|Home | About | Journals | Submit | Contact Us | Français|
Development of a fully effective vaccine against the pre-erythrocytic stage of malaria infection will likely require induction of both humoral and cellular immune responses. Protein based vaccines can elicit such broad-based immunity depending on the adjuvant and how the protein is formulated. Here to assess these variables, non-human primates (NHP) were immunized three times with Plasmodium falciparum (Pf) circumsporozoite protein (CSP) or CSP cloned into MG38, a monoclonal antibody that targets DEC-205 (αDEC-CSP), an endocytic receptor on dendritic cells (DCs). Both vaccines were administered with or without poly(I:C) as adjuvant. Following three immunizations, the magnitude and quality of cytokine secreting CD4+ T cells were comparable between CSP + poly(I:C) and αDEC-CSP + poly(I:C) groups with both regimens eliciting multi-functional cytokine responses. However, NHP immunized with CSP + poly(I:C) had significantly higher serum titers of CSP-specific IgG antibodies and indirect immunofluorescent antibody (IFA) titers against Pf sporozoites. Furthermore, sera from both CSP or αDEC-CSP + poly(I:C) immunized animals limited sporozoite invasion of a hepatocyte cell line (HC04) in vitro. To determine whether CSP-specific responses could be enhanced, all NHP primed with CSP or αDEC-CSP + poly(I:C) were boosted with a single dose of 150,000 irradiated Pf sporozoites (PfSPZ) intravenously. Remarkably, boosting had no effect on the CSP-specific immunity. Finally, immunization with CSP + poly-ICLC reduced malaria parasite burden in the liver in an experimental mouse model. Taken together, these data showing that poly(I:C) is an effective adjuvant for inducing potent antibody and Th1 immunity with CSP based vaccines offers a potential alternative to the existing protein based pre-erythrocytic vaccines.
Malaria is an infectious disease of tropical regions infecting approximately 500 million people and causing 1–2 million deaths each year . Adults living in malaria endemic region acquire a form of adaptive immunity over time which provides protection against clinical disease, but is dependent upon continuous exposure to the parasite for its maintenance . Among the five species of Plasmodium known to infect humans, P. falciparum (Pf) is the leading cause of morbidity and mortality. While public health measures such as insecticide treated bed nets and anti-malarial therapy have significant effects on morbidity and mortality, vaccines offer the most compelling intervention for effective and durable prevention of this infection.
At present, the most clinically advanced pre-erythrocytic malaria vaccine candidate uses circumsporozoite protein (CSP), which is expressed abundantly on the surface of the sporozoite stage of the parasite [3, 4]. CSP-specific antibodies , CD8+ and CD4+ T cells [6–8] have been shown to elicit protective immunity in mouse models of malaria. In humans it seems clear that antibodies against CSP may be necessary but not entirely sufficient for the protection seen. Indeed, CSP-specific Th1 responses have also been suggested to correlate with protection in humans following vaccination or natural infection [9, 10]. Based on these findings there has been substantial effort to develop vaccines using CSP as an antigen. Currently, a phase III efficacy trial is underway using the RTS,S vaccine. RTS,S is a complex formulation comprised of two polypeptide chains of Pf CSP (amino acid 207 to 395) linked to hepatitis B surface antigen (HBsAg) to form a particle. This is then mixed with the TLR4 ligand, MPL and QS-21 in an oil-in-water (AS02A) or liposome (AS01B) formulation. Immunization of malaria naïve individuals with RTS,S and AS02A or AS01B induces CSP-specific CD4+ T cells and humoral immune responses with ~ 30–50% efficacy . Importantly, Th1 and CSP-specific humoral immunity are increased with AS01B compared to AS02A [11–14] suggesting that the vaccine formulation may have a critical role in optimizing immunity. Finally, protective immunity induced following immunization with RTS,S appears to wane over time and is not boosted upon natural infection . Thus, developing alternative CSP based vaccines with improved adjuvants, formulations or both may improve the durability of humoral and T cell immunity and enhance protection.
