Background
Despite recent progress in malaria prevention and control, the disease continues to take a heavy toll [
1]. It is believed that the development of a malaria vaccine, which is effective over a wide range of human populations and covers a vast genetic diversity of the parasite, would be essential for complete eradication of malaria. At present, the best available vaccine is RTS,S/AS01, which received a positive opinion from European regulators for the first time in 2015 [
2], is a pre-erythrocytic vaccine that targets sporozoites and protects by curtailing liver infection [
3]. In recent field trials, this vaccine had shown modest efficacy in protection [
4‐
6] and is unlikely to meet the goals for complete eradication of malaria. Efforts to develop a vaccine against the asexual blood stages of the parasite (which causes the clinical symptoms of the disease and against which natural immunity evolves) have led to identification of several antigens that could induce protective response. Some of these have been tested for their protective activity without much success [
7‐
12]. Two major hurdles in the path for the development of a blood stage vaccine have been the presence of a high degree of antigenic polymorphism in the parasite and the high threshold levels of antibodies needed for protection [
13,
14].
Plasmodium, being an intracellular parasite, needs to invade host cells to establish infection. There are three invasive stages (sporozoites, merozoites, ookinetes) in the life cycle of
Plasmodium, two of which (sporozoites and merozoites) get briefly exposed to the humoral branch of the human immune system, rendering the molecular machinery of merozoites and sporozoites involved in invasion as attractive targets for anti-malarial vaccine.
Current approaches to circumvent the barriers imposed by the genetic diversity in
Plasmodium and its multistage complex life cycle are to combine multiple antigens that are valid targets at various stages in the parasite life cycle as well as their orthologues from different species/strains to obtain an effective multistage, species and strain transcending malaria vaccine [
15‐
17]. An alternative approach will be to identify antigens or epitopes that have cell surface expression at multiple stages, do not exhibit polymorphism, have critical non-redundant physiological function(s) and have high immunogenicity.
Pfeno has recently been identified to be a target of parasite neutralizing antibodies. This antigen is unusual in exhibiting cell surface expression at all the three invasive stages [
18‐
20]. Structurally, Pfeno is distinct from the host (human and mosquito) enolases in having a plant-like insert, EWGWS [
21,
22]. Enolase in merozoites and ookinetes is a target for parasite neutralizing antibodies [
19,
20,
23]. Anti-rPfeno antibodies showed strong growth inhibitory effect on blood stage in vitro cultures of
Plasmodium falciparum. Antibodies induced by active immunization with rPfeno resulted in significantly reduced parasitemia that prolonged the survival of the infected mice [
20]. These mice had a large fraction of their IgGs targeted against the EWGWS epitope of parasite enolase [
24]. Immunization with a peptide containing EWGWS epitope also resulted in control of parasitemia. Antibody titers in EWGWS immunized animals showed a positive correlation with survival and a negative correlation with parasitemia [
23]. Partial protection observed in immunized mice suggested that either the induced antibodies were at sub-threshold levels (poor immunogenicity) or there was a functional redundancy in the invasion step at which enolase was involved. Here, a monoclonal antibody based approach is taken to obtain further insight and assess the merozoite invasion neutralizing potential of the antibody directed against EWGWS epitope.
Methods
Materials
Protein A-Sepharose was obtained from Thermo Scientific, Rockford, IL, USA. ABTS was supplied by Sigma-Aldrich, St. Louis, MO, USA. Synthetic peptides were supplied by USV India Ltd., Mumbai, India. All the chemicals used were of analytical grade.
Halobacterium gas vesicle nanoparticles (wild type and recombinant) were prepared as described earlier [
23]. Recombinant particles had a peptide with a sequence ASKN
EWGWSKSKS cloned in one of the gas vesicle proteins (gvpC).
Purification of rPfeno and activity measurements
rPfeno was purified using the over expression system as described earlier [
21]. Briefly, full-length enolase gene was cloned from a gametocytic cDNA library made from NF54 strain of
P. falciparum. The cloned gene was over expressed in
E. coli BL21 and the 6XHis tagged recombinant protein was purified using Ni–NTA metal affinity chromatography. Cloning resulted in addition of 18 aminoacid residues (MRGSHHHHHHGSACELGT-) to the N-terminus and 7 residues (-LQPSLIS) to the C-terminus. Enolase activity was measured using 2-phosphoglycerate (2-PGA) as the substrate and monitoring the formation of phosphoenolpyruvate by increase in OD at 240 nm. The assay mixture consisted of 500 μl of 50 mM Tris–HCl pH 7.5, 1.5 mM Mg(II) and 1.5 mM 2-PGA [
24].
