Robust Bioconjugated Antigens Induce Immune Responses Preventing Malaria Infection and its Transmission

ABSTRACT

Malaria remains a major global health challenge, with current vaccines providing only limited reductions in case numbers. This study introduces an innovative approach that utilizes engineered Escherichia coli to bioconjugate and assemble the pre‐erythrocytic circumsporozoite protein‐derived RTS,S antigen and the sexual‐stage Pfs47 subdomain into biopolymer particles (BPs), generating the dual‐antigen Pfs47‐RTS,S‐BP vaccine. Compared to single‐antigen BPs, the fused Pfs47‐RTS,S‐BP formulation shows synergistic enhancement in immunogenicity and protective efficacy. Pfs47‑RTS,S‑BP retains its physical properties and antigenicity after storage at 37°C. Although functional in vivo assessment of these thermostable formulations has not yet been performed, the findings indicate promising thermostability that could help overcome cold‑chain limitations in malaria‑endemic regions. Pfs47‐RTS,S‐BP induces durable antibody and T cell responses alongside crucial liver‐resident memory T cells for frontline defence. Vaccine‐induced antibodies efficiently neutralize sporozoites and inhibit parasite transmission. Pfs47‐RTS,S‐BP generates RTS,S‐specific antibody levels 5.3‐fold higher than the protective thresholds of the current RTS,S/AS01 vaccine. The Pfs47‐RTS,S‐BP formulation also achieves 80.4% protection against mosquito bite challenge and 68.15% transmission‐reducing activity, demonstrating strong potential as a robust, dual‐stage malaria vaccine candidate. Overall, this work underscores the potential of Pfs47‑RTS, S‑BP to overcome key limitations of current vaccines and to substantially advance global malaria control efforts.

Engineered malaria antigens enable in vivo bioconjugation to submicron biopolymer particles in E. coli. Two antigens are co‐tethered at high density and defined stoichiometry. The resulting thermostable particles elicit strong, durable antibody and T‐cell responses, conferring robust protection and blocking mosquito transmission. This scalable, cold chain‐independent platform addresses key limitations of current malaria vaccines.

1. Introduction

Malaria remains one of the world's most devastating infectious diseases, with an estimated 249 million cases and 608,000 fatalities reported in 2022, predominantly affecting children under the age of five in sub‐Saharan Africa [1]. Despite decades of control efforts, progress against malaria has stagnated in recent years due to factors such as drug resistance, insecticide resistance, and disruptions to healthcare systems caused by the COVID‐19 pandemic [2, 3]. This alarming situation emphasizes the urgent need for novel approaches, with effective vaccines playing a crucial role in comprehensive malaria control and elimination strategies [4]. Plasmodium falciparum, the causative agent of the most severe type of malaria, has a complex life cycle spanning human and mosquito hosts, presenting numerous potential targets for vaccine development [5]. Traditional approaches have generally targeted individual life‐cycle stages, such as pre‐erythrocytic vaccines targeting sporozoites and liver stages, blood‐stage vaccines targeting merozoites and infected erythrocytes, and transmission‐blocking vaccines targeting sexual stages [6]. Each strategy has its advantages and limitations. Pre‐erythrocytic vaccines aim to prevent infection by targeting sporozoites and liver stages, but near‐perfect efficacy is required to prevent blood‐stage infections [7]. Blood‐stage vaccines seek to lower parasite burden and clinical symptoms, but they face challenges such as antigenic diversity and redundant invasion pathways [8]. Transmission‐blocking vaccines target the sexual stage to prevent transmission by inhibiting parasite development in mosquitoes, but they do not provide direct clinical protection to vaccinated individuals [9].

The RTS,S/AS01 (Mosquirix), the most advanced malaria vaccine, works by inducing antibodies against the circumsporozoite protein (CSP) on sporozoite surfaces to target the pre‐erythrocytic stage [10, 11]. Although RTS,S/AS01 demonstrated moderate efficacy (30%–40%) in phase III trials, the protection significantly decreased over time, with follow‐up studies showing waning immunity over a 4‐year period [10, 12, 13]. Despite this breakthrough, RTS,S/AS01 faces challenges, including a complex four‐dose regimen, substantial cold chain requirements, and higher manufacturing costs limiting accessibility in endemic regions [14, 15]. In contrast, the more recently developed R21/Matrix‐M vaccine, which also targets CSP, has shown higher initial efficacy rates of 75%–80% in preventing clinical malaria over 12 months in clinical trials in children, with a simplified three‐dose primary series plus annual booster [16]. However, studies show that this immunity wanes over time, requiring additional booster doses to maintain protection [17]. While R21/Matrix‐M offers potential advantages such as higher efficacy relative to RTS,S/AS01 [15], the long‐term protection remains to be fully established through ongoing phase III trials and post‐implementation surveillance [18]. Furthermore, the deployment of this vaccine in resource‐constrained settings faces additional hurdles, particularly due to the requisite robust healthcare infrastructure and stringent cold chain requirements [17]. These limitations underscore the need for next‐generation approaches that can boost both the magnitude and longevity of protective immunity while addressing practical deployment challenges in endemic regions. A promising approach to overcome these limitations is the development of multi‐stage vaccines that simultaneously target different phases of the parasite life cycle [6, 19, 20]. By combining pre‐erythrocytic antigens with transmission‐blocking components, such vaccines could potentially offer personal protection while reducing community transmission, thus increasing the epidemiological impact [21, 22]. This dual‐stage approach is in line with the World Health Organization's Malaria Vaccine Technology Roadmap, which advocates vaccines that reduce clinical disease and interrupt transmission [23, 24].

The selection of appropriate target antigens is critical to vaccine efficacy. The CSP has been validated as a pre‐erythrocytic target through the development of RTS,S/AS01, which combines the central repeat and C‐terminal regions of CSP [25]. A few candidate antigens, such as Pfs25, Pfs48/45, and Pfs47 have emerged targeting transmission blocking [9, 26, 27, 28, 29, 30]. Among these, Pfs47 is particularly promising as it plays a crucial role in parasite evasion of the mosquito immune system and exhibits comparatively limited polymorphism in contrast to other transmission‐blocking candidate antigens [29, 31]. In addition to antigen selection, delivery systems are important to enhance antigen immunogenicity and efficacy. Particulate delivery platforms have proven to be more immunogenic than soluble antigens by facilitating efficient uptake by antigen‐presenting cells, enhancing antigen processing, and offering sustained antigen presentation [32, 33, 34, 35, 36]. Various particulate systems have been investigated for malaria vaccines, including virus‐like particles, liposomes, and polymeric nanoparticles, each has unique benefits and drawbacks with regard to manufacturing scalability, stability, and immune activation profiles [37, 38, 39, 40, 41].

Biopolymer particles (BPs) represent an emerging class of particulate delivery systems with several advantageous properties for vaccine applications [42, 43]. These BPs are assembled inside engineered Escherichia coli and are made of biocompatible polyhydroxybutyrate (PHB) densely coated with PHB synthesizing enzyme that can be engineered to serve as a BP anchor for covalent display of antigens of interest [44, 45]. This approach enables precise control over antigen display and orientation while maintaining native epitope conformation, both of which are essential for inducing protective immunity [43]. Additionally, the particulate nature of BPs facilitates efficient uptake by antigen‐presenting cells and provides an adjuvant‐like effect, potentially enhancing immunogenicity [42, 44, 46]. A significant advantage of BP‐based vaccines is that they can be produced in bioengineered bacteria using established fermentation processes, which are scalable, enabling cost‐effective manufacturing [42, 47, 48]. This is particularly relevant for malaria vaccines, which need to be used in resource‐limited settings [49]. Furthermore, preliminary studies suggest that BP‐based vaccines may exhibit enhanced thermostability compared to conventional protein vaccines, which could potentially reduce the need for a cold chain, a critical consideration for deployment in tropical areas where malaria is endemic [47, 48]. While several studies have investigated the use of particulate delivery systems for malaria antigens [21, 37, 41, 50, 51, 52, 53], the application of BPs for multi‐stage malaria vaccine development remains unexplored. This approach has the potential to combine the advantages of particulate delivery with the strategic targeting of multiple stages of the parasite life cycle, potentially addressing several limitations of current malaria vaccine candidates.

Importantly, assessing the functional efficacy of vaccine‐induced antibodies, such as their capacity to inhibit sporozoite motility or transmission, remains a major challenge in malaria vaccine research. To address this, we applied a novel quantitative software tool (SMOOT), enabling objective and reproducible analysis of antibody‐mediated inhibition of parasite motility [54]. This functional evaluation complements immunogenicity data and provides critical insight into vaccine‐induced protection mechanisms.

In this study, we developed a dual‐antigen malaria vaccine using the BP platform, targeting both pre‐erythrocytic stage infection and transmission. We provide a comprehensive characterization of this vaccine candidate, including physicochemical properties, thermal stability, immunogenicity, durability of memory responses, and protective efficacy in challenge models. Our findings demonstrate that this BP‐based dual‐antigen approach effectively disrupts the parasite life cycle at two crucial stages while offering benefits in safety, cost‐efficiency, thermal stability, scalability, robust immunogenicity, and strong protection.

2. Results

2.1. Development and Characterization of BP‐Based Malaria Vaccine Candidates

For vaccine development, we selected key epitopes from P. falciparum antigens that target pre‐erythrocytic and sexual stages of the malaria life cycle (Table S1). The malaria epitope RTS,S (targeting pre‐erythrocytic stage) and a domain of Pfs47 (targeting sexual‐stage) were translationally fused to the BP anchor and expressed in bioengineered E. coli (Figure 1a) as previously described [42, 43]. As shown in Figure 1a, we developed a production platform utilizing bioengineered E. coli for the expression of malaria epitope‐coated BPs through a multistep process, involving gene expression in modified bacteria, self‐assembly of particles, followed by cell disruption and purification. This approach resulted in four distinct BP‐based vaccines, including a BP construct with only the surface‐exposed BP anchor (64.2 kDa), two single‐antigen BPs, namely RTS,S‐BP (75.3 kDa) and Pfs47‐BP (77.6 kDa), and a dual‐antigen BP, Pfs47‐RTS,S‐BP (88.4 kDa) (Figure 1b).

FIGURE 1.

Design, development, and characterization of malaria vaccine candidates targeting pre‐erythrocytic stage infection and transmission blocking using biopolymer particle (BP)‐based delivery system. (a) Schematic representation depicting the production of malaria epitope‐coated BPs in engineered E. coli. The malaria epitopes RTS,S and Pfs47 were genetically fused to BPs expressed in the bioengineered bacteria. This multistep process involves gene expression in the modified E. coli, self‐assembly of the particles, followed by cell disruption and purification. The purified BPs then undergo formulation and sterilization, ensuring their quality and safety, before being subjected to animal studies evaluating immunogenicity, thermostability, longevity, and protection. (b) Schematic illustrating the vaccine constructs, highlighting the fusion protein compositions and their corresponding molecular weights: BP (64.2 kDa), RTS,S‐BP (75.3 kDa), Pfs47‐BP (77.6 kDa), and Pfs47‐RTS,S‐BP (88.4 kDa). The BP particles were engineered to display either the RTS,S or Pfs47 epitopes, or a combination of both, the developed a dual‐antigen vaccine candidate Pfs47‐RTS,S‐BP. (c) Structural model of the dual‐antigen malaria vaccine, Pfs47‐RTS,S‐BP, was predicted using AlphaFold2 showing the RTS,S and Pfs47 epitopes displayed on the surface of the BP core, forming a dense epitope layer. The RTS,S epitope is depicted in green, while Pfs47 is shown in blue, and the BP anchor is represented in grey. The Cys319 amino acid within the BP anchor covalently links the fusion protein to the polymer chains that make up the BP core. (d) SDS‐PAGE analysis of purified BPs of the vaccine constructs, showing protein bands at the expected molecular weights. Comparison of the lanes demonstrating successful recombinant protein expression and purification of the BP‐epitope fusions. SDS‐PAGE image presented is representative of three independent vaccine production batches. (e) TEM images of the vaccine constructs, produced in both whole cells (top row; scale bar, 200 nm) and as purified BPs (bottom row; scale bar, 200 nm), demonstrating the structural integrity of the particles. These representative Images were obtained from three independent experiments. (f) The surface charge characteristics of the various BP formulations, both with and without the Quil‐A adjuvant (Q), were assessed through zeta (ζ) potential measurements. (g) Particle size distribution profiles of BP formulations with and without Quil‐A adjuvant, were obtained through dynamic light scattering measurements, showing particle diameter distributions. The particle size distribution and ζ‐potential measurements were determined using the Litesizer 500 instrument. Data are means of three replicates ± SEM.

Structural modelling of the Pfs47‐RTS,S‐BP vaccine using AlphaFold2 [55] revealed the strategic arrangement of RTS,S and Pfs47 epitopes displayed on the surface of the BP core, with a covalent linkage via the Cys319 amino acid in the BP anchor to the polymer chains comprising BP core (Figure 1c and Figure S1), consistent with established structural characteristics of BPs⁵⁶. This arrangement creates a dense epitope/antigen layer accessible for immune recognition, a feature crucial for effective interaction with the immune system [32] and contributes to the observed synergistic effect. SDS‐PAGE analysis of proteins attached to BPs confirmed successful expression and purification of all BP‐epitope fusions at their expected molecular weights, with distinct bands corresponding to each fusion protein (Figure 1d). Purified protein yields were quantified using densitometric analysis (Figure S2a) to determine the concentration of RTS,S, Pfs47, and RTS,S‐Pfs47 in the vaccine preparations, enabling precise dosing for subsequent experiments. Transmission electron microscopy (TEM) demonstrated the structural integrity of the particles in both whole cells and purified preparations, with clearly defined morphology consistent across all vaccine constructs (Figure 1e) and aligned with previous findings [44, 56].

Polyhydroxybutyrate (PHB) contributed to approximately 40%–60% of the whole‐cell mass and 60%–80% of the purified BP mass as determined by HPLC analysis (Figure S2b). Quantification of the fusion proteins using densitometry revealed the proportional representation of RTS,S, and Pfs47 components in the malaria vaccine candidates (Figure S2c). Surface charge characteristics measured via zeta (ζ) potential measurements showed consistent negative charge profiles (‐15 to ‐25 mV) across all vaccine formulations, with minor increases observed upon addition of Quil‐A adjuvant (Figure 1f). This surface charge range is favorable for interaction with antigen‐presenting cells and has been associated with enhanced immunogenicity [57]. The particle size distribution analysis and diameter measurements (Figure 1g; Figure S2d) demonstrated that the dual‐stage formulation Pfs47‐RTS,S‐BP had the largest diameter (∼0.8 µm) among unadjuvanted particles, likely due to the incorporation of both malaria antigens. When formulated with Quil‐A adjuvant, the diameter of Pfs47‐RTS,S‐BP+Q decreased to approximately 0.6 µm while maintaining a narrow size distribution, suggesting that adjuvant addition may contribute to enhanced dispersion stability. These particle sizes (0.3–0.6 µm) fall within the ideal range for efficient uptake by antigen‐presenting cells, particularly dendritic cells, which is crucial for initiating robust immune responses [36, 58, 59]. The polydispersity index, a measure of particle size distribution (Figure S2e), showed that BP formulations with Quil‐A adjuvant displayed reduced polydispersity indices (0.2–0.4) compared to unadjuvanted particles (0.3–0.6), indicating that adjuvant addition improved particle homogeneity, a factor known to influence vaccine performance [34].

