Abstract
Introduction:
Recently, hantaviruses have been discovered in insectivores in Europe, Asia, Africa, and North America. Imjin virus (MJNV) was first isolated from the lung tissues of Ussuri white-toothed shrew (Crocidura lasiura) from South Korea in 2009. We aim to detect the species and prevalence of insectivore- and rodent-borne hantaviruses in shrews and rodents.
Materials and Methods:
Shrews and rodents were captured in Jiaonan County of Shandong Province, China, in 2014. RT-PCR was used to amplify viral RNA of Hantavirus species, including insectivore-borne Imjin virus (MJNV), rodent-borne Hantaan virus (HTNV), and Seoul virus (SEOV) from shrews and rodents.
Results and Discussion:
We found that MJNV infected 10.7% (19/178) of Crocidura shrews, but it infected none of rodents (0/475); we also found that 2 of 178 (1.1%) Crocidura shrews were PCR positive to SEOV. This study indicated that the major animal hosts of Imjin virus are shrews, and rodent-borne SEOV can infect shrews.
Introduction
H
Imjin virus (MJNV) was the second insectivore-borne hantavirus that was isolated from shrews in 2009 after Thottapalayam virus (TPMV) (Carey et al. 1971, Song et al. 2009). Although many studies had detected insectivore-borne hantaviruses in shrews and rodent-borne hantaviruses in rodents, to our knowledge, insectivore-borne hantaviruses were not investigated for whether they could infect rodents or vice versa. In China HFRS is caused by Hantaan virus (HTNV) and Seoul virus (SEOV). MJNV was also identified in shrews in Zhejiang Province in Southern China (Lin et al. 2014). We found that shrews and rodents were often captured in the same area in Jiaonan County, Shandong Province in East China, suggesting that shrews and rodents in East China might be exposed to both insectivore-borne and rodent-borne hantaviruses. The aim of this study was to determine whether shrews in Shandong Province carried both insectivore-borne MJNV and rodent-borne HTNV and SEOV by detecting viral RNA of these viruses in shrews and rodents captured in Jiaonan County.
Materials and Methods
Study site and animal trapping
Jiaonan County (longitude 119°30′–120°30′, latitude 35°35′–36°08′) in Shandong Province was one of the high-incidence areas of HFRS in China (Qiu and Su 2005). Rodents and shrews were trapped once in the middle of each month (except for September) for 3 consecutive days using snap traps with peanut bait in Jiaonan County in 2014. Everyday 200 traps in total were set before sunset in rural areas surrounding farmlands and collected in the morning in different sites in Jiaonan County. The traps were strategically placed along animal traffic routes to maximize trap success. The trap sites were in close proximity to laboratory-confirmed cases of HFRS and were placed where the numbers of rodents were abundant according to the local farmers. The lung tissues of the rodents and shrews were collected aseptically and frozen at −80°C until use. We recorded the basic information about the shrews and rodents such as sex and appearance. Shrews and rodents were classified according to appearance (hair color) and body structures. All shrews were further confirmed by molecular typing described below. Specimens were archived for future study.
Ethics statement
The collection and use of the animals were approved by the Bioethics Committee of School of Public Health, Shandong University.
RNA and DNA extraction
Total RNA and DNA of shrews and rodents were extracted from 30 mg of lung tissues of each animal with the AllPrep DNA/RNA Mini Kit (Qiagen, Germany) according to the manufacturer's instruction. RNA was used as a template for virus amplification by using the RT-PCR System (Promega, Madison, WI).
RT-PCR amplification of Imjin virus
For amplification of MJNV in shrews and rodents, outside and inside primers for RT-PCR were designed from the S and L segments of the MJNV genome in this study (Table 1). We used one step RT-PCR and nested-PCR. The RT-PCR cycle included an initial denaturation at 95°C for 3 min, followed by 30 cycles of 95°C 30 s, 53°C 60 s, and 68°C 90 s and a final extension of 10 min at 68°C. We used one step RT-PCR products as template for nested PCR. The nested PCR cycle included an initial DNA denaturation at 95°C for 5 min, followed by 30 cycles of 95°C 30 s, 53°C 30 s, and 72°C 90 s and a final extension of 10 min at 72°C. We also used Hantaan virus (HTNV) and Seoul virus (SEOV) common primer pairs to detect HTNV and SEOV in shrews by nested RT-PCR as described previously (Fang et al. 2015).
HTNV, Hantaan virus; MJNV, Imjin virus; SEOV, Seoul virus.
The PCR products were separated by agarose gel electrophoresis and the band with the expected size was excised from the gel and purified with the E.Z.N.A. Gel Extraction Kit (Promega, Madison, WI). The purified PCR products were cloned into pMD19-T vector (TaKaRa, Dalian, China), and two clones derived from each animal were sequenced on both strands.
