoplasmosis was developed by analyzing the rate of maternofetal transmission of the parasite after infection during guinea pig pregnancy. The infectious status of pups delivered from guinea pigs was evaluated after brain homogenate injection in mice, animals that are very susceptible to T. gondii infection. Pregnant guinea pigs were intradermally challenged with 5 䡠 105 T. gondii tachyzoites after 7 weeks of gestation. To minimize the risk of contamination during delivery, pups were sacrificed within 48 h following birth.

4949

Previous experiments demonstrated that mice died between 5 and 18 days after infected brain homogenate injections. Consequently, mice which did not survive from 21 days onwards after injection were considered infected, and their mortality indicated the infection status of the pups; we considered a pup infected once one of the two injected mice died. In this animal model, the protective effect of recombinant SAG1 vaccination against congenital toxoplasmosis was assessed. Guinea pigs were immunized with 10 ␮g of recombinant SAG1 formulated with the SBAS1 adjuvant before mating. A schematic representation of the congenital model of toxoplasmosis is given in Fig. 1. Humoral immune response induced by SAG1 vaccination. Guinea pigs were immunized subcutaneously with SAG1 adjuvanted with SBAS1c or with adjuvant alone. Table 1 shows that, for two individual experiments, three injections of adjuvanted SAG1 elicited a strong humoral immune response. All the animals seroconverted; the individual titers ranged from 2,000 to 230,000 (Table 1). As expected, no specific anti-SAG1 antibodies were detected in the mock-immunized group. Sera from seropositive guinea pigs were also able to recognize RH strain tachyzoites. However, the titers were lower than those obtained for anti-SAG1 antibodies, probably due to the lower sensitivity of the ELISA. Moreover, natural SAG1 represents 3 to 5% of the total parasite protein (20). Nevertheless, our data indicated that recombinant SAG1 displayed an overall conformation similar to that of SAG1 present on the surfaces of tachyzoites. Moreover, this suggested that glycosylations carried by recombinant SAG1 did not interfere with the immunogenicity of the protein. For the second experiment, we also determined the mean titers of specific antibodies at the moment of delivery, corresponding, respectively, to 9 to 10 weeks after the third SAG1 injection and 2 to 3 weeks after parasite infection. The results showed that the challenge with tachyzoites did not increase the mean titer of anti-SAG1 specific antibody in the SAG1-vaccinated animals, whereas infection elicited antibodies to SAG1 in the control group (Table 1). Anti-T. gondii antibody titers increased markedly in both groups. Protection against maternofetal transmission. To determine whether SAG1 vaccination induced protection against congenital toxoplasmosis, female guinea pigs were mated with males and challenged with purified tachyzoites. The maternofetal transmission of parasites was assessed as described above. Preliminary experiments revealed a greater mortality in litters from mock-immunized guinea pigs than in those from SAG1-vaccinated mothers. Indeed, mothers from the mockimmunized group (n ⫽ 15) produced 55 pups, of which 12 were retrieved from dead mothers and 18 were stillborn. In contrast, out of 53 pups derived from the SAG1-vaccinated group (n ⫽ 14), only 4 and 9, respectively, were retrieved from dead mothers or stillborn (Fisher’s test, P ⬍ 0.002). In both groups, none of the pups retrieved from dead mothers were infected. More than 78% of stillborn pups from the nonvaccinated group were infected with tachyzoites, whereas the incidence of T. gondii infection among stillborn pups from the SAG1-vaccinated group reached only 30% (Fisher’s test, P ⬍ 0.01). Since the fetal mortality of the mock-immunized group was important in the preliminary experiments, housing conditions for pregnant guinea pigs were improved by adding hay to the animal feed and vitamin C to the drinking water. In two independent experiments, 130 (69 for experiment 1; 61 for experiment 2) and 126 (63 for experiment 1; 63 for experiment 2) pups were born from SAG1- and mock-immunized female guinea pigs, respectively. Out of this total, 58 (40 for experi-

4950

HAUMONT ET AL.

INFECT. IMMUN.

