Infection, Genetics and Evolution 17 (2013) 269–276

Contents lists available at SciVerse ScienceDirect

Infection, Genetics and Evolution

journal homepage: www.elsevier .com/locate /meegid

Multilocus genotyping of Giardia duodenalis in Malaysia

Choy Seow Huey a, Mohammed A.K. Mahdy a,b,⇑, Hesham M. Al-Mekhlafi a,b, Nabil A. Nasr a,Yvonne A.L. Lim a, Rohela Mahmud a, Johari Surin a

a Department of Parasitology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysiab Department of Parasitology, Faculty of Medicine, Sana’a University, Sana’a, Yemen

a r t i c l e i n f o

Article history:Received 18 December 2012Received in revised form 6 April 2013Accepted 9 April 2013Available online 25 April 2013

Keywords:Multilocus genotypingGiardiatpibggdhMalaysia

1567-1348/$ - see front matter � 2013 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.meegid.2013.04.013

⇑ Corresponding author at: Department of ParasiUniversity of Malaya, 50603 Kuala Lumpur, Malaysia+60 3 7949 4754.

E-mail address: alsharaby9@yahoo.com (M.A.K. M

a b s t r a c t

Giardia duodenalis is considered the most common intestinal parasite in humans worldwide. In Malaysia,many studies have been conducted on the epidemiology of giardiasis. However, there is a scarcity ofinformation on the genetic diversity and the dynamics of transmission of G. duodenalis. The present studywas conducted to identify G. duodenalis assemblages and sub-assemblages based on multilocus analysisof the glutamate dehydrogenase (gdh), beta-giardin (bg) and triose phosphate isomerase (tpi) genes. Fae-cal specimens were collected from 484 Orang Asli children with a mean age of 7 years and examinedusing light microscopy. Specimens positive for Giardia were subjected to PCR analysis of the three genesand subsequent sequencing in both directions. Sequences were edited and analysed by phylogeneticanalysis. G. duodenalis was detected in 17% (84 of 484) of the examined specimens. Among them, 71 weresuccessfully sequenced using at least one locus. Genotyping results showed that 30 (42%) of the isolatesbelonged to assemblage A, 32 (45%) belonged to assemblage B, while discordant genotype results wereobserved in 9 specimens. Mixed infections were detected in 43 specimens using a tpi-based assemblagespecific protocol. At the sub-assemblages level, isolates belonged to assemblage A were AII. High nucle-otide variation found in isolates of assemblage B made subtyping difficult to achieve. The finding ofassemblage B and the anthroponotic genotype AII implicates human-to-human transmission as the mostpossible mode of transmission among Malaysian aborigines. The high polymorphism found in isolates ofassemblage B warrants a more defining tool to discriminate assemblage B at the sub-assemblage level.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Giardia duodenalis (syn. G. intestinalis; Giardia lamblia) is a fla-gellate enteric parasite that infects the intestinal tract of humansand a wide variety of other mammals through ingestion of infec-tive cysts (van der Giessen et al., 2006). It is a major cause ofnon-viral/bacterial diarrhea affecting both developed and develop-ing countries and more frequently encountered in areas with sub-standard environment, poor hygienic practices and inadequatewater treatment system (Daly et al., 2010; Robertson et al.,2009). G. duodenalis infections are most often asymptomatic. Forsymptomatic cases, the degree of symptoms and severity variesamong different individuals (Buret, 2008).

Knowing the genetic diversity of Giardia is an essential compo-nent in enhancing our understanding on the taxonomy, epidemiol-ogy and population genetics of this parasite. Therefore, molecularcharacterization and in the recent years, multilocus genotyping

ll rights reserved.

tology, Faculty of Medicine,. Tel.: +60 3 7949 4746; fax:

ahdy).

(MLG) have become the prevailing tool used for genotyping andsubtyping of G. duodenalis. This has in turn facilitated in outbreaksurveillance, contamination source-tracking, and unveiling zoo-notic potential, dynamic transmission, and relationship of geno-types and hosts (Feng and Xiao, 2011). Molecular analysesrevealed that G. duodenalis isolated from humans and animals isa species complex with at least 8 distinct assemblages designatedas A to H, which demonstrate similarity in morphologic character-istics but are phenotypically and genotypically heterogeneous(Plutzer et al., 2010). Of these, only assemblages A and B can causeinfection in humans with the exception of a small fraction of caseswhere animal host specific assemblages C–F were reported in hu-man (Sprong et al., 2009). Besides, both assemblages A and B arealso capable of infecting animals. The subdivision into AI–AIII hasprovided greater insights on the dynamic of transmission of theseassemblages. Sub-assemblage AII has been regarded as anthropo-notic whereas AI and AIII are predominant in livestock and wildlife,respectively (Feng and Xiao, 2011; Nolan et al., 2010; Sprong et al.,2009; Thompson et al., 2000).

In Malaysia, giardiasis is an endemic disease and is associatedwith malnutrition among children in the rural areas resulting instunting, wasting and vitamin A deficiency (Al-Mekhlafi et al.,

270 C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276

2010, 2005). The prevalence of human Giardia infection varies be-tween 0.2% and 29.2% (Lim et al., 2008; Noor Azian et al., 2007).Most of the epidemiological studies detected Giardia on the basisof microscopic examination without employing molecular ap-proach. Data on genotypes of G. duodenalis up to the assemblage le-vel remains scarce. In a previous genotyping study using SSU rRNAlocus, one specimen was identified as assemblage A in 42 speci-mens and the rest were assemblage B (Mahdy et al., 2009). In astudy on immunocompromised patients, assemblage A was identi-fied in four of the microcopy-positive Giardia specimens using tpigene (Lim et al., 2011). Assemblage A was also isolated from envi-ronmental samples including recreational lake water and waterbodies in a zoo (Lim et al., 2009a,b). In addition, genotyping studywas conducted on animals and assemblages A and E were detectedamong goats (Lim et al., 2013). However, subtyping of assemblagesA and B in the previous studies was not conducted which limit ourunderstanding on the transmission dynamics and the source ofinfection of giardiasis in this country. Thus, the present studywas aimed to identify G. duodenalis assemblages and sub-assem-blages based on multiloci genes which included gdh, bg and tpigenes to attain better understanding of the genetic diversity andtransmission of giardiasis in Malaysia. The study also aimed atdetermining the occurrence of mixed infections using primers tar-geting tpi gene specific for assemblages A and B.

