https://orcid.org/0000-0001-7405-0404
Figures
Abstract
Severe fever with thrombocytopenia syndrome virus (SFTSV) is a recently identified tick-borne virus that has emerged in the twenty-first century. Its primary clinical manifestations include fever and thrombocytopenia, and its high morbidity and mortality rates have garnered significant attention. It is crucial to have a comprehensive understanding of the spatial and temporal characteristics of SFTSV migration in order to prevent and control this disease. The SFTSV strains from East Asian countries in GenBank during 2017-2023 were collected and analyzed with phylogenetic and Bayesian methods. Phylogenetic analysis showed that SFTSV can be categorized into five genotypes (A, B, C, D, and E), with 24 recombination events and 15 reassortment events identified. This represented a higher number than previously observed. The results of our study indicated that SFTSV first diverged around 1785. We categorized the migration of SFTSV into two distinct periods, and identified the centers of spread and migration routes of SFTSV in each period. We propose that the eastern migration routes of migratory birds played a pivotal role during the initial stages of virus transmission, functioning as a primary conduit for the dispersal of the virus across the sea. The eastern and central migratory routes were similarly pivotal in subsequent phases of virus transmission. The results of the study showed that Japan was the first region where the virus originated and became endemic, and that the virus spread widely among countries. Elucidating the spatial and temporal characteristics of SFTSV migration will help prevent and control SFTS.
Author summary
In Asia, SFTSV is a pathogen that can trigger a disease with a high fatality rate. Our research comprehensively utilized a large number of SFTSV gene sequences to conduct phylogenetic analysis, explore rearrangement and recombination events, as well as perform discrete phylogeographic analysis. Remarkably, the number of rearrangement and recombination events identified in this study surpassed previous observations. Moreover, the migration process of the virus could be divided into two main stages, and we also determined the transmission centers and migration routes of SFTSV in each stage. Based on these findings, it is speculated that the virus originated in Japan. The migration routes of migratory birds highly coincide with those of the virus, suggesting that migratory birds play an extremely crucial role in the transoceanic spread of the virus. Future research could focus on further investigating the specific mechanisms by which migratory birds facilitate virus transmission, potentially leading to more effective prevention and control strategies.
Citation: Sheng R, Cheng T, Wang Y, Wen H (2025) Molecular evolution and geographic migration of severe fever with thrombocytopenia syndrome virus in Asia. PLoS Pathog 21(3): e1012970. https://doi.org/10.1371/journal.ppat.1012970
Editor: Ronald Swanstrom, University of North Carolina at Chapel Hill, UNITED STATES OF AMERICA
Received: September 16, 2024; Accepted: February 10, 2025; Published: March 7, 2025
Copyright: © 2025 Sheng et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are in the manuscript and/or supporting information files.
Funding: This work was supported by the Special Key Project of Biosafety Technologies (2022YFC2604000 to HLW) for the National Major Research & Development Program of China and the Natural Science Foundation of Shandong Province (ZR2022MH130 to HLW). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Severe Fever with Thrombocytopenia Syndrome (SFTS) is a recently identified tick-borne infectious disease that emerged at the beginning of the 21st century [1,2]. It presents with main clinical manifestations, including fever, thrombocytopenia, leukopenia, central nervous system symptoms, and gastrointestinal symptoms [3–5]. In 2009, experts and scholars first identified the causative agent of SFTS in China, which was subsequently named the Severe Fever with Thrombocytopenia Syndrome Virus (SFTSV) [5]. SFTSV is a single-stranded negative RNA virus whose genome consists of three segments, S, M, and L [6], and belongs to the order Bandavirus genus, Phenuiviridae family, Bunyavirales order [7]. The CFR of SFTS was initially reported to be 12–30% in 2010 in China [8]. As of 2021, SFTS cases have been identified in at least 20 provinces in China, with 12,953 confirmed cases [9]. Furthermore, the disease has been documented in other East Asian countries, including South Korea [10], Japan [11], and Thailand [12]. In 2018, the World Health Organization (WHO) identified SFTS as one of the ten most critical infectious diseases [13]. To gain a deeper comprehension of SFTS outbreaks and epidemics and to anticipate their prospective developments, comprehensive investigation into the multifaceted aspects of SFTSV is imperative.
