Plasmodium species undergo an obligatory initial developmental stage in the liver that leads to the pathogenic erythrocytic phase of the infection. Effective inhibition of sporozoites (Spz) and hepatic parasites, the preerythrocytic (PE) stages, prevents blood infection and consequently disease and transmission. However, immunity against PE parasites must be total, because productive infection of a single hepatocyte can lead to a patent blood infection. Three vaccination protocols have been shown to confer sterile immunity (absence of blood stage parasites after challenge with sporozoites) against PE stages. First, immunization with radiation-attenuated sporozoites (RAS) has long been the gold standard for the induction of sterile immunity in rodents, monkeys, and humans (
12,
19). Second, immunization with a smaller number of live sporozoites under chloroquine prophylaxis (Spz plus CQ) equally conferred sterile protection against sporozoite challenge in mice (
2,
3). Recently, full protection was also obtained in human volunteers immunized by the bites of 15
Plasmodium falciparum-infected mosquitoes while under chloroquine prophylaxis (
23). Finally, immunization with genetically attenuated sporozoites (GAS) was shown to be effective at fully protecting mice from sporozoite infections (
16,
31). In hosts immunized with irradiated sporozoites, persistence of developmentally arrested early liver stages is thought to be important, because primaquine treatment abolishes protection (
25). For vaccination with live sporozoites under drug cover, few if any hepatic parasites are expected to persist, but still, the liver parasites are necessary for the induction of immunity (
3). In vaccination with GAS, liver parasite numbers waned rapidly, becoming undetectable 4 to 5 days postinoculation (
16). In all three models of protection, antigens expressed during the liver stages appeared to be crucial for the induction and maintenance of sterile immunity. The circumsporozoite protein (CSP), a protein expressed by sporozoites and early liver forms, has long been considered the antigen that contributes most to the protective responses induced, justifying its status as the major vaccine candidate (reviewed in reference
18). Nonetheless, two recent studies demonstrated that protection induced by RAS could be obtained in the absence of significant responses against CSP (
10,
15). We wished to characterize the immune responses induced against CSP in mice immunized by Spz plus CQ and to ascertain whether they played an important role in the sterile protection obtained. In order to do so, we used two rodent malaria parasite species,
Plasmodium berghei and
Plasmodium yoelii, and a transgenic line of
P. berghei in which the endogenous
csp gene had been replaced by
csp of
P. yoelii (
P. berghei[
PyCS]). In this manner, the contribution of CSP alone to protection could be assessed.
DISCUSSION
All immunization protocols based on live sporozoites (attenuated or otherwise) have led to the induction of sterile protection. These observations, initially made for RAS in mice more than 40 years ago (
19) and more recently for Spz plus CQ (
2,
3) and GAS (
16,
31), gained in importance and relevance to vaccine development when they were repeated in humans in the 1970s for RAS (
4,
12,
22) and last year for Spz plus CQ (
23). Clinical trials to test the efficacy of immunization with recently produced
P. falciparum GAS (
30) are imminent. In RAS-immunized hosts, the immunodominance of the humoral responses to the major sporozoite surface protein CSP (
13) and the crucial cellular responses (
18) directed at the hepatic parasite in which CSP is also expressed during the early stages have justified the selection and subsequent concerted focus on CSP as a vaccine candidate. The immunodominance of CSP was also observed in mice immunized with GAS. Here, we show that humoral responses against CSP are also dominant in animals immunized by Spz plus CQ.
The immunodominance of CSP notwithstanding, we have previously demonstrated, using a parasite of one species made transgenic for CSP of another species, that immune responses specifically induced against the CS by RAS immunization in mice did not contribute to the consequent sterile protection (
10,
15). Here, we have demonstrated that the full protection induced by Spz plus CQ in inbred BALB/c mice is obtained with a minimal role for responses against CSP. In a single experiment, where outbred CD1 mice (5 mice per group) were immunized with
P. berghei or
P. berghei[
PyCS] Spz and challenged with Spz of
P. berghei, substantial sterile protection (60%) was observed in both groups, suggesting that the mouse background did not influence the outcome. This supports the conclusion that CSP is minimally involved in the protection induced by immunization with Spz plus CQ.
Cross-reactive responses against the different CSPs were induced in some combinations, as expected from the relative homology between the conserved domains of the gene in the two parasite species, but their presence could not account for the protection observed. It must be noted that we did not detect cross-species sterile protection in the Spz-plus-CQ model, as observed for the models of RAS (
15,
26) or GAS (
6). This clearly suggests that the immune mechanisms involved in protection differ between the models but are nevertheless independent of CSP.
In the course of this work, we observed that the levels of immune responses against CSP depended in part on the parasite genetic background in which the antigen was presented (the same CSP gene in P. berghei sporozoites or P. yoelii sporozoites induced different levels of T-cell responses and antibody titers). This phenomenon, though of immunological interest, does not affect the conclusions discussed above.
The demonstration that induced responses to CSP do not lead to full protection should not be taken as inimical to efforts to develop vaccines based on CSP. Indeed, such vaccine formulations have invariably shown some protective efficacy, culminating in RTS, S, where the association of CSP-containing particles with a powerful new-generation adjuvant has significantly reduced malaria morbidity in African adults and children (
1) despite short-lived maintenance of sterile immunity (
27). By comparing the results of this study with our previously published data using the same parasites but the RAS immunization protocol (
15), it appears that CSP plays a less important role in the Spz-plus-CQ immunization protocol than in that based on RAS. A factor that we consider most likely to account for this difference concerns the levels and duration of exposure of CSP to the immune system, particularly when in the liver. CSP is continuously expressed by the Spz, whether normal or radiation attenuated, and by liver stage parasites, but not by blood stage parasites. In contrast to Spz that develop normally, liver stages derived from radiation-attenuated sporozoites can persist for long periods, during which they continue to express CSP (
25). We suggest that this allows the induction of responses to CSP higher than those induced by a shorter, transient expression of the antigen.
Ultimately, the main conclusion from this work and our previous observations using another immunization scheme that also induces sterile immunity against sporozoite infection (
10,
15) is that one or more antigens other than CSP are actually responsible for the induction of sterile immunity against preerythrocytic malaria parasites. Although other preerythrocytic antigens have been studied (TRAP, STARP, SALSA, LSA1, and LSA3 for
P. falciparum) (
11), the identities and natures of most antigens expressed by the sporozoites and, possibly more pertinently, by the hepatic parasite have remained inaccessible to detailed investigations until recently (
9,
24,
28,
32). The implication of other preerythrocytic antigens in our Spz-plus-CQ immunization model would be important to investigate, using an approach similar to ours and those of others (
5,
17,
20). Nevertheless, studies of preerythrocytic parasites remain technically challenging, and knowledge of the biology and immunology of these stages has consequently lagged behind that of the erythrocytic stages.
Acknowledgments
This work was supported by intramural grants from the Agency for Science, Technology and Research (A*STAR) and Institut National de la Sante et de la Recherche Medicale (INSERM) and by a grant from the European community (MALINV contract number LSH-CT-2005-01299) (L.R.).
A.C.G and L.R. equally directed this work. A.C.G., G.S., L.R., M.M., and R.T. conceived and designed the experiments. A.C.G., E.L., M.K., M.M., and N.D. performed the experiments. R.T. constructed the transgenic parasites. A.C.G., G.S., J-.M.C., and M.M. produced or provided parasites and materials. A.C.G., G.S., L.R., and M.M. analyzed the data. A.C.G., G.S., L.R., and M.M. wrote the paper. All authors read and corrected the manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.