2. Materials and methods

2.1. Source of samples

A total of 484 children (51% males, 49% females) with the meanage of 7 years were involved in the present study. The sample col-lection was conducted from April to September 2011 in 13 OrangAsli villages located in Lipis district, Pahang state, PeninsularMalaysia. The study protocol was approved by The Medical EthicsCommittee of University of Malaya Medical Center, Kuala Lumpur,Malaysia under the MEC Ref. No.: 788.74. Informed consents wereobtained from the participants’ parents or guardians prior to thecollection. Single sample was collected from each participant.Upon receipt, the samples were transferred to the Department ofParasitology, University of Malaya, preserved in 2.5% potassiumdichromate and stored in cold room (4 �C) until further analysis.

2.2. Microscopy

The faecal samples were concentrated based on formal-ethertechnique. Briefly, a small amount of stool samples (pea size)was emulsified with 7 ml of 10% formalin and sieved through adouble-layered gauze and collected in a beaker. The suspensionwas transferred into 15 ml conical centrifuge tube and topped upwith three ml of diethyl ether. The centrifuge tube with the sus-pension was capped and mixed by shaking before centrifugingfor 5 min at 2500 rpm. The top three layers (ether, debris and for-malin water) were removed and the sediment was examined bylight microscope under x100 and x400 magnification for the pres-ence of Giardia cysts and other intestinal parasites after stainingwith iodine.

2.3. Molecular analysis

DNA was extracted directly from samples positive for Giardiacyst using PowerSoil�DNA Isolation Kit (MoBio Laboratories Inc.,Carlsbad, California) following the manufacturer’s instructions.Elution step was accomplished by adding reduced volume of solu-tion C6 (10 mM Tris) to obtain a final volume of 50 ll and the DNAwas stored in freezer under �20 �C.

A partial sequence of gdh gene (�432-bp) was amplified usingsemi-nested PCR as described by Read et al. (2004). For primaryPCR, the forward primer GDHeF (50-TCA ACG TYA AYC GYG GYTTCC GT-30) and the reverse primer GDHiR (50-GTT RTC CTT GCACAT CTC C-30) were used. In the secondary PCR, the forward primerGDHiF (50-CAG TAC AAC TCY GCT CTC GG-30) and the reverse primerGDHiR were used. Primary and secondary PCR reactions were per-formed in a 50 ll PCR master mix comprising 0.5 lM of each primer(Bioneer Q-Oligos, Korea), 2.5 U of HotStarTaq� Plus DNA Polymer-ase (Qiagen, Hilden, Germany),1� PCR buffer (Qiagen, Hilden, Ger-many), 200 lM of dNTP (Fermentas, Ontario, Canada), 1.5 mMMgCl2 (Qiagen, Hilden, Germany), 5% dimethyl sulfoxide (Sigma–Al-drich, USA) and 0.4 mg/ml BSA (New England Biolabs, Ipswich, USA).2 ll of DNA template were used in both amplifications that were runin the MyCycler thermal cycler (Bio-Rad, Hercules, USA) under thefollowing conditions: Initial activation at 95 �C for 5 min, 1 cycle at94 �C for 2 min, 56 �C for 1 min and 72 �C for 2 min, followed by 55amplification cycles at 94 �C for 30 s, 56 �C for 20 s, 72 �C for 45 s,and a final extension at 72 �C for 7 min. For secondary PCR, the num-ber of cycles was reduced to 33.

A partial sequence of tpi (�530-bp) was amplified using nested-PCR protocol according to Sulaiman et al. (2003). Primary PCR wasrun using forward primer AL3543 (50-AAA TIA TGC CTG CTC GTC G-30) and reverse primer AL3546 (50-CAA ACC TTI TCC GCA AAC C-30).For secondary PCR, forward primer AL3544 (50-CCC TTC ATC GGIGGT AAC TT-30) and reverse primer AL3545 (50-GTG GCC ACC ACICCC GTG CC-30) were used. Primary and secondary PCRs were per-formed in a 50 ll PCR mix comprising 0.2 lM of each primer (Bio-neer Q-Oligos, Korea), 1 U of HotStarTaq� Plus DNA Polymerase(Qiagen, Hilden, Germany), 1� PCR buffer (Qiagen, Hilden, Ger-many), 200 lM dNTP (Fermentas, Ontario, Canada), 1.5 mM MgCl2

(Qiagen, Hilden, Germany), and 0.2 mg/ml BSA (New England Bio-labs, Ipswich, USA). 2 ll of DNA template were used and the pre-pared master mix was incubated in the MyCycler thermal cycler(Bio-Rad, Hercules, USA) under the following conditions: Initialhot start at 95 �C for 5 min, 35 amplification cycles at 94 �C for45 s, 50 �C for 45 s (58 �C for secondary PCR), 72 �C for 60 s and afinal extension at 72 �C for 10 min.

In addition, the first PCR product of the reaction described bySulaiman et al. (2003) underwent further amplification using aset of separate A (Geurden et al., 2007) and B (Geurden et al.,2009) assemblage-specific primers. Presence of mixed infectionwas detected by visualizing the occurrence of bands in the agarosegel at 332 bp for assemblage A amplified using primers AssAF (50-CGC CGT ACA CCT GTC-30) and AssAR (50-AGC AAT GAC AAC CTCCTT CC-30) and at 400 bp for assemblage B amplified using primersAssBF (50-GTT GTT GTT GCT CCC TCC TTT -30) and AssBR (50-CCGGCT CAT AGG CAA TTA CA-30). The PCR reaction mix consisted of0.2 lM (0.4 lM for assemblage B) of each primer (Bioneer Q-Oli-gos, Korea), 1.25 U of HotStarTaq� Plus DNA Polymerase (Qiagen,Hilden, Germany), 1� PCR buffer (Qiagen, Hilden, Germany),200 lM dNTP (Fermentas, Ontario, Canada), 1.5 mM MgCl2 (Qia-gen, Hilden, Germany) and 0.1 mg/ml BSA (New England Biolabs,Ipswich, USA) to a final volume of 25 ll. 1 ll of DNA templatewas added for assemblage A and 2 ll was added for assemblageB for the PCR amplifications following the cycle parameter: Initialhot start at 95 �C for 5 min, initial denaturation at 94 �C for 10 min,and 35 amplification cycles at 94 �C for 45 s, 64 �C for 45 s (62 �Cfor secondary PCR) and 72 �C for 45 s.

