Female BALB/c mice and New Zealand white rabbits were purchased from the Beijing Animal Institute. The P. berghei ANKA strain 2.34 was maintained by serial passage and used for challenge infection as described previously (Bai et al., 2023). The Δpbg37 parasite used for generating a transgenic parasite expressing Pvg37 was from an earlier study (Liu et al., 2018). Adult (3-5 days old) Anopheles stephensi and An. dirus mosquitoes were fed on a 10% (w/v) glucose solution and kept in an insectary under 25°C ± 1°C and 50 – 80% relative humidity, with a 12 h light and dark cycle. All animal procedures were carried out per the welfare and ethical review standards of China Medical University.

The TrPv37Pb line was generated by inserting the complete Pvg37 open reading frame (ORF, 1053 bp) tagged with 3×HA at its C-terminus into the pL0034 vector at the ApaI and XhoI sites. The Pvg37 ORF was flanked by the 5’ UTR and 3’ UTR of the Pbg37 gene. Ten micrograms of the plasmid were digested with ApaI and XhoI and then electroporated into purified Δpbg37 schizonts using a Nucleofector system. Subsequently, the parasites were injected intraperitoneally into two mice. After 24 h, the TrPv37Pb line was selected through gavage in mice using 5-fluorocytosine (20 mg/mL in water). To confirm the proper integration of the Pvg37 gene at the Pbg37 locus in the P. berghei genome, PCR analysis was performed using specific primers ( Supplementary Table S1 ). The TrPv37Pb parasites were then cloned using a limiting dilution technique.

To investigate the impact of Pvg37 on parasite development, we compared the development among the (wild-type) WT P. berghei, TrPvg37Pb, and Δpbg37 lines. Three groups of BALB/c mice (3 mice per group) were intraperitoneally injected with 1×10⁶ infected red blood cells (iRBCs) of the respective parasite clones. Asexual parasitemia levels were monitored on days 3, 5, 7, 9, and 11 post-infections by examining Giemsa-stained thin blood smears. The number of mature gametocytes per 100 parasites was determined during the parasitemia range of 10–20%. In each mouse, 100 mature gametocytes were differentiated into male and female gametocytes based on morphological characteristics to establish the gametocyte sex ratio. Following induction for gametogenesis at 25°C for 15 min, the culture was transferred onto a coverslip, and exflagellation centers were counted under a phase-contrast microscope at 400× magnification. To observe ookinete formation, 10 μL of infected blood containing equal gametocyte counts were mixed with the ookinete culture medium in a total volume of 50 μL and maintained at 19°C for 24 h. The gametocyte counts were normalized according to gametocytemia. The ookinete number in 0.5 μL of culture was counted using an IFA, with ookinetes stained with a monoclonal antibody (mAb) against Pbs21.

The Pvg37 protein fragments spanning amino acids 25 to 38 and 55 to 68 were synthesized as polypeptides (Genescipt, China), namely Pvg37-P1 and Pvg37-P2, respectively, which were conjugated to keyhole limpet hemocyanin (KLH) for immunization. Three rabbits were subcutaneously immunized with 500 μg of Pvg37 peptides emulsified in Freund’s complete adjuvant. Three booster immunizations were performed at weeks 2, 5, and 8 after emulsification with 250 μg of Pvg37 peptides and incomplete Freund’s adjuvant. The immune serum was collected 10 days after the last immunization. IgGs were purified from Pvg37 immune and pre-immnue sera, respectively, using Protein A columns. The concentrations of anti-Pvg37 and pre-immune control antibodies were determined using the BCA Protein Assay kit.

