1. The study developed a new PCR/RFLP technique to identify the 3 genotypes of Plasmodium vivax circumsporozoite protein (VK210, VK247, and P. vivax-like) using DNA extracted from blood samples. 2. The technique uses PCR amplification of the central immunodominant region of the CSP gene followed by restriction enzyme digestion and fragment analysis to distinguish the genotypes. 3. Testing demonstrated the technique could accurately identify the genotypes using plasmid controls for each variant, and that it had high sensitivity detecting parasitemia levels as low as 0.0069 parasites per microliter.

1. Available online at www.sciencedirect.com
Diagnostic Microbiology and Infectious Disease 59 (2007) 415 – 419
www.elsevier.com/locate/diagmicrobio
Parasitology
A new polymerase chain reaction/restriction fragment length
polymorphism protocol for Plasmodium vivax circumsporozoite protein
genotype (VK210, VK247, and P. vivax-like) determination
Renata Tomé Alves a,d , Marinete Marins Póvoa b , Ira F. Goldman c , Carlos Eugênio Cavasini d ,
Andréa Regina Baptista Rossit d,e , Ricardo Luiz Dantas Machado d,e,⁎
a
Universidade Estadual Paulista Júlio de Mesquita Filho, 15054-000 São José do Rio Preto, São Paulo State, Brazil
b
Instituto Evandro Chagas, MS/SVS, 67030-000 Ananindeua, Pará State, Brazil
c
Centers for Disease Control and Prevention, Atlanta, GA 30333, USA
d
Centro de Investigação de Microrganismos, Departamento de Doenças Dermatológicas, Infecciosas e Parasitárias, Faculdade de Medicina de São José do
Rio Preto, 15090-000 São José do Rio Preto, São Paulo State, Brazil
e
Fundação Faculdade de Medicina de São José do Rio Preto- FUNFARME, 15090-000 São José do Rio Preto, São Paulo State, Brazil
Received 20 March 2007; accepted 27 June 2007
Abstract
For the molecular diagnosis of Plasmodium vivax variants (VK210, VK247, and P. vivax-like) using DNA amplification procedures
in the laboratory, the choice of rapid and inexpensive identification products of the 3 different genotypes is an important prerequisite. We
report here the standardization of a new polymerase chain reaction/restriction fragment length polymorphism technique to identify the 3
described P. vivax circumsporozoite protein (CSP) variants using amplification of the central immunodominant region of the CSP gene
of this protozoan. The simplicity, specificity, and sensitivity of the system described here is important to determine the prevalence and
the distribution of infection with these P. vivax genotypes in endemic and nonendemic malaria areas, enabling a better understanding of
their phylogeny.
© 2007 Published by Elsevier Inc.
Keywords: Plasmodium vivax genotypes; PCR/RFLP; Diagnosis; Circumsporozoite protein; Plasmodium vivax-like; VK210; VK247
1. Introduction the CSP gene (Gopinath et al., 1994). Rosenberg et al. (1989)
described a variant of P. vivax in Thailand (VK247), and Qari
Plasmodium vivax is the second most prevalent malaria et al. (1993) reported on the P. vivax-like variant in Papua
parasite affecting more than 75 million people each year New Guinea, which morphologically resembles the classic
(Imwong et al., 2005). The circumsporozoite protein (CSP) of form (VK210) and VK247 but has a distinctive repeated
the infective sporozoite is the main target for the development portion of the central region of the CSP gene. Another
of recombinant malaria vaccines (Qari et al., 1993; Gonzales important issue generated by the existence of these genotypes
et al., 2001; Herrera et al., 2007). Nevertheless, the data is the possibility of differential variant-linked responses to
generated require special considerations because of the treatment. The first evidence that P. vivax develops resistance
discovery of sequence variations in the central portion of to chloroquine (CQ) was reported in Papua New Guinea
(Rieckmann et al., 1989) and, consequently, studies carried
out by Kain et al. (1993) suggested that the response to CQ
⁎ Corresponding author. Centro de Investigação de Microrganismos, can vary depending on the P. vivax genotypes. In addition,
Departamento de Doenças Dermatológicas, Infecciosas e Parasitárias, Machado et al. (2003) confirmed a significant correlation
Faculdade de Medicina de São José do Rio Preto, 15090-000 São José do
Rio Preto, São Paulo State, Brazil. Tel.: +55-17-3201-5913; fax: +55-17-
between parasite clearance and its 3 genotypes.