A partial sequence of bg gene (�511-bp) was amplified usingPCR protocols described by Caccio et al. (2002) and Lalle et al.(2005). The primers for primary PCR were G7 (50-AAG CCC GACGAC CTC ACC CGC AGT GC-30) and G759 (50-GAG GCC GCC CTGGAT CTT CGA GAC GAC-30). For secondary PCR, BG511F (50-GAACGA ACG AGA TCG AGG TCC G-30) and BG511R (50-CTC GAC GAGCT TCG TGT T-30) were used. The PCR master mix consisted of

Table 1The distribution of assemblages A and B based on the different loci and mixedinfection.

Genes Assemblages n (%)

A B A + B Total

tpi 34 (58) 25 (42) – 59gdh 7 (18) 31 (82) – 38b-giardin 18 (64) 10 (36) – 28MLGsa 3 (27) 8 (73) – 11tpi-Mixedb 1 (2) 23 (34) 43 (64) 67

a Multilocus genotypes (isolates with the same assemblage identified using tpiand gdh and b-giardin).

b tpi-based PCR used assemblage-specific primers to identify the mixed infection.

C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276 271

0.4 lM of each primer (Bioneer Q-Oligos, Korea), 2.5 U HotStarTaq�

Plus DNA Polymerase (Qiagen, Hilden, Germany), 1� PCR buffer(Qiagen, Hilden, Germany) and 200 lM dNTP (Fermentas, Ontario,Canada) in a total volume of 50 ll. 2 ll of DNA template wereadded and the mixture was run in the MyCycler thermal cycler(Bio-Rad, Hercules, USA) under the following conditions: Initialhot start at 95 �C for 5 min, initial denaturation at 95 �C for15 min, followed by 35 amplification cycles at 95 �C for 30 s,65 �C for 30 s (55 �C for secondary PCR), 72 �C for 60 s, and a finalextension at 72 �C for 7 min.

In all the PCR reactions, a Giardia-positive DNA specimenand distilled water were used as positive and negative control.The PCR products were analyzed using 2% agarose gel (Vivantis)electrophoresis stained with SYBR� Safe DNA (Invitrogen, Auck-land, New Zealand).

Table 2The distribution of assemblages A and B based on the different loci and mixed infection.

Isolate tpi gdh bg tpi-Mixed

TB11.6 A (A2) A (A2) A (A2) A + BTG2.5 A (A2) A (A2) A (A3) BTG13.2 A(A2) A (A2) A (A3) BTB13.3 A (A2) A (A2) – A + BSB14.1 A (A2) A (A2) – –ST15.1 A (A2) A (A2) – A + BKK16.4 A (A2) – A (A2) –KK16.5 A (A2) – A (A2) A + BTB7.3 A (A2) A (A2) A + BSE10.1 A (A2) – A (A3) BSE17.1 A (A2) – A (A2) ATB15.3 A (A2) – A (A2) A + BTG9.2 A (A2) – A (A2) A + BPD1.1 A (A2) – A (A3) A + BPD11.2 A (A2) – A (A2) A + BSE29.3 A (A2) – A (A2) –KK11.3 A (A2) – A + BKM4.1 A (A2) – – –TB3.3 A (A2) – – A + BTB13.2 A (A2) – – A + BSE25.2 A (A2) – – A + BKK1.6 A (A2) – – A + BCK1.2 A (A2) – – BCK10.2 A (A2) – – A + BSB1.1 A (A2) – – A + BSE25.4 A – – A + BST10.3 – A (A2) – BTB12.3 – – A (A2) –KK7.5 – – A (A3) –SE29.4 – – A (A2) –KM15.1 B B B BTB2.3 B B B BTB4.1 B B B A + BTB5.1 B B B A + BCK20.2 B B B A + BSE7.1 B B B A + BSE12.1 B B B BSE24.2 B B B BKN7.4 B B – A + B

2.4. DNA sequencing and sequences analysis

Successfully amplified PCR products with the nested-PCR foreach locus (except for tpi assemblage-specific PCR) were sequencedin both directions using Applied Biosystems 3730 xl DNA Analyzer(Applied Biosystems, USA). The chromatograms and sequences gen-erated from this study were viewed and assembled using BioEdit Se-quence Alignment Editor Programme (http://www.mbio.ncsu.edu).Preliminary similarity comparison of the consensus sequence withthe sequences in GenBank database was made using Basic LocalAlignment Search Tool (BLAST) (http://blast.ncbi.nlm.nih.gov). Theisolate sequences were genotyped into assemblage and sub-assem-blage using multiple alignments implemented by ClustalW (Thomp-son et al., 1994) with previously defined reference sequencesretrieved from GenBank database. Phylogenetic analysis was per-formed in MEGA 5 (www.megasoftware.net) using neighbour-join-ing (NJ) algorithms with evolutionary distances calculated byKimura-2 parameter method (Kimura, 1980) and 1000 bootstrap va-lue. Sequences from this study were deposited in the GenBank underaccession numbers KC313907- KC313948.