The molecular evolution of SFTSV has progressed somewhat at a slower rate than the pace of investigation into other aspects of SFTSV, including antiviral mechanisms and epidemiological characteristics [14]. Genotyping is a valuable tool for investigating discrepancies in viral immunogenicity, antigenicity, and pathogenicity [15]. Nevertheless, there is currently no recognized genotyping method for SFTSV. Some studies had categorized it into two clades and eight genotypes [16], while others had identified five [17] or six genotypes [18–20]. A molecular phylogenetic analysis of SFTSV revealed that SFTSV strains could be categorized into Chinese and Japanese branches, indicating that SFTSV strains from disparate regions exhibit disparate evolutionary and genetic diversity [21]. Recombination and reassortment represent two essential mechanisms through which RNA viruses generate recombinant genomes [22]. Previous studies had identified homologous recombination in the M [23] and L [24] segments of SFTSV, whereas recombination events in the S segment had been observed less frequently [25]. Moreover, studies have demonstrated high reassortment rates in select segmented RNA viruses, including the influenza A virus [26]. This phenomenon was postulated to underlie the observed genetic evolution and emergence of new genotypes. Similarly, reassortment events were identified in the S, M, and L segments of SFTSV [27]. However, the data used in these studies are relatively limited, which consequently limits the support for the analysis results. With regard to the provenance and migratory pathways of SFTSVs, there is considerable inconsistency between the findings of different studies. Based on the genotype distribution and Bayesian analysis of SFTSV, Liu et al. presented evidence indicating that SFTSV originated in central China, subsequently spreading to eastern and northeastern China, and then crossing the sea to Japan and Korea [19]. Fu et al. demonstrated that strains of the SFTSV identified in the Zhoushan Islands were of Central Chinese and Korean origin, while those isolated in Japan were determined to have originated in Korea [18]. Liu et al. posited that Zhejiang Province was the most probable source of SFTSV, with a likelihood of 48.66% [28]. Additionally, they proposed that Korea and Japan were also potential origins of SFTSV, with probabilities of 19.73% and 14.24%, respectively. Furthermore, the data regarding the temporal origin of SFTSV exhibit considerable variability, with estimated ranging from 50 to 215 years ago [19,24]. The relatively limited amount of full-length genome sequence data for SFTSV in GenBank, coupled with the significant uneven distribution of sequences across locations and time, raises concerns about the representativeness of the data. In terms of geographical distribution, there is a greater prevalence of sequences from Henan and Hubei Provinces, with fewer sequences from Japan and Korea, as well as endemic provinces such as Shandong, Liaoning, and Anhui. This suggests a high probability of bias in the conclusions drawn.
The recent accumulation of additional sequence data of SFTSV in GenBank has facilitated a more comprehensive understanding of the molecular evolution and spread of SFTSV, with an increased sampling of regions and years. In the present study, we obtained SFTSV strains from East Asian countries in GenBank between 2017 and 2023 and conducted a phylogenetic analysis of their complete genomes. The results indicate that SFTSV may have experienced recombination and reassortment during its evolutionary history. Subsequently, we analyzed the approximate year of origin of the various SFTSV genotypes, the provenance of the virus, and the migration pathways. Furthermore, we discussed the potential factors influencing SFTSV migration and evolution.
Materials and methods
Sequence dataset
The SFTSV sequences (including those isolated from non-human animals) from 2017 to 2023 utilized in this study were obtained directly from GenBank and designated as S1 Table. The dataset comprised a total of 1,645 sequences (L: 554; M: 577; S: 514), of which 485 strains exhibited complete L, M, S segments. Subsequently, data pertaining to the selected SFTSV strain were collated, comprising information on the host, the site of collection, and the date of collection. The complete methods are detailed on this internet site for protocol sharing (link: https://www.protocols.io/blind/FD49B277E46F11EFACCC0A58A9FEAC02).