In terms of formulations, recombinant proteins can be administered as soluble antigens or as a particle such as the RTS,S vaccine. Alternatively, more efficient processing and presentation of proteins with increase in immunogenicity can be achieved by targeting the protein directly to dendritic cells (DCs) through monoclonal antibodies against cell surface receptors . In this regard, DEC-205, an endocytic receptor expressed at high levels on lymphoid tissue DCs has been extensively characterized for targeting protein antigens in mice [16–18]. Indeed, DEC-205 mediated delivery of protein antigens improves the induction of both Th1 and CD8+ T cell responses in mouse models [16, 19]. T cell immunity with αDEC-205 requires poly(I:C) as an adjuvant . Poly(I:C), a synthetic double stranded RNA, is a potent inducer of IL-12 and type I IFNs through activation of innate immunity via endosomally expressed TLR3 and the cytoplasmic receptor MDA-5 . Moreover, poly(I:C) through induction of type I IFNs enhances DC maturation and B cell activation leading to induction of potent CD4+ T cell and humoral immune responses, respectively, in mice with protein antigens [21–23]. Finally, type I IFN is critical for cross presentation of protein antigens to generate CD8+ T cell responses in mice [24, 25]. Collectively, these data strongly support poly(I:C) as an adjuvant for improving humoral and cellular immunity with protein based vaccines. While the ability of poly(I:C) to induce broad-based immunity in mice has been established with protein and αDEC vaccines, there is only initial data on the potency of poly (I:C) as an adjuvant in NHP . As innate immune mechanisms are far more similar between humans and NHP than mice, evaluation of NHP may provide a more predictive model for what would be observed in humans.
The primary aim of this study was to compare the adaptive immune responses generated in NHP following immunization with CSP or αDEC-CSP (CSP cloned into the carboxyl terminus of the heavy chain of mAb against DEC-205) with or without poly(I:C) as adjuvant. In addition, as prime-boost immunization with heterologous vaccine formulations has been shown to enhance immunity in a variety of experimental settings compared to either vaccine modality alone, all animals immunized with CSP or αDEC-CSP with or without poly(I:C) were boosted with irradiated PfSPZ. As irradiated PfSPZ are the gold standard for eliciting complete protection in humans, this study provides the first assessment of how they would influence an existing CSP-specific response in NHP. Finally, to determine whether this vaccine approach could have a biologic effect in vivo, C57BL/6 mice were immunized with CSP with or without poly-ICLC  and challenged with chimeric P. berghei sporozoites expressing Pf CSP (Pb-Pf) . Overall, this study shows that poly(I:C) is an effective adjuvant for inducing potent multi-functional CD4+ T cells and antibody responses in mice and NHP with CSP. Thus, poly(I:C) offers a promising adjuvant for application in humans as part of a pre-erythrocytic Pf vaccine regimen with protein based vaccines.
Lab bred Indian rhesus macaques were stratified into comparable groups based on age, weight and sex of the animals. Animals were maintained at the animal facility of BIOQUAL Inc. (Rockville, MD). Female C57BL/6 mice were obtained from The Jackson Laboratory. Mice were maintained in the Vaccine Research Center Animal Care Facility (Bethesda, MD) under pathogen-free conditions. All experiments were conducted according to the guidelines of the National Research Council, under protocols approved by the Institutional Animal Care and Use Committee at the National Institutes of Health.
Groups of NHP (n=4 to 5/group) were immunized three times s.c. in the axillary region with 400 μg of recombinant Pf circumsporozoite protein (CSP) with or without 4 mg of poly(I:C) (Invivogen, San Diego, CA) or poly(I:C) alone at two separate sites. Additional groups of animals were immunized three times with 400 μg of αDEC-CSP (CSP cloned into the carboxyl terminus of heavy chain of αDEC-205 mAb clone MG38) with or without 4 mg of poly(I:C) given in a total volume of 2 ml divided into two sites s.c. [28, 29]. Injections were done at weeks 0, 5 and 13. At week 27 all the animals were boosted with 150,000 irradiated PfSPZ i.v.. For mouse experiments, C57BL/6 mice were immunized with 20 μg of CSP with or without 50 μg of poly-ICLC (Oncovir, Washington, DC) or poly-ICLC alone at weeks 0, 3 and 6. All injections were done s.c.
