Protective Effect of Vaccination with a Combination of Recombinant Surface Antigen 1 and Interleukin-12 against Toxoplasmosis in Mice

Abstract

We studied the immune response induced in mice by recombinant Toxoplasma gondii surface antigen 1 (rSAG1) protein, alone or combined with interleukin-12 (IL-12) as an adjuvant, and the protective effect against toxoplasmosis. Immunization with rSAG1 alone induced a specific humoral type 2 immunity and did not protect the animals from infection. In contrast, immunization with rSAG1 plus IL-12 redirected humoral and cellular immunity toward a type 1 pattern and reduced the brain parasite load by 40%.

Toxoplasma gondii infection is usually asymptomatic or mild, but it can be life threatening in immunodeficient patients and fetuses (17). The immune response induced in immunocompetent subjects by primary T. gondii infection protects throughout life. This protection is associated with a type 1 response, the main effectors of which are T lymphocytes (12) and gamma interferon (IFN-γ) (32). The following general mechanisms are widely accepted (11). Macrophages infected by T. gondii secrete interleukin-12 (IL-12), which activates T cells and NK cells to produce IFN-γ. IFN-γ, in the presence of cofactors such as tumor necrosis factor alpha (TNF-α), in turn activates macrophage toxoplasmicidal activity. Similarly, stimulated T cells secrete IL-2 and IFN-γ, leading to a type 1 cellular immune response. Nevertheless, a moderate type 2 immune response is not necessarily detrimental, because it would counter immunopathologic phenomena linked to excessive inflammatory and type 1 responses (13).

The molecular structure of T. gondii is complex. Several antigens involved in these immunological mechanisms have been identified (10), and these form the basis for work on candidate vaccines. The 30-kDa surface antigen 1 (SAG1) protein is the major surface antigen of tachyzoites (4, 19) and is highly conserved among virulent strains of T. gondii (35). It induces high antibody titers in humans and is recognized by all sera from seropositive subjects (28). The protective value of SAG1 has also been evaluated in vivo in animal models of T. gondii infection and showed only partial protection (3, 7, 8, 20, 24). The model developed in our laboratory since 1995 is based on the rapid immunization protocol designed by Khan et al. (20). Because the use of experimental adjuvants in humans is not authorized, we chose the cytokine IL-12, which is undergoing clinical trials in cancer patients (33). Moreover, IL-12 rapidly and strongly orients the immune response toward a type 1 pattern (34, 36) and plays an essential role in the acute phase of toxoplasmosis (18, 21), and its value as a Th1-inducing immunoadjuvant has clearly been shown in several experimental models of infection (1, 26, 27). These results led us to replace the natural protein with a recombinant SAG1 (rSAG1) protein produced in Escherichia coli and then renatured. This recombinant protein does not contain the glycosyl phosphatidyl inositol anchor, the immunological role of which is, at present, unknown (30). A study by Harning et al. (16) showed that this rSAG1 was well recognized by the serum of seropositive subjects and that it induced antibodies recognizing the natural protein in mice, indicating that the B conformational epitopes were present.

The aim of this work was to characterize, in mice, the immune response induced by T. gondii rSAG1 protein administered alone or combined with IL-12 as an adjuvant and to assess the protective effect of this vaccination against T. gondii infection.

Parasites.

Cysts of the avirulent PRU strain of T. gondii were obtained from the brains of orally infected Swiss OF1 mice (CERJ, Le Genest St. Isle, France) and prepared as previously described (2). Tachyzoites of the virulent RH strain of T. gondii were harvested from peritoneal fluid of Swiss OF1 mice infected intraperitoneally and were used to prepare the T. gondii lysate antigen (TLA) as previously described (31).

Mouse immunization.

Female inbred CBA/J mice (CERJ) were used at 8 to 10 weeks of age. rSAG1 protein expressed in E. coli was kindly provided by Roche-Diagnostics, Basel, Switzerland. Recombinant murine IL-12 was from Genzyme, Cergy Saint-Christophe, France. Mice were immunized twice a week for 2 weeks (days 1, 4, 8, and 13) with rSAG1 alone (cumulative dose, 4 μg) or with rSAG1 plus IL-12 (cumulative dose, 4 μg of each). Control groups were injected with IL-12 alone (cumulative dose, 4 μg) or with the vehicle. Each dose of 100 μl was injected subcutaneously in sterile water. For immunological studies, three mice per group were sacrificed on day 21. Blood was obtained by retro-orbital puncture, and the spleens were removed under sterile conditions. Each experiment was repeated three times, and the experiments were reproducible. The results shown here are from one representative experiment.

