Transmission-reducing and -enhancing monoclonal antibodies against Plasmodium vivax gamete surface protein Pvs48/45

ABSTRACT

Gamete surface protein P48/45 has been shown to be important for male gamete fertility and a strong candidate for the development of a malaria transmission-blocking vaccine (TBV). However, TBV development for Plasmodium vivax homolog Pvs48/45 has been slow because of a number of challenges: availability of conformationally suitable recombinant protein; the lack of an in vivo challenge model; and the inability to produce P. vivax gametocytes in culture to test transmission-blocking activity of antibodies. To support ongoing efforts to develop Pvs48/45 as a potential vaccine candidate, we initiated efforts to develop much needed reagents to move the field forward. We generated monoclonal antibodies (mAbs) directed against Pvs48/45 and characterized putative functional domains in Pvs48/45 using recombinant fragments corresponding to domains D1–D3 and their biological functionality through ex vivo direct membrane feeding assays (DMFAs) using P. vivax parasites from patients in a field setting in Brazil. While some mAbs partially blocked oocyst development in the DMFA, one mAb caused a significant enhancement of the infectivity of gametocytes in the mosquitoes. Individual mAbs exhibiting blocking and enhancing activities recognized non-overlapping epitopes in Pvs48/45. Further characterization of precise epitopes recognized by transmission-reducing and -enhancing antibodies will be crucial to design an effective immunogen with optimum transmission-reducing potential.

INTRODUCTION

In 2022, there were an estimated 249 million cases of malaria globally with an estimated 608,000 deaths (1). About 2.5 billion people worldwide are at risk for malaria caused by Plasmodium vivax, one of the four well-established Plasmodium species known to infect humans. Plasmodium falciparum has been predominantly associated with malaria-related mortality; however, a large section of the world’s population lives in P. vivax endemic areas, and it has remained a much less studied species (2–6). More research effort needs to be expended on both these species to advance toward the goal of malaria elimination or eradication and increased vaccine efforts directed against P. vivax at the same time (7, 8). In regions of the world with both P. falciparum and P. vivax malaria, there is decreased transmission of P. falciparum and increased P. vivax transmission (9). Recent observations that Duffy negative individuals, who are reportedly refractory to P. vivax infection, have been found to be infected with P. vivax suggest that P. vivax may be evolving to infect reticulocytes through additional novel surface receptors (2, 4). Increasing cases of severe or fatal P. vivax malaria cases (6, 10) and drug resistance have also been recently reported (11). These recent findings, in combination with unique aspects of P. vivax biology that include the ability to persist as dormant hepatic hypnozoites and faster development of transmission-mediating sexual stages, may make eliminating P. vivax malaria more challenging than P. falciparum.

During transmission of malaria, infective sporozoites from a female Anopheles mosquito vector transmitted to humans during a blood meal invade the hepatocytes to form merozoites which are released into the bloodstream. Merozoites invade the red blood cells and initiate asexual reproduction when a small percentage of parasites differentiate into intra-erythrocytic male and female gametocytes, which, upon ingestion by a new vector during its blood meal, initiate the next transmission cycle. In the mosquito midgut, the gametocytes develop into extracellular male and female gametes followed by fertilization and transformation of zygotes into motile ookinetes that traverse the mosquito midgut wall to form oocysts. Over time, these oocysts mature and produce sporozoites which traffic to the salivary gland of the mosquito, where they are ready to infect a new host during the next blood meal. Several approaches are currently under consideration as effective methods to reduce or eliminate malaria transmission to diminish the global burden of the disease. The transmission-blocking vaccines (TBVs) that induce antibodies directed against gamete-specific (male and female) surface proteins and prevent sexual replication within the mosquito vector represent an important tool among other vaccines targeting malaria infections in the human host (8, 12).

Gamete surface proteins (P230 and P48/45) and those expressed on zygote and ookinete cell surfaces (P25 and P28) have been identified as candidate target antigens for TBVs in P. falciparum and P. vivax (8, 12, 13). Antibodies against P230 and P48/45 have been shown to effectively inhibit sexual development of parasites in the mosquito midgut. P230 and P48/45 form a stable membrane-bound complex on the surface of gametes, and P48/45 has been shown to be critical for male gamete fertility (14, 15). Unlike progress on Pfs48/45, there are inherent challenges in developing a Pvs48/45 TBV. For example, P. vivax cannot be cultured in the lab to produce mature gametocytes and evaluation of transmission-blocking antibodies in direct membrane feeding assays (DMFAs) requires regular access to blood from infected patients in the field. Additionally, there is no effective animal model for evaluating P. vivax TBV. To address these issues for P. vivax, research efforts are underway to generate transgenic Plasmodium berghei, a rodent malaria species, expressing P. vivax antigens (16). Use of these transgenic parasites may offer an effective laboratory-based model for a given Plasmodium species, which can then be used to optimize candidate vaccine immunogenicity parameters, as well as provide an in vivo challenge model.