The fusion protein identity was further confirmed by tryptic peptide fingerprinting using MALDI‐TOF/MS (Table S3). Notably, the yield of purified particles and antigen content per liter of culture (Table S4) demonstrated that the dual‐antigen Pfs47‐RTS,S‐BP vaccine produced the highest antigen yield (770.30±0.2 mg L⁻¹), suggesting cost‐effective manufacturability of this vaccine candidate. These results collectively established the successful development of BP‐based malaria vaccine candidates with defined physicochemical properties, incorporating pre‐erythrocytic‐and sexual‐stage targeting antigens.

2.2. Immune Activation by BP‐Based Malaria Vaccines

Before assessing immunostimulatory capacity, we first evaluated the safety profile of our vaccine formulations through cytotoxicity testing. The Alamar Blue reduction assay demonstrated that none of the BP‐based vaccine constructs exhibited significant cytotoxicity in either HEK‐293 cells or rat bone marrow‐derived cells (BMDCs) after 72 h of incubation when tested at antigen‐equivalent concentrations corresponding to 25–200 µg mL⁻¹ (Figure S3). This favorable safety profile, even at concentrations exceeding the therapeutic range, suggests the BP‐based vaccine has excellent biocompatibility with mammalian cells, consistent with the reported biocompatibility of polyhydroxyalkanoate‐based materials [60]. Importantly, the absence of cytotoxicity in rat BMDCs is particularly significant, as these immune progenitor cells are highly sensitive to toxic compounds [61] and represent an important indicator of potential systemic safety in vivo.

The in vitro immunostimulatory capacity of the vaccine formulations was evaluated using dendritic cell (DC) activation assays. Flow cytometry analysis showed that BP‐based vaccines significantly upregulated DC activation markers CD80, CD86, and MHCII compared to soluble antigens or controls (Figure 2a–c; Figures S4 and S5a), consistent with enhanced stimulation of innate immunity similar to that observed with other particulate antigen delivery systems [32, 34]. Although BP formulations carrying single‐antigen (RTS,S‐BP or Pfs47‐BP) enhanced DC activation compared to soluble antigens, the dual‐antigen Pfs47‐RTS,S‐BP elicited the strongest expression of DC activation markers and significantly higher levels of the proinflammatory cytokine IFN‐γ secretion (>500 pg mL⁻¹) (Figure 2d). This indicates a synergistic enhancement of innate immune activation resulting from the fusion of the individual antigens, highlighting its potential to more effectively prime adaptive immunity against malaria [62, 63].

FIGURE 2.

In vitro immune activation and T cell antigen presentation elicited by BP‐based malaria vaccine candidates. Flow cytometric quantification of (a) CD80‐positive (b) CD86‐positive and (c) MHCII‐positive DCs as a percentage of parent cells across all experimental groups after overnight in vitro treatment with the different vaccine formulations. The graphs display the percentage of positive cells for each marker across nine experimental conditions: Untreated (cells only/ negative control), RTS,S‐Soluble (10 µg mL^{− 1}), Pfs47‐Soluble (10 µg mL^{− 1}), BP alone (10 µg mL^{− 1}), ConA (5 µg mL^{− 1}; positive control), recombinant CSP (rCSP; 10 µg mL^{− 1}), RTS,S‐BP (10 µg mL^{− 1} antigen‐equivalent), Pfs47‐BP (10 µg mL^{− 1} antigen‐equivalent), and Pfs47‐RTS,S‐BP (10 µg mL^{− 1} antigen‐equivalent). Data are means of six replicates ± SEM, and p values were determined with ordinary two‐way ANOVA Dunnett's multiple comparisons test. (d) The concentration of the proinflammatory cytokine, IFN‐γ measured in cell culture supernatants following overnight in vitro treatment with the various vaccine formulations. Data are means of six replicates ± SEM, and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (e) The overlaid cell trace violet histograms demonstrate the proliferation profiles of CD4⁺ and CD8⁺ T cells. These T cells were co‐incubated in vitro for 5 days with DC pulsed with various vaccine formulations. Cell count is represented on the y‐axis and fluorescence intensity on the x‐axis. Each row corresponds to a different treatment condition, with the leftward shift in histogram peaks indicating increased T cell proliferation. The overlaid histograms shown are representative of six replicates per group. Flow cytometry analysis was used to determine the degree of activated T cell proliferation. (f) The proliferation of CD4+ and CD8+ T cells was quantified across all treatment groups, with the results presented as proliferation index. The proliferation indices were calculated using FlowJo's proliferation modeling tool (FlowJo v10.8 Software, BD Life Sciences). Data are means of six replicates ± SEM, and p values were determined with two‐way ANOVA Sidak's multiple comparisons test. (g) IFN‐γ production (pg mL⁻¹) by CD4⁺ and CD8⁺ T cells following in vitro stimulation with the various DC pulsed vaccine formulations for 2 days. The graph shows concentration measurements from cell culture supernatants. Data are means of six replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's multiple comparisons test. In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

T cell proliferation assays demonstrated that both CD4⁺ and CD8⁺ T cells showed robust proliferation when co‐cultured with DCs pulsed with BP‐based vaccines compared to those exposed to soluble antigens or controls (Figure 2e,f; Figure S5b), confirming the immunological advantages of particulate delivery systems [59, 64]. The dual‐antigen Pfs47‐RTS,S‐BP induced significantly higher proliferation indices in both T cell populations compared to either RTS,S‐BP or Pfs47‐BP alone (Figure 2f). This synergistic effect extended to functional capacity, with proliferating CD4⁺ and CD8⁺ T cells producing substantial IFN‐γ levels, with CD8⁺ T cells stimulated by Pfs47‐RTS,S‐BP‐pulsed DCs secreting the highest levels (∼750 pg mL⁻¹) (Figure 2g), indicating a cooperative immunological interplay between the two antigens that potentiates cellular immunity beyond what either antigen could achieve independently. This highlights a potent enhancement of malaria‐specific effector T cell responses critical for malaria protection [7, 65].

2.3. Temperature Stability of Vaccine Formulations

Thermostability of all vaccine constructs was evaluated over 28 days at 4°C, 25°C, 37°C, and 50°C, following WHO guidelines [66], to evaluate suitability for deployment in malaria‐endemic regions with limited cold chain infrastructure [67, 68]. Additionally, the thermostability and immunogenicity of Pfs47‐RTS,S‐BP were assessed in suspension (S) and freeze–dried (FD) formats under the same conditions. Based on the temperature stability profiles, immunogenicity of 37°C‐treated S and FD formulations was evaluated in Wistar rats (n = 8 per group) using a 10 µg antigen‐equivalent dose in a prime‐boost‐boost regimen (days 0, 14, and 28), followed by blood collection (days 0, 21, 29, and 42) and euthanization on day 42, allowing comprehensive assessment of vaccine‐induced immune responses after temperature treatment (Figure S6a). SDS‐PAGE confirmed preserved protein integrity at 88.4 kDa for both formats after 28 days (Figure S6b), and densitometric analysis was used to quantify the antigen concentrations for precise in vivo evaluation (Figure S6c).

Both S and FD formats of Pfs47‐RTS,S‐BP demonstrated exceptional thermal stability, maintaining protein integrity at 4°C, 25°C, and 37°C for 28 days, with only slight degradation at 50°C, as evidenced by minimal changes in the SDS‐PAGE profile; however, a slightly greater degree of protein degradation was observed in the FD form compared to the S form (Figure 3a,b (left panel)). Densitometric analysis of protein band intensity confirmed this stability (Figure 3a,b (right panel)), which exceeds many conventional protein‐based vaccines requiring strict cold chain maintenance [67, 69].

FIGURE 3.

Comparative thermostability and immunological assessment of suspension (S) versus freeze–dried (FD) malaria vaccine candidate Pfs‐47‐RTS,S‐BP over a 28‐day period. SDS‐PAGE analysis and corresponding protein band intensity quantification of (a) suspension and (b) freeze–dried Pfs‐47‐RTS,S‐BP formulations stored at different temperatures (4°C, 25°C, 37°C, and 50°C) for 7, 14, 21, and 28 days. Gel images (left panels) show the protein profiles, while quantitative analysis (right panels) demonstrates retention of protein integrity under various storage conditions. Samples were analyzed in triplicates with band intensity quantified using Image Lab software. Data are means of triplicates ± SEM, and P values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (c) RTS,S‐specific and (d) Pfs‐47‐ specific total IgG antibody titres tested against the serum of 37°C‐treated suspension and freeze–dried Pfs‐47‐RTS,S‐BP groups collected at days 21 and 42 post‐immunization. Data are means of eight or ten biological replicates ± SEM, and p values were determined with two‐way ANOVA Tukey's multiple comparisons test. (e) RTS,S‐specific and (f) Pfs‐47‐ specific total IgM antibody titres tested against the serum of 37°C‐treated suspension and freeze–dried Pfs‐47‐RTS,S‐BP groups collected at days 21 and 42 post‐immunization. Data are means of eight or ten biological replicates ± SEM, and p values were determined with two‐way ANOVA Tukey's multiple comparisons test. Flow cytometric assessment showing the proportions of CD4⁺ and CD8⁺ memory T cell subsets (resident memory (RM), central memory (CM), and effector memory (EM)) in the (g) spleen, and (h) liver at day 42 post‐immunization across the temperature‐ treated S and FD forms of Pfs‐47‐RTS,S‐BP. Data are means of four or eight biological replicates ± SEM and, p values were determined with two‐way ANOVA Tukey's multiple comparisons test. (i) Flow cytometric assessment showing the proportions of B memory cell subsets (class‐switched, conventional memory B cells (IgD⁻ IgG⁺) and non‐switched memory B cells (IgD⁻ IgM⁺)) in the spleen at day 42 across the temperature‐ treated S and FD forms of Pfs‐47‐RTS,S‐BP. Data are means of four or eight biological replicates ± SEM, and p values were determined with two‐way ANOVA Tukey's multiple comparisons test. In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

Immunogenicity assessments confirmed that both freeze–dried Pfs47‐RTS,S‐BP stored at 37°C (FD/37°C) and suspension stored at 37°C (S/37°C) BP elicited high RTS,S‐ and Pfs47‐specific IgG levels at days 21 and 42, with no significant difference between them (Figure 3c,d). However, the cold‐stored suspension (S/4°C) group showed significantly higher of RTS,S‐ and Pfs47‐specific IgG titres than the FD/37°C and S/37°C groups at both days 21 and 42. IgM levels were comparable across all groups (Figure 3e,f), while IgE remained low (Figure S7a,b), indicating preserved early humoral responses and minimal allergic sensitization risk.

Serum IFN‐γ levels measured on day 29 post‐immunization showed that S/37°C maintained IFN‐γ levels similar to S/4°C, indicating preserved systemic Th1‐type responses. In contrast, FD/37°C induced significantly lower IFN‐γ, suggesting that both freeze‐drying and heat exposure induced impairment of T cell activation capacity (Figure S7c). Splenocytes from FD/37°C and S/37°C groups showed reduced antigen‐specific IFN‐γ upon peptide restimulation compared to S/4°C (Figure S7d). Flow cytometry on day 42 revealed comparable CD4⁺ and CD8⁺ memory T cell subsets (central memory (CM; CD44⁺CD62L⁺), effector memory (EM; CD44⁺CD62L⁻), and resident memory (RM; CD69⁺CD44⁺)) in spleen and liver and memory B cells (class‐switched (IgD⁻IgG⁺) and non‐switched (IgD⁻IgM⁺)) in spleen for FD/37°C and S/37°C groups (Figure 3g–i). However, their percentages were significantly or moderately lower than S/4°C, suggesting thermal stress impacts memory cell development or maintenance.

Antigenicity, assessed by epitope‐specific IgG binding, remained consistent up to 37°C, with a significant decline only at 50°C (Figures S8a and S9a). IgG endpoint titres also remained stable, indicating preserved immunoreactivity across temperature conditions.

Quantitative analysis of particle size distribution revealed distinct differences at baseline. FD formulation had a slightly larger initial particle size than the S form, likely due to partial aggregation induced during the freeze‐drying process. The FD form maintained size stability across temperatures, while the S form showed progressive size increase at 37°C and 50°C, indicating temperature‐induced aggregation (Figures S8b and S9b). Zeta potential remained relatively consistent for both formats across all conditions, reflecting stable colloidal and thermal properties (Figures S8c and S9c).

The S format of single‐antigen RTS,S‐BP and Pfs47‐BP vaccines showed similar stability profiles (Figures S10 and S11). Among all three formulations, the dual‐antigen Pfs47‐RTS,S‐BP exhibited the most favorable temperature stability, particularly at 37°C, essential for deployment in malaria‐endemic regions. This enhanced stability may result from mutual stabilization of the two antigens on BPs and protein stabilization through molecular crowding [70]. These findings highlight the dual‐antigen BP‐based vaccine's potential to withstand short‐term temperature excursions during transport and storage in resource‐limited settings [67, 68].

2.4. Humoral Immune Responses Induced by BP‐based Malaria Vaccines

To comprehensively assess both the immediate immunogenicity and long‐term durability of immune responses to our vaccine ca ndidates, we implemented a carefully structured in vivo study design (Figure 4a). Wistar rats (n = 14 per group) received a three‐dose immunization regimen consisting of a prime dose on day 0, followed by two boost doses on days 14 and 28. Our study included a comprehensive set of vaccine formulations to allow direct comparison between BP‐based and soluble antigen approaches, as well as to assess potential synergistic effects between PfS47 and RTS,S. The experimental groups received one of the following formulations: 25 µg Quil‐A adjuvant alone (Q), 10 µg of BP formulated with 25 µg of Quil‐A (BP+Q), 10 µg antigen‐equivalent doses of BP‐based and soluble peptide vaccine formulations, all formulated with 25 µg Quil‐A (RTS,S‐BP+Q, Pfs‐47‐BP+Q, Pfs‐47‐RTS,S‐BP+Q, RTS,S‐Soluble+Q, Pfs‐47‐Soluble+Q, and Pfs‐47‐RTS,S‐Soluble+Q), and 10 µg of recombinant CSP protein with 25 µg of Quil‐A (rCSP+Q) as a reference control.

FIGURE 4.

