Abstract

Introduction

The core desired attribute of a malaria vaccine is the ability to reliably induce long-lasting and sterile immunity against infection. Targeting the pre-erythrocytic sporozoite stage of Plasmodium is thought to combat infection between the site of the mosquito bite and the liver, thereby preventing the onset of morbidity and stopping the parasite’s transmission cycle. Immunity induced by a pre-erythrocytic vaccine is sterile and, thus, drastically differs from natural immunity whereby parasites persist in the blood of infected individuals without invoking severe symptoms of the disease.

The circumsporozoite protein (CSP) is one of the main targets of anti-sporozoite antibodies when immunity is induced by whole-sporozoite vaccine candidates, such as sporozoite immunization under chloroquine cover¹ or radiation-attenuated sporozoites². CSP is the most abundantly expressed protein on the surface of the sporozoite. It has been the leading malaria vaccine antigen for decades, albeit with variable success depending on the vaccine platform^3–5. RTS,S/AS01, which is the current lead recombinant vaccine candidate against malaria, is based on a pseudo-particle consisting of the hepatitis B surface antigen and the central repeat and C-terminal regions of CSP. The vaccine consistently induces 50–60% sterile protection in malaria-naïve individuals⁶; however, for it to become a viable commercial product, higher efficacy levels are needed in light of the continuing morbidity and mortality caused by malaria. Moreover, a highly efficacious anti-infection vaccine would play an important role in accelerating parasite eradication.

To date, there are no confirmed immune correlates of protection against malaria infection. The identification of immune correlates would greatly assist in vaccine design and down-selection of vaccine candidates. There is, however, mounting evidence that antibodies to the repeat region in RTS,S are associated with protection against malaria⁶, although it is unknown how these antibodies may contribute to protection. Previous studies have shown that while CSP-specific antibodies may be a surrogate marker for vaccine efficacy⁶, CSP-specific antibody titers alone are insufficient for predicting protection^{3, 7, 8}, suggesting that while CSP-specific antibodies are essential for vaccine efficacy, some qualitative aspects of RTS,S-induced antibody responses, such as their fine specificity, avidity, isotype, and functional activity, play a major role in determining protection.

Our laboratory is developing functional assays to evaluate humoral and cellular responses induced by vaccination in an effort to reveal immune correlates of protection. We previously reported on the inverse association between serum opsonophagocytic activity (OPA) mediated by RTS,S-induced antibodies and protection against sporozoite challenge, employing a sensitive, high-throughput in vitro OPA assay⁷. The study analyzed OPA in serum samples from an RTS,S trial, in which roughly 50% of the study participants were protected from a controlled human malaria infection (CHMI). These experiments guided the development of the OPA index used here, which is expressed as a log ratio of the OPA titer and the ELISA titer. We found that the OPA index was significantly lower in protected individuals. The data encouraged us to apply this readout tool to additional RTS,S studies in an effort to validate the OPA index as a surrogate marker of protection.

A recent clinical trial⁸ showed that a delayed fractional dose regimen of RTS,S, in which the standard 0–1–2 month immunization schedule was altered to a 0–1–7 schedule, with the third immunization at 20% of the full dose. This change in the standard immunization regimen results in a significant increase in vaccine efficacy, reproducing a result first observed two decades ago⁹. The clinical trial revealed that modifying the vaccine regimen alone, without altering the vaccine formulation, can have a dramatic effect on protective immunity: compared to the standard regimen (012M), the delayed fractional dose regimen (Fx017M) demonstrated a significant increase in vaccine efficacy from 63% (95% CI, 29.4–80.1%) to 86.7% (95% CI, 66.8–94.6%). Furthermore, the study showed that the Fx017M regimen resulted in an increase in CSP-specific antibody avidity and higher somatic hypermutation frequency in CSP-specific B cells⁸. Here, we used serum samples from this trial to expand on our previous findings that the OPA index may be a potential surrogate marker of protection⁷.