Development of specific monoclonal antibodies (mAbs) was key to identification of P230 and P48/45 as target antigens for the development of TBVs (17, 18). Availability of blocking mAbs has also been instrumental in revealing conformational restrictions of target epitope recognition. There is a glaring lack of mAb reagents against Pvs48/45 for similar studies and the lack of specific reagents has impeded the progress in understanding the biological features of P. vivax transmission. Availability of recombinant Pvs48/45 offers a unique opportunity to develop mAb reagents to enable several aspects of vaccine development, including mAbs that are directed against blocking as well as non-blocking epitopes, and assist with evaluating antigenic polymorphism and structure of epitopes as targets of natural immune responses (19, 20). The motivation for developing and characterizing Pvs48/45 mAbs was to rationally design future TBVs for redirecting immune responses to blocking epitopes by avoiding non-blocking or enhancing or other immune distracting epitopes. We describe the isolation and characterization of Pvs48/45-specific mAbs that provide the impetus for further studies to facilitate in vitro functional assays and vaccine development efforts for P. vivax.

RESULTS

Isolation and characterization of mAbs

Splenocytes from mice immunized with rPvs48/45 after one or two booster doses were used for fusion. Stable hybridomas were obtained from fusion numbers designated 3, 5, and 7. Mouse for fusion 3 received a single booster dose on day 21 after primary immunization, whereas for fusions 5 and 7, mice received two booster doses on days 21 and 51. Immunization and fusion details are summarized in Table 1, including characterization of antibody isotype, avidity, and fragment reactivity of interaction with rPvs48/45 of five mAbs from fusion 3 (designated 38, 42, 44, 47, and 48), four from fusion 5 (designated 5.4.6, 5.4.10, 5.55.2, and 5.55.5), and two from fusion 7 (designated 7.42.10 and 7.59.11). All the mAbs were specific for rPvs48/45 and did not show any cross-reactivity with rPfs48/45 (data not shown). Immunofluorescence analysis using gametes of P. vivax revealed distinct reactivity patterns. Figure 1 shows reactivity patterns for one representative mAb analyzed from each fusion: weak patchy surface and diffuse cytoplasmic reactivity for mAb 38 and mAb 7.59.11, and strong surface and diffuse cytoplasmic reactivity for mAbs 47 and 5.55.5. There was no detectable reactivity with IgG from normal mouse sera (NMS).

TABLE 1.

Characterization of mAbs and antigen specificity

Fusion #	Booster 1 (day)	Booster 2 (day)	Fusion (day)	mAb clone	IgG isotype	Avidity index (M)	Fragment reactivity
3	21	–a	46	38	IgG2b	1.3	D1, D1D2
3	21	–	46	42	IgG2b	1.5	D1, D1D2
3	21	–	46	44	IgG2b	1.25	D1, D1D2
3	21	–	46	47	IgG2b	1.3	D1, D1D2
3	21	–	46	48	IgG2b	1.3	D1, D1D2
5	21	51	116	5.55.2	IgG2a	2.1	D1, D1D2
5	21	51	116	5.55.5	IgG2a	2.2	D1, D1D2
5	21	51	116	5.4.6	IgG2a	2.1	D1, D1D2
5	21	51	116	5.4.10	IgG2a	1.7	D1, D1D2
7	21	51	124	7.42.10	IgG1	4.2	D2, D1D2, D2D3
7	21	51	124	7.59.11	IgG1	4.2	D2, D1D2, D2D2

^a The dashes indicate that a second booster dose was not given.

Fig 1.

Immunofluorescence reactivity of representative mAbs. Parasites were fixed with 4% paraformaldehyde (not permeabilized) prior to incubation with various mAbs. Purified IgG from NMS was used as a negative control. Cells were counter stained with Hoechst 33342 prior to fluorescent examination. Shown are corresponding representative panels (Hoechst, Alexa Fluor 488, and Merged). The inset in the NMS Alexa Fluor panel also shows a bright field image of the parasite.