Antigen‐specific antibody responses and their longevity after vaccination with various malaria vaccine candidates. (a) Schematic of detailed experimental design and timeline. Wistar rats (n = 14) were immunized with 10 µg of rCSP formulated with 25 µg of Quil‐A (rCSP+Q), 10 µg of BP formulated with 25 µg of Quil‐A (BP+Q), and BP‐based vaccine formulations administered at antigen‐equivalent doses of 10 µg for each construct: RTS,S‐BP, Pfs47‐BP, and Pfs47‐RTS,S‐BP, all formulated with 25 µg of Quil‐A (RTS,S‐BP+Q, Pfs‐47‐BP+Q, and Pfs‐47‐RTS,S‐BP+Q, respectively). The animals were then boosted twice at 2‐week intervals at days 14 and 28, and blood samples were collected (indicated by red droplets) at intervals throughout the study duration. The investigation was structured in two phases: an initial immunogenicity assessment (days 0–42) with 4 animals euthanized at day 42, followed by an extended longevity study (days 42–196) with the remaining 10 animals euthanized at day 196 to evaluate long‐term immune persistence. (b) ELISA analyses presenting the comprehensive kinetics of RTS,S‐specific total IgG antibody responses and their durability tested against the serum of all five groups collected at eight sequential timepoints (days 21, 42, 56, 84, 112, 140, 168, and 196). Antibody levels were compared to those at day 42 to assess durability, and differences between Pfs47‐RTS,S‐BP+Q and RTS,S‐BP+Q at various time points were evaluated. Data are means of 10 biological replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's multiple comparisons test. (c) ELISA analyses showing the Pfs‐47‐specific total IgG antibody responses and their durability tested against the serum of all five groups collected at regular intervals (days 21‐196). Antibody levels were compared to those at day 42 to assess durability, and differences between Pfs47‐RTS,S‐BP+Q and Pfs‐47‐BP at various time points were evaluated. Data are means of 10 biological replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's multiple comparisons test. (d) ELISA analyses were conducted to profile the early RTS,S‐specific IgM antibody responses in serum samples collected at days 21 and 42, capturing the initial wave of antibody production before isotype switching takes place. Data are means of 10 biological replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's and Sidak's multiple comparisons tests. (e) ELISA analyses examining the early Pfs‐47‐specific IgM antibody responses in serum samples collected on days 21 and 42 across Pfs‐47‐BP+Q and Pfs‐47‐RTS,S‐BP+Q vaccine variants. Data are means of 10 biological replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's and Sidak's multiple comparisons tests. (f) Assessment of RTS,S‐specific IgE antibody levels in serum collected on day 29 using ELISA analyses. Data are means of 10 biological replicates ± SEM, and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (g) Pfs‐47‐specific IgE antibody levels in serum collected on day 29 using ELISA analyses. Data are means of 10 biological replicates ± SEM, and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (h) Quantification of cellular immune activation by measuring IFN‐γ concentrations in serum samples collected on day 29 across five distinct vaccine formulations (BP+Q, RTS,S‐BP+Q, Pfs‐47‐BP+Q, Pfs‐47‐RTS,S‐BP+Q, and rCSP+Q). Data are means of ten biological replicates ± SEM, and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

The study was designed in two interconnected phases: an initial immunogenicity assessment phase (days 0–42) and an extended longevity phase (days 42–196). This study design allows for evaluation of both immediate immune responses and their long‐term persistence, addressing a critical limitation in previous malaria vaccine studies [10, 49]. Blood samples were collected at predetermined timepoints (indicated by red droplets in Figure 4a) on days 0, 21, 29, 42, 56, 84, 112, 140, 168, and 196 to track the development, peak, and potential waning of immune responses. This sequential sampling approach allowed us to monitor antibody kinetics continuously throughout the entire six‐month post‐immunization period. To balance comprehensive immunological assessment with longitudinal monitoring, we employed a split‐cohort strategy where four animals from each group were euthanized at day 42 (the anticipated peak of immune response) for detailed analysis of cellular responses in lymphoid tissues. The remaining ten animals per group continued in the study until day 196, allowing us to evaluate the persistence of both humoral and cellular immunity over an extended period. This experimental design enabled us to address several critical questions about our malaria vaccine candidates: (1) the ability to induce robust initial immune responses, (2) the comparative immunogenicity of BP‐based versus soluble antigen formulations, (3) the potential advantages of dual‐antigen versus single‐antigen approaches, (4) the durability of antibody responses over a six‐month period, and (5) the establishment and maintenance of memory T and B cell populations critical for long‐term protection against malaria.

The comparison between BP‐based vaccines (BP+Q, RTS,S‐BP+Q, Pfs47‐BP+Q, and Pfs47‐RTS,S‐BP+Q) and soluble antigen formulations (RTS,S‐Soluble+Q, Pfs47‐Soluble+Q, and Pfs47‐RTS,S‐Soluble+Q) revealed striking differences in antibody production and persistence (Figure 4; Figure S12). ELISA analyses revealed that BP‐based formulations (RTS,S‐BP+Q and Pfs47‐RTS,S‐BP+Q) induced significantly higher RTS,S‐specific IgG titres compared to soluble antigen formulations (Figure 4b; Figure S12a). The kinetics of antibody responses showed an early detection at day 21 and a peak at day 42 (post third immunization), followed by a gradual decline but with substantial titres maintained even at day 196, suggesting long‐term humoral immunity. This pattern of antibody kinetics with sustained levels is highly favorable compared to the rapid waning of antibody responses observed with the RTS,S/AS01 vaccine in clinical trials [10, 25]. Similar patterns were observed for Pfs47‐specific total IgG antibodies, with both Pfs47‐BP+Q and Pfs47‐RTS,S‐BP+Q vaccines inducing robust and persistent antibody responses compared to soluble counterparts (Figure 4c; Figure S12b). The dual‐antigen Pfs‐47‐RTS,S‐BP+Q vaccine elicited robust and sustained total IgG responses against RTS,S, and Pfs‐47, which remained high throughout the 196‐day study period. Notably, these responses were significantly higher compared to the single‐antigen formulations (RTS,S‐BP+Q and Pfs‐47‐BP+Q) at multiple timepoints (days 21–196) (Figure 4b,c), demonstrating a synergistic immunogenicity without any antigenic interference.

BP‐based vaccines induced strong early IgM responses (Figure 4d,e) and moderate IgE production (Figure 4f,g) against RTS,S, and Pfs47, indicating balanced Th1 and Th2 immune activation conducive to malaria immunity [71, 72]. In contrast, soluble antigen formulations showed minimal or undetectable IgM and IgE responses, which declined rapidly (Figure S12c–f). The dual‐antigen Pfs47‐RTS,S‐BP+Q vaccine generated significantly higher RTS,S‐ and Pfs47‐specific IgM responses than single‐antigen formulations, suggesting enhanced immune recognition and processing due to synergistic epitope presentation. Moderate IgE levels at day 29 further supported the vaccine's favorable safety profile. Compared to soluble antigens, BP‐based vaccines consistently demonstrated superior antibody titers, breadth, and persistence, particularly for the dual‐antigen Pfs47‐RTS,S‐BP+Q, which maintained strong responses against pre‐erythrocytic—and sexual‐stage antigens.

Additionally, BP‐based vaccines induced significantly higher IFN‐γ levels (∼300–400 pg mL⁻¹) on day 29 than soluble formulations (∼20 to 80 pg mL⁻¹) (Figure 4h; Figure S12g), with the dual‐antigen vaccine achieving the highest IFN‐γ response. This robust IFN‐γ production, critical for liver‐stage protection [63, 73], highlights comprehensive immune engagement and demonstrates the synergistic effect of combining RTS,S, and Pfs47 epitopes on the BP platform.

2.5. Cytokine Profiling of Recall Responses

To assess the functional capacity of vaccine‐induced memory responses, we performed ex vivo restimulation of splenocytes with various malaria antigens. At day 42 post‐vaccination, splenocytes from animals immunized with BP‐based vaccines produced significantly higher levels of multiple cytokines compared to those from animals receiving soluble antigens or controls (Figures S13 and S14). These cytokines included pro‐inflammatory cytokines (IFN‐γ, TNF‐α, IL‐1β, IL‐6, IL‐12p70, IL‐18, and IL‐17), T cell growth factor (IL‐2), Th2‐associated cytokines (IL‐4, IL‐5), regulatory cytokine (IL‐10), and chemokine (MCP‐1), indicating a balanced immune activation profile. This diverse cytokine response is characteristic of effective malaria immunity, which requires both Th1 and Th2 components [63, 71].

Remarkably, this enhanced cytokine production capacity was maintained until day 196, with splenocytes from BP‐vaccinated animals retaining their ability to produce elevated levels of multiple cytokines upon antigen re‐encounter (Figures S13 and S14). This sustained functionality indicates the successful establishment of long‐lived memory cells capable of rapid reactivation, a critical feature for effective malaria vaccination strategies.

The dual‐stage Pfs47‐RTS,S‐BP vaccine elicited particularly robust and durable cytokine responses, suggesting superior functional immune memory establishment and potential synergistic effects between Pfs47 and RTS,S antigens. This addresses a critical limitation of previous malaria vaccine candidates, which often showed rapid waning of functional immunity [10, 49].

The balanced cytokine profile observed across different functional categories suggests that our BP‐based vaccines elicited coordinated activation of both humoral and cellular immune responses, indicating an effective engagement of the complementary components of the immune system. IFN‐γ and TNF‐α levels were particularly elevated in response to BP‐based vaccines containing RTS,S, aligning with their known roles in protecting against liver‐stage malaria [63, 73]. Similarly, robust IL‐4 production in Pfs47‐containing formulations supports Th2 activation essential for transmission‐blocking immunity [22, 74]. These findings demonstrate the ability of BP‐based vaccines to establish long‐lived, multifunctional immune memory, addressing a critical limitation of current malaria vaccines.

2.6. Cellular Immune Memory Development

Flow cytometric analysis revealed the development and persistence of memory T cell populations in circulation following immunization with the various vaccine formulations. We systematically tracked both CD4⁺ and CD8⁺ T cell memory subsets throughout the 196‐day study period, enabling comprehensive assessment of memory T cell formation, contraction, and long‐term maintenance. The BP‐based vaccine candidates induced robust and durable memory T cell responses in peripheral blood. CD4⁺ T_EM cells were detected as early as day 42 post‐ immunization and reached peak levels of total circulating CD4⁺ T cells followed by a contraction phase and then a relatively stable maintenance phase, characteristic of a typical T cell memory response to vaccination. Similarly, circulatory CD8⁺ T_EM cells also showed significant expansion. The dual‐antigen Pfs47‐RTS,S‐BP+Q vaccine consistently elicited the highest proportion of both CD4⁺ and CD8⁺ T_EM cells, with significantly elevated levels maintained from day 42 through day 196 post‐immunization compared to other formulations (Figure 5a–d; Figures S15a–d and S19a). This pattern was consistent across all BP‐based vaccine formulations, though with quantitative differences. CD4⁺ and CD8⁺ T_CM cells, critical for long‐term protection and rapid secondary responses, followed similar kinetics but were present at different proportions within the T cell population, indicating the expected maturation of the memory T cell compartment. In contrast, BP alone and soluble antigen formulations induced significantly lower or undetectable levels of memory T cells (Figure S16a–e).

FIGURE 5.

Cellular immune responses induced by distinct BP‐based malaria vaccine formulations: longitudinal analysis of circulatory and resident CD4⁺ and CD8⁺ T cell memory subsets in blood, spleen, and liver. Flow cytometric analyses demonstrating the longitudinal kinetics of circulating CD4⁺ and CD8⁺ T cell memory populations, including effector memory (EM) and central memory (CM), in blood samples from Wistar rats immunized with (a) RTS,S‐BP+Q, (b) Pfs‐47‐BP+Q, (c) Pfs‐47‐RTS,S‐BP+Q, and (d) rCSP+Q, respectively. T cell memory responses were assessed in peripheral blood mononuclear cells (PBMCs) collected at days 42, 56, 84, 112, 140, 168, and 196 post‐immunizations, and the responses were compared to day 42. Flow cytometric assessment showing the proportions of CD4⁺ and CD8⁺ memory T cell subsets (resident memory (RM), central memory (CM), and effector memory (EM)) in the spleen at (e) day 42 and (f) day 196 across the different vaccine formulations (BP+Q, RTS,S‐BP+Q, Pfs‐47‐BP+Q, Pfs‐47‐RTS,S‐BP+Q, and rCSP+Q. Flow cytometric assessment depicting the proportions of CD4⁺ and CD8⁺ memory T cell subsets (RM, CM, and EM) in the liver at (g) day 42 and (h), day 196 across the different vaccine formulations. Flow cytometric quantification of T cell functionality through proliferation index measurements of T cells isolated from the liver at (i) day 42 and (j) day 196. T cells were isolated from Wistar rats immunized with different vaccine formulations, and their proliferation index was measured in response to stimulation with various antigens, including untreated (negative control), QuilA (Q) (adjuvant control), BP (vaccine platform), CpG (positive control), and T‐epitope (malaria‐specific T cell epitope (IQNSLSTEW)) derived from liver‐stage infection), evaluating antigen‐specific T cell responses in the liver. The proliferation indices were calculated using FlowJo's proliferation modeling tool (FlowJo v10.8 Software, BD Life Sciences). Data are means of four or ten biological replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's multiple comparisons test. In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

Memory T_RM cells, essential for providing frontline defense at potential sites of pathogen entry or replication [75, 76], were extensively characterized in both secondary lymphoid organ spleen and the liver, a critical site for malaria parasite development. In the spleen, we identified substantial populations of CD4⁺ and CD8⁺ T_RM cells at both day 42 and day 196 timepoints, with a notable increase in CD4⁺ and CD8⁺ T_RM frequencies at day 196 compared to day 42 (Figure 5e,f; Figure S17a,d). This increase likely reflects the progressive establishment of tissue‐resident memory compartments following the initial expansion and contraction phases, consistent with the maturation of protective immunity. The dual‐antigen Pfs47‐RTS,S‐BP+Q vaccine induced the highest percentages of splenic CD4⁺ and CD8⁺ T_RM cells at both timepoints (Figure 5e,f). Similar patterns were observed in liver tissue, a critical site for protection against pre‐erythrocytic stage malaria [65, 77], with significant populations of CD4⁺ and CD8⁺ T_RM detected throughout the study period and enhanced accumulation by day 196 (Figure 5g,h; Figure S17g,j). This liver‐specific enrichment of T_RM cells over time suggests preferential retention and potential local proliferation of antigen‐specific T cells at this crucial site of parasite development. Notably, parallel increases in CD4⁺ and CD8⁺ T_CM and T_EM populations were also observed in both spleen and liver at day 196 (Figure 5e–h; Figures S17 b,c,e,f,h,i,k,l and Figure S19b), indicating coordinated development of diverse memory T cell subsets across different tissue compartments. In contrast, soluble antigen formulations induced significantly lower or undetectable levels of splenic and liver‐resident memory T cells across all subsets (T_RM, T_CM, and T_EM) (Figure S16f–i).