The Escherichia Coli derived recombinant Pf CSP comprises of amino acid 123–411 from the T4 isolate [18, 30]. This construct contains 3 repeats NANPNVDP and 21 repeats NANP plus the carboxyl terminus domain. For generation of αDEC-CSP, the exact same CSP sequence comprising of amino acid 123–411 was cloned in fusion with the carboxyl terminus of the heavy chain of MG38 antibody (IgG1) . Approximately 30% of the weight of the αDEC-CSP construct comprises of CSP. Poly(I:C) was purchased from Invivogen (San Diego, CA) and Poly-ICLC from Oncovir, Inc. (Washington, DC). Poly-ICLC is poly(I:C) stabilized with poly lysine and carboxymethylcellulose to prevent rapid inactivation in vivo by natural enzymes (Oncovir, Inc., Washington, DC). Radiation attenuated (150 Gy) Pf sporozoites strain NF54 (PfSPZs) were provided by Sanaria, Inc. (Rockville, MD). Pf CSP 15-mer peptides overlapping by 11 amino acids and spanning the entire length of the protein were synthesized by Henry Zebroski in the Rockefeller University proteomics facility.
PBMCs were isolated from fresh blood by Ficoll density centrifugation using Leucosep™ tubes (greiner bio-one). PBMCs were used immediately or after cryopreservation for ELISPOT analysis or intracellular FACS staining.
The frequency of IFN-γ producing cells from PBMCs was determined by ELISPOT assay as previously described . Briefly, 2 X 105 PBMCs were added in triplicate to 96-well plates coated with anti-human IFN-γ (Bender MedSystems, Vienna, VA). Pf CSP pooled peptides (2 μg of 15-mer peptides overlapping by 11 amino acids spanning the entire protein) were added per well and incubated for 18 h at 37°C. The number of spot-forming cells was determined by using the Axioplan 2 imaging system (Zeiss, Germany).
2 × 106 PBMCs were stimulated in complete RPMI 1640 for 5 h with 5 μg/ml α-CD28-Ax680, α-CD49d (BD Pharmingen, San Diego, CA) and Brefeldin A (BFA; Sigma-Aldrich, St. Louis, MO) at 10 μg/ml each with or without 2 μg/ml CSP pooled peptides (15-mer peptides overlapping by 11 amino acids). After stimulation, cells were stained as described previously . Briefly, cells were surface stained for CCR7 at 37°C for 20 min. This was followed by 15 min surface staining at room temperature for CD4, CD8, CD95 (BD Pharmingen, San Diego, CA) and CD45RA (Beckman Coulter, Chaska, MN). In addition, Aqua Blue (Molecular Probes, Carlsbad, CA) was added to identify dead cells. After washing, fixing, and permeabilizing, cells were stained intracellularly for IFN-γ, IL-2, TNF-α, and CD3 (all purchased from BD Pharmingen, San Diego, CA). Approximately 1X106 cells were acquired on a BD LSR II (BD Biosciences) and FACS data was analyzed using FlowJo software (Tree Star) and SPICE. Antibodies specific for CD28 (clone CD28.2), CCR7 (clone 150503), CD4 (clone M-T477), and CD8 (clone OKT8) were labeled with fluorescent conjugates at Molecular Probes. Multi-parameter flow cytometry for mouse spleen cells was performed as described previously . Antibodies specific for CD3, CD4, IFN-γ, IL-2 and TNF-α were purchased from BD Pharmingen (San Diego, CA). Anti-CD8 antibody and cell viability dye ViViD were purchased from Biolegend (San Diego, CA) and Molecular Probes (Carlsbad, CA) respectively.
Pf CSP-specific IgG antibodies were detected in plasma or serum of immunized animals at indicated time points. Ninety-six well plates were coated with CSP overnight. Plasma or serum from immunized animals were added in serial dilutions to CSP coated plates and incubated for 2 h. Following 1 hour incubation with horseradish-peroxidase-conjugated anti-monkey IgG antibody at room temperature (BD Pharmingen, San Diego, CA), plates were developed using 3,3′,5,5′ – tetramethylbenzidine substrate-chromogen (DAKO, Carpinteria, CA).