Determination of Pvs48/45 domains recognized by mAbs

mAbs were tested using single- and double-domain recombinant fragments of the Pvs48/45 protein expressed in Escherichia coli by Western blotting. D1 (aa 46–181), D2 (aa 186–291), and D3(aa 297–429) have five, four, and six cysteine residues, respectively. Each single domain was trimmed (2-3 aa at N- and C-termini) to optimize recombinant expression and refolding of proteins. However, those sequences were retained in each double domain to maintain the native sequence. It is expected that single-domain fragments D2 and D3 with all Cys paired are likely to refold better than D1 with one unpaired Cys (Fig. 2). Expression of all three single-domain and double-domain fragments was verified by Western blot analysis using anti-(6×)His and polyclonal antisera from mice immunized with recombinant Pvs48/45 (not shown). The presence of numerous disulfide bands contributes to oligomerization/aggregation of purified protein which upon reduction shows a predominant monomeric protein band (Fig. S1). As shown for two mAbs, both non-reduced aggregated and reduced monomeric proteins were recognized by the mAbs (Fig. S1). In Western blot analysis with various sub-fragments, mAbs 7.42.10 and 7.59.11 from fusion #7 reacted predominantly with the D2 domain spanning aa 186–291. mAbs from fusion #3 and #5 reacted with D1 domain (aa 45–185). No differences were seen with reduced and non-reduced westerns.

Fig 2.

Schematic representation of single and double domains of Pvs48/45. All the cysteine residues were predicted to form disulfide bonds, except for one unpaired cysteine in D1 and D1D2 fragments. Signal sequence (SS, residues 1–27) and glycosylphosphatidylinositol (GPI) membrane anchor (residues 438–450) are shown in full-length Pvs48/45 scheme.

Functional activity in DMFA

Purified IgG from ascites were evaluated for transmission-reducing activity (TRA) by DMFA using blood collected from 11 microscopy-confirmed P. vivax-infected donors. Purified IgG from pooled normal mouse serum was used as a negative control. Due to the limited blood volume available from each donor at any point of blood draw, only three to four mAbs could be tested in each DMFA. IgGs were tested at a final concentration of 0.5 mg/mL and each mAb was tested in multiple DMFAs using blood from multiple donors. Figures 3 and 4 show the DMFA data for all the mAbs. There was significant donor-to-donor differences in the baseline infectiousness of gametocytes in Anopheles darlingi mosquitoes. The asexual and sexual parasite density varied between 640 and 22,840 for asexual stage parasites and 40 and 920 sexual stage parasites per microliter blood resulting in wide variations not only in the baseline infectivity (in the presence of normal mouse IgG) but also infectivity in the presence of test mAbs. When analyzed for correlation between the parasitemia values and mosquito infectivity, only the gametocyte density showed a density-dependent correlation with mosquito midgut oocyst numbers (Pearson correlation: r = 0.8379; P = 0.0007). Figure 3 shows DMFA results for mAbs that revealed minimal effect on the infectivity of parasites (oocyst per midgut), averaged for two to four donors used for each mAb. In general, the infectivity in the presence of these nine mAbs shown in Fig. 3 either did not significantly differ from the NMS negative control or showed inconsistent infectivity patterns. DMFA results for the other two mAbs (mAb 47 and mAb 5.55.5) are shown in Fig. 4. mAb 47 was tested against four donors and, in each case, the oocyst numbers in the presence of the mAb were enhanced (average %TRA −124.6%) as compared to NMS IgG negative control. mAb 5.55.5 consistently resulted in a reduction in the number of oocysts for all six donors (average TRA 44.9%). These results also revealed the complexity of outcome differences among donors and blood samples during different transmission seasons; some were more susceptible to inhibition, while others were either enhanced or unaffected.

Fig 3.

DMFA with mAbs revealing mixed minimal effects. All the mAbs (+) and NMS IgG (−) were tested at 0.5 mg/mL final concentration using blood from different donors identified by D number above each set. The total number of mosquitoes dissected varied between 29 and 31. The values below each data panel also present mean number of oocysts and percent prevalence of infected mosquitoes. Results were analyzed using Kruskal-Wallis test followed by Dunn’s multiple comparison test, and P-values are indicated.

Fig 4.

