Introduction

The surface membrane of all plasmodial sporozoites is covered by an antigen designated the circumsporozoite (CS) protein. The CS proteins have a central immunodominant region consisting of short tandem repeat amino acid sequences that contain multiple copies of the immunodominant B cell epitope (Arnot et al. 1985). Because they are highly immunogenic and can induce a protective response in sporozoite-immunized experimental animals and men, the CS proteins are being investigated as candidates for a human malaria vaccine. The immunodominant B cell epitopes of the CS proteins of a large number of isolates of Plasmodium falciparum of diverse geographical origin and of a smaller number of isolates of P. vivax have been examined and were found to be conserved within the species (Shi et al. 1992). It was first reported that a strain of P. vivax containing a variant repeat in its CS protein was isolated in Thailand (Rosenberg et al. 1989). The CS repeat of this variant strain (Thai VK247) differs at six of the nine amino acids contained in the repeat sequence found in all previously described P. vivax CS proteins. After that discovery, several studies have been conducted to evaluate the global distribution of variant VK247; it was detected in indigenous populations of China (Han et al. 1999), Brazil (Branquinho et al. 1993), Mexico (Kain et al. 1992), Peru (Need et al. 1993), and Papua New Guinea (Kain et al. 1992). In South Korea, P. vivax has also prevailed from 1993 on after a two decade absence (Chai et al. 1994; Cho et al. 1994; Paik et al. 1988; Park et al. 2003). Knowledge of the type of P. vivax that has prevailed in South Korea has been crucial for proper antimalaria treatment, because it is known that drug susceptibility is slightly different between VK210, which has GDRA(D/A)GQPA repeats (Arnot et al. 1985), and VK247 (Rosenberg et al. 1989). It is also known that Anopheles albimanus and An. pseudopunctipennis differ in their susceptibilities to P. vivax circumsporozoite phenotypes. An. albimanus is more susceptible to VK210 type and An. pseudopunctipennis is more susceptible to VK247 (González-Cerón et al. 1999). An. sinensis, An. pullus, An. kleini, An. lesteri, and An. yatsushiroensis are vector mosquitoes in Korea (Ree et al. 1967; Shin et al. 2002; Lee et al. 2007). However, the susceptibility of different mosquitoes to the subtypes of P. vivax is not well known.

To develop the diagnostic systems for identification of malaria patients in Korea based on presence of the CS protein dominant and variant forms of CS protein are needed. In order to do this, in a previous study we constructed the recombinant CS protein from a Korean isolate of VK210 (Lee et al. 1999, 2000; Kim et al. 2002). Thus, in this work, we cloned the VK247 gene expressed it in Escherichia coli and characterized its antigenicity to determine its usefulness for diagnosis of patients infected with the VK247 variant subtype of P. vivax.

Materials and methods

Malaria survey in Myanmar

Patients with clinically suspected malaria attending the Central VBDC (Vector Borne Diseases Control) clinic in Yangon, Myanmar, were examined for malaria parasites. Thin and thick blood smears were prepared from the blood collected for microscopic examination (magnification 7 × 100) from the fingertips of individuals. A local health unit employee read the Giemsa-stained blood smears, and treatment was administered to those individuals who tested positive for malaria, based on the guidelines of the Department of Health, the Union of Myanmar. Before treatment, additional blood samples of approximately 3 ml were collected from each individual whose infection was confirmed by microscopic examination. The blood samples were transferred to the National Institute of Health, Korea Centers for Disease Control and Prevention, Korea, for further polymerase chain reaction (PCR) and genotyping. Informed consent was obtained from all patients. The study protocol was approved by the Department of Health (Upper Myanmar), the Union of Myanmar.

