Identification of a novel merozoite surface antigen of Plasmodium vivax, PvMSA180

Although a number of Plasmodium vivax proteins have been identified, few have been investigated as potential vaccine candidates. This study characterized the Plasmodium vivax merozoite surface antigen 180 (PvMSA180, PVX_094920), a novel P. vivax antigenic protein. The target gene was amplified as four overlapping domains (D1, D2, D3 and D4) to enable expression of the recombinant protein using cell-free and bacterial expression systems. The recombinant PvMSA180 proteins were used in protein microarrays to evaluate the humoral immune response of 72 vivax-infected patients and 24 vivax-naïve individuals. Antibodies produced in mice against the PvMSA180-D1 and -D4 domains were used to assess the subcellular localization of schizont-stage parasites with immunofluorescence assays. A total of 51 pvmsa180 sequences from 12 countries (41 sequences from PlasmoDB and 6 generated in this study) were used to determine the genetic diversity and genealogical relationships with DNAsp and NETWORK software packages, respectively. PvMSA180 consists of 1603 amino acids with a predicted molecular mass of 182 kDa, and has a signal peptide at the amino-terminus. A total of 70.8% of patients (51/72) showed a specific antibody response to at least one of the PvMSA180 domains, and 20.8% (15/72) exhibited a robust antibody response to at least three of the domains. These findings suggest that PvMSA180 is targeted by the humoral immune response during natural infection with P. vivax. Immunofluorescence analysis demonstrated that PvMSA180 is localized on the merozoite surface of schizont-stage parasites, and pvmsa180 sequences originating from various geographic regions worldwide showed low genetic diversity. Twenty-two haplotypes were found, and haplotype 6 (Hap_6, 77%) of pvmsa180 was detected in isolates from six countries. A novel P. vivax surface protein, PvMSA180, was characterized in this study. Most of P. vivax-infected patients had specific antibodies against particular antigenic domains, indicating that this protein is immunogenic in naturally exposed populations. Genetic analysis of worldwide isolates showed that pvmsa180 is less polymorphic than other well-known candidates and that some haplotypes are common to several countries. However, additional studies with a larger sample size are necessary to evaluate the antibody responses in geographically separated populations, and to identify the function of PvMSA180 during parasite invasion.


Background
Malaria is a public health problem in countries in which it is endemic [1]. In 2015, the number of cases and deaths was 214 million and 438,000, respectively [2]. Plasmodium vivax causes 50% of all malaria cases globally [3], and is prevalent in the tropics and subtropics [4]. A malaria vaccine shows promise for controlling malaria [5]; however, the antigenic diversity and immune-evasion ability of Plasmodium has hampered vaccine development [6]. Molecules expressed on the merozoite surface, such as apical membrane antigen-1 (AMA1), merozoite surface protein-1 (MSP1), and Duffy binding protein, have been the focus of vaccine development efforts [7].
Bioinformatic and genome analysis of P. vivax have led to the identification of malaria antigens, few of which have been investigated as vaccine candidates [8][9][10]. MSPs, such as MSP-1, MSP-9, MSP-4 and MSP-5, have been identified as vaccine candidates [11]. Some hypothetical proteins have been identified as vaccine candidates based on coiled coil structure [10]. Moreover, several proteins of P. vivax that are expressed on the surface or in apical organelles, including MSPs, rhoptryassociated membrane antigen, glycosylphosphatidylinositol (GPI)-anchored micronemal antigen and AMA1, have been proposed as vaccine candidates due to their involvement in merozoite invasion or the longevity of the antibody response [12][13][14][15][16]. Due to the limitations of P. vivax in vitro culture systems, fewer surface proteins have been identified in this pathogen than in P. falciparum. In fact, a number of P. vivax surface proteins have been identified based on their orthologues in P. falciparum [9,10,15,17], and the antibody responses to them have been investigated [18][19][20].
One of hypothetical proteins, named P. vivax merozoite surface antigen 180 (PvMSA180) was previously identified [21]. Of the 96 P. vivax blood-stage proteins, 18 (including PvMSA180) elicited robust antibody responses [21]. Thus, this study has characterized PvMSA180, which is immunogenic in naturally exposed populations, and determined its subcellular localization in P. vivax.

Sample and serum information
Blood samples were collected from 72 patients in 2012 in the Shwe Kyin area of Myanmar (mean age, 24.5 years; range, 18-42 years). The patients were confirmed to have P. vivax malaria using the malaria rapid diagnostic test (SDFK80; Standard Diagnostics, Gyeonggi, Korea) and microscopy. Samples were centrifuged and the serum was separated. Serum samples from 24 healthy malarianaïve individuals residing in non-endemic areas in the Republic of Korea (ROK) were also collected and used as controls.