To measure the functional capacity of monkey sera to block Pf sporozoite invasion of a hepatocyte cell line, an inhibition of liver stage development assay (ILSDA) was performed. HC-04 (1F9) cells (NMRC, Silver Spring, MD) were seeded in 8-well LabTek chamber slides (Thermo Fisher Scientific, Waltham, MA) at a concentration of 40,000 cells/well. Twenty-four hours later, 25,000 irradiated Pf sporozoites were added to each well, in triplicate, with or without the specified serum diluted 1:20. MAb anti-PfCSP [3, 34] (100 μg/ml) was used as a positive control. After 3 hours, wells were washed 3 times with medium and incubated for 72 hours with daily medium changes. At the end of the 3-day incubation, cells were fixed with ice-cold methanol, and stained with rabbit anti-Pf Liver Stage Antigen-1 (LSA-1) sera (Sanaria Inc., Rockville, MD), followed by Alexa Fluor 488 goat anti-rabbit IgG (Invitrogen, Carlsbad, CA). The total number of liver stage parasites per well were counted using a BX51 fluorescence microscope (Olympus) at 400× magnification.
Sporozoite specific IgG from monkeys were measured by adding serially diluted sera to air dried P. falciparum sporozoites mounted on slides, followed by a fluorescein conjugated anti-monkey IgG antibody. The endpoint titer was defined as the greatest serum dilution that yielded a positive fluorescence above the pre-immune control levels.
Immunized mice were challenged with 10,000 chimeric P. berghei sporozoites expressing Pf CSP comprising of the repeat domain (Pb-Pf). These parasites were generated as described elsewhere . 40 hours after challenge with sporozoites, the mice were euthanized and the liver excised to evaluate parasite development. For this purpose, total liver RNA was purified and the presence of malaria specific 18S rRNA measured by using real time PCR, as described . The total number of 18S rRNA copies were determined using a standard curve obtained with purified plasmodial rRNA.
Based on the established role that CSP-specific humoral and cellular immunity has on protection in mice and humans [4–8, 36–38], we compared two distinct protein based vaccine platforms using CSP and poly(I:C) as an adjuvant in NHP. Animals were immunized three times with CSP or αDEC-CSP with or without poly(I:C) or poly(I:C) alone. The magnitude of CSP-specific IFN-γ producing cells in peripheral blood was assessed throughout the course of the study by ELISPOT analysis. As shown in Fig. 1, there was a progressive increase in the frequency of IFN-γ producing cells in animals immunized with CSP or αDEC-CSP + poly(I:C) after three immunizations. The magnitude of IFN-γ producing cells was comparable between animals immunized with CSP + poly(I:C) or αDEC-CSP + poly(I:C) at 2 weeks post each immunization. In terms of durability, both vaccine groups had sustained IFN-γ responses albeit at low frequencies when assessed at week 26 (Fig. 1). These data demonstrate that poly(I:C) is an effective adjuvant for generating strong CSP-specific IFN-γ producing cells in NHP with CSP or αDEC-CSP.
Vaccination with irrradiated PfSPZ is the current gold standard for complete protection in humans [39–41]. Since multiple immunizations are required to elicit protective immunity, our aim was to determine whether a single dose of irradiated PfSPZ could boost CSP-specific cellular and humoral immunity as well as induce responses to other malarial antigens. For this study, we obtained irradiated PfSPZ that had been isolated from the mosquito for the first time under GMP conditions. This provided a unique opportunity to assess their immunogenicity in NHP. All animals immunized with CSP or αDEC-CSP with or without poly(I:C) were boosted at week-27 (12 weeks after the 3rd immunization) via the intravenous route with 150,000 irradiated PfSPZ. An additional group of naïve NHP was added and received only the irradiated PfSPZ intravenously thereby allowing a comparison between prime-boost and boost alone. Interestingly, there was no significant increase in the frequency of IFN-γ producing cells in the PBMCs of any of the vaccine groups following the irradiated PfSPZ boost compared with the last assessment done at week-26. Moreover, a single dose of irradiated PfSPZ did not generate measurable CSP-specific IFN-γ producing cells (Fig. 1). To determine whether the irradiated PfSPZ immunization elicited malarial specific T cell responses to non-CSP antigens, irradiated PfSPZ themselves were used for re-stimulation in vitro in ELISPOT and ICS assays. Using this assay we were unable to detect any malaria specific responses following a single immunization with irradiated PfSPZ (data not shown). As this assay is able to measure malaria specific T cell responses in NHP and humans that had received multiple immunizations with the same irradiated PfSPZ (manuscript in preparation) it suggests that, a single immunization of irradiated PfSPZ is not sufficient in NHP to boost CSP-specific responses or elicit detectable responses to other malarial antigens.