The present study was designed to profile the serological response induced by the different regimens through the assessment of the functional activity of CSP-specific antibodies and by integrating the results with other serological measures, such as titer, avidity, and IgG subclasses, to determine the mechanistic basis for the dramatic increase in vaccine efficacy. Although we were unable to confirm the OPA index as a surrogate marker of protection, in part because of the low number of non-protected individuals in this study, we identified two additional factors that differentiate the Fx017M cohort from the 012M cohort: 1) induction of IgG4 antibodies and 2) significantly lower OPA. Importantly, the finding that CSP-specific IgG4 titers were predominantly detected in the Fx017M cohort prompted additional mechanistic experiments to determine the interplay between the IgG subclasses and their impact on biological activity. Using linear regression modeling, we showed that CSP-specific IgG1 and IgG3 antibodies were positively associated with OPA, whereas IgG4 antibodies were negatively associated with OPA. We subsequently carried out IgG subclass depletion experiments and found that depletion of IgG4 antibodies resulted in significantly higher OPA, particularly in sera from the Fx017M cohort. Our results suggest that CSP-specific IgG4 can directly reduce the OPA of other IgG subclasses. The study further explores the utility of OPA as a surrogate marker for efficacy in CSP-based malaria vaccines and its value in identifying potentially new mechanisms of protection.

Results

Serum OPA

We employed a previously developed in vitro assay to measure the serum OPA mediated by immune sera from subjects immunized with the RTS,S vaccine⁷. Briefly, CSP-coated fluorescent beads were incubated with the serum sample at five different dilutions (1:200, 1:400, 1:800, 1:1600, 1:3200) for 1 hr, washed, and then added to THP-1 cells, a human monocytic cell line which expresses receptors comparable to competent macrophages¹⁰. During the 1 hr incubation with the THP-1 cells, opsonized CSP-coated beads were phagocytosed. OPA was then quantified by flow cytometry through two related measures: macrophage frequency (Mfreq), which measures the percentage of the THP-1 population that phagocytosed one or more CSP-coated beads; and mean fluorescent intensity (MFI), which measures the average number of fluorescent beads taken up by phagocytosing THP-1 cells (see Methods). Using the OPA at each serum dilution, we calculated the MFI and Mfreq OPA titer, as the serum dilution that corresponds to 60% of peak opsonization activity.

We tested serum samples from subjects in the 012M (n=16) and Fx017M (n=30) cohorts of a recent RTS,S clinical trial⁸. The MFI and Mfreq OPA titers of each subject are shown in Fig. 1, along with the protection status. As reported previously⁷, we found a strong correlation between MFI and Mfreq (R=0.85), suggesting that both measures share similar underlying mechanisms.

Open in a separate window

Figure 1

Phagocytic activity by subjects in 012M and Fx017M cohorts. OPA, as measured by MFI (red) and Mfreq (blue) OPA titers for each subject in the 012M and Fx017M cohorts, denoted by their protection status as protected (‘P’, solid bars) or not protected (‘NP’, hatched bars). Subjects selected for low and high phagocytic activity pools are shown as ‘*’ and ‘**’, respectively, for both cohorts.

Overall, the Fx017M cohort showed significantly lower OPA titers than the 012M cohort, using both MFI and Mfreq as measures (p<0.05, and p<0.01, respectively) (Fig. 2). We previously reported that a relative measure of OPA, known as the OPA index, was significantly lower in protected individuals in another RTS,S vaccine study. The OPA index expresses the functional activity of sera as the OPA titer relative to the CSP-specific antibody titer⁷, which is determined by ELISA. In the present study, the MFI OPA index was significantly lower in the 017M cohort than in the 012M cohort. The Mfreq OPA index tended to be lower in the Fx017M cohort, although the difference was not statistically significant.

Open in a separate window

Figure 2

Opsonophagocytic activity in 012M and Fx017M cohorts. OPA titers for Mfreq (A) and MFI (B) for the 012M and Fx017M cohorts. OPA indices for Mfreq (C) and MFI (D), relative to CS antibody titer, for 012M and Fx017M cohorts. Protected and non-protected subjects are shown in blue and red, respectively.

In a previous RTS,S study, we reported an association between OPA and C-terminal antibodies⁷, based on the observation that OPA correlated with full-length CSP (FL-CSP) and C-terminal region ELISA titers, but not with repeat region ELISA titers. In this study, we could not reproduce this aspect; instead, OPA correlated with ELISA titers from all three antigens in both cohorts, and unlike in the previous study, the correlation between each ELISA measure was also higher (data not shown). It is unclear, however, whether this was due to differences in the ELISA assay conditions between the two studies.