DMFA with mAbs 47 and 5.55.5 exhibiting enhancing and reducing activities. The mAbs (+) and NMS IgG (−) were tested at 0.5 mg/mL final concentration using blood from different donors identified by D number above each set. The total number of mosquitoes dissected varied between 29 and 31 for each feed. The values below each data panel also present mean number of oocysts and percent prevalence of infected mosquitoes. Results were analyzed using Kruskal-Wallis test followed by Dunn’s multiple comparison test, and P-values are indicated.

Competition enzyme-linked immunosorbent assay (ELISA)

Competition-binding ELISAs were conducted to determine binding specificity of the mAbs to rPvs48/45 protein. Horseradish peroxidase (HRP)-labeled mAb 5.55.5 at a fixed concentration was mixed with varying concentrations of either unlabeled mAb 5.55.5 or unlabeled mAb 47 and tested for binding to Pvs48/45 by ELISA. As shown (Fig. 5), the mAbs 5.55.5 and 47 do not compete for binding to Pvs48/45 and are directed against unique sites. As expected, there was a concentration-dependent (2- to 60-fold molar excess) inhibition of binding of HRP-labeled 5.55.5 (panel A) and 47 (panel B) by unlabeled 5.55.5 and 47, respectively. We tested all the other nine mAbs for their ability to compete with HRP-mAb 5.55.5 and HRP-mAb 47. The binding of the mAb 5.55.5 was inhibited only by self and not any of the other mAbs. Similar patterns were obtained when unlabeled mAbs were tested at 25- and 50-fold excess (Fig. 6 panel A shows data for 50-fold excess of all the competing mAbs). When binding of HRP-mAb47 was tested in the presence of 25- and 50-fold excess of competing mAbs, in addition to mAb 47 itself, stronger inhibition was revealed by two additional mAbs, 38 and 42. The other two mAbs 44 and 48 derived from the same fusion #3 (spleen cells after one booster dose on day 21 after primary immunization) like the mAbs 38, 42, and 47 did not show any competition (Fig. 6, panel B).

Fig 5.

Competition between mAb 47 and mAb 5.55.5 for binding to Pvs48/45 in an ELISA. Panel (A) shows binding of HRP-conjugated mAb 5.55.5 in the presence of increasing ratio of unlabeled mAb 47 (dotted line) or unlabeled mAb 5.55.5 (solid line). Panel (B) shows binding of HRP-conjugated mAb 47 in the presence of increasing molar ratio of unlabeled mAb 5.55.5 (dotted line) or unlabeled mAb 47 (solid line). The assays were done in duplicate wells and repeated two times and the panels show data from a representative experiment.

Fig 6.

Binding of HRP-mAb 47 and HRP-mAb 5.55.5 to Pvs48/45 in an ELISA in the presence of excess of all the mAbs. All the competing unlabeled mAbs identified on x-axis were tested at 25- to 50-fold excess concentration. Panel (A) shows binding of HRP-conjugated mAb 5.55.5 in the presence of unlabeled competing mAbs. Panel (B) shows binding of HRP-conjugated mAb 47 in the presence of competing unlabeled mAbs. Figure shows results with 50-fold excess of competing mAbs with similar pattern revealed when competing mAbs were tested at 25-fold molar excess.

DISCUSSION

Antibodies against sexual stages of Plasmodium have been shown to interrupt parasite lifecycle in the mosquito midgut leading to reduction of overall transmission potential (8). Such antibodies can be induced during parasite infection, as well as by vaccination aimed at interrupting malaria transmission. The latter forms the basis for the development of TBV and much progress has been made in P. falciparum. However, similar progress in P. vivax has lagged substantially due to inherent challenges in working with P. vivax, one of which is the lack of much needed reagents to explore immune responses to P. vivax and characterize functional epitopes, and another is the lack of cell culture-available P. vivax parasites.

