Rodent-Borne Hantaviruses in Cambodia, Lao PDR,and Thailand

Kim Blasdell,1,2 Jean Francois Cosson,3 Yannick Chaval,4 Vincent Herbreteau,5

Bounneuang Douangboupha,6 Sathaporn Jittapalapong,7 Ake Lundqvist,8 Jean-Pierre Hugot,9

Serge Morand,2 and Philippe Buchy1

1Virology Unit, Institut Pasteur du Cambodge, 5 Monivong Boulevard, BP 983, Phnom Penh, Cambodia2Institut des Sciences de l’Evolution, CNRS, IRD, Universite Montpellier 2, CC065, 34095 Montpellier Cedex 05, France3INRA, UMR CBGP (INRA/IRD/Cirad/Montpellier SupAgro), Campus International de Baillarguet, CS 30016, 34988 Montferrier sur Lez, France4Centre de Biologie et de Gestion et des Populations (CBGP), International de Baillarguet, CS 30016, 34988 Montferrier sur lez, France5CIRAD, UR AGIRs (Animal et Gestion Integree des Risques), Campus International de Baillarguet, Montpellier, France6National Agricultural Research Centre, National Agricultural and Forestry Research Institute, Vientiane, Lao PDR7Department of Parasitology, Faculty of Veterinary Medicine, Kasetsart University, Bangkok 10900, Thailand8Swedish Institute for Infectious Disease Control, Stockholm, Sweden9Origine, Structure et Evolution de la Biodiversite, Museum National d’Histoire Naturelle, Paris, France

Abstract: In order to evaluate the circulation of hantaviruses present in southeast Asia, a large scale survey of

small mammal species was carried out at seven main sites in the region (Cambodia, Lao People’s Democratic

Republic, and Thailand). Small scale opportunistic trapping was also performed at an eighth site (Cambodia).

Using a standard IFA test, IgG antibodies reacting to Hantaan virus antigens were detected at six sites.

Antibody prevalence at each site varied from 0 to 5.6% with antibodies detected in several rodent species

(Bandicota indica, B. savilei, Maxomys surifer, Mus caroli, M. cookii, Rattus exulans, R. nitidius, R. norvegicus,

and R. tanezumi). When site seroprevalence was compared with site species richness, seropositive animals were

found more frequently at sites with lower species richness. In order to confirm which hantavirus species were

present, a subset of samples was also subjected to RT-PCR. Hantaviral RNA was detected at a single site from

each country. Sequencing confirmed the presence of two hantavirus species, Thailand and Seoul viruses,

including one sample (from Lao PDR) representing a highly divergent strain of Seoul virus. This is the first

molecular evidence of hantavirus in Lao PDR and the first reported L segment sequence data for Thailand

virus.

Keywords: hantavirus, southeast Asia, rodents, serology, phylogeny, seoul virus variant

INTRODUCTION AND PURPOSE

Hantaviruses are single-stranded, tri-segmented, negative

sense RNA viruses belonging to the family Bunyaviridae.

Electronic supplementary material: The online version of this article

(doi:10.1007/s10393-011-0725-7) contains supplementary material, which is available

to authorized users.

Published online: November 29, 2011

Correspondence to: Philippe Buchy, e-mail: pbuchy@pasteur-kh.org

EcoHealth 8, 432–443, 2011DOI: 10.1007/s10393-011-0725-7

Original Contribution

� 2011 International Association for Ecology and Health

At present, 45 distinct hantavirus species are recognized, all

associated with either rodent or soricomorph hosts (Dearing

and Dizney 2010). The genus is further subdivided into

species hosted by Sigmodontinae, Neotominae, Murinae,

and Arvicolinae rodent families and by the Soricidae and

Talpidae insectivorous families (Henttonen et al. 2008).

Hantaviruses are found throughout Eurasia, the

Americas, and Africa, where pathogenic variants are jointly

responsible for up to 200,000 human clinical cases annually

(Dearing and Dizney 2010; Song et al. 1984). Two syn-

dromes are caused by hantaviruses in humans: hantavirus

cardiopulmonary syndrome (HCPS), which is restricted to

the Americas and hemorrhagic fever with renal syndrome

(HFRS), with a much wider distribution including Eurasia,

Africa, and a few isolated cases in the Americas. Hantavirus

species hosted by Sigmodontinae and Neotominae rodents

are responsible for HCPS, while HFRS-associated hantavi-

ruses are generally hosted by Murinae rodents. However,

infections with Puumala virus, an arvicoline-borne virus,

are also associated with HFRS (Terajima et al. 2004). Tula

virus, another arvicoline-borne virus, is also potentially

linked with human infections, although disease has not

been confirmed (Schultze et al. 2002; Klempa et al. 2003).

With the exception of China, hantaviruses have been

poorly studied in southeast (SE) Asia. In China, 40,000–

60,000 human cases are reported annually (Bi et al. 2008)

but only isolated cases have been reported from other

countries in the region (e.g., Thailand, Suputthamongkol

et al. 2005; Singapore, Wong et al. 1989). However, pre-

vious studies have identified at least six hantavirus species

in rodent hosts in the region. These include Hantaan virus,

hosted by Apodemus agrarius (Lee et al. 1978), restricted

primarily to China; Seoul virus hosted by Rattus norvegicus

(Lee et al. 1982), which probably has a worldwide distri-

bution; Thailand virus, hosted by Bandicota indica and

present in Thailand (Xiao et al. 1994; Hugot et al. 2006);

Serang virus, hosted by R. tanezumi and recently identified

in Indonesia (Plyusnina et al. 2009); Thottapalayam virus,

hosted by Suncus murinus and found in India, Thailand,

and Indonesia (Carey et al. 1971; Okumura et al. 2007);

Cao Bang virus, hosted by Anurosorex squamipes and found

in Vietnam (Song et al. 2007). Other hantavirus species

may also be present, including partially characterized spe-

cies hosted by Rattus species in Cambodia (Reynes et al.

2003) and R. tanezumi in Singapore (Johansson et al. 2010)

which appear to be variants of Serang virus. Distinct

genetic variants of Seoul virus have also been identified in

China (Zou et al. 2008).