Serum IgG isotype subclass MFIs

We measured the IgG subclass MFIs for each serum sample, using a Luminex-based assay against three CSP test antigens: full-length recombinant CSP, a peptide representing the repeat region (NANP), and a peptide representing the C-terminal region (PF16) (Fig. 3). Overall, there was no significant difference in IgG1, IgG2, or IgG3 titers between the 012M and Fx017M cohorts, for all three CSP antigens. However, the Fx017M cohort showed significantly higher IgG4 MFIs than the 012M cohort (p<10⁻⁴) when sera were tested against the full-length CSP test antigen. This effect was also seen with the NANP and PF16 antigens: no subject in the 012M cohort showed detectable IgG4 responses to NANP, whereas a majority of subjects from the Fx017M cohort (14 out of 20) showed robust IgG4 MFIs to NANP. For the C-terminal PF16 antigen, only 50% of subjects in the 012M cohort showed detectable IgG4 titers, whereas all subjects in the Fx017M cohort did. In conclusion, the Fx017M cohort differed serologically from the 012M cohort, with its significantly enhanced IgG4 response against all three CSP test antigens.

Open in a separate window

Figure 3

IgG subclass titers for 012M and Fx017M cohorts. Antibody titers for IgG1, IgG2, IgG3, and IgG4 for the 012M and Fx017M cohorts against three CSP test antigens: full-length recombinant CSP (left column), NANP repeat region peptide (middle column), and C-terminal PF16 peptide (right column). Protected and non-protected subjects are colored in blue and red, respectively.

IgG isotype subclass depletion experiments

The linear regression model identified correlations or associations that served as the basis for our hypothesis on the contributions of the different IgG subclasses to OPA. To experimentally validate our hypothesis, we carried out IgG isotype subclass depletion experiments, whereby IgG1, IgG3, or IgG4 antibodies were selectively depleted from a serum sample prior to testing for OPA. Because of limitations in the amount of serum available from each subject and the need for a relatively large quantity of serum for these experiments, we needed to pool serum samples for the depletion experiments. One risk with using an aggregate readout from a pooled sample is that the contribution of any single individual sample to the pooled sample may be masked by other samples in the pool. For example, a serum sample with low OPA might be masked by a serum sample with high OPA if they were combined into a single pooled sample and assessed for OPA. For this reason, we created pooled subsets of low- and high-OPA subjects into a low-OPA pool (low-OPA) and a high-OPA pool (high-OPA), respectively, for the 012M and Fx017M cohorts (see Fig. 1), which resulted in four test samples. This way, we could assess the role of IgG subclass depletion with respect to both overall serum OPA (low vs. high) and cohort (012M and Fx017M), independently.

We carried out the depletion experiments two times, and present both the average results as well as the lower and higher values for each measure from the two independent runs. We found that the depletion experiment resulted in high reproducibility for Mfreq OPA measures between the two runs (R²=0.85), but produced poor reproducibility for the MFI OPA measure (R²=0.23), particularly for the low OPA pools (R²=0.05). We have previously reported that Mfreq OPA measure shows more reliability than the MFI OPA measure⁷, but the reasons for the particularly poor reproducibility of the MFI measure of OPA following depletion of the serum sample are unclear. Therefore, we restricted our analysis of the effects of IgG subclass depletion to Mfreq measures of OPA.

We measured the relative change in OPA in terms of Mfreq, relative to the same sample without depletion, at a serum dilution of 1:250 (Fig. 4). Depletion of either IgG1 or IgG3 alone led to a 0 to 50% reduction in OPA, whereas the combined IgG1+IgG3 depletion led to a reduction of 80 to 100%, suggesting that at a serum dilution of 1:250, IgG1 and IgG3 antibodies act in a redundant manner to contribute to OPA. Furthermore, IgG4 depletion led to an increase in OPA in the Fx017M cohort, particularly for the low-PA pool (+80%), suggesting that in this cohort, IgG4 antibodies may be strongly inhibiting OPA and, thus, are associated with protection in the Fx017M regimen. These results confirm the linear regression modeling that IgG1 contributes to OPA and that IgG4 inhibits OPA, but also shows that IgG3 plays a major role in OPA, a finding not predicted by the modeling. Subsequent analysis of the modeling results showed that IgG1 and IgG3 titers are correlated (R²=0.25), and that combining the two measures resulted in a significant positive association with OPA titers (p<0.001) using both Model 1 and Model 2.

Open in a separate window

Figure 4

Serum opsonophagocytic activity following depletion of IgG isotype subclass antibodies. The change in peak OPA relative to the non-depleted sample, as measured by Mfreq, when the low-OPA and high-OPA pools from the 012M and Fx017M cohorts were depleted of IgG1, IgG3, IgG4, or IgG1+IgG3.

