Vaccines in Congenital Toxoplasmosis: Advances and Perspectives

Congenital toxoplasmosis has a high impact on human disease worldwide, inducing serious consequences from fetus to adulthood. Despite this, there are currently no human vaccines available to prevent this infection. Most vaccination studies against Toxoplasma gondii infection used animal models in which the infection was established by exogenous inoculation. Here, we review recent research on potential T. gondii vaccines using animal models in which infection was congenitally established. Endeavors in this field have so far revealed that live or subunit vaccines previously found to confer protection against extrinsically established infections can also protect, at least partially, from vertically transmitted infection. Nevertheless, there is no consensus on the more adequate immune response to protect the host and the fetus in congenital infection. Most of the vaccination studies rely on the assessment of maternal systemic immune responses, quantification of parasitic loads in the fetuses, and survival indexes and/or brain parasitic burden in the neonates. More research must be carried out not only to explore new vaccines but also to further study the nature of the elicited immune protection at the maternal-fetal interface. Particularly, the cellular and molecular effector mechanisms at the maternal-fetal interface induced by immunization remain poorly characterized. Deeper knowledge on the immune response at this specific location will certainly help to refine the vaccine-induced immunity and, consequently, to provide the most effective and safest protection against T. gondii vertical infection.


INTRODUCTION
Toxoplasma gondii is an obligate intracellular protozoan parasite and the etiologic agent of congenital toxoplasmosis. T. gondii is considered one of the most successful parasites worldwide, infecting over 30% of the human population, with high associated disease burden (1,2). Seroprevalence varies greatly from region to region, ranging from approximately 30% in the American, European, and Asiatic regions, to more than 60% in the African continent (3,4). The disease is potentially dangerous in women who become infected during pregnancy, as it can lead to transplacental transmission of the parasite upon primary infection or re-infection with highly virulent strains (5,6). The incidence of congenital toxoplasmosis varies according to the timing of infection during pregnancy. The transmission rate is greater in the final stages of pregnancy, but the severity of infection is greater in early gestation (7). In the more severe cases, hydrocephalus, chorioretinitis, and cerebral calcification may occur, according to the parasite's brain and ocular tropism (5). An association between congenital infection and the development of neurological and psychiatric disorders later in life, including schizophrenia, Alzheimer's disease, bipolar disease, and even suicidal tendencies has also been suggested (8,9).
Innate and adaptive immunity determines protection against T. gondii infection. An effective immune response must control parasite growth while avoiding immunopathology. In both mice and humans, the IL-12-IFN-g axis is the main immune mechanism responsible for parasite control. Protection mediated by IFN-g produced by NK and Th1 cells induces the expression of immunity-related GTPase and/or guanylatebinding proteins, indoleamine-2,3-dioxygenase, and NO production. TNF-a has also been associated with host protection, as highlighted in patients with defective IFN-g signaling (10). TGF-b, IL-6, and IL-23 promote the production of IL-17 that may also play a host protective role in toxoplasmosis by avoiding excessive IFN-g-dependent inflammation (11). T cells producing IL-10, which include T regulatory cells and Foxp3 -T-bet + Th1 cells, can limit excessive inflammation driven by T. gondii (12). A regulatory role for IL-4, and IL-27 in minimizing host tissue injury due to exacerbated inflammation has also been shown (10,13). A good vaccine candidate would thus induce not only IL-12 and IFN-g, but also counterbalancing cytokines such as IL-4, IL-10, and IL-27 (14).
It is important to note that there are no licensed human vaccines able to prevent toxoplasmosis (15). The lack of effective treatment makes the development of a vaccine against congenital toxoplasmosis one of the main objectives in the management of this disease. Here, recent findings in vaccination approaches to congenital toxoplasmosis using animal models of vertical T. gondii infection will be reviewed and the possible implications in the quest for a vaccine protecting from congenital toxoplasmosis will be discussed.

