Cells and viruses

HEK293T and baby hamster kidney (BHK) cells were cultivated in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco, Life Technologies, New York, USA) enriched with 10% heat-inactivated fetal bovine serum (FBS), non-essential amino acids, penicillin, and streptomycin. Similarly, rabbit kidney (RK13) cells were maintained in RPMI 1640 medium (Gibco, Life Technologies), supplemented with 10% heat-inactivated FBS, penicillin, and streptomycin. The study used an improved version of a highly attenuated replication-competent Vaccinia virus LC16m8Δ, known as LC16m8Δ2, in which sequences for restriction enzyme sites present in the original LC16m8Δ genome have been removed^29,30,31,32.

Anopheles spp. mosquito colonies, parasites, and animals

The two Anopheles spp. (An. darlingi and An. stephensi) used in the experiment were reared as previously reported¹⁷. The genetically modified Plasmodium berghei parasite lines used in this study, which included engineered PfCSP and PvCSP-VK210, were originally developed by the Laboratory of Vaccinology and Applied Immunology at Kanazawa University^33,34. An. stephensi mosquitoes (SDA 500 strain) served as hosts for these transgenic parasites after feeding on Plasmodium-infected female BALB/c mice that were 6 weeks old before initiation of the challenge studies. Female BALB/c mice, over 6–8 weeks of age, were obtained from Japan SLC (Hamamatsu, Shizuoka, Japan) and were routinely used in this study.

Vector construction

To enable homologous recombination, the transfer plasmid vector, m8Δ-mPH5-Pv(s25-CSP-VK210/247)-iRFP670-A46R, was constructed using the pUC57-simple plasmid, which contains the mPH5 promoter and the fluorescent reporter gene iRFP670. To construct the transfer vector, the left (A46R-L) and right (A46R-R) homologous arms were amplified from LC16m8Δ genomic DNA by PCR (Supplementary Table 1). The fluorescent reporter gene iRFP670 was also PCR-amplified from the pMXs-PT2A-iRFP670-GFP-TdTomato.These PCR products were cloned into the pENTR-D-TOPO vector to generate pENTR-A46R-L, pENTR-A46R-R, and pENTR-iRFP670-P11. The pUC57-simple-A46R-mPH5-P11 plasmid was also employed to facilitate the construction process.

The development of the transfer vector involved multiple cloning steps. Initially, the pUC57-simple-A46R-mPH5-P11 plasmid was digested with PstI and FseI, and the resulting fragment was inserted into the corresponding sites of pENTR-A46R-L, forming pUC57-A46R-mPH5-P11. Subsequently, the pENTR-iRFP670 plasmid was digested with KpnI and XbaI, and the iRFP670 fragment was inserted into the corresponding sites of pUC57-A46R-L-mPH5-P11, resulting in pUC57-A46R-L-mPH5-P11-iRFP670.

To further enhance the vector, the P. vivax antigen Pv(s25-CSP-VK210/247) was incorporated by digesting the pVR1-P7.5-Pv(s25-CSP-VK210/247) plasmid with EcoRI and XmaI. The fragment was then cloned into the EcoRI and XmaI sites of pUC57-A46R-L-mPH5-P11-iRFP670, yielding pUC57-A46R-L-mPH5-Pv(s25-CSP-VK210/247)-P11-iRFP670. Finally, the right arm of the A46R region (A46R-R) was added to this construct by digesting the pENTR-A46R-R plasmid with NotI and HindIII. The fragment was then inserted into the corresponding sites of pUC57-A46R-L-mPH5-Pv(s25-CSP-VK210/247)-P11-iRFP670, resulting in the final transfer vector, pUC57-A46R-L-mPH5-Pv(s25-CSP-VK210/247)-P11-iRFP670-A46R-R.

Vaccine construction and production

Using m8Δ-p7.5-Pf(s25-CSP)-HA as a parental strain, the transfer plasmid vector pVR1-mPH5-Pv(s25-CSP-VK210/247)-iRFP670-A46R was inserted into the A46R region of the parental strain to develop m8Δ-Pf(s25-CSP)-HA-Pv(s25-CSP-VK210/247)-iRFP670-A46R¹⁶. The fusion gene cassette encoding P. vivax antigens, mPH5-Pv(s25-CSP-VK210/247)-iRFP670, was integrated into the A46R locus of the parental vector through homologous recombination. This genetic modification allowed the expression of P. vivax antigens under the control of the mPH5 promoter.

