A

lrmrfa1awpa©

K

1

m[brc

MN

tEEl

0d

Molecular & Biochemical Parasitology 153 (2007) 178–185

Extensive heterozygosity at four microsatellite loci flankingPlasmodium vivax dihydrofolate reductase gene�

Mohammad Tauqeer Alam, Richa Agarwal, Yagya D. Sharma ∗Department of Biotechnology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India

Received 31 December 2006; received in revised form 6 March 2007; accepted 6 March 2007Available online 12 March 2007

bstract

Only limited but contrasting reports are available on microsatellites based population structure of Plasmodium vivax. Further, there is completeack of information on microsatellites in the flanking regions of the P. vivax drug resistance genes to trace the origin and spread of the drugesistant vivax malaria. Therefore, we scanned ±300 kb flanking sequences of the P. vivax dihydrofolate reductase (pvdhfr) gene for di-nucleotideicrosatellite repeats with minimum of 8 unit array length. Only 13 such repeats were detected in this region as compared to 738 di-nucleotide

epeats present in ±300 kb flanking region of P. falciparum dhfr gene. We have analyzed here the nucleotide sequence of 110 Indian P. vivax isolatesor four of these di-nucleotide microsatellites (two in the nearest regions at −38.83 kb and +6.15 kb, and two in the farthest regions at −230.54 kbnd +283.28 kb). All the four microsatellites were found to be highly polymorphic in the population where number of alleles varied from 4 to0 with the median values of 9–11 at these loci. The expected heterozygosity (He) at these loci ranged from 0.50 to 0.82. We did not find anyssociation between pvdhfr mutations and the flanking microsatellite alleles. There was a regional variation in the microsatellites polymorphism

hich was not associated with the reported prevalent rates of drug resistance or malaria transmission. In conclusion, the level of microsatelliteolymorphism in P. vivax is as high as in P. falciparum. These results will be valuable in understanding the evolutionary history of the pvdhfrlleles as well as for designing the malaria control strategies.

2007 Elsevier B.V. All rights reserved.

Linka

bpwno

eywords: P. vivax malaria; pvdhfr; Pyrimethamine resistance; Microsatellite;

. Introduction

Plasmodium vivax is a widely distributed species of humanalaria parasites, responsible for 70–80 million cases annually

1]. Although vivax malaria is less lethal than the disease caused

y P. falciparum, several cases of severe vivax malaria are beingeported recently [2]. The numbers of drug resistant P. vivaxases are also increasing which may further increase the global

Abbreviations: pvdhfr, Plasmodium vivax dihydrofolate reductase; MP,adhya Pradesh; A&N, Andaman and Nicobar; UP, Uttar Pradesh; TN, Tamiladu

� Note: Nucleotide sequence data reported in this paper are available inhe GenBank, EMBL and DDJB databases under the accession numbersF442131–EF442139 for +6.15 kb; EF442148–EF442163 for +283.28 kb;F442140–EF442147 for −38.83 kb and EF442164–EF442171 for 230.54 kb

oci.∗ Corresponding author. Tel.: +91 11 26588145; fax: +91 11 26589286.

E-mail address: ydsharma aiims@yahoo.com (Y.D. Sharma).

tca

rdfawrowlh

166-6851/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.molbiopara.2007.03.003

ge disequilibrium; Selective sweep

urden of human malaria [3]. Genetic characterization of thearasite population is required for epidemiological studies asell as for designing the malaria control strategies. Although aumber of polymorphic genetic markers in P. vivax are previ-usly being described [4–7], limited information is available onhe microsatellites of the parasites for this purpose [8]. This is inontrast to P. falciparum where numerous microsatellite markersre available for parasite population genetic studies [9,10].

Microsatellites are highly polymorphic simple sequenceepeats (SSR) of 1–5 base pairs. They are also called Short Tan-em Repeats (STRs). The length of the repeat arrays may varyrom 10 to 100 base pairs. Slippage errors during DNA replicatonre frequent at the microsatellites and mutation rates increasesith the increased number of repeat array length [8]. The AT-

ich (∼80% AT) genome of P. falciparum contains abundance

f (AT)n or (TA)n repeats than those of the P. vivax (∼55% AT),hose genome is more GC-rich [11,12]. Recently, microsatellite

oci close to the drug resistance marker genes of P. falciparumave been characterized [13–18]. However, only three studies

emical Parasitology 153 (2007) 178–185 179

worr

d[moHaauPplvogimeP

2

2

pjaatGfa(tm

2t

upDam1utwuoT

Table 1Different type of repeats in 5′ and 3′ flanking regions of the P. vivax dhfr gene

S. no. Positions (bp)from pvdhfr

Repeat unit sequence Number ofrepeats

1 −279,321 TCCGCCTGCCCTC 10.382 −237,106 AAAAAGAAG 8.223 −234,874 AAATGGGCAGGTAATA 8.254 −230,544 TA 9.55 −207,499 TTTTTTGCA 10.336 −188,602 TAGCGGTGGA 14.27 −128,819 AT 8.58 −93,931 TA 109 −85,958 TA 9.5