Generation of mouse mAbs against rPfeno
For the generation of mAbs directed against various epitopes of rPfeno, 6 weeks old, female BALB/c mice were injected intraperitoneally with 50 µg of purified rPfeno emulsified with Freund’s complete adjuvant. This was followed by booster injections at an interval of 3 weeks for the next 2 months. The best responder mouse was immunized with 250 μg of immunogen (rPfeno) in phosphate buffer saline (PBS) (10 mM Na-Phosphate, 137 mM NaCl, 2.7 mM KCl, pH 7.4). Five days later, the splenocytes from this mouse were fused with mouse myeloma SP2/O-Ag14 cells (Sigma-Aldrich) using polyethylene glycol (Merck) as fusogen. After selection in medium containing Hypoxanthine-Aminopterin-Thymidine (HAT, Invitrogen) for a week, the resulting hybrid clones were screened for antibody secretion wherein binding of hybridoma cell culture supernatants to rPfeno was tested by ELISA. Of the fifty seven hybrid clones that showed reactivity to rPfeno in ELISA, thirty four were re-cloned by limiting dilution to obtain pure clones. Generation of hybridomas and production of hybridoma supernatants were carried out by Bioklone Biotech Pvt. Ltd, Chennai, India.
Purification and isotyping of mAbs
mAbs from the hybridoma supernatants were purified using the ammonium sulfate precipitation method [
25]. Briefly, the IgG (mAb) was precipitated from hybridoma supernatant by adding powdered ammonium sulfate to 45% saturation using standard protocol. The precipitate was collected by centrifugation, dissolved in phosphate saline buffer (PBS) and dialyzed to remove the salt. The dialyzed solution was passed through a Protein-A affinity column (Thermo Scientific Pierce 20356) and eluted with a low pH buffer (0.1 M glycine–HCl, pH 2.0). Alternatively, the ascites fluid was filtered through glass wool and ~ 15 ml filtrate was passed through 5 ml Protein-A Sepharose column. The flow through was collected and stored for later testing of any unbound residual antibody. The column was washed with three column volumes of PBS. Following this, 5 ml of elution solution (0.1 M glycine–HCl pH 2.0) was added and 1 ml fractions of the eluate were collected. Each 1 ml fraction was neutralized with 100 μl of 1 M Tris, pH 9.0. 10 μl of each of the five aliquots, dissolved in Laemmli buffer [
26], were run on a 10% SDS-PAGE to evaluate the purity of the preparation. IgGs were also purified from anti-rPfeno antisera (polyclonal) as well as pre-immune sera using ammonium sulfate precipitation and Protein A affinity chromatography as described above. Purified IgGs were sterilized by filtration through a 0.22-μm filter and the solution was tested in LB medium for any bacterial contamination.
For determining the IgG subtype of mAb, 100 μl of goat antibodies to mouse isotypes viz., IgG1, IgG2a, IgG2b, IgG3, IgM and IgA were coated in the wells of a microtitre plate at 1:1000 dilution. Following this, the hybridoma culture fluid containing anti-rPfeno mAb was added and incubated overnight at 4 °C. Mouse anti-rPfeno serum (1:1000) was used as positive control. The wells were later treated with rabbit anti-mouse IgG-HRP (that binds to mouse IgG, IgM and IgA) and ABTS substrate was added to measure reactivity at 405 nm [
21]. The OD values obtained for the positive control were IgG1: 0.61; IgG2b:0.69; IgG3:0.57; IgM:2.24; IgA:0.93 indicating the ability of secondary antibody to bind to all three immunoglobulins. Isotyping of mAbs was performed by Bioklone Biotech Pvt. Ltd, Chennai, India.
