Rodent-Borne Orthohantaviruses in Vietnam, Madagascar and Japan

Abstract

Hantaviruses are harbored by multiple small mammal species in Asia, Europe, Africa, and the Americas. To ascertain the geographic distribution and virus-host relationships of rodent-borne hantaviruses in Japan, Vietnam, Myanmar, and Madagascar, RNAlater™-preserved lung tissues of 981 rodents representing 40 species, collected in 2011–2017, were analyzed for hantavirus RNA by RT-PCR. Our data showed Hantaan orthohantavirus Da Bie Shan strain in the Chinese white-bellied rat (Niviventer confucianus) in Vietnam, Thailand; orthohantavirus Anjo strain in the black rat (Rattus rattus) in Madagascar; and Puumala orthohantavirus Hokkaido strain in the grey-sided vole (Myodes rufocanus) in Japan. The Hokkaido strain of Puumala virus was also detected in the large Japanese field mouse (Apodemus speciosus) and small Japanese field mouse (Apodemus argenteus), with evidence of host-switching as determined by co-phylogeny mapping.

1. Introduction

Formerly classified in the genus Hantavirus of the family Bunyaviridae, hantaviruses have been reclassified recently into four genera (Orthohantavirus, Loanvirus, Mobatvirus, and Thottimvirus) of the family Hantaviridae [1]. Hantaviruses possess a tripartite, negative-sense, single-stranded RNA genome comprising S, M, and L segments that encode a nucleocapsid protein, Gn and Gc envelope glycoproteins, and an RNA-dependent RNA polymerase, respectively [2,3]. Rodents (order Rodentia) have long been known to serve as reservoir hosts of hantaviruses [4]. Recently, the geographic landscape of hantaviruses has been disrupted by the identification of genetically divergent hantaviruses in multiple species of shrews and moles (order Eulipothyla) and bats (order Chiroptera) in Europe, Asia, Africa, and the Americas [5].

Several rodent-borne orthohantaviruses, notably Hantaan virus (HTNV), Seoul virus (SEOV), Puumala virus (PUUV), and Dobrava virus (DOBV), cause hemorrhagic fever with renal syndrome (HFRS) in Europe and Asia [4]. The seminal discovery of HTNV in the striped field mouse (Apodemus agrarius) in Korea [6] made possible subsequent discoveries of other rodent-borne hantaviruses. For example, PUUV was originally identified in the bank vole (Myodes glareolus), collected in the Puumala region around Lake Finland in October 1977 [7]. Genetic variants of PUUV, namely Hokkaido orthohantavirus (PUUV Hokkaido strain) and Muju orthohantavirus (PUUV Muju strain), have been identified in the grey-sided vole (Myodes rufocanus) in Japan [8,9,10] and the Korean red-backed vole or royal vole (Myodes regulus) in South Korea [11,12], respectively.

Another rodent-borne hantavirus, Thailand orthohantavirus (THAIV), initially identified in the greater bandicoot rat (Bandicota indica) in Thailand [13,14], has not been definitively shown to cause HFRS, but seroepidemiological data suggest that THAIV may be pathogenic in humans [15]. Genetic variants of THAIV include Serang virus in the Asian house rat (Rattus tanezumi) in Indonesia [16] and Singapore [17] and Anjozorobe virus in the black rat (Rattus rattus) and Major’s tufted-tailed rat (Eliurus majori) in Madagascar [18]. Instead of the previous belief that each hantavirus species is hosted by a single mammalian host species, a single hantavirus species may be carried by multiple reservoir host species [5]. Mounting evidence suggests that preferential host switching and local adaptation account for the complex evolutionary history of hantaviruses [19,20,21].

The objective of the present study was to determine the geographic distribution and virus-host relationships of rodent-borne hantaviruses in Japan, Vietnam, Myanmar, and Madagascar. Our data showed three distinct rodent-borne orthohantaviruses: Hantaan orthohantavirus Da Bie Shan strain (HTNV DBS) in the Chinese white-bellied rat (Niviventer confucianus) in Vietnam; Thailand orthohantavirus Anjozorobe strain (THAIV ANJO) in the black rat in Madagascar; and Puumala orthohantavirus Hokkaido strain (PUUV HOK) in the grey-sided vole in Japan. As determined by co-phylogeny mapping, PUUV HOK infection in the large Japanese field mouse (Apodemus speciosus) and small Japanese field mouse (Apodemus argenteus) was consistent with host-switching.