Measurement of antibody responses.

Specific anti-rSAG1 immunoglobulin G1 (IgG1) and IgG2a were measured by using an enzyme-linked immunosorbent assay (ELISA). Ninety-six-well Maxisorb microtiter plates (Nunc, Roskilde, Denmark) were coated overnight at 4°C with rSAG1 at 5 μg/ml in 0.1 M carbonate sodium buffer (pH 9.6) (100 μl/well). The plates were then saturated for 1 h at 37°C with 200 μl of phosphate-buffered saline (PBS)–1% bovine serum albumin. Serum samples diluted 1:20 in PBS–0.05% Tween 20 (100 μl/well) were added in duplicate, and the plates were incubated for 1 h at 37°C. Peroxidase-labelled anti-IgG1 (clone LO-MG 1-2; Biosoft, Paris, France) and biotinylated anti-IgG2a (clone LO-MG 2a-3; Sigma, St. Louis, Mo.) were added at a dilution of 1/1,000 (100 μl/well) and incubated for 1 h at 37°C. To detect IgG2a, 100 μl of streptavidin-peroxidase (Amersham, Les Ullis, France) was added (dilution 1/1,000), and this mixture was then incubated for 30 min at room temperature. ABTS [2,2′-azinobis(3-ethylbenzthiazolinesulfonic acid)] substrate solution (KPL, Gaithersburg, Md.) was then added (100 μl/well) for 20 min at room temperature, and the reaction was stopped by addition of 1% SDS (100 μl/well). Plates were read for A₄₀₅ in a microplate reader (Dynex Thermobio Analysis, Issy les Moulineaux, France). Results are expressed as indices of antibody titers (optical density at 405 nm [OD₄₀₅] of the sample/mean OD₄₀₅ of three negative controls).

Measurement of cytokine production.

Spleen cells were prepared as described before (5) and cultured at a density of 4 × 10⁵ per well in 96-well flat-bottom plates (Nunc, Roskilde, Denmark) in RPMI 1640-HEPES supplemented with 10% fetal calf serum (FCS) and with penicillin (100 U/ml) plus streptomycin (100 μg/ml) plus amphotericin B (Fungizone) (0.25 μg/ml) solution (Life Technology Ltd., Paisley, Scotland). Cells were stimulated with 1 μg of TLA per ml. This optimal TLA concentration was determined in preliminary experiments with a range of concentrations (0.1 to 10 μg/ml) (data not shown). Positive controls were assayed with concanavalin A in all experiments (data not shown). Culture medium alone was used for negative controls. Levels of IL-2, IFN-γ, IL-4, IL-6, and IL-10 were measured in the supernatant of spleen cells by using a double-monoclonal antibody (MAb) sandwich ELISA. All MAbs were obtained from Pharmingen (San Diego, Calif.). Preliminary studies showed that the optimal incubation times were 6 h for IL-4 measurement and 5 days for other cytokines (data not shown). Ninety-six-well microtiter plates (Nunc) were coated overnight at 4°C with a capture antibody at a concentration of 2 μg/ml in 0.1 M NaHCO₃ buffer (pH 8.2) (50 μl/well) (anti-IL-2 MAb, clone JES6-1A12; anti-IFN-γ MAb, clone R4-6A2; anti-IL-4 MAb, clone BVD4-1D11; anti-IL-6 MAb, clone MP5-20M3; and anti-IL-10 MAb, clone JES5-2A5). The plates were then blocked overnight with PBS–10% FCS. Samples diluted 1:2 in PBS–10% FCS (triplicate wells) and serial dilutions of recombinant standards (duplicate wells) were incubated overnight at 4°C (100 μl/well). Biotinylated conjugates were then added at 1 μg/ml in PBS–10% FCS (100 μl/well), and the plates were incubated for 45 min at room temperature (anti-IL-2 MAb, clone JES6-5H4; anti-IFN-γ MAb, clone XMG1.2; anti-IL-4 MAb, BVD6-24G2; anti-IL-6 MAb, clone MP5-32C11; anti-IL-10 MAb, clone SXC-1). Avidin-peroxidase (Sigma) was added at 25 μg/ml (100 μl/well) for 30 min at room temperature, and the ABTS substrate was added as described above. A₄₀₅ was measured, and the cytokine concentrations were calculated with RMS software (Dynex Thermobio Analysis, Issy les Moulineaux, France) by interpolation of standard curves. TNF-α was assayed by using a commercial ELISA kit (Duoset mouse TNF-α ELISA kit; Genzyme, Cambridge, Mass.) with the supernatants of cells cultured for 5 days. IL-12 was assayed in cell culture supernatants after 6 h, 48 h, and 5 days of culture, with an antibody pair kindly provided by M. Gately, Hoffman-La Roche, Nutley, N.J. (capture MAb, clone 9A5; peroxidase-labelled MAb, clone 5C3-POD). The detection limits of the assays were as follows: IL-2, 10 pg/ml; IFN-γ, 150 pg/ml; IL-4, 8 pg/ml; IL-6, 10 pg/ml; IL-10, 15 pg/ml; TNF-α, 20 pg/ml; and IL-12, 5 pg/ml.