Amplification of the CS protein gene

For the purpose of expression of the variant CS protein gene, genomic DNA was extracted from the whole blood of a malaria patient using a QIAamp Blood Kit (Qiagen, Hilden, Germany). PCRs were performed using AccuPower PCR Premix (Bioneer, Taejeon, South Korea), 50 ng purified genomic DNA, 40 pmol each of forward (F1; 5′-TCCCCACGCACTGCGGGCACAAT-3′) and reverse primers (R1; 5′-TTAATATGCACCGTGAGGACGCC-3′), and the total volume was adjusted to 20 μl with distilled water. The F1 primer contained nucleotides coding for the signal peptide and for the first four amino acids of the mature protein, and the R1 primer contained the 21 bases 3′ to the stop codon, so this primer set can amplify the entire mature CS protein. The thermocycler conditions were as follows: denaturation at 94°C for 5 min, 35 cycles of 1 min at 94°C, 1 min at 49°C and 2 min at 72°C, and finally incubation at 72°C for 5 min. The second PCRs were performed with the same conditions, except for adding the first PCR products (2–5 μl) instead of genomic DNA and a different forward (F2; 5′-AAAAAGGATGGAAAGAAAG-3′) and reverse (R2; 5′-GACTTTTCATTTGGGGCA-3′) primer set (Mann et al. 1994). To exclude the blood samples that were coinfected with P. falciparum, we also performed PCR with a P. falciparum-specific primer set to amplify ring-infected erythrocyte surface antigen (RESA) (Wooden et al. 1992), with forward (5′-GATCAAGGAGGAGAGAACC-3′) and reverse primers specific for that gene (5′-CAGCATTAACACCAACACC-3′). All PCR products were analyzed on a 1.2% agarose gel, confirmed under a UV transilluminator and purified with a NucleoSpin Extract Kit (Macherey-Nagel, Duren, Germany). The purified PCR products were ligated with pCR2.1 cloning vector (Invitrogen, Carlsbad, CA, USA) and then transformed into E. coli INVα F′ according to the procedures of Invitrogen.

Genotyping of PCR products by DNA hybridization

Ten microliters of each 20 μl PCR reactions was blotted onto a nylon membrane using a slot blotter. Two oligonucleotides, AL116 (5′-GGTGATAGAGCAGATGGA-3′) for VK210 and AL114 (5′-ATCAACCAGGAGCAAATG-3′), for VK247 were used as probes for the DNA hybridization (Qari et al. 1994). These probes were end-labeled with 35S-α-dATP using a 3′-end labeling kit (Amersham Pharmacia Biotech, Uppsala, Sweden) for 1 h at 37°C. The radiolabeled oligonucleotides were separated from unincorporated nucleotides on a Sephadex column (Amersham Pharmacia Biotech). Nylon membranes were soaked in denaturation buffer for 10 min, washed with distilled water, and then incubated with neutralization buffer for 15 min and twice with freshly prepared buffer. After the nylon membranes were incubated in 5 × SSC (0.75 M sodium chloride and 0.075 M sodium citrate) for 5 min, they were prehybridized for 1 h at 42°C in 5 × SSC, 1% SDS, 1 × Denhart’s solution (1% BSA, 1% Ficoll 400, and 1% polyvinyl pyrrolidone), and 5% sodium pyrophosphate. Hybridizations were done in the same buffer for each radiolabeled nucleotide for 12 h at 42°C. Each nylon membrane was coreacted with anti-AL116 (5′-CCACTATCTCGTCTACCT-3′) and anti-AL114 (5′-TAGTTGGTCCTCGTTTAC-3′) as positive controls. After hybridization, nylon membranes were washed three times for 5 min each at 42°C in 5 × SSC and 1% sodium pyrophosphate. These nylon membranes were air-dried and exposed to X-ray film (Kodak X-Omat) with an intensification screen at −80°C for 72 h (Sallenave-Sales et al. 2000; Torres et al. 2006).

DNA sequencing and analysis

After genotyping of P. vivax, several candidates containing the variant CS protein gene were selected for DNA sequencing. After gel extraction with a Qiagen gel extraction kit, PCR products were ligated with pCR2.1 cloning vector and transformed into E. coli INVαF′. The PCR product inserted E. coli was selected on ampicillin and on X-gal containing medium (Na et al. 2005). To confirm the transformants, gel electrophoresis was done with EcoRI digestion products after preparation of the plasmid with a Qiagen plasmid isolation kit, according to the protocol supplied by the manufacturer. The variant CS protein gene sequence was determined using ABI PRISM dye terminator cycle sequencing ready reaction kit FS (Perkin Elmer, Cambridge, MA, USA) according to the supplied manual. M13 reverse and M13 forward (−20) primers were used in sequencing. Nucleotide and deduced amino acid sequences were analyzed using EditSeq and Clustal in the Megalign program, a multiple alignment program in the DNASTAR package (DNASTAR, Madison, WI, USA). The internet-based BLAST search program of the National Center for Biotechnology Information (NCBI) was used to search protein databases. The sequence of YM19 from Myanmar isolate was deposited in GenBank (Accession No. AF316584).