The isolation and characterization of mAbs against P. vivax sexual stage protein Pvs48/45 described herein presents evidence for antibodies mediating transmission reduction as well as enhancement. These opposite activities were observed with two different mAbs recognizing distinct epitopes on Pvs48/45. As shown in Table 1, mAb 5.55.5 that resulted in ~45% transmission reduction was derived from mouse splenocytes after two booster immunizations (fusion 5) as compared to transmission-enhancing mAb 47 derived after one booster dose (fusion 3). Additionally, two other mAbs (38 and 42) derived from the same fusion 3 and strongly competing with mAb 47 in an ELISA did not show transmission-enhancing activity. We also sought to assess if there were quantitative differences in Pvs48/45 binding among these mAbs. When tested at a concentration of 500 ng/mL in ELISA, the absorbance values were 2.13 (mAb 5.55.5), 1.89 (mAb 38), 1.81(mAb 42), and 1.93 (mAb 47) for the different mAbs. Even though blocking mAb 5.55.5 and enhancing mAb 47 exhibited isotype and avidity differences, it is not clear if these differences account for opposite functional activities. TBV target antigens including P230 and P48/45 exist in a stable membrane-bound complex in both male and female gametes (14), and it is possible that during mosquito infection, the various mAbs may preferentially bind to their corresponding epitopes on different male and female gametes leading to modulation of biological outcomes (reduction versus enhancement).

In a test for reactivity of the mAbs to bind to distinct regions of the full-length Pvs48/45, our data showed that mAbs from fusions 3 and 5 reacted with the D1 fragment comprising aa 45–181 and mAbs from fusion 7 reacted to fragment D2. It is worth stressing that mAbs 47 and 5.55.5 exhibiting opposite biological effects recognize a highly conserved region of the molecule with no reported single nucleotide polymorphism (21). We also show that two other mAbs competed with mAb 47 in the competition ELISA, suggesting that they either compete for the same epitope as mAb 47 or maybe sterically hinder the binding of mAb 47 to its epitope. These mAbs 38 and 42 do not exhibit transmission-enhancing activity similar to mAb 47 and the reason for this can at best be only speculative. In addition to the lack of information of the precise epitope specificity, we note that in the competition ELISA, the antibodies are interacting with recombinant Pvs48/45 bound to a solid support, whereas in the biological DMFA, the native form of the antigen is present in association with P230 on the surface of gametes of both sexes. It is plausible that the presentation of the epitopes recognized by various mAbs is greatly affected by such differences and may account for why mAb 47 showed enhancement whereas mAbs 38 and 42 competing for binding in ELISA did not.

We also noted that the isotype specificity and the avidity index grouped with the different fusions. As shown in Table 1, the three different fusions from which the hybridomas were derived had different booster regimens and time intervals prior to the spleens being removed for fusion. mAbs from fusion 3 were IgG2a with avidity index around 1.5. mAbs from fusion 5 were IgG2b with avidity index around 2.0, and finally, mAbs from fusion 7 were IgG1 and avidity index >4.0. It is of interest to note that enhancing mAb 47 had the lower avidity index and a different isotype as compared to mAb 5.55.5, derived from different fusions. Intuitively, it can be inferred that avidity and affinity are related and raise the possibility of affinity of antibodies as a factor contributing to enhancement of transmission. It is possible that weakly interacting antibodies readily dissociate from gamete surface leading to unperturbed fusion between male and female gametes to become zygotes. It is also possible that transmission-enhancing antibodies promote key interactions of Pvs48/45 facilitating fertilization and subsequent oocyst formation. Previous studies that identified transmission-enhancing effects of sera obtained from P. vivax patients as well as enhancement of oocyte numbers with mAbs did not focus on identification of precise epitopes recognized by such antibodies (22, 23). Not much is known about the mechanisms of transmission enhancement by antibodies, and our observations, complementing and extending previous studies, open several new lines of investigation.

There are a number of studies demonstrating not only transmission-reducing but also transmission-enhancing antibodies for malaria, and thus raising biologically significant questions with future implications for designing effective vaccines (24). These opposite outcomes have been observed with polyclonal antisera from P. vivax-infected humans and mAbs directed against gamete surface and recognizing protein of 36 and 42 kDa in the gametes of P. vivax. As reported, these dual effects were dependent upon the amount of antibodies used during evaluation in mosquitoes with higher concentration, giving blocking and lower concentration of the same resulting in two- to fourfold enhancement (22, 23). While this phenomenon has been investigated in some details for P. vivax, similar antibody-mediated enhancement of infection has also been reported for P. falciparum (8, 24, 25) and animal malaria parasites (26, 27). It is also important to point that antibody-mediated enhancement of viral infections has been studied intensely for nearly four decades (28).