Phylogenetic analysis
To reduce redundancy and prevent over-representation of genomic information, we uploaded the sequences in S1 Table to CD-HIT [29] and removed sequences with greater than 99.9% similarity, thereby obtaining S2 Table for phylogenetic analysis. This dataset included 1,253 sequences from GenBank (L: 434; M: 458; S: 361).
Phylogenetic analyses were separately performed on the three segments, employing the ClustalW method in BioEdit [30] for sequence alignment. Phylogenetic trees were constructed by maximum likelihood (ML) method using IQtree-2.3.1 [31] comparison of the completed sequences, where the best fitting model for nucleotide substitutions was GTR+F+I+R3 for the L segment, GTR+F+I+R4 for the M segment, and TVMe+R for the S segment. The plausibility of the phylogenetic tree structure was evaluated through the application of the Bootstrap method, with 1,000 ultrafast bootstrap replicates.
The “ape” package in R 4.1.0 was utilized to conduct a comparative analysis of the topological structures of the phylogenetic trees of the L, M, and S segments, and the classification of genotypes was carried out based on the results of the comparative analysis. Since there is no standardized method for SFTSV genotype classification [16,17], we used the phylogenetic tree of the S segment as a reference and named the branches from top to bottom as genotypes A - E. Moreover, the intra-group and inter-group genetic distances of each virus genotype (S3 Table) were calculated by using MEGA 11.0 [32].
Recombination and reassortment analysis
Based on the genotype data obtained in the “phylogenetic analysis” section, we compared the three sequences of the same virus strain. If the genotypes inferred from the L, M and S gene segments are different, it is considered to be a potential reassortment strain.
The occurrence of recombination events within the SFTSV genome was subjected to analysis using Recombination Detection Program v4.101 (RDP4) package (https://h3abionet.org/categories/tools-services/recombination-detection-program). The occurrence of recombination breakpoints at the terminal position of a segment is defined as a recombination event [33]. Following a comparison of the three segments in Fasta format, the data were imported into RDP4 using the seven available methods: RDP, 3Seq, GENECONV, SiScan, Chimaera, Bootscan, and Maxchi. A recombination event was deemed authentic if it was identified by a minimum of three of the seven analytical techniques, with a p-value threshold of 0.05 (S5 Table). The recombinant strain’s associations with the major and minor parent were then observed. The aforementioned analytical techniques were integrated to ascertain the dependability of the recombination events indicated by the software and to formulate the ultimate conclusions. Following the identification of reassortment and recombination strains, a statistical analysis was conducted on the genotypes and geographic distribution of these strains to examine the patterns of emergence of SFTSV.
SFTSV discrete phylogeographic analysis
BEAST 2.4.4 [34] was employed to conduct a series of evolutionary analyses of SFTSV, encompassing the evolutionary rate of genes, time-of-origin analyses, and investigations into the place of origin and transmission pathways.
Given the probability of a recombination event in the S segment being minimal, it was selected for the construction of the Maximum Clade Credibility (MCC) tree and the analysis of geographic migration. On the basis of S1 Table, the sequences lacking collection locations and collection times as well as those with recombination events were removed, and the remaining sequences were organized to create S4 Table (498 sequences). The collection time and GenBank number of each sequence were incorporated into the sequence name, and the GTR+G+I nucleic acid substitution model, Relaxed Clock Log Normal molecular clock model, and Coalescent Constant Population species model were selected for the BEAST operation of the dataset. Bayesian Markov chain Monte Carlo analysis was run for 800 million steps, 10% of which were removed as burn-in and sampled every 80,000 steps to achieve parameter convergence and adequate effective sample sizes (ESS > 200). Sampling was conducted at 80,000-step intervals and documented in the 1og and tree files. Following the operation, the operation log file was opened with Tracer (version 1.7.2) (https://beast.community/tracer) to observe the effective sampling of the main parameters. The BEAST post-operation tree file was initially used with TreeAnnotator (v1.10.4) (https://beast.community/treeannotator) to integrate all the information, including all the branch lengths, branching times, posterior probabilities, and so forth, of all the sampled evolutionary trees into a single evolutionary tree. The burn-in percentage was selected as 10% to generate the MCC tree. The Bayes factors relevant to the transmission paths are presented in the S6 Table. A topological structure comparison between the MCC tree and the ML tree generated for the S segment using the “ape” package in R 4.1.0.