We measured the change in serum OPA following IgG isotype subclass depletion at several concentrations (1:250, 1:1250, 1:6250). Figure 5 shows the change in Mfreq relative to the non-depleted sample for depletion of IgG1, IgG3, and IgG4 within the range of serum concentrations tested. As the serum concentration decreased, the enhancement of OPA following IgG4 depletion increased, particularly in the Fx017M cohort. In the low-OPA Fx017M pool, IgG4 depletion led to an increase in OPA from+70% at 1:250 dilution to+185% at 1:6250 dilution. In the high-OPA Fx017M pool, IgG4 depletion increased OPA from+15% at 1:250 dilution to+100% at 1:6250 dilution. Finally, the low-OPA pool showed a greater increase in OPA following IgG4 depletion than did the high-OPA pool for both the 012M and Fx017M cohorts. Depletion of IgG1 and IgG3 antibodies led to a 20% to 70% reduction in OPA in all cohorts. The reason for this concentration-dependent effect is unclear; it is possible that at lower serum concentrations, the higher-avidity IgG4 outcompetes the lower-avidity cytophilic IgG1 and IgG3 for antigen binding, or that IgG1/IgG3-mediated OPA is more sensitive to changes in concentration than is IgG4-mediated inhibition, for example, if OPA requires the formation of a multivalent Fc-receptor engagement whereas inhibition does not. In conclusion, IgG4 antibodies substantially inhibit OPA, especially in the Fx017M cohort.

Open in a separate window

Figure 5

Serum opsonophagocytic activity following depletion of IgG isotype subclasses at a range of serum concentrations. The change in Mfreq relative to the non-depleted sample is shown following depletion of IgG1, IgG3, and IgG4 for the low-OPA and high-OPA pools for the 012M and Fx017M cohorts.

Discussion

The identification of immunological markers that reliably predict protective, sterile immunity against many infectious diseases has been a major challenge¹¹. Knowing what immune measures are predictive of protection would greatly accelerate vaccine development and reduce the number of clinical trials necessary to identify protective vaccine formulations. In a previous study, we found that low serum OPA in individuals vaccinated with RTS,S was associated with protection⁷, based on the analysis of serum samples from a clinical trial with 20 subjects immunized with the standard regimen of RTS,S/AS01B (0, 1, 2 months; 012M) given at full dose¹².

In the present study, we pursued several objectives. First, we sought to determine whether the candidate surrogate marker of OPA applied broadly to CSP-based malaria vaccines and would still apply when the immunization regimen was changed. Second, we designed the study to better understand the mechanism underlying the significant increase in protection against malaria associated with a change in the vaccination regimen. We obtained samples from a recently published clinical trial that had two arms: a standard regimen (012M) and a delayed fractional dose regimen (Fx017M). The subsequent CHMI demonstrated that vaccine efficacy was significantly improved in the delayed fractional dose regimen, which prompted us to use samples from this clinical trial for serological profiling.

We profiled the humoral responses of 36 study participants by (1) quantitatively assessing antibody titers against FL-CSP, the repeat region, and the C-terminus⁸; (2) measuring the avidity of CSP-specific antibodies⁸; (3) determining CSP-specific IgG1, IgG2, IgG3, and IgG4 MFIs; and (4) quantifying serum OPA. Using linear regression modeling, we identified serological factors that significantly contributed to serum OPA in the study. We identified key differences between the two vaccine regimen cohorts and hypothesized that the enhanced IgG4 responses in the Fx017M cohort was responsible for the reduced OPA compared to the 012M cohort, which we subsequently confirmed using IgG isotype depletion experiments. This is of particular interested in light of our finding in a previous study that low OPA was a surrogate marker of protection in CSP-based vaccines⁷ (which was also reported by Boyer et al. unpublished observation).

The clinical study and the associated analyses highlight the impact of vaccine dose and schedule on the type and quality of the induced immune response and on protective efficacy. The immunological assessment of the recent clinical trial demonstrated that the Fx017M cohort had a significantly higher rate of somatic hypermutation in the RTS,S-vaccine-induced plasma cells as well as higher CSP-specific antibody avidities⁸. Further clinical studies are needed to determine whether the fractional booster dose alone, or the extension of the interval between the second and third immunizations, or both are responsible for the increase in vaccine efficacy and the underlying immunological changes.