Clearly, more studies are needed to explore the breadth and scope of transmission-blocking versus -enhancing antibodies during immune responses activated during natural infection as well as during vaccination. It is well established that sera from naturally infected people in endemic areas do exhibit transmission-reducing activity. A careful analysis that evaluates transmission-reducing activity and comparative ratio of antibodies against blocking and enhancing epitope will be key to understanding such immune response dynamics. mAbs 5.55.5 and 47 can facilitate such analysis of blocking versus enhancing antibodies. Likewise, similar lines of investigations may be undertaken during vaccination with Pvs48/45 and the impact of various vaccine platforms. Even though the studies described pertain to Pvs48/45, it is also suggested that future studies explore similar possibilities with Pfs48/45 TBV candidate. In the end, the goal of such studies is to develop vaccination strategies that optimize protective responses, and it may be possible to either switch vaccine platforms that minimize enhancing responses or engineer molecules by deleting enhancing epitope for immuno-focusing responses against blocking epitopes. For greatest impact of a TBV vaccine, it will be critical to achieve highest blocking activity with minimum to no enhancing responses.

MATERIALS AND METHODS

Expression and purification of Pvs48/45

Recombinant full-length Pvs48/45 (PlasmoDB accession number PVX_083235) was prepared as described (19). Briefly, the codon-harmonized sequence of Pvs48/45 (GenScript) was cloned between NdeI and NotI restriction sites of the expression vector pET(K-) modified to allow expression with an in-frame C-terminal 6× histidine tag. Other details for isopropyl β-d-1-thiogalactopyranoside (IPTG) induction and purification using Ni²⁺NTA column (QIAGEN) and characterization of purity were as described (19). The eluted protein was dialyzed using phosphate buffered saline (PBS) (pH 8.0) containing 350 mM NaCl and stored at −80°C. Protein purity and immunoreactivity was assessed by SDS-PAGE and Western blot analyses under non-reducing and reducing conditions.

Mouse immunizations

Female BALB/c mice (5–6 weeks) purchased from Charles River were housed and cared for in facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. Mice were immunized with 10 µg protein formulated in complete Freund’s adjuvant (Sigma-Aldrich) and administered (100 µL total volume) intraperitoneally (IP). Mice received booster doses administered IP with 10 µg of recombinant Pvs48/45 in 100 µL incomplete Freund’s adjuvant at specified intervals (Table 1) after priming. Bleeds before immunization and 2 weeks after each booster immunization were evaluated by ELISA for specific antibody production. Spleens from antibody positive responders were used for developing hybridomas. Two immunized mice were boosted once, bled after 2 weeks, and then used for the hybridoma generation. Three mice were boosted a second time and then used for hybridoma generation. The rationale for doing so was that a single boost will generate a different repertoire of specific antibodies in comparison to the repertoire after an additional boost.

Hybridoma generation and monoclonal antibody isolation

The medium used for maintenance of SP2/0 myeloma and hybridomas was Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS), 10% non-essential amino acids, 10% NCTC-109, glutamine, and antibiotics (penicillin and streptomycin). To generate robust antigen-specific B cell proliferation in the spleen and increase the likelihood of antigen-specific B cell hybridoma generation after fusion, mice were given a final boost 3 days prior to fusion using 10 µg of protein in 20 µL PBS, pH 7.4, administered via tail vein injection.

Splenocytes were released by gentle abrasion of the spleens between the frosted ends of two glass slides in a sterile Petri dish, and red blood cells (RBCs) were removed using ammonium chloride-potassium (ACK) lysing solution. Splenocytes and SP2/0 myeloma cells (>90% viability) were washed three times with serum-free medium, and fused using 50% Hybri-Max polyethylene glycol (MW 1,500) (Sigma-Aldrich) at a ratio of 1:1–1:3 myeloma cells to splenocytes using standard methods (29, 30) and selected on hypoxanthine-aminopterin-thymidine selective medium. Ten to 14 days post fusion, supernatants were screened by ELISA using recombinant Pvs48/45 protein. Positive hybridomas were expanded and cloned by limiting dilution using irradiated splenocytes as feeder cells. Selected antigen-specific clones were expanded in culture, screened for Pvs48/45 reactivity, and were injected (intraperitoneal) into BALB/c mice primed with pristane for ascites production. Antibodies from ascites were purified by Protein G-Sepharose affinity chromatography for further studies.

