Ethics statement

All small mammals were handled in accordance with the ethical guidelines approved by the Korea University Institutional Animal Care and Use Committee (Approval number: KUIACUC #2022–0034). All necropsies and laboratory experiments were performed in a biosafety level 3 facility at the Korea University College of Medicine, Seoul, ROK.

Sample collection

Between 2022 and 2023, small mammals—including rodents and shrews—were captured at five locations on Jeju Island, ROK: Hogeun-dong and Seohong-dong in Seogwipo-si, and Bongseong-ri, Sangdae-ri, and Ora-dong in Jeju-si. The procedures for trapping and transportation have been described previously [38]. Briefly, field sampling was conducted once annually at each site over the study period, with each trapping session lasting exactly two consecutive nights. At each location, 20–25 collapsible, live-capture Sherman traps (7.7 × 9 × 23 cm; H.B. Sherman, Tallahassee, FL, USA) were deployed each afternoon along transects spaced at 4–5 meter intervals and retrieved the following morning. Traps were placed in ecotones and transitional zones across a variety of habitats, including unmanaged grasslands, herbaceous vegetation, farmlands, and forest edges. Given the five sites sampled over two years, with two-night trapping sessions at each visit, the total trapping effort was estimated at approximately 400–500 trap-nights, depending on the number of traps deployed per site. Captured animals were individually labeled, placed in secure transport containers, and transferred to Korea University in Seoul. Upon arrival, individuals were humanely euthanized and identified to species based on external morphological characteristics; sex and body weight were also recorded. Whole blood and internal organs (lung, liver, spleen, and kidney) were aseptically collected in accordance with protocols approved by the KUIACUC and stored at −80°C until further analysis.

Mitochondrial DNA (mtDNA) analysis

Total DNA was extracted from the homogenized liver tissues using either the High Pure PCR Template Preparation Kit (Roche, Basel, Switzerland) or TRI Reagent Solution (Ambion, Austin, Texas, USA) according to the manufacturer’s instructions. To verify the taxonomic classification of small mammals, PCR amplification was performed targeting the mtDNA cytochrome b (CYTB) gene. The PCR employed universal primers for mammals and followed the cycling conditions outlined previously [39].

Indirect immunofluorescence antibody test

Serum and heart fluids were diluted with phosphate buffered saline (PBS) to final dilution factors of 1:32 and 1:2, respectively. The diluted samples were then applied to ten-well slides (PTFE Printed Slide 10 well; Electron Microscopy Science, USA) containing acetone-fixed HTNV-infected Vero E6 cells and incubated at 37°C for 30 min. Following incubation, the plates were washed sequentially with PBS and distilled water (D.W.). The cells were then treated with fluorescein isothiocyanate (FITC)-conjugated goat antibodies specific to mouse immunoglobulin G (IgG) for rodents or mouse or rat IgG for shrews (MP Bio, CA, USA). The slides were then incubated at 37°C for 30 min and washed. Hantavirus-specific fluorescence was identified using a fluorescence microscope (Axio Scope; Zeiss, Berlin, Germany). Antibody titers were quantified as the highest dilution of the sample that exhibited weak yet specific fluorescence, as determined through a two-fold serial dilution.

Reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was isolated from 50–100 mg of homogenized rodent lung tissue using TRI Reagent Solution (Ambion), in accordance with the manufacturer’s protocol. cDNA was synthesized from 1 µg of total RNA employing the High-Capacity RNA-to-cDNA Kit (Applied Biosystems, Foster City, CA, USA), using a random hexamer primer in conjunction with OSM55 (5′–TAG TAG TAG ACT CC–3′). Subsequent amplification of the target regions was conducted through first-round and nested PCR in a total volume of 25 µL. Each reaction mixture comprised 0.625 U of Ex Taq DNA polymerase (TaKaRa Bio, Japan), 2.5 µL of 10 × reaction buffer, 2 µL of a 2 mM dNTP mix, 10 pmol of each primer (final concentration: 0.4 µM), and 1.5 µL of the cDNA template. Thermal cycling conditions included an initial denaturation at 94°C for 5 min, followed by six cycles of denaturation at 94°C for 30 s, annealing at 37°C for 40 s, and extension at 72°C for 1 min. This was succeeded by 32 cycles of denaturation at 94°C for 30 s, annealing at 42°C for 40 s, and extension at 72°C for 1 min, concluding with a final extension step at 72°C for 5 min. PCR amplifications were carried out using the ProFlex PCR System (Life Technologies, CA, USA). Amplified products were purified with the MinElute PCR Purification Kit (Qiagen, Hilden, Germany). Bidirectional Sanger sequencing was performed using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) on an ABI 3730XL DNA Analyzer (Applied Biosystems). The primers used for amplification have been previously described [9].