10 −48,013 GCCCAAATGGAGGT 811 −38,836 AT 912 −30,054 GAG 8.6613 −2,640 AAATT 8.814 +6,151 AT 915 +30,812 ACACCACCGCC 11.9016 +67,084 CTC 1117 +72,451 TA 818 +95,353 CT 8.519 +118,994 CTGCTTCGTTG 12.9020 +126,924 TG 821 +145,489 TCT 10.6622 +174,247 AT 823 +175,303 TA 824 +185,042 GCA 825 +205,349 TGTA 15.7526 +237,532 AT 10.527 +283,284 AT 8.5

PrD

dr1caTeaAuM1awp

2

a(Bioneer Corporation, Korea) following manufacturer’s instruc-

M.T. Alam et al. / Molecular & Bioch

ith contrasting data have reported the microsatellite sequencesf P. vivax [12,19,20] but none of them being around drugesistance genes, although drug resistant vivax malaria is beingeported [3,21].

Pyrimethamine resistance associated mutations in P. vivaxhfr have been reported from various countries including India21–26]. Although antifolates are used to treat P. falciparumalaria, the P. vivax parasite gets exposed to the drug because

f mixed species infection leading to pvdhfr mutations [22].owever, very little information is available about the origin

nd spread of this drug resistance in P. vivax. Allelic diversityt flanking microsatellite loci of the dhfr gene will be helpful innderstanding the origin and spread of antifolate resistance in. vivax populations as well as the evolutionary history of thearasite species. Therefore, we planned to study the microsatel-ites around dhfr gene of P. vivax among Indian isolates fromarious geographical locations. We report here the presencef fewer microsatellites in the close proximity of the pvdhfrene which did not show any association with the mutationsn this gene. The heterozygosity (He) values at these P. vivax

icrosatellites were similar to any other microsatellite presentlsewhere in the genome as well as to the microsatellites of. falciparum.

. Materials and methods

.1. Parasite

A total of 110 field isolates of P. vivax were colleted fromatients attending malaria clinics at Kamrup (Assam), Pan-im (Goa), Panna (Madhya Pradesh), Car Nicobar (Andamannd Nicobar Islands), Ghaziabad and Aligarh (Uttar Pradesh)nd Chennai (Tamil Nadu). The samples were screened forhe presence of malaria parasite by light microscopy afteriemsa staining. About 200 �l heparinized blood was collected

rom the P. vivax positive patients. Patients were treated withntimalarial drugs as per the national drug policy of Indiahttp://www.namp.gov.in). Informed consent was obtained fromhe patients prior to blood collection. Blood collections were

ade as per the institutional ethical guidelines.

.2. Extraction of parasite DNA and PCR amplification ofhe microsatellites

Parasite DNA was extracted from 200 �l of patient bloodsing AccuPrep® Genomic DNA extraction kit (Bioneer Cor-oration, Korea) following manufacturer’s instructions. TheNA was eluted in 100 �l of elution (TE) buffer (pH 8.0)

nd a fraction was used for PCR amplification of the fouricrosatellites. The flanking nucleotide sequences (contig

370936 bp) of the pvdhfr gene were obtained from thenfinished P. vivax genome database (http://www.tigr.org). Aotal of 300 kb upstream and 300 kb downstream sequences

ere scanned for the presence of any microsatellite locising Perfect Tandem Repeat Finder program availablenline (http://sgdp.iop.kcl.ac.uk/nikammar/repeatfinder.html).he limit was set at minimum repeat unit length two (minimum

tpts

erfect Tandem Repeat Finder program (http://sgdp.iop.kcl.ac.uk/nikammar/epeatfinder.html) was used to scan the ±300 kb sequences flanking pvdhfr.i-nucleotide microsatellites used in this study are boldfaced.

i-nucleotide), minimum unit length of the array 8 and maximumepeat unit length 100 bp. Here, we have analyzed only 4 out of1 di-nucleotide (AT or TA) repeats, two from upstream (onelosest and one farthest) and two from downstream (one closestnd one farthest) flanking regions of the pvdhfr gene (Table 1).he details of the primer sequences and the PCR cycling param-ters for the amplification of all four microsatellite loci are givens supplementary information in Tables S1 and S2, respectively.ll PCR amplifications were carried out in a volume of 25 �lsing 0.2 units of Taq DNA polymerase (MBI Fermentas-Inc.,D, USA), 0.2 mM of each dNTP, 0.3 �M of each primer and

.5 mM MgCl2. PCR products were electrophoresed on a 1.2%garose gels. Same sets of primers did not yield any producthen genomic DNA from human and the cultured P. falciparumarasite was used for the PCR amplification.