FIG. 1. Schematic representation of the guinea pig model of congenital toxoplasmosis. Toxo, T. gondii; IP, intraperitonealy; SC, subcutaneously; ID, intradermally; w, weeks.

ment 1; 18 for experiment 2) animals were stillborn and were excluded from the readout. Pups retrieved from dead mothers were not counted, because in the previous experiments, these fetuses, even those from the mock-immunized group, were never infected with tachyzoites. The results presented in Table 2 show parasite transmission in 36 of 42 fetuses and 45 of 54 fetuses from the litters of guinea pigs vaccinated with adjuvant alone in experiments 1 and 2, respectively. All the litters were totally or partially infected. Very interestingly, immunization with SAG1 significantly protected fetuses from maternofetal transmission: only 7 of 50 pups in experiment 1 and 18 of 52 pups in experiment 2 were found positive, representing 14 and 34% congenital in-

fection, respectively P ⬍ 0.0001 for both experiments by Fisher’s test). Only one litter in each experiment was fully infected by tachyzoites. In both experiments, the mean birthweights of pups in the two groups were the same, and infected pups did not express any clinical symptoms of the disease. It must be pointed out that, in all three experiments performed, brains from stillborn pups or from pups retrieved from dead mothers displayed an abnormal appearance: the organs were not compact but were reduced to a pulp. To confirm the infection rates in pups determined by the in vivo mouse assay, a nested PCR assay was developed to detect the presence of T. gondii DNA in brain homogenates. DNA fragments were amplified with primers derived from the T.

TABLE 1. Titers of specific IgG antibodies directed against SAG1 and T. gondii tachyzoite total lysate in female guinea pigs Titer of antibody (mean ⫾ SD)a Expt and group

Anti-T. gondii

Anti-SAG1 Before challenge

At delivery

Expt 1 SAG1 immunized (n ⫽ 20) Mock immunized (n ⫽ 20)

72,862 ⫾ 64,697 0

NT NT

Expt 2 SAG1 immunized (n ⫽ 17) Mock immunized (n ⫽ 18)

25,508 ⫾ 26,487 0

19,811 ⫾ 13,808 9,988 ⫾ 9,266

a

Measured as described in Materials and Methods. NT, not tested.

Before challenge

389 ⫾ 426 0 989 ⫾ 2,357 0

At delivery

NT NT 2,897 ⫾ 1,676 2,056 ⫾ 1,039

VOL. 68, 2000

VACCINE AGAINST CONGENITAL TOXOPLASMOSIS

4951

TABLE 2. Maternofetal transmission of T. gondii tachyzoitesa Expt and group

No. of litters

No. (%) of pups

Total

Negative

Partially positive

Fully positive

Total

Stillborn

Viable

Positive

Expt 1 SAG1 immunized Mock immunized

17 15

11 0

5 7

1 8

69 63

19 (27) 21 (34)

50 (73) 42 (66)

7 (14) 36 (86)

Expt 2 SAG1 immunized Mock immunized

16 17

5 0

10 6

1 10

61 63

9 (15) 9 (14)

52 (85) 54 (86)

18 (34) 45 (83)

a The incidence of congenital infection in the litters of guinea pigs was measured only for viable pups. A viable pup was considered positive when one of the two mice died within a period of 21 days after brain homogenate injection. A litter was defined as partially or fully positive once one or all of the pups, respectively, were considered positive.

gondii B1 gene. The results obtained from the readout in vivo were well correlated with the presence of total T. gondii DNA in brain homogenates (␬ ⫽ 0.72). Indeed, except for one fetus, no T. gondii DNA was detected by PCR in the brains of fetuses for which homogenate injections did not kill mice. However, PCR was shown to be less sensitive than the in vivo readout in detecting infection, as it failed to demonstrate infection in some fetuses for which the presence of parasites was clearly demonstrated after brain homogenate injections in mice (P ⫽ 0.013 by McNemar’s test). To determine whether the results from the in vivo readout were correlated with the titers of antibodies directed against SAG1, we evaluated the specific humoral response in newborn pups at the moment of sacrifice. The results shown in Table 3 clearly supported the lack of correlation between anti-SAG1 specific antibody titers and protection. Among SAG1-vaccinated animals, there was indeed no statistically significant difference in humoral response between infected and noninfected pups (P ⫽ 0.44 by ANOVA). Strikingly, one fetus from the vaccinated group was completely uninfected although it had no detectable specific anti-SAG1 antibodies. In contrast, high titers of specific antibodies did not systematically impair maternofetal transmission of tachyzoites. DISCUSSION Toxoplasma infection is a significant risk for pregnant women, inducing abortion, stillbirth, or congenital malformations and lesions involving the central nervous system. Several models of congenital toxoplasmosis in rodents such as mice and rats have been described. The main advantage of the rat model lies in the fact that, in contrast to mice, rats, like humans, are resistant to T. gondii infection (17). Moreover, fetuses from chronically infected mothers are protected from congenital toxoplasmosis in the rat model (37), whereas parasite transmission to the offspring can occur during chronic infection in mice (29). The guinea pig model of T. gondii infection shows several features which are relevant for the study of the human congenital toxoplasmosis. The guinea pig is an outbred animal, resistant to parasite inoculation, and in these respects is closer to humans than are inbred mice or rats. A high rate of transmission was reported whatever the gestation period at which infection occurred (36). Among the hemochorial placentas of rodents, the guinea pig placenta is histologically closer to that of humans than the mouse or rat placenta. Indeed, only one layer of tissue separates the fetus from the mother in humans and guinea pigs, whereas this interface is composed of three layers of tissues in mice and rats (28). The present study describes the development of a model of