Furthermore, the propagation paths of the virus were rendered using SpreaD3 [35], then we merged the visualization results of the SFTSV migration and the migration routes of migratory birds onto a blank map for subsequent analysis.
Results
Phylogenetic and reassortment analysis
S1 Table included 485 SFTSV strains with full-length genome sequences from Japan, Korea, Thailand, and nine provinces in China: Liaoning, Beijing, Shandong, Henan, Zhejiang, Jiangxi, Hubei, Anhui, and Chinese Taiwan, as indicated by strain names in GenBank and related reports. The ML method was employed to construct three phylogenetic trees based on full-length S, M, and L segment sequences. Although the evolutionary positions of these trees differed in the S, M, and L segments, they exhibited a comparable topology, with the viral strains distinctly clustered into five branches (Fig 1). The five branches were designated as A, B, C, D, and E. A notable clustering of strains was observed, predominantly originating from Japan, Thailand, Korea, and Zhejiang Province, China (Fig 2). These strains exhibited a predominant distribution within genotypes A and D. The findings indicated that the strains from Zhejiang Province were more closely related to those from Japan, Korea, and Thailand in terms of evolutionary history.
A, B, and C were the topological comparison results of the S - M, M - L, and S - L segments of SFTSV, respectively. When the names of the leaf nodes were consistent, they were connected by blue dashed lines. Genotyping was carried out based on the topological comparison results, and types A - E were represented by different colors in the figure.
ML trees were constructed using S (A), M (B) and L (C) segment sequences and tested by bootstrap analysis over 1000 iterations. The five genotypes represented by cyan, green, purple, blue, and orange are classified as ABCDE genotypes, while black is designated as the unassigned type. The geographic distribution of these genotypes is illustrated by color blocks of varying hues. In instances where Liaoning province collected data exclusively with M segments and Jilin province collected data exclusively with L and S segments, these genotypes are represented by a single color to facilitate comparison.
Genotype A is mostly from China, such as Henan Province, with strains from Japan and Thailand. Genotype B exhibited disparate behavior across the three trees, comprising strains from Japan and Chinese Taiwan in the S-tree, strains from Japan and Hubei Province only in the M-tree, and strains from Hubei Province only in the L-tree. It is hypothesized that the primary cause of this phenomenon is the presence of incomplete gene sequences, and the secondary cause is the occurrence of reassortment events. Genotype C encompassed strains originating from Shandong Province and multiple other inland provinces within China. Genotype D was predominantly represented by the Japanese, Korean, Thailand, and Zhejiang Province strains. Genotype E represented the largest branch and encompasses the majority of strains originating from China’s inland provinces. It is notable that the majority of SFTSV strains (96.9%, 470/485) showed consistent genotype classification among the three genomic segments, whereas 15 strains (3.1%, 15/485) showed inconsistent genotype results, suggesting reassortment. Fig 3A illustrates the sampling locations of all strain sequences in S1 Table, as well as the distribution locations of reassorted strains. Of the 15 strains, 12 originated from central China (Henan, Hubei, and Jiangxi Provinces), two were from Japan, and the remaining one lacked geographical location information. Furthermore, we analyzed the data on common genotypes in each region, and the results are shown in Fig 3B. The genotypes of reassorted strains are illustrated in Fig 4. A total of 15 reassorted strains were classified into 10 types. The strains were named in the L-M-S order according to the genotypes of the three segments in their respective phylogenetic trees. These included AEE (with 3 sequences), AAB (with 2 sequences), DAD (with 2 sequences), EAA (with 2 sequences), AAC, BBE, CEE, ECC, EEC and a unique reassortment genotype (CGC, where G represents unassigned).