3201-5736. By serological and/or molecular approaches, different
E-mail address: ricardomachado@famerp.br (R.L.D. Machado). authors evaluated the global occurrence of these variants.
0732–8893/$ – see front matter © 2007 Published by Elsevier Inc.
doi:10.1016/j.diagmicrobio.2007.06.019

2. 416 R.T. Alves et al. / Diagnostic Microbiology and Infectious Disease 59 (2007) 415–419
The proportion of positive sera, specific for the VK210 and fresh blood samples collected in different endemic areas of
VK247 variants, ranged from 28% to 66% in Thailand the Brazilian Amazon region, all with positive results for
(Wirtz et al., 1990). Nevertheless, VK247 genotype was P. vivax thick blood films (TBFs). TBFs were examined by
identified in 58% of all patients infected with both genotypes independent experienced microscopists who were unaware of
(Kain et al., 1992, 1993). In Brazil, all variants were each result as recommended by the World Health Organiza-
genotyped, but only VK210 was found as a single agent of tion. Furthermore, molecular confirmation of P. vivax was
infection, whereas the other 2 occurred as mixed infections performed for all samples according to the method described
(Machado and Póvoa, 2000; Silva et al., 2006). On the other by Kimura et al. (1997). The protocol for this study was
hand, serological approaches had shown higher levels of reviewed and approved by the Ethics Research Board of the
positivity for antibodies against the 3 variants in Brazilian Medicine School in São José do Rio Preto, Brazil.
endemic and nonendemic areas (Arruda et al., 1996;
Oliveira-Ferreira et al., 2004). VK247 variant was mainly 2.2. Target DNA sequences and design of synthetic
found as a single infection in West Africa and the Indian oligonucleotides
subcontinent. In addition, the majority of the studied DNA was extracted from blood samples by the phenol–
individuals had mixed infections with both variants, the chloroform method (Pena et al., 1991). To amplify the CSP
predominant and VK210 (Kain et al., 1991; Gonzales et al., gene, 2 sets of forward and reverse primers were designed
2001). In Southern Mexico, it was observed that all patients based on the conserved central portion of the CSP gene.
were infected with VK210 and most of them also had VK247 The CSP sequences are available in the GenBank database
(Rodriguez et al., 2000). All variants were detected in field (VK210, accession number 11926; VK247, accession
isolates from malarious regions of Papua New Guinea, number M69061; P. vivax-like, accession number
Indonesia, and Madagascar, although no pure P. vivax-like L13724). The sequences were amplified using the follow-
isolate was verified (Qari et al., 1991, 1993). ing set of primers: PR1 (5′-ACT TTT ATT CGA CTT TGT
Kain et al. (1992) developed a genotype-specific poly- TGG TC-3′) and PR2 (5′-ATG GAC TCC ATG CAG TGT
merase chain reaction (PCR) technique by 32P-end–labeled AAC C-3′). The optimal specificity was achieved using the
oligoprobes to detect the VK210 and VK247 variants, BLAST program (http://www.ncbi.nlm.nih.gov/BLAST/).
whereas Kho et al. in 1999 investigated the polymorphisms A conformational analysis was made to investigate the
of the CSP gene in isolates from Korea by the PCR/ possibility of secondary structure formations (primer
restriction fragment length polymorphism (RFLP) technique. dimer). All oligonucleotide primers were synthesized by
The first methodology developed to identify P. vivax the Integrated DNA Technologies (Coralville, IA).