guinea pig congenital toxoplasmosis. Experiments consisted of challenging pregnant guinea pigs in order to assess maternofetal transmission by subinoculation with fetus brain homogenates in mice. Since the maximal risk of fetal infection appears in humans during the last trimester of pregnancy (35), pregnant female guinea pigs were infected at this period. Our results clearly showed that more than 80% of fetuses from mock-immunized control pregnant guinea pigs were infected. The incidence of fetal infection was approximately 20% higher than in humans, at the same stage of pregnancy, probably because intradermal injection of tachyzoites in guinea pigs is not the physiological route of infection. For two independent experiments, all the litters were partially or totally infected, indicating that not all fetuses from the same infected litter were contaminated. Similar results, observed in the rat congenital toxoplasmosis model (37), can be explained by the fact that guinea pig placentas from the same litter are independent. It must be pointed out that, in two pairs of human twins with congenital toxoplasmosis, only one twin was infected by the parasite (6). Although the in vivo readout in mice after fetus brain homogenate injection was well correlated with T. gondii DNA detection by PCR in brain fetuses, mouse inoculation was more sensitive for diagnosis of congenital toxoplasmosis than nested PCR. Discrepancies between results obtained from PCR and mouse inoculation of tissue homogenates have been reported previously (13, 19) and could be explained by inhibitory factors affecting the PCRs (15). The present animal model was used to evaluate the protective effect of vaccination with a recombinant form of SAG1 before pregnancy on the maternofetal transmission of T. gondii. This tachyzoite major surface antigen was chosen because previous studies demonstrated significant protection after SAG1 vaccination in animal models. Because protective immunity against T. gondii is mainly a Th1-biased immune response (9), the recombinant protein was formulated with SBAS1, a Th1 response inducer adjuvant. Vaccination with recombinant SAG1 induced major protection against maternofetal transmission, as shown by the fact that 66 to 86% of fetuses from SAG1-immunized guinea pigs were not infected. The heterogeneity of the percentage of protection in the two experiments could result from the outbred character of guinea pigs. Moreover, it is noteworthy that immunological responses to SAG1-derived peptides were clearly different between strains of mice with different haplotypes and had a direct influence on protection (33). From preliminary results, it was tempting to conclude that vaccination with SAG1 reduced fetal mortality. However, in

4952

HAUMONT ET AL.

INFECT. IMMUN.

TABLE 3. Correlation between the readout in vivo and the presence of specific IgG antibodies in fetuses from SAG1-immunized or mock-immunized guinea pigsa Antibody titer in mother Group

Litter no.

Before challenge Anti-SAG-1

Vaccinated

Control

At delivery

Anti-T. gondii

Anti-SAG1

Anti-T. gondii

Fetus no.