3. Results

3.1. Identification of G. duodenalis assemblages and mixedassemblages

From a total of 484 participants, 84 (17.0%) were found in-fected with Giardia using microscopy. The 84 microscopy-posi-

Isolate tpi gdh bg tpi-Mixed

SM1.2 B B – A + BKM13.2 B B – A + BTB9.2 B B – A + BCK7.1 B B – A + BSE13.1 B B – BTB15.1 B B – A + BSE9.3 B – B A + BKN6.4 B – – A + BTB9.1 B – – A + BUM3.1 B – – A + BSE8.4 B – – A + BSE27.6 B – – BSE27.7 B – – BKK2.3 B – – A + BKK2.5 B – – A + BKM7.3 – B – –KM8.1 – B – –KM15.2 – B – BST13.2 – B – BST20.3 – B – BKM8.4 – B – –CK1.4 – B – A + BSB4.1 – – B –ST8.1 A (A2) B A (A2) BKN7.2 A (A2) B – BSM6.2 A (A2) B – A + BKM3.2 A (A2) B – BKM12.3 A (A2) B – A + BTB11.2 A (A2) B – A + BSB15.1 A (A2) B – A + BPD20.1 B B A (A3) A + BSE18.4 A B – A + BKK6.1 – – – A + BCK20.1 – – – A + BST1.3 – – – BST3.2 – – – BST21.1 – – – BTB2.5 – – – BSE13.2 – – – B

272 C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276

tive specimens were analyzed at three loci (tpi, gdh, and bg) anda set of assemblage-specific PCRs was used to detect mixedinfection. PCR amplicons from 78 specimens using at least oneof the molecular markers were amplified and subsequently se-quences were generated for 71 specimens. The alignment analy-sis of the sequences obtained from tpi, gdh, and bg genes withreference sequences from the GenBank database identifiedassemblages A and B in 30 (42%) and 32 (45%) specimens,respectively. Discordant genotype results were noted in 9 (13%)specimens. At tpi gene, PCR amplicons were produced from70% of the examined specimens (59/84). Among them, assem-blages A and B were found in 34 (58%) and 25 (42%) specimens,respectively. At gdh gene, 38 (45%) specimens were positive withassemblage B being more frequently observed compared toassemblage A (31 vs 7). PCR analysis of bg gene had the lowestdetection rate [33% (28/84)]. The distribution of assemblage Aand B for bg gene was 18 (64%) and 10 (36%), respectively. MLGswere successful in 11 specimens, which were identified asAssemblages A and B, in 3 and 8 specimens, respectively (Ta-ble 1). As mentioned above assemblage discordant results wereobserved in 9 specimens. One isolates (ST8.1) was identified asassemblage A at tpi and bg genes and as assemblage B at gdhgene. The specimen PD20.1 was identified as assemblage B attpi and gdh genes and as assemblage A at bg gene. Seven speci-mens (KN7.2, SM6.2, KM3.2, KM12.3, TB11.2, SB15.1 and SE18.4)were found to harbour assemblage A at tpi gene and assemblageB at gdh gene. The tpi-mixed protocol showed that 6 of the 9specimens harboured mixed assemblages A and B (Tables 2).

The tpi-based PCR protocol for mixed infection using assem-blage-specific primers was conducted on the 84 Giardia-positivespecimens. Of these, 67 specimens showed amplicons for G.duodenalis assemblages; 43 specimens had mixed infectionswith assemblages A and B, 23 specimens had single infectionwith assemblage B and one case was identified as assemblageA (Table 1 and 2). The specificity of this protocol was confirmedby sequencing selected cases representing assemblages A and B.When the chromatogram was scrutinized at the positions wherenucleotide polymorphism could differentiate assemblage A andB, double peaks were observed in 14 of the mixed isolates.

Fig. 1. Phylogram of G. duodenalis genotypes constructed by neighbour-joininganalysis, based on the nucleotide sequences of tpi retrieved from this studycompared with reference sequences of known assemblages from Genbank. Boot-strap values obtained from 1000 replicates are indicated on branches in percentage.

3.2. Subtyping of G. duodenalis Assemblage A

Based on analysis targeting tpi gene, 34 of the assemblage A iso-lates (Table 2) were classified as subtype A2 based on the phyloge-netic analysis (Fig. 1) and the substitution pattern (Table 3). Theneighbor-Joining tree placed the four representative sequences(TB11.2, KK16.4, TB13.2 and SM6.2) in one cluster with AII se-quence references with high bootstrap support.

At gdh gene, all sequences of the assemblage A showed com-plete homologous to the reference sequence of subtype A2(Accession No. L40510) except for isolate ST10.3 where substitu-tion at two positions were observed (A–G at position 341 and C–T at position 485) (Table 4, Fig. 2). Subtype A2 based on tpi orgdh belongs to sub-assemblage AII (Caccio et al., 2008; Fengand Xiao, 2011).

At bg gene, 18 isolates were identified as assemblage A. Twelvewere distinguished as A2 from A1 at position 606 (C–T) and fromA3 at positions 460 (C–T) and 468 (T–C). Three were typed as A3and the remaining 3 isolates have the subtype similar to A3 but dif-ferent by one position at 460 (C–T/Y). Phylogenetic analysis dem-onstrated that six isolates were typed as A3 (100% bootstrapvalue) while the remaining 12 isolates were grouped as A2(Fig. 3). It should also be noted that subtypes A2 and A3 belongto sub-assemblage AII (Caccio et al., 2008; Feng and Xiao, 2011).

3.3. Subtyping of G. duodenalis assemblage B

At tpi gene, 16 sequences representing 23 isolates which wereidentified as assemblage B, were multiple aligned with 5 referencessequence representing BIV, BIV-like, BIII and BIII-like and analysedfor BIV/BIII specific substitutions according to Wielinga andThompson (2007). The 16 representative sequences had heteroge-neous nucleotides at positions defining the sub-assemblages, mak-ing proposing specific sub-assemblages for assemblage B isolatesnot possible (Table 3). Phylogenetic analysis confirmed the mono-phyletic group of assemblage B (bootstrap = 99%). The Neighbor-Joining formed tree sub-clusters of assemblage B. However, sub-clustering was not supported by bootstrap values (Fig. 1).