Several theories may explain why the delayed fractional dose results in increased IgG4 responses and decreased OPA, and how this may be linked to protection. The reduced antigen load in the third delayed dose may have induced resting memory B cells expressing high-avidity IgG to start differentiating into high-avidity IgG4-secreting B cells. This assumption contrasts with the prevalent view that induction of IgG4 is favored by extended or chronic exposure to antigen in a non-infectious setting¹³. It has been postulated that the typically high-affinity IgG4 subclass outcompetes other subclasses and thereby interferes with any Fc-receptor mediated activations/events¹⁴. Unlike the switch to other antibody isotypes, the induction of IgG4 cannot be explained simply by the cytokines produced by T helper subsets. Instead, other factors such as the prolonged presence of antigen¹⁴ or inflammation as well as the presence of IL-12 during a TH2 response favor the induction of IgG4¹⁵. Extending the immunization regimen may represent the prolonged exposure of the immune system to the CSP antigen. This theory is consistent with our model-based prediction and subsequent experimental confirmation that IgG1 and IgG3 antibodies were responsible for antibody OPA, whereas IgG4 antibodies inhibited IgG1- and IgG3-mediated OPA. Finally, it is important to note that a protective role of IgG4 against infection has been previously reported for a potentially therapeutic monoclonal antibody capable of blocking sporozoite invasion of hepatocytes¹⁶.

Currently, information on the role of antibody subclasses in mediating immunity against pre-erythrocytic stages of Plasmodium is scarce. However, numerous studies have described the anti-parasitic effects of antibodies against erythrocytic stages, indicating that IgG4 antibodies may either block the invasion of host cells or inhibit the intracellular development of the parasites through various mechanisms^{17, 18}. Similarly, the binding of CSP-specific antibodies to sporozoites has been shown to inhibit their motility¹⁹. In a study utilizing in vivo imaging in mice, we previously demonstrated that CSP-specific antibodies, induced by DNA immunization, block invasion of the liver²⁰. It remains to be determined whether other vaccine platforms that elicit CSP-based immunity induce similar immune mechanisms. RTS,S-mediated immunity has been associated with antibodies as well as CD4⁺ T cells⁶, consistent with our rationale in profiling the humoral immune response.

While the hypothesis that opsonized sporozoites are targeted for phagocytosis is plausible as a mechanism of protection, data from our previous study showed a modest, but statistically significant, inverse relationship between OPA and protection⁷, indicating that any possible role that OPA plays in protection is more complex. The current study was underpowered in terms of the sample size needed to reproduce the previously-observed relationship between OPA and protection based on the effect size observed in the previous study. However, we did find that OPA was significantly lower in the sera from subjects in the Fx017M cohort, which had significantly higher efficacy than the 012M cohort. This regimen induced significantly higher efficacy than the 012M regimen, which is consistent with a link between lower OPA and protection. It is possible that phagocytosis mediated by CSP-specific antibodies may not contribute to protection, but instead constitute an immune escape mechanism, similar to that shown in another apicomplexan, i.e., Leishmania ²¹, in which parasite-specific proteins convert the complement factor C3b and thereby promote enhanced binding to complement receptor (CR) 3. The preferential binding to CR3 rather than to CR1 prevents the oxidative burst response during phagocytosis and ultimately results in the escape of the parasite^{22, 23}. Finally, it is important to note that the in vitro assay used to assess OPA in this study has certain differences from possible in vivo serum OPA during sporozoite infection, including the differences in antigen presentation between the CSP-coated fluorescent beads and sporozoites, differences in phagocytic activity between THP-1 macrophages and host macrophages, and the role of other immune components not captured by the in vitro assay, such as complement or T cells.

Studies investigating the role of the CSP antigen in sporozoite infections of host cells demonstrate that the protein downregulates the immune function of liver-resident Kupffer cells^{24, 25}, which are involved in the infection of the liver by sporozoites^{26, 27}. The role of antibodies in the interaction of Kupffer cells and sporozoites, which leads to the uptake and safe passage sporozoites into the liver, has not yet been investigated. It is possible that CSP-specific IgG4 antibodies, which may have high avidity for CSP, outcompete cytophilic IgG1 and IgG3, preventing IgG1- and IgG3-mediated phagocytosis of the sporozoite. This, in turn, could prevent the sheltering of the sporozoite in phagocytic cells and its subsequent immune evasion or regulation of activity that interferes with protection.