Antibody titer in fetus Anti-SAG1

Anti-T. gondii

Readout in vivo

1

61,820

500

18,320

2,499

1 2 3

8,332 9,046 4,987

1,900 2,000 918

⫹ ⫺ ⫺

2

24,370

400

9,578

1,356

1 2 3 4

50 5,941 1,596 3,722

85 661 246 591

⫺ ⫺ ⫺ ⫹

3

14,680

0

25,010

4,155

1 2 3 4

19,100 22,920 30,320 17,430

4,447 4,400 4,800 4,620

⫺ ⫹ ⫺ ⫺

4

20,400

1,500

41,910

3,244

1 2 3

13,400 18,140 16,660

1,878 2,173 1,827

⫹ ⫹ ⫺

1

0

0

5,370

2,556

1 2 3

1,716 1,900 1,932

609 502 734

⫹ ⫹ ⫹

2

0

0

3,359

2,949

1 2 3 4

899 4,331 1,837 1,638

287 1,288 707 530

⫹ ⫹ ⫹ ⫹

3

0

0

5,009

1,858

1 2 3

3,221 1,460 618

1,121 584 333

⫺ ⫹ ⫹

4

0

0

37,530

2,493

1 2 3

12,570 15,570 14,350

785 720 919

⫺ ⫺ ⫹

5

0

0

28,190

2,759

1 2 3

2,220 678 1,438

284 73 115

⫺ ⫹ ⫹

a Individual titers of anti-SAG1 and anti-T. gondii antibodies are shown for four litters (4 mothers and 14 pups) from the SAG1-immunized group and five litters (5 mothers and 16 pups) from the mock-immunized group.

two independent experiments, the stillbirth level of the mockimmunized group was close to that of the vaccinated group. Improvements in animal housing conditions, particularly the addition of vitamin C to water, seemed crucial to decreasing fetal mortality in the mock-immunized group, but not in the vaccinated group. The humoral immune response induced by SAG1 in pregnant guinea pigs and fetuses demonstrated that there was no quantitative correlation between anti-SAG1 antibody titers and protection against maternofetal transmission. This result was in agreement with the findings of previous reports showing that antibody production did not play a crucial protective role. Indeed, although specific antibody and complement can inactivate T. gondii in vitro (31), data on protection after passive transfer of serum or monoclonal antibody have been conflicting (12, 18, 23, 32). On the other hand, it is generally well accepted that protective immunity against T. gondii is mainly

T-cell mediated, and that production of gamma interferon (IFN-␥) by CD8⫹ T lymphocytes is required (14, 25). No lymphoproliferative assays were performed in our model, as no immunological tools are available for phenotyping cell-mediated immunity in guinea pigs. The recently published partial mRNA sequence of guinea pig IFN-␥ (GenBank accession no. AF058395) will allow us to determine the presence of this cytokine in proliferative assays by reverse transcription-PCR. In summary, this work demonstrates that the guinea pig model is suitable for studying congenital toxoplasmosis and for testing the protective effect of vaccine candidates against maternofetal transmission. Moreover, this is the first demonstration that a subunit vaccine based on SAG1 confers a high degree of protection against congenital T. gondii infection. Because toxoplasmosis is a significant cause of abortion and neonatal mortality in sheep, an animal which is also an important epidemiological vector for human infection, the protective