Although hantaviruses were previously thought to

share co-phylogeny with their rodent hosts (Khaiboullina

et al. 2005), recent findings of multiple new hantaviruses

and in particular, the sorciomorph-borne viruses, pose

some problems for the co-phylogeny theory and have called

this idea into debate (Henttonen et al. 2008; Ramsden et al.

2009; Schmidt-Chanasit et al. 2010). Each hantavirus spe-

cies or lineage is thought to be hosted predominantly by a

single rodent or soricomorph species or group of closely

related species. However, spill-over infections into other

species also commonly occur [spill-over of Dobrava virus

A. agrarius lineage into A. flavicollis (Schlegel et al. 2009);

Hantaan virus, usually hosted by Apodemus species but also

detected in R. norvegicus and Niviventer spp. (Yao et al.

2002; Li et al. 2005; Zou et al. 2008); Seoul virus infecting

several Rattus species and Mus musculus (Wang et al. 2000;

Jiang et al. 2008)]. These spill-over events may even have

led to host-switching events in some cases, further com-

plicating co-phylogeny analyses. Evidence for host-switch-

ing events has been proposed for both rodent- and

soricomorph-borne hantaviruses (Vapalahti et al. 1999;

Kang et al. 2009).

As SE Asia is a biodiversity hotspot with numerous

rodent and soricomorph species, unrecognized virus spe-

cies and/or hantavirus host species and lineages may be

present in the region. Our study aimed to determine the

prevalence of hantaviruses in a variety of SE Asian habitats

(including both disturbed and relatively undisturbed areas)

and to assess the virus species present. The data were also

assessed to establish if hantavirus presence was related to

the level of species richness in the rodent communities.

METHODS

Rodent Samplings

This research is part of the CERoPath (Community Ecology

of Rodents and their Pathogens in a changing environ-

ment) project, which aims to investigate the dynamics of

murine rodents and their pathogens (macro and micro-

parasites) in SE Asia. This region is characterized by high

biodiversity and growing modification of habitats due to

rapid economic development and insertion into the global

economy. Seven main sampling locations were selected:

Nan (19�150N; 100�830E), Loei (17�390N; 101�770E) and

Buriram (14�890N; 103�010E) in Thailand; Luang Prabang

(19�620N; 102�050E) and Champasak (15�120N; 105�800E)

in Laos; and Preah Sihanouk (10�710N; 103�820E) and

Hantaviruses in Southeast Asian Rodents 433

Mondolkiri (12�120N; 106�890E) in Cambodia (see Fig. 1).

These locations represent a variety of habitats, in relation to

human pressures and land usage (see below). Opportunistic

trapping was also carried out on a much smaller scale in the

environs of Phnom Penh, Cambodia (11�330N; 104�550E).

For the seven main localities, two sampling sessions

were conducted, one during the wet season and one during

the dry season. At each of these locations, rodents were

sampled from four habitat types: forest (including natural

forest and plantations); non-flooded habitats (including

shrub, orchards and upland agriculture); lowland flooded

habitats (rice paddies); and peridomestic (houses and

immediate surrounding areas). Trapping took place within

an area of approximately 100 km2. For each session 30 lines

of ten traps were set over 4 days, amounting to 1,200 trap

nights. These trap nights were evenly distributed between

the first three habitat types (forest, non-flooded habitats,

and lowland flooded habitats) with 400 trap nights set in

each habitat. In addition, animals were also trapped by

local hunters in peridomestic habitats, ranging from 334

(Loei) to 822 (Buriram) trap nights per site with a mean of

546 trap nights. Greater use of local hunters was employed

at sites with low species richness in an attempt to reach

local rodent species richness. Local handmade wire live

traps, baited with cassava, banana, and sticky rice were set

for 4 days at each line and checked daily. Animals were

euthanized and a blood pellet, serum, and organs were

collected, stored temporarily in liquid nitrogen and then

transferred to long term storage at -80�C.

In Phnom Penh, animals were captured at several loca-

tions within and around the city from both domestic

dwellings and agricultural land on an opportunistic basis.

Traps, baits, and animal processing were as described above.

Rodent Species Determination

Rodent species were determined in the field when possible

using morphological criteria as described in the literature

(Marshall 1988; Aplin et al. 2003; Pages et al. 2010). Most

Bandicota were determined using the criteria used by Aplin

et al. (2003) but ambiguous specimens (mainly subadults)

were subjected to a PCR targeting the D-loop gene on the

mitochondrial DNA (Chaval et al. 2010). As morphological

criteria were not fully discriminant between some species

within the genera Rattus, Niviventer, and Mus, a molecular

approach was thus carried out by sequencing the COI gene.

Species determination was performed using the freely

available molecular tool developed for SE Asian rodent

species molecular identification (http://www.ceropath.org/

barcoding_tool/rodentsea).

Within the seven main sites, rodent species richness was

estimated both by the observed number of species (Sobs) and

the estimated number corrected for sample size (Sk). Sk was

calculated by the rarefaction procedure in BiodivR v. 1.1

(Hardy 2009) and corresponds to the expected number of

species in a sample of k rodents. In our case, k was fixed to 170,

i.e., the lowest sample size recorded among our seven sites.

Serology

Hantaan virus-infected Vero E6 cells fixed on slides were

used to perform indirect fluorescent antibody (IFA)

serology using previously described methods (Schilling

et al. 2007). As hantavirus antibodies are known to cross

react between hantavirus species (Hujakka et al. 2003),

species determination was not possible. Serum samples

were diluted 1:30 with PBS. Secondary antibody,

Figure 1. Geographic locations of trapping sites.

434 Kim Blasdell et al.

comprising a mixture of goat anti mouse IgG and goat anti-

rat IgG, both conjugated with fluorescein isothiocyanate

(FITC) (Sigma, Steinheim, Germany), was used at a dilu-

tion of 1:50 in PBS. Fluorescence was visualized using ultra

violet microscopy.

Hantavirus Molecular Identification

Lung samples from seropositive animals, from additional

individuals belonging to species with high seroprevalence

and from individuals where serum samples were not

available for screening by IFA, were selected for molecular

testing. Lung samples were homogenized (50 mg sample in

500 ll PBS) using the MagNA Lyzer Instrument (230 V,

Roche, cat. no. 03358976001) and bead system, then RNA

extracted using the Qiagen Viral RNA extraction kit (cat.

no. 52960). To determine hantavirus species, samples were

screened for hantavirus RNA using previously described

reverse-transcription nested RT-PCRs targeting the L

segment (Klempa et al. 2006) and S segment (Papa et al.