Quantitative PCR (qPCR)

qPCR was conducted in a 10 µL reaction mixture consisting of 5 µL of SYBR Green PCR Master Mix (Applied Biosystems), 3 µL of D.W., 1 µL of cDNA, and 5 pmol of each primer (final concentration: 0.4 µM), using the QuantStudio 5 Flex Real-Time PCR System (Applied Biosystems). The cycling conditions and primer sequences have been previously described [9]. Viral copy numbers were quantified by generating a standard curve from recombinant plasmid DNA containing the S segment of HTNV, as outlined in prior studies [40].

Multiplex PCR-based nanopore sequencing

cDNA was amplified using previously described multiplex HTNV-specific primer mixtures designed for southern Korean strains, in combination with the Solg 2X Uh-Taq PCR Smart Mix (Solgent, Daejeon, ROK), following the protocol established by Prayitno et al (2024) [27]. Library preparation was performed using a Native Barcoding Kit 24 V14 (SQK-NBD 114.24; ONT). Following end preparation, the libraries were purified using AMPure XP beads (Beckman Coulter, CA, USA). The purified barcoded libraries were pooled, ligated to sequencing adapters, and subsequently sequenced on a MinION MK1C device (ONT) using a FLO-MIN114 (R10.4.1) flow cell for 14 h.

Raw reads were demultiplexed and adapter-trimmed using Porechop (v0.9.0), and only reads with a minimum Q-score of 7 were retained. Index sequences were removed, and to minimize the risk of artificial primer-derived sequence similarity, the first and last 20 nucleotides at the 5′ and 3′ ends of each read were trimmed using CLC Genomics Workbench (v22.0.2; Qiagen, Hilden, Germany). Filtered reads were then mapped to the HTNV reference genome (strain Ac20–5) using the following alignment parameters: match score = 2, mismatch cost = 3, length fraction = 0.7, similarity fraction = 0.8, and non-specific match handling = ignore. Genome coverage depth was subsequently calculated, and consensus sequences were generated based on majority-rule nucleotide calling at positions with a minimum depth of 5 × ; positions with coverage <5 × were excluded and denoted as ‘N’. Indels and frameshift errors within open reading frames were manually corrected according to previously established error correction criteria by comparing the mapped reads with the reference genome [25]. Genome regions not covered by the reads were amplified and confirmed using conventional nested RT-PCR. Final consensus sequences were deposited in GenBank.

Rapid amplification of cDNA ends (RACE) PCR

To acquire the 3’ and 5’ terminal sequences of the HTNV genome, RACE PCR was carried out using the 3’- and 5’-RACE System for Rapid Amplification of cDNA Ends, Version 2.0 (Invitrogen, Carlsbad, CA, USA), following the manufacturer’s protocol.

Phylogenetic analysis

The genomic sequences of HTNV tripartite segments were aligned using the ClustalW algorithm in MegAlign (v5; DNASTAR, Madison, WI, USA). The most appropriate substitution model was determined to be GTR + G + I using the MEGA 7 software. Phylogenetic trees for each segment were constructed using Bayesian Evolutionary Analysis by Sampling Trees (BEAST v2.6.7) under a strict molecular clock model [41]. The analysis was conducted over 80 million iterations, ensuring an effective sample size of greater than 200.