A shows the sampling locations of all strain sequences and the distribution locations of reassorted strains in S1 Table. We filled each province with orange, and the shade of the color indicates the quantity of sequences. B presents a pie chart illustrating the genotype percentage of sequences in L, M, and S in each region. It should be noted that only the M segment is available for the data collected in Liaoning Province, while only the L and S segments are available for the data collected in Jilin. The base layer of the map was made with Natural Earth (naturalearthdata.com), distributed openly under the following terms and conditions: https://www.naturalearthdata.com/about/terms-of-use/.
Each genotype is represented by a distinct color, with the L, M, and S segments indicated by varying line lengths. The black line represents unassigned. The name of each sequence with a potential reassortment event is provided below the corresponding graphic.
Recombination analysis
We analyzed the homologous recombination phenomenon in all sequences of S, M and L segments separately by RDP. Table 1 presents the host information for all sequences of the three segments, along with statistical data on the number of recombination events occurring in different hosts. Our findings indicated that the regions exhibiting the highest levels of recombination activity were widely distributed across both L and M. The comprehensive data set pertaining to all recombination events is presented in S7 Table, while Fig 5 illustrates the range of breakpoints where these events occurred across all segments.
The blue circle represents the initial position of the recombination breakpoint, the orange circle represents the final position of the recombination breakpoint, and the middle line represents the recombination range. The figure also provides the number of each recombination sequence and the corresponding GenBank number at the bottom.
In the S segment, there were only three sequences in which homologous recombination had occurred, all of which were from inland China. This indicated that during the evolutionary process, the S segment seldom relied on the mechanism of recombination which can rapidly introduce large-scale sequence variations.
A total of nine sequences underwent homologous recombination in all M segments analyzed, all from inland China, with most recombination occurring in the genotype C. Eight of these sequences were all typed the same as their major parental sequences, but one recombinant sequence, M2 (OQ388971), was typed the same as its minor parental sequences. M2 was from Hubei Province, with a Thailand strain in its minor parental sequences. M3 (OQ388989) was from Hubei Province, but had a Japanese virus sequence in its minor parental sequence. M5 (OM452787) was from Henan Province, but its major parental sequences had the involvement of Korean viral sequences, and the minor parental sequences were all from mainland China. M8 (OM452726) was also from Henan Province, but the Japanese viral sequences were found in its minor parental sequences.
A total of 12 sequences were subjected to homologous recombination in all L segments under analysis. Of these, only one sequence (L7:MT413432) originated from the coast of China, while the remainder were derived from inland China. L7 was a strain of Shandong Province, with a tick host and classified as type C as its major parental sequences. Its minor parental sequences demonstrated the involvement of Japan, Korea, and Thailand, and was classified as type D. Among the strains from inland, L4 (0M453367) exhibited greater specificity, with L4 originating from the Henan Province and its major parental host being ticks. The minor parental strain was derived from mainland China and Thailand.
Analysis of SFTSV time of origin and migration pathways
Least homologous recombination events were predicted within the S segment by the PHI test and RDP analysis, and its internal structure was more stable with smaller variability. The S segment was selected for discrete geographic system analysis. Consequently, we conducted an analysis of viral migration based on a comparison of the sequences of the S segments of SFTSV. The MCC tree was constructed using 498 full-length S segments (S4 Table), which included a substantial number of sequences isolated from patients and animal hosts.
The topology of this S-tree was found to be consistent with that of the S-tree constructed based on the ML method (S1 Fig), thereby providing further confirmation that SFTSV strains can be categorized into five genotypes. The MCC tree for all sampled SFTSV strains, as analyzed, supported a TMRCA of 1785 (95% HPD = 1625 to 1897). Genotype C differentiated around 1898 (95% HPD = 1841 to 1937), genotype D differentiated around 1868 (95% HPD = 1812 to 1917), genotype E differentiated around 1891(95% HPD = 1840 to 1934), and genotypes A and B, the latest of the five genotypes, differentiated around 1932 (95% HPD = 1888 to 1964). Although the E genotype emerged subsequently, it had the greatest number of sequences in the SFTSV gene database and was the most prevalent genotype in China.