ELISA was utilized to determine antibody titers of sera. A 96-well plate was coated with polypeptides Pvg37-P1 and Pvg37-P2 in 0.05 M sodium carbonate buffer (pH 9.6) and incubated overnight at 4°C. The samples were then washed three times with PBST (0.05% Tween-20, 0.1 M PBS) and incubated with 1% BSA (Sigma) for 1 h at 37°C. Following another round of washing with PBST, the anti-Pvg37 peptide sera and negative control sera were diluted in PBS containing 1% BSA at multiple proportions ranging from 1:1000 to 1:512000. The samples were incubated at 37°C for 2 h and washed three times with PBST. HRP-labeled sheep anti-rabbit IgG, diluted in 3% BSA (1:5000), was added to the 96-well plates, and the samples were incubated at 37°C for 1 h. Subsequently, the plates were washed five times, and tetramethyl-benzidine was added for color development in the dark for 10 min. The reaction was stopped by adding 2 mM H₂SO₄, and the absorbance value at 490 nm was measured. The final dilution value was considered to be higher than the mean + 3 × standard deviation (cut-off value) of the pre-immune control sera.

IFA was performed on gametocytes, gametes, zygotes, retorts, and ookinetes. TrPvg37Pb parasites were fixed with 4% paraformaldehyde and 0.0075% glutaraldehyde in PBS for 30 min at room temperature. Then, the parasites were washed twice with PBS. After being permeabilized with 0.1% Triton X-100, parasites were blocked with PBS containing 3% BSA for 1 h at 37°C. The rabbit anti-Pvg37-P2 sera (1:200) in PBS containing 3% BSA were added into parasites for 1 h at 37°C. All parasites were co-incubated with mouse antisera against PbMSP1 (1:500), Pbs47 (1:500), Pbα-tubulin (1:500), and Pbs21 (1:500) as specific markers for schizonts, female gametocytes/gametes, male gametocytes/gametes, and zygotes/ookinetes, respectively. These antisera were self-made and have been previously reported (Zheng et al., 2024). After washing the slides with PBS, Alexa Fluor 488-conjugated anti-rabbit IgG secondary antibodies (1: 500, Invitrogen) and Alexa-555 conjugated goat anti-mouse IgG secondary antibodies (1: 500, Abcam) were added into the parasites for 1 h at 37°C. WT ookinetes were used as the negative control. Images were acquired using a Leica SP8 confocal laser scanning microscope. For comparison, mouse anti-HA mAb (1:500, abclone) was also used to probe parasites to determine the expression stage of Pvg37. Furthermore, the localization of Pvg37 protein on the P. vivax gametocytes was confirmed using anti-Pvg37-P2 IgGs and isolating gametocytes from clinical samples of P. vivax.

Different gradient Nycodenz was used to isolate and purify TrPvg37Pb parasites at various stages. When the parasitemia reached 3-5%, mouse blood was collected and mixed with schizont culture medium (RPMI 1640, 50 mg/L penicillin, 50 mg/L streptomycin, 100 mg/L neomycin, 25% fetal bovine serum, and 6 U/mL heparin). The mixture was cultured at 37°C for 20 h. Schizonts were subsequently isolated and purified on a 56% (v/v) Nycodenz gradient. For purifying sexual stages, mice were treated with sulfadiazine (Sigma, Burlington, USA, 20 mg/L) for 2 days to eliminate asexual blood stages when the parasitemia ranged between 10 and 20%. When parasites reached the gametocyte stage, blood was collected and mixed with PBS at 4°C to prevent gametocyte activation. Gametocytes were then isolated and purified on a 48% (v/v) Nycodenz gradient. To obtain ookinetes, 1 mL of blood was mixed with 9 mL of ookinete medium (50 mg/L penicillin, 50 mg/L streptomycin, 100 mg/L neomycin, 20% fetal bovine serum, 1 mg/L heparin, pH 8.3) and cultured at 19°C for 24 h. Ookinetes were then isolated and purified using a 62% (v/v) Nycodenz gradient. Finally, purified parasites from each stage mentioned above were washed twice with PBS.

