Journal of Vector Ecology · 2012 · 10 citations · 17 references
Life SciencesImmunologyInfectious DiseasesMammalogyMolecular VirologyVirologyAnimal VirusesViral Diseases Of WildlifeViral EpidemiologyViral InfectionsViral ZoonosesZoonotic VirusesHantavirus InfectionsImmunofluorescence AssaysImmunofluorescence DiagnosisIndirect Immunofluorescence AssayRodent-borne Viral DiseasesRodent-borne Viral ZoonosesRodent-borne ZoonosesTrapped Populations
The seroprevalence of Hantaan virus (HTNV) in wild rodents in South Korea was analyzed. Wild rodents were trapped in 18 cities in eight provinces during 2005–2007 and on three islands and four mountains during 2008–2010. Sera were collected from 629 out of 933 trapped wild animals and examined for immunoglobulin G antibodies to HTNV using indirect immunofluorescence assays. Apodemus agrarius (80.1%) was the most frequently captured species at almost all trapping sites. The overall prevalence of HTNV antibodies was 0.26 (162/629). Seropositive individuals were more frequent in cities (32.2%, n=410) than on islands (14.0%, n=57) or mountains (13.6%, n= 162). HTNV antibody-positive rate was higher in the fall (29.6%, n=253) than in the spring (23.1%, n=376). A. agrarius had the highest prevalence of HTNV antibodies (26.9%, n=561) of all tested species. Considering all the individuals, the prevalence of HTNV antibodies was higher in males (29.2%, n=250) than in females (22.3%, n=305). Our results show that HTNV is widely distributed throughout South Korea, and that HTNV infection of wild rodents is affected by their habitat, species, sex, and season. Hantaviruses (family Bunyaviridae, genus Hantavirus) are negative-sense, enveloped RNA viruses. Hantaviruses are persistently maintained in infected rodents, and each hantavirus appears to have co-evolved with a specific rodent host according to the hantavirus species and rodent gene phylogenies (Jonsson et al. 2010). Hantaviruses are shed in saliva, urine, and feces of infected rodents, and human infection occurs by inhalation of aerosolized virus in association with rodent excreta (Xu et al. 1985). At least 45 different hantavirus species have been found worldwide (Dearing and Dizney 2010). They cause two different syndromes in humans: haemorrhagic fever with renal syndrome (HFRS) and hantavirus pulmonary syndrome, depending on the hantavirus species (Dearing and Dizney 2010). The rodent reservoirs that carry HFRS-causing viruses in Europe and Asia are as follows: (i) the striped field mouse (Apodemus agrarius), carrying Dobrava–Belgrade (DOBV) and Saaremaa viruses in Europe and Hantaan virus (HTNV) in Asia (Plyusnina et al. 2011, Lee et al. 1978); (ii) the yellow-necked mouse (Apodemus flavicollis), DOBV in Europe (Sibold et al. 2001); (iii) Norway rat (Rattus norvegicus), Seoul virus (SEOV) in Asia (Lee et al. 1982); and (iv) bank vole (Myodes glareolus), Puumala virus (PUUV) in Europe (Vapalahti et al. 2003). In South Korea, the Korean field mouse (Apodemus peninsulae), royal vole (Myodes regulus), and Ussuri white-toothed shrew (Crocidura lasiura) have recently been identified as hosts of the hantaviruses, Soochong virus (SOOV) (Baek et al. 2006), Muju virus (MUJU) (Song et al. 2007), and Imjin virus (Song et al. 2009), respectively, but whether they cause illness in humans is unknown (Jonsson et al. 2010). In South Korea, approximately 400 HFRS cases are reported annually, and about 40 deaths caused by HFRS were registered over the past ten years (KCDC 2011). About 70% of HFRS cases are caused by HTNV and the remaining 30% by SEOV and other unidentified agents (Noh et al. 2006). HTNV has the potential to cause HFRS anywhere in South Korea because A. agrarius is the most abundant and widespread rodent on the Korean peninsula (Ryou et al. 2011, Song et al. 2006). Here we report the seroprevalence of infection with HTNV in wild rodents in South Korea and analyze the characteristics according to trapping region, season, species, and sex of rodent host. Trapping of rodents was conducted in 18 cities across eight provinces during 2005–2007 and on three islands and four mountains during 2008–2010 (Figure 1). We conducted trapping in spring or fall and the temperature range of trapping sites was 9°– 26° C. The average altitudes of trapping sites were 163.2 m and 177.0 m at cities and islands, respectively. The range of mountain sites’ altitudes was 177–1,102 m. Trapping was performed for one, three, and two nights per year at each city, island, and mountain, respectively. We used Sherman live traps (8×9×23 cm; H. B. Sherman Trap Co., Tallahassee, FL, U.S.A.) baited with a mixture of oats and peanut butter. Two hundred traps per night were set at 16:00–18:00 and collected on the next