IN VIVO MODELS TO STUDY CONGENITAL INFECTION
In the study of toxoplasmosis, animal models are used to better understand the disease pathology and the immunological mechanisms induced by infection, as well as to assess the effectiveness of experimental vaccination. However, no single animal model has, so far, been able to mimic all clinical symptoms and signs developed by humans in response to T. gondii infection (16). For congenital toxoplasmosis, the murine model is commonly used as it allows a short pregnancy period and mimics features of human congenital toxoplasmosis, namely the co-localization of inflammatory cells and necrosis at the maternal-fetal interface after primo-infection during pregnancy (17). Further, primo-infection confers resistance to maternal fetal transmission throughout later infections (16). However, the immune response to T. gondii in mice and humans presents distinct features and this should be adequately considered (18,19).
Most vaccination studies in congenital infection reviewed here used Kunming, BALB/c, Swiss OF1, and CBA/J mice. No study was found using the C57BL/6 mouse strain. This might be due to excessive susceptibility to disease exhibited by T. gondii infected C57BL/6 mice. In contrast, BALB/c mice present higher resistance to infection, more closely resembling humans, rendering this strain more suitable for vaccination studies in congenital infection models (20)(21)(22). In murine congenital toxoplasmosis, C57BL/6 mice exhibited higher abortion rate compared to BALB/c due to exacerbated proinflammatory cytokines such as TNF-a (23). Mouse strain differences in major histocompatibility complex haplotypes (e.g., H-2 b , H-2 d , H-2 k for C57BL/6, BALB/c, and CBA/J, respectively), and therefore, antigenic presentation, could explain different susceptibility to T. gondii infection and induced immunopathology (20).
Other animal models can be used such as sheep, rats, guinea pig, or hamster (24)(25)(26)(27)(28). Rats and sheep are widely used in studies addressing drug effectiveness to T. gondii (18). Rats have placental development and hemochorial placentation identical to humans (29). In sheep, congenital toxoplasmosis is very similar to what occurs in humans (15). Thus, sheep is an adequate animal model to study congenital toxoplasmosis, not only because it shares important aspects with the human infection, but also because it directly contributes to the study of new disease control measures in livestock, also severely affected by T. gondii (15). There are well established models of toxoplasmosis in pregnant sheep that provide a starting point for the preparation and testing of new vaccines (15).

VACCINES IN VERTICAL INFECTION
This section aims to describe the vaccines already tested in congenital infection models, and to reveal the gap of analysis concerning immune cells and related mechanisms induced by immunization at the level of maternal-fetal interface as wells as in the neonates. Literature research was performed through a PubMed search, using the query "[(Toxoplasma gondii) AND (vaccine)] AND (congenital toxoplasmosis [title/abstract])". The results presented pertain to studies using pregnant mice where protection and immune responses of pregnant mice and their offspring was evaluated.