Construction of the variant CS protein expression vector

For the expression of variant CS proteins in E. coli, the variant CS protein gene fragment was amplified from a blood sample that was confirmed to be infected with variant type of P. vivax described as above, except for adding the Fex2 (5′-ggatccAAAAAGGATGGAAAGAAAG-3′) and Rex2 (5′-aagcttGACTTTTCATTTGGGGCA-3′) primer set, which have the BamHI and HindIII sites on their 5′ ends, respectively, instead of the F2 and R2 primer set in the second PCR reaction. Amplified PCR products were digested with BamHI and HindIII, purified with a Qiagen Gel Extraction Kit after being run on an agarose gel, and were then integrated into the BamHI and HindIII cleavage sites of the pQE30 expression vector (Qiagen). The resulting plasmid was subsequently used for the expression of CS-(His)6 fusion protein in E. coli SG13009 cells (Barfod et al. 2009). Transformants were confirmed both by gel electrophoresis of plasmid DNA after restriction enzyme digestion with BamHI and HindIII and by DNA sequencing.

Expression and purification of recombinant CS protein

Expression of the recombinant protein was induced in E. coli with isopropyl-1-thio-β-d-galactopyranoside (IPTG) (Lee et al. 1999) and purification of the CS-(His)6 fusion protein was carried out using immobilized metal ion affinity chromatography (Lim et al. 2002). The purification was done under native conditions according to the supplier’s protocol (Qiagen). Proteins were analyzed on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in each purification step.

Western blot analysis

Recombinant CS-(His)6 fusion protein was separated on a 12% SDS-PAGE gel and was then transferred to a nitrocellulose membrane. After the transfer, the membrane was cut into strips and blocked for nonspecific binding with 3% skim milk for 12 h at 4°C. The membrane was then washed in PBS with 0.15% Tween 20 for 3 × 10 min. The strips were allowed to react with sera of malaria patients or of uninfected people (diluted 1:100, vol/vol) for 4 h and then washed using the procedure described above. After this the membrane was incubated with diluted peroxidase-conjugated goat anti-human IgG secondary antibody (1:1,000, v/v) (Sigma) for 3 h at room temperature. For color development, a solution containing 0.2% diaminobenzidine and 0.02% H2O2/PBS was applied to each well (Tsang et al. 1983; Gao et al. 2009).

Results

Selection of VK247 P. vivax

We collected 100 blood samples from Yangon (a) and Mandalay (b) in Myanmar (Fig. 1). Most patients were infected with P. falciparum and/or P. vivax dominant form VK210 (Kim et al. 2010). Among them, only one patient was positive for variant form VK247 circumsporozoite protein by DNA hybridization assay. This gene (YM19, Yangon, Myanmar isolate) was studied in this research. This sequence was deposited in Genbank (Accession No. AF316584).