.3. Nucleotide sequencing

Individual band of each PCR product was excised from thegarose gel and purified using AccuPrep® gel purification kit

ions. Sequencing of the PCR product was done using samerimers those used for their nested-PCR. The cycling parame-ers for sequencing PCR and other downstream protocols wereame as described earlier [18].

1 emic

2

Swtsvge(si

3

3

ugbl13drmb

idfrtlnwndrWvmat

liabalMa

Fi

80 M.T. Alam et al. / Molecular & Bioch

.4. Microsatellite sequence analysis

The electropherograms were analyzed using BioEditequence Alignment Editor [27]. The nucleotide sequencesere aligned using GeneDoc Multiple Sequence Alignment Edi-

or and Shading Utility (Version 2.6.002). All microsatelliteequences were compared with the sequences of the Sal-ador 1 strain of P. vivax (http://www.plasmodb.org). Theenetic variation in terms of expected heterozygosity (He) forach microsatellite locus was calculated using the formulaHe)=[n/(n − 1)][1 − ∑

pi2], where n is the number of P. vivax

amples genotyped for that locus and pi is the frequency of theth allele.

. Results

.1. Analysis of microsatellites flanking pvdhfr gene

We started our study by scanning the complete 300 kbpstream and 300 kb downstream sequences of the pvdhfrene with the default search parameters (i.e, minimum num-er of repeats considered was three; minimum repeat unitength two and maximum repeat unit length was kept at00 bp). A total of 1378 perfect repeats were observed in the00 kb upstream region, of which 1051 and 203 were perfect

i-nucleotide and tri-nucleotide repeats respectively, and theemaining 124 were larger repeats (data not shown). When weade our search more stringent by limiting the minimum num-

ers of repeats to 8, only 13 perfect repeats were observed

w+fo

ig. 1. Distribution of microsatellite alleles at −230.54 kb (A), −38.83 kb (B), +6.1solates. Number of di-nucleotide repeats in each allele in shown on X-axis.

al Parasitology 153 (2007) 178–185

n the 300 kb upstream region, out of which, only 5 werei-nucleotide repeats (Table 1). Similarly, a total of 1403 per-ect repeats were initially observed in the 300 kb downstreamegion, of which 991 and 301 were perfect di-nucleotide andri-nucleotide repeats respectively and the remaining 111 werearger repeats (data not shown). When the limit for minimumumbers of repeat was increased to 8, only 14 perfect repeatsere observed in this region. Out of these, only 8 were di-ucleotide microsatellite repeats (Table 1). Thus a total of 13i-nucleotide repeats (AT, TA, CT and TG) with more than 8epeat arrays were observed in this ±300 kb region (Table 1).

e selected four di-nucleotide microsatellites to study theirariation among Indian P. vivax population. Two of theseicrosatellites were in the nearest (−38.83 kb and +6.15 kb)

nd two in the farthest (−230.54 kb and +283.28 kb) regions ofhe pvdhfr gene.

A total of 110 P. vivax isolates were subjected to microsatel-ite analysis. We were able to analyze 101, 95, 93 and 92 P. vivaxsolates for microsatellites at −230.54 kb, −38.83 kb, +6.15 kbnd +283.28 kb, respectively. Equal number of isolates could note amplified for each of the locus because PCR did not work forll the four loci for some of the isolates. All the four microsatel-ite loci were found to be extremely polymorphic (Fig. 1).

aximum numbers of ten alleles were observed at −230.54 kbnd minimum number of four alleles at +6.15 kb loci. There

ere eight and seven microsatellite alleles at −38.83 kb and283.28 kb loci, respectively. Distribution of each allele variedrom each other as some of them were more predominant whilethers were rare.

5 kb (C) and +283.28 kb (D) loci flanking pvdhfr gene. n; number of P. vivax

emic

3

p[f

asnslfaAAi((olaG(

nag(wpMafms

oavc((Of

wl(lwlGrafA

3p

m5iw(ambotDpdttlwhlDp(cma

3l

csT0+Hat

4

t[osla,

M.T. Alam et al. / Molecular & Bioch

.2. Regional distribution of microsatellite alleles

The above 110 P. vivax isolates were collected from differentarts of the country having different level of pvdhfr mutations22]. Therefore, we have analyzed here these isolates region wiseor flanking microsatellite polymorphisms.