Live-Attenuated Vaccines
Live-attenuated vaccines have been the most studied in the context of congenital transmission ( Table 1). These vaccines consist of parasites with reduced virulence but are nevertheless capable of inducing an immune response (36). Alternatively, Barros et   -Ex-vivo splenocyte specific cytokine production (exp#1) -Brain cyst load (exp#2) -Litter size and survival of delivered pups at birth and 5 days after birth (exp#2).
-Brain cyst burden in pups at 35 days after birth.
-Decreased maternal brain parasite load and no clinical signs in immunized dams.
-Higher litter sizes of viable neonates, survival rate and body weight.
-Decreased pup brain parasite load. attenuated virulent strains can also be used, albeit, in this case, the attenuation must be complete to ensure that the vaccine will not cause the disease (37). These vaccines present several advantages, such as using whole parasites, meaning that multiple antigens are available simultaneously. Live vaccines also do not usually require repeated immunizations or the use of adjuvants. However, a concern in using this type of vaccination is the possibility of reversing the parasite to a virulent state causing infection (36). Moreover, live vaccines are not recommended to be used in immunocompromised hosts. Toxovax ® , the only licensed vaccine for toxoplasmosis, administered to avoid abortion in sheep, is a live-attenuated vaccine, using the strain S48 tachyzoites, originally isolated from an aborted lamb in New Zealand (14). This vaccine is not licensed for humans due to the possibility of parasite reversion to its virulent form (38). Moreover, it has a short shelf life and does not lead to full parasite elimination (14). Live-attenuated vaccines may be produced by gene targeted approaches. Such is the case of a modified RH strain lacking two genes, respectively encoding micronemal protein 1 (MIC1), which associates to MICs 4 and 6, rendering them active, and MIC3, a micronemal protein necessary for host cell invasion and MIC8 function. The double deletion of these genes (Dmic1-3) resulted in the loss of function of these five proteins. Female Swiss OF1 mice immunized with Dmic1-3 strain exhibited higher levels of IFN-g and IL-2 and a smaller number of brain cysts compared to non-immunized mice when infected with T. gondii. Also, all pups born from immunized animals survived compared to 64% of non-immunized mice. Moreover, 55% of the pups born from immunized mice did not present brain cysts and those with brain cysts exhibited a 91% reduction of cyst burden (30). This vaccine was also tested in Bizet, Romanov, and Solognot ewes, using the same experimental design but adapted to the ewes' pregnancy length. It showed to be effective by both subcutaneous (sc) and intraperitoneal (ip) routes, inducing protection against abortion, a higher rate of viable lambs and a decrease of brain parasite cysts in the lambs born from vaccinated ewes (31). Wang et al. attempted immunization before pregnancy with a live attenuated vaccine, using tachyzoites of the RH strain with a deletion of the dense granule protein 17 (GRA17) gene (Dgra17) that had previously shown to protect mice from lethal infection (32). The deletion entailed a defective parasitophorous vacuole (PV) and decreased intravacuolar tachyzoite proliferation, due to interference with protein transport across the PV membrane. Ip immunization with Dgra17 strain elicited the production of Th1type response cytokines, IL-12, and IFN-g, as well as of IL-10 in Kunming mice (32). Thereafter, the same authors have tested this vaccine against vertical transmission using both acute and chronic infection models. No maternal clinical signs of infection and abortion were found and the litter sizes of viable neonates in immunized and RH inoculated dams were higher. In both models of infection, pups presented a higher survival rate. Maternal spleen T. gondii-induced cytokine production was evaluated at day 18 of pregnancy, 6 days after infection with Pru strain. Higher levels of Th1-type cytokines IFN-g, IL-12, and IL-2, and of IL-10 were detected. Significantly lower parasite -Healthy dams and decreased maternal brain parasite load.
-Higher litter sizes of viable neonates. burden was found in the brain of immunized dams. Further, partial protection was observed, concerning brain parasite load in the progeny of these animals (32). Recently, a live-attenuated vaccine of the RH strain was engineered to harbor a deletion of Gra17 and of novel putative transporter 1 (NPT1) gene, encoding a selective arginine transporter (33). The virulence of this strain (RHDgra17Dnpt1) was completely attenuated in vivo. The brain cyst burden of immunized dams was significantly lower and no abortions were observed compared with non-immunized infected mice. All pups born from immunized infected mice had about 15 times fewer brain cysts than non-immunized infected mice pups (33). A live-attenuated vaccine using tachyzoites of the Pru strain with a deletion of the calcium-dependent protein kinase 2 (CDPK2) gene (Dcdpk2) was also developed by Wang et al. (34). CDPKs harbored by T. gondii are required for cell invasion and gliding motility and are important virulence factors. Specifically, CDPK2 prevents accumulation of amylopectin to toxic levels in the cell, that would cause the parasite to be morphologically defective and unable to form cysts. Therefore, these parasites were incapable of establishing chronic infection since they were not able to form tissue cysts. Dam brain cyst burden was, in average, 43 times lower than that of non-immunized challenged dams. Splenocytes from immunized dams produced higher levels of IFN-g, IL-2, IL-12, and IL-10 compared with non-immunized mice when stimulated in vitro with soluble tachyzoite antigen. These results indicated that this vaccination approach led to a balanced pro-and counter-inflammatory maternal response, useful to control infection but also to avoid potentially harmful excessive inflammation. Pups from non-immunized infected mice harbored in average 919 ± 339 brain cysts, whereas only 41.4% of the pups from immunized infected mice harbored cysts, averaging 60 ± 33 cyst per brain (34). Recently, Wang et al. created a live-attenuated vaccine from the T. gondii RH strain with a deletion of the tyrosine kinase-like 1 (TKL1) gene (RHDtkl1) (35). Vaccinated Kunming mice remained without clinical signs of infection and showed significant decrease in brain cyst burden 30 days after delivery. No abortions occurred and litter size was unaltered in immunized mice when infected, while all non-immunized infected mice suffered abortions. A decreased brain cyst load was observed in the pups from immunized infected dams indicating reduced vertical transmission (35).