Materials and Methods

Electronic supplementary material

Acknowledgements

Author Contributions

S.C. performed the data analysis and computational portion of the work, and compiled parts of the manuscript; C.D. performed the serological assessment of the samples; S.D. provided the reagents for the assay and edited the manuscript; F.L. and E.J. provided scientific discussions; F.L., J.R., A.W., N.W., and E.J. consulted on the analysis and edited the manuscript; E.B.-L. designed the experiments, performed the wet-lab portion of the work, and compiled the manuscript. All authors read and approved the final manuscript.

Notes

Footnotes

References

1. Felgner PL, et al. Pre-erythrocytic antibody profiles induced by controlled human malaria infections in healthy volunteers under chloroquine prophylaxis. Scientific Reports. 2013;3:3549. 10.1038/srep03549. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

2. Epstein JE, et al. Protection against Plasmodium falciparum malaria by PfSPZ Vaccine. Journal of Clinical Investigation: Insight. 2017;2:e89154. [Europe PMC free article] [Abstract] [Google Scholar]

3. Ockenhouse CF, et al. Phase I/IIa safety, immunogenicity, and efficacy trial of NYVAC-Pf7, a pox-vectored, multiantigen, multistage vaccine candidate for Plasmodium falciparum malaria. Journal of Infectious Diseases. 1998;177:1664–1673. 10.1086/515331. [Abstract] [CrossRef] [Google Scholar]

4. Sedegah M, et al. Identification of minimal human MHC-restricted CD8+T-cell epitopes within the Plasmodium falciparum circumsporozoite protein (CSP) Malaria Journal. 2013;12:185. 10.1186/1475-2875-12-185. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

5. Stoute JA, et al. Long-term efficacy and immune responses following immunization with the RTS,S malaria vaccine. Journal of Infectious Diseases. 1998;178:1139–1144. 10.1086/515657. [Abstract] [CrossRef] [Google Scholar]

6. White MT, et al. The relationship between RTS,S vaccine-induced antibodies, CD4(+) T cell responses and protection against Plasmodium falciparum infection. PLoS One. 2013;8:e61395. 10.1371/journal.pone.0061395. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

7. Chaudhury S, et al. The biological function of antibodies induced by the RTS,S/AS01 malaria vaccine candidate is determined by their fine specificity. Malaria Journal. 2016;15:301. 10.1186/s12936-016-1348-9. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

8. Regules JA, et al. Fractional Third and Fourth Dose of RTS,S/AS01 Malaria Candidate Vaccine: A Phase 2a Controlled Human Malaria Parasite Infection and Immunogenicity Study. Journal of Infectious Diseases. 2016;214:762–771. 10.1093/infdis/jiw237. [Abstract] [CrossRef] [Google Scholar]

9. Stoute JA, et al. A preliminary evaluation of a recombinant circumsporozoite protein vaccine against Plasmodium falciparum malaria. RTS,S Malaria Vaccine Evaluation Group. The New England Journal of Medicine. 1997;336:86–91. 10.1056/NEJM199701093360202. [Abstract] [CrossRef] [Google Scholar]

10. Chanput W, Mes JM, Wichers HJ. THP-1 cell line: An in vitro cell model for immune modulation approach. International Immunopharmacology. 2014;23:37–45. 10.1016/j.intimp.2014.08.002. [Abstract] [CrossRef] [Google Scholar]

11. Plotkin SA. Correlates of protection induced by vaccination. Clinical Vaccine Immunology. 2010;17:1055–1065. 10.1128/CVI.00131-10. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

12. Ockenhouse CF, et al. Ad35.CS.01-RTS,S/AS01 heterologous prime boost vaccine efficacy against sporozoite challenge in healthy malaria-naive adults. PLoS One. 2015;10:e0131571. 10.1371/journal.pone.0131571. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

13. Vidarsson G, Dekkers G, Rispens T. IgG subclasses and allotypes: from structure to effector functions. Frontiers in Immunology. 2014;5:520. 10.3389/fimmu.2014.00520. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

14. Collins AM, Jackson KJ. A Temporal Model of Human IgE and IgG Antibody Function. Frontiers in Immunology. 2013;4:235. 10.3389/fimmu.2013.00235. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