morning. Captured animals were carried to local laboratories and euthanized according to an approved animal use protocol (KCDC 2005). The species, sex, body length, and weight of each captured animal as well as date and location of capture were recorded. We collected blood samples from the rodents which were alive when they were carried to local laboratories. Blood samples, collected with separate sterile instruments, were stored at –70° C until use. Map of trapping sites in South Korea. GG, Gyeonggi province, GW, Gangwon province; CB, Chungcheongbuk province; CN, Chungcheongnam province; GB, Gyeongsangbuk province; GN, Gyeongsangnam province; JB, Jeollabuk province; JN, Jeollanam province; JJ, Jeju special autonomous province (•, city; □, island; ▴, mountain). Because almost all captured rodents were A. agrarius, which is the primary host of HTNV, we examined immunoglobulin G (IgG) antibodies against HTNV by an indirect immunofluorescence assay (IFA) technique (Lee et al. 1985). Briefly, Vero E6 cells (CRL 1586, ATCC, Manassas, VA, U.S.A.) were maintained in Eagle's minimal essential medium containing 5% heat-inactivated foetal bovine serum, 500 IU/l penicillin, and 500 mg/ml streptomycin. HTNV strain 76–118 was inoculated and cultivated on a confluent monolayer of Vero E6 cells for seven days at 37° C in the presence of 5% CO2. The HTNV-infected cells were detached with trypsin and air-dried on slide spots. The slides were fixed with 80% cold acetone and stored at –70° C until use. Aliquots of a two-fold serial dilution (from 1:16 to 1:4096) of sera from captured animals were added to slides and incubated in a moist chamber for 30 min at 37° C. The slides were washed three times for 5 min each with PBS and incubated with FITC-anti-mouse IgG conjugate (Jackson Immuno Research Laboratory, West Grove, PA, U.S.A.) for 30 min at 37° C. The slides were washed three times for 5 min each with PBS and then air-dried. The slides were examined using a fluorescence microscope (400×, Axioskop2 plus, Carl Zeiss, Hallbergmoos, Germany). The sera from captured animals were considered positive if typical fluorescence was detected at a dilution of 16 or more. Seroprevalence was estimated as the percentage of rodents that tested positive in the IFA. Chi-square analysis was used to compare the frequencies of seropositive individuals according to trapping region and sex. Over six years, 933 small mammals were trapped across South Korea (Table 1). A. agrarius (80.1%) was the most frequently captured species at almost all trapping sites in this study. Other individuals belonging to the Rodentia order were M. regulus (4.0%), Micromys minutus (2.0%), R. norvegicus (1.8%), and A. peninsulae (1.8%). In addition, two species in the Soricomorpha order were trapped: C. lasiura (8.1%) and Crocidura suaveolens (0.3%). A. agrarius represented 86.9% of all individuals trapped in cities. On the other hand, A. agrarius represented only 65.2% of an otherwise diverse range of species on islands and mountains. From 933 wild animals, 629 sera (67.4%) were collected and examined for antibodies to HTNV (Table 2A). The overall prevalence of HTNV antibodies was 0.26 (162/629). The seroprevalence differed significantly among regions studied (c2=25.55, df=2, P<0.01). Seropositive individuals were more frequent in cities than on islands or mountains (cities: 32.2%, n=410; islands: 14.0%, n=57; mountains: 13.6%, n=162). By season, small mammals of cities and islands had more HTNV antibodies in fall than in spring. In the mountain regions, however, the opposite pattern was detected. The overall seroprevalence was higher in fall (29.6%) than in spring (23.1%) but not statistically significant. (c2=3.348, df=1, P=0.07). Among 629 analyzed sera from across South Korea, species of 627 small mammals were distinguished (Table 2B). A. agrarius showed the highest prevalence of HTNV antibodies (26.9%), followed by A. peninsulae (21.4%), R. norvegicus (18.8%), and M. regulus (17.2%). HTNV antibodies were not detected in M. minutus and C. lasiura, which had low sample sizes (1 and 6, respectively). Antibodies against HTNV were mainly found in A. agrarius (32.4%) in cities, but they were also detected in other species in mountains (A. peninsulae, 25.0%; M. regulus, 17.2%) and islands (R. norvegicus, 14.3%). From 629 analyzed sera, the sexes of 555 small mammals (45.0% of males) were recorded (Table 2C). The prevalence of HTNV antibodies was higher in males (29.2%, n=250) than in females (22.3%, n=305). This difference was approaching significance (c2=3.456, df=1, P=0.06). Comparison of seropositivity for captured animals by region showed that males in cities and mountains were more often seropositive than females, but