To confirm the functional capacity of liver T cells, we performed ex vivo restimulation assays with T cells isolated from liver tissue at days 42 and 196. T cell were restimulated with T epitope (IQNSLSTEW; MHC‐I) derived from liver‐stage parasites. T cells from BP‐based vaccine recipients showed robust proliferative capacity upon stimulation with cognate antigens, with proliferation indices of 2‐3 at day 42 (Figure 5i). This functional responsiveness, while significantly reduced compared to day 42 levels, was maintained at day 196 (Figure 5j; Figure S17m), confirming the establishment of durable and functional tissue‐resident memory, suggesting the potential for long‐term protection against hepatic stages of malaria infection [77, 78]. Notably, the dual‐stage Pfs47‐RTS,S‐BP+Q vaccine elicited the most robust proliferative capacity at both early and late timepoints (Figure 5i,j), suggesting potential synergistic effects between the two antigens in establishing functional liver‐resident T cell immunity. In contrast, T cells from recipients of soluble antigen formulations showed minimal to undetectable proliferation upon restimulation (Figures S5b, S16j,k and S18).

Memory B cell responses were also comprehensively characterized. Flow cytometric analysis of B cell memory populations in blood and spleen revealed that BP‐based vaccines induced significant development of memory B cells. While circulating memory B cells, including class‐switched (IgD⁻IgG1⁺) and non‐switched (IgD⁻IgM⁺) subsets, gradually declined over time in blood, there was evidence of substantial proportions maintained at day 196 (Figure 6a–d). The dual‐antigen Pfs47‐RTS,S‐BP+Q formulation consistently generated the highest frequencies of both memory B cell subsets in circulation throughout the study period (Figure S20a,b).

FIGURE 6.

Longitudinal analysis of B cell memory subsets in blood and spleen following immunization with various BP‐based malaria vaccine candidates. Flow cytometric evaluations showing longitudinal kinetics of the B cell memory subsets generated by different malaria vaccine formulations, including class‐switched, conventional memory B cells (IgD⁻ IgG⁺) and non‐switched memory B cells (IgD⁻ IgM⁺), in circulation (blood) of Wistar rats immunized with (a) RTS,S‐BP+Q, (b) Pfs‐47‐BP+Q, (c) Pfs‐47‐RTS,S‐BP+Q, and (d) rCSP+Q, respectively. B‐cell memory responses were assessed in peripheral blood mononuclear cells collected at days 42, 56, 84, 112, 140, 168, and 196 post‐immunizations, and the responses were compared to day 42. Flow cytometric assessment showing the proportions of B memory cell subsets (IgD⁻IgG1⁺ and IgD⁻IgM⁺) in the secondary lymphoid organ (spleen) at (e) day 42, and (f) day 196 across all vaccine formulations. Data are means of four or ten biological replicates ± SEM, and p values were determined with two‐way ANOVA Dunnett's multiple comparisons test. In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

Importantly, memory B cells continued to develop and accumulate in the spleen through day 196, with the Pfs47‐RTS,S‐BP+Q formulation inducing the highest frequencies of splenic memory B cells at both days 42 and 196 (Figure 6e,f), consistent with the known dynamics of memory B cell compartmentalization compartmentalization [79, 80]. Remarkably, these elevated frequencies were maintained and remained unchanged at day 196 (Figure S20c,d), highlighting the durability of vaccine‐induced B cell memory in secondary lymphoid organs. The presence of a reservoir of memory B cells in these organs suggests a capacity for rapid recall responses upon parasite exposure, even without high circulating antibody levels. Soluble antigen formulations induced significantly lower or undetectable levels of memory B cell responses in blood and spleen (Figures S21 and S22).

The persistence of functional memory T and B cells observed at day 196 strongly suggests the establishment of long‐lived immunological memory capable of mediating protection against parasite challenge long after initial vaccination. Notably, the Pfs47‐RTS,S‐BP+Q group exhibited particularly elevated levels of memory B and T cells, indicating superior establishment of high‐quality memory by this dual‐antigen formulation, likely due to synergistic effects between the Pfs47 and RTS,S epitopes. This finding is particularly significant in the context of malaria vaccine development, where durable protection has been a major challenge [10, 49].

2.7. Functional Evaluation of Vaccine‐Induced Antibody Responses

To investigate the functional activity of antibodies generated by our BP vaccine candidates, we evaluated their ability to inhibit sporozoite motility and block parasite transmission, two critical mechanisms for malaria protection. Sporozoite motility assays, performed using the SMOOT [54], revealed striking differences in parasite movement patterns when exposed to sera from different vaccination groups. Image processing of microscopic videos showed extensive motility tracks (yellow trajectories) in the soluble peptides and negative control groups, while markedly reduced movement was observed in sera from animals immunized with RTS,S‐BP+Q and Pfs47‐RTS,S‐BP+Q, approaching the level of inhibition seen with the 2A10 monoclonal antibody positive control (Figure 7a). Quantitative analysis of sporozoite inhibition demonstrated that sera from animals immunized with Pfs47‐RTS,S‐BP+Q achieved the highest proportion of stationary sporozoites (∼75%), followed by RTS,S‐BP+Q (∼60%), both significantly outperforming BP+Q control (∼30%) and soluble antigen formulations (Figure 7b). This pronounced immobilization effect represents a key protective mechanism, as sporozoite motility is essential for hepatocyte invasion and subsequent liver‐stage infection [81, 82]. Velocity distribution analysis provided further insights into the inhibitory effects, with sera from BP‐based vaccine groups (RTS,S‐BP+Q and Pfs47‐RTS,S‐BP+Q) shifting the sporozoite velocity profiles toward lower speeds. While negative control or soluble peptide sera exposed sporozoites exhibited a wide distribution of velocities, sporozoites exposed to Pfs47‐RTS,S‐BP+Q sera showed markedly reduced speeds with narrower distributions, indicating substantial functional impairment (Figure S23). This significant reduction in motility demonstrates the potent neutralizing capacity of antibodies generated by our dual‐antigen vaccine approach.

FIGURE 7.

In vitro functional evaluation of vaccine‐induced antibodies against sporozoite motility and transmission of Plasmodium falciparum. The figure demonstrating the functional activity of antibodies generated by various malaria vaccine formulations, highlighting their ability to inhibit malaria infection and transmission. (a) Image processing of microscopic videos to analyze motility patterns of sporozoites gliding on a glass surface. Sporozoites were incubated with medium (negative control), 15 ng µL⁻¹ humanized 2A10 antibody (positive control), and pooled sera collected on day 42 from Wistar rats immunized with BP+Q, Pfs‐47‐BP+Q, Quil‐A, Pfs‐47‐Soluble+Q, RTS,S‐Soluble+Q, Pfs‐47‐RTS,S‐Soluble+Q, rCSP+Q, RTS,S‐BP+Q, and Pfs‐47‐RTS,S‐BP+Q. The sporozoite tracks were color‐coded to indicate their velocity using a colour gradient: purple (low velocity) to yellow (high velocity). The distinctive yellow trajectories illustrate the movement of individual sporozoites, with substantially reduced motility visible in vaccine groups, RTS,S‐BP+Q and Pfs‐47‐RTS,S‐BP+Q compared to the extensive movement observed in other groups. (b) Quantitative analysis of sporozoite inhibition, displaying the proportion of stationary (coloured) versus motile (white) sporozoites for each vaccine formulation. Pooled sera from each group were used, and data are means of two replicates ± SEM. (c) Evaluation of transmission‐blocking potential of vaccine‐induced antibodies against P. falciparum using Standard Membrane Feeding Assay (SMFA). Each data point represents oocyst counts in individual mosquitoes fed on infectious blood meals containing pooled sera collected on day 42 from Wistar rats immunized with different vaccination groups. Oocyst counts were compared to negative control and Pfs‐47‐RTS,S‐BP+Q. Data are means of a range of biological replicates ± SEM, and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (d) The table summarizing the key SMFA metrics across all vaccine formulations, including infection prevalence (infected/fed mosquitoes ratio), parasite burden (mean oocyst counts), transmission‐reducing activity (TRA%), and transmission‐blocking activity (TBA%). In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

To assess the transmission‐blocking potential of induced antibodies, we performed standard membrane feeding assays (SMFA) using P. falciparum. Mosquitoes fed on blood meals containing parasites mixed with sera from animals vaccinated with Pfs47‐BP+Q and Pfs47‐RTS,S‐BP+Q showed significantly reduced oocyst development compared to the negative control (Figure 7c). Individual oocyst counts revealed that while negative control mosquitoes harbored high parasite numbers (mean count 41.44), those exposed to Pfs47‐BP+Q and Pfs47‐RTS,S‐BP+Q sera demonstrated dramatically reduced oocysts levels (mean count 17.3 and 8.04, respectively). Comprehensive analysis of transmission parameters across all vaccine formulations revealed that the dual‐antigen Pfs47‐RTS,S‐BP+Q vaccine achieved the highest transmission‐reducing activity (TRA) at 80.6% and transmission‐blocking activity (TBA) at 46.18% (Figure 7d). This substantially exceeded the performance of single‐antigen formulation Pfs47‐BP+Q (TRA 58.25%, TBA 24.16%), BP+Q, and soluble Pfs47 formulations. Notably, the dual‐stage Pfs47‐RTS,S‐BP+Q vaccine reached or outperformed the transmission‐blocking efficacy of the 4B7 positive control monoclonal antibody, highlighting the exceptional potency of our vaccine‐induced antibodies. This dual‐antigen formulation also displayed significantly reduced infection prevalence, with only 22 mosquitoes infected out of 50 fed (44%), compared to 112/137 (81.8%) in the negative control group.

These functional assay results demonstrated that antibodies induced by our dual‐antigen Pfs47‐RTS,S‐BP vaccine effectively inhibit two critical processes in the malaria life cycle: sporozoite motility (essential for liver infection) and sexual‐stage development in mosquitoes (required for transmission). This dual functionality of our dual‐antigen vaccine demonstrates enhanced protection by targeting both the pre‐erythrocytic (RTS,S) and sexual (Pfs47) stages of the malaria parasite, offering broader protection than single‐antigen approaches.

2.8. In Vivo Protection Studies

To evaluate the protective efficacy of our proposed dual‐antigen vaccine candidate, we conducted challenge studies in female C57BL/6 mice following a prime‐boost‐boost immunization regimen (days 0, 14, and 28) with either a 10 µg antigen‐equivalent dose of Pfs47‐RTS,S‐BP formulated with 25 µg Quil‐A (Pfs47‐RTS,S‐BP+Q) or 10 µg of BP formulated with 25 µg Quil‐A (BP+Q; control) (Figure 8a). The experimental design included both direct sporozoite challenge and mosquito bite challenge to assess protection against liver‐stage infection, followed by monitoring of blood‐stage parasitemia to evaluate levels of protection. Before challenge studies, we confirmed the identity of or vaccine constructs (Figure S24a,b).

FIGURE 8.

In vivo evaluation of dual‐antigen malaria vaccine efficacy in C57BL/6 mice against liver infection and parasite transmission. (a) Schematic representation of the experimental challenge design and timeline. C57BL/6 female mice were immunized with 10 µg of BP formulated with 25 µg of Quil‐A (BP+Q) and 10 µg of antigen‐equivalent dose of Pfs‐47‐RTS,S‐BP formulated with 25 µg of Quil‐A (Pfs‐47‐RTS,S‐BP+Q). Following a prime‐boost‐boost immunization regimen (days 0, 14, 21), vaccinated mice were challenged either intravenously (IV) with sporozoites or through mosquito bites on day 42. Liver burden measurements were taken on day 44 post‐vaccination. Parasitemia was evaluated in mosquito bite challenged mice from days 46 to 52. Blood samples were collected (indicated by red droplets) at predetermined intervals. (b) Measurements of liver infection, expressed as total flux in photons per second, were obtained following sporozoite IV challenge. The data shows individual mouse values for the BP+Q, Pfs‐47‐RTS,S‐BP+Q, naïve infected, and naïve non‐infected control groups. The naïve non‐infected control group was used to detect the background level in the assay. Data are means of eight biological replicates ± SEM, and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (c) The bar graph displays the calculated percent inhibition of liver infection following sporozoite IV challenge for the BP+Q and Pfs‐47‐RTS,S‐BP+Q vaccination formulations. (d, e) The similar data for mosquito bite challenge are shown, where Pfs‐47‐RTS,S‐BP+Q achieved 80.4% inhibition compared to ‐34.3% for BP+Q. (f) ELISA titration curves showing the RTS,S specific total IgG antibody titres measured by ELISA for sporozoite IV challenge experimental group vaccinated with Pfs‐47‐RTS,S‐BP+Q. Data are means of eight biological replicates ± SEM. The sera collected on day 41 from mice immunized with Pfs‐47‐RTS,S‐BP+Q was used in ELISA analyses. (g) The table presents detailed quantification of RTS,S specific total IgG antibody responses, displaying individual 2A10 equivalence values for each mouse following sporozoite IV challenge. Data are means of eight biological replicates ± SEM. (h) ELISA titration curves showing the RTS,S specific total IgG antibody titres measured by ELISA for mosquito bite challenge experimental group vaccinated with Pfs‐47‐RTS,S‐BP+Q. Data are means of eight biological replicates ± SEM. The sera collected on day 41 from mice immunized with Pfs‐47‐RTS,S‐BP+Q was used in ELISA analyses. (i) The table presents detailed quantification of RTS,S specific total IgG antibody responses, displaying individual 2A10 equivalence values for each mouse following mosquito bite challenge. Data are means of eight biological replicates ± SEM. j, The graph showing the transmission‐blocking efficacy of dual‐antigen Pfs‐47‐RTS,S‐BP+Q through SMFA, comparing negative control, 4B7 (positive control), and BP+Q, Each data point represents individual oocyst counts in mosquitoes and data are means of a range of biological replicates ± SEM. Oocyst counts were compared to negative and positive controls and p values were determined with ordinary one‐way ANOVA Dunnett's multiple comparisons test. (k) A summary table with mean oocyst counts and TRA% across all the groups. In all panels with p values: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and ns, not significant.

Following intravenous sporozoite challenge on day 42, we measured parasite liver burden by in vivo imaging of luciferase activity on day 44 post‐vaccination. Mice vaccinated with Pfs47‐RTS,S‐BP+Q showed significantly reduced parasite burden compared to naïve infected controls (Figure 8b). Quantitative analysis revealed 63.6% inhibition of liver infection in Pfs47‐RTS,S‐BP+Q‐vaccinated mice compared to only 7.5% inhibition in BP+Q controls (Figure 8c). This substantial reduction in liver‐stage parasite development demonstrates the potent pre‐erythrocytic protection conferred by the dual‐antigen vaccine. More impressively, following challenge via mosquito bites, which more closely mimics natural infection conditions, the Pfs47‐RTS,S‐BP+Q vaccine achieved 80.4% inhibition of liver infection compared to −34.3% for the BP+Q control (Figure 8d,e). This marked improvement in protection under physiologically relevant challenge conditions highlights the vaccine's potential efficacy in natural transmission settings. The superior protection observed in the mosquito bite challenge compared to direct sporozoite injection suggests that vaccine‐induced antibodies may be particularly effective at neutralizing sporozoites during their natural migration from the skin to the liver, consistent with observations in other pre‐erythrocytic vaccine studies [77, 82].