Amplification of full-length pvmsa180
Parasite genomic DNA was extracted using a Qiagen DNA Extraction Kit (Hilden, Germany). The pvmsa180 (PVX_094092) sequence was obtained from PlasmoDB (http://plasmodb.org/). Full-length pvmsa180 was amplified from five Myanmar and one South Korean isolate using the forward primer 5′-GATGACGACACAA ACAAAAGGG-3′ and reverse primer 3′-CGCGGCGTAG TTGATGTG-5′. Full-length pvmsa180 was amplified by PCR using high-fidelity Thermococcus kodakaraensis (KOD) DNA polymerase (Toyobo, Osaka, Japan) under the following conditions: 2.0 μL DNA template, 0.4 U KOD DNA polymerase, 0.25 mM of each primer and 500 μM of each dNTP, in a final volume of 20 mL. The cycling conditions were 94 °C for 2 min, followed by 35 cycles at 94 °C for 15 s, at 58 °C for 30 s, at 68 °C for 4.5 min, and a final extension at 68 °C for 10 min.

Protein microarray
Slides were coated with amine solution [21] and used to screen the serum of 72 vivax-infected patients and 24 healthy individuals by protein microarray. The crude recombinant proteins expressed using the wheat germ cell-free system were applied in duplicate and incubated for 2 h at 37 °C, followed by blocking in 5% BSA in phosphate-buffered saline (PBS) with 0.1% Tween 20 (PBS-T) and incubation for 1 h at 37 °C. Diluted serum (1:25) was added to a slide harbouring recombinant proteins and incubated for 1 h, followed by addition of an Alexa 647-conjugated goat-anti-human IgG (10 µg/mL, Invitrogen). Then slides were imaged using a fluorescence scanner (ScanArray Express, PerkinElmer, Boston, MA, USA). The cut-off value was equal to the mean fluorescence intensity (MFI) plus two standard deviations (SDs) of the negative samples. Normalized MFI values were calculated from the MFI/cut-off values.

Production of mice antibody
Female BALB/c mice at 6-8 weeks of age (DBL, Seoul, ROK) were injected intraperitoneally with 30 µg recombinant PvMSA180-D1 and -D4 in PBS with complete Freund's adjuvant (Sigma-Aldrich, St. Louis, MO) in a final volume of 100 µL, and subsequently boosted with incomplete Freund's adjuvant. Three mice were injected three times with each of the D1 and D4 fragments of PvMSA180 at 2-week intervals. Sera were collected 2 weeks after the final boost. All the animal experimental protocols were approved by the Institutional Ethics Committee and were conducted in accordance with the Ethical Guidelines for Animal Experiments of Kangwon National University (ROK).

Sequence diversity and haplotype analysis of pvmsa180 in isolates obtained worldwide
The pvmsa180 sequences obtained in this study were compared to those of 46 isolated worldwide in the Plas-moDB database. Pvmsa180 sequences from 12 countries (South Korea, Myanmar, Salvador, Peru, Colombia, Mexico, Thailand, Brazil, India, Papua New Guinea, North Korea, and China) were included in the analysis (Additional file 1). All the raw sequences were analysed and trimmed using the SeqMan software, Lasergene ver. 7.0 (DNASTAR). Sequences were aligned using the CLUSTAL-W program in MegAlign Lasergene ver. 7.0 (DNASTAR) and exported in FASTA format. Sequence diversity (π), defined as the average number of nucleotide differences per site between two sequences, number of polymorphic sites was determined using the DNAsp ver. 5.0 software. The relationships among the haplotypes of pvmsa180 were evaluated with the me-dian-joining method using the NETWORK software ver. 4.6.1.2. (Fluxus Technology Ltd., Suffolk, UK). The C-terminus amino acid sequences of PvMSA180 were also separately aligned using Clustal W2 to assess the sequence homology among two human and simian malaria parasites.

Statistical analysis
Simple scatter regression was applied to construct a standard curve using SigmaPlot (Systat Software Inc., San Jose, CA, USA). Data were analysed using GraphPad Prism (GraphPad, Software, San Diego, CA, USA). The Student's t test was used to assess differences between means. Values of p < 0.05 were considered statistically significant. The MeV software (MultiExperiment Viewer, http://www.tm4. org/mev.html) was used to visualize the antibody responses of individual patients to the PvMSA180 domains.

Schematic structure of PvMSA180
The protein consists of 1603 amino acids with a predicted molecular mass of 182 kDa, and lacks a transmembrane domain and GPI anchor, as determined using the deduced amino acid sequence of P. vivax (Sal-1). The cysteine residues are adjacent to the C-terminus domain. Proteins were expressed without a signal peptide. The D1 domain of PvMSA180 (207-307 amino acids) from worldwide isolates contains a polymorphic region (Fig. 1a). In addition, the C-terminus sequence of PvMSA180 has high sequence homology with the human (P. falciparum) and two simian (Plasmodium knowlesi and Plasmodium cynomolgi) malaria parasites. The amino acid alignment revealed 70% amino acid identity among human and simian malaria parasites (Fig. 1b).