1998). The Superscript� III one-step RT-PCR system with

Platinum� Taq (Invitrogen, cat. no. 12574026) was used

for first round PCR and Promega GoTaq� Flexi DNA

polymerase (cat. no. M8295) was used for nested PCRs.

Amplicons of the expected size (324 and 945 bp for L and S

segments, respectively) were sequenced using a MJ

Research PTC-225 Peltier Thermal Cycler and ABI PRISM

BigDyeTM Terminator Cycle Sequencing Kits with Amp-

liTaq DNA polymerase (FS enzyme: Applied Biosystems).

For phylogenetic analysis, sequences for other hantavi-

ruses were obtained from GenBank. Specifically, published

data documented from the Asian region was collated (Sup-

plementary Table). Because the rodents collected in this

study almost entirely consisted of species belonging to the

Murinae subfamily, viruses isolated from murines (rats and

mice) and from human patients were focused on. Hantavi-

ruses from voles, sigmodontines, and insectivores were not

included in the analysis due to their distant phylogenetic

relationship to murine-borne hantaviruses. Data sets of S and

L segments were analyzed separately. Sequences were aligned

using the Muscle option in Seaview software v.4.2.12 (Gouy

et al. 2010). As both partial segments corresponded to coding

sequences, predicted amino acid sequences were also ana-

lyzed. Percentage differences between sequences (constitut-

ing the number of nucleotides/amino acids differing between

two sequences, by sequence length) were calculated in MEGA

4.0 (Tamura et al. 2007). The best model of evolution for each

data set was determined using jModelTest 0.1.1 (Posada

2008). Phylogenetic reconstructions were performed using

the Maximum Likelihood method of Phyml 3.0 (Guindon

and Gascuel 2003), implemented in Seaview using the model

of evolution selected by jModelTest (GTR model), and

visualized in FigTree v1.3.1 (http://tree.bio.ed.ac.uk/). Nodal

supports were determined using the approximate likelihood

ratio test approach.

RESULTS

Trapping Effort

During sessions where species richness was low, hunter

trapping pressure was increased. Hence there was a sig-

nificant negative correlation between collected rodent

species richness and local hunter trapping nights (in log;

slope = -0.37, df = 7, adjusted R2: 0.45, P 0.028), indi-

cating that increasing trapping by local hunters did not

significantly increase rodent species richness.

Hantavirus Seroprevalence

A total of 2,364 small mammals were collected during this

study, of which 1,640 rodents were tested by IFA for IgG

antibodies to Hantaan virus and 50 tested positive. Anti-

Hantaan virus antibodies were detected in nine species,

namely (in order of prevalence) R. nitidius, B. indica,

R. norvegicus, B. savilei, R. exulans, R. tanezumi, M. caroli,

Maxomys surifer, and M. cookii (Table 1). Four species did

not test positive despite a significant number of samples

analyzed (�35): Berylmys berdmorei, R. argentiventer,

R. losea, and M. cervicolor. Six species captured and tested

in low numbers (�23) also all tested negative: B. bowersi,

Leopoldamys edwardsi, Niviventer fulvescens, R. andaman-

ensis and the non-murine species Cannomys badius and

Rhizomys pruinosus. Given the low number of captures for

these species, we would be unlikely to be able to detect a

low prevalence of hantavirus infection in these populations.

Between zero and five species were found seropositive at

any given site. Seropositive R. exulans were identified at five

sites; R. tanezumi at three sites; B. indica and B. savilei at

two sites; and R. nitidus, R. norvegicus, M. surifer, M. cookii,

and M. caroli each at a single site (Table 1).

Anti-hantavirus antibodies were detected in every

country and at six of the seven sites studied. All main sites

were sampled twice, but serum was only available from the

second trapping session in Loei, Thailand. Overall hanta-

virus antibody prevalence at the trapping sites was 0% for

Hantaviruses in Southeast Asian Rodents 435

Phnom Penh, 3.9% for Mondulkiri in March 2009 but 0%

in November 2009; 0% in April 2009 and 3.2% in

November 2009 for Champasak; 4.4% for Luang Prabang

in July 2008 and 3.9% in February 2010; 5.1% for Nan in

November 2008 and 6.4% in March 2010; 1.2% for Buri-

ram in November 2008 and 0.8% in June 2009; 4.5% in

November 2008 and 5.1% in July 2009 for Preah Sihanouk.

Serological testing of samples did not reveal any seroposi-

tive individuals in Loei despite a large sampling size (196

rodents tested from session two only) (Table 1).

Overall, serological evidence of previous exposure to

hantavirus was found to be negatively correlated with

rodent richness (Kendall’s rank correlation, Sobs:

P = 0.020; Sk: P = 0.011) (Table 1; Fig. 2).

Molecular Analysis

As the serological data could not determine the infecting

hantavirus species, a subset of 520 samples were subjected

to the partial L segment nested RT-PCR and of these, 132

were also tested by the partial S segment nested RT-PCR

(data not shown) to allow species identification. All sero-

positive samples were subjected to this test, along with a

partially random subset of other samples. Due to the

unavailability of sera from the first trapping session in Loei,

a large number of the samples from this session were tested

by RT-PCR (159 in total) to compensate for the lack of

serological testing. Of the 520 samples tested, ten were

confirmed positive by both RT-PCRs (although sequences

for the partial S segment were only successfully obtained

from nine samples). The majority of these positive samples

(eight samples) were obtained from B. indica from the Nan

site in Thailand (sample IDs R5320, R5370, R6083, R6098,

R6107, R6108, R6109, and R6153). One sample was from a

R. tanezumi captured in Luang Prabang, Lao PDR (L0199),

and one was from a R. norvegicus captured in Phnom Penh,

Cambodia (PP21).