Monitoring of parasitemia in mosquito bite‐challenged mice revealed that 25% (2/8) of mice vaccinated with Pfs47‐RTS,S‐BP+Q were completely protected from blood‐stage infection, demonstrating the potential for sterile immunity. Whereas all mice in the BP+Q and naïve infected groups developed patent parasitemia (Figure S25a). This achievement of sterile protection represents a significant advancement, as it indicates the vaccine's capacity to completely prevent malaria infection rather than merely reducing disease severity. This synergistic effect between pre‐erythrocytic and sexual‐stage components illustrates how targeting multiple parasite life stages can enhance overall vaccine efficacy beyond what would be expected from either component alone.

To assess potential correlates of protection, we measured antigen‐specific antibody responses in individual mice vaccinated with Pfs‐47‐RTS,S‐BP+Q. ELISA titration curves for RTS,S‐specific antibodies from sporozoite challenge mice (#556‐563) and mosquito bite challenge mice (#580‐587) showed robust responses across all vaccinated animals (Figure 8f,h). Quantification of these responses as 2A10 monoclonal antibody equivalents, revealed high RTS,S‐specific antibody titres, with an average of 557.70 ± 285.43 µg mL⁻¹ in the sporozoite challenge group and 488.71 ± 170.86 µg mL⁻¹ in the mosquito bite challenge group (Figure 8g,i). Importantly, both groups significantly exceeded the established protective thresholds from RTS,S/AS01 benchmark studies. The sporozoite challenge group showed a 5.3‐fold increase over the 105 µg mL⁻¹ level required for 50% liver infection reduction, while the mosquito bite challenge group surpassed the more stringent 285 µg mL⁻¹ required for 50% sterile protection by 1.7‐fold [83, 84, 85]. This quantitative association between elevated antibody titers and robust protection highlights the superior immunogenicity of the dual‐antigen approach, positioning Pfs47‐RTS,S‐BP+Q as a promising next‐generation malaria vaccine candidate with the potential to outperform the current gold standard.

To evaluate the simultaneous induction of transmission‐blocking antibodies, we SMFA using blood containing P. falciparum gametocytes mixed with sera from immunized mice used in the challenge study. Analysis of oocyst counts in individual mosquitoes revealed a significant reduction in parasite development with sera from Pfs47‐RTS,S‐BP+Q‐vaccinated animals compared to negative controls (Figure 8j). Quantification showed that while negative control mosquitoes harboured high parasite burdens (mean oocyst count 18.26), those exposed to sera from 4B7 monoclonal anti‐Pfs25 antibody positive control and Pfs47‐RTS,S‐BP+Q groups showed significantly reduced oocyst number, with mean counts of 3.33 and 5.81, respectively (Figure 8k). These results translate to a transmission‐reducing activity (TRA) of 81.74%, 37.12%, and 68.15% for 4B7, BP+Q, and Pfs47‐RTS,S‐BP+Q, respectively. The transmission‐blocking effect of the dual‐antigen Pfs47‐RTS,S‐BP+Q vaccine, approaching that of the established 4B7 monoclonal antibody, demonstrated its potent activity against the sexual stages of the parasite within the mosquito vector, comparable to other leading transmission‐blocking vaccine candidates [9, 74].

Furthermore, antibody analysis demonstrated high Pfs47‐specific IgG levels in Pfs47‐RTS,S‐BP+Q vaccinated mice, both from sporozoite challenge (#556‐563) and mosquito bite challenge (#580‐587) groups, indicating the potential of the Pfs47‐RTS,S‐BP+Q vaccine candidate to elicit strong transmission‐blocking antibody responses in vivo (Figure S25b,c).

Collectively, these in vivo protection studies established that our dual‐antigen Pfs47‐RTS,S‐BP vaccine confers significant protection against both liver‐stage infection and parasite transmission, providing comprehensive targeting of two critical phases of the P. falciparum life cycle and supporting its potential as a dual‐stage malaria vaccine candidate. The level of protection observed compares favourably with other advanced malaria vaccine candidates [12, 16], with the added advantage of transmission‐blocking activity that could contribute to community‐wide malaria control [6, 22]. Additionally, the synergistic targeting of multiple stages of the parasite life cycle represents a significant advancement in malaria vaccine development, offering a more comprehensive approach to disease control than traditional single‐stage vaccines.

3. Discussion

The development of an effective vaccine against malaria remains one of the most pressing challenges in global public health. Our study offers an innovative approach utilizing BPs as a versatile platform for delivering malaria antigens that target pre‐erythrocytic and sexual stages of the parasite life cycle. The findings demonstrate that this technology has a number of competitive advantages over conventional vaccine development approaches.

The dual‐antigen Pfs47‐RTS,S‐BP vaccine exhibited remarkable immunological properties that address key limitations of previous malaria vaccine candidates. The BP‐based delivery system induced significantly stronger and more durable immune responses compared to their soluble counterparts. Robust antibody titres maintained for over six months and functional memory T and B cells persisting in circulation, lymphoid organs, and liver tissue, demonstrating enhanced immunogenicity in both humoral and cellular responses. Notably, the establishment of tissue‐resident memory T cells in the liver is particularly crucial, as this represents a first line of defense at the site of initial parasite replication.

A significant advantage of our dual‐antigen vaccine approach is the synergistic targeting of multiple parasite life stages, which creates a comprehensive protective strategy that surpasses what could be achieved by targeting individual stages separately. This synergism is evident in the superior performance of the Pfs47‐RTS,S‐BP formulation compared to single‐antigen vaccines, with the dual‐antigen construct demonstrating enhanced efficacy in both pre‐erythrocytic protection and transmission blocking. By simultaneously targeting sporozoite invasion of hepatocytes through RTS,S‐induced antibodies and blocking parasite development in mosquitoes through Pfs47‐mediated immunity, our vaccine creates a two‐pronged attack that addresses both the cause of clinical disease and the cycle of transmission.

A key challenge in malaria vaccine development has been the rapid waning of vaccine‐induced immunity, as seen with the RTS,S/AS01 vaccine in clinical use [12]. Our longitudinal assessment showed that the BP‐based vaccines, particularly the dual‐antigen formulation, retained significant antibody titres even at day 196 post‐immunization. This extended maintenance of high antibody levels may be attributed to the particulate nature of the BP delivery system, which efficiently targets antigen‐presenting cells and generates a depot effect for continuous antigen presentation [86, 87]. Furthermore, the physical characteristics of our particles (0.3‐0.6 µm diameter) fall within the optimal range for dendritic cell uptake, which may contribute to their ability to enhance immunogenicity of attached antigens [58].

Another advantage of our dual‐antigen approach is the induction of a balanced immune activation profile, which includes the production of Th1, Th2, and Th17‐associated cytokines. This balanced response promotes both cellular immunity, which is necessary for targeting liver‐stage parasites, and humoral immunity, which is required for inhibiting sporozoite liver traversal and transmission blocking [88, 89, 90].

Here, we incorporated quantitative functional antibody evaluation by using the SMOOT software, which enabled for precise, objective analysis of sporozoite motility inhibition. Traditional assessments of antibody efficacy often rely on subjective observation or semi‐quantitative measures, limiting reproducibility and comparability across studies. By implementing SMOOT, we were able to rigorously quantify the degree of sporozoite immobilization and velocity reduction, demonstrating that antibodies induced by the Pfs47‐RTS,S‐BP vaccine potently impair sporozoite motility, a critical step for preventing liver infection. This adds an additional layer of validation to our immunogenicity data and underscores the importance of including functional assessments in vaccine evaluation. Furthermore, functional assays confirmed that vaccine‐induced antibodies also significantly reduced oocyst development in mosquitoes, establishing a dual mechanism of protection that targets both liver infection and mosquito borne transmission. This dual‐stage efficacy reinforces the potential of our vaccine to contribute meaningfully to malaria elimination strategies.

The thermostability of our vaccine constructs provides a considerable practical advantage for deployment in resource‐constrained tropical environments where maintaining cold chain infrastructure is challenging. The exceptional retention of protein integrity and immunoreactivity in both suspension and freeze–dried formats of Pfs47‐RTS,S‐BP at temperatures up to 37°C for 28 days suggests that these formulations could potentially withstand temperature excursions during transportation and storage. This attribute is especially valuable for malaria vaccines, which must be distributed in endemic regions with limited refrigeration infrastructure.

Another important consideration is that all Plasmodium falciparum strains used in this study were laboratory‐standardized strains. While these provide controlled and reproducible conditions for evaluating immunogenicity and functional activity, they do not capture the extensive antigenic polymorphism found in circulating field isolates. As CSP and Pfs47 exhibit sequence variation across endemic regions, evaluating the breadth of protection against genetically diverse parasite strains will be important for determining the field applicability of the Pfs47‐RTS,S‐BP vaccine.

Our challenge studies revealed compelling evidence of vaccine efficacy, with the dual‐antigen Pfs47‐RTS,S‐BP formulation providing significant protection against both direct sporozoite challenge (63.6% inhibition) and mosquito bite challenge (80.4% inhibition). The higher protection observed in the mosquito bite model shows that vaccine‐induced immune responses may be particularly effective against naturally delivered sporozoites, which must pass through many tissue barriers before reaching the liver. Furthermore, achieving sterile protection in 25% of vaccinated animals represents a significant achievement, as complete prevention of blood‐stage infection is the ideal outcome for pre‐erythrocytic vaccines.

These protective outcomes are underpinned by remarkably robust antibody responses that far exceed current vaccine benchmarks. When quantified using 2A10 monoclonal antibody equivalents, our dual‐antigen vaccine elicited RTS,S‐specific titres represent substantial improvements over established RTS,S/AS01 protective thresholds, with 5.3‐fold and 1.7‐fold enhancements above the critical protective levels. The strong correlation between these elevated antibody responses and enhanced protective efficacy validates the immunological superiority of our dual‐antigen BP platform over conventional approaches.

The TRA of 68.15% obtained with our dual‐antigen vaccine is comparable to the efficacy of established monoclonal antibodies, highlighting its potential impact on parasite transmission dynamics. The synergistic effect of combining pre‐erythrocytic and transmission‐blocking components is particularly important as it creates a dual benefit, protecting vaccinated individuals from disease while simultaneously reducing community transmission, potentially benefiting even unvaccinated individuals.

The bioengineered E. coli production platform has various advantages for vaccine manufacturing, including scalability, cost‐effectiveness, and consistent product quality. The high yield of purified Pfs47‐RTS,S‐BP (770.30 ± 0.2 mg L⁻¹) making this approach economically viable for large‐scale manufacturing. Furthermore, the bacterial expression system eliminates the requirement for complex eukaryotic cell culture facilities, potentially enabling regional manufacturing in malaria‐endemic nations.

Despite these promising results, several limitations should be acknowledged. The challenge studies were conducted in mouse models, which may not accurately represent human immune responses or parasite‐host interactions. Future studies should also assess the vaccine's performance under genetically diverse and epidemiologically relevant conditions, including variations in parasite biology and host immunity encountered in real‐world transmission settings. Additionally, while we demonstrated antibody persistence and memory cell maintenance for 196 days, longer‐term studies are needed to determine the ultimate durability of protection. The potential requirement for booster doses beyond six months remains to be established, and further optimization of antigen dose, adjuvant formulation, and immunization schedules could potentially enhance protective efficacy. Evaluation of alternative epitopes or additional antigens targeting blood‐stage parasites might also broaden the protective scope of the vaccine.

Furthermore, the current study evaluated vaccine performance exclusively in healthy adult rodents. However, malaria disproportionately affects vulnerable populations such as infants, young children, pregnant women, and immunocompromised individuals, whose immune responses may differ markedly from those of healthy adults. Evaluating safety and immunogenicity in model systems that reflect these physiological states will therefore be important for future clinical development.

We also acknowledge that the temperature‐treated vaccine formulations evaluated in our thermostability study were not used in the in vivo challenge experiment, which was intentionally designed to measure peak protective efficacy using freshly prepared vaccine. Although thermostability immunogenicity data showed that strong antibody and T cell responses were preserved after 28 days of storage at 37°C, we did not functionally evaluate whether these antibodies retained their ability to inhibit pre‐erythrocytic infection or block transmission at day 42. Similarly, we did not assess whether antibodies collected at day 196 maintained these functional properties. These evaluations represent logical next steps to bridge thermostability findings and the immune response durability data with real‐world protective performance under field‐relevant temperature conditions.

Finally, while the day 42 challenge model is the standard time point for assessing peak vaccine‐mediated protection in murine malaria models, a delayed challenge at later time points would further determine whether the observed durable immune responses correlate with protection against infection and inhibition of transmission. Our current data justify the future set‐up of such a long term evaluation of functional immune responses toward vaccine development.

Despite these limitations, our findings represent a significant advancement in malaria vaccine research. The synergistic approach of combining pre‐erythrocytic and transmission‐blocking antigens on a single BP platform creates a vaccine candidate with multifaceted protective mechanisms that collectively enhance overall efficacy beyond what could be achieved by individual components. This synergism coupled with enhanced immunogenicity, protective immunity, transmission blocking, thermostability, and manufacturing advantages, positions the dual‐antigen Pfs47‐RTS,S‐BP vaccine as a promising candidate for further development. This strategy aligns with the WHO Malaria Vaccine Technology Roadmap's goal of developing vaccines that have at least 75% efficacy against clinical malaria and transmission‐blocking capabilities.

4. Conclusion

Our study demonstrates that the BP‐based dual‐antigen malaria vaccine, targeting both pre‐erythrocytic stage infection and mosquito borne transmission, represents a significant advancement that addresses multiple challenges in malaria vaccine development. By synergistically combining pre‐erythrocytic and sexual‐stage antigens on a single particulate platform, our Pfs47‐RTS,S‐BP vaccine fills a critical gap lacking in current vaccine candidates, which typically focus on a single‐stage of the parasite's complex life cycle. The vaccine's superior performance is evidenced by RTS,S‐specific antibody levels that exceed current RTS,S/AS01 protective benchmarks by 5.3‐fold and 1.7‐fold, demonstrating enhanced immunogenicity that translates into robust protective efficacy. This vaccine has the potential to provide effective and long‐lasting protection against this devastating global disease. Additionally, the enhanced thermostability of our formulations addresses practical deployment challenges in tropical malaria‐endemic regions with limited cold chain infrastructure. Its cost‐effective production in an E. coli and scalable manufacturing design facilitates affordable distribution, while enabling rapid production to meet global demand. These promising findings warrant further development and progression toward clinical evaluation, with the goal of contributing to crucial malaria control and elimination efforts worldwide.