At gdh gene, 15 sequences representing 30 isolates had highnucleotide variations, limiting the classification of assemblage Bto sub-assemblages (Table 4). Similar to tpi gene, phylogenetic

Table 3Multiple alignments of tpi sequences from this study with reference sequences obtained from GenBank, representing sub-assemblages of assemblages A and B.

Isolates GenBank accession no. Nucleotide position from the start of the gene

Assemblage A 129 399AI L02120 T CAII (A2) JH U57897 C TAII (A2) TB11.6a KC313923 C T

Assemblage B 39 45 91 100 109 121 162 165 168 210 216 247 271 280 297 354 402 429 438 483BIV Ad 19 AF069560 A T T C G G G T T A C A C A A G A A T ABIV GS/M L02116 . . . . . . . . . . . . . . . . . G . .BIV-Like 7115 AY368167 . . . . . . . . . G . . . . . . . . . .BIII 2434 AY368165 G . C . . . . C C G . . . . . . . G . ABIII-Like 2436 AY368153 G C . . . . . . . . T . . . . . . . . G

CK7.1, SE8.4, TB5.1 KC313907 . . . . . . . . . . . . . . . . . G . .SE24.2, KN7.4,TB4.1 KC313908 . . . . . . . . . G . . . . . . . G . .TB2.3 KC313909 G C C . . . . . . G . . . . . . . . . .TB15.1 KC313910 G C C . . . . . . G . . t . . . . G . .KM13.2 KC313911 G C . . . . . . . G . . t g . . . G . .CK20.2 KC313912 G . C . . . . . C G . . . . G . g G . GTB9.2 KC313913 G C . . . . . . C G . . . . . . . G . .SE9.3 KC313914 . . . . . . . . C G . . . . . . . G . .SE13.1 KC313915 . . . . . . . . C . . . . . . . . G . GUM3.1 KC313916 . C . . . . . . C G . . . . . . . G . .KN6.4 KC313917 . . . . . . . . C G . . . . . . g G . .SE12.1, 27.6, TB9.1, SE7.1 KC313918 . . C . . . . . C G . . . . . . . G . .KK2.3 KC313919 G C C . . . . . C G . . . . . . . G . .SE27.7 KC313920 . . C . . . . C C G . . . . . . . G . .SM1.2 KC313921 G . C . . . . C C . . g . . . . . G . .KM15.1 KC313922 . . . t . a . C C G . . . . . a . G . .

Numbers in bold represent nucleotide substitutions from the start of the gene, which differentiate between subtypes according to Weilinga and Thompson (2007). Dots (.)indicate nucleotides identity. Nucleotides in small letter indicate substitution not in the subtypes-defining positions.

a Representative sequence of isolates identified as A2 (Listed in Table 2). Accession numbers in bold are reference sequences from the GenBank.

Table 4Multiple alignments of gdh sequences from this study with reference sequences obtained from GenBank, representing sub-assemblages of assemblages A and B.

Isolates GenBank accessionno.

Nucleotide position from the start of the gene

Assemblage A 237 246 341 485 603 621AI (Prtland1) M84604 C C A C T CAII (A2) L40510 C C . . C TAII (A2) TB11.6,TG2.5,TG13.2, TB13.3, SB14.1, T15.1 KC313924 ? ? . . C T

ST10.3 KC313925 ? ? G T C T

Assemblage B 309 357 360 429 447 519 540 561 570 576 582 597 612BIII (BAH-12) AF069059 C T G T T C C C C G G C GBIV (Ad-7) L40508 T . . C C . T T . . . . A

SM1.2 KC313926 T . . . C T . . . . . . .CK1.4 KC313927 T . . . . T . . . . . . .KM7.3,TB11.2,KM8.1, SM6.2 KC313928 T . . C C T . . . . . . .KM15.2,ST8.1,KM15.1, ST13.2,KM8.4,ST20.3,TB2.3

KC313929 T . . C C . . . . . . . .

SE13.1 KC313930 T C . C T . . . . . . .KM3.2 KC313931 T C . C C T . . . . . . .TB15.1, SB15.1 KC313932 T C . C C . . . . . . . .SE12.1, TB4.1 KC313933 T C . C C . T . . . . . .TB5.1, SE7.1 KC313934 T C . C C . . . . . . . AKM13.2,CK7.1,TB9.2, SE24.2,SE18.4 KC313935 T C . C C . T . . . . . APD20.1, KN7.4 KC313936 . . . C C . T . . . . . .CK20.2 KC313937 . C . . C . . . . . . . .KM12.3 KC313938 . . . . C . . . . . . . .

Numbers in bold represent nucleotide substitutions from the start of the gene, which differentiate between subtypes according to Weilinga and Thompson (2007). Dots (.)indicate nucleotides identity. Question mark (?) indicates missing nucleotides. Accession numbers in bold are reference sequences from the GenBank.

C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276 273

analysis formed a monophyletic group for assemblage B (Fig. 2).Similar results were observed in bg gene (Table 5, Fig. 3).

4. Discussion

To the best of our knowledge, the present study was the firststudy of giardiasis using multi-locus genotyping (MLG) in Malay-sia. In this study, the prevalence of giardiasis in the Orang Aslicommunities was 17.0%. This is comparable with a recent study

conducted by Anuar et al. (2012) in Orang Asli communities inSelangor state, which showed an overall prevalence of giardiasisat 20.0% (n = 500).

Amplifications of the three PCR assays (tpi, gdh, and bg) wereperformed differently at different loci. Of the 84 Giardia microscop-ically-positive specimens, only 11 isolates were able to be ampli-fied at three loci. The tpi locus achieved the highest percentageof amplicons produced (70%), followed by gdh (45%) and bg(33%). Similar occurrences were also reported in previous studies

Fig. 2. Phylogram of G. duodenalis genotypes constructed by neighbour-joininganalysis, based on the nucleotide sequences of gdh retrieved from this studycompared with reference sequences of known assemblages from Genbank. Boot-strap values obtained from 1000 replicates are indicated on branches in percentage.

Fig. 3. Phylogram of G. duodenalis genotypes constructed by neighbour-joininganalysis, based on the nucleotide sequences of bg retrieved from this studycompared with reference sequences of known assemblages from Genbank. Boot-strap values obtained from 1000 replicates are indicated on branches in percentage.