genotypes was PCR/hybridization, which also uses radi-
olabeled oligoprobes, but the technique is expensive and 2.3. PCR standardization
time-consuming, and also requires an adequate laboratorial
structure for elimination of the oligoprobes proper disposal Different PCR conditions were tested, varying PCR mixer
(Qari et al., 1993). Six years ago, Machado and Póvoa (2000) concentrations, primer annealing, and number of cycles.
optimized the Glass Fiber Membrane (GFM)/PCR/enzyme- After optimization, DNA (1.5 μL) was amplified in a total
linked immunosorbent assay (ELISA) method; however, it reaction volume of 25 μL consisting of 1× PCR buffer
needs much time, as well as uses initiating biotinylated (10 mmol/L Tris–HCl, pH 8.3, 50 mmol/L KCl), 1.5 mmol/L
primers and digoxigenin-labeled probes, raising the cost of of MgCl2, 1.0 μmol/L of each primer, 200 μmol/L
the procedure. In 2006, a protocol of nested-PCR/RFLP was deoxyribonucleotide triphospate (dNTPs), 2.5 U ampli-Taq
standardized for the diagnosis of 2 of the 3 genotypes: DNA polymerase, 1% betaine, and water (25 μL). Twenty-
VK210 and VK247 (Zakeri et al., 2006). five cycles of amplification were performed in a thermo-
The analysis of RFLPs of PCR products is a fast and cycler (DNA MasterCycler, Eppendorf, Hamburg, Germany)
simple technique (Trost et al., 2004) normally used in after initial denaturation of DNA at 94 °C for 5 min. Each
molecular biology laboratories in malaria endemic countries, cycle consisted of a denaturation step at 93 °C for 60 s, an
requiring only basic equipment (Tahar et al., 1998). Here, we annealing step at 41 °C for 90 s, and an extension step at
report on the standardization of a new PCR/RFLP for the 72 °C for 2 min, with a final extension at 72 °C for 10 min
identification of the 3 described P. vivax CSP gene variants. after the last cycle. The PCR products were analyzed by
electrophoresis using 1.5% agarose gels and stained with
ethidium bromide.
2. Materials and methods
2.4. Restriction digests of PCR products
2.1. Samples
The selected enzymes were required to have at least 1
For PCR standardization, we used 3 different plasmids cleavage site in the amplification of each variant, resulting in
(BlueScript, Stratagene, La Jolla, CA), one for the DNA fragments that are easily visible in polyacrylamide gel.
characteristic CSP repetitive region of each variant Restriction digests were set up with 10 μL of PCR product
(VK210, VK247, and P. vivax-like), and 45 frozen plus 10 and 1 U of the respective enzyme (AluI and DpnI, Promega,

3. R.T. Alves et al. / Diagnostic Microbiology and Infectious Disease 59 (2007) 415–419 417
San Diego, CA), incubated for 1 h at 37 °C. Restriction
fragments were separated by electrophoresis in 12.5%
polyacrylamide gels. The gels were stained with ethidium
bromide and analyzed with a Gel Doc 2000 illuminator (Bio-
Rad, Hercules, CA).
2.5. PCR sensitivity threshold
Three P. vivax blood samples from patients with
parasitemia ranging from 300 to 12,500 parasites per
microliter were used. These samples were serially diluted
in blood from an uninfected donor to a final level of
parasitemia corresponding to 10−6 and further processed
for PCR amplification. After that, a parasitologic evalua-
tion was performed to compare the sensitivity among the Fig. 2. P. vivax CSP gene RFLP patterns after enzymatic digestion with AluI
(from lanes 2 to 4) and DpnI (from lanes 5 to 7). Lane 1, 50-bp DNA ladder
PCR products. (Invitrogen); lanes 2 and 5, VK210 plasmid; lanes 3 and 6, VK247 plasmid;
lanes 4 and 7, P. vivax-like plasmid; lane 8, 100-bp DNA ladder.