ELISA
ELISA was performed as described earlier [
20]. Briefly, wells in immunoplates (Nunc, Denmark) were coated with 50–100 ng of purified rPfeno and incubated at 37 °C for 2 h. After a single wash, the wells were blocked by the addition of 5% skimmed milk in PBST (phosphate buffer saline containing 0.05% Tween-20, pH 7.4) and incubated for 1 h at room temperature or 4 °C overnight. The blocking solution was discarded and the wells were washed with PBST 3 times for 5 min each. The wells were then coated with 100 μl of primary antibody (dilution varied according to experiment) and incubated at room temperature for 1 h. The solution was discarded and the wells were washed with PBST three times to remove any unbound antibody. This was followed by coating the wells with HRP conjugated mouse secondary antibody at a dilution of 1:1000 and incubating it at room temperature for 45 min. The solution was discarded and the wells were washed with PBST thrice; 200 μl of ready-to-use ABTS substrate was added into each well. The colour was allowed to develop for 10–15 min and the OD was measured at 405 nm on a Tecan Plate Reader.
For a competitive ELISA, a cyclized 13-mer peptide with a sequence NH2-SCKNEWGWSKSCS-COOH (CPS1920) containing EWGWS was synthesized. ELISA plates were coated with 50 ng of rPfeno (1 pmol); 100 µl of buffer containing 1 pmol of antibody and different concentrations of CPS1920 peptide (0, 5, 10 and 100 µM) were incubated overnight at 20 °C. Pre-incubated peptide plus mAb was then added to the ELISA plate wells in triplicates and incubated for 10 min and the plates were developed using the ABTS substrate as described above. An irrelevant peptide NH2-SWPLPSHTAVWG-COOH (Peptide CPS1916-1) was used in place of CPS1920 as a control to ensure specificity of competitive displacement.
SDS-PAGE and western blotting
Proteins were resolved on a 10 or 12% SDS-PAGE [
26] and visualized by staining with Coomassie Brilliant Blue R-250. For Western blotting, proteins separated by SDS-PAGE were transferred to a PVDF membrane using semi-dry western transfer apparatus (Trans-blot SD-cell, Bio-Rad Laboratories, Inc., Hercules, CA, USA) at a constant voltage of 18 V for 50 min. The membranes were blocked with 5% skimmed milk in PBST for 2 h. The blots were treated with primary antibody followed by washing and incubation with HRP conjugated secondary antibody. The immunoblots were developed using ECL substrate (Pierce).
In vitro culture and synchronization of Plasmodium falciparum 3D7
Plasmodium falciparum 3D7 was cultured as described earlier [
27]. Fresh human RBCs were obtained from the blood samples drawn from consenting informed volunteers by a professional at the Pathology Department, Health Promotion Facility of Tata Institute of Fundamental Research, Mumbai, India.
The ring stage parasite cultures were synchronized by sorbitol treatment as described earlier [
28]. Briefly, 5% sorbitol was added to the infected RBC pellet and incubated at 37 °C for 10 min followed by centrifugation (1500 g, 5 min). The cells in the pellet were cultured for 48 h (i.e., one cycle of multiplication) before subjecting them to a second round of sorbitol treatment. Such a double sorbitol treatment resulted in a high degree of synchrony (> 98% ring stage parasites) [
29].
Growth inhibition assays
Asexual stages of
P. falciparum 3D7 were cultured in vitro [
30] and growth/invasion inhibition assay was performed on a 96-well plate as described earlier [
16,
20,
31].
Plasmodium falciparum 3D7 cultures maintained on O
+ erythrocytes at 2% haematocrit were synchronized at the ring stage using sorbitol treatment [
29]. For each assay, 200 μl of 0.5–3% parasitized cells were used in three identical wells for each time point or antibody concentration. The culture plates were incubated for 48 h and parasitemia was assessed by examining the culture smears using Field stain [
32]. Percent parasitemia was determined by counting at least > 1000 erythrocytes [
20]. Differences in mean parasitemia between control (no mAb) and experimental (mAb added) samples were analysed with Student’s t-test using GraphPad (InStat, San Diego, CA, USA). Differences were considered significant if the p-values were < 0.05.
For the data presented in Fig.