5. Experimental Section

5.1. Cells and Reagents

Human embryonic kidney cells 293T (HEK293T, ATCC) were cultured in Dulbecco's Modified Eagle Medium (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA), 2 mm GlutaMAX (Gibco, USA), 1 mm Sodium Pyruvate (Gibco, USA), 10 mm N‐2‐hydroxyethylpiperazine‐N‐2‐ethane sulfonic acid (HEPES; Gibco, USA), 100 U mL⁻¹ Penicillin–Streptomycin (Gibco, USA), 1x MEM Non‐Essential Amino Acids Solution (Gibco, USA). The cells were maintained in an incubator at 37°C with 5% CO₂.

5.2. Plasmid DNA Constructs, Bacterial Strains, Primers, and Growth Conditions

The bacterial strains and plasmids employed in this investigation are detailed in Table S2. The selected P. falciparum epitope sequences were optimized for codon usage in E. coli and synthesized by Integrated DNA Technologies (IDT, Coralville, IA, USA), then utilized for subsequent cloning. Primers used for sequencing were procured from IDT and are listed in Table S2.

The plasmid pET14b‐PhaC, which encodes the PHB synthase (PhaC; BP anchor), was used to construct the plasmids pET14b‐RTS,S‐BP, pET14b‐Pfs‐47‐BP, and pET14b‐Pfs‐47‐RTS,S‐BP. These plasmids encode P. falciparum‐derived epitopes, which were utilized in this study for the development of malaria vaccines. All cloning procedures were performed in the E. coli Top10 strain (Invitrogen, USA). For the production of BP, an endotoxin‐free E. coli BL21 (DE3) (Lucigen, Middleton, WI, USA) strain containing the pMCS69 plasmid was employed.

For plasmid propagation, the E. coli Top10 strain was cultured in Luria broth (LB; Difco, Detroit, MI) supplemented with the appropriate selection antibiotic, 100 µg mL⁻¹ ampicillin (ThermoFisher Scientific, USA) at 37°C and 200 rpm. For the production of BP, an endotoxin‐free variant of E. coli BL21 (DE3) was grown in LB miller medium (LB; Difco, Detroit, MI) containing 100 µg mL⁻¹ ampicillin and 50 µg mL⁻¹ chloramphenicol (Glentham Life Sciences, UK), also at 37°C and 200 rpm.

5.3. Antigen or Epitope Selection

The antigens or epitopes utilized in this investigation were identified from previously published research, selected based on their experimentally verified immunogenic properties and relevance to P. falciparum‐induced malaria. The list of these antigens or epitopes and their sources is provided in Table S1.

5.4. Molecular Cloning for the Production of BPs

Cloning techniques were carried out as previously described [91]. The DNA constructs purchased from IDT were subjected to restriction digestion using enzymes such as XbaI, NotI, BamHI, and XhoI (New England Biolabs, USA), depending on the gene location. The target DNA fragments were then separated via agarose gel electrophoresis with GelRed (Biotium, USA) staining, followed by DNA gel purification (New England Biolabs, USA). The purified DNA fragments were ligated using T4 DNA ligase (New England Biolabs, USA) into the pET14b vector to generate the plasmid constructs: pET14b‐RTS,S‐BP, pET14b‐Pfs‐47‐BP, and pET14b‐Pfs‐47‐RTS,S‐BP. The target DNA sequence of the successfully generated plasmids was confirmed by the Griffith University DNA Sequencing Facility (Griffith University, Australia). The confirmed plasmids were then transformed into the chemically competent production host, E. coli BL21, for the respective BP production.

5.5. Production and Purification of BP Vaccines

For the production of BPs, LB miller medium was supplemented with 100 µg mL⁻¹ ampicillin, 50 µg mL⁻¹ chloramphenicol and 1% w/v glucose (Chem‐Supply, Australia). This medium was inoculated with a 20 mL overnight culture of E. coli BL21 (DE3) harboring the pMCS69 plasmid, which encodes the PhaA and PhaB enzymes, as well as the respective pET14b plasmid encoding either just the BP or the fusion proteins, including RTS,S‐BP, Pfs‐47‐BP, and Pfs‐47‐RTS,S‐BP. The inoculated flasks were then incubated at 37°C and 200 rpm for 3 h, after which 1 mM isopropyl β‐D‐1‐thiogalactopyranoside (IPTG; Glentham Life Sciences, UK) was added, and the cultures were allowed to grow for an additional 48 h at 25°C and 200 rpm. The cells were harvested by centrifugation at 8000 × g for 20 min at 4°C and mechanically disrupted using a Microfluidizer M‐110P (Microfluidics, USA). The disrupted cell lysate was then centrifuged at 8000 × g for 20 min at 4°C to sediment the BPs. The sedimented BPs were washed and purified as previously described [91, 92, 93, 94]. The purified BPs were sterilized with 10 µg mL⁻¹ amphotericin B (Sigma‐Aldrich, USA) and 1 mg mL⁻¹ ciprofloxacin hydrochloride hydrate (Thermo Scientific Chemicals, USA), and then washed three times with 1x Tris‐buffered saline (TBS; pH 7.5). The sterile BPs were stored at 4°C in 1x TBS (pH 7.5) until they were formulated for animal trials and analysis.

5.6. Characterization of BP Vaccines

BPs were detected and analyzed by sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE), as described elsewhere [95, 96]. The concentration of the antigen and the percentage of the fusion protein relative to the total protein in the particle fraction were determined by densitometry (Image Lab Software, Bio‐Rad Laboratories, USA), comparing the protein band intensities to a standard curve generated from known concentrations of bovine serum albumin (BSA; 62.5–500 ng), as previously described [97]. The target protein bands were excised from the gel, and their identity as RTS,S‐BP, Pfs‐47‐BP, and Pfs‐47‐RTS,S‐BP constructs was confirmed using Q‐TOF/MS at the University of Queensland's Mass Spectrometry Facility (Queensland, Australia).

The morphology of purified BPs and whole cells was examined using a Hitachi HT7700 transmission electron microscope (TEM) to confirm the production, shape, and size of BPs, as described previously [95]. The size distribution and surface charge (Zeta potential/ ζ) of purified BPs were measured using Litesizer500 (Anton Paar, Australia) at 25°C. All measurements were performed in triplicate. The purified BPs were diluted 1000‐fold in Tris buffer (pH 7), sonicated for 2 min and the particle size was evaluated by dynamic light scattering (DLS) analysis. The zeta potential was determined by electrophoretic light scattering coupled with phase analysis light scattering. The polyhydroxybutyrate (PHB) content analysis of BPs was performed as previously described [98, 99]. Briefly, the purified BPs and whole cells were freeze–dried and weighed. The dry BPs were boiled in 1 mL of concentrated sulfuric acid for one hour at 90°C to hydrolyze the PHB to crotonic acid. The samples were cooled, diluted with water, and filtered using a 0.22 µm filter (Merk Millipore). The diluted samples (1:500) were analyzed for PHB concentration by high‐performance liquid chromatography (HPLC) using an Agilent 1200 HPLC system. Crotonic acid was used as a standard, and the PHB content (%w/w) was calculated based on the amount of crotonic acid produced in the samples.

5.7. Freeze‐Drying of BPs

BPs were rapidly frozen in liquid nitrogen and subsequently vacuum‐dried at −52°C for 12 h to remove both bound and unbound water molecules, resulting in freeze–dried BPs (FD‐BPs). Protein profile retention was assessed using SDS‐PAGE, with untreated BPs serving as controls.

5.8. Thermostability Assessment of BPs

To assess the thermostability of the sterile BPs, they were exposed to various temperatures (4, 25, 37, and 50°C) for 28 days. Subsequently, the changes in particle size distribution and zeta potential were measured using a Litesizer500, with all measurements conducted in triplicate. Furthermore, the variations in protein profiles and band intensities were determined through SDS‐PAGE analysis to evaluate the retention of antigens. Additionally, an enzyme‐linked immunosorbent assay (ELISA) was performed to assess the preservation of antigenicity after the temperature treatment. In the ELISA, the temperature‐treated BPs were coated at a concentration of 5 µg mL⁻¹ and tested against serum samples from immunized Wistar rats with the highest antibody titres against the corresponding BP‐based vaccines. The ELISA procedure is described in a section below, and the final antibody titres were represented as endpoint titres.

5.9. Animal Trial‐ Immunological Assessment of 37°C‐Treated Suspension and Freeze–Dried Format of Pfs‐47‐RTS,S‐BP

Adult female Wistar rats weighing between 180–220 grams were obtained from the Animal Resources Centre (Perth, Australia) and acclimated for two weeks before the study interventions. The rats were housed in groups of four per cage in a well‐ventilated animal facility, with access to rat pellets and water ad libitum. Eight animals were assigned to each experimental group. The animal procedures and experiments for the immunogenicity study were approved by the Animal Ethics Committee at Griffith University, Queensland, Australia (Ethics number: GRIDD/06/22/AEC). This study was conducted in compliance with the recommendations outlined in the Guide for the Care and Use of Laboratory Animals of the Griffith University Animal Ethics Committee.

To evaluate the immunogenicity of the 37°C‐treated suspension (S) and freeze–dried (FD) formats of Pfs‐47‐RTS,S‐BP, the rats were administered intramuscular injections in the right and left calf muscles (100 µL in each) using a 10 µg antigen‐equivalent dose per immunization. All groups received a prime immunization followed by two booster vaccinations (days 0, 14, and 28), and blood samples were collected on days 0, 21, 29, and 42 via tail vein. The serum was separated from the blood and stored at −80°C until further analysis. On day 42, rats from each group were humanely euthanized following blood collection via cardiac puncture, and their tissues, including the spleen and liver, were collected for immune response analysis. Antigen‐specific antibody responses (total IgG, IgM, and IgE) were measured using ELISA, as detailed in a subsequent section. The proportions of CD4⁺ and CD8⁺ memory T and B cell subsets in the spleen and liver were analyzed using flow cytometry, with tissue processing and flow cytometric procedures also described later. IFN‐γ levels were assessed in serum collected on day 29 and in restimulated splenocytes harvested on day 42. IFN‐γ concentrations were measured using a commercial Rat IFN‐gamma Quantikine ELISA Kit (R&D systems, Hong Kong), following the manufacturer's instructions. Absorbance was recorded at 450 nm using a BioTek Synergy 2 microplate reader (BioTek, USA).

5.10. In Vitro DC Activation

Wistar rats were euthanized, and their bone marrow was harvested from the femurs and tibias. The immune cells from the bone marrow were isolated as previously described [100]. Red blood cells (RBCs) were lysed using an RBC Lysis Buffer (Invitrogen, USA). The remaining cells were collected following centrifugation. To produce immature Bone Marrow‐Derived Dendritic Cells (iBMDCs), the bone marrow cells were resuspended in complete Roswell Park Memorial Institute1640 medium (RPMI; Gibco, USA) supplemented with 2 mm GlutaMAX, 100 U mL⁻¹ Penicillin–Streptomycin, 10% FBS, 20 ng mL⁻¹ rat GM‐CSF (BioLegend, San Diego, CA, USA) and 5 ng mL⁻¹ rat IL‐4 (BD Biosciences, USA), and incubated at 37°C with 5% CO₂ for 7 days. On day 8, the iBMDCs were collected, seeded in 24‐well plates, and activated by incubating them with 10 ug mL⁻¹ CpG ODN Class C 2395 (InvivoGen, USA) at 37°C with 5% CO₂ for 24 h to obtain mature BMDCs (mBMDCs). Excess CpG was removed by washing with 1x PBS, pH 7.5, twice. On day 9, 1 × 10⁶ mBMDCs were stimulated in 1 mL of complete RPMI 1640 media supplemented with 10 ug mL⁻¹ antigen‐equivalent BPs or 10 ug mL⁻¹ soluble peptides (ChinaPeptides, Shanghai, China) or 10 ug mL⁻¹ recombinant circumsporozoite protein (rCSP; Creative Diagnostics, USA) or 5 ug mL⁻¹ Concanavalin A (Con A; Invitrogen, USA) or left unstimulated (cells only), and incubated overnight at 37°C with 5% CO₂. On day 10, the supernatants were collected and frozen at −80°C for cytokine analysis. The concentration of IFN‐γ in the supernatants was measured using a Rat IFN gamma SimpleStep ELISA Kit (Abcam, UK), and the results were analyzed with a BioTek microplate reader.

The stimulated and unstimulated BMDCs were collected and stained with 7‐Aminoactinomycin D (7‐AAD; Miltenyi Biotech, USA), following the manufacturer's instructions, to identify the live cells. The cells were then stained with the following fluorochrome‐conjugated antibodies for 30 min at 4°C: anti‐rat CD11c‐APC‐Vio770, CD80‐PE, CD86‐PE‐Vio770, and MHC‐II‐PerCP‐Vio700 to assess the maturation of the dendritic cells. After washing the cells twice with 1x PBS buffer, they were fixed with 1% paraformaldehyde (ThermoFisher Scientific, USA) and analyzed using a flow cytometer within 24 h. All the antibodies were purchased from Miltenyi Biotech, USA, and used according to the manufacturer's instructions. The data were analyzed using FlowJo v10.8 software (BD Life Sciences, USA).

5.11. In Vitro Antigen Presentation Assay

Wistar rats were euthanized, and their spleens were harvested to prepare a single‐cell suspension of splenocytes, as described previously [101]. This procedure was adapted from the protocol [101], which detailed spleen cell isolation in mice, and was performed based on our specific experimental needs. From the single‐cell splenic suspension, CD4⁺ T cells were enriched and isolated using a CD4⁺ T Cell Isolation Kit (Miltenyi Biotech, USA), while CD8+ T cells were enriched and isolated using a CD8⁺ T Cell Isolation Kit (Miltenyi Biotech, USA). The enriched CD4⁺ and CD8⁺ T cells were washed with 1x PBS, pH 7.5, supplemented with 0.1% (w/v) BSA. The T cells were then incubated with 5 µM Cell Trace Violet (CTV; Invitrogen) for 20 min at 37°C in the dark. After the incubation, the cells were washed twice in RPMI 1640 containing 2% FBS. Wistar rat mBMDCs were pulsed with BPs, soluble peptides, or rCSP overnight at 37°C with 5% CO₂, as described earlier. mBMDCs stimulated with ConA served as a positive control, while medium alone (unstimulated) was used as a negative control.