Expression of recombinant PvMSA180 proteins
The four domains of PvMSA180 were expressed using a wheat germ cell-free system, resulting in a single band of ~50 to 75 kDa; all the domains were predicted to be <50 kDa (Fig. 2a). PvMSA180-D1 was successfully Red bars indicate the conserved aa in the four Plasmodium species, green bars in three species, sky-blue bars in two species, and dark-blue bars in one species. Asterisks denote cysteine residues conserved in the four Plasmodium species expressed as a 70.3 kDa GST fusion protein, and PvMSA180-D4 was expressed in E. coli as a 6× Histagged, 52.7 kDa recombinant protein (Fig. 2b, c). SDS-PAGE and Western blotting showed that recombinant PvMSA180-D1 and PvMSA180-D4 have molecular weights of ~80 and ~60 kDa, respectively (Fig. 2b, c). Purified PvMSA180-D1 and PvMSA180-D4 were used to immunize Balb/c mice with the aim of generating polyclonal antibodies.

Recognition of recombinant PvMSA180 proteins
To confirm the immunoreactivity of PvMSA180, purified recombinant PvMSA180-D1 and PvMSA180-D4 were incubated with pooled sera of vivax-infected patients. Bands at ~80 and ~60 kDa for PvMSA180-D1 and PvMSA180-D4, respectively, were detected by Western blotting (Fig. 2d). In addition, to confirm antibody production in mice against the D1 and D4 domains of PvMSA180, recombinant proteins were subjected to Western blotting using polyclonal anti-PvMSA180-D1 and PvMSA180-D4 antibodies (Fig. 2d).

Humoral immune response to PvMSA180
The humoral immune response of 72 vivax-infected patients and 24 healthy individuals to the four domains of PvMSA180 was evaluated by protein microarray. Sera from vivax-infected patients showed significantly higher reactivity than that from healthy individuals (Fig. 3a,  p < 0.0001). The number of healthy individuals and vivaxinfected patients showing a serum immune response to PvMSA180 was determined ( Table 1). The antibody response to the four domains did not differ markedly. However, antibodies to PvMSA180 showed positive reactivity against PvMSA180, and there was no correlation between the humoral immune response and patient age or the presence of parasitaemia.

Discussion
Genomics, proteomics, and transcriptomics have been used to identify target antigens in Plasmodium spp. [10,[21][22][23][24]. This has resulted in the identification of proteins expressed by infected red blood cells on the surface or in apical organelles; however, few such P. vivax proteins have been reported [13]. A small number of promising merozoite antigens has been investigated as vaccine candidates, some of which have shown partial efficacy in clinical trials [13]. MSPs are of interest for the development of vaccines as well as drugs that inhibit blood-stage replication of P. falciparum [25][26][27]. In this study, PvMSA180, a novel surface protein of P. vivax, was characterized.
Most of the cysteine residues are in the C-terminus domain of PvMSA180, which has no transmembrane domain or GPI-anchor. The results of the immunofluorescence assay showed that the PvMSA180 protein is localized on the merozoite surface; thus, it might be categorized as being peripherally associated with the surface of P. vivax merozoites as it lacks transmembrane domains and a GPI-anchor [13]. PvMSA180 orthologs in other species have a conserved domain at the C-terminus that shows ~70% sequence similarity with simian malaria parasites (P. knowlesi and Plasmodium chabaudi), but a lower level of identity to Plasmodium falciparum. Nevertheless, P. vivax is genetically, phenotypically, and biologically more similar to P. chabaudi and P. knowlesi than to P. falciparum [28][29][30][31]. Unlike erythrocyte-binding ligand (ebl) and reticulocytebinding protein (rbp), which are crucial for the invasion of erythrocytes by P. vivax, P. knowlesi, and P. chabaudi, the function of msa180 in other species is unclear. The amplification and sequence analysis of clinical isolates from diverse geographic regions showed the presence of a short polymorphic domain at the N-terminus region of pvmsa180. The polymorphic region of invasion genes is considered under immune selection pressure [32] and plays a role in protective immunity against clinical isolates [32][33][34]. However, in this study, the polymorphic and conserved domains of PvMSA180 showed similar immunogenicity. Furthermore, 20.8% of the patients (15/72) showed a robust antibody response to more than three domains of PvMSA180, and only 30% did not respond to any of the antigens (Fig. 3b). The development of a malaria vaccine based on merozoite surface antigens has been hindered by their high sequence diversity [35,36]. In this study, haplotype network analysis resulted in the detection of 22 haplotypes, 1 of which (Hap_6, 17/22) was shared by six countries. However, a greater number of samples are required to confirm this finding, as serum samples from only Myanmar and the ROK were used in this study. These data will facilitate the development of a subunit vaccine based on the domains conserved among strains isolated worldwide. The functional activity of this antibody could not be performed due to the limitation of P. vivax culture system. It remains to be determined as further study in the future.

Conclusions
In this study, a novel P. vivax surface protein, PvMSA180, was identified. Its immunogenicity in a naturally exposed population suggests that it elicits a humoral immune response, and its localization suggests that it may be involved in merozoite attachment to, and invasion of red blood cells. However, further functional and biological investigations of PvMSA180 protein are required prior to its development as a subunit vaccine. Genetic analysis indicated that the gene is less polymorphic than other well-known candidates and a haplotype common to six countries was identified.