Sequencing and subsequent genetic analysis indicated

that these samples represented at least two species of han-

tavirus, namely Thailand virus (found at the Nan site,

Thailand) and Seoul virus (in Phnom Penh, Cambodia and

Luang Prabang, Lao PDR). Details of sequences included in

the p-distance and phylogenetic analysis can be found in

the Supplementary Table. Nucleotide sequence compari-

sons (P distances, data not shown) found PP21, from

Phnom Penh, Cambodia, to be closely related to several

strains of Seoul virus (up to 99 and 94% sequence

homology for S and L segments, respectively). The isolates

from Nan, Thailand, showed high sequence homology for

the S segment with previous isolates of Thailand virus. This

is the first known study to include L segment sequences of

Thailand virus. The virus L0199 from Luang Prabang, Lao

PDR, appeared to be a highly divergent variant of Seoul

virus. The S segment of this virus showed highest nucleo-

tide homology to a previously isolated Cambodian strain of

Seoul virus (AJ427506, GenBank) and to PP21, which was

also isolated in Cambodia (945 nucleotides: 85%). How-

ever, for the partial L segment (324 nucleotides) L0199 was

found most closely related (86% nucleotide identity) to two

viruses from China, CGRn8316 (GenBank accession

number EF990931) and CGRn9415 (GenBank accession

number EF990930), isolated from R. norvegicus and

believed to be natural recombinants between Seoul and

Hantaan viruses (Zou et al. 2008).

Phylogenetic analyses supported the genetic analysis

for all viruses. The divergent Laos virus grouped distantly

with other Seoul virus isolates at the nucleotide level for

both S and L segments and more closely for the partial

nucleoprotein sequence. S segment analysis indicated that

this Laotian isolate was the most basal of the Seoul samples

(Fig. 3), while L segment analyses showed L0199 forming a

sister taxa with the two suspected Seoul–Hantaan

recombinant viruses (Fig. 4). In contrast, when predicted

amino acid identities were compared, strain L0199 grouped

closely with other isolates of Seoul and was found to share

approximately 97 and 100% homology for L protein and

nucleoprotein, respectively.

DISCUSSION

The findings from this study increase the number of areas in

SE Asia where hantaviruses are known to circulate. Of the

sites studied, anti-hantavirus antibodies and hantaviral RNA

have previously only been detected at Buriram (Nitatpattana

et al. 2002; Pattamadilok et al. 2006) so this is the first evi-

dence of the presence of hantavirus at Luang Prabang,

Champasak, Mondulkiri, Preah Sihanouk, and Nan and the

first molecular detection of hantaviruses in Lao PDR.

This study supports previous findings that many SE

Asian rodent species can be infected with hantaviruses

(Ibrahim et al. 1996; Nitatpattana et al. 2000, 2002), with

nine species found seropositive. The antibodies detected in

some of the rodent species here may represent spillover

infections of known hantavirus species into non-reservoir

436 Kim Blasdell et al.

Tab

le1.

Ro

den

td

ata

and

han

tavi

rus

sero

pre

vale

nce

for

the

seve

ntr

app

ing

loca

tio

ns

Lu

ang

Pra

ban

gN

anL

oie

Ch

amp

asak

Bu

rira

mP

reah

Sih

ano

uk

Mo

nd

olk

iri

Ph

no

mP

enh

To

tal

(sp

ecie

s)

Ban

dic

ota

ind

ica

6/6/

2001

105/

103/

1116

/1/0

1/1/

2000

3/3/

2000

3/3/

2000

134/

117/

12

(10.

7%)

(10.

3%)

Ban

dic

ota

savi

lei

25/3

/017

/15/

122

/22/

083

/82/

41/

1/20

0014

8/12

3/5

(6.7

%)

(4.8

%)

(4.1

%)

Ber

ylm

ysbe

rdm

orei

1/1/

2000

14/1

3/0

13/4

/06/

4/20

009/

9/20

004/

4/20

0047

/35/

0

(0.0

%)

(0.0

%)

Ber

ylm

ysbo

wer

si3/

2/20

003/

3/20

0023

/1/0

29/6

/0

(0.0

%)

Leo

pold

amys

edw

ard

si1/

1/20

0012

/0/0

2/2/

2000

15/3

/0

(0.0

%)

Max

omys

suri

fer

29/6

/03/

3/20

0070

/66/

235

/35/

013

7/11

0/2

(3.0

%)

(0.0

%)

(1.8

%)

Niv

iven

ter

fulv

esce

ns

77/1

0/0

1/1/

2000

2/1/

2000

14/1

1/0

94/2

3/0

(0.0

%)

(0.0

%)

(0.0

%)

Rat

tus

and

aman

ensi

s5/

3/20

005/

3/20

00

(0.0

%)

Rat

tus

arge

nti

ven

ter

2/2/

2000

48/4

6/0

50/4

8/0

(0.0

%)

(0.0

%)

Rat

tus

exu

lan

s75

/52/

282

/30/

011

9/11

5/2

74/7

1/2

116/

98/9

64/6

3/1

9/4/

2000

539/

433/

16

(3.8

%)

(0.0

%)

(1.7

0%)

(2.8

%)

(9.2

%)

(1.6

%)

(3.7

%)

Rat

tus

lose

a13

8/39

/010

/10/

02/

2/20

0015

0/51

/0

(0.0

%)

(0.0

%)

(0.0

%)

Rat

tus

nit

idu

s16

/15/

21/

0/0

17/1

5/2

(13.

3%)

(13.

3%)

Rat

tus

nor

vegi

cus

28/2

6/2

8/6/

2000

36/3

2/2

(7.7

%)

(6.3

%)

Rat

tus

tan

ezu

mi

122/

97/4

39/3

7/1

40/5

/013

/13/

035

/33/

087

/87/

346

/46/

038

2/31

8/8

(4.1

%)

(2.7

%)

(0.0

%)

(0.0

%)

(3.4

%)

(0.0

%)

(2.5

%)

Mu

sca

roli

35/3

2/2

12/8

/049

/16/

01/

1/20

0039

/38/

01/

1/20

0013

7/96

/2

(6.3

%)

(0.0

%)

(0.0

%)

(2.1

%)

Mu

sce

rvic

olor

7/5/

2000

67/3

6/0

38/3

3/0

112/

74/0

(0.0

%)

(0.0

%)

(0.0

%)

Mu

sco

okii

73/7

2/1

33/2

0/0

113/

38/0

1/1/

2000

220/

131/

1

(1.4

%)

(0.0

%)

(0.0

%)

(0.8

%)

Hantaviruses in Southeast Asian Rodents 437

host species, or may reflect exposure to a yet unidentified

hantavirus species or lineage (as suggested by the molecular

evidence for Seoul-like viruses in at least two Rattus species

(Wang et al. 2000). Also, the overall seroprevalence in

R. exulans (3.7%) was actually higher than for R. tanezumi

(2.5%), which is a known hantavirus host. The presence of

antibodies in R. exulans at several sites may also indicate

that this species potentially supports an as yet unidentified

hantavirus species or lineage.