The allelic diversity at −230.54 kb microsatellite locus wasnalyzed among the 101 P. vivax isolates collected from allix study sites. Although we could not amplify the identicalumber of isolates from each region, there have been exten-ive allelic variation among the isolates for this microsatelliteocus (Table S3). The number of TA repeats at this locus rangedrom 5 to 21 with 10 alleles and a median value of 9. Thellele with (TA)5 repeats was predominant in all regions exceptndaman and Nicobar (A&N) Islands and Tamil Nadu (TN).lleles with (TA)5, (TA)9 and (TA)10 repeats were present

n all the six regions whereas alleles with (TA)11, (TA)16,TA)19 and (TA)21 repeats were region specific. Alleles withTA)16 and (TA)19 repeats were rare as they were presentnly in one isolate each. The maximum numbers of eight alle-es were found in A&N (total eight alleles) followed by sixlleles in Uttar Pradesh (UP). Other regions had four (Assam,oa, TN) and five (Madhya Pradesh (MP)) alleles at this locus

Table S3).The −38.83 kb locus is characterized by (AT)n repeats. The

umber of AT repeats at this locus, ranged from 7 to 15 with 8lleles and a median value of 11(Table S4). Out of 95 samplesenotyped for this locus, majority of them exhibited (AT)1123.1%) and (AT)7 (22.1%) repeats. None of the eight allelesas present in all the regions. Except (AT)15 allele, which wasresent only in Goa others were present in more than one region.aximum number of alleles were found in MP (seven alleles)

nd Goa (six alleles) whereas Assam, A&N and UP showedour alleles each. Only three alleles were observed in TN whichay also be due to the fact that lesser numbers of isolates were

equenced from this area.As compared to other three loci, less allelic variations were

bserved at +6.15 kb locus (Table S5). The number of AT repeatst this locus ranged from 7 to 11 with only 4 alleles with a medianalue of 9. The isolates from all the six regions were found toontain the (AT)9 repeat allele. This allele was predominant66.60%, n = 93) among the isolates. On the other hand, alleleAT)7 was rare and found only in Goa among two (10%) isolates.ther two alleles (AT)10 and (AT)11 were present in five and

our regions, respectively.Seven alleles with variable AT repeats ranging from 6 to 12

ith a median value of 9 were present among 92 P. vivax iso-ates at +283.28 kb locus (Table S6). Three alleles with (AT)932.6% isolates), (AT)8 (30.4% isolates) and (AT)6 (26.0% iso-ates) were predominant and present in all the five regions (weere not able to analyze the P. vivax samples of TN for this

ocus due to the limited amount of DNA). P. vivax isolates fromoa and MP showed maximum number of seven and five alleles

espectively for this locus whereas isolates from Assam, A&Nnd UP showed four alleles each. The alleles with (AT)11 wereound only in Goa and MP while (AT)12 only in Goa and A&N.llele (AT)7 was rare and found only in Goa.

snvfi

al Parasitology 153 (2007) 178–185 181

.3. Association of four-loci microsatellite haplotypes withvdhfr alleles

We have successfully amplified and sequenced all fouricrosatellite loci among 67 P. vivax isolates. There were

3 different combined four-loci haplotypes among thesesolates (Table 2). Out of 67 isolates, 43 were havingild-type (F57S58T61S93S117) DHFR codons, two had single

F57S58T61S93N117), 14 exhibited double (F57R58T61S93N117),nd remaining 8 had quadruple (L57R58M61S93T117) DHFRutations [22]. It may be stated here that the parasite har-

ouring these different DHFR mutations show different levelf pyrimethamine resistance [21]. In fact patients infected withhe P. vivax parasites bearing quadruple (L57R58M61S93T117)HFR mutations show very poor therapeutic response toyrimethamine while those with single (F57S58T61S93N117) orouble (F57R58T61S93N117) DHFR mutations are cleared byhe drug but at a relatively higher concentrations than that ofhe wild-type allele (F57S58T61S93S117) [21]. Among 43 iso-ates with wild-type DHFR, we found 32 four-loci haplotypeshereas each of the two isolates with single DHFR mutationad a different four-loci haplotype (Table 2). Similarly, 13 four-oci haplotypes were present among 14 isolates with doubleHFR mutations while each of the 8 isolates with quadru-le DHFR mutations contained different four-loci haplotypeTable 2). Therefore, we have not observed any kind of asso-iation between a particular microsatellite allele and the DHFRutations. In addition, correlations were also not seen between

ny two-loci microsatellite haplotypes and the DHFR mutations.

.4. Expected heterozygosity in P. vivax population at fouroci

The level of genetic variation at each of these four loci wasalculated among the P. vivax population. Each of the four locihowed extensive genetic variation in the parasite population.he expected heterozygosity (He) values varied from 0.50 to.82, with maximum variation at −38.83 kb and minimum at6.15 kb locus (Table 3). There were regional differences in thee values at each of the four loci. However, the He values for

ll the four loci combined together were almost similar amonghe regions.