Network analysis

Phylogenetic networks were constructed using the median-joining algorithm implemented in Network software (version 10.2.0.0; Fluxus Engineering, Cambridge, UK), with the epsilon set to 0 to reduce reticulations and simplify network topology. The networks were generated from multiple sequence alignments of HTNV genomes, including 55 sequences for the S and M segments and 51 sequences for the L segment. Analyses were based on aligned nucleotide positions spanning the S segment (49–1,343 nt), the M segment (1–3,431 nt), and the L segment (28–6,506 nt). The median-joining method integrates minimum spanning trees with inferred median vectors to represent potential ancestral relationships and mutational steps among sequences. Lineages were delineated based on genetic distance, resulting in the identification of two major groups. Network visualization was performed using Network Publisher (version 2.1.2.5) and finalized in Adobe Illustrator 2021.

Amino acid substitution analysis

The amino acid sequences of the three HTNV genes were aligned using the ClustalW method in MegAlign (v5; DNASTAR). Amino acid mutations in HTNV were identified for each variant by comparison with the reference HTNV Ac20–5. The lollipop plot was visualized using the dplyr and ggplot2 packages in R version 4.3.3.

Field trapping for small mammals on Jeju Island, 2022–2023

We collected 50 small mammals from five locations: across Seogwipo-si (Hogeun-dong and Seohong-dong) and Jeju-si (Bongseong-ri, Sangdae-ri, and Ora-dong), Jeju Island, ROK, during 2022–2023 (Fig 1). The captured animals included 47 Apodemus chejuensis (94%) and three Crocidura shantungensis individuals (6%) (S1 Table). The species identification of A. chejuensis was validated by phylogenetic analysis based on mitochondrial sequencing (S1 Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Geographic locations of Hantaan virus (HTNV) carried by Apodemus chejuensis captured on Jeju Island, Republic of Korea, in 2022–2023.

The map delineates the five sampling sites: (A) Hogeun-dong, (B) Bongseong-ri, (C) Sangdae-ri, (D) Seohong-dong, and (E) Ora-dong. Locations where HTNV was detected are marked by red and blue circles, whereas black circles denote sites where HTNV was not detected. The initial map was generated using Quantum Geographical Information System (QGIS) 3.10 for Windows and subsequently refined in Adobe Illustrator 2020 (Adobe Systems Incorporated, California, USA). The base layer of the map was sourced from https://www.naturalearthdata.com/ and is freely available for use in any project without the need for permission.

https://doi.org/10.1371/journal.pntd.0013459.g001

Epidemiological surveillance of HTNV

Serological testing of A. chejuensis revealed a positivity rate of 34% (16/47), whereas HTNV RNA was detected in 27.7% (13/47) (Table 1). HTNV RNA positivity rates were 27.6% (8/29) in Hogeun-dong, Seogwipo-si; 57.1% (4/7) in Bongseong-ri, Jeju-si; and 12.5% (1/8) in Sangdae-ri, Jeju-si. Males exhibited a higher viral positivity rate (37.5%, 9/24) than females (17.4%, 4/23). The prevalence also increased with weight: 17.4% (4/23) for 21–30 g, 54.5% (6/11) for 31–40 g, and 75% (3/4) for >40 g. No HTNV RNA was detected in rodents weighing 11–20 g. The epidemiological characteristics of HTNV-infected A. chejuensis are shown in S2 Table. In this study, the shrews tested negative in both serological and molecular assays.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Serological and molecular prevalence of Hantaan virus (HTNV) in Apodemus chejuensis collected on Jeju Island, Republic of Korea, 2022– 2023.

https://doi.org/10.1371/journal.pntd.0013459.t001

Determination of HTNV RNA copy numbers from A. chejuensis lung tissues

HTNV RNA loads in the lung tissues of 13 A. chejuensis were measured (S3 Table). Five rodents (Ac23–18, Ac23–20, Ac23–15, Ac23–19, and Ac22–24) exhibited the highest viral RNA loads, with the Ct values ranging from 15.9 to 20.3, corresponding to 10⁵–10⁶ copies/μL. Three individuals (Ac23–1, Ac23–17, and Ac23–19) showed Ct values ranging from 23.0 to 24.2, indicating viral RNA amounts of 10⁴–10⁵ copies/μL. Ac23–12 and Ac23–22 had viral RNA quantities of 10²–10³ copies/μL, with Ct values between 27.2 and 29.0. Additionally, three A. chejuensis strains (Ac23–14, Ac22–20, and Ac22–23) exhibited the lowest viral loads, ranging from 0 to 1 copy, with Ct values >38.0.