Recombinant Protein Vaccines
Recombinant T. gondii surface antigen 1 (rSAG1) protein (39) was assessed in two models of congenital infection ( Table 2). Haumont et al. immunized Dunkin-Hartley guinea pigs subcutaneously (sc) with rSAG1 three times at 3-week intervals and intradermally challenged 3 weeks after breeding (25). Vaccination induced protection against maternal-fetal transmission as assessed by the brain parasite load in the live pups. However, the SAG1-specific IgG levels in newborn pups did not correlate with protection, while cellular responses were not evaluated (25). In another study, BALB/c and CBA/J mice (H2 d and H2 k background, respectively) were sc immunized twice with rSAG1 (40). A reduction of 50% of maternal-fetal transmission in BALB/c, but not in CBA/J mice, was observed. Protection found in immunized BALB/c mice correlated with a maternal increase in rSAG1-specific IgG1 and a decrease in rSAG1-specific IgG2a. IFN-g and IL-10 levels were increased in serum and in supernatants of T. gondii lysate antigen (TLA)stimulated splenocytes obtained from vaccinated animals. In contrast, the immunized CBA/J mice showed no protection and significantly increased serum IL-10 and IL-4 levels. Further, no differences were observed concerning serum IFN-g or IFN-g levels in the supernatants of TLA ex-vivo stimulated spleen cells from both rSAG1-vaccinated or control CBA/J animals. These observations suggest that the Th1-/Th2type responses induced by the immunization used were affected by the host genetic background, such as the major histocompatibility complex, leading to the different outcomes after immunization (40).

DNA Vaccines
DNA vaccines are among the most promising in T. gondii research. These vaccines have numerous advantages such as ease of development, low-cost production, stable storage, and shipping. More than 50 vaccine variants have been experimentally produced and tested and have shown positive results in their protective capacity using exogenous infection models (36,37). Vaccination with a DNA plasmid encoding SAG1, which previously showed to protect BALB/c mice against infection with the avirulent Beverly type-2 strain, upon intra-muscular immunization with PltPASAG1 plasmid, was tested in a congenital infection model. However, maternalfetal transmission was not reduced as compared to shamimmunized control mice (41) This observation led to the conclusion that different immune mechanisms could mediate protection in adult-acquired infection and congenital parasite transmission (41). Another approach was performed by Mevelec et al. combining DNA plasmids encoding SAG1, T. gondii dense granule antigen 4 (GRA4) and murine granulocyte-macrophage colonystimulating factor (GM-CSF) (42). The survival rates of pups from immunized infected dams was significantly higher compared to non-immunized infected dams. These results indicate that DNA plasmid multiantigen vaccine works better than a single antigen vaccine (42). Further studies are however necessary to highlight the suitability and efficacy of this type of vaccines in vertical infection.

Exosome Vaccines
Exosomes are nano-sized vesicles released by most eukaryotic cells (14). These vesicles can contain a wide variety of molecules, such as proteins, lipids, and nucleic acids, able to activate cellular and humoral responses altering the outcome of parasite infections (45).They can transfer mRNA, miRNA, and proteins between cells, representing a communication path between cells, necessary for immune homeostasis (14,45). Studies showed that immunization with exosomes released from T. gondii-pulsed dendritic cells (DCs) induced protection against congenital toxoplasmosis, associated with IFN-g and IL-10 responses in Barros et   -Brain cyst burden (6 weeks after birth) -Specific IgG isotypes and IgA response and ex-vivo spleen and lymph node specific cytokine production (6 weeks after birth).
-Higher survival rate and pups body weight.
Barros et al.

Vaccination in Congenital Toxoplasmosis
Frontiers in Immunology | www.frontiersin.org February 2021 | Volume 11 | Article 621997 the pups (43). This vaccine, when administered before pregnancy, provided strong fetus protection against infection. Low cyst burden was observed in both immunized dams and pups delivered from immunized dams (43). This was the first description, so far, studying immune responses in the offspring from vaccinated females.