Challenge infection.

Four weeks after the last immunization, five mice per group were infected orally with 20 cysts of the PRU strain. The mice were killed, and the brains were recovered on day 21 postinfection. Each brain was weighed and homogenized in 1 ml of PBS. The number of cysts was determined microscopically by counting four samples of each homogenate (20 μl each) and was expressed per gram of brain tissue.

rSAG1 alone induces specific humoral type 2 immunity.

The results obtained after immunization of mice by rSAG1 alone were globally comparable to those obtained with the purified natural protein (6). The presence of specific anti-rSAG1 IgG1 (titer, 3.8 ± 1.1) and the absence of IgG2a suggest that rSAG1 preferentially induces type 2 humoral immunity (Fig. 1). The cytokine pattern produced ex vivo by splenocytes in response to TLA stimulation indicates a decreased type 1 response with reduced IL-2 (P = 0.0011) and IFN-γ (P < 0.0001) production relative to that of the vehicle-treated mice (Table 1). However, IL-4 production by splenocytes was not significantly modified by rSAG1 immunization. Addition of IL-12 as an adjuvant abolished the IL-4 production. No IL-10 was detected. These results are reminiscent of those reported by Godard et al., who fragmented SAG1 into five peptides and found that four peptides induced type 2 humoral immunity (exclusive IgG1 production) when they were administered subcutaneously or intravenously to mice (15).

FIG. 1.

Specific anti-rSAG1 IgG1 and IgG2a titers after immunization. Results are mean antibody titers (n = 3) ± standard deviation, expressed as the indices of triplicate wells. The indices were calculated as mean OD₄₀₅ of triplicate samples/mean OD₄₀₅ of three vehicle controls.

TABLE 1.

Cytokine production by splenocytes from mice immunized with rSAG1 alone or combined with IL-12

Immunization regimen	Production of cytokine (pg/ml)a
	IL-2	IFN-γ	IL-4	IL-6	TNF-α
Unstimulated control cells (RPMI)
Vehicle	<10	400 ± 30	23.9 ± 9.0	211.4 ± 9.2	328.0 ± 7.5
IL-12	<10	500 ± 50	<8	55.9 ± 9.0	68.3 ± 3.3
rSAG1	<10	200 ± 40	13.6 ± 1.6	81.5 ± 7.0**	78.8 ± 9.9**
rSAG1–IL-12	<10	1,500 ± 60***	<8	302.5 ± 9.0**	916.0 ± 7.8**
TLA stimulation
Vehicle	26.7 ± 2.0	1,600 ± 50	19.0 ± 4.0	146.9 ± 7.0	627.8 ± 7.7
IL-12	30.1 ± 5.0	600 ± 70	<8	92.2 ± 3.0	273.0 ± 13.0
rSAG1	12.0 ± 2.3**	200 ± 30**	8.2 ± 0.0	111.3 ± 6.8	181.0 ± 0.0**
rSAG1–IL-12	33.1 ± 4.7*	3,100 ± 20**	<8	158.2 ± 2.0	572.0 ± 9.3

^aResults are expressed as means ± standard deviations of triplicate wells. *, P < 0.05; **, P < 0.001 (Student’s t test for comparisons between vehicle control group and rSAG1 or rSAG1–IL-12 treatment group). 

The rSAG1–IL-12 combination reorients the immune response toward the type 1 pattern.