. Discussion

Microsatellites are found in every eukaryote studied so far andhey are considered important markers for population genetics28]. The high mutation frequency rates in the microsatellitesccur due to the strand slippage during DNA replication anduch mispairing errors are more frequent at di-nucleotides withonger repeats [8,29]. The di-nucleotide microsatellite markersre used more frequently in the literature than the tri-, tetra-penta-, or other imperfect repeats [12,14–20]. Therefore, we

elected those microsatellite loci which had eight or more di-ucleotide repeats. Furthermore, microsatellites present in theicinity of those genes which are under selection tend to getxed because of genetic hitchhiking. They show linkage dise-

182 M.T. Alam et al. / Molecular & Biochemical Parasitology 153 (2007) 178–185

Table 2Distribution of four-loci microsatellite haplotypes among pvdhfr genotypes

DHFR genotype Types Four-loci microsatellite haplotypes Number ofisolates (n = 67)−230.54 kb (TA)n −38.83 kb (AT)n +6.15 kb (AT)n +283.28 kb (AT)n

F57S58T61S93S117 H1 5 7 9 6 4H2 5 7 9 8 1H3 5 7 9 9 1H4 5 7 9 11 1H5 5 9 9 6 1H6 5 9 9 9 1H7 5 9 9 10 1H8 5 9 11 6 1H9 5 9 11 7 1H10 5 11 10 8 1H11 5 12 9 9 3H12 5 12 10 6 1H13 5 12 10 11 1H14 5 12 11 8 1H15 5 13 10 8 1H16 5 13 10 9 1H17 9 7 9 8 1H18 9 10 10 9 1H19 9 11 9 9 3H20 9 12 9 11 1H21 10 7 9 9 4H22 10 9 9 6 2H23 10 9 9 8 1H24 10 9 9 10 1H25 10 9 10 8 1H26 10 11 9 9 1H27 10 12 9 8 1H28 10 14 10 9 1H29 12 9 7 6 1H30 13 12 9 8 1H31 19 11 9 10 1H32 20 11 10 9 1

F57S58T61S93N117 H33 5 11 9 8 1H34 9 11 9 8 1

F57R58T61S93N117 H35 5 7 7 8 1H1 5 7 9 6 1H2 5 7 9 8 1H36 5 7 10 8 1H37 5 7 10 12 1H38 5 9 11 9 1H39 5 10 9 6 1H40 5 11 9 6 2H41 9 11 9 6 1H42 9 12 9 8 1H43 10 12 9 6 1H44 12 13 10 8 1H45 20 9 10 6 1

L57R58M61S93T117 H46 5 11 9 6 1H47 5 11 9 12 1H48 9 7 10 9 1H49 10 12 10 8 1H50 13 13 9 6 1H51 13 13 11 9 1H52 16 11 9 6 1H53 21 13 9 8 1

n: number of isolates; mutated amino acids are bold faced.

M.T. Alam et al. / Molecular & Biochemical Parasitology 153 (2007) 178–185 183

Table 3Regional distribution of the expected heterozygosity (He) at each of the four loci

Area Expected heterozygosity (He) at each locus (kb) Average (He)

−230.54 kb locus −38.83 kb locus +6.15 kb locus +283.28 kb locus

Assam 0.75 (n = 8) 0.78 (n = 8) 0.00 (n = 5) 0.74 (n = 11) 0.57Goa 0.54 (n = 23) 0.83 (n = 21) 0.65 (n = 20) 0.54 (n = 16) 0.64MP 0.70 (n = 25) 0.85 (n = 24) 0.42 (n = 23) 0.71 (n = 23) 0.67AN 0.93 (n = 14) 0.67 (n = 15) 0.63 (n = 17) 0.66 (n = 22) 0.72UP 0.72 (n = 25) 0.72 (n = 21) 0.40 (n = 22) 0.74 (n = 20) 0.65TN 0.87 (n = 6) 0.61 (n = 6) 0.33 (n = 6) ND 0.60

A 0

N

qidwgsAflvlthnumwcu

orsatos+t(b

vveTmi+gitif

atshanmtohlabrtpad

mafiPawinls[vr[waahtK

verage 0.72 ± 0.14 S.D. (n = 101) 0.82 ± 0.093 S.D. (n = 95)

D: not determined. n: number of isolates

uilibrium with the mutations in the affected gene. This resultsn a selective sweep of a particular genotype [9]. Hence, weecided to analyze those di-nucleotide microsatellites whichere present in the flanking region of the pvdhfr gene. Thisene has recently been shown to be under pyrimethamine pres-ure resulting in the mutations at its various codons [22,24].lthough 13 di-nucleotide repeats were detected in the ±300 kbanking regions of pvdhfr, none of them was in its very closeicinity (Table 1). In fact, the nearest di-nucleotide repeats wereocated at the distance of −38.83 kb and +6.15 kb with respecto the pvdhfr gene. In other words, +30 kb region around pvd-fr had only one di-nucleotide repeat (AT)9 at +6.1 kb whereasone in the −30 kb region (Table 1). This is contrary to the sit-ation in P. falciparum where large numbers of di-nucleotideicrosatellites are present around dhfr [9,15,17]. Indeed, whene scanned ±300 kb sequences around dhfr gene in the P. fal-

iparum genome, we observed 738 such microsatellites (386 inpstream and 352 in downstream region) (Figs. S1 and S2).