4, the following method was used to measure the GIA of the mAb. Chloroquine (CQ) sensitive (3D7) and resistant (INDO) strains of
P. falciparum were used for in vitro cultures. Parasite strains were cultivated by the method of Trager and Jensen [
33] with minor modifications. Cultures were maintained in fresh O
+ human erythrocytes (obtained from Rotary Blood Bank, 56-57, Tughlakabad Institutional Area, New Delhi, 110062). Cultures were maintained at 4% haematocrit in complete medium (RPMI 1640 with 0.2% sodium bicarbonate, 0.5% Albumax II, 45 mg/l hypoxanthine, and 50 mg/L gentamicin) at 37 °C under reduced O
2 (gas mixture 5% O
2, 5% CO
2, and 90% N
2). Stock solutions (1 mM) of CQ were prepared in water (MilliQ grade), and lyophilized powders of antibodies were dissolved in filter sterilized MilliQ water to a final concentration of 1 mg/ml. 0–6 µl antibody solutions were transferred to wells of microtitre plate taking each dose in triplicates and final volume was made to 6 µl in each well using PBS. Chloroquine (1 µM) was used as positive control. Ring stage Sorbitol synchronized culture (94 µl) was aliquoted to wells of 96-well plate at 2% haematocrit and 1% parasitemia in a final volume of 100 µl. After 48 h of incubation under standard culture conditions, plates were harvested and read by the SYBR Green I fluorescence-based method [
34,
35] using a 96-well fluorescence plate reader (Victor, Perkin-Elmer), with excitation and emission wavelengths of 485 and 535 nm, respectively. The fluorescence readings were plotted against antibody concentration and EC
50 values were obtained by visual matching the mAb concentration corresponding to 50% growth inhibition. In cases where parasitemia was determined microscopically, > 2000 cells were counted using Auto count software [
36].
Passive immunization
Passive transfer of immunity was tested using 6–8 weeks old Swiss male mice. In two different experiments, mice were challenged either with
P. yoelii 17XL or with
P. berghei ANKA. Frozen stocks of parasite infected RBCs (pRBCs) were thawed and injected into the mice. The pRBCs obtained from the second round of passage were used to challenge the mice. In the first experiment (Fig.
5a), all mice (five animals in experimental group, three each in pre-immune and no IgG control groups) were administered with ~ 10
5 P. yoelii 17XL infected pRBCs. On the same day, purified antibody (0.75 mg H12E1 in 100 µl) was administered intravenously. Control groups were injected either with 0.75 mg of pre-immune IgGs or with 100 µl of buffer.
In the second experiment (Fig.
5b), two groups of mice (five animals in experimental group and five in control group) were challenged with ~ 10
5 P. berghei ANKA infected mouse pRBCs. H12E1 antibody (1 mg in 100 µl) was injected in experimental group of animals while the control group received 1 mg of purified pre-immune IgGs. In both the experiments, thin blood smears were prepared (by obtaining blood via the tail bleeding method) every alternate day and the parasitemia was measured using
Plasmodium Auto count software [
36]. At each time point > 2000 cells were counted. Percentage parasitemia was determined as an average of three samples and plotted with standard deviation at each time point. Differences in mean parasitemia between the control and experimental groups were analysed with Student’s t-test as described above. If parasitemia rose to ≥ 50–60% and the mouse showed severe clinical symptoms, the animal was euthanized.
mAb-rPfeno interaction using Surface Plasmon Resonance
Surface Plasmon Resonance (SPR) measurements were made using Biacore T20 machine version 2 control and evaluation software (GE Healthcare Life Sciences, Sweden) located at IIT, Mumbai. All measurements were made at 25 °C. Running buffer consisted of 10 mM HEPES containing 150 mM NaCl, 3 mM EDTA and 0.005% (w/v) P20 surfactant, pH 7.4. rPfeno was immobilized covalently on the surface of CM5 sensor chip using amine coupling with the target response unit of 1000 Rmax. The process involving steps of activation, immobilization and blocking were carried out. Briefly, using a flow rate of 10 μl/min, the chip surface was activated by injecting freshly prepared 1:1 mixture of EDC and NHS (both dissolved in water as per the manufacturer’s instructions). Subsequently, 30 μg/ml rPfeno in 10 mM sodium acetate (pH 5.5) was passed through the active flow cell for 90 s at a flow rate of 30 μl/min. The remaining activated carboxy methyl groups on the surface were blocked by a 7-min injection of 1M-ethanolamine-HCl, pH 8.5. An unmodified flow cell surface was used as a reference for each analysis to check for the non-specific binding response to dextran matrix.