HEK293T cells were infected with m8Δ-Pf(s25-CSP)-HA at a multiplicity of infection (MOI) of 0.2 and incubated at 33 °C in DMEM to produce the recombinant virus. Following infection, the cells were transfected with the transfer vector, pVR1-mPH5-Pv(s25-CSP-VK210/247)-iRFP670-A46R, using Lipofectamine 3000. The recombinant virus, m8Δ-Pf(s25-CSP)-HA-Pv(s25-CSP-VK210/247)-iRFP670-A46R, was harvested after freeze-thaw cycles and purified using plaque purification based on iRFP670 fluorescence.

Each recombinant m8Δ virus was isolated and purified by plaque purification using iRFP670 fluorescence. For AAV1-based vaccines, AAV1-Pf(s25-CSP)¹² and AAV1-Pv(s25-CSP-VK210/247)^16,17 were used as previously described. Henceforth, m8Δ-Pf(s25-CSP)-HA-Pv(s25-CSP-VK210/247)-iRFP670-A46R is described as m8Δ-Pf/Pv, m8Δ-Pf(s25-CSP) as m8Δ-Pf, m8Δ-Pv(s25-CSP-VK210/247) as m8∆-Pv, AAV1-Pf(s25-CSP) as AAV1-Pf, and AAV1-Pv(s25-CSP-VK210/247) as AAV1-Pv.

Immunoblotting

HEK293T cells were infected with various recombinant m8Δ viruses, including m8Δ-Pf/Pv, m8Δ-Pf, and m8Δ-Pv, at an MOI of 5. Additionally, cells were infected with AAV1-Pf or AAV1-Pv at an MOI of 10⁵. Cell lysates were collected in Laemmli buffer 24 or 48 h post-infection and analyzed using immunoblotting³⁵. The lysates were subjected to electrophoresis on a 10% sodium dodecyl sulfate-polyacrylamide gel (SDS-PAGE) under reducing conditions, followed by probing with specific monoclonal antibodies (mAbs), including anti-PfCSP mAb 2A10 (BEI Resources, #MRA-183) and anti-PvCSP-VK210 mAb 2F2 (BEI Resources, #MRA-184). For non-reducing conditions, anti-Pfs25 mAb 4B7 (BEI Resources, #MRA-315) or anti-Pvs25 mAb N1-1H10 (BEI Resources, #MRA-471) was used. Detection was carried out using goat anti-mouse IRDye 800-conjugated secondary antibodies (Rockland Immunochemicals, Limerick, PA, USA), and the results were visualized using an Odyssey infrared imaging system (LI-COR, Lincoln, NE, USA)³⁶.

Immunofluorescence assay

Immunofluorescence assays were conducted to assess protein expression using fluorescently-labeled antibodies. RK13 or HEK293T cells, seeded at a density of 5 × 10⁴ cells per well, were infected with the m8Δ vaccine through serial dilution. After 72 h (RK13) or 24 h (HEK293T), live-cell staining was conducted. For this, R-Phycoerythrin LK23 (Dojindo, Japan)-conjugated anti-PfCSP mAb (2A10) and Fluorescein LK01 (Dojindo, Japan)-conjugated anti-PvCSP-VK210 mAb (2F2) were added directly to the plaque-forming cells. A BZ-X710 fluorescence microscope (Keyence Corp., Tokyo, Japan) was used for image acquisition.

Immunization

For the prime immunization, 1 × 10⁷ plaque-forming units of the m8Δ vaccine, including m8Δ-Pf/Pv, m8Δ-Pf, and m8Δ-Pv, were administered via tail scarification. Six weeks after the initial priming, 1 × 10¹⁰ viral genomes of AAV1-Pf and AAV1-Pv were delivered intramuscularly as a boost. The animals in the control group received phosphate-buffered saline (PBS) injections following the same 6-week interval, serving as controls for the immunization study. Mice were anesthetized using ketamine–xylazine and the final bleed was collected by cardiac puncture, before been cervically dislocated inside a biosafety cabinet.