The overall prevalence of antibodies detected at our

sites is comparable to findings from previous studies in

Thailand with values of between 1.1 and 4.0% (Nitatpat-

tana et al. 2002; Kantakamalakul et al. 2003; Pattamadilok

et al. 2006). The prevalences found for the different rodent

species were also similar to previous findings in the region

(Ibrahim et al. 1996; Nitatpattana et al. 2000; Reynes et al.

2003). A seroprevalence of 8.2% was detected in rodents

trapped in and around Phnom Penh in Cambodia in 1998

(Reynes et al. 2003). Higher prevalences (up to 29% at one

site) were also detected around certain ports in northern

Vietnam (Truong et al. 2009). This may be a result of the

rodent species present in these communities, as the two

species captured most frequently, R. norvegicus and R. rat-

tus (latter probably R. tanezumi as R. rattus is not known to

occur in the region (Pages et al. 2010)) are both known to

host hantavirus species.

The variation in hantavirus seroprevalence between

and within sites is probably due to a combination of fac-

tors. Seasonal differences and abiotic factors are known to

influence the survival of hantaviruses in the environmentTab

le1.

con

tin

ued

Lu

ang

Pra

ban

gN

anL

oie

Ch

amp

asak

Bu

rira

mP

reah

Sih

ano

uk

Mo

nd

olk

iri

Ph

no

mP

enh

To

tal

(sp

ecie

s)

Mu

sfr

agil

icau

da

1/1/

2000

1/1/

2000

(0.0

%)

Oth

ersp

ecie

s(n

ot

incl

ud

edin

div

ersi

tyan

alys

is)

7/5/

2000

10/9

/022

/6/0

01/

0/0

60/0

/01/

1/20

0010

/0/0

111/

21/0

(0.0

%)

To

tal

(sit

e)26

9/23

4/10

298/

250/

1470

8/19

6/0

171/

163/

321

7/20

5/2

420/

333/

1625

3/24

8/5

28/1

1/0

2364

/164

0/50

(4.3

%)

(5.6

%)

(0.0

%)

(1.8

%)

(1.0

%)

(4.8

%)

(2.0

%)

(0.0

%)

(3.0

%)

N26

228

868

617

121

636

025

2N

/A

Sob

s9

815

99

79

N/A

Sk8.

257.

9313

.43

8.98

8.69

6.72

8.53

N/A

Num

ber

sar

eth

enum

ber

ofca

ptu

res/

num

ber

ofsa

mple

ste

sted

by

IFA

/num

ber

ofpositives

and

the

han

tavir

usse

ropre

val

ence

(giv

enin

par

enth

eses

).H

anta

vir

usse

ropre

val

ence

isonly

giv

enfo

rsp

ecie

sw

ith

ten

or

more

indiv

idual

ste

sted

sero

logic

ally

per

site

.So

bs

isth

en

um

ber

of

ob

serv

edsp

ecie

sin

the

cap

ture

dat

afo

rea

chst

ud

ysi

tean

dSk

isth

en

um

ber

of

spec

ies

corr

ecte

dfo

rsa

mp

lesi

ze.

Figure 2. Relationship between rodent community characteristics

(X-axis) and seroprevalence (Y-axis). Hantavirus prevalence versus

observed number of species (Sobs) or expected number of species

(Sk).

438 Kim Blasdell et al.

itself (Kallio et al. 2006; Linard et al. 2007; Dearing et al.

2009) and the difference in primary host density (and

population and community demographics) will also have

an effect (Calisher et al. 2007; Suzan et al. 2009). Previous

studies have also identified a trend for hantavirus preva-

lence to increase with decreasing species diversity (Mills

2006; Clay et al. 2009; Dizney and Reudas 2009; Suzan et al.

2009). It has been postulated that, because in many cases

the rodent species remaining in communities of low species

richness are those that are the reservoir hosts of hantavi-

ruses, this would allow for hantaviruses to circulate at a

higher prevalence in these communities (Mills 2006). It has

been suggested that this negative correlation may in part be

due to the dilution effect (Ostfeld and Keesing 2000), where

increasing species diversity is predicted to decrease patho-

gen prevalence through several mechanisms including

reduced encounters between hosts and decreased host

density (Keesing et al. 2006). Due to limitations in our data

set, we were unable to explore this finding further.

In this study, hantaviruses were found to infect animals

living in many different habitats, and particularly species

associated with human habitats (settlements and agricul-

ture). Similar findings have been observed for hantaviruses

circulating in South America (Mills 2006; Armien et al.

2009). As these areas also probably pose the greatest

potential for human transmission it is here that transmis-

sion prevention needs to be implemented in priority.

Of the three murine species found positive for hanta-

virus RNA in this study, two were the presumed natural

hosts of the viruses detected [Thailand virus in B. indica

(Xiao et al. 1994) and Seoul virus in R. norvegicus (Lee et al.

1982)]. However, Seoul virus was also identified in

R. tanezumi, a species previously associated with Serang

virus (Plyusnina et al. 2009). At least one isolate of Seoul

virus (Gou3 strain) has been collected from a rodent of the

R. rattus species complex previously (Wang et al. 2000) and

this isolate was also found to be genetically divergent from

other Seoul virus strains. It is conceivable that R. tanezumi

is the reservoir host for a distinct lineage of Seoul virus in

this area of Laos, as no R. norvegicus were encountered

during extensive trapping in this region. The divergent

nature of the nucleotide sequence of this sample may reflect

adaptation to R. tanezumi, as may also be the case for Gou3

virus to R. rattus.

Seoul virus regularly causes human infections, pri-

marily in east and SE Asia (Wong et al. 1989; Plyusnina

Figure 3. Maximum likelihood

tree for the partial S segment of

murine hantaviruses, showing the

position of sequences identified in

this study. Bootstrap values for the

major clades are indicated.

