The blood samples used in this study were obtained from Korean patients infected with P. vivax in Korea between 1998–2013 [14]. All the patients inhabited in malaria endemic areas, i.e., Ilsan, Kimpo, Paju, Gangwha, or Yoncheon, and have not been abroad at least in recent 2 years when their blood samples were collected. All the patients were confirmed to be infected with P. vivax by microscopic examinations and PCR. Monoclonal P. vivax infections were confirmed by genotyping of P. vivax positive samples at the highly polymorphic loci MS16 and MSP-1F3 as described previously [16]. Blood samples were collected under protocols reviewed and approved by the Ethics Committees of Inha University School of Medicine and Gachon University of Medicine and Science. Blood samples were stored as aliquots of heparinized blood frozen at −70°C.

Genomic DNA was extracted from 200 μL of each blood sample using a QIAamp DNA Blood Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions.

Polymorphic regions of MSP-1 (block 5 to 6) and MSP-3α (block I and II) were amplified with PCR using specific primers for each gene [17–19]. ExTaq DNA polymerase (Takara, Otsu, Japan) was used in all PCR amplifications to minimize possible amplification error.

Based on nucleotide sequences of 255 MSP-1 and 255 MSP-3α genes (GenBank accession nos. KU893351 to KU893860) amplified from Korean P. vivax isolates [14], we surveyed the restriction enzyme sites in each subtype of MSP-1 and MSP-3α using the Restriction Mapper program (http://www.restrictionmapper.org/). The restriction enzymes with digestion sites that were conserved within a subtype in a given sequence were selected. The selected restriction enzymes and their predicted cleavage sites in each subtype of MSP-1 and MSP-3α were shown in Fig. 1A and Fig. 3A. PCR products of each subtype MSP-1 were digested with PvuII, SpeI, and StyI (each 3 U; Takara) in a total volume of 30 μL at 37°C for 2 hr. RFLP analysis of each subtype of MSP-3α was carried out by digesting the PCR products with the PvuII, StyI, and EcoRI (each 3 U; Takara) in a total volume of 30 μL at 37°C for 2 hr. The resulting digested fragments were separated in 3% agarose gels by electrophoresis in 0.5×Tris-acetate-EDTA (TAE) buffer, stained with RedSafe Nucleic Acid Staining Solution (Intron, Seongnam, Korea) and visualized under ultraviolet light. The 100-bp DNA ladder or 1-kb DNA ladder (Life Technologies, Rockville, Maryland, USA) was used as the DNA size marker.

To evaluate the applicability of the developed PCR-RFLP methods for subtyping of MSP-1 and MSP-3α in Korean P. vivax isolates, a blind test was performed with 50 Korean P. vivax isolates. The PCR reactions were set to the same conditions as above, and the PCR products were subsequently digested with the corresponding restriction enzymes for MSP-1 and MSP-3α. The reactants were analyzed by 3% agarose gel electrophoresis as described above and observed the profiles of restriction fragments. Sequencing analysis of amplified MSP-1 and MSP-3α was also performed for each P. vivax isolate by automatic nucleotide sequencing to evaluate the accuracy of the PCR-RFLP methods.

On the basis of computational restriction fragment analysis of currently known 7 MSP-1 subtypes in Korean P. vivax isolates, restriction enzymes PvuII, SpeI, and StyI were chosen, as they were expected to produce fragments of different sizes and subtype-specific restriction profiles (Fig. 1A). PCR amplifications of polymorphic regions (block 5 to 6) of 7 distinct Korean P. vivax MSP-1 subtypes (subtypes A to G) revealed that the amplified products showed only slightly different length polymorphisms between the subtypes (subtype A, 588 bp; subtype B, 591 bp; subtype C, 591 bp; subtype D, 495 bp; subtype E, 495 bp; subtype F, 498 bp; subtype G, 507 bp), which did not clearly discriminated each other on a gel (Fig. 1B upper panel). Digestion of the amplified PCR products with PvuII, SpeI, and StyI resulted in different sets of digestion profiles between the subtypes as expected (Fig. 1B, lower panel). Subtypes A and B generated 2 fragments of approximate sizes of 450 bp and 140 bp, while subtype C yielded 3 fragments of 360, 140, and 90 bp in their approximate sizes. Subtype D produced 3 different sized fragments with approximate 260, 140, and 100 bp. Subtype E generated 4 digested fragments with approximate sizes of 170, 140, 100, and 90 bp. In the cases of subtypes F and G, 3 fragments with approximate lengths of 250, 170, and 90 bp were produced. These results showed that the 7 MSP-1 subtypes in Korean P. vivax isolates produced different RFLP profiles that can be distinguished each other, even though Sal I type subtypes (subtypes A and B) and Belem type subtypes (subtypes F and G) did not clearly differentiated due to their high sequence identities. To evaluate the applicability of the PCR-RFLP, a blind analysis using 50 Korean P. vivax isolates was performed. As expected, distinct RFLP profiles for each subtype were successfully obtained from the analyzed samples (Fig. 2). The MSP-1 amplicons were sequenced for validation of the method and confirmed that the accuracy of this method was 100% in the analyzed 50 samples.

For subtype-specific PCR-RFLP method for MSP-3α, PvuII, StyI, and EcoRI restriction enzymes were selected by in silico restriction fragment analysis (Fig. 3A). PCR amplifications of polymorphic regions (block I and II) of the 7 Korean P. vivax MSP-3α subtypes (subtypes 1 to 7) resulted in different sizes of amplicons, but they were not clearly discriminated into the corresponding subtypes on a gel (Fig. 3B, upper panel). Amplifications of MSP-3α followed by RFLP analysis using the 3 restriction enzymes generated subtype-specific fragments of different numbers and sizes that can easily be discriminated into each subtype on an agarose gel (Fig. 3B, lower panel). Subtype 1 produced 3 fragments with approximate lengths of 1,280, 340, and 310 bp. Subtype 2 generated 2 different sized fragments of about 1,600 and 310 bp. Subtype 3 produced 2 digested fragments with approximate sizes of 790 and 340 bp. Subtype 4 yielded 3 distinct sized fragments of about 1,030, 530, and 330 bp, while subtype 5 generated only a band of approximate size with 1,140 bp since it had not any cleavage site by the 3 restriction enzymes. In the case of subtype 6, 2 different sized restriction fragments with approximate sizes with 240 and 1,030 bp were produced. Meanwhile, subtype 7 produced 2 fragments with approximate sizes of 900 and 1,020 bp. A blind analysis of 50 Korean P. vivax isolates for MSP-3α showed that they are well discriminated into each subtype (Fig. 4). Further sequencing analysis of each MSP-3α amplicon to validate the accuracy of this method confirmed that the method showed 100% concordance for the 50 tested P. vivax Korean isolates.