Purified CD4⁺ and CD8⁺ T cells labelled with CTV were co‐cultured with stimulated or unstimulated mBMDCs at BMDC‐to‐T cell ratio of 1:10 in 96‐well round‐bottom plates. The co‐cultures were maintained in complete RPMI‐1640 medium supplemented with 2 mm GlutaMAX, 10% FBS, 100 U mL⁻¹ Penicillin‐Streptomycin, and incubated for 5 days at 37°C with 5% CO₂. After 48 h, the culture supernatants were collected and analyzed for IFN‐γ levels using a Rat IFN gamma SimpleStep ELISA Kit (Abcam, UK). On day 6, the T cells were harvested, stained with 7‐AAD to exclude dead cells, and their proliferation was assessed by flow cytometry and CTV dilution. The samples were acquired on a MACSQuant Analyzer 16 flow cytometer (Miltenyi Biotech, USA), and FlowJo's proliferation modeling tool was used to determine the number of proliferating T cells and proliferation indices, thereby evaluating the in vitro antigen presentation by BP‐pulsed BMDCs to CD4⁺ and CD8⁺ T cells.

5.12. Cytotoxicity Assessment of BPs

HEK293T cells and Wistar rat BMDCs were used to determine the non‐cytotoxic concentrations of BPs and soluble peptides in vitro. Briefly, the cells were seeded in 96‐well plates at a density of 5 × 10³ cells per 100 µL and incubated at 37°C with 5% CO₂ for 24 h. After 24 h, the plates were washed three times with 1x PBS (pH 7.5), and the cells were treated with different antigen‐equivalent concentrations of BPs (ranging from 25 to 200 µg mL⁻¹), 100 µg mL⁻¹ of soluble peptides, or 100 µg mL⁻¹ of rCSP for 72 h at 37°C with 5% CO₂. Untreated cells (Cells only) and cells treated with 50% DMSO were used as the negative and positive controls, respectively. Following the incubation, the alamar blue reagent (Invitrogen, USA) was directly added to the wells at a final concentration of 10% (v/v) and incubated at 37°C with 5% CO₂ for 6 h. After the 6 h incubation, the absorbance was measured at 570 and 600 nm using a BioTek microplate reader.

5.13. Vaccine Formulation and Immunizations

In vivo evaluation of Immunogenicity and longevity: Vaccines were formulated with the adjuvant, 25 µg of Quil‐A (InvivoGen, USA), with 20 µg of antigen‐equivalent dose in a final volume of 200 µL in sterile 1x TBS at pH 7.5. Vaccines included in this study were 10 µg of BP formulated with 25 µg of Quil‐A (BP+Q), 10 µg of antigen‐equivalent BP‐based vaccine formulations, such as RTS,S‐BP, Pfs‐47‐BP, and Pfs‐47‐RTS,S‐BP, all formulated with 25 µg of Quil‐A (RTS,S‐BP+Q, Pfs‐47‐BP+Q, and Pfs‐47‐RTS,S‐BP+Q, respectively), soluble peptide counterparts, such as 10 µg of RTS,S‐Soluble, 10 µg of Pfs‐47‐Soluble, and 10 µg of Pfs‐47‐RTS,S‐Soluble, all formulated with 25 µg Quil‐A (RTS,S‐Soluble+Q, Pfs‐47‐Soluble+Q, and Pfs‐47‐RTS,S‐Soluble+Q, respectively), 10 µg of rCSP formulated with 25 µg of Quil‐A (rCSP+Q) and 25 µg Quil‐A (Q) alone.

In vivo evaluation of dual‐stage malaria vaccine efficacy: Vaccines were formulated with the adjuvant, 25 µg of Quil‐A, with 10 µg of antigen‐equivalent dose in a final volume of 100 µL in sterile 1x TBS at pH 7.5. Vaccines included in this study were 10 µg of BP formulated with 25 µg of Quil‐A (BP+Q) and 10 µg of antigen‐equivalent Pfs‐47‐RTS,S‐BP formulated with 25 µg of Quil‐A (Pfs‐47‐RTS,S‐BP+Q). The negative control cohort received injections of 100 µL 1x TBS on the same schedule (naïve).

5.14. Animal Trial‐ Immunogenicity and Longevity Studies

Adult female Wistar rats weighing between 180–220 grams were obtained from the Animal Resources Centre (Perth, Australia) and acclimated for two weeks prior to the study interventions. The rats were housed in groups of five per cage in a well‐ventilated animal facility, with access to rat pellets and water ad libitum. Fourteen animals were assigned to each experimental group. To evaluate the immunogenicity of the BPs, the rats were administered intramuscular injections in the right and left calf muscles (100 µL in each). All groups received a prime immunization followed by two booster vaccinations (days 0, 14, and 28), and blood samples were collected on days 0, 21, 29, and 42 via tail vein. The serum was separated from the blood and stored at −80°C until further analysis. On day 42, four rats from each group were humanely euthanized following blood collection via cardiac puncture, and their tissues, including the spleen and liver, were collected for immune response analysis.

To evaluate the longevity of the immune responses induced by the BP‐based vaccines, the remaining ten rats were maintained under the same conditions for 6 months (days 56 to 196). Blood samples were collected via tail vein at predetermined intervals (days 56, 84, 112, 140, 168, and 196) throughout this extended study period. The serum was separated from the blood and stored at −80°C until analysis. On day 196, all rats were humanely euthanized following blood collection via cardiac puncture, and their spleen and liver tissues were collected for further immune response analysis.

The animal procedures and experiments for the immunogenicity and longevity studies were approved by the Animal Ethics Committee at Griffith University, Queensland, Australia (Ethics number: GRIDD/06/22/AEC). This study was conducted in compliance with the recommendations outlined in the Guide for the Care and Use of Laboratory Animals of the Griffith University Animal Ethics Committee.

5.15. Serum Cytokine Analysis

Wistar rat serum samples collected on day 29 were analyzed for cytokine levels. Specifically, the concentration of IFN‐γ in the serum was determined using a commercial Rat IFN‐gamma SimpleStep ELISA Kit (Abcam, UK), following the manufacturer's protocol. The absorbance was measured at 450 nm using a BioTek microplate reader.

5.16. ELISA Using Purified BPs, Recombinant E1 and E2 Proteins

The Wistar rat serum samples collected on days 21, 42, 56, 84, 112, 140, 168, and 196 were analyzed using ELISA to quantify the total IgG levels against the soluble peptides RTS,S and Pfs‐47. Briefly, 100 µL of BP, RTS,S, and Pfs‐47 diluted in 1x PBS (pH 7.5) at a concentration of 5 µg mL⁻¹ were coated onto high‐binding ELISA plates (Greiner Bio‐One, Germany) overnight at 4°C. The plates were washed with 1x PBS containing 0.05% (v/v) Tween 20 (pH 7.5) and blocked with 3% skimmed milk (w/v) in PBST for 1 h at 25°C. After three washes with PBST, the individual Wistar rat samples were added at a 1:200 dilution, followed by two‐fold serial dilution, and incubated for 1 h at 25°C. Following three washes with PBST, HRP‐conjugated goat anti‐rat IgG H&L (Abcam, UK) diluted 1:10,000 in PBST was added as the secondary antibody and incubated for 1 h at 25°C. The plates were washed three times with PBST, and then o‐Phenylenediamine dihydrochloride substrate (SIGMAFAST OPD; Sigma‐Aldrich, USA) was added to the wells. The reaction was incubated at room temperature (RT) in the dark for signal development and then stopped by adding 50 µL of 2N H₂SO₄. The results were measured at 450 nm using a BioTek microplate reader.

The Wistar rat serum samples collected on days 21 and 42 were analyzed using ELISA to quantify the total IgM levels, while the samples collected on day 29 were analyzed by ELISA to measure the total IgE levels against the soluble peptides RTS,S and Pfs‐47. Mouse monoclonal [MARM‐4] Anti‐Rat IgM H&L‐HRP (Abcam, UK) and Goat anti‐Rat IgE‐HRP (Abcam, UK) were employed as the secondary antibodies, respectively. The serum titres were represented as endpoint titres.

5.17. Cytokine and Chemokine Analysis of Restimulated Splenocytes

Spleens were harvested from the euthanized Wistar rats on days 42 and 196 and placed in cold 1x PBS. Single‐cell suspensions were prepared by gently disrupting the tissue through a 70 µm cell strainer (Miltenyi Biotech, USA). RBCs were lysed using RBC Lysis Buffer for 5 min at RT. After washing, the splenocytes were resuspended in complete RPMI 1640 medium supplemented with 10% FBS, 2 mm L‐glutamine, 100 U mL⁻¹ Penicillin‐Streptomycin, 55 µm 2‐mercaptoethanol (Sigma‐Aldrich, USA), and 1 mm sodium pyruvate. Cell viability was assessed by trypan blue exclusion, and the concentration was adjusted to 2 × 10⁶ cells mL⁻¹.

Splenocytes were seeded at a density of 4 × 10⁵ cells per well in 96‐well plates and restimulated with either a specific antigen (10 µg mL⁻¹) or ConA (5 µg mL⁻¹) as a positive control, or medium alone as a negative control. The cells were incubated at 37°C in a humidified atmosphere containing 5% CO₂ for 48 h to analyze cytokine and chemokine secretion. The cell culture supernatants were collected after 48 h of restimulation and stored at ‐80°C until analysis. Cytokine and chemokine levels (IFN‐γ, TNF‐α IL‐12 p70, IL‐1β, IL‐2, IL‐4, IL‐5, IL‐6, IL‐10, IL‐17, IL‐8, and MCP‐1) were determined using the Bio‐Plex Pro Rat Cytokine 23‐Plex Assay kit (Bio‐Rad, USA) according to the manufacturer's instructions, and the results were analyzed with the Bio‐Plex 200 system (Bio‐Rad Laboratories, USA). Cytokine concentrations were calculated using standard curves generated with standards.

5.18. Preparation of Tissue Samples for Flow Cytometric Analysis

Isolation of Immune Cells from spleen: Spleens were harvested from the euthanized Wistar rats following vaccination on days 42 and 196 and placed in tubes containing MACS Tissue Storage Solution (Miltenyi Biotech, USA). The spleen samples were filtered through 70 µm cell strainers and gently disrupted using a syringe plunger into a new 15 mL falcon tube. The RBCs were lysed using RBC Lysis Buffer, and the remaining cells were collected after centrifugation at 300 × g for 10 min at RT.

Isolation of Immune Cells from liver: Livers were harvested from euthanized Wistar rats following immunization on days 42 and 196. The rats were euthanized, and their abdominal cavities were opened. The livers were then perfused with 10 mL of cold 1x PBS through the portal vein to remove circulating blood cells. The livers were carefully excised, placed in a petri dish containing cold RPMI‐1640 medium, and minced into small pieces using sterile scissors. The liver tissue was transferred to a 50 mL conical tube containing 10 mL of digestion buffer (RPMI‐1640 supplemented with 0.05% collagenase IV, 0.002% DNase I, and 1% FBS) and incubated at 37°C for 30 min with gentle agitation. After digestion, the suspension was passed through a 100 µm cell strainer (Sigma‐Aldrich, USA), and the cells were washed twice with cold RPMI‐1640 containing 5% FBS by centrifugation at 400 × g for 5 min at 4°C. The cell pellet was resuspended in 10 mL of 40% Percoll solution (Sigma‐Aldrich, USA) in 1x PBS and gently overlaid onto 5 mL of 70% Percoll solution in a 15 mL conical tube. The gradient was centrifuged at 750 × g for 20 min at room temperature with no brake. Immune cells were collected from the interface between the 40% and 70% Percoll layers using a sterile Pasteur pipette. The harvested cells were washed twice with cold 1x PBS containing 2% FBS. RBC contamination was removed using RBC lysis buffer for 3 min at RT. After a final wash, the cells were resuspended in complete RPMI‐1640 medium supplemented with 10% FBS, 2 mM L‐glutamine, 100 U mL⁻¹ Penicillin‐Streptomycin, and 50 µm 2‐mercaptoethanol. Cell viability was assessed by trypan blue exclusion, and the concentration was adjusted to 2 × 10⁶ viable cells ml⁻¹ for subsequent analyses. The procedure was performed as previously described [102] with minor changes based on the specific experimental needs.

Isolation of Peripheral Blood Mononuclear Cells (PBMCs) from Blood: Blood samples were collected from Wistar rats on days 42, 56, 84, 112, 140, and 168 via tail vein bleeds, and on day 196 via cardiac puncture under terminal anesthesia with isoflurane (Mediquip, Australia). The blood from each rat was drawn using a 25‐gauge needle into tubes containing sodium heparin as an anticoagulant (Becton Dickinson, USA). The collected blood was diluted 1:1 with sterile 1x PBS at RT. In 15 mL conical tubes, 3 mL of Histopaque‐1077 (Sigma‐Aldrich, USA) was added, and the diluted blood was carefully layered on top of the Histopaque using a Pasteur pipette, taking care not to disrupt the interface. Samples were then centrifuged at 400 × g for 30 min at RT with the centrifuge brake turned off.

After centrifugation, the plasma layer was carefully aspirated to within 0.5 cm of the PBMC layer. The PBMC layer was collected using a Pasteur pipette and transferred to a new 15 mL conical tube. The collected cells were washed twice with 10 mL of 1x PBS by centrifugation at 300 × g for 10 min at RT. To remove residual RBCs, the cell pellet was resuspended in 2 mL of RBC lysis buffer and incubated for 3 min at RT. The lysis was stopped by adding 8 mL of 1x PBS, and the cells were centrifuged at 300 × g for 5 min. The supernatant was discarded, and the cell pellet was resuspended in complete RPMI‐1640 medium supplemented with 10% FBS, 2 mm L‐glutamine, and 100 U mL⁻¹ penicillin‐Streptomycin. Cell viability was assessed using trypan blue exclusion, and cell concentration was determined using a hemocytometer. The isolated PBMCs were either used immediately for downstream applications or cryopreserved in freezing medium (90% FBS, 10% DMSO) and stored in liquid nitrogen. The procedure was carried out as described in the previous study [103], with minor adjustments to meet our specific experimental requirements and in accordance with the manufacturer's guidelines for Histopaque‐1077.

5.19. Flow Cytometric Staining and Immunophenotypic Analysis

To determine the circulatory and tissue‐resident T memory cells, single‐cell suspensions from Wistar rat spleen and immune cells from Wistar rat liver were stained with 7‐AAD to assess cell viability. After washing with 1x PBS, the cells were incubated with rat‐specific Fc block (anti‐rat CD32/CD16 antibody, clone D34‐485) diluted according to the manufacturer's instructions for 15 min at 4°C to minimize non‐specific Fc‐mediated binding. The cells were incubated with the following antibody cocktail for 30 min at 4°C in the dark: anti‐rat CD3‐Viobright720, CD4‐APC‐Vio770, CD8‐Viogreen, CD25‐PE, CD44‐PerCP‐Vio700, and CD62L‐FITC. Following this, cells were washed twice with 1× PBS, then incubated with the unconjugated rabbit anti‐CD69 primary antibody diluted appropriately in staining buffer, for 30 min at 4°C in the dark. After washing twice with 1× PBS, cells were incubated with goat anti‐rabbit IgG‐APC antibody for 30 min at 4°C in the dark. Cells were then washed twice with 1x PBS and fixed with 1% paraformaldehyde for 15 min at 4°C. After fixation, the cells were washed and resuspended in FACS buffer to analyze the proportions of CD69⁻ CD44⁺ CD62L⁺ (central memory), CD69⁻ CD44⁺ CD62L⁻ (effector memory), and CD69⁺ CD44⁺ CD62L⁻ (resident memory) cells in the CD4⁺ T and CD8⁺ T cell populations.