In response to TLA stimulation, splenocytes of mice treated with rSAG1 plus IL-12 produced significantly more IL-2 (P = 0.048) and IFN-γ (P = 0.0002) than the vehicle control group (Table 1). In contrast, the production of the type 2 cytokine IL-4 was totally abolished in both control mice treated with IL-12 alone and mice treated with the combination rSAG1 plus IL-12. In the IL-12 control group, the production of IL-2 was unchanged relative to that in the vehicle control group, whereas IFN-γ production was curiously reduced. Similar intriguing results have been observed by Afonso et al. in an experimental immunization model with an antigen from Leishmania major in combination with IL-12 (1). In response to in vitro TLA stimulation, we did not observe any exacerbated inflammatory response in the rSAG1–IL-12-treated group, because IL-6 and TNF-α production remained unchanged relative to that in the vehicle control group (P = 0.056 and P = 0.060, respectively). No IL-12 production was detected under our experimental conditions, even when a short incubation time (6 h) and different TLA concentrations (0.1 to 10 μg/ml) were used. Nevertheless, this does not rule out very early production after initial contact with the antigen. Some observations indeed suggest that this cytokine is produced within the first few hours of infection, and then production ceases (22). This orientation toward a type 1 immune response in the presence of a type 2 response-inducing antigen has been documented in other models of infection by parasites (1, 26), bacteria (27), and fungi (25).

The precise mechanisms underlying the Th1 reorientation are uncertain. IL-12 would act very rapidly on NK cells and undifferentiated naive Th0 T cells, orienting them towards a Th1 secretory phenotype with IFN-γ production (36). IL-12 itself and the IFN-γ thereby produced would then have an inhibitory effect on the production and effects of IL-4 (29, 34).

The study of specific anti-rSAG1 antibody subclasses (Fig. 1) showed that IgG1 titers were halved in the rSAG1–IL-12 group relative to that in the rSAG1 group (2.0 ± 0.4 and 3.8 ± 1.1, respectively), although the difference was not significant (P = 0.055), probably because of the large standard deviation relative to the sample size. The persistence of IgG1 despite the absence of measurable IL-4 production and the presence of IFN-γ can be explained in several ways. IL-4 concentrations may be below the detection limit but adequate to maintain IgG1 production despite the inhibitory action of IFN-γ. Conversely, IgG1 production may be at least partially independent of IL-4, as suggested by the persistence of IgG1 production in IL-4 gene-deficient mice (23).

Surprisingly, we observed no detectable production of IgG2a, although most published data suggest that IL-12 induces the synthesis of this isotype (14), mainly via IFN-γ (9). Our previous studies with natural SAG1 combined with IL-12 have shown IgG2a production; however, the dose of IL-12 injected was double that used with the recombinant SAG1. In the present experiment, the absence of detectable IgG2a would therefore seem to show an imbalance, in favor of IL-4, between the inhibitory effects of IL-4 and the activation effects of IFN-γ on IgG2a synthesis.

rSAG1 plus IL-12 immunization reduces brain parasite load after peroral infection.

Immunization with rSAG1 alone did not modify the number of cysts compared with that in control mice (1,777 ± 112 and 1,817 ± 536 cysts/g of brain, respectively; P = 0.875) (Fig. 2). In contrast, in the group treated with rSAG1 plus IL-12, parasite load was significantly reduced by 40% (1,083 ± 309 cysts/g of brain; P = 0.029). Our results are difficult to compare with those from published data, because most authors used either survival models or very different models of infection to evaluate the protective effect of their vaccine protocols. Nevertheless, our previous experiments with purified natural SAG1 protein showed similar results with a brain parasite load significantly lower in the SAG1–IL-12 group than that in the SAG1 group despite identical survival rates (6). Debard et al. confirmed the lack of protection provided by the natural SAG1 protein in terms of brain parasite load (8).

FIG. 2.

Brain cyst count on day 21 postinfection. Results are expressed as the mean number of cysts per gram of brain (n = 5) ± standard deviation. ∗, P < 0.05 (Student’s t test for comparisons between the vehicle control group and the rSAG1 or the rSAG1–IL-12 treatment groups).

In conclusion, this study shows that rSAG1 protein administered subcutaneously at a total dose of 4 μg induces specific immunity characterized by a predominantly humoral type 2 pattern and does not protect the animals against infection by a cyst-forming strain of T. gondii. In contrast, combined immunization with IL-12 and rSAG1 protein redirects immunity toward a predominantly humoral and cellular type 1 pattern and reduces the brain parasite load by 40%. Because this protocol protected mice only partially against infection, other injection time points (one injection every 2 weeks for 2 months and one injection monthly for 4 months) were evaluated and showed identical results. Other adjuvants are now under investigation.