The results of the SpreaD3 visualization were plotted on a map, with Fig 6B illustrating the transmission path of SFTSV from 1785 to 1980 and Fig 6C depicting the transmission path of SFTSV from 1980 to 2023. Prior to 1980, the transmission path of the virus was relatively straightforward, with Japan serving as the primary source of transmission and the virus spreading from Japan to South Korea, Chinese Taiwan, and Henan Province. However, after 1980, the transmission path exhibited a diversified trend, with the center of transmission shifting to the three regions of Japan, Henan and Hubei Provinces, and the virus spreading to the surrounding provinces in mainland China, with Henan and Hubei Provinces serving as the center.
A: Maximum clade credibility tree with branches colored by discrete geographic regions is presented. Color bars indicate hosts and subtypes. B: Discrete systematic geographic reconstruction of SFTSV propagation during 1785-1980. C: Discrete systematic geographic reconstruction of SFTSV propagation during 1980-2023. The direction of the arrows in B and C indicate the propagation routes, and the colors indicate the different export areas (the color scheme is consistent with that used in A). The mauve and light blue areas indicate the migratory routes of birds in central and eastern China. The base layer of the map was made with Natural Earth (naturalearthdata.com), distributed openly under the following terms and conditions: https://www.naturalearthdata.com/about/terms-of-use/.
Discussion
The phylogenetic analysis yielded evidence of at least five distinct genotypes, which is consistent with previous findings by other researchers in this field [17,19]. The phylogenetic trees established by L, M, and S, when analyzed together, exhibited a similar topology. The origin and number of strains in the four genotypes were essentially analogous, with the exception of the B genotype, which we postulated was attributable to the fact that the B genotype involved a smaller number of strains and the majority of them lacked intact S, M, and L segments. It was observed that clusters of sequences sampled from the same geographic location were interspersed with clusters of sequences sampled from other geographic locations within the same genotype. The results of the present study are consistent with those of previous studies in those strains from Japan, Thailand, South Korea, and Zhejiang Province, China, were clustered to form genotypes A and D [36]. This suggests a close relationship between strains from China, Japan, and South Korea. Further investigation is required to ascertain the underlying causes of this phenomenon. Furthermore, although the results suggest that Japan may be the origin of the virus, the variety of genotypes present in China (genotypes A-E) exceeds that of Japan (genotypes A, D). This discrepancy can be hypothesized to be due to China’s vast size, climatic and geographic diversity, and the widespread distribution of Haemaphysalis Iongicornis in China. These factors provide more space and opportunities for the evolution of the virus. Additionally, the free range farming model accounts for a large proportion in China’s animal husbandry, which increases the likelihood of ticks biting humans and livestock.
The phylogenetic results offer a hypothesis that reassortment events may be a potential driving force in the evolution of SFTSV. It is well established that the evolution of RNA viruses is profoundly affected by reassortment events. These events are distinct from homologous recombination and are widespread in viruses with segmented genomes, including orthomyxoviruses and bunyaviruses [37]. In instances where two or more segmented viruses simultaneously infect a single cell, the genomic segments may be randomly packaged into progeny viruses. The progeny may subsequently inherit genomic segments from multiple parental sources, thereby increasing the overall genetic variability. This process is fundamental to the survival and diversification of viruses [38]. The current study revealed the occurrence of reassortment events across all three segments of the SFTSV strain, resulting in the generation of ten distinct reassortment phenotypes. The L segment exhibited the highest frequency of reassortment events, while the M segment demonstrated the lowest. The typing of reassortment strains can be classified into two principal categories. The first category comprised type A, B, and D strains, which were predominantly found in Japan and Korea. The second category encompassed type C and E strains, which were primarily distributed in China. Our findings revealed instances where the two types of segments were observed in conjunction with each other in Central China. Recombination is a pivotal process in the evolutionary genetics of RNA viruses. However, it is commonly assumed to be uncommon in negative-strand RNA viruses [39]. Nevertheless, our findings indicated the presence of recombination in the L, M, and S segments of SFTSV, with the highest number of recombination events occurring in the L segment and the lowest in the S segment. It was noteworthy that all of the recombinant sequences originated from China, yet the major or minor parental sequences of some of the sequences were derived from Japan and Korea. The hosts of some of the recombinant sequences and their parents were identified as ticks, which suggests that recombination of SFTSV may occur in either humans or ticks. Furthermore, the majority of the data collected originated from human patients and ticks, with a smaller number of strains derived from other animals. Therefore, it is not possible to definitively exclude the possibility of recombination events of SFTSV in other animal species. The aforementioned phenomenon points to the migration of SFTSV over considerable distances between China and Japan and Korea. Additionally, it is evident that recombination and reassortment of virus segments have occurred in multiple regions, which may contribute to the expansion of genetic diversity and the accelerated evolution of SFTSV.