To establish whether the IFN-γ producing cells detected in Fig. 1 by ELISPOT assay were CD4+, CD8+ T cells or both, PBMCs were analyzed by multi-parameter flow cytometry. In addition to IFN-γ, this analysis included assessment of IL-2 and TNF-α as well as cell surface markers to delineate their phenotype into central and effector memory cells. At 2 weeks after the third immunization, there were comparable frequencies of CSP-specific IFN-γ, IL-2, and TNF-α producing CD4+ T cells in CSP or αDEC-CSP + poly(I:C) groups with no cytokines detected in CSP or αDEC-CSP immunized animals (Fig. 2A). We could not detect CSP-specific CD8+ T cells in any vaccine group with this immunization regimen. We next determined the quality of the response as defined by any combination of IFN-γ, IL-2 and TNF-α at the single cell level [32, 33]. The actual frequency of each of the potential combinations of cytokines is shown in Fig 2B and C for CSP or αDEC-CSP + poly(I:C) respectively. The relative proportion of such responses are depicted by pie charts in Fig. 2D and E from individual animals with the mean for all 5 animals highlighted in the box. The mean proportion of CD4+ T cells producing IFN-γ/IL-2/TNF-α is ~ 25–35% following immunization with CSP + poly(I:C) or αDEC-CSP + poly(I:C). Of note, IFN-γ producing CD4+ T cells were split among CCR7− and CCR7+ effector and central memory T cells while IL-2 and TNF-α were higher in CCR7+ cells (data not shown). Thus, both CSP and αDEC-CSP administered with poly(I:C) elicited multi-functional central and effector memory CD4+ T cell responses.
A notable feature of CSP is a central region comprised of a repetitive sequence of four amino acids (NANP) forming a major B cell epitope [4, 42, 43]. Antibodies specific against this region are believed to be a central component of the protection conferred by the RTS,S vaccine . Thus, we assessed the CSP-specific humoral immune responses generated in NHP following immunization with CSP or αDEC-CSP with or without poly(I:C). As shown in Fig. 3A and B, significantly higher titers of CSP-specific IgG antibodies were measured following the third immunization with CSP + poly(I:C) compared to αDEC-CSP + poly(I:C). Two doses of either vaccine were sufficient to attain the peak titers of CSP-specific IgG antibodies with no further boosting upon third immunization (Fig. 3C). A gradual decline in the level of CSP-specific antibodies was observed following the third immunization with either vaccine formulation (Fig. 3D and E). However, measurable CSP-specific IgG antibodies were still detected even at week 38 post primary immunization with CSP + poly(I:C) or αDEC-CSP + poly(I:C) (Fig. 3E).
We next assessed whether irradiated PfSPZ immunization would boost CSP-specific IgG antibodies generated following immunization with CSP or αDEC-CSP + poly(I:C). Consistent with the data in Fig. 1, there was no increase in CSP-specific IgG antibodies following immunization with irradiated PfSPZ in any of the primed vaccine groups or when given alone (Fig. 3D and E). Of note, in our other NHP study, we did detect CSP-specific IgG antibodies after two immunizations i.v. (manuscript in preparation). Therefore, a single dose of PfSPZ injected i.v. does not boost CSP-specific CD4+ T cell and humoral immune responses.