2. Materials and Methods

2.1. Trapping and Sample Collection

From 2011 to 2017, small mammals were trapped at various locations in Vietnam, Myanmar, Madagascar, and Japan using Sherman live traps (H. B. Sherman Traps, Inc., Tallahassee, FL, USA). Trapped animals were euthanized according to guidelines of the American Society of Mammalogists [22,23]. Lung tissues were preserved in RNAlater™ RNA Stabilization Reagent (Qiagen Inc., Valencia, CA, USA) and stored at −30 °C for RNA and DNA extraction. To aid in morphological identification, weight and body measurements were recorded. Field investigation procedures and protocols were approved by the Institutional Animal Care and Use Committee of the National Institute of Infectious Diseases (permission numbers: 108074, 111126, 112152, 115162, 118180).

2.2. RNA Extraction and RT-PCR Analysis

Total RNA was extracted from RNAlater™-preserved lung tissues using the MagDEA RNA 100 Kit and Magtration systems 12GC PLUS (Precision System Science, Matsudo, Japan) and then reverse transcribed using PrimeScript II 1st strand cDNA Synthesis Kit (Takara Bio, Otsu, Japan) with oligonucleotide primer (OSM55F, 5′–TAGTAGTAGACTCC−3′), designed from the conserved 5′-ends of the S, M, and L segments of hantaviruses [24,25,26,27,28]. Oligonucleotide primers used to amplify the S-, M-, and L-genomic segments are provided in Supplementary Table S1. PCR cycling condition and reagent concentrations were described previously [29]. PCR products were confirmed by MultiNA with DNA-12000 kit (Shimazu, Kyoto, Japan). PCR products were treated with ExoSAP enzyme (New England Biolab, Ipswich, MA, USA) and sequenced directly using an Applied Biosystems 3730 × l DNA Analyzer (Applied Biosystems, Foster City, CA, USA).

2.3. Host Identification

Genomic DNAs were prepared from RNAlater™-preserved lung tissues using the MagDEA DNA 200 and Magtration systems 12GC PLUS (PSS). DNA samples were amplified by PCR with the primer sets and sequence protocol described previously [24,30,31]. Analysis of cytochrome b (CYTB) and cytochrome c oxidase subunit I (COI) and phylogenetic relationship were conducted by ClustalW [32] in BioEdit [33] and Genetyx Ver 15 (Genetyx Corporation, Shibuya, Tokyo, Japan).

2.4. Genetic and Phylogenetic Analysis

The entire and partial length of coding regions (CDS) of the S-, M-, and L-segment nucleotide and amino acid sequences of rodent-borne hantaviruses were aligned with hantavirus sequences available on GenBank, using ClustalW in BioEdit. Pair-wise comparisons were performed to ascertain the degree of sequence homology.

Phylogenetic trees were generated using the Markov chain Monte Carlo (MCMC) methods MrBayes 3.1.2 [34] under the best-fit general time-reversible model of nucleotide evolution with gamma-distributed rate heterogeneity and invariable sites (GTR + I + Γ) [35]. The best-fit model was selected with jModelTest version 2.1.7 [36] for phylogenetic trees. Two replicate Bayesian Metropolis–Hastings MCMC runs, each consisting of six chains of 10 million generations sampled every 100 generations with a burn-in of 25,000 (25%), resulted in 150,000 trees overall. In addition, to ascertain the co-evolutionary history of hantaviruses and their hosts, phylogenetic trees were reconstructed for co-phylogeny mapping [37] from a virus tree into a host tree [38].

3. Results

3.1. Trapping Surveys

Trapping of small mammals was conducted in Japan, Madagascar, Myanmar, and Vietnam (Figure 1). The results of field surveys are summarized in

Table 1. Summary of collected rodent samples.