Whole-genome sequencing of HTNV

Using amplicon-based nanopore sequencing, nearly complete genome sequences of HTNV were obtained from lung tissues of A. chejuensis collected from Jeju Island, ROK, between 2022 and 2023 (Table 2). Five rodents (Ac23–18, Ac23–20, Ac23–15, Ac23–19, and Ac22–24) exhibited the highest viral loads, containing 10⁵ to 10⁶ copies/µL, with genome coverages of 96.4%–98.5%, 97.5%–99.3%, and 97.1%–99.6% for the S, M, and L segments, respectively. Three individuals (Ac23–1, Ac23–17, and Ac23–19) with RNA copy numbers ranging from 10⁴ to 10⁵ copies/µL showed coverages of 98.5%, 95.3%–99.8%, and 95.8%–98.6% for the S, M, and L segments, respectively. Two variants (Ac23–12 and Ac23–22) with 10² to 10³ copies/µL displayed coverages of 95.7%–96.3%, 87.6%–91.6%, and 93.7%–93.8% for the S, M, and L segments, respectively. The three samples with the lowest viral RNA amounts (0–1 copy/µL; Ac23–14, Ac22–20, and Ac22–23) exhibited genome coverages of 86.0%–96.3%, 80.8%–91.2%, and 71.8%–93.0% for the S, M, and L segments, respectively. The average number of mapped viral reads and sequencing depths are listed in S4 Table.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Genome coverage rate of Hantaan virus (HTNV) in lung tissues of Apodemus chejuensis collected from Jeju Island, based on viral RNA copy number.

https://doi.org/10.1371/journal.pntd.0013459.t002

Nine full-length genomic sequences of HTNV were recovered from rodents captured at three locations: Hogeun-dong in Seogwipo-si and Sangdae-ri and Bongseong-ri in Jeju-si, Jeju Island, ROK, in 2022–2023. Gaps in the HTNV tripartite RNA genome sequences were filled using conventional nested RT-PCR. To achieve a whole-genome sequence of HTNV, 5′ and 3′ RACE PCR was conducted on HTNV Ac23–18, a specimen from Seogwipo-si on Jeju Island with the highest viral copy number. The 5′ terminal sequence of HTNV tripartite RNA was determined to be 5′-GGA GUC UAC UAC UA-3′, whereas the 3′ end sequence was identified as 5′-UAG UAG UAU GCU CC-3, ’ corresponding to conserved termini sequences of the reference genome sequence (HTNV Ac20–5) [9]. Terminal sequences of the remaining HTNV specimens were recovered empirically.

Genome comparison of HTNV

Nucleotide comparison analysis revealed that the HTNV genome shared 99.2% identity in the S segment, 99.1%–99.2% identity in the M segment, and 99.4–99.8% identity in the L segment with the reference HTNV Ac20–5, which was originally detected in A. chejuensis captured in Ora-dong, Jeju-si, ROK (S5 Table). The amino acid sequences showed similarities of 99.4%–99.6% for the N protein, 99.5%–99.6% for the glycoproteins, and 99.5%–99.6% for the RdRp.

Phylodynamic analysis of HTNV

A phylodynamic analysis was conducted to explore the genetic diversity and mutation dynamics among the tripartite genomes of HTNV derived from A. chejuensis collected from Jeju Island, ROK, from 2022 to 2023 (Fig 2). Phylogenetic analysis of nine HTNV variants, including Ac23–15, Ac23–17, Ac23–18, Ac23–19, Ac23–20, and Ac23–22 from Hogeun-dong in Seogwipo-si; Ac23–1 from Sangdae-ri; and Ac22–19 and Ac22–24 from Bongseong-ri in Jeju-si, demonstrated consistent clustering into a homologous genetic group along with other HTNV variants collected from Jeju Island, ROK. However, discordance in evolutionary divergence was observed within the S segment of the HTNV variant Ac23–22 from Hogeun-dong in Seogwipo-si compared to the M and L segments.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Phylodynamic analysis of Hantaan virus (HTNV) S, M, and L segments from Apodemus chejuensis captured on Jeju Island, Republic of Korea (ROK), in 2022–2023.