The temporal and spatial dynamics of RNA viruses are generally reflected in their phylogenetic structure [40]. Bayesian analysis showed that the time to most recent common ancestor (TMRCA) for all sampled SFTSV strains was 1785, we predict that genotype D is the earliest to differentiate. Genotypes A and B represent the most recent of the five genotypes to be differentiated, with differentiation occurring around 1932. Several scholars have previously suggested that the virus originated 50–150 years ago [17,18,24], and the data they used were generally from before 2015, while the data we used were from 2017-2023. These data are more accurate and have a wider range of hosts and geographical sources. Therefore, we believe that the reason for this discrepancy is mainly due to the bias in the gene sequence data.
The geographic migration analysis revealed that the dissemination of SFTSV could be delineated into two distinct phases: the initial period from 1785 to 1980, during which the virus was initially transmitted from Japan to China; the subsequent period from 1980 to 2023, during which the virus began to proliferate extensively across East Asia. In the initial phase, the transmission pathway of the virus exhibited a relatively homogeneous pattern, with Japan serving as the primary source of transmission. However, in the subsequent phase, there was a notable diversification in the transmission pathway, with the epicenter shifting to three distinct regions: Japan, Henan and Hubei Provinces. In general, it is our contention that SFTSV first originated in Japan and subsequently spread to Korea, Chinese Taiwan, and central China. From there, it is believed to have spread from central China to neighboring regions. The natural barrier of the ocean between countries such as China, Japan, and South Korea precludes direct transmission of the virus between countries. Transmission typically occurs indirectly through a specific host. Ticks serve as both key hosts and vectors of SFTSV, transmitting SFTSV and other tick-borne pathogens to other animals, including mammals, land birds, and seabirds, primarily through their bites [41–43]. Consequently, we posit that international travel and bird migration are likely to be significant conduits for SFTSV transmission.
The migratory routes of waterbirds in China’s wetlands are primarily classified into three distinct migratory routes: the eastern, central, and western migratory routes [44–46]. The Eastern Migratory Route constitutes a significant component of East Asian–Australasian Flyway. The migratory patterns of wetland waterbirds breeding in Russia, Japan, the Korean Peninsula, and northeastern and eastern North China indicate a north-south trajectory, with the majority of individuals traversing the eastern coast of China during the spring and fall seasons. In spring, the northward migrating birds arrive in Chinese Taiwan and divide into two branches, one spreading along the Chinese mainland or moving northward along the east coast, and the other going to Japan or continuing northward via the Ryukyu Islands. Wetland waterbirds migrating north along the east coast of mainland China are divided into two northward migration routes at the mouth of the Yangtze River. One goes through Jiangsu and Shandong Provinces to the northeast and Russia, while the other crosses the sea to the Korean Peninsula or Japan. In the fall, its southern migration route is roughly similar to the northern route in the spring. The Central Migration Route is a migration route within China where migratory birds fly south along the Yellow River Basin, Luliang Mountains, and Taihang Mountains in the fall to winter in central China or further south. Our findings revealed that four transmission routes, namely Japan-Korea, Japan-Chinese Taiwan, Japan-Inland China, and Korea-Inland China, exhibited a high degree of overlap with the East Asia-Australia migration routes. Meanwhile, the virus-endemic regions of inland provinces in China, including Henan, Hubei, Jiangxi, Anhui, and Beijing, demonstrated a markedly high degree of overlap with the Central China migration routes [41,47]. It is postulated that the eastern migratory route was of significant importance during the initial stages of virus transmission, acting as a principal conduit for the virus’s dispersal across the sea. The eastern and central migratory routes were similarly instrumental in the subsequent phases of virus transmission. It can be posited that migratory birds play a pivotal role in the dissemination of SFTSV, thereby offering an explanation for the observed spread of SFTSV in a number of coastal regions across East Asia. Furthermore, the impact of international travel, import and export trade, and other factors on the subject matter cannot be discounted.