2.6. PCR specificity
As a negative control, blood samples obtained from 20
molecularly diagnosed Plasmodium falciparum-infected with human DNA alone or with samples containing only
patients and from 10 blood donors living in the same areas P. falciparum parasites. The sensitivity of PCR was determined
with negative molecular results for Plasmodium were used. by serial dilutions of P. vivax blood samples with known
parasitemia. PCR of the P. vivax CSP gene detected levels of
2.7. P. vivax CSP gene amplification control parasitemia corresponding to 0.0069 parasites per microliter.
A single amplification of a CSP gene fragment using a set 3.2. RFLP analysis
of previously described oligonucleotide primers (AL60 5′-
GTC GGA ATT CAT GAA GAA CTT CAT TCT C-3′and To distinguish among the 3 P. vivax genotypes, RFLP
AL61 5′-CAG CGG ATC CTT AAT TGA ATA ATG CTA using AluI identified fragments of 10, 27, 38, 54, 106, and 135
GG-3′) was performed for all DNA samples (Machado and bp for VK210 and 10, 38, and 673 bp for VK247, whereas
Póvoa, 2000). P. vivax-like showed an unique fragment of 10 and 726 bp. In
respect to RFLP, using DpnI, we observed fragments with
sizes of 27, 42, 54, 81, 108, and 301 bp (VK247), whereas
3. Results P. vivax-like presented as 2 fragments (39 and 697 bp). The
3.1. Amplification of the P. vivax CSP gene fragment second enzyme has no restriction site for VK210 (Fig. 2).
Other fragments below 38 bp, not considered for variant
As shown in Fig. 1, DNA from all samples of P. vivax determination, were also formed after the RFLP procedure.
included in this study were amplified with the PR1 and PR2
primers. PCR products had lengths of 694 bp (VK210),
4. Discussion
721 bp (VK247), to 736 bp (P. vivax-like), as expected from
the BLAST program analysis. No amplifications were observed
Malaria is one of the most prevalent severe infectious
diseases in tropical and subtropical regions worldwide.
Because P. vivax malaria has been endemic in many
countries and its CSP genotypes are found worldwide, its
effective diagnosis is very important. Indeed, P. vivax
malaria variants may have different characteristics with
respect to the intensity of symptoms, the response to drugs,
and vector preference, which could cause drug resistance and
failure of control measures (Gopinath et al., 1994).
A new PCR/RFLP system was developed to identify
P. vivax genotypes. PCR primers were designed to amplify
the central immunodominant region of the CSP gene of this
protozoan. In our method, PCR primers were optimized to
Fig. 1. P. vivax CSP gene PCR products. Lane 1, 100-bp DNA ladder achieve easily distinguishable restriction fragments. The
(Invitrogen, Carlsbad, CA); lane 2, VK210 plasmid; lane 3, VK247 plasmid;
lane 4, P. vivax-like plasmid: lanes 5–6, P. vivax DNA from blood samples;
choice of restriction enzymes was also influenced by our
lanes 2–6, P. vivax CSP gene amplified with PR1 and PR2 primers; lanes objective of creating an efficient test with optimal resolution
7–8, P. vivax CSP gene amplified with AL60 and AL61. of restriction profiles. Based on the sequence analysis of

4. 418 R.T. Alves et al. / Diagnostic Microbiology and Infectious Disease 59 (2007) 415–419
P. vivax variants available in GenBank, the AluI and DpnI parasite in globally collected blood samples. J Infect Dis 170:
endonucleases were found to be the most suitable enzymes 1630–1633.
Herrera S, Corradim G, Arévalo-Herrera M (2007) An update on the search
for this purpose. AluI showed optimal discriminatory power for a Plasmodium vivax vaccine. Trends Parasitol :122–127.
to distinguish VK210 and P. vivax-like, but it was not adequate Imwong M, Pukrittayakamee S, Gruner AC, Renia L, Letourneur F,
to identify VK247 in mixed infections with the P. vivax-like Looareesuwan S, White NJ, Snounou G (2005) Practical PCR
variant. We solved this problem by adding a second genotyping protocols for Plasmodium vivax using Pvcs and Pvmsp1.
restriction step of the PCR product using DpnI, thereby Malar J 27:20.