Species	Japan							Madagascar		Myanmar				Vietnam			Total
	2011	2012	2013	2014	2015	2016	2017	2014	2015	2012	2013	2014	2015	2012	2013	2014
Apodemus argenteus	43	13	62	18	30		2										168
Apodemus speciosus	43	9	87	7	29	1	13										189
Bandicota bengalensis											18	2	4				24
Bandicota indica												2			10		12
Bandicota savilei											4						4
Berylmys berdmorei															3		3
Berylmys bowersi														1	2	1	4
Dacnomys millardi															1		1
Eothenomys andersoni						2											2
Eothenomys smithi					4												4
Eothenomys sp.																1	1
Leopoldamys sabanus															1		1
Maxomys moi															1		1
Maxomys surifer														1	11		12
Micromys minutus						2											2
Microtus montebelli						1											1
Mus caroli												10			1		11
Mus cookii												1					1
Mus fragilicauda												1					1
Mus lepidoides													1				1
Mus musculus			9	4	25		1		7								46
Mus musculus castaneus	1							1			5		3				10
Mus musculus musculus									3								3
Mus nitidulus												11	3				14
Mus pahari															1		1
Myodes rex	1				1												2
Myodes rufocanus bedfordiae	5	7	27	3	36	3	12										93
Myodes rutilus mikado				1	1		4										6
Niviventer cf. confucianus															2	32	34
Niviventer fulvescens												1				1	2
Niviventer sp.																1	1
Niviventer huang															6		6
Niviventer langbianis															1		1
Rattus andamanensis														13			13
Rattus exulans											3	8	7		22		40
Rattus nitidus															2		2
Rattus norvegicus		28	26	4		2			5								65
Rattus rattus species complex	1	8	2			110	3	1	7			4					136
Rattus tanezumi species complex							44			1	4	5		5	1	1	61
Rattus tiomanicus															2		2
Total	94	65	213	37	126	121	79	2	22	1	34	45	18	20	67	37	981

. A total of 981 mammals representing 40 species, including two species of Apodemus, three species of Bandicota, two species of Berylmys, one species of Dacnomys, three species of Eothenomys, one species of Leopoldamys, two species of Maxomys, one species of Micromys, one species of Microtus, seven species of Mus, three species of Myodes, five species of Niviventer, and seven species or species complex of Rattus.

3.2. Hantavirus Screening and Sequence Analyses

Lung tissues were analyzed for hantavirus RNA using S-, M-, and L-segment primer sets. Sequencing of the PCR products and pair-wise alignment and comparison of the S-, M-, and L-segment nucleotide and amino acid sequences indicated that the newly detected hantaviruses were genetically similar to prototypic hantavirus species. That is, HTNV DBS strain was detected in two Niviventer cf. confucianus from Vietnam, THAIV ANJO strain in one Rattus rattus from Madagascar, and PUUV HOK strain in nine Myodes rufocanus, one Apodemus argenteus, and two Apodemus speciosus from Japan. Hantaviruses, host species, and countries are listed in Supplementary Table S2.

3.3. Phylogenetic Analysis and Host Species Identification

Phylogenetic trees, constructed using the Bayesian method and based on each genomic segment of HTNV strains VN3973M589 and VN4004M620, THAIV strain MDG3887MG9, and PUUV strains UA1818B74, KT3011KTF49, KT3028KTF66, KT3120KTF116, KT3122KTF118, JA4032KTF597, JA4277KTF637, JA5171KTF862, JA5274KTF935, JA5277KTF945, JA6551KTF153, and JA6557KTF159 showed that each of the hantaviruses clustered with the prototypic hantavirus species (Figure 2 and Figure 3).

Molecular confirmation of the rodent host species was based on the sequence analyses of the CYTB and COI genes of mitochondrial DNA. One Niviventer sp. was confirmed as Niviventer species using morphological and genetic analysis, but it was not clearly N. confucianus based on morphology. Therefore, we have decided to tentatively classify it as N. cf. confucianus. PUUV Hokkaido strains were confirmed as the grey-sided vole (My. r. bedfordiae). The subspecies, My. r. bedfordiae, is a geographically different species from My. r. rufocanus. My. r. bedfordiae is distributed only on Hokkaido Island and small areas in Sakhalinskaya and Far East Russia. THAIV ANJO strain was detected in R. rattus species complex in Madagascar. THAIV ANJO-positive rodent was captured in the Tsimbazaza Zoo in Antananarivo, Madagascar.