To analyze the circulatory T memory cells, PBMCs were initially stained with 7‐AAD to exclude the dead cells. After staining, the cells were washed with 1× PBS and incubated with an anti‐rat CD32/CD16 antibody for 15 min at 4°C. Following Fc blocking, cells were incubated with the following antibody cocktail for 30 min at 4°C in the dark: CD3‐Viobright720, CD4‐APC, CD8‐BV570, CD44‐PerCP‐Vio700, and CD62L‐FITC. After staining, the cells were washed, fixed with 1% paraformaldehyde, and then washed again before being resuspended in FACS buffer to quantify the proportions of CD44⁺ CD62L⁺ (central memory) and CD44⁺ CD62L⁻ (effector memory) within the CD4⁺ T and CD8⁺ T cell populations.

To analyze the circulatory B memory cells, single‐cell suspensions from rat spleen and PBMCs were stained with 7‐AAD to determine the percentage of viable cells. After washing with 1× PBS, cells were incubated with an anti‐rat CD32/CD16 antibody for 15 min at 4°C. To assess the proportions of class‐switched, conventional memory B cells (IgD⁻ IgG⁺) and non‐switched memory B cells (IgD⁻ IgM⁺), the immune cells from blood and spleen were stained with an antibody cocktail containing CD45R‐PEVIO770, IgD‐APC, IgG‐APC‐Vio770, and IgM‐FITC for 30 min at 4°C in the dark. In a separate incubation, cells were also stained with unconjugated rat anti‐CD19, appropriately diluted in staining buffer, for 30 min at 4°C in the dark. After washing twice with 1× PBS, cells were incubated with a goat anti‐rat IgG‐PE antibody and highly cross‐adsorbed, for 30 min at 4°C in the dark. After washing and fixing, the cells were resuspended in FACS buffer for flow cytometric analysis.

All antibodies were obtained from Miltenyi Biotech, USA or Abcam, UK or Invitrogen, USA or BD Biosciences, USA, and used according to the manufacturer's instructions. The samples were acquired on a MACSQuant Analyzer 16 flow cytometer, and the data were analyzed using FlowJo v10.8 software.

5.20. Antigen‐Specific Proliferation of Liver‐Resident T Cells Following T Cell‐Epitope Stimulation

The immune cells were isolated from the livers of euthanized Wistar rats following immunization on days 42 and 196 as previously described. T cells were extracted and purified from the liver immune cells using a Pan T Cell MicroBeads, rat kit (Miltenyi Biotech, USA) per the manufacturer's instructions. The purified liver T cells were resuspended in 1x PBS at a concentration of 1 × 10⁶ cells mL⁻¹. CellTrace Violet was added to a final concentration of 5 µm, and the cells were incubated for 20 min at 37°C in the dark. The labeling was quenched by adding 5 volumes of complete RPMI‐1640 medium containing 10% FBS and incubating for 5 min on ice. The cells were washed three times with complete medium and resuspended at 1 × 10⁶ cells mL⁻¹ in complete RPMI‐1640 medium supplemented with 10% FBS, 2 mM L‐glutamine, 100 U mL⁻¹ Penicillin‐Streptomycin, 1 mM sodium pyruvate, and 55 µm 2‐mercaptoethanol. The CTV‐labeled liver T cells (5 × 10⁴ cells/well) were cultured in 96‐well round‐bottom plates, and a malaria‐specific T cell epitope (IQNSLSTEW; ChinaPeptides, Shanghai, China) was added at a concentration of 10 µg mL⁻¹. The cultures were incubated at 37°C in a humidified atmosphere containing 5% CO₂.

Positive control cells were stimulated with CpG (5 µg mL⁻¹). Negative control cells were cultured without peptide. Additional controls included cells stimulated with Quil‐A (25 µg) and BP (10 µg). After 72 h, cells were harvested and stained with 7‐AAD to determine the viable cell percentage. T cell proliferation was assessed by measuring CTV dilution. The proliferation index and the percentage of dividing cells were calculated using FlowJo v10.8 software.

5.21. Assessment of Sporozoite Motility Inhibition by Post‐Immunization Sera In Vitro

Transgenic P. falciparum (NF54 strain) sporozoites (SPZ) that express the fluorescent reporter protein GFP under the CSP promotor were used for these experiments (NF54 − ΔPf47–5’csp‐GFP‐Luc, kindly provided by Dr. Koen Dechering; TropIQ Health Sciences). To obtain these sporozoites, P. falciparum gametocytes were cultured using standard culture conditions in a semi‐automated culture system. Subsequently, Anopheles stephensi female mosquitoes were infected by standard membrane feeding of these gametocyte cultures as described previously [104, 105]. To obtain the sporozoites, the salivary glands of these mosquitoes were manually dissected at day 14 or 15 post‐infection. The salivary glands were crushed to release sporozoites in RPMI medium enriched with 10% FBS. The free sporozoites were counted in a Bürker counting chamber using phase‐contrast microscopy (Zeiss, Germany).

To assess the quality of the antibodies generated by the vaccine platform on in vitro sporozoite motility, sporozoites were incubated with pooled serum from the immunized Wistar rat groups for 30 min at RT at a concentration of 1:15. Alternatively, for a negative control, the sporozoites were incubated in medium, while for a positive control, they were incubated with 15 ng µL⁻¹ 2A10 antibody [106]. Thereafter, the sporozoites and serum mix were added to a 96 wells optical glass bottom plate (ThermoFisher Scientific, USA). The plate was centrifuged for 3 min at 1200 rpm at RT. Immediately after centrifugation, 200 frames were measured by Leica TCS SP5 (Leica Microsystems, Wetzlar) at 37°C and 5% CO₂. Each condition was measured in triplicates, and all groups were assessed in two independent experiments, resulting in the analysis of a total of 23 425 sporozoite tracks, with an average of 1415 per condition.The movies were analyzed using SMOOT as previously described [54, 107].

5.22. Standard Membrane Feeding Assays (SMFA) Using P. falciparum Culture

Mature gametocytes of P. falciparum (NF54) were cultured [108] using human erythrocytes with the approval of the Internal Review Board (IRB) of the Johns Hopkins University Bloomberg School of Public Health (#NA 00019050). For the SMFA, serum from immunized groups was diluted to the indicated final concentrations using complement‐active or ‐inactive normal human sera and combined with human RBCs and P. falciparum gametocytes (∼0.3 % gametocytemia and 50 % hematocrit). This mixture was immediately fed to A. stephensi mosquitoes for 20 min through water‐jacketed glass feeders maintained at 37°C using a circulating water bath, as previously described [109, 110]. The blood‐fed mosquitoes were maintained in an incubator (27°C, 80 % relative humidity (RH)) with 10 % sucrose for 7–9 days. Finally, the mosquito midguts were dissected and stained with 0.5% mercurochrome to enumerate the oocysts.

5.23. Animal Study Design‐In Vivo Evaluation of Vaccine Efficacy

The study utilized 6‐8‐week‐old female C57BL/6 mice (Charles River Labs, Frederick, MD, USA), which were housed at the animal facility of the Johns Hopkins Bloomberg School of Public Health. The mouse housing conditions were maintained at 40%–60% relative humidity and a temperature range of 68–79°F, with at least ten air changes per hour and a 14/10‐hour light/dark cycle. The animals were divided into groups of 8 per group. Challenge studies were conducted to assess the reduction in liver infection and protection from parasitemia. The mice were immunized intramuscularly in the calf muscles of the right and left legs with BP+Q and Pfs‐47‐RTS,S‐BP+Q. All groups received one prime and two booster vaccinations (day 0, 14, and 28), and blood samples were collected at predetermined intervals via retro‐orbital bleeding. Sera samples collected on day 41 were used to perform SMFA assay and ELISA. Following immunization, on day 42, the mice were challenged either intravenously with sporozoites or through mosquito bites. Liver burden measurements were taken on day 44 post‐vaccination. Parasitemia was evaluated in the mosquito bite challenged mice from days 46 to 52.

Experiments were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Animal Care and Use Committee of Johns Hopkins University (protocol numbers MO21H417). For in vivo procedures such as bleeding and imaging, mice were under partial anesthesia, (isoflurane), for the mosquito bite challenge, mice were under deep anesthesia (2%avertin). After the final readout, mice were euthanized by CO₂ exposure (5 min), followed by cervical dislocation, following guidelines at JHU.

5.24. Antibodies and Parasites

Antibodies: 2A10 is a murine monoclonal antibody (mAb) that specifically recognizes the repeat domain of the P. falciparum CSP36 [108].

Parasites: Transgenic P. berghei sporozoites expressing P. falciparum CSP, green fluorescent protein, and luciferase reporter gene were used in all studies, as described previously [111]. The parasite preparation process is detailedl [111]. Transgenic P. berghei sporozoites expressing P. falciparum CSP, green fluorescent protein, and luciferase reporter gene were used in all studies, as described previously. The parasite preparation process is detailed. Briefly, 5‐day old A. stephensi mosquitoes were fed on mice harboring 1%–2% transgenic parasites. After 20–22 days post‐blood meal, the transgenic sporozoites were harvested from the mosquito salivary glands and used within 60 min for intravenous infection in liver burden studies. For parasitemia studies, the infectious mosquitoes were used directly.

5.25. Reduction in Liver Infection Assay

The parasite liver burden was assessed using bioluminescence as previously described [111]. Briefly, transgenic sporozoites freshly harvested from mosquito salivary glands were suspended in a 2% FBS‐HBSS medium and adjusted to a concentration of 10 000 parasites/mL. Mice were challenged by intravenous injection of 2000 sporozoites that express the full‐length P. falciparum CSP and GFP‐Luciferase enzyme. 42 h after the parasite challenge, the mice were injected intraperitoneally with 100 µL of D‐Luciferin (30 mg mL⁻¹) to measure the parasite load in the liver. The mice were anesthetized in an isoflurane chamber, and once immobilized, their liver bioluminescence was measured using an IVIS Spectrum in vivo imaging system (PerkinElmer). Groups of five anesthetized mice were placed in the imager, and the radiance measurements were recorded using the live imager software v4.5.1. The total flux reading for each mouse was recorded individually. To verify the background reading, four naive mice that received only the D‐luciferin substrate were included in each study.

5.26. Protection From Mosquito Bite Challenge Assay

Mosquito bite challenge in mice was performed as previously described [111]. On day 42, the immunized mice were exposed to bites from five infected mosquitoes, drawn from a population with an 80% infection rate. The mice were anesthetized with 2% Avertin and placed on top of cages containing the infected mosquitoes for 10 min. 42 h post‐challenge, the mice were injected intraperitoneally with 100 µL of D‐Luciferin and anesthetized in an isoflurane chamber. Once immobilized, bioluminescence in their livers was measured. The number of mosquitoes that fed on blood was determined by observing a red abdomen. Days 4 to 10 after challenge, blood smears were collected from the tip of the mouse tails, stained with 10% Giemsa, and examined under a light microscope to detect the presence of blood‐stage parasites.

5.27. Determination of Antibody Titer Using ELISA

To prepare the ELISA plates, 100 µL of RTS,S and Pfs‐47 soluble peptides were coated at a concentration of 5 µg mL⁻¹ onto 96‐well MaxiSorp plates, which were then incubated overnight at RT. The plates were washed three times with 1x PBS and blocked with 1x PBS‐1% BSA for 1 h. After blocking, the plates were washed three times with 1x PBS, followed by a 1 h incubation with 100 µL of serially diluted serum samples from mice immunized with BP+Q and Pfs‐47‐RTS,S‐BP+Q. The initial serum dilution was 1:100, with seven subsequent three‐fold dilutions. Additionally, antibody 2A10 was used as a positive control, and 1X PBS‐1% BSA as a negative control on each plate. The plates were then washed twice with 1x PBS‐0.5% Tween‐20 and three times with 1x PBS. After washing, each well was incubated with 100 µl of peroxidase‐labelled goat anti‐mouse IgG at a concentration of 250 ng mL⁻¹ for 1 h. This was followed by three washes with 1x PBS‐0.5% Tween‐20 and three washes with 1x PBS. 100 µL of horseradish peroxidase substrate was added to each well, and the plates were allowed to develop in the dark for 20 min. The substrate reaction was stopped by adding 50 µL of 1% SDS to each well. Each plate was then read in a spectrophotometer at OD 405 nm.

5.28. CSP Antibody Serum Levels‐ 2A10 Equivalence

The collected antibody titer data were used to select multiple serum dilutions that fell within the linear range of the titration curve. For each ELISA plate, a standard curve was generated for the monoclonal antibody 2A10 through three‐fold dilutions starting at 333.3 ng mL⁻¹. After measuring the absorbance values, a four‐parameter, non‐linear regression analysis was performed on the 2A10 standard curve using GraphPad Prism software. The absorbance data from the individual serum dilutions were then plotted onto the standard curve, and the dilution factor was applied to estimate the 2A10 equivalents for each serum sample. The chosen data points were averaged to obtain the final results.

5.29. Statistical Analysis

All raw datasets were assessed for quality before analysis. Unless otherwise specified, data are presented as mean ± standard error of the mean (SEM), consistent with the figure legends. Sample sizes (n) for each experiment, including technical replicates for in vitro assays, biological replicates (animals) per group in rat and mouse studies, and the number of mosquitoes per SMFA replicate, are reported in the corresponding figure legends. For experiments involving multiple groups or multiple conditions, statistical significance was assessed using ordinary one‐way ANOVA or two‐way ANOVA as appropriate. Post hoc analyses incorporated Dunnett's, Sidak's, or Tukey's multiple comparison tests depending on experimental design, number of comparisons, and control group structure. A p‐value less than 0.05 was considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). No randomization was performed, and investigators were blinded for the SMFA and Sporozoite Motility Inhibition assays. All statistical analyses and graphing were performed using GraphPad Prism v8.0 (GraphPad Software, USA).

Conflicts of Interest

N.S. and B.H.A.R. are named inventors on the Australian Provisional Patent Application No. 2026900573 that describes the composition and use of BP‐based Malaria vaccines. All other authors declare no conflicts of interest.

Supporting information

Supporting File: smll73179‐sup‐0001‐SuppMat.pdf.

Data Availability Statement

The data that support the findings of this study are present in the paper or the supportive information and they are also available from the corresponding author upon reasonable request. The plasmid maps and formulations can be made available under a material transfer agreement upon request to B.H.A.R.