We next determined the functional properties of CSP-specific antibodies generated following immunization with CSP or αDEC-CSP plus poly(I:C) by two assays. As shown in Table IA, there was a four-fold increase in indirect immunofluorescent antibody (IFA) titers against Pf sporozoites from pooled sera collected at week-4 after the 3rd immunization with CSP + poly(I:C) compared to αDEC-CSP + poly(I:C). Moreover, IFA titers against Pf sporozoites from sera of individual animals immunized with CSP + poly(I:C) show mean titers of ~ 400,000 at week-4 post third immunization which declined to ~ 25,000 by week 13 (Table IB). It should be noted that the assessment of IFA titers in Tables IA and B were done on separate days thus accounting for a modest discrepancy in titers from pooled sera and individual sera from CSP + poly (I:C) immunized animals 4 weeks after the 3rd immunization. Nevertheless, such animals had the highest titers on both days. To assess the ability of CSP-specific antibodies to block Pf sporozoite entry into a liver cell line in vitro, a functional assay to inhibit liver stage development (ILSDA) was performed. At 4 weeks after the 3rd immunization, pooled sera from CSP or αDEC-CSP + poly(I:C) immunized NHP inhibited Pf sporozoite entry into the hepatocyte cell line by 56% and 43% respectively whereas sera from CSP, αDEC-CSP or poly(I:C) immunized animals showed only minimal inhibition in this assay (Table II). Additional analysis using sera from individual animals immunized with CSP + poly(I:C) showed inhibition of Pf sporozoite invasion in a range between 33.9 – 71.34 % at week-4 following 3rd immunization (Table III). These data show that the magnitude of CSP specific IgG antibodies and IFA titers against Pf sporozoites were significantly higher in CSP + poly(I:C) compared to αDEC-CSP + poly(I:C) immunized animals. In addition, antibodies generated by either vaccine platform could inhibit invasion of a hepatocyte cell line by Pf sporozoites in vitro in a functional assay.
As Pf CSP + poly(I:C) appears to be effective for eliciting functional antibody and CD4+ T cell responses, we next assessed whether this vaccine approach could limit malaria parasite burden in the liver in vivo. As NHP are susceptible to Plasmodium knowlesi but not Pf, we used a mouse challenge model using chimeric Plasmodium berghei sporozoites expressing Pf CSP (Pb-Pf) . C57BL/6 mice were immunized three times at three-week intervals with Pf CSP with or without poly-ICLC or poly-ICLC alone. It should be noted that for these studies, poly-ICLC which is poly(I:C) formulated with poly lysine and carboxymethylcellulose, was used based on recent studies showing greater potency than the poly(I:C) used in the NHP study (data not shown). In order to assess protection at a memory time-point, all animals were challenged with 10,000 Pb-Pf i.v. two months following the last immunization.
As shown in Fig. 4A high titers of CSP-specific IgG antibodies were generated in the serum of CSP + poly-ICLC immunized animals as compared to CSP or poly-ICLC alone groups. CSP-specific T cell responses were assessed in the spleen 40 hrs post-challenge following ex vivo stimulation of splenocytes with overlapping peptides. As shown in Fig 4B, a high frequency of CSP-specific CD4+ T cells producing IFN-γ, IL-2 and TNF-α were present in mice immunized with CSP + poly-ICLC even 2 months after the 3rd immunization but not in CSP or poly-ICLC alone groups. In terms of T cell quality, Fig. 4D shows the frequency of CD4+ T cells producing various combinations of cytokines and Fig. 4C depicts a pie chart representation of the relative proportion of such responses for individual animals. The mean for all 5 animals shows that ~ 30% of the responses were multi-functional secreting all three cytokines (Fig. 4C). Thus, immunization with CSP + poly-ICLC elicits high titers of CSP-specific IgG antibodies as well as multi-functional CD4+ T cells which are durable. Similar immunogenicity data was observed in BALB/c mice (data not shown). Finally, following challenge with 10,000 Pb-Pf parasites i.v, a significant reduction in parasite burden was assessed in the livers of CSP + poly-ICLC immunized animals as compared with CSP only group (Fig. 4E). Based on the detection of parasite 18S rRNA in the liver it would not be expected that sterilizing immunity would be observed with this high challenge dose and route. Nevertheless, these data provide evidence that this vaccine regimen can have an effect on parasite burden in vivo.