To determine the expression of Pvg37 at different stages, purified schizonts, gametocytes, and ookinetes were treated with 0.2% saponin to lyse erythrocytes. After three washings with PBS, parasites were treated with RIPA lysis buffer containing phenylmethylsulfonyl fluoride three times to extract total proteins. Protein concentrations were determined using the BCA Protein Assay kit. Equal parasite proteins (20 μg/lane) were separated by 10% SDS-PAGE and transferred to a 0.22 μm PVDF membrane. Then, the PVDF membrane was blocked with TBST containing 5% skim milk for 1 h at 37°C. After blocking, the PVDF membrane was probed with anti-Pvg37 rabbit immune sera (1:200) and anti-rHSP70 sera (1: 1000), and after three washing with TBST, HRP-conjugated goat anti-rabbit IgG antibodies (1: 10000, Invitrogen) as the secondary antibodies were added. The blots were then detected by an ECL Western Blotting Kit (Beyotime).

The TB potential of Pvg37 was evaluated using two assays: an in vitro ookinete formation assay and a direct mosquito feeding assay. The in vitro assay employed various dilutions of the immune sera. Groups of BALB/c mice (three mice per group) were injected intraperitoneally with 1×10⁶ iRBCs of the TrPv37Pb line. The exflagellation of male gametocytes was quantified using the method mentioned above. Purified pre-immune IgGs or anti-Pvg37-P2 IgGs were added into the ookinete culture medium at a final concentration of 0.2, 1.0, and 2.0 µg/µL and mixed with 10 μL of blood from mice infected with the TrPvg37Pb parasites, resulting in a total volume of 50 μL. The number of exflagellation centers was observed in each microscope field at 400x magnification. Ookinete cultures were incubated at 19°C for 24 h, and mature ookinetes in 0.5 μL of culture were counted using a fluorescence microscope (100× oil objective). For antibody transfer experiments, 150 µL of purified anti-Pvg37-P2 IgGs were injected into the tail veins of the mice one hour prior to An. stephensi mosquito direct feeding. Twelve days after feeding, mosquitoes were dissected to assess the number of oocysts using a compound microscope at 200× magnification.

We conducted DMFA using blood samples from volunteers infected with P. vivax. Prior to participation, informed written consent was obtained from four volunteers. Parasitemia was estimated using Giemsa-stained films. The anti-Pvg37-P2 antibody and the negative control antibody were diluted 1:1 with 90 μl of heat-inactivated (complement negative) healthy human AB+ serum, resulting in a total volume of 180 μl. The diluted antibodies were then mixed with RBCs collected from P. vivax malaria patients in a 1:1 ratio. Pooled blood samples were incubated at 37°C for 15 min and then introduced into a membrane feeder. Approximately 100 An. dirus mosquitoes (starved for 12 h before the experiment) were fed with the blood samples for 30 min using the membrane feeder maintained at 37°C. Unfed mosquitoes were subsequently removed, and fed mosquitoes were allowed to feed on cotton pads dipped in a 10% sucrose solution at 27°C and 80% relative humidity for one week. For each group, 20 mosquitoes were dissected to count oocysts.

Genomic DNA from the four P. vivax isolates used in the DMFA was extracted using a QIAamp DNA Blood Mini kit (Qiagen, Germany). The pvg37 DNA fragment encoding aa 1–182 was amplified by PCR with primers designed based on the Sal-I sequence ( Supplementary Table S1 ). The purified PCR products were sequenced using the ABI Prism BigDye™ cycle sequencing kit (Applied Biosystems, Thermo Fisher Scientific). The sequences were aligned using ClustalW in MEGA7.0.26.

Statistical analyses were conducted using SPSS software version X (SPSS Inc., USA). Ordinary one-way ANOVA was used to compare the groups in terms of asexual parasitemia, gametocytes, sex ratio, exflagellation, and ookinete numbers. Mann-Whitney U test was employed to analyze oocyst density (oocyst number per midgut), while Fisher’s exact test was used to analyze infection prevalence. The results are presented as the mean ± SD. A significance level of 0.05 was considered statistically significant.