PCR amplification of shrew mtDNA
All shrews were molecularly typed using cytochrome b region of mtDNA by PCR with primers (L14724: CGAA GCTTGATATGAAAAACCATCGTTG and H15149: AAA CTGCAGCCCCTCAGAATGATATTTGTCCTCA) (Linacre and Lee 2005). PCR products were cloned and sequenced as described above.
Phylogenetic analyses
Sequences were aligned using the Clustal W method with MEGA 5 (Tamura et al. 2011). The nearly complete sequences of the S (1518 nucleotides), M (3369 nucleotides), and L (6039 nucleotides) segments of Imjin virus and the partial sequences of S (241 nucleotides) segments of Seoul virus were used for phylogenetic analyses. Phylogenetic trees were constructed using the neighbor-joining method available at MEGA 5.
Statistical analysis
Pearson's chi-squared test was used to evaluate differences in the prevalence of hantavirus infection of different seasons and sex using SPSS 18.0 (SPSS, Chicago, IL).
Results
Captured shrew and rodent species
From January to December 2014, a total of 178 shrews (54 male and 124 female) were captured in Jiaonan County. Among the shrews, 164 were Crocidura lasiura, 7 were Crocidura attenuata, and 7 were Crocidura shantugensis. We first calculated the sample size for random sampling at average prevalence of MJNV of rodents as 0.05, allowable error as 0.02, and α value as 0.05. The results suggested 465 is the suitable value for our sample size. The whole number of our captured rodents is 950. We randomly chose 475 rodents (176 male and 299 female) from the total rodent population, which was captured from January to December 2014 (except for September). The rodents included seven species with 163 striped field mice (Apodemus agrarius), 102 house mice (Mus musculus), 88 brown rats (Rattus norvegicus), 62 greater long-tailed hamsters (Cricetulus triton), 29 Chinese hamsters (Cricetulus griseus), 29 Chinese striped hamsters (Cricetulus barabensis), and 2 buff-breasted rats (Rattus flavipectus).
Hantaviruses species detected in shrews and rodents
We first screened insectivore-borne MJNV and rodent-borne HTNV and SEOV in the captured shrews. We found that 19 of 178 (10.6%) shrews, including 18 C. lasiura shrews and 1 C. shantugensis shrew were positive to MJNV with S-segment primers, and 18 of 178 (10.1%) shrews, including 17 C. lasiura shrews and 1 C. shantugensis shrew, were positive to MJNV with L-segment primers (Table 2). The MJNV-positive C. shantungensis shrew and C. lasiura shrews were captured at the different sites. Eighteen of 19 shrews that were positive for S segments were also positive for the L segment. The partial sequences of S and L segments of MJNV were highly homologous among the shrews, which were 99.1–99.9% homologous for S segment sequences and 99.3–100.0% homologous for L segment sequences. We also amplified RNA sequences from 2 of 178 (1.1%) shrews (C. lasiura) with HTNV and SEOV common primers, and DNA sequencing showed that both sequences were SEOV. The two partial sequences (241-bp) were 95% homologous to each other and were 95.4–100.0% homologous to the S segment of SEOV in GenBank. No shrew was infected with both MJNV and SEOV. The prevalence of MJNV infection among shrews had no statistical difference between different sexes (p = 0.514). We then tested for insectivore-borne MJNV on captured rodents and found that none of the 475 rodents was positive to MJNV. The rodents were captured at the same sites as shrews simultaneously. The sequences of MJNV and SEOV detected in shrews were deposited in GenBank (acc. nos. KY094128, KY094129, and KU997695–KU997712).
Phylogenic analysis of the MJNV and SEOV
We amplified 1518 nucleotides of the S segment of MJNV from 19 MJNV-positive shrews, which was almost the complete sequence of S segment of MJNV (expected to be 1567 nucleotides). The sequences of the S segment among the 19 shrews were highly homologous with 99.1–99.9% homology. The partial sequences (692 nucleotides) of L segments of MJNV from shrews were also highly homologous with 99.3–100.0% homology. We expected that the completed M or L segments of MJNV in the shrews should be highly homologous, and therefore, we only further sequenced the M and L segments in one MJNV-positive shrew (no. 705) by genome walking. We obtained a 3369-nucleotide sequence for the M segment of MJNV from the shrew, which was expected to be 3649 nucleotides, and a 6039-nucleotide sequence for the L segment of MJNV from the shrew, which was expected to be 6581 nucleotides. Comparison of these sequences of nearly complete S, M, and L segments from Shrew no. 705 with known hantaviruses indicated that the viral strain was closely clustered together with MJNV from Zhejiang Province in Southern China and with MJNV from South Korea (Fig. 1). The nucleotide differences between the MJNV in this study and the strain Cixi-Cl-23 from Zhejiang Province were 4% for S segment, 3% for M segment, and 2% for L segment.

Phylogenetic trees were generated by NJ method based on the S
Phylogenetic analysis with two 241-nucleotide sequences of the S segment from C. lasiura indicated that the viruses in shrews were clustered together with SEOV (Fig. 2).