Fully effective vaccines against malaria will likely require generation of broad-based humoral and cellular immunity. Thus, developing non-live vaccine formulations that can elicit potent antibody and T cell responses offers a platform that can itself induce substantial protection such as the RTS,S vaccine, and also be used with other vaccines to further optimize responses and protection. In this report, we compare the immunogenicity of Pf CSP as a soluble antigen versus targeting the CSP to DCs through the DEC-205 receptor using poly(I:C) as adjuvant in NHP. The data show that targeting CSP to DCs through DEC-205 receptor does not enhance the magnitude or alter the quality of CSP-specific CD4+ T cells generated post-immunization compared to CSP with the same adjuvant. In addition, we failed to detect measurable CSP-specific CD8+ T cells in NHP immunized with CSP or αDEC-CSP + poly(I:C). However, significantly higher titers of Pf sporozoite binding CSP-specific IgG antibodies were generated in NHP immunized with CSP + poly(I:C) as compared to αDEC-CSP + poly(I:C) group. Antibodies from both vaccine groups showed neutralizing activity by limiting Pf sporozoite entry into a hepatocyte cell line in an in vitro assay. Taken together, the data show that CSP delivered as soluble antigen elicits comparable Th1 immunity and more potent antibody response as compared to CSP targeted to DCs trough DEC-205 receptor when adjuvanted with poly(I:C).
The lack of demonstrable differences in Th1 immunity with CSP compared to the αDEC-205 targeting approach and the failure to elicit CD8+ T cell responses differs from data obtained in mice [16, 18]. Several explanations may explain these findings. First, in this study we used a high dose (400 μg) of CSP to immunize NHP. Mouse studies reveal that very low amounts of αDEC-antigen are far more efficient than substantially higher amounts of protein . It is possible that a dose response of CSP and αDEC-CSP might have revealed a difference in their relative immunogenicity. Second, the lack of CD8+ T cell responses by the DEC targeting construct may reflect differences between mouse and NHP or a requirement of anti-CD40 antibody that was used in the mouse study [16, 18]. In addition, there may be a limited number of Pf CSP CD8+ T cell epitopes in NHP. Third, the avidity of the αDEC antibody may be critical for optimizing its effects. In this regard, clone MG38 used in this study has a much lower binding affinity for DEC-205 receptor as compared to the newer generation mAb clone 3G9 (data not shown). Indeed, in ongoing experiments in NHP using a higher affinity αDEC-205 (clone 3G9) with an HIV Gag protein administered with Poly-ICLC shows induction of CD8+ T cells that is not observed with HIV Gag protein and Poly-ICLC (manuscript in preparation). Thus, future studies using a higher affinity αDEC-205 antibody will assess whether DC targeting of CSP is advantageous for the development of a pre-erythrocytic vaccine against malaria.
An important aspect of any vaccine study is the durability of the responses. In this regard, low frequency CSP-specific CD4+ T cells were detected in the peripheral blood of both CSP or αDEC-CSP + poly(I:C) immunized groups at approximately 6 months post primary immunization. Moreover, there was also a decline in the titers of CSP-specific IgG antibodies over time. These results are similar to data obtained following immunization with adjuvanted RTS,S . Therefore, a CSP based vaccination regimen may require annual boosting to maintain a sufficient threshold of antibody to mediate protection. Accordingly, CSP-specific CD4+ T cell responses were boosted upon an additional immunization with a single dose of CSP + poly-ICLC at week-73 post primary immunization (data not shown). Finally, in terms of having a biologic effect in vivo, we show that mice immunized with CSP + poly-ICLC had a decrease in the parasite burden in the liver upon i.v. challenge with a high dose of 10,000 chimeric Pb-Pf parasites two months following the last immunization. These data establish that a truncated Pf CSP with poly-ICLC as an adjuvant, has some protective effect by limiting parasite burden in the liver upon Pf infection. Ongoing studies using a full-length Pf CSP and poly-ICLC have shown more substantial reduction in parasite burden in the liver (manuscript in preparation). In comparing this study with other platforms, a prior report using QS-21 as an adjuvant with a multiple antigen peptide based vaccine conferred sterilizing immunity in mice using a mosquito challenge with Pb-Pf parasites. In this study, transfer of serum from immunized animals into naïve mice was sufficient to mediate sterile protection upon challenge with infected mosquito bite . Thus, it will be of interest to assess whether our vaccine regimen used here would have a greater effect on protection with this challenge model rather than with a high dose of parasites administered i.v..