the opposite was indicated for those of islands. In this study, the HTNV seroprevalence of wild rodents was investigated throughout South Korea. Consistent with previous studies (Ryou et al. 2011, Song et al. 2006), A. agrarius (80.1%) was the most abundant wild rodent species at almost all trapping sites in this study. In fact, approximately 70% of HFRS patients in South Korea are infected by HTNV, which is carried by A. agrarius. Most of the trapping sites in the 18 cities were near farmhouses where wild rodents had easy access to abundant food. The four mountain sampling sites were in designated national parks containing grassland, broadleaf forest, and coniferous forest environments. Wild animals on islands were trapped in grassland, fields, or embankments around the end of fields. Antibodies to HTNV were found in almost all sampling sites, suggesting that HTNV infection is widespread in South Korea. Wild rodents in cities showed higher seropositivity (32.2%) than those on islands and mountains. We found a significant spatial difference in HTNV seroprevalence between cities and islands or mountains. The reason for this difference is uncertain. It may be that city environments are preferred by seropositive rodents, or that rodents living in cities have more chances to transmit HTNV to one another. However, there was no direct relationship between the HTNV seropositivity rate and the HFRS incidence rate (data not shown). According to previous studies, the probability of hantavirus transmission from wild rodents to humans is influenced by several factors, including climate and human settlements, occupation, and activity (Jonsson et al. 2010). Our data show higher overall HTNV antibody-positive rates in the fall season compared with that in the spring, which are consistent with another study (Douglass et al. 2007). In fall, A. agrarius have high reproductive capacity, which may affect high transmission of HTNV among them (Sames et al. 2009). However, when the data was stratified by trapping regions, the island's host animals increased in HTNV seropositivity rates in spring. This result may be caused by other factors such as environmental conditions and movement. Among A. agrarius individuals captured in this study, 26.9% had antibodies to HTNV. Although only HTNV antigen slides were used in this study, the seropositivities of R. norvegicus, A. peninsulae, and M. regulus were 18.8%, 21.4%, and 17.2%, respectively. These species are known to be reservoirs of SEOV, SOOV, and MUJU, respectively, which have been isolated in South Korea (Lee et al. 1982, Baek et al. 2006, Song et al. 2007). Multiplex RT-PCR assays to amplify HTNV, SEOV, and PUUV simultaneously were conducted with total RNAs from lung samples collected during 2005–2010, using the oligonucleotide primers designed in previous study (Ryou et al. 2011). Three viral genomes could be amplified to HTNV amplicon size from rodent species other than A. agrarius, but virus isolation failed (data not shown). Therefore, further studies on HTNV-seropositive samples of rodent species other than A. agrarius are necessary. Considering all the individuals, the prevalence of HTNV antibodies was higher in males than in females in cities and mountains, whereas females on islands had higher seropositivity than males. It was suggested that males mainly contributed to HFRS infection because of their wounding during competition for mating (Lee et al. 1981). In addition, the seroprevalence of SEOV among adult rats has been higher in males than in females (Cueto et al. 2008, Hinson et al. 2004). However, according to Nitatpattana et al. (2002), HTNV seroprevalence was not significantly different in age class or sex of all tested species. Our study did not discriminate among rodents by age but rather analyzed individuals of all ages. Therefore, there is a need for seroprevalence analysis according to sex and maturity in order to characterize HTNV infection in wild rodents in South Korea. Our study analyzed the seroprevalence and characteristics of infection with HTNV in wild rodents in cities and on islands and mountains across South Korea. HTNV seropositivity in rodents which are not the natural hosts of HTNV are thought to be due to cross-reactivity in serological assays, and we suspect that they were not infected with HTNV but with other types of hantaviruses. Additional RT-PCR screening with various primers and sequencing is needed to more deeply understand the biodiversity and molecular evolution of hantaviruses in South Korea. We thank all the researchers who were engaged in the animal trapping and treatment. This research was funded by the National Institute of Health, Korea Centers for Disease Control and Prevention.
17