For the measurement of kinetics of interaction between rPfeno and mAb, the analyte (mAb H12E1) was diluted with HEPES buffer and injected over the rPfeno-immobilized chip for 90 s at a flow rate of 30 µl/min followed by a final 240 s dissociation phase. Several concentrations of analyte ranging between 0.78 and 100 nM were used. Regeneration was done with 10 mM NaOH at a flow rate of 30 µl/min for 30 s. Data were evaluated with BIA evaluation software (GE Healthcare Life Sciences, Sweden version 2.0). The sensograms obtained at each antibody concentration was fitted with 1:1 bivalent kinetics for both the association and dissociation phases. Equilibrium dissociation constant (KD) was then calculated from the dissociation and association rate constants.
Molecular modelling
3D-structure of Pfeno was modelled based on the X-ray crystallographic structure of
T. gondii Enolase1 (TgENO1; PDB: 3OTR) [
37] using the Automated Mode of Swiss Model PDB viewer 8.05 (
http://swissmodel.expasy.org/). UniProt/Swiss-Prot database was used to obtain the amino acid sequence of Pfeno (UniProt Accession No. Q27727, 446 amino acids). Energy minimization of the optimized model structure was achieved using the software: MOE 2016.08 (Chemical Computing Group, Canada). PyMOL 1.3 [
38] was used to visualize the modelled structures.
Discussion
Enolase is a glycolytic enzyme with a highly conserved structure and multiple moonlighting functions [
47,
48]. Several pathogens have enolase expressed on their cell surface where it assists in host tissue invasion [
49‐
54]. In
Plasmodium, enolase localizes in multiple organelles (diverse sub-cellular localization) where it is likely to participate in different physiological functions [
18]. Presence of Pfeno on the surface of sporozoites, merozoites and ookinetes raised the possibility of its involvement in host cell invasion. Recently, anti-Pfeno antibodies have been shown to inhibit growth at erythrocytic and mosquito stages presumably by blocking the invasion by merozoites and ookinetes [
19,
20,
23,
50]. Antibody induced inhibition of ookinete was mediated through the plasminogen receptor function of Pfeno, [
19,
50] while the merozoite inhibition appeared to be mediated by blocking a unique epitope, EWGWS [
23,
24]. Data presented here further validated that EWGWS is the target for the merozoite neutralizing antibodies. The inhibitory antibodies exert their effect on ookinete by blocking the binding of plasminogen to
277DKSLVK
282 domain of Pfeno, while merozoite inhibitory antibody (H12E1) functions through the
104EWGWS
108 epitope. The underlying molecular mechanisms in the two cases are likely to be quite different. In ookinetes, Pfeno functions as a receptor for plasminogen as well as a ligand for mosquito mid-gut epithelial cell receptor [
19], whereas in merozoites, it is the EWGWS epitope of Pfeno that serves as a ligand for some yet to be identified receptor on human erythrocyte surface. The effect of anti-Pfeno antibodies on sporozoite invasion of hepatocytes has not yet been examined. It will be interesting to see whether anti-Pfeno antibodies can disrupt sporozoite invasion of hepatocytes.
The observed inhibition of growth shown by H12E1 mAb was ≥ 90% at < 100 µg/ml. There are very few reports where such potent inhibition of merozoite invasion in in vitro cultures has been reported. Some of the mAbs raised against
P. falciparum reticulocyte binding protein homolog 5 (PfRH5) had shown similar inhibition [
55]. H12E1 antibody was quite effective against multiple species of
Plasmodium (
P. yoelii, P. berghei and
P. falciparum) and different strains of
P. falciparum (3D7 and INDO). Given the highly conserved nature of the target sequence and its location in the polypeptide (Fig.