ELISA

Pvs48/45-specific antibody in immune serum and hybridoma supernatants were determined by standardized ELISA (19). Briefly, Immulon 4HBX plates were coated with 1.0 µg/mL Pvs48/45 in carbonate-bicarbonate buffer (pH 9.6) overnight at 4°C, blocked with 5% milk in PBS, and incubated with primary antibody either at room temperature (22°C–25°C) for 1 h or overnight at 4°C. The plates were washed five times in PBS-0.05% Tween-20 followed by further incubation with 1∶10,000 dilution of HRP-conjugated anti-mouse Ig (G, A, and M) antibody mixture for 1 h at room temperature. After washing as above, bound antibodies were detected using one-component ABTS substrate (Seracare, Milford, MA) for 20 min in the dark at room temperature and plates read at 405 nm using VersaMax plate reader (Molecular Device).

Antibody isotype analysis

For IgG subtype analysis, primary antibodies were tested as above except that various IgG isotype-specific HRP-conjugated secondary antibodies were used (anti-mouse IgG1, IgG2a, IgG2b, and IgG3 from Southern Biotech, Birmingham, AL) followed by color development.

Antibody avidity analysis

For comparing relative avidity of antigen-antibody interaction, Pvs48/45-coated ELISA plates incubation with mAbs were treated with sodium thiocyanate (NaSCN, 0 to 8 M serial dilutions) for 15 min. After washing, the plates were incubated with the secondary antibody (HRP-labeled anti-mouse IgG). Binding of antibodies after NaSCN treatment was expressed as a percentage of total binding (absorbance in the wells without NaSCN). The avidity index for each mAb is expressed as the NaSCN concentration resulting in 50% dissociation of bound antibodies.

Competition ELISA

mAbs purified using Protein G Sepharose beads were dialyzed against carbonate-bicarbonate buffer, pH 9.4, using buffer exchange protocol (Amicon Ultra-4, Sigma-Aldrich) for conjugation with HRP. The conjugation protocol was exactly as described for EZ-Link Plus Activated Peroxidase kit (Thermo Scientific). The HRP-conjugated mAbs (HRP-mAb 5.55.5 and HRP-mAb 47) were desalted into PBS using Amicon Ultra-4 spin tubes and stored at −20°C after adding 50% glycerol.

For competition ELISA (31–33), Immulon 4HBX 96-well plates were coated with 100 µL per well of Pvs48/45 (1.5 µg/mL) in 0.1 M carbonate-bicarbonate buffer, pH 9.6, overnight. After washing with PBST (PBS, pH 7.4 containing 0.1% Tween-20) (three times), plates were blocked for 1–2 h with 5% nonfat milk in PBST at 25°C. For competition ELISA, blocked plates were washed (three times) with PBST and incubated with antibodies as described below. First, the optimum concentration of each HRP-mAb conjugate was determined to give an absorbance (405 nm) value around 1.0. During the blocking step, each competing mAb was serially diluted using 1% BSA in PBST and combined with a pre-determined fixed amount of HRP-conjugated mAb to generate a series of mixtures of target (HRP-conjugated mAb) and competing mAbs. Aliquots of each mix were then added to the ELISA plate in duplicate wells and incubated for 60 min at 25°C. The plates were washed (six times) with PBST and developed using 100 µL of ABTS substrate (SeraCare, Milford, MA) at room temperature in the dark. Starting at 10 min, the absorbance (405 nm) was recorded at 5-min intervals using VersaMax (Molecular Device) plate reader. Total binding of HRP-conjugated mAb without any competing mAb was used to calculate percent competition. The background absorbance in the wells without any antibodies was subtracted from the absorbance of antibody-treated wells. The absorbance of bound HRP-mAb in the presence of serially diluted competing mAb was converted to a percentage of absorbance in the wells without any competing mAb using the formula: (absorbance in the well incubated with serially diluted test mAb plus HRP-mAb / absorbance with HRP-mAb alone)× 100.

Production of recombinant Pvs48/45 fragments for Western blot analysis

Codon-harmonized DNA sequences of single- (D1, D2, and D3) and double-domain (D1D2 and D2D3) fragments (Fig. 2) were synthesized (GenScript) and cloned in the expression vector pET(K-). Each sequence contained 6×His residues at the C-terminus separated by a Gly-Pro linker. Details of expression in BL21 E. coli and characterization are similar to those described for full-length Pvs48/45 (19). Expression of proteins was verified by Western blot (WB) analysis using anti-His antibody. Crude bacterial lysates were used for SDS-PAGE and WB analysis to characterize the reactivity of mAbs to various domains.