Enzyme-linked immunosorbent assay (ELISA)

Serum samples were collected from the tail veins of mice to evaluate humoral immunity. Blood samples were continuously collected from the tail vein over 8 months, and antibody titers were determined using ELISA. Specific IgG titers against PfCSP and truncated-PfCSP (N-terminal, repeat, and C-terminal), Pfs25, PvCSP-VK210, PvCSP-VK247, and Pvs25 were quantified as described previously^17,37.

In brief, 96-well microtiter plates (EIA/RIA plate, Coster^®, Kennebunk, CA, USA) were coated with 100 µl of PfCSP protein (200 ng/well)³⁸, truncated-PfCSP protein (N-terminal, repeat, and C-terminal) (each 400 ng/well)³⁷, Pfs25 protein (400 ng/well)³⁹, PvCSP-VK210 protein (200 ng/well)⁴⁰, PvCSP-VK247 peptide (200 ng/well, NH₂-GAGNQPGANGAGNQPGANGAGNQPGAN-COOH, Eurofins, Kyoto, Japan)¹⁷, or Pvs25 protein (400 ng/well)³⁹ overnight at 4 °C. Then, the plates were blocked with PBS containing 1% bovine serum albumin. All proteins used in the study were expressed using the E. coli production system. Serum samples from tail vein blood were serially diluted and added to the wells, followed by incubation at room temperature for 1 h. After incubation, the plates were washed, and horseradish peroxidase-conjugated goat anti-mouse IgG (H + L) antibodies (Bio-Rad, Tokyo, Japan) were added to the wells. A substrate solution containing hydrogen peroxidase (H₂O₂) and 2,2ʹ-azino-bis(3-ethylbenzothiazoline-6-sulfonate) was used to develop the color. The endpoint titers were calculated as the inverse of the highest dilution of serum that produced an optical density (OD) reading of 0.15 units at 414 nm, above the OD of the negative controls (below 0.1 OD units). Before the vaccination study, all mice were confirmed to be seronegative for the measured antibodies.

Challenge infection with PfCSP and PvCSP transgenic sporozoites

Mice were intravenously challenged with PfCSP/Pb or PvCSP-VK210/Pb sporozoites suspended in RPMI 1640 medium. The sporozoites were prepared following established protocols^16,17. Each mouse received 100 µl of media containing 1000 sporozoites through tail vein injection. The progression of malaria infection was assessed daily from days 4 to 14 post-challenge by examining tail blood in thin blood smears stained with Giemsa. Protection was defined as the absence of blood-stage parasitemia on day 14 post-challenge. For mice that developed blood-stage infections, the time taken to reach 1% parasitemia was recorded as described previously¹⁷. In addition, mice that exhibited sterile protection against PfCSP/Pb or PvCSP-VK210/Pb sporozoites were rechallenged with 1000 sporozoites of the alternate strain. Protective efficacy was determined by detectable blood-stage parasites in peripheral blood smears until day 14 post-rechallenge. Furthermore, a predictive model was generated to estimate the time required to reach 1% parasitemia. Surviving mice were anesthetized using ketamine-xylazine followed by cervical dislocation in a biosafety cabinet.

Direct feeding assay (DFA) using Pfs25/Pb

The transmission blocking effect against P. falciparum was evaluated using a DFA with Pfs25DR/Pb as previously described²⁰. Transgenic P. berghei expressing Pf antigens were used based on their reliable infectivity in mice and established use in preclinical TB vaccine studies.

Mice were treated intraperitoneally with phenylhydrazine 28 days after the booster immunization to induce reticulocytosis. Three days later, the mice were injected intraperitoneally with 10⁶P. berghei Pfs25DR3-parasitized red blood cells to establish infection. Three days post-phenylhydrazine, 7–10-day-old An. stephensi mosquitoes were allowed to feed on the infected blood. The next day, non-feeding female mosquitoes were removed. Dissection of all remaining mosquitoes was performed on days 11 and 12 post-feeding to determine the number of oocysts in their midguts.