274 C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276

(David et al., 2011; Lalle et al., 2009). Discordant genotyping re-sults at the three loci (tpi, gdh, and bg) have been observed in 9 iso-lates (13%) in this study, which is in agreement with previousreports (Caccio et al., 2008; Sprong et al., 2009; Thompson et al.,2009; Yang et al., 2010a). An analysis of sequences from four ge-netic loci (tpi, gdh, bg and rDNA) has been conducted on 61 humanisolates and 29 animal isolates of G. duodenalis (Caccio et al., 2008).In this study, incongruent assignment of five human isolates andone macaque isolates to G. duodenalis assemblages was reported(Caccio et al., 2008). Yang et al. (2010) genotyped 124 human iso-lates of G. duodenalis at gdh and rDNA genes, and reported incon-gruent genotyping results in five isolates, where three isolateswere classified as assemblage A at rDNA gene and assemblage Bat gdh gene, and two isolates were classified as assemblage B atrDNA gene and assemblage A at gdh gene. Inconsistency in geno-typing results of G. duodenalis has also been reported among dogsand cats isolates, where 7 isolates (3 dogs and 4 cats) were classi-fied as assemblage D at rDNA gene, and as assemblage B (3 iso-lates), assemblage C (3 isolates) and assemblage E (1 isolate) atgdh gene. One dog isolate was classified as assemblage C at rDNAand assemblage B at gdh gene (Read et al., 2004). Thompsonet al. (2009) genotyped G. duodenalis isolated from 70 cayotes atrDNA and gdh, and reported four incongruent genotyping results.

The implication of this inconsistency has provoked questionsabout the strength and validity of genotyping relying on single lo-cus and hence, the true picture of Giardia diversity has a highchance of being masked by the data generated from analysis usingsingle marker (Feng and Xiao, 2011). Although the underlyingmechanisms remain uncertain, Caccio and Ryan (2008) have re-viewed a number of factors that could contribute to the incongru-ent assignment of assemblages by different makers. These factorsinclude meiotic recombination, which suggests the potential ofsexual reproduction in Giardia, introgression that involves backcrossing of hybrids with parental species and retention of ancestralpolymorphism that could lead to presence of identical alleles ingenetically distinct groups. Another contributing factor whichwould be discussed further in this context is the occurrence ofmixed infections. With the use of the assemblage-specific assays,high prevalence (64%) of mixed infections was detected and furtherproven the biasness of using a standard PCR assay alone as themost plentiful assemblage would be amplified in preference (Wiel-inga and Thompson, 2007). Thus, the feature of mixed infectionscould be applied in the present study to explain the discordantgenotype results.

The isolates with assemblage A were all identified as A2, whichbelongs to sub-assemblage AII. This is similar with other studies

Table 5Multiple alignments of bg sequences from this study with reference sequences obtained from GenBank, representing sub-assemblages of A and assemblages B.

Isolates Genbank accession no.

Assemblage A 450 460 468 606 729AI-A1 Portland 1 X85958 C C T C AAII-A2 KC8 AY072723 . . . T GAII-A3 AY072724 . T C T AA2 KK16.4a KC313945 . . . T ?A3 KK7.5, TG2.5 KC313946 . T C T ?A3 SE10.1 KC313947 . Y C T ?A3 PD20.1, PD1.1, TG13.2 KC313948 . . C T ?

Assemblage B 210 228 354 378 438 540 564B AY072727 C A C C C G T

SB4.1, SE7.1 KC313939 T . T . . A .TB2.3 KC313940 T . . . A .SE9.3 KC313941 . . . . . A .TB4.1 KC313942 Y . T . . A .SE24.2 KC313943 . . T . . . .TB5.1, CK20.2, SE12.1 KC313944 . . T . . A .

Numbers in bold represent nucleotide substitutions from the start of the gene, which differentiate between subtypes according to Wielinga and Thompson (2007). Dotsindicate nucleotides identity. Question mark (?) indicates missing nucleotides. The dash (–) indicates deletion.

a Representative sequence of isolates identified as A2. Accession numbers in bold are reference sequences from the GenBank.

C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276 275

where AII was the predominant sub-assemblage in human (Bonho-mme et al., 2011; Hussein et al., 2009; Lebbad et al., 2011). Sub-assemblage AI which was absent from this study was reported tohave preference to cause infection in livestock or companion ani-mals as opposed to AII which was found mainly in human casesthan animals (Sprong et al., 2009). The isolation of anthroponoticassemblages and sub-assemblages (B and AII) in this study impli-cates human as a potential source of the infection. In accordancewith our findings, previous studies conducted in Orang Asli com-munities in Malaysia found that the presence of other family mem-bers infected with G. duodenalis was either the only or the mainrisk factor for giardiasis (Norhayati et al., 1998; Anuar et al.,2012). However, the postulation of human-to-human transmissionis limited by the fact that animals were not included for analysis inthis study.

On the other hand, sub-assemblages were not able to assign toisolates belonged to assemblage B due to high degree of nucleo-tides variation which was not uncommon among other molecularstudies (Bonhomme et al., 2011; Caccio et al., 2008; Lalle et al.,2009; Lebbad et al., 2011; Levecke et al., 2009). This phenomenonled us back to the focus of mixed infection, which is known to takeplace at both the inter- and at the intra-assemblage levels (Caccioand Ryan, 2008). Mixed infection can happen when a host ingestsGiardia cysts of different genetic profiles or subsequent infection ofan infected host by genetically different Giardia cysts. This is espe-cially common in areas where giardiasis is endemic (Caccio andRyan, 2008; Lebbad et al., 2011; Sprong et al., 2009). Therefore,mixed subtype infections (intra-assemblage level) can stand as afactor that causes the high heterogeneity. Besides, high polymor-phism in assemblage B can also be attributed to intra-assemblagerecombination and allelic sequence heterozygosity (ASH). In orderto resolve the question whether the mixed genotype is due tomixed infections or ASH, Ankarklev et al. (2012) had conductedan investigation on single Giardia trophozoite and cyst. The resultbased on molecular analysis positively exhibited the occurrenceof ASH at single cell level. However, high numbers of mixed sub-genotypes infections (demonstrated by variable sequence pat-terns) were also observed in different cysts isolated from the samesample indicating that mixed infection is also playing a role for thehigh heterogeneity.