15. de Boer BA, Kruize YC, Rotmans PJ, Yazdanbakhsh M. Interleukin-12 suppresses immunoglobulin E production but enhances immunoglobulin G4 production by human peripheral blood mononuclear cells. Infection and Immunity. 1997;65:1122–5. [Europe PMC free article] [Abstract] [Google Scholar]

16. Chappel JA, Hollingdale MR, Kang AS. IgG(4) Pf NPNA-1 a human anti-Plasmodium falciparum sporozoite monoclonal antibody cloned from a protected individual inhibits parasite invasion of hepatocytes. Human Antibodies. 2004;13:91–96. [Abstract] [Google Scholar]

17. Duncan EH, Bergmann-Leitner ES. Miniaturized Growth Inhibition Assay to Assess the Anti-blood Stage Activity of Antibodies. Methods in Molecular Biology (Clifton, N.J.) 2015;1325:153–165. 10.1007/978-1-4939-2815-6_13. [Abstract] [CrossRef] [Google Scholar]

18. Bergmann-Leitner, E. S., Duncan, E. H. & Angov, E. in Malaria Parasites (ed Okwa O. O.) (InTech, 2012).

19. Mishra S, Nussenzweig RS, Nussenzweig V. Antibodies to Plasmodium circumsporozoite protein (CSP) inhibit sporozoite’s cell traversal activity. Journal of Immunological Methods. 2012;377:47–52. 10.1016/j.jim.2012.01.009. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

20. Bergmann-Leitner ES, et al. Protective immune mechanisms against pre-erythrocytic forms of Plasmodium berghei depend on the target antigen. Trials in Vaccinology. 2014;3:6–10. 10.1016/j.trivac.2013.11.002. [CrossRef] [Google Scholar]

21. Olivier M, Gregory DJ, Forget G. Subversion mechanisms by which Leishmania parasites can escape the host immune response: a signaling point of view. Clinical Microbiology Reviews. 2005;18:293–305. 10.1128/CMR.18.2.293-305.2005. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

22. Kane MM, Mosser DM. Leishmania parasites and their ploys to disrupt macrophage activation. Current Opinions in Hematolology. 2000;7:26–31. 10.1097/00062752-200001000-00006. [Abstract] [CrossRef] [Google Scholar]

23. Mosser DM, Edelson PJ. The third component of complement (C3) is responsible for the intracellular survival of Leishmania major. Nature. 1987;327:329–331. 10.1038/327329b0. [Abstract] [CrossRef] [Google Scholar]

24. Bertolino P, Bowen DG. Malaria and the liver: immunological hide-and-seek or subversion of immunity from within? Frontiers in Microbiology. 2015;6:41. [Europe PMC free article] [Abstract] [Google Scholar]

25. Usynin I, Klotz C, Frevert U. Malaria circumsporozoite protein inhibits the respiratory burst in Kupffer cells. Cellular Microbiology. 2007;9:2610–2628. 10.1111/j.1462-5822.2007.00982.x. [Abstract] [CrossRef] [Google Scholar]

26. Ishino T, Yano K, Chinzei Y, Yuda M. Cell-passage activity is required for the malarial parasite to cross the liver sinusoidal cell layer. PLoS Biology. 2004;2:E4. 10.1371/journal.pbio.0020004. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

27. Tewari K, et al. Poly(I:C) is an effective adjuvant for antibody and multi-functional CD4+T cell responses to Plasmodium falciparum circumsporozoite protein (CSP) and alphaDEC-CSP in non human primates. Vaccine. 2010;28:7256–7266. 10.1016/j.vaccine.2010.08.098. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

28. Kester KE, et al. Randomized, double-blind, phase 2a trial of falciparum malaria vaccines RTS,S/AS01B and RTS,S/AS02A in malaria-naive adults: safety, efficacy, and immunologic associates of protection. Journal of Infectious Diseases. 2009;200:337–346. 10.1086/600120. [Abstract] [CrossRef] [Google Scholar]

29. Schwenk R, et al. IgG2 antibodies against a clinical grade Plasmodium falciparum CSP vaccine antigen associate with protection against transgenic sporozoite challenge in mice. PLoS One. 2014;9:e111020. 10.1371/journal.pone.0111020. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

30. Porter MD, et al. Transgenic parasites stably expressing full-length Plasmodium falciparum circumsporozoite protein as a model for vaccine down-selection in mice using sterile protection as an endpoint. Clinical Vaccine Immunology. 2013;20:803–810. 10.1128/CVI.00066-13. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]