Nanoparticle-Based Vaccines
Advances in research have revealed the use of nanoparticles (NP) as antigen delivery systems, thus setting the basis for a new type of vaccines. This antigen delivery system avoids antigen degradation and increases bloodstream life span, internalization, and presentation by antigen-presenting cells, such as DCs (14). Dipalmitoyl phosphatidyl glycerol-loaded nanoparticles (DGNP) loaded with T. gondii total antigen extract were shown to deliver parasitic antigens to mucosa after intranasal immunization, inducing a specific Th1/Th17 response in vivo (46). Further work has tested the safety and efficiency of DGNP in congenital toxoplasmosis (44). Placental levels of cytokine production were analyzed, which revealed no signs of inflammation exacerbation in immunized mice, even though, there was an increase in IFN-g concentration. IL-10 and IL-6 levels were significantly raised. Survival of offspring and dams was 100% and mean litter size and pup weight was not diminished in infected immunized mice. The parasite burden on the fetus was 86% reduced from females immunized with DGNT/TE compared to controls (44).

CONCLUSION
Most of the research work that addresses vaccination using vertical T. gondii infection models assessed the immune response to vaccination, determining the IgG isotypes and cytokines produced in response to parasite antigen stimulation. Typically, production of IFN-g, IL-2, IL-12, IL-10, and, occasionally, IL-4 was found elevated in response to vaccination in non-pregnant mice until 60-70 days postvaccination. This might explain why this time-point has been referred in the achievement of pregnant animals, since the balance between these cytokines is essential in successful pregnancy as reported in almost all the studies referred here. Most reports gather data from acute, chronic, and congenital infection experiments. However, the immune status during pregnancy is altered (7), therefore, the data from non-pregnant mice must be carefully discussed and not directly extrapolated to pregnancy due its specific immune status.
In two studies using live attenuated vaccines, mixed maternal Th1-/Th2-type responses were induced, being discussed as Th1 crucial against T. gondii congenital infection and Th2 essential for pregnancy maintenance (32,34). Currently, a balance between several subsets of T cells must be considered such as Th1, Th2, Th9, Th17, Th22, and follicular Th cells (Tfh) for a successful human pregnancy (47). Indeed, these T cell subsets contribute to the immune response occurring at the maternal-fetal interface, known to be important not only in protecting against infection and controlling inflammatory response, but also contributing to immune homeostasis, implantation, decidualization, maternal immune tolerance and acceptance of the fetus, and parturition (47). A heterogeneity in the experimental designs was found in the studies reported. The difficulty of performing research work with congenital infection models is very high and may explain the scarcity of research in this area. It would be useful to choose a standard experimental design in the study of vaccines, using congenital models able to get valid and robust results. Murine models are the most chosen because they offer simplified logistics, including the facility to monitor physiologic parameters and the short length of pregnancy. Further, mice allow experiments with higher animal numbers, availability of immunological reagents and genetically modified hosts. Mice were largely validated as an adequate model to study congenital toxoplasmosis, by reproducing many features of human infection (21).
None of the vaccines described so far managed to fully protect against T. gondii vertical transmission, even if providing multiantigenic stimulus. The lack of more effective vaccination approaches may be a consequence of the scarce knowledge on the host protective immune molecular and cellular mechanisms operating at the maternal-fetal interface, specifically at decidua and placenta. Indeed, only one work testing nanoparticles containing T. gondii total extracts, analyzed the cytokine profile in placentas from vaccinated dams (44). This study found that a reduction in IFN-g and an increase in IL-6 and IL-10 production was associated with protection against vertical transmission and ocular toxoplasmosis in the offspring (44). Determining the type of immune response at maternal-fetal interface that can correlate with protection will be useful to refine vaccination approaches, by selecting adjuvants that could adequately polarize T cell responses, thereby leading to protection against congenital infection. On the other hand, it would be also important to understand the immune response developed in the pups born from vaccinated dams and infected during pregnancy. To our knowledge, only one study analyzed the immune response in surviving pups born from vaccinated dams, having found a mixed Th1-and Th2-type response associated with a high survival rate, high weight mean, and low cyst brain burden (43). The development of adaptive fetal immune responses are observed in neonates exposed to an infection environment in uterus but not necessarily infected themselves (48). Thus, it will be worthy to get new insights into how T cell responses and related mediators operate in fetuses and in neonates. Novel vaccine design formulations and delivery systems can also be improved, concerning parasite antigen determinants and elicited immune mechanisms involved in protection against T. gondii vertical infection (36).

AUTHOR CONTRIBUTIONS
MBa, DT, and MBo wrote the initial manuscript. All authors contributed to the article and approved the submitted version.