Hantaviruses in Southeast Asian Rodents 439

et al. 2004; Kariwa et al. 2007) and sporadically elsewhere

(Cueto et al. 2008; Heyman et al. 2009). There is also

serological data suggesting human infection by Thailand

virus (Pattamadilok et al. 2006) although this virus has

never been isolated from a patient. However, all three

rodent host species found positive for hantavirus RNA in

this study are associated with human habitats, indicating

the potential for zoonotic transmission. Rattus norvegicus is

primarily associated with human urban habitation and

both R. tanezumi and B. indica are found in a variety of

habitats including agricultural land and around (and in the

case of R. tanezumi, in) human habitation in rural areas

(Aplin et al. 2003).

This association with human environments has been

demonstrated in at least one case to increase the reproduc-

tive capacity of rodent species by increasing the duration of

the breeding season (Kuenzi et al. 2001). Most known out-

breaks of hantavirus disease have occurred in anthropo-

logically disturbed habitats, where natural biodiversity has

been reduced to low levels (Mills 2006). As with elsewhere,

most of the rodent species that host hantaviruses in SE Asia

are highly adaptable to many environments and they often

constitute one of only two or three rodent species present

within certain urban sites. These species can often reach high

densities due to the lack of competitive pressure. The low

rodent species richness (as a measure of biodiversity) asso-

ciated with high seroprevalence observed in our study sup-

ports previous findings from Vietnam and Cambodia

(Reynes et al. 2003; Truong et al. 2009). This finding may

have serious implications for SE Asian habitats, as many are

undergoing modification and no doubt, subsequent reduc-

tions in their natural species richness.

CONCLUSION

Nine murine rodent species were found to have been in-

fected with hantavirus in this study, with molecular analysis

confirming the presence of at least two species of hanta-

virus. This study presents the first molecular evidence for

hantaviruses in Lao PDR and indicates the presence of a

divergent strain of Seoul virus circulating in this country.

This study also presents the first L segment sequence data

for Thailand hantavirus. Our findings also suggest that

hantavirus seroprevalence is related to rodent species

richness, which is a potentially important finding in a

Figure 4. Maximum likelihood

tree for the partial L segment of

murine hantaviruses, showing the

position of sequences identified in

this study. Bootstrap values for the

major clades are indicated.

440 Kim Blasdell et al.

region where species richness is changing rapidly in many

areas.

ACKNOWLEDGMENTS

The authors wish to thank Mr. Hul Vibol and Mr. Y

Bunthin for their technical assistance in both the laboratory

and the field and to Mr. Kim Aun for his assistance in the

field. Also to Maria Walhstrom for kindly preparing the IFA

slides used in this study. The CERoPath project (Commu-

nity ecology of rodents and their pathogens in South East

Asia: effects of biodiversity changes and implications in

health ecology/ANR 07 BDIV 012) is funded by the Agence

Nationale de la Recherche (ANR, France).

REFERENCES

Aplin KP, Brown PR, Jacob J, Krebs CJ, Singleton GR (2003) FieldMethods for Rodent Studies in Asia and the Indo-Pacific, Mel-bourne: ACIAR

Armien AG, Armien B, Koster F, Pascale JM, Avila M, Gonzalez P,et al. (2009) Hantavirus infection and habitat associationsamong rodent populations in agroecosystems of Panama:implications for human disease risk. American Journal ofTropical Medicine and Hygiene 81:59–66

Bi Z, Formenty PBH, Roth CE (2008) Hantavirus infection: areview and global update. Journal of Infection in DevelopingCountries 2:3–21

Calisher CH, Wagoner KD, Amman BR, Root JJ, Douglass RJ,Kuenzi AJ, et al. (2007) Demographic factors associated withprevalence of antibody to Sin Nombre virus in deer mice in thewestern United States. Journal of Wildlife Disease 43:1–11

Carey DE, Reuben R, Panicker KN, Shope RE, Myers RM (1971)Thottapalayam virus: a presumptive arbovirus isolated from ashrew in India. Indian Journal of Medical Research 59:1758–1760

Chaval Y, Dobigny G, Michaux J, Pages M, Corbisier C, CossonJF, et al. (2010) A multi-approach survey as the most reliabletool to accurately assess biodiversity: an example of Thai murinerodents. Kasetsart Journal of Natural Science 44:590–603

Clay CA, Lehmer EM, Jeor SS, Dearing MD (2009) Sin nombrevirus and rodent species diversity: a test of the dilution andamplification hypotheses. Public Library of Science One 4:e6467.doi:10.1371/journal.pone.0006467

Cueto GR, Cavia R, Bellomo C, Padula PJ, Suarez OV (2008)Prevalence of hantavirus infection in wild Rattus norvegicus andR. rattus populations of Buenos Aires City, Argentina. TropicalMedicine and International Health 13:46–51. doi:10.1111/j.1365-3156.2007.01968.x

Dearing MD, Dizney L (2010) Ecology of hantaviruses in achanging world. Annals of the New York Academy of Sciences1195:99–112

Dearing MD, Previtali MA, Jones JD, Ely PW, Wood BA (2009)Seasonal variation in Sin Nombre virus infections in deer mice:preliminary results. Journal of Wildlife Diseases 45:430–436

Dizney LJ, Reudas LA (2009) Increased host species diversity anddecreased prevalence of Sin Nombre virus. Emerging InfectiousDiseases 15:1012–1018

Gouy M, Guindon S, Gascuel O (2010) SeaView version 4: amultiplatform graphical user interface for sequence alignmentand phylogenetic tree building. Molecular Biology and Evolution27:221–224

Guindon S, Gascuel O (2003) A simple, fast, and accurate algo-rithm to estimate large phylogenies by maximum likelihood.Systems Biology 52:696–704

Hardy OJ (2009) BiodivR 1.1. A program to compute indices ofspecies diversity within sample and species similarity betweensamples using rarefaction principles to reduce sampling bias.Available via http://ebe.ulb.ac.be/ebe/Software.html

Henttonen H, Buchy P, Suputtamongkol Y, Jittapalapong S,Herbreteau V, Laakkonen J, et al. (2008) Recent discoveries ofnew hantaviruses widen their range and question their origins.Animal Biodiversity and Emerging Diseases 1149:84–89