The TrPvg37Pb transgenic strain was constructed using the Gene Insertion and Marker Out (GIMO) technique. This approach involves replacing the drug resistance gene in the obtained Δpbg37 strain with the pvg37 gene, which is tagged with a 3×HA tag, under negative selection with 5-fluorocytosine ( Figure 1A ). The genotypes of WT, Δpbg37, and TrPvg37Pb were identified by PCR amplification using primers 1 and 2, 1 and 3, and 1 and 4, respectively ( Figure 1A ). Diagnostic PCR results confirmed successful recombination of the pvg37 gene as specific bands were only amplified from corresponding parasite gDNA samples ( Figure 1B ). Additionally, Pvg37-HA protein expression in transgenic P. berghei parasite lines was confirmed by Western blotting using the anti-HA mAb. A protein band of approximately 39 kDa was detected in the TrPvg37Pb parasites but not found in WT parasites ( Figure 1C ).

The impact of complementing Pvg37 on the growth and development of P. beighei was assessed through phenotype analysis performed on TrPvg37Pb, Δpbg37, and WT strains. Equal numbers of parasite-infected RBCs for each strain were injected into BALB/c mice via the tail vein. The results showed no significant difference in asexual-stage parasitemia among TrPvg37Pb, Δpbg37, and WT parasites ( Figure 2A ). However, the Δpbg37 strain exhibited significantly lower gametocytemia and a higher female/male gametocyte ratio ( Figures 2B, C ). Furthermore, compared to WT parasites, the Δpbg37 strain displayed a significant reduction of 27.8% in exflagellation centers and 34.4% in ookinete formation ( Figures 2D, E ), consistent with previously published data (Liu et al., 2018). Notably, no significant differences were observed between the TrPvg37Pb strain and WT strain in terms of gametocyte formation, exflagellation centers, or ookinete production ( Figures 2B-E ). These findings indicate that substituting Pbg37 with Pvg37 does not affect the development of both asexual and sexual stages of P. berghei. Moreover, Pvg37 compensates for the abnormal phenotype caused by pbg37 deletion, suggesting a functional equivalence between Pvg37 and Pbg37.

The Pvg37 protein possesses seven transmembrane domains similar to Pbg37, and multiple sequence alignment revealed a high degree of conservation among Plasmodium species. Two highly conserved regions in Pvg37 were identified at amino acids 25-38 and 55-68, which showed abundant B-cell antigenic epitopes (Liu et al., 2018). Therefore, we opted to utilize these two highly conserved regions, namely Pvg37-P1 and Pvg37-P2, for peptide synthesis. Polyclonal antibodies against Pvg37-P1 and Pvg37-P2 were generated by immunizing rabbits separately. ELISA showed that the final antibody titers for Pvg37-P1 and Pvg37-P2 reached 1:8000 and 1: 128000, respectively ( Figure 3A ). Remarkably, the anti-Pvg37-P2 sera exhibited significantly higher antibody titers than the anti-Pvg37-P1 sera. Thus, we selected the Pvg37-P2 antibodies for subsequent experiments. IgG concentrations purified from the Pvg37-P2 immunized serum and pre-immunized serum were 11.667 and 14.168 μg/μL, respectively, which were adjusted to 10 mg/mL with PBS for transmission-blocking assessment.

To confirm the expression and localization of the Pvg37 protein in transgenic parasites, we purified TrPvg37Pb parasites at different stages, including schizonts, gametocytes, gametes, and ookinetes. Western blot using the rabbit anti-Pvg37-P2 sera detected a ~37 kDa protein band in TrPvg37Pb gametocytes and ookinetes but not schizonts. The expression level in gametocytes appeared higher than in ookinetes ( Figure 3B ).