In conclusion, MLG was conducted to analyze the genetic diver-sity of Giardia infections in Orang Asli communities in Malaysia.While assemblages A and B have almost equal frequency of infec-tions, more than half of the isolates were mixed infections. The

presence of assemblage B and sub-assemblage AII in the samplessuggest that the mode of transmission of giardiasis in Malaysiamay be human-to-human. Nevertheless, the role of animals inthe dynamic of transmission needs to be further investigatedinvolving multilocus genotyping of parasites from animals andhumans.

Acknowledgments

We are grateful to the Orang Asli children and their parents fortheir voluntary involvements and would like to acknowledge thefinancial support from UMRG grants RG302-11HTM and RG439/12HTM, and the student grant (PV063- 2012A).

References

Al-Mekhlafi, H.M., Azlin, M., Nor Aini, U., Shaik, A., Sa’iah, A., Fatmah, M.S., Ismail,M.G., Ahmad Firdaus, M.S., Aisah, M.Y., Rozlida, A.R., Norhayati, M., 2005.Giardiasis as a predictor of childhood malnutrition in Orang Asli children inMalaysia. Trans. R. Soc. Trop. Med. Hyg. 99, 686–691.

Al-Mekhlafi, H.M., Surin, J., Sallam, A.A., Abdullah, A.W., Mahdy, M.A.K., 2010.Giardiasis and poor vitamin A status among aboriginal school children in ruralMalaysia. Am. J. Trop. Med. Hyg. 83, 523–527.

Ankarklev, J., Svard, S.G., Lebbad, M., 2012. Allelic sequence heterozygosity in singleGiardia parasites. BMC Microbiol. 12, 65.

Anuar, T.K., Al-Mekhlafi, H.M., Abdul Ghani, M.K., Osman, E., Mohd Yasin, A., Nordin,A., Nor Azreen, S., Md Salleh, F., Ghazali, N., Bernadus, M., Moktar, N., 2012.Giardiasis among different tribes of Orang Asli in Malaysia: highlighting thepresence of other family members infected with Giardia intestinalis as a mainrisk factor. Parasitol. Int. J. Parasitol. 42, 871–880.

Bonhomme, J., Goff, L.L., Lemee, V., Gargala, G., Ballet, J.-J., Favennec, L., 2011.Limitations of tpi and bg genes sub-genotyping for characterization of humanGiardia duodenalis isolates. Parasitol. Int. 60, 327–330.

Buret, A.G., 2008. Pathophysiology of enteric infections with Giardia duodenalius.Parasite 15, 261–265.

Caccio, S.M., Ryan, U., 2008. Molecular epidemiology of giardiasis. Mol. Biochem.Parasitol. 160, 75–80.

Caccio, S.M., De Giacomo, M., Pozio, E., 2002. Sequence analysis of the beta-giardingene and development of a polymerase chain reaction-restriction fragmentlength polymorphism assay to genotype Giardia duodenalis cysts from humanfaecal samples. Int. J. Parasitol. 32, 1023–1030.

Caccio, S.M., Beck, R., Lalle, M., Marinculic, A., Pozio, E., 2008. Multilocus genotypingof Giardia duodenalis reveals striking differences between assemblages A and B.Int. J. Parasitol. 38, 1523–1531.

Daly, E.R., Roy, S.J., Blaney, D.D., Manning, J.S., Hill, V.R., Xiao, L., Stull, J.W., 2010.Outbreak of giardiasis associated with a community drinking-water source.Epidemiol. Infect. 138, 491–500.

David, E.B., Coradi, S.T., Oliveira-Sequeira, T.C.G., Ribolla, P.E.M., Katagiri, S.,Guimaraes, S., 2011. Diagnosis of Giardia infections by PCR-based methods inchildren of an endemic area. J. Venom. Anim. Toxins 17, 209–215.

Feng, Y.Y., Xiao, L.H., 2011. Zoonotic potential and molecular epidemiology ofGiardia Species and giardiasis. Clin. Microbiol. Rev. 24, 110.

276 C.S. Huey et al. / Infection, Genetics and Evolution 17 (2013) 269–276

Geurden, T., Geldhof, P., Levecke, B., Martens, C., Berkvens, D., Casaert, S.,Vercruysse, J., Claerebout, E., 2007. Mixed Giardia duodenalis assemblage Aand E infections in calves. Int. J. Parasitol. 38, 259–264.

Geurden, T., Levecke, B., Caccio, S.M., Visser, A., De Groote, G., Casaert, S., Vercruysse,J., Claerebout, E., 2009. Multilocus genotyping of Cryptosporidium and Giardia innon-outbreak related cases of diarrhoea in human patients in Belgium.Parasitology 136, 1161–1168.

Hussein, A.I.A., Yamaguchi, T., Nakamoto, K., Iseki, M., Tokoro, M., 2009. Multiple-subgenotype infections of Giardia intestinalis detected in Palestinian clinicalcases using a subcloning approach. Parasitol. Int. 58, 258–262.

Kimura, M., 1980. A simple method for estimating evolutionary rates of basesubstitutions through comparative studies of nucleotide sequences. J. Mol. Evol.16, 111–120.

Lalle, M., Pozio, E., Capelli, G., Bruschi, F., Crotti, D., Caccio, S.M., 2005. Geneticheterogeneity at the beta-giardin locus among human and animal isolates ofGiardia duodenalis and identification of potentially zoonotic subgenotypes. Int. J.Parasitol. 35, 207–213.

Lalle, M., Bruschi, F., Castagna, B., Campa, M., Pozio, E., Caccio, S.M., 2009. Highgenetic polymorphism among Giardia duodenalis isolates from Sahrawichildren. Trans. R. Soc. Trop. Med. Hyg. 103, 834–838.