Microsatellite markers located within 20 kb flanking regionf the pvdhfr gene should be ideal to detect linkage disequilib-ium (LD), but in the present situation where such markers arecanty (Table 1), we decided to analyze the next nearest locust −38.83 kb in the upstream region besides +6.15 kb locus inhe downstream region. Since our aim was to analyze the extentf microsatellite polymorphism flanking the pvdhfr gene, weelected two nearest di-nucleotide repeats loci at −38.83 kb and6.15 kb as no other di-nucleotide repeat was present in between

hese two loci. In addition, we also selected two far placed loci−230.54 kb and +283.28 kb) in order to compare the variationetween closest and farthest markers flanking the pvdhfr gene.

Absence of the di-nucleotide microsatellites in the closeicinity of pvdhfr gene is probably an indication that there is aery little possibility of hitchhiking of the flanking neutral mark-rs and LD between the microsatellites and pvdhfr mutations.his indeed was evident from our results on the di-nucleotideicrosatellites where an enormous heterogeneity was observed

n the parasite population at all of the four selected loci, except6.15 kb locus, irrespective of their distance from the pvdhfrene (Table 3). The average He value also varied among regional

solates but did not show any correlation either with drug resis-ance or with the level of malaria transmission rates prevalentn these areas [30]. For example, the average He values at allour loci among UP and MP isolates were in the similar range

rplr

.50 ± 0.24 S.D. (n = 93) 0.73 ± 0.083 S.D. (n = 92) 0.69

s observed in Assam and A&N (Table 3), although, the latterwo regions show higher drug resistance and malaria transmis-ion rates than UP and MP [22,30–34]. Further, the amount ofeterozygosity around sensitive and resistant pvdhfr alleles atll four loci was almost same, thereby indicating that there iso association between any of the microsatellites and the pvdhfrutations (Table 2). The lack of di-nucleotide microsatellites in

he closest proximity of pvdhfr gene is mostly due to the naturef P. vivax genome. The observed linkage equilibrium couldave arisen due to several factors, including: (a) high microsatel-ite mutation rates, (b) microsatellite mutations were present forlong time (old event) allowing recombination mechanism to

reakdown the chromosomal segment, (c) high recombinationates, and (d) multiple founder mutations. This is contrary tohe situation in P. falciparum where markers around sensitivefdhfr allele show more variations while those around resistantlleles show less variations (fixed) and thus have strong linkageisequilibrium with the mutations in the gene [16,17,35].

Very little information is currently available about theicrosatellite based genetic structure of P. vivax populations

s compared to the P. falciparum [8–10]. Indeed this is for therst time when any microsatellite around drug resistance gene of. vivax is being investigated. Although our aim was to identifynd analyze microsatellites linked to the pvdhfr gene, the resultsere unexpected as no di-nucleotide microsatellite was detected

n the close vicinity and the studied microsatellite markers didot show any association with mutant pvdhfr alleles. Neverthe-ess, the extensive polymorphism at these microsatellite loci wasimilar to the microsatellites present elsewhere in the genome12,19,36]. There are contrasting reports on the heterozygosityalues for the P. vivax microsatellites due to usage of differentepeat array lengths [12,19,20,36]. For this reason, Leclerc et al.20] reported lower level of genetic variation in P. vivax as theyere using very short repeat array lengths whereas Imwong et

l. [12], reported higher He values since they used longer repeatrrays (median 5.5 and range 4–13). Similarly Gomez et al. [19]ave also observed a higher allelic diversity (7–24 AT repeats) ashey studied the loci with longer repeats. In a very recent study,arunaweera et al. [36] have used longer tri- and tetra-nucleotide

epeats and reported an extensive allelic diversity (6–13 alleleser locus) and expected heterozygosity (0.627–0.913) in Sri-ankan P. vivax parasite population. As mentioned above, theate of variation at microsatellite loci depends on the repeat

1 emic

awmtl(lpTiao[nts

tlaTauwtavs

A

KhsRsRGB(

A

i

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

84 M.T. Alam et al. / Molecular & Bioch

rrays length [12,37,38], the results of the present study alongith previous reports [12,19,20] thus strongly suggest that theicrosatellites with longer repeats should essentially be used

o estimate the level of genetic diversity in the P. vivax popu-ation. The expected heterozygosity reported here in our study0.50–0.82) is closer to these reports because we also selectedonger repeat (more than 8) arrays. Our results thus provide sup-ort to the fact that genetic diversity in P. vivax is not meager.he SNP studies have also shown high level of genetic diversity

n P. vivax [6]. Furthermore, the degree of microsatellite vari-tions we found in our study is also comparable with the levelf variations observed for the microsatellites of P. falciparum17,18,39–42]. Although P. vivax contains lesser number of di-ucleotide repeats than P. falciparum (Table 1, Fig. S1 and S2),he microsatellite repeats of equal length from both the specieshow comparable heterozygosity (Table 3).