The transmission-reducing activity (TRA) and transmission-blocking activity (TBA) were evaluated by comparing the oocyst counts in the midguts of mosquitoes fed on immunized and control mice. TRA and TBA were determined using the following formulas: TRA (%) = 100 × [1 − (average number of oocysts in the test group/average number of oocysts in the control group)]; and TBA (%) = 100 × [1 − (proportion of mosquitoes with no oocysts in the test group/proportion of mosquitoes with oocysts in the control group)].

Collection of P. vivax-infected blood samples in Brazil

Blood samples from P. vivax-infected patients were collected as part of this study, following ethical guidelines and written informed consent from all participants. The study was approved by the Ethical Review Board of Fundação de Medicina Tropical Dr Heitor Vieira Dourado (FMT-HVD) under approval number 4.768.634 (CAAE: 47000221.3.0000.0005). Heparinized tubes were obtained with 9 ml of venous blood samples from patients. Patient recruitment was conducted at FMT-HVD in Manaus, Amazonas, Brazil, and met specific inclusion criteria. Participants were aged 18 years or older, with a confirmed diagnosis of P. vivax mono-infection by positive microscopy (parasitemia exceeding 1000 parasites/µl) and positive results from rapid diagnostic tests (Tokyo Future Style, Tokyo, Japan). Patients who had received antimalarial treatment within the last 30 days were excluded.

The percentages of parasitemia (P) and gametocytemia (G) for each P. vivax-infected blood sample isolated were recorded (ID1; P 0.02%, G 0.0018%, ID2; P 0.21%, G 0.0019%, ID3; P 0.02%, G 0.0018%). After blood collection, all participants received treatment in accordance with the Brazilian Malaria Treatment Guidelines⁴¹.

Direct membrane feeding assay (DMFA) on sera from P. vivax-infected patients from Brazil

To evaluate whether serum collected 28 days (short-term) or 224 days (long-term) post-boost could block P. vivax transmission to An. darlingi mosquitoes, a DMFA was performed as described elsewhere⁴². This assay used a membrane feeding system connected to a 37 °C water circulation unit with an inlet to introduce blood to feed the mosquitoes. Blood samples were processed by centrifugation to remove plasma, and erythrocytes were washed three times with equal volumes of RPMI 1640 medium. The washed erythrocytes were then reconstituted with inactivated human serum mixed with either control sera from unimmunized mice (n = 30) at a 1:5 dilution or sera from immunized mice (collected 28 or 224 days post-boost, n = 10) at dilutions of 1:5, 1:10, or 1:50, achieving a final hematocrit concentration of 50% in a total volume of 500 µl.

The reconstituted blood was used to feed the four groups of 100–120 female An. darlingi mosquitoes via a Parafilm^® membrane for 60–120 min. After feeding, engorged mosquitoes from each group were transferred to a larger cage and maintained at 27 °C with 70%–80% relative humidity. Seven days post-infection, the midguts of mosquitoes from each group were dissected, and the number of oocysts in each midgut was recorded. Infection intensity, represented as the mean number of oocysts per midgut, was calculated and used to determine TBA and TRA, as described in the DFA analysis above.

Statistical analysis

The protective efficacy of the m8∆/AAV1-Pf/Pv vaccine was assessed by comparing the survival times with 1% parasitemia of protected and infected mice among the vaccinated groups and the PBS or naïve control groups. Statistical significance was determined using the log-rank (Mantel–Cox) test. The differences between groups were evaluated using the Kruskal–Wallis test to compare antibody responses, oocyst intensity, TRA, and TBA, followed by Dunn’s multiple comparisons test. For TB efficacy of the vaccine against P. falciparum, the Mann–Whitney U-test was used. All analyses were conducted using GraphPad Prism software version 9 (GraphPad Software, San Diego, CA, USA), with p < 0.05 considered statistically significant. The figures indicate statistical significance as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Ethics statement

All patients with P. vivax infection included in the project provided written informed consent to protocols approved by the Fundacao de Medicina Tropical Heitor Vieira Dourado (FMT-HVD) ethical board committee (CAAE: 47000221.3.0000.0005, approval number 4.768.634), Brazil. All animal care and handling procedures were performed under the approved guidelines of the Animal Care and Ethical Review Committee of Kanazawa University (No. AP-214212), Japan. The study was conducted in accordance with the local legislation and institutional requirements.