VOL. 68, 2000

VACCINE AGAINST CONGENITAL TOXOPLASMOSIS

effect of SAG1 vaccination, particularly against fetal abortion, will be assessed in sheep by using this protocol. ACKNOWLEDGMENTS We thank Najat Loukh and Willy Haid for skillful technical assistance and Ste´phane Veenstra and Sylvie Cayphas (SmithKline Beecham Biologicals, Rixensart, Belgium) for helpful discussions. This work was supported in part by SmithKline Beecham Biologicals and by the Walloon Region of Belgium (Direction Ge´ne´rales des Technologies, de la Recherche et de l’Energie). REFERENCES 1. Angus, C. W., D. Klivington-Evans, J. P. Dubey, and J. A. Kovacs. 2000. Immunization with a DNA plasmid encoding the SAG1 (P30) protein of Toxoplasma gondii is immunogenic and protective in rodents. J. Infect. Dis. 181:317–324. 2. Biemans, R., D. Gre´goire, M. Haumont, A. Bosseloir, L. Garcia, A. Jacquet, C. Dubeaux, and A. Bollen. 1998. The conformation of purified Toxoplasma gondii SAG1 antigen, secreted from engineered Pichia pastoris, is adequate for serorecognition and cell proliferation. J. Biotech. 66:137–146. 3. Bulow, R., and J. C. Boothroyd. 1991. Protection of mice from fatal Toxoplasma gondii infection by immunization with p30 antigen in liposomes. J. Immunol. 147:3496–3500. 4. Buxton, D., K. Thomson, S. Maley, S. Wright, and H. J. Bos. 1991. Vaccination of sheep with a live incomplete strain (S48) of Toxoplasma gondii and their immunity to challenge when pregnant. Vet. Rec. 129:89–93. 5. Buxton, D., and J. Finlayson. 1986. Experimental infection of pregnant sheep with Toxoplasma gondii: pathological and immunological observations on the placenta and foetus. J. Comp. Pathol. 96:319–333. 6. Couvreur, J., G. Desmonts, and J. Y. Girre. 1976. Congenital toxoplasmosis in twins. A series of 14 pairs of twins: absence of infection in one twin in two pairs. J. Pediatr. 89:235–240. 7. Darcy, F., P. Maes, H. Gras-Masse, C. Auriault, M. Bossus, D. Deslee, I. Godard, M. F. Cesbron, A. Tartar, and A. Capron. 1992. Protection of mice and nude rats against toxoplasmosis by multiple antigenic peptide construction derived from Toxoplasma gondii P30 antigen. J. Immunol. 149:2636– 2641. 8. Debard, N., D. Buzoni-Gatel, and D. Bout. 1996. Intranasal immunization with SAG1 protein of Toxoplasma gondii in association with cholera toxin dramatically reduces development of cerebral cysts after oral infection. Infect. Immun. 64:2158–2166. 9. Denkers, E. Y., and R. T. Gazzinelli. 1998. Regulation and function of T-cell-mediated immunity during Toxoplasma gondii infection. Clin. Microbiol. Rev. 11:569–588. 10. Dubey, J. P., and C. P. Beattie. 1988. Toxoplasmosis of animals and man. CRC Press, Boca Raton, Fla. 11. Dubey, J. P., and S. K. Shen. 1991. Rat model of congenital toxoplasmosis. Infect. Immun. 59:3301–3302. 12. Foster, B. G., and W. F. McCulloch. 1968. Studies of active and passive immunity in animals inoculated with Toxoplasma gondii. Can. J. Microbiol. 14:103–110. 13. Fricker-Hidalgo, H., H. Pelloux, C. Racinet, I. Grefenstette, C. Bost-Bru, A. Goullier-Fleuret, and P. Ambroise-Thomas. 1998. Detection of Toxoplasma gondii in 94 placentae from infected women by polymerase chain reaction, in vivo, and in vitro cultures. Placenta 19:545–549. 14. Gazzinelli, R. T., F. T. Hakim, S. Hieny, G. M. Shearer, and A. Sher. 1991. Synergistic role of CD4⫹ and CD8⫹ T lymphocytes in IFN-gamma production and protective immunity induced by an attenuated Toxoplasma gondii vaccine. J. Immunol. 146:286–292. 15. Guy, E. C., H. Pelloux, H. Lappalainen, H. Aspock, A. Hassl, K. K. Melby, M. Holberg-Pettersen, E. Petersen, J. Simon, and P. Ambroise-Thomas. 1996. Interlaboratory comparison of polymerase chain reaction for the detection of Toxoplasma gondii DNA added to samples of amniotic fluid. Eur. J. Clin. Microbiol. Infect. Dis. 15:836–839. 16. Israelski, D. M., and J. S. Remington. 1993. Toxoplasmosis in the nonAIDS-immunocompromised host. Curr. Clin. Top. Infect. Dis. 13:322–356.

Editor: W. A. Petri, Jr.