7c), it is not surprising that the antibody has species and strain transcending parasite neutralization activity. The ability of this antibody to block erythrocyte invasion with high efficiency and inhibit growth in heterologous species and strains other than
P. falciparum 3D7 makes it a promising candidate for a blood stage anti malaria vaccine. One of the major limitations of this epitope is the poor immunogenicity of EWGWS. As all plant enolases have this epitope [
22,
56], humans (or other vertebrate hosts) get exposed to this antigen early on in their life (through food). There have been suggestions that oral tolerance to food antigens could induce systemic suppression [
57]. Such suppression could be one of the factors responsible for poor immunogenicity of EWGWS. Observation of partial protection in EWGWS immunized mice was largely due to low titers of induced antibodies [
23]. Thus, for greater efficacy in protection against malaria, this epitope will need to be presented in a highly immunogenic context, employing novel methods of antigen presentation [
55,
58‐
61].
Several antigens derived from the three stages of parasite life cycle (i.e. pre-erythrocytic, erythrocytic and mosquito stages) have been tested for their ability to protect against malaria and block its transmission [
8,
9,
62]. Some of the erythrocytic stage antigens, mostly of merozoite origin, have undergone phase I/II level trials [
12]. Apical membrane antigen 1 (AMA-1) that is expressed by both sporozoites and merozoites plays a key role in invasion of hepatocytes and erythrocytes, showed strain specific reduction in malaria cases, but offered no clinical protection [
13,
63,
64]. Other vaccine candidates such as Merozoite Surface Protein 1 (MSP 1), Erythrocyte Binding Antigen Region II (EBA 175) etc. have also undergone Phase I and Phase II level trials [
65‐
67]. In all these efforts, the antibody response failed to cover antigenic polymorphism and yielded only low efficacy allele specific protective response [
12]. Multi-stage vaccines containing candidate antigens from multiple stages have also been tested [
68,
69]. Extensive efforts to cover all genetic variations in field strains have not succeeded mostly due to the failure to overcome the issue of polymorphism. Stage specific expression, high degree of polymorphism and requirement for very high titers of induced antibodies has hindered any meaningful outcome from these efforts. Recently, PfRH5 has emerged as a leading antigen for erythrocytic stage vaccine [
70]. PfRH5 is a merozoite adhesin essential for erythrocyte invasion and is susceptible to vaccine inducible strain transcending parasite neutralizing antibodies [
71‐
73]. PfRH5 has critical non-redundant function during merozoite invasion of erythrocytes [
74] and does not appear to be a target of antibodies acquired in natural immunity [
72]. Results from a recent Phase Ia trial have been very promising [
75]. There are some interesting parallels between Pfeno and PfRH5. Both antigens have a high degree of sequence conservation (none or low genetic polymorphism) [
73,
76‐
78], both are targets for strain transcending parasite neutralizing antibodies, mAbs directed towards certain specific epitopes in these proteins have shown strong growth inhibition [
55] and both appear to be under low level natural immune pressure. However, an added advantage with Pfeno is that it has cell surface expression at all the three invasive stages [
18], two (merozoites and ookinetes) of which are known targets of parasite neutralizing antibodies [
19,
20,
23]. Thus, it could be developed into a ‘two stage’ (erythrocytic and transmission blocking) vaccine candidate. The possibility of anti-Pfeno antibodies targeting sporozoites should also be explored.
For the development of a multivalent vaccine that is effective against multiple species and strains of the parasite, with an ability to neutralize growth at multiple stages, one will need to combine several antigens [
15,
17,
62]. Epitope based approaches for each target antigens are also anticipated in near future. Hence, antigens like Pfeno with identified protective epitopes, ability to neutralize multiple species/strains and with potential to target multiple stages in parasite life cycle could be the promise of future.
Authors’ contributions
GKJ: conceived the project and designed the study, planned laboratory work, data analysis, writing and editing of the manuscript: DS: study design, planned laboratory work, data analysis and editing; SD: planned experiments, mAb screening, IgG purification, mAb specificity, nanoparticle experiment, GIAs, data analysis, figure preparations; AT: mAb screening; CB: IgG purification, passive immunization, coordinating collaboration, statistical analysis; MR and PA: mAb concentration dependence of GI, GIA strain effects; RV: ELISA, SPR measurements, analysis; AM: Molecular modelling, ELISA measurements with variants, statistical analysis, figure preparations, manuscript editing. All authors read and approved the final manuscript.