Lebbad, M., Petersson, I., Karlsson, L., Botero-Kleiven, S., Andersson, J.O.,Svenungsson, B., Svard, S.G., 2011. Multilocus genotyping of human giardiaisolates suggests limited zoonotic transmission and association betweenassemblage B and flatulence in children. PLoS Negl. Trop. Dis. 5 (8), e1262.

Levecke, B., Geldhof, P., Claerebout, E., Dorny, P., Vercammen, F., Caccio, S.M.,Vercruysse, J., Geurden, T., 2009. Molecular characterisation of Giardiaduodenalis in captive non-human primates reveals mixed assemblage A and Binfections and novel polymorphisms. Int. J. Parasitol. 39, 1595–1601.

Lim, Y.A.L., Ahmad, R.A., Smith, H.V., 2008. Current status and future trends inCryptosporidium and Giardia epidemiology in Malaysia. J. Water Health 6, 239–254.

Lim, Y.A.L., Lai, M.M., Mahdy, M.A.K., Mat Naim, H.R., Smith, H.V., 2009a. Moleculardetection of Giardia contamination in water bodies in a zoo. Environ. Res. 109,857–859.

Lim, Y.A.L., Ramasame, S.D., Mahdy, M.A.K., Sulaiman, W.Y., Smith, H.V., 2009b.Detection and molecular characterization of Giardia isolated from recreationallake water in Malaysia. Parasitol. Res. 106, 289–291.

Lim, Y.A.L., Iqbal, A., Surin, J., Sim, B.L., Jex, A.R., Nolan, M.J., Smith, H.V., Gasser, R.B.,2011. First genetic classification of Cryptosporidium and Giardia from HIV/AIDSpatients in Malaysia. Infect. Genet. Evol. 11, 968–974.

Lim, Y.A.L., Mahdy, M.A.K., Tan, T.K., Goh, X.T., Jex, A.R., Nolan, M.J., Sharma, R.S.,Gasser, R.B., 2013. First molecular characterization of Giardia duodenalis fromgoats in Malaysia. Mol. Cell. Probes 27, 28–31.

Mahdy, M.A.K., Surin, J., Mohd-Adnan, A., Wan, K.L., Lim, Y.A.L., 2009. Molecularcharacterization of Giardia duodenalis isolated from Semai Pahang Orang Asli(Peninsular Malaysia aborigines). Parasitology 136, 1237–1241.

Nolan, M.J., Jex, A.R., Pangasa, A., Young, N.D., Campbell, A.J., Stevens, M., Gasser,R.B., 2010. Analysis of nucleotide variation within the triose-phosphateisomerase gene of Giardia duodenalis from sheep and its zoonoticimplications. Electrophoresis 31, 287–298.

Noor Azian, M.Y., San, Y.M., Gan, C.C., Yusri, M.Y., Nurulsyamzawaty, Y., Zuhaizam,A.H., Maslawaty, M.N., Norparina, I., Vythilingam, I., 2007. Prevalence ofintestinal protozoa in an aborigine community in Pahang. Malaysia Trop.Biomed. 24, 55–62.

Norhayati, M., Penggaben, M., Oothuman, P., Fatmah, M.S., 1998. Prevalence andsome risk factors of Giardia duodenalis infection in a rural community inMalaysia. Southeast Asian J. Trop. Med. Public Health 29, 735–738.

Plutzer, J., Ongerth, J., Karanis, P., 2010. Giardia taxonomy, phylogeny andepidemiology: facts and open questions. Int. J. Hyg. Environ. Health 213, 321–333.

Read, C.M., Monis, P.T., Thompson, R.C., 2004. Discrimination of all genotypes ofGiardia duodenalis at the glutamate dehydrogenase locus using PCR-RFLP. Infect.Genet. Evol. 4, 125–130.

Robertson, L., Gjerde, B., Hansen, E.F., Stachurska-Hagen, T., 2009. A watercontamination incident in Oslo, Norway during October 2007; a basis fordiscussion of boil-water notices and the potential for post-treatmentcontamination of drinking water supplies. J. Water Health 7, 55–66.

Sprong, H., Caccio, S.M., van der Giessen, J.W., 2009. Identification of zoonoticgenotypes of Giardia duodenalis. PLoS Negl. Trop. Dis. 3, e558.

Sulaiman, I.M., Fayer, R., Bern, C., Gilman, R.H., Trout, J.M., Schantz, P.M., Das, P., Lal,A.A., Xiao, L., 2003. Triosephosphate isomerase gene characterization andpotential zoonotic transmission of Giardia duodenalis. Emerg. Infect. Dis. 9,1444–1452.

Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving thesensitivity of progressive multiple sequence alignment through sequenceweighting, position-specific gap penalties and weight matrix choice. NucleicAcids Res. 22, 4673–4680.

Thompson, R.C., Hopkins, R.M., Homan, W.L., 2000. Nomenclature and geneticgroupings of Giardia infecting mammals. Parasitol. Today 16, 210–213.

Thompson, R.C., Colwell, D.D., Shury, T., Appelbee, A.J., Read, C., Njiru, Z., Olson, M.E.,2009. The molecular epidemiology of Cryptosporidium and Giardia infections incoyotes from Alberta, Canada, and observations on some cohabiting parasites.Vet. Parasitol. 159, 167–170.

van der Giessen, J.W., de Vries, A., Roos, M., Wielinga, P., Kortbeek, L.M., Mank, T.G.,2006. Genotyping of Giardia in Dutch patients and animals: a phylogeneticanalysis of human and animal isolates. Int. J. Parasitol. 36, 849–858.

Wielinga, C.M., Thompson, R.C., 2007. Comparative evaluation of Giardia duodenalissequence data. Parasitology 134, 1795–1821.

Yang, R., Lee, J., Ng, J., Ryan, U., 2010. High prevalence Giardia duodenalis assemblageB and potentially zoonotic subtypes in sporadic human cases in WesternAustralia. Int. J. Parasitol. 40, 293–297.