DMFA

Blood was drawn from adult male and female volunteers (≥18 years old) recruited from patients diagnosed with P. vivax malaria at the Centro de Pesquisa em Medicina Tropical in Porto Velho (CEPEM), Rondonia, Brazil, under an approved protocol (CEPEM #28176720.9.0000.0011). Patients who had used antimalarial treatment in the previous month and/or those who presented severe symptoms of malaria were excluded from this study. Malaria transmission in Porto Velho shows seasonal peaks of incidence following the rainy periods (from October to April). Most cases (~95%) are caused by P. vivax.

Malaria diagnosis was performed by microscopy in thick smears prepared using finger-prick blood collected immediately before the test, stained with 10% Giemsa, and examined for the presence of sexual and asexual parasites. Results were independently confirmed by two well-trained microscopists, and inconsistencies were solved by a senior microscopist. The blood was transported from the clinic to the insectary at Plataforma de Produção e Infecaçã de Vetores da Malaria, Fiocruz Rondonia, Porto Velho, Rondonia, Brazil using a Thermos flask containing water at 37°C and fed to mosquitoes within 15 min of collection. Anopheles darlingi were reared (34, 35) at 26°C ± 1°C with a relative humidity of 70% ± 10%, light/dark period of 12 h:12 h and provided with 15% honey. Adult mosquitoes starved overnight were used for DMFA using Hemotek chambers maintained at 37°C for 30 min. Briefly, the blood was centrifuged (1,300 × g, 10 min at 37°C) and cells were resuspended (1:0.6, vol/vol) using AB blood group serum (heat inactivated, 56°C, 30 min). Resuspended blood (400 µL) was mixed with 100 µL of purified IgG (2.5 mg/mL) from mAbs and the contents added to the Hemotek chambers were maintained at 37°C. After 30 min of blood feeding, unfed and partially fed mosquitoes were removed, and fully engorged mosquitoes were maintained on 15% honey in incubators (26°C ± 1°C, 70% ± 10% relative humidity). Seven days post blood feeding, mosquito midguts were dissected and stained with 0.2% mercurochrome to enumerate the numbers of oocysts. Purified IgG from NMS was used as a negative control. TRA is defined as the percentage reduction in the oocyst numbers, calculated using the formula: [1 − (mean number of oocysts in test IgG / mean number of oocysts in negative control)] × 100.

Immunofluorescence assay (IFA)

Mosquitoes (An. darlingi) (~100) were infected via a membrane feeder with blood from P. vivax-infected volunteers. Mosquito midguts were dissected 4 h after the blood meal and gently homogenized to free parasites and the midgut content was treated with 3% acetic acid (20 µL per midgut) for 2–3 min at room temperature. Acetic acid-treated midgut content was centrifuged (5 min at 1,000 rpm) and washed three times with PBS. Pelleted cells were resuspended in PBS and used to prepare slides using 12-well glass slides. Air-dried slides were stored at −80°C in Ziplock bags containing desiccant. For IFA, frozen slides were removed into a desiccator to warm them up for 2–3 h at room temperature. Cells were fixed using 4% paraformaldehyde for 20 min and washed with PBS (three times). Fixed non-permeabilized cells were incubated with various antibodies diluted to 100 µg/mL in PBS in a humid box, at room temperature for 1 h. Wells were washed (five times) with PBS and incubated with 1:100 dilution of goat anti-mouse immunoglobulins labeled with Alexa Fluor 488 (Invitrogen) for 1 h in dark. After washing with PBS, the wells were counter stained with Hoechst 33342 (10 ng/mL), washed, and mounted using Everbrite mounting medium (Biotium). Slides were examined by light microscopy (Nikon Eclipse 80i) with a 100× oil immersion objective.

AFTER EPUB

[This article was published on 30 January 2024 with a duplicate reference. The references were updated in the current version, posted on 31 January 2024.]

ETHICS APPROVAL

All the in vivo animal experiments were approved by the Tulane University IACUC and adhered to the Guide for the Care and Use of Laboratory Animals by the National Research Council.

DATA AVAILABILITY

Data will be made available upon request.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/iai.00374-23.

Figure S1. iai.00374-23-s0001.tif.

SDS-PAGE and Western blot analyses of rPvs48/45.

Figure S1 Legend. iai.00374-23-s0002.docx.

Legend for supplemental Figure S1.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.