We have identified and analyzed microsatellite loci close tohe pvdhfr gene. All four loci analyzed here showed completeinkage equilibrium with the pvdhfr mutations and behaved likeny unlinked microsatellites present elsewhere in the genome.he degree of genetic variation at these microsatellite loci wass high as in P. falciparum. Our findings will be valuable innderstanding the evolutionary history of the pvdhfr alleles asell as for designing the malaria control strategies. Nevertheless,

he other microsatellite repeats around the pvdhfr gene shouldlso be investigated to provide valuable information on geneticariation as each microsatellite has its own characteristics andhows different rates of mutations.

cknowledgements

We wish to thank Drs. A.P. Dash, Wajihullah, Ashwaniumar, Neeru Singh, Vas Dev and Manoj Das for their kindelp. We also thank Ms. Sumiti Vinayak for fruitful discus-ions. MTA acknowledge Council of Scientific and Industrialesearch (CSIR), for Senior Research Fellowship. Financial

upport for the work granted by Indian Council of Medicalesearch (ICMR) and Department of Biotechnology (DBT),overnment of India, is acknowledged. We are also grateful toioinformatics facilities of Biotechnology Information System

BTIS).

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at doi:10.1016/j.molbiopara.2007.03.003.

eferences

[1] Mendis K, Sina BJ, Marchesini P, Carter R. The neglected burden ofPlasmodium vivax malaria. Am J Trop Med Hyg 2001;64S:97–106.

[2] Kochar DK, Saxena V, Singh N, Kochar SK, Kumar SV, Das A. Plasmodiumvivax malaria. Emerg Infect Dis 2005;11:132–4.

[3] Baird JK. Chloroquine resistance in Plasmodium vivax. Antimicrob AgentsChemother 2004;48:4075–83.

[4] Galinski MR, Barnwell JW. Plasmodium vivax: merozoites, invasion ofreticulocytes and considerations for malaria vaccine development. ParasitolToday 1996;12:20–9.

[

al Parasitology 153 (2007) 178–185

[5] Cui L, Escalante AA, Imwong M, Snounou G. The genetic diversity ofPlasmodium vivax populations. Trends Parasitol 2003;19:220–6.

[6] Feng X, Carlton JM, Joy DA, et al. Single-nucleotide polymorphismsand genome diversity in Plasmodium vivax. Proc Natl Acad Sci USA2003;100:8502–7.

[7] Lim CS, Tazi L, Ayala FJ. Plasmodium vivax: recent world expansionand genetic identity to Plasmodium simium. Proc Natl Acad Sci USA2005;102:15523–8.

[8] Russell B, Suwanarusk R, Lek-Uthai U. Plasmodium vivax genetic diver-sity: microsatellite length matters. Trends Parasitol 2006;22:399–401.

[9] Anderson TJC. Mapping drug resistance genes in Plasmodium falci-parum by genome-wide association. Curr Drug Targets Infect Disord2004;4:65–78.

10] Su XZ, Wellems TE. Toward a high-resolution Plasmodium falciparumlinkage map: polymorphic markers from hundreds of simple sequencerepeats. Genomics 1996;33:430–44.

11] Carlton J. The Plasmodium vivax genome sequencing project. Trends Par-asitol 2003;19:227–31.

12] Imwong M, Sudimack D, Pukrittayakamee S, et al. Microsatellite variation,repeat array length, and population history of Plasmodium vivax. Mol BiolEvol 2006;23:1016–8.

13] Anderson TJC, Roper C. The origins and spread of antimalarial drug resis-tance: lessons for policy makers. Acta Trop 2005;94:269–80.

14] Wootton JC, Feng X, Ferdig MT, et al. Genetic diversity and chloroquineselective sweeps in Plasmodium falciparum. Nature 2002;418:320–3.

15] McCollum AM, Poe AC, Hamel M, et al. Antifolate resistance in Plasmod-ium falciparum: multiple origins and identification of novel dhfr alleles. JInfect Dis 2006;194:189–97.

16] Nair S, Williams JT, Brockman A, et al. A selective sweep driven bypyrimethamine treatment in southeast Asian malaria parasites. Mol BiolEvol 2003;20:1526–36.

17] Nash D, Nair S, Mayxay M, et al. Selection strength and hitchhiking aroundtwo anti-malarial resistance genes. Proc Biol Sci 2005;272:1153–61.

18] Vinayak S, Mittra P, Sharma YD. Wide variation in microsatellitesequences within each Pfcrt mutant haplotype. Mol Biochem Parasitol2006;147:101–8.

19] Gomez JC, McNamara DT, Bockarie MJ, Baird JK, Carlton JM, Zimmer-man PA. Identification of a polymorphic Plasmodium vivax microsatellitemarker. Am J Trop Med Hyg 2003;69:377–9.

20] Leclerc MC, Durand P, Gauthier C, et al. Meager genetic variability ofthe human malaria agent Plasmodium vivax. Proc Natl Acad Sci USA2004;101:14455–60.