We examined the subcellular localization of Pvg37 protein in TrPvg37Pb parasites by IFA with rabbit anti-Pvg37-P2 IgGs. Fluorescent signals were observed in the cytosol and at the plasma membrane of gametocytes ( Figure 3C ). During the gametogenesis of microgametocytes, signals were prominently observed along the flagellas. In subsequent development, Pvg37 was specifically associated with the plasma membranes of gametes, zygotes, retorts, and ookinetes ( Figure 3C ). No signal indicating Pvg37 expression was detected in TrPvg37Pb schizonts or WT ookinetes. Since the Pvg37 protein was C-terminally tagged with a 3×HA tag in the TrPvg37Pb parasite, we also performed IFA with a mouse anti-HA mAb and obtained similar results as with anti-Pvg37-P2 IgGs ( Supplementary Figure S1 ). Overall, these findings demonstrate that Pvg37 exhibited a plasma membrane localization pattern during gamete–ookinete transition, similar to Pbg37.

Initially, we investigated the inhibitory activity of anti-Pvg37 IgGs on the formation of exflagellation centers and ookinetes using in vitro assays. The inhibitory effects of the anti-Pvg37-P2 IgGs were concentration-dependent. Compared with control IgGs, anti-Pvg37-P2 IgGs at 0.2 μg/μL had no inhibitory effect on exflagellation or ookinete formation ( Figures 4A, B ). However, at 1.0 and 2.0 μg/μL concentrations, anti-Pvg37-P2 IgGs inhibited the number of exflagellation centers by 28.6% and 60.0%, respectively ( Figure 4A ). Similarly, these concentrations of the anti-Pvg37-P2 IgGs reduced the number of ookinetes by 43. 7% and 69.7%, respectively ( Figure 4B ).

The TB potential of anti-Pvg37 IgGs was further assessed through mosquito feeding assays. In a passive antibody transfer experiment, anti-Pvg37-P2 IgGs significantly reduced the number of oocysts per midgut in mosquitoes by 80.2% compared to the control group ( Figure 4C ), although we did not observe noticeable TBA for the anti-Pvg37-P2 IgGs ( Figure 4D ).

We studied Pvg37 expression and localization in clinical P. vivax samples. In both male and female gametocytes, Pvg37 exhibited cytoplasmic localization and a punctate distribution along the plasma membrane. In contrast, the control IgGs did not show any staining in P. vivax gametocytes. These findings demonstrate that anti-Pvg37-P2 IgGs specifically reacted with P. vivax gametocytes ( Figure 5A ).

To investigate the efficacy of anti-Pvg37-P2 IgGs in blocking P. vivax transmission, DMFA was performed using P. vivax samples collected from four Thai patients with P. vivax mono-infection, as confirmed by PCR analysis targeting the 18S rRNA using species-specific primers. In DMFA, An. dirus mosquitoes feeding on all four blood samples mixed with pre-immune antibodies displayed a mean midgut oocyst intensity of 4.3, 75.4, 111.2, and 69.4, respectively ( Table 1 ). In comparison, mosquitoes feeding on the same blood samples mixed with the anti-Pvg37-P2 IgGs at a 2.5 μg/μL concentration showed a mean infection intensity of 2.3, 50.6, 55.2, and 49.6 oocysts per midgut, corresponding to a respective reduction in oocyst density by ~5.9%, 32.9%, 50.4%, and 28.6% (P < 0.05, Figure 5B ; Table 1 ). However, compared to the control antibodies, the anti-Pvg37-P2 IgGs did not show a significant reduction in infection prevalence ( Figure 5B ; Table 1 ).

The genetic diversity of malaria vaccine candidates in endemic parasites poses a challenge to vaccine development (Takala and Plowe, 2009). To investigate whether the variability of TRA among the different isolates might be attributed to genetic polymorphisms of the pvg37 gene, we sequenced the pvg37 gene fragments from the four P. vivax isolates used in DMFA. Our results showed that these samples had identical amino acid sequences of Pvg37 with the Sal-I strain ( Supplementary Figure S2 ).