The analysis of host-virus relationship using co-phylogeny mapping showed congruent topologies for nearly all hantaviruses and their reservoir host species (Figure 4). For example, the phylogenetic positions of PUUV strains UA1818B74, JA6557KTD158, and JA5274KTF953, based on the NP, GP, and LP, segregated with other PUUV strains and Myodes host species. On the other hand, the reservoir host species phylogenies of the small Japanese field mouse (A. argenteus) and large Japanese field mouse (A. speciosus) showed evidence of host switching. That is, PUUV UA1818B74, PUUV JA6557KTF159, and PUUV JA5274KTF935 had switched from its natural rodent reservoir host and become well established in a rodent host of a different genus. THAIV ANJO strain in Madagascar was also detected in the Major’s tufted-tailed rat (E. majori), a different host species from R. rattus.

4. Discussion

The present study demonstrated three orthohantavirus species: one each in Japan, Vietnam, and Madagascar. PUUV was detected in My. rufocanus, A. specious, and A. argenteus in Hokkaido, Japan. On the other hand, three Myodes species (My. rufocanus, M. rex, and M. rutilus) are distributed in Hokkaido, Japan; we were able to detect hantavirus RNA only in My. rufocanus. Several research groups have conducted seroepidemiological and/or molecular epidemiological surveys nationwide in Japan thus far. PUUV is distributed in only Hokkaido, Japan [39,40]. On the other hand, A. speciosus and A. argenteus are distributed nationwide in Japan. However, PUUV has not been detected in either Apodemus species elsewhere in Japan. Collectively, the data suggest that My. rufocanus is the principal reservoir host species of PUUV in Japan and that spill-over events have occurred in A. speciosus and A. argenteus.

We detected HTNV RNA in N. cf. confucianus in Vietnam. N. confucianus is widely distributed from the Indo-China Peninsula and adjacent mountains near southeastern Qinghai-Tibetan Plateau to Loess Plateau and northern China [41]. Recent genetic analysis of Niviventer species suggests that there are 16 species and one possible species complex. In addition, a morphologically similar species, Chiromyscus, is proposed to belong to Niviventer. The Niviventer-Chiromyscus complex is still unclear [42]. Our HTNV sequences, strains VN3973M589 and VN4004M620, were closely related to HTNV Da Bie Shan (DBS) strain that was detected in N. confucianus in Yunnan province, China [43]. The CYTB sequence of the host species of HTNV DBS strain was not reported in that study, but the results suggested that our newly detected strains (VN3973M589 and VN4004M620) and DBS were all harbored by N. confucianus.

In this research, we detected THAIV RNA in R. rattus captured in the Botanical and Zoological Garden of Tsimbazaza in Antananarivo, Madagascar. The same genotype of THAIV has been identified thus far [18,44]. Prototype THAIV was initially identified in B. indica in Thailand. Interestingly, the origin of Malagasy is constructed by Southeast Asia and East Africa based on Y-chromosomal sequences [45]. On the other hand, the house mouse, Mus musculus, in Madagascar is similar to the Yemen linage, suggesting that the Madagascar lineage was introduced from Yemen in Arabian Peninsula [46]. The precise origin of R. rattus in Madagascar is still unclear, but it may be associated with the peopling of Madagascar from Southeast Asia and Indonesia [47].

Co-circulation or host-sharing of rodent-borne hantaviruses has been reported for PUUV in bank voles (My. glareolus) [7], grey-sided voles (My. rufocanus) [8,9,10], and royal voles (My. regulus) [11,12] and for DOB/BGDV in yellow-necked field mice (A. flavicollis) [48], striped field mice (A. agrarius) [49], and Black Sea field mice (A. ponticus) [50]. Similar findings have been reported for non-rodent-borne hantaviruses, such as Seewis virus (SWSV) in the Eurasian shrew (Sorex araneus) [26], Laxmann’s shrew (Sorex caecutiens) [24], tundra shrew (Sorex tundrensis) [51], and Eurasian pygmy shrew (Sorex minutus) [52]. The present survey, showing the detection of PUUV infection in A. speciosus and A. argenteus, and a recent investigation of THAIV in E. majori [18] represent examples of spill-over or host-switching events. THAIV infection in R. rattus may also represent host switching from B. indica. Further investigations are warranted to better clarify the evolutionary origins and worldwide expansion of rodent-borne hantaviruses.