Kain KC, Keystone J, Franke ED, Lanar DE (1991) Global distribution of a
unequivocally separating the VK247 and P. vivax-like variant of the circumsporozoite gene of Plasmodium vivax. J Infect Dis
variants, which allowed us the detection of mixed infections. 164:208–210.
The high cost and the need for adequate laboratory Kain KC, Brown AE, Webster HK, Wirtz RA, Keystone JS, Rodriguez MH,
conditions are the most frequently used arguments against Kinahan J, Rowland M, Lanar DE (1992) Circumsporozoite genotyping
using PCR in developing countries (Torres et al., 2006). of global isolates of Plasmodium vivax from dried blood specimens.
J Clin Microbiol 30:1863–1866.
However, PCR-based assays have advantages over micro- Kain KC, Brown AE, Lanar DE, Ballou WR, Webster HK (1993) Response
scopic tests because of their great capacity to distinguish P. of Plasmodium vivax variants to chloroquine as determined by
vivax genotypes, as all 3 variants are morphologically similar microscopy and quantitative polymerase chain reaction. Am J Trop
(Qari et al., 1993). Our method is sensitive and specific to Med Hyg 49:478–484.
Kho WG, Park YH, Chung JY, Kim JP, Hong ST, Lee WJ, Kim TS, Lee
detect P. vivax variants in both fresh and frozen samples. A
JS (1999) Two new genotypes Plasmodium vivax circumsporozoite
previous PCR/RFLP assay described by Zakeri et al. (2006) protein found in the Republic of Korea. Korean J Parasitol 37:
can only detect VK210 and VK247 and is inadequate for 265–270.
complete large-scale studies. In addition, our methodology Kimura M, Kaneko O, Liu Q, Zhou M, Kawamoto F, Wataya Y,
saves time and reduces the cost by more than one-half of the Otani S, Yamaguchi Y, Tanabe K (1997) Identification of the four
price of the GFM/PCR/ELISA method developed by species of human malaria parasites by nested PCR that targets
variant sequences in the small subunit rRNA gene. Parasitol Int
Machado and Póvoa (2000). Moreover, it does not use 46:91–95.
oligonucleotide primers labeled with radioisotopes as Machado RLD, Póvoa MM (2000) Distribution of Plasmodium vivax
described by Qari et al. (1993). variants (VK210, VK247 and P. vivax-like) in three endemic areas of
RFLP is more useful in distinguishing P. vivax genotypes Amazonian Brazil and their correlation with chloroquine-treatment.
than classifying them by using the CSP gene sequencing Trans R Soc Trop Med Hyg 94:377–381.
Machado RLD, Figueriredo-Filho AF, Calvosa VSP, Figueredo MC,
technique (Kho et al., 1999). Finally, the simplicity, Nascimento JM, Póvoa MM (2003) Correlation between Plasmodium
specificity, and sensitivity of the PCR/RFLP system vivax variants in Belém, Pará State, Brazil and symptoms and clearance
described here should be sufficient to determine the of parasitemia. Braz J Infect Dis 7:175–177.
prevalence and the distribution of infections of these Oliveira-Ferreira J, Pratt-Riccio LR, Santos M, Ribeiro F, Goldberg CT,
Banic AC (2004) HLA class II and antibody responses to
P. vivax genotypes in endemic and nonendemic malaria
circumsporozoite protein repeats of P. vivax (VK210, VK247 and P.
areas, enabling a better understanding their phylogeny. vivax-like) in individuals naturally exposed to malaria. Acta Trop 92:
63–69.
Acknowledgments Pena SDJ, Macedo AM, Gontijo NF (1991) DNA bioprints: simple non-
isotopic DNA fingerprints with biotinylated probes. Electrophoresis 12:
The authors thank Ana Carolina Silva, Gustavo Capatti, 14–52.
Qari SH, Goldman IF, Povoa MM, Oliveira S, Alpers MP, Lal AA (1991)
Valéria Fraga, and Luciana Moran for help in laboratory Wide distribution of the variant form of the human malaria parasite
work. Financial support was provided by FAPESP (Funda- Plasmodium vivax. J Biol Chem 266:16297–16300.