Nine whole-genome sequences of HTNV were newly obtained in this study. Phylogenetic inference was performed using an alignment based on the HTNV (A) S (49–1,343 nt), (B) M (1-3,431 nt), and (C) L (28-6,506 nt) genomic sequence, using the strict clock method within the Bayesian Evolutionary Analysis Sampling Trees (BEAST, v2.6.7). A total of 80 million iterations were conducted to ensure an effective sample size greater than 200. Posterior probabilities are displayed at the tips of each node, with colored circles indicating nodes with posterior values. The arrowhead marks the branch point of differentiation associated with a key amino acid variation in each segment, which was notably found in Seogwipo-si, Jeju Island, ROK. Sequence colors correspond to their respective collection sites at the city level, with the newly acquired HTNV sequences from this study denoted by stars. Accession numbers for the hantaviral sequences from this study are listed in S6 Table.

https://doi.org/10.1371/journal.pntd.0013459.g002

Moreover, four unique amino acid substitutions were identified in the tripartite genomes of HTNV variants from Seogwipo-si (Table 3 and S2 Fig). These mutations comprised one in the S segment (N protein; R199K), two in the M segment (glycoproteins; V563L and E1019K), and one in the L segment (RdRp; F439S) when compared to the corresponding sequences of HTNV variants from Jeju-si.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Key amino acid substitutions in the S, M, and L segments of Hantaan virus (HTNV) carried by Apodemus chejuensis collected on Jeju Island, Republic of Korea.

https://doi.org/10.1371/journal.pntd.0013459.t003

Phylogeographic network analysis of HTNV

Network analyses revealed a well-supported evolutionary divergence and genetic diversity among the HTNV S, M, and L genomes across various geographic regions of Jeju Island and the Korean Peninsula (Fig 3). Two distinct phylogenetic lineages were identified in the ROK. Lineage 1 encompasses regions in the northern ROK, including Pocheon-si, Yeoncheon-gun, Dongducheon-si, and Paju-si in Gyeonggi Province, and Chuncheon-si, Hwacheon-gun, Cheorwon-gun, and Yanggu-gun in Gangwon Province. Conversely, lineage 2 was associated with the southern regions of the ROK, including Seogwipo-si and Jeju-si on Jeju Island, and Gwangju-si and Boseong-gun in Jeollanam Province. These two lineages were defined based on consistent phylogeographic clustering and regional separation observed across all three genomic segments. The mean nucleotide divergence between Lineage 1 and Lineage 2 ranged from approximately 12% to 18%, depending on the segment.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Phylogeographic network analysis of Hantaan virus (HTNV) from Apodemus chejuensis collected on Jeju Island, Republic of Korea (ROK), in 2022–2023.

The phylogeographical network was generated from alignments of HTNV tripartite genomes for 55 genomic sequences (for S and M segments) and 51 genomic sequences (for L segment). (A) The geographic distributions of HTNV variants are indicated by colored circles at the collection sites on the map. The initial map was generated using Quantum Geographical Information System (QGIS) 3.10 for Windows. The base layer of the map was sourced from https://www.naturalearthdata.com/ and is freely available for use in any project without the need for permission. Networks were constructed using the median-joining method in Network software (version 10.2.0.0; Fluxus Engineering, Cambridge, UK) with an epsilon value of 0, based on the HTNV (B) S (49–1,343 nt), (C) M (1–3,431 nt), and (D) L (28–6,506 nt) segments. Color-coded symbols represent individual HTNV variants according to their collection sites, with circle sizes proportional to the connection frequency between viruses. The nodes illustrate the relationships among viral strains, with red numbers on the nodes indicating the total nucleotide mutations between strains. Lineage 1 spans regions in northern ROK, including Pocheon-si, Yeoncheon-gun, Dongducheon-si, and Paju-si in Gyeonggi Province, and Chuncheon-si, Hwacheon-gun, Cheorwon-gun, and Yanggu-gun in Gangwon Province. Lineage 2 covers areas in southern ROK, including Seogwipo-si and Jeju-si on Jeju Island, as well as Gwangju-si and Boseong-gun in Jeollanam Province. HTNV genomes from China were used as the outgroup in this study. Accession numbers for the hantaviral sequences from this study are listed in S6 Table.

https://doi.org/10.1371/journal.pntd.0013459.g003