Heyman P, Baert K, Plyusnina A, Cochez C, Lundkvist AEsbroeckMV, et al. (2009) Serological and genetic evidence for thepresence of Seoul hantavirus in Rattus norvegicus in Flanders,Belgium. Scandinavian Journal of Infectious Diseases 41:51–56

Hugot JP, Plyusnina A, Herbreteau V, Nemirov K, Laakkonen J,Lundkvist A, et al. (2006) Genetic analysis of Thailand hanta-virus in Bandicota indica trapped in Thailand. Virology Journal3:72–81. doi:10.1186/1743-422X-3-72

Hujakka H, Koistinen V, Kuronen I, Eerikainen P, Parviainen M,Lundkvist A, et al. (2003) Diagnostic rapid tests for acutehantavirus infections: specific tests for Hantaan, Dobrava andPuumala viruses versus a hantavirus combination test. Journal ofVirological Methods 108:117–122

Ibrahim IN, Sudomo M, Morita C, Uemura S, Muramatsu Y,Ueno H, et al. (1996) Seroepidemiological survey of wild rats forSeoul virus in Indonesia. Japanese Journal of Medical Science andBiology 49:69–74

Jiang JF, Zuo SQ, Zhang WY, Wu XM, Tang F, De Vlas SJ, et al.(2008) Prevalence and genetic diversities of hantaviruses inrodents in Beijing, China. American Journal of Tropical Medicineand Hygiene 78:98–105

Johansson P, Yap G, Low HT, Siew CC, Kek R, Ng LC, et al.(2010) Molecular characterization of two hantavirus strainsfrom different Rattus species in Singapore. Virology Journal7:15–24. doi:10.1186/1743-422X-7-15

Kallio ER, Klingstrom J, Gustafsson E, Manni T, Vaheri A,Henttonen H, et al. (2006) Prolonged survival of Puumalahantavirus outside the host: evidence for indirect transmissionvia the environment. Journal of General Virology 87:2127–2134.doi:10.1099/vir.0.81643-0

Kang HJ, Bennett SN, Dizney L, Sumibcay L, Arai S, Ruedas LA,et al. (2009) Host switch during evolution of a genetically dis-tinct hantavirus in the American shrew mole (Neurotrichusgibbsii). Virology 388:8–14 . doi:10.1016/j.virol.2009.03.019

Kantakamalakul W, Siritantikorn S, Thongcharoen P, Singchai C,Puthavathana P (2003) Prevalence of rabies virus and Hantaanvirus infections in commensal rodents and shrews trapped inBangkok. Journal of the Medical Association of Thailand 86:1008–1014

Kariwa H, Yoshimatsu K, Arikawa J (2007) Hantavirus infectionin East Asia. Comparative Immunology, Microbiology and Infec-tious Diseases 30:341–356

Keesing F, Holt RD, Ostfeld RS (2006) Effects of species diversityon disease risk. Ecology Letters 9:485–498

Hantaviruses in Southeast Asian Rodents 441

Khaiboullina SF, Morzunov SP, St Jeor SC (2005) Hantaviruses:molecular biology, evolution and pathogenesis. Current Molec-ular Medicine 5:773–790

Klempa B, Fichet-Calvet E, Lecompte E, Auste B, Aniskin V,Meisel H, et al. (2006) Hantavirus in African wood mouse,Guinea. Emerging Infectious Diseases 12:838–840

Klempa B, Meisel H, Rath S, Bartel J, Ulrich R, Kruger DH (2003)Occurrence of renal and pulmonary syndrome in a region ofnortheast Germany where Tula hantavirus circulates. Journal ofClinical Microbiology 41:4894–4897

Kuenzi AJ, Douglass RJ, White D Jr, Bond CW, Mills JN (2001)Antibody to sin nombre virus in rodents associated with peri-domestic habitats in west central Montana. American Journal ofTropical Medicine and Hygiene 64:137–146

Lee HW, Baek LJ, Johnson KM (1982) Isolation of Hantaan virus,the etiological agent of Korean hemorrhagic fever from wildurban rats. Journal of Infectious Diseases 146:638–644

Lee HW, Lee PW, Johnson KM (1978) Isolation of the etiologicalagent of Korean hemorrhagic fever. Journal of Infectious Diseases137:298–308

Li L, Yang DJ, Chen JY, Ding JQ, Su X, Tian YQ, Wei JF (2005)Cloning and sequencing of S segment of hantavirus strain TJJ16isolated in Tianjin. Chinese Journal of Microbiology and Immu-nology 25:652–655 (In Chinese)

Linard C, Tersago K, Leirs H, Lambin EF (2007) Environmentalconditions and Puumala virus transmission in Belgium. Inter-national Journal of Health Geographics 6:55–66 . doi:10.1186/1476-072X-6-55

Marshall JT (1988) Family Muridae. In: Mammals of Thailand,Lekagul B, McNeely JA (editors), Bangkok: Darnsutha Press, pp397–487

Mills JN (2006) Biodiversity loss and emerging infectious disease:an example from the rodent-borne hemorrhagic fevers. Biodi-versity 7:9–17

Nitatpattana N, Henrich T, Palabodeewat S, Tangkanakul W,Poonsuksombat D, Chauvancy G, et al. (2002) Hantaan virusantibody prevalence in rodent populations of several provincesof northeastern Thailand. Tropical Medicine and InternationalHealth 7:840–845

Nitatpattana N, Chauvancy G, Dardaine J, Poblap T, JumronsawatK, Tangkanakul W, et al. (2000) Serological study of hantavirusin the rodent population of Nakhon Pathom and NakhonRatchasima Provinces Thailand. Southeast Asian Journal ofTropical Medicine and Public Health 31:277–282

Okumura M, Yoshimatsu K, Kumperasart S, Nakamura I, OginoM, Taruishi M, et al. (2007) Development of serological assaysfor Thottapalayam virus, an insectivore-borne Hantavirus.Clinical Vaccine Immunology 14:173–181 . doi:10.1128/CVI.00347-06

Ostfeld RS, Keesing F (2000) Biodiversity and disease risk: the caseof Lyme disease. Conservation Biology 14:722–728