ção de Amparo à Pesquisa do Estado de São Paulo, São Qari SH, Shi YP, Goldman IF, Udhayakumar V, Alpers MP, Collins WE, Lal
Paulo, Brazil, process number 04/15486-7) and CNPq AA (1993) Identification of Plasmodium vivax-like human malaria
(Conselho Nacional de Desenvolvimento, Científico e parasite. Lancet 341:780–783.
Rieckmann KH, Davis DR, Hutton DC (1989) Plasmodium vivax resistance
Tecnológico, Brasília, DE, Brazil, process number 475524/ to chloroquine? Lancet 18:1183–1184.
2004-7). R.T.A. is a Masters student from the Genetic Rodriguez MH, Gonzalez-Ceron L, Hernandez JE, Nettel JA, Villarreal C,
Program Pos Graduation of the IBILCE/UNESP and has Kain KC, Wirtz RA (2000) Different prevalence of Plasmodium vivax
received research studentship from CNPq. phenotypes VK210 e VK247 associated with the distribution of Ano-
pheles albimanus and Anopheles pseudopunctipenis in Mexico. Am J
Trop Med Hyg 62:122–127.
References Rosenberg R, Wirtz RA, Lanar DE, Sattabongkot J, Hall T, Waters
AP, Prasittisuk C (1989) Circumsporozoite protein heterogeneity in
Arruda ME, Aragaki C, Gagliardi F, Halle RW (1996) A seroprevalence and the human malaria parasite Plasmodium vivax. Science 245:
descriptive epidemiological study of malaria among Indian tribes of the 973–976.
Amazon basin of Brazil. Ann Trop Med Parasitol 90:135–143. Silva AN, Santos CCB, Lacerda RN, Machado RLD, Wirtz R, Póvoa MM
Gonzales JM, Hurtado S, Arevalo-Herrera M, Herrera S (2001) Variants of (2006) Comparative Susceptibility of Anopheles aquasalis and An.
the Plasmodium vivax circumsporozoite protein (VK210 and VK247) in darlingi to Plasmodium vivax VK210 and VK247. Mem Inst Oswaldo
Colombian isolates. Mem Inst Oswaldo Cruz 96:709–712. Cruz 101:547–550.
Gopinath R, Wongsrichanalai C, Cordón-rosales C, Mirabelli L, Kyle D, Tahar R, De Pecoulas PE, Mazabraud A, Basco LK (1998)
Kain KC (1994) Failure to detect a Plasmodium vivax-like malaria Plasmodium vivax: rapid detection by polymerase chain reaction

5. R.T. Alves et al. / Diagnostic Microbiology and Infectious Disease 59 (2007) 415–419 419
and restriction fragment length polymorphism of the key Trost A, Graf B, Eucker J, Sezer O, Possinger K, Gobel UB, Adam T (2004)
mutation in dihydrofolate reductase-thymidylate synthase gene Identification of clinically relevant yeasts by PCR/RFLP. J Microbiol
associated with pyrimethamine resistance. Exp Parasitol 89: Methods 56:201–211.
343–346. Wirtz RA, Rosenberg R, Sattabongkot J, Webster HK (1990) Prevalence of
Torres KL, Figueiredo DV, Zalis MG, Daniel-Ribeiro CT, Alecrim W, antibody to heterologous circumsporozoite protein of Plasmodium vivax
de Ferreira-da-Cruz MF (2006) Standardization of a very specific in Thailand. Lancet 336:593–595.
and sensitive single PCR for detection of Plasmodium vivax in low Zakeri S, Abouie Mehrizi A, Djadid ND, Snounou G (2006) Circumspor-
parasitized individuals and its usefulness for screening blood ozoite protein gene diversity among temperate and tropical Plasmodium
donors. Parasitol Res 98:519–524. vivax isolates from Iran. Am J Trop Med Hyg 11:729–737.