Fig. 1
figure 1

Geographical locations of sample collections of Myanmar: a Yangon, b Mandalay

Full size image

DNA sequence of VK247 Myanmar isolate

The variant CS protein gene that was amplified by PCR from genomic DNA was analyzed on a 1.2% agarose gel. Amplification of the variant CS protein gene yielded an approximately 720-bp DNA fragment which, after purification, was ligated with pCR 2.1 cloning vector (3.9 kb). The transformants were confirmed as bearing PCR inserts by EcoRI digestion. The plasmid containing the PCR product was named pVK247 (4.6 kb) and was used for DNA sequence analysis. As a results of DNA sequencing, the cloned variant CS protein gene (YM19) was shown to be 717 bp and consisted of 239 amino acids that were deduced by DNASIS (Fig. 2). The nonapeptide repeat unit consisted of AN(G/E)A(G/D)(N/D)QPG and was repeated 19 times. In particular, this Myanmar isolate has a different amino acid in the third position of the nonapeptide repeat, Usually the third amino acid is a conserved G that is coded for by GGA or GGG, but that isolate has an E instead a G, which is coded by GAG. The amino acid sequence of YM19 has shown 98.6% homology with a Thai isolate (Genbank accession No. M28745), 93.8% with a PNG isolate (M69059), 95.7% with BzlB19-2, a Brazil isolate (M69061), 47.9% with CH-5, a China isolate (U08979), 40.5% with Ph-79, a Philippines isolate (U08981) and 39.7% with SOL101, a Solomon isolate (U08983) (Figs. 3 and 4).

Fig. 2
figure 2

The gene sequence and deduced amino acid sequence of the circumsporozoite protein gene (VK247 type) of a Myanmar isolate of P. vivax

Full size image
Fig. 3
figure 3

Multiple amino acid sequence alignment. The deduced amino acid sequence of YM19 (AF316584) was aligned with those from other P. vivax isolates. The dashes represent gaps introduced to maximize alignment. PNG: P. vivax PNG strain (PNG, Accession; M69059), BzlB19-2: P. vivax BzlB19-2 strain (Brazil, Accession; M69061), Thai: P. vivax Thai strain (Thailand, Accession; M28745), SOL101: P. vivax SOL101 strain (Solomon, Accession; U08983), CH-5: P. vivax CH-5 strain (China, Accession; U08979), Ph-79: P. vivax Ph-79 strain (Philippines, Accession; U08981)

Full size image
Fig. 4
figure 4

Phylogenetic relationships between the circumsporozoite protein of several strains of P. vivax. Computer analysis was performed using the multiple sequences alignment of MegAlign. All amino acid sequences were obtained from GenBank BLAST (http://WWW.ncbi.nlm.nih.gov). YM19: CSP of P. vivax Myanmar isolate, PNG: P. vivax PNG strain (PNG, Accession; M69059), BzlB19-2: P. vivax BzlB19-2 strain (Brazil, Accession; M69061), Thai: P. vivax Thai strain (Thailand, Accession; M28745), SOL101: P. vivax SOL101 strain (Solomon, Accession; U08983), CH-5: P. vivax CH-5 strain (China, Accession; U08979), Ph-79: P. vivax Ph-79 strain (Philippines, Accession; U08981)

Full size image

Expression of variant repeat CS protein in E. coli

To generate the expression plasmid, the variant repeat region of the CS protein gene was amplified from YM19 patient DNA digested with BamHI and HindIII, and subcloned into the same restriction enzyme sites of expression vector pQE30 (Qiagen). The resultant plasmid pCS247 contained the variant repeat region of the CS protein gene fused to a (His)6-tag. The recombinant plasmid pCS247 was then transferred into E. coli SG13009. Next, 1 mM IPTG was added to cultures of E. coli SG13009 (pCS247) grown to logarithmic phase in liquid LB plus 100 μg/ml ampicillin and 25 μg/ml kanamycin to induce expression of the target protein. As analyzed on SDS-PAGE followed by Coomasie blue staining, the molecular weight of variant CS recombinant protein was 50 kDa under native purification conditions (Fig. 5). The molecular weight of the target protein was shown to be twice as large as the expected molecular weight (24 kDa), which was deduced by DNASIS.

Fig. 5
figure 5

Patterns of purified circumsporozoite protein by SDS-PAGE under native conditions using Ni-NTA agarose resin. Lanes: 1 crude cell lysate, 2 flow-through, 3 20 mM imidazole wash, 4 250 mM imidazole elution, M protein size marker

Full size image

Antigenicity of variant CS recombinant protein

To determine the antigenicity of this recombinant protein, a Western blot was done. Twenty-one of 29 malaria patients (72.4%) were positive by Western blot while the sera from the normal control patients (n = 15) who had never been exposed to malaria tested negative (Fig. 6).