21] Hastings MD, Porter KM, Maguire JD, et al. Dihydrofolate reductase muta-tions in Plasmodium vivax from Indonesia and therapeutic response tosulfadoxine plus pyrimethamine. J Infect Dis 2004;189:744–50.

22] Alam MT, Bora H, Bharti PK, et al. Similar trends of pyrimethamineresistance associated mutations in Plasmodium vivax and P. falciparum.Antimicrob Agents Chemother 2007;51:857–63.

23] Brega S, De Monbrison F, Severini C, et al. Real-time PCR for dihydrofo-late reductase gene single-nucleotide polymorphisms in Plasmodium vivaxisolates. Antimicrob Agents Chemother 2004;48:2581–7.

24] Eldin de Pecoulas P, Tahar R, Ouatas T, Mazabraud A, Basco LK. Sequencevariations in the Plasmodium vivax dihydrofolate reductase-thymidylatesynthase gene and their relationship with pyrimethamine resistance. MolBiochem Parasitol 1998;92:265–73.

25] Imwong M, Pukrittayakamee S, Renia L, et al. Novel point mutationsin the dihydrofolate reductase gene of Plasmodium vivax: evidence forsequential selection by drug pressure. Antimicrob Agents Chemother2003;47:1514–21.

26] Kaur S, Prajapati SK, Kalyanaraman K, Mohmmed A, Joshi H, ChauhanVS. Plasmodium vivax dihydrofolate reductase point mutations from theIndian subcontinent. Acta Trop 2006;97:174–80.

27] Hall TA. BioEdit: a user-friendly biological sequence alignment editor

and analysis program for Windows 95/98/NT. Nucl Acids Symp Ser1999;41:95–8.

28] Li YC, Korol AB, Fahima T, Beiles A, Nevo E. Microsatellites: genomicdistribution, putative functions and mutational mechanisms: a review. MolEcol 2002;11:2453–65.

emic

[

[

[

[

[

[

[

[

[

[

[

[

M.T. Alam et al. / Molecular & Bioch

29] Kashi Y, King DG. Simple sequence repeats as advantageous mutators inevolution. Trends Genet 2006;22:253–9.

30] Sharma VP. Current scenario of malaria in India. Parasitologia 1999;41:349–53.

31] Ahmed A, Das MK, Dev V, Saifi MA, Wajihullah, Sharma YD. Quadruplemutations in dihydrofolate reductase of Plasmodium falciparum iso-lates from Car Nicobar Island, India. Antimicrob Agents Chemother2006;50:1546–9.

32] Ahmed A, Lumb V, Das MK, Dev V, Wajihullah, Sharma YD. Prevalenceof mutations associated with higher levels of sulfadoxine-pyrimethamineresistance in Plasmodium falciparum isolates from Car Nicobar and Assam,India. Antimicrob Agents Chemother 2006;50:3934–8.

33] Misra SP. In-vivo resistance to chloroquine and sulfa-pyrimethamine com-bination in Plasmodium falciparum in India. Proc Natl Acad Sci India1996;66:123–8.

34] Mittra P, Vinayak S, Chandawat H, et al. Progressive increase in point muta-

tions associated with chloroquine resistance in Plasmodium falciparumisolates from India. J Infect Dis 2006;193:1304–12.

35] Roper C, Pearce R, Bredenkamp B, et al. Antifolate antimalarialresistance in Southeast Africa: a population-based analysis. Lancet2003;361:1174–81.

[

[

al Parasitology 153 (2007) 178–185 185

36] Karunaweera ND, Ferreira MU, Hartl DL, Wirth DF. Fourteen polymorphicmicrosatellite DNA markers for the human malaria parasite Plasmodiumvivax. Mol Ecol Notes 2007;7:172–5.

37] Ellegren H. Microsatellites: simple sequences with complex evolution. NatRev Genet 2004;5:435–45.

38] Lai Y, Sun F. The relationship between microsatellite slippage muta-tion rate and the number of repeat units. Mol Biol Evol 2003;20:2123–31.

39] Conway DJ, Machado RLD, Singh B, et al. Extreme geographical fix-ation of variation in the Plasmodium falciparum gamete surface proteingene Pfs48/45 compared with microsatellite loci. Mol Biochem Parasitol2001;115:145–56.

40] Ferreira MU, Nair S, Hyunh TV, Kawamoto F, Anderson TJC. Microsatel-lite characterization of Plasmodium falciparum from cerebral anduncomplicated malaria patients in southern Vietnam. J Clin Microbiol2002;40:1854–7.

41] Roper C, Pearce R, Nair S, Sharp B, Nosten F, Anderson T. Intercon-tinental spread of pyrimethamine-resistant malaria. Science 2004;305:1124.

42] Ferdig MT, Su XZ. Microsatellite markers and genetic mapping in Plas-modium falciparum. Parasitol Today 2000;16:307–12.