Pages M, Chaval Y, Herbreteau V, Waengsothorn S, Cosson JF,Hugot JP, et al. (2010) Revisiting the taxonomy of the Rattinitribe: a phylogeny-based delimitation of species boundaries.BioMed Central Evolutionary Biology 10:184–210. doi:10.1186/1471-2148-10-184

Papa A, Johnson AM, Stockton PC, Bowen MD, Spiropoulou CF,Alexiou-Daniel S, et al. (1998) Retrospective serological andgenetic study of the distribution of hantaviruses in Greece.Journal of Medical Virology 55:321–327

Pattamadilok S, Lee BH, Kumperasart S, Yoshimatsu K, OkumuraM, Nakamura I, et al. (2006) Geographical distribution of

hantaviruses in Thailand and potential human health signifi-cance of Thailand virus. American Journal of Tropical Medicineand Hygiene 75:994–1002

Plyusnina A, Ibrahim IN, Plyusnin A (2009) A newly recognizedhantavirus in the Asian house rat (Rattus tanezumi) in Indo-nesia. Journal of General Virology 90:205–209. doi:10.1099/vir.0.006155-0

Plyusnina A, Ibrahim IN, Winoto I, Porter KR, Gotama IB,Lundkvist A, et al. (2004) Identification of Seoul hantavirus inRattus norvegicus in Indonesia. Scandinavian Journal of Infec-tious Diseases 36:356–359

Posada D (2008) jModelTest: phylogenetic model averaging.Molecular Biology and Evolution 25:1253–1256

Ramsden C, Holmes EC, Charleston MA (2009) Hantavirusevolution in relation to its rodent and insectivore hosts: noevidence for codivergence. Molecular Biology and Evolution26:143–153. doi:10.1093/molbev/msn234

Reynes JM, Soares JL, Hue T, Bouloy M, Sun S, Kruy SL, et al.(2003) Evidence of the presence of Seoul virus in Cambodia.Microbes and Infection 5:769–773

Schilling S, Emmerich P, Klempa B, Auste B, Schnaith E, SchmitzH, et al. (2007) Hantavirus disease outbreak in Germany: lim-itations of routine serological diagnostics and clustering of virussequences of human and rodent origin. Journal of ClinicalMicrobiology 45:3008–3014 . doi:10.1128/JCM.02573-06

Schlegel M, Klempa B, Auste B, Bemmann M, Schmidt-Chanasit J,Buchner T, et al. (2009) Dobrava-Belgrade virus spilloverinfections, Germany. Emerging Infectious Diseases 15:2017–2020

Schmidt-Chanasit J, Essbauer S, Petraityte R, Yoshimatsu K,Tackmann K, Conraths FJ, et al. (2010) Extensive host sharingof central European Tula virus. Journal of Virology 84:459–474.doi:10.1128/JVI.01226-09

Schultze D, Lundkvist A, Blauenstein U, Heyman P (2002) Tulavirus infection associated with fever and exanthema after a wildrodent bite. European Journal of Clinical Microbiology andInfectious Disease 21:304–306

Song JW, Kang HJ, Song KJ, Truong TT, Bennett SN, Arai S, et al.(2007) Newfound hantavirus in Chinese mole shrew, Vietnam.Emerging Infectious Diseases 13:1784–1787

Song G, Hang CS, Liao HX, Fu JL, Gao GZ, Qiu HL, et al. (1984)Antigenic difference between viral strains causing classical andmild types of epidemic hemorrhagic fever with renal syndromein China. Journal of Infectious Diseases 150:889–894

Suputthamongkol Y, Nitatpattana N, Chayakulkeeree M, Palab-odeewat S, Yoksan S, Gonzalez JP (2005) Hantavirus infectionin Thailand: first clinical case report. Southeast Asian Journal ofTropical Medicine and Public Health 36:700–703

Suzan G, Marce E, Giermakowski JT, Mills JN, Ceballos G, OstfeldRS, et al. (2009) Experimental evidence for reduced rodentdiversity causing increased hantavirus prevalence. Public Libraryof Science One 4:e5461. doi:10.1371/journal.pone.0005461

Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: molecularevolutionary genetic analysis (MEGA) software version 4.0.Molecular Biology and Evolution 24:1596–1599

Terajima M, Vapalahti O, Van Epps HL, Vaheri A, Ennis FA(2004) Immune responses to Puumala virus infection and thepathogenesis of nephropathia epidemica. Microbes and Infection6:238–245

Truong TT, Yoshimatsu K, Araki K, Lee BH, Nakamura I, Endo R,et al. (2009) Molecular epidemiological and serological studiesof hantavirus infection in northern Vietnam. Journal of Veteri-nary Medical Science 71:1357–1363

442 Kim Blasdell et al.

Vapalahti O, Lundkvist A, Fedorov V, Conroy CJ, Hirvonen S,Plyusnina A, et al. (1999) Isolation and characterization of ahantavirus from Lemmus sibiricus: evidence for host switchduring hantavirus evolution. Journal of Virology 73:5586–5592

Wang H, Yoshimatsu K, Ebihara H, Ogino M, Araki K, Kariwa H,et al. (2000) Genetic diversities of hantaviruses isolated in Chinaand characterization of novel hantaviruses isolated from Nivi-venter fulvescens and Rattus rattus. Virology 278:332–345

Wong TW, Chan YC, Joo YG, Lee HW, Lee PW, Yanagihara R(1989) Hantavirus infections in humans and commensal ro-dents in Singapore. Transactions of the Royal Society of TropicalMedicine and Hygiene 83:248–251

Xiao SY, Leduc JW, Chu YK, Schmaljohn CS (1994) Phylogeneticanalyses of virus isolates in the genus Hantavirus, family Bun-yaviridae. Virology 198:205–217

Yao ZH, Dong GM, Yu YX, Zhang JK, Yan DY, Liu XC, Zhang LL(2002) A new subtype of hantavirus SN7 isolated from Nivi-venter confucianus in Sichuan province, China. Chinese Journalof Infectious Diseases 20:79–82 (in Chinese)

Zou Y, Hu J, Wang ZX, Wang DM, Yu C, Zhou JZ, et al. (2008)Genetic characterization of hantaviruses isolated from Guizhou,China: evidence for spillover and reassortment in nature. Jour-nal of Medical Virology 80:1033–1041. doi:10.1002/jmv.21149

Hantaviruses in Southeast Asian Rodents 443