Fig. 6
figure 6

Western blot analysis of VK247 recombinant protein. a Malaria patients, b normal person

Full size image

Discussion

P. vivax has presumably been prevalent in Korea for a long time. However, as a result of a national malaria eradication program with help from the World Health Organization (WHO), the incidence of vivax malaria has rapidly decreased (National Malaria Eradication Service 1966; Paik et al. 1988). After the latest report of two malaria patients in 1985 (Soh et al. 1985), there were no additional reported cases until 1993 when one case was reported (Chai et al. 1994). Malaria cases then rapidly increased until around 2000 (Cho et al. 1994; Park et al. 2003; Lee et al. 1998). After that, the reported malaria cases were reduced due to several efforts to limit the incidence of malaria for several years. However, it is very regrettable that malaria was not thoroughly eradicated in the Korean peninsula because 2–3% of patients experience failed drug treatment every year, and many travelers and workers come from malaria-prevalent areas, including North Korea. For these reasons, serological diagnostic tools are needed to support both traditional microscopic diagnosis and seroepidemiological studies for estimating the prevalence of malaria in Korea. Thus, we tried to set up a serological diagnostic method based on CS protein because it has several merits. First, it tests for antibodies that have a very short half-life. Most Korean malaria patients lose this antibody within 3 months after contact with infectious mosquitoes (unpublished data), so presence of the antibody might be used as an index to indirectly determine the inoculation rate of parasite-infected mosquitoes each year. Second, it might be used to discriminate between patients who experience a long-term vs. a short-term incubation period, by duration of antibody production against CS protein; this is due to the absence of malarial vectors over the long winter in Korea. According to our data, patients who show the long-term incubation period do not have an antibody to combat CS protein from either VK210 or VK247, but patients with short-term incubation have the antibody (data not shown). Over a 5-month period from late October to late March, mosquitoes cannot survive due to low temperatures (under 16°C). It is very important for estimating the extent of malaria prevalence for the next year. Therefore, we attempted to set up a diagnostic method based on CS protein. The CS protein of VK210, which is a main genotype of malaria in Korea, had already been produced by the authors alongside the genome sequence of the Korean isolate (Lee et al. 1999). However, it was suggested that VK247 might prevail in Korea by Lee et al. (2002) and Coleman et al. (2002), but they showed this possibility only in mosquitoes and not in human hosts. Thus, we tried to isolate this genotype from Korean isolates, but did not find it easily. Therefore, we choose to survey the malaria-infected population of Myanmar. Malaria is ranked as the first public health problem in Myanmar, and nearly 600,000 malaria patients attend health institutions in the country annually. Among the malaria species in Myanmar, P. falciparum accounts for 80% of cases, while P. vivax makes up 17.8% and the remaining cases are due to P. malariae with mixed infections (Department of Health, Myanmar 1997). Fortunately, we could find one VK247 CS protein gene. Most samples we obtained were highly contaminated with other species or genotypes (Kim et al. 2010). Additionally, these samples have complex antibodies against P. vivax, P. falciparum, and the subtypes of P. vivax, VK210 and VK247. This means that malaria prevalence in Myanmar is a complicated situation. These were the main problems in finding a pure VK247 genotype among 100 blood samples. As a results of sequence analysis of YM19, we found that, although VK247 of Myanmar has high homology to the Thai strain (Figs. 3 and 4), it showed a different number of nonapeptide repeats. That is, the Thai strain showed 18 repeats of ANGA(G/D)(N/D)QPG (Rosenberg et al. 1989), but the Myanmar samples showed 19 repeats of AN(G/E)A(G/D)(N/D)QPG. We could not state that this characteristic repeat pattern was an absolutely unique feature of Myanmar isolates because it was difficult to obtain pure VK247 CS protein gene. Therefore, more intensive studies should be done in Myanmar to more clearly characterize variant types of CS protein genes. To express the variant CS protein gene in E. coli, we used the defined YM19 that was isolated from Yangon, Myanmar. The recombinant CS protein expressed well in E. coli SG13009 and was successfully purified with Ni-NTA agarose chromatography (Fig. 5). Additionally, it had antigenicity in malaria patients (Fig. 6). Therefore, this recombinant protein can play a crucial role in determining the prevalence of VK247 malaria in Korea.