The other major aim of this study was to assess whether a heterologous prime-boost immunization with CSP + poly(I:C) and irradiated PfSPZ could enhance immunity compared to either vaccine modality alone. Multiple immunizations with radiation attenuated or genetically modified Plasmodium sporozoites generate sterilizing immunity across multiple species [39–41, 44–46]. Thus, establishing a combination vaccine regimen that could limit the number of immunizations with irradiated PfSPZ may have important clinical application. Here we sought to determine how boosting with a single immunization with irradiated PfSPZ influenced existing CSP-specific antibody and Th1 responses. CSP-specific CD4+ T cells and humoral immune responses generated following immunization with CSP/αDEC-CSP + poly(I:C) were not boosted by a single dose of 150,000 irradiated PfSPZ administered intravenously. Moreover, we failed to detect any CSP-specific antibody or T cell responses to a single immunization with irradiated PfSPZ. These data suggest that more than one boost with irradiated PfSPZ may be required to enhance the CSP-specific humoral and cellular responses. Indeed, data from human vaccine studies show that several immunizations with irradiated PfSPZ showed only a modest induction of CSP-specific IFN-γ producing cells . Similarly, we have shown that CSP-specific antibodies and malaria-specific T cell responses can be induced in NHP with irradiated PfSPZ but this required at least two immunizations given i.v. (manuscript in preparation). Additional possibilities to explain the lack of boosting by irradiated PfSPZ are that pre-existing CSP-specific antibodies from the primary immunization may bind to the irradiated PfSPZ thus effectively neutralizing any boosting effect. In summary, the inability of irradiated PfSPZ to boost CSP-specific CD4+ T cell and humoral immune responses generated in NHP following immunization with CSP or αDEC-CSP + poly(I:C) suggest that either these heterologous vaccine platforms do not optimize immunity, the order of prime-boosting needs to be changed or there needs to be a greater number of immunizations with irradiated PfSPZ.
In conclusion, this NHP immunogenicity study shows that CSP-specific antibody and CD4+ T cell responses can be elicited using an unformulated Pf CSP with poly(I:C). Thus, poly(I:C) offers a potential alternative to AS01B as an adjuvant with protein based vaccines. A critical question is whether the efficacy seen with the RTS,S/AS01B vaccine in humans is due to antibodies, CD4+ T cells or both. If poly(I:C) is a more effective adjuvant for inducing CD4+ T cell responses than AS01B and such responses have a direct effector role or lead to enhanced antibody responses, it offers the potential to improve upon the clinical efficacy of RTS,S/AS01B. In this regard, the data presented here in NHP with CSP and a non optimized formulation of poly(I:C) induced a significantly higher number of CSP-specific ELISPOT responses than detected in another NHP study following three immunizations with RTS,S/AS01B . The frequency of multi-functional CSP-specific CD4+ T cells secreting IFN-γ, IL-2 and TNF-α reported here is also higher than observed in NHP immunized with RTS,S/AS01B . While the role of multi-functional Th1 responses in malaria infection is unclear, a recent study reported a strong association of multi-functional CSP-specific CD4+ T cells with a better outcome upon malaria challenge following immunization with RTS,S/AS01B in a phase 2 clinical trial . In terms of humoral immunity, the antibody responses generated in this study are measured differently than the other NHP studies with RTS,S/AS01B. However, they are clearly robust and functional. An important caveat to making such comparisons amongst different studies is the variability associated with NHP studies and the assays used to measure adaptive immunity. Ultimately, a direct comparative study would be required to assess differences amongst RTS,S and CSP with poly(I:C) as adjuvant. Finally, since induction of CD8+ T cells has been shown to have a major role in controlling liver stage of malaria infection [6, 7] optimization of the protein and poly(I:C) formulation may be required to enhance cross-priming. As poly-ICLC used in the mouse studies is a more potent formulation than the poly(I:C) used in the NHP study, it is possible that it would enhance priming for CD8+ T cells. The induction of such responses may require additional formulation of the CSP, which is being evaluated. Overall, having a new non-live protein based platform that can induce potent antibody and Th1 responses combined with a recombinant live viral vector vaccine to induce CD8+ T cell responses could provide improved immunity and have a major impact upon the development of a pre-erythrocytic vaccine.
We are grateful to Dr. Fidel Zavala for assistance with the mouse challenge study.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.