4953

17. Jacobs, L. 1956. Propagation, morphology and biology of Toxoplasma. Ann. N. Y. Acad. Sci. 64:154–159. 18. Jauniaux, E., D. Jurkovic, B. Gulbis, C. Liesnard, C. Lees, and S. Campbell. 1995. Materno-fetal immunoglobulin transfer and passive immunity during the first trimester of human pregnancy. Hum. Reprod. 10:3297–3300. 19. Jenum, P. A., B. Stray-Pedersen, K. K. Melby, G. Kapperud, A. Whitelaw, A. Eskild, and J. Eng. 1998. Incidence of Toxoplasma gondii infection in 35,940 pregnant women in Norway and pregnancy outcome for infected women. J. Clin. Microbiol. 36:2900–2906. 20. Kahn, A. I., K. H. Ely, and L. H. Kasper. 1991. A purified parasite antigen (p30) mediates CD8⫹ T cell immunity against fatal Toxoplasma gondii infection in mice. J. Immunol. 147:3501–3506. 21. Kasper, L. H., I. A. Khan, K. H. Ely, R. Buelow, and J. C. Boothroyd. 1992. Antigen-specific (p30) mouse CD8⫹ T cells are cytotoxic against Toxoplasma gondii-infected peritoneal macrophages. J. Immunol. 148:1493–1498. 22. Kasper, L. H., M. S. Bradley, and E. R. Pfefferkorn. 1984. Identification of stage-specific sporozoite antigens of Toxoplasma gondii by monoclonal antibodies. J. Immunol. 132:443–449. 23. Krahenbuhl, J. L., J. Ruskin, and J. S. Remington. 1972. The use of killed vaccines in immunization against an intracellular parasite: Toxoplasma gondii. J. Immunol. 108:425–431. 24. Luft, B. J., and J. S. Remington. 1992. Toxoplasmosis encephalitis in AIDS. Clin. Infect. Dis. 15:211–222. 25. McCabe, R. E., B. J. Luft, and J. S. Remington. 1984. Effect of murine interferon gamma on murine toxoplasmosis. J. Infect. Dis. 150:961–962. 26. Nielsen, H. V., S. N. Laeumøller, L. Christiansen, S. Buus, A. Fomsgaard, and E. Petersen. 1999. Complete protection against lethal Toxoplasma gondii infection in mice immunized with a plasmid encoding the SAG1 gene. Infect. Immun. 67:6358–6363. 27. Petersen, E., H. V. Nielsen, L. Christiansen, and J. Spenter. 1998. Immunization with E. coli-produced recombinant T. gondii SAG1 with alum as adjuvant protects mice against lethal infection with Toxoplasma gondii. Vaccine 16:1283–1289. 28. Pijnenborg, R., W. B. Robertson, I. Brosens, and G. Dixon. 1981. Trophoblast invasion and the establishment of haemochorial placentation in man and laboratory animals. Placenta 2:71–92. 29. Remington, J. S., L. Jacobs, and L. Melton. 1961. Congenital transmission of toxoplasmosis from mother animals with acute and chronic infection. J. Infect. Dis. 108:163–173. 30. Roberts, C. W., and J. Alexander. 1992. Studies on a murine model of congenital toxoplasmosis: vertical disease transmission only occurs in BALB/c mice infected for the first time during pregnancy. Parasitology 104:19–23. 31. Schreiber, R. D., and H. A. Feldman. 1980. Identification of the activator system for antibody to Toxoplasma as the classical complement pathway. J. Infect. Dis. 141:366–369. 32. Sharma, S. D., J. M. Hofflin, and J. S. Remington. 1985. In vivo recombinant interleukin 2 administration enhances survival against a lethal challenge with Toxoplasma gondii. J. Immunol. 135:4160–4163. 33. Velge-Roussel, F., M. Moretto, D. Buzoni-Gatel, I. Dimier-Poisson, M. Ferrer, J. Hoebeke, and D. Bout. 1997. Differences in immunological response to a T. gondii protein (SAG1)-derived peptide between two strains of mice: effect on protection in T. gondii infection. Mol. Immunol. 34:1045–1056. 34. Wastling, J. M., S. Nicoll, and D. Buxton. 1993. Comparison of two gene amplification methods for the detection of Toxoplasma gondii in experimentally infected sheep. J. Med. Microbiol. 38:360–365. 35. Wong, S.-Y., and J. S. Remington. 1994. Toxoplasmosis in pregnancy. Clin. Infect. Dis. 18:853–861. 36. Wright, I. 1972. Transmission of Toxoplasma gondii across the guinea pig placenta. Lab. Anim. 6:169–180. 37. Zenner, L., F. Darcy, M. F. Cesbron-Delauw, and A. Capron. 1993. Rat model of congenital toxoplasmosis: rate of transmission of three Toxoplasma gondii strains to fetuses and protective effect of a chronic infection. Infect. Immun. 61:360–363. 38. Zenner, L., J. Estaquier, F. Darcy, P. Maes, A. Capron, and M. F. CesbronDelauw. 1999. Protective immunity in the rat model of congenital toxoplasmosis and the potential of excreted-secreted antigens as vaccine components. Parasite Immunol. 21:261–272.