Phylogenetic tree was generated by NJ method based on the S segments of the SEOV and other hantaviruses in GenBank using MEGA5 software with 1000 replicates for bootstrap testing. Numbers (>50) above or below branches are posterior node probabilities. The GenBank number and the name of viral species were labeled on each line. Triangles indicated MJNV sequences obtained in this study. Scale bar indicates nucleotide substitutions per site. SEOV, Seoul virus.
Discussion
Hantaan virus was first identified in the striped filed mouse (A. agrarius) in 1978 in South Korea (Lee et al. 1978). Then a variety of species or serotypes of hantaviruses were identified in different rodent animal hosts, including SEOV in rats (R. norvegicus), Dobrava virus (DOBV) in yellow-necked field mice (Apodemus flavicollis), and Puumala virus (PUUV) in bank voles (Myodes glareolus) (Song et al. 2007a). These findings suggested that hantavirus associated with rodents until discovery of the insectivore-borne hantavirus Thottapalayam virus (TPMV) from the Asian house shrew (Suncus murinus) (Carey et al. 1971). TPMV has been placed in genus Hantavirus of the family Bunyaviridae by virtue of its morphologic features and overall genetic similarities to well-characterized rodent-borne hantaviruses (Zeller et al. 1989, Chu et al. 1994, Xiao et al. 1994). Over the past years, many novel hantaviruses have been found in shrews, moles, and bats, including Imjin virus (MJNV) in C. lasiura from Korea (Song et al. 2009), Tanganya virus (TGNV) in C. theresae from Guinea (Klempa et al. 2007), Qian Hu Shan virus (QHSV) in Sorex cylindricauda from China (Zuo et al. 2014), Cao Bang virus (CBNV) and Lianghe virus (LHEV) in Anourosorex squamipes from Vietnam and China (Song et al. 2007b, Guo et al. 2013), Yakeshi (YKSV) virus in Sorex isodon from China (Guo et al. 2013), and Kenkeme virus (KKMV) in Sorex roboratus from Russia and China (Kang et al. 2010, Wang et al. 2014).
Although many hantaviruses were found in shrews and other animals, the knowledge about the host specificity and host ranges of these viruses are difficult to find. Insectivore-borne hantaviruses were not studied in rodents and vice versa. We found that MJNV had a high infection rate (10.7%) in shrews captured in Jiaonan County of Shandong Province. However, we found that no rodent was infected with MJNV. The rodents and shrews were captured in the same areas at the same time; we expected that both species can be exposed to MJNV. These results confirm earlier studies that C. lasiura is the reservoir host of MJNV. We found that a small percentage of shrews (1.1%) were infected with SEOV. This is consistent with previous studies, which indicated SEOV infection of shrews in China (Fang et al. 2015, Wang et al. 2015). Another study did not find SEOV in C. lasiura and S. murinus shrews in Zhejiang Province, China (Lin et al. 2014). The inconsistency in detection of SEOV in shrews in different reports may be caused by low infection rate of SEOV in shrews, sample size differences, PCR protocol variations, and/or different study sites.
We did not analyze the infection rate of rodents to Hantaan virus and Seoul virus. A previous study carried out a year earlier in the same area indicated that rodents were seroprevalent (13.7%) to hantavirus and were viral RNA positive to HTNV (2.5%) and SEOV (2.2%), suggesting the presence of rodents infected with hantavirus in that area (Fang et al. 2015). Although more and more hantaviruses have been found in shrews, moles, and bats, they were investigated only on the hosts from which they were discovered originally. The aim of our study was to determine whether shrews in Shandong Province carried both insectivore-borne MJNV and rodent-borne HTNV and SEOV. Rodent-borne hantaviruses HTNV and SEOV are capable of causing disease in humans, raising the possibility that newfound hantaviruses detected in shrews may similarly cause serious disease in humans. A study provided that serology evidence of shrew-borne hantaviruses can infect humans in Africa (Heinemann et al. 2016), suggesting that shrew-borne hantaviruses could cause disease in humans.
MJNV can cause lethal disease in infant and juvenile Syrian hamsters (Gu et al. 2015), suggesting that rodents are susceptible to MJNV. Our study indicated that rodents were negative to MJNV RNA, suggesting either MJNV caused a lethal disease in rodents or rodents were not exposed to MJNV in nature.
Conclusions
We demonstrated that MJNV is highly prevalent in shrews collected in Shandong Province, and we also detected SEOV, but not HTNV in shrews, suggesting that the major animal hosts of Imjin virus are shrews; and SEOV can infect both rodents and shrews.
Footnotes
Acknowledgments
This study was supported by grants from National Natural Science Funds of China (nos. 31570167 and 81401368) and Shandong Province Science and Technology Development Program (no. 2014GSF121004).
Author Disclosure Statement
No competing financial interests exist.