In conclusion, a phylogenetic and Bayesian analysis of SFTSVs from East Asian countries was accomplished over the period spanning 2017 to 2023. Our study identified a greater number of recombination and reassortment events than ever before, which provides compelling evidence for the migration of the virus between countries in East Asia. Our findings indicate that the place of origin of SFTSV is Japan. Moreover, our analysis suggests that the spread of SFTSV can be divided into two distinct periods, with the migration routes of SFTSV changing over time. It is significant to acknowledge that migratory birds serve as a crucial vector for the transoceanic migration of SFTSV. These findings contribute to a more profound comprehension of the genesis and dissemination of SFTSV, and have significant ramifications for the prevention and management of SFTS.
Supporting information
S1 Table. The information of included sequences.
https://doi.org/10.1371/journal.ppat.1012970.s001
(DOCX)
S2 Table. Sequence data used in phylogenetic analysis.
https://doi.org/10.1371/journal.ppat.1012970.s002
(DOCX)
S3 Table. Genetic distance data of the phylogenetic trees of L, M and S segments.
https://doi.org/10.1371/journal.ppat.1012970.s003
(XLSX)
S4 Table. Sequence data of S segment used for constructing the Maximum Clade Credibility Tree.
https://doi.org/10.1371/journal.ppat.1012970.s004
(DOCX)
S5 Table. The information of the RDP4 homologous recombination assay.
https://doi.org/10.1371/journal.ppat.1012970.s005
(DOCX)
S6 Table. The information of the Bayes factors relevant to the transmission paths.
https://doi.org/10.1371/journal.ppat.1012970.s006
(XLSX)
S7 Table. SFTSV recombination events detected using the RDP package.
https://doi.org/10.1371/journal.ppat.1012970.s007
(XLSX)
S1 Fig. Comparison of the topological structures of the ML tree and the MCC tree of the S segment.
https://doi.org/10.1371/journal.ppat.1012970.s008
(TIF)
Acknowledgments
We would like to express our gratitude to all the data contributors (authors and their laboratories) who were responsible for obtaining specimens, generating gene sequences, and sharing them through GenBank.
References
- 1. Chen Z, Zhang J, Wang J, Tong H, Pan W, Ma F, et al. N6-methyladenosine RNA modification promotes Severe Fever with Thrombocytopenia Syndrome Virus infection. PLoS Pathog. 2024;20(11):e1012725. pmid:39585899
- 2. Li J-C, Zhao J, Li H, Fang L-Q, Liu W. Epidemiology, clinical characteristics, and treatment of severe fever with thrombocytopenia syndrome. Infect Med (Beijing). 2022;1(1):40–9. pmid:38074982; PMCID: PMC10699716
- 3. Liu R, He F, Chen S, Wang J, Yang C, Zhan Z, et al. Pathogen isolation and traceability analysis of a fatal case of severe fever with thrombocytopenia syndrome virus (SFTSV) infectious encephalitis in China. Virol J. 2024;21(1):300. pmid:39578877; PMCID: PMC11585235
- 4. Zhou C-M, Qi R, Qin X-R, Fang L-Z, Han H-J, Lei X-Y, et al. Oral and ocular transmission of severe fever with thrombocytopenia syndrome virus. Infect Med (Beijing). 2022;1(1):2–6. pmid:38074978; PMCID: PMC10699656
- 5. Yu X-J, Liang M-F, Zhang S-Y, Liu Y, Li J-D, Sun Y-L, et al. Fever with thrombocytopenia associated with a novel bunyavirus in China. N Engl J Med. 2011;364(16):1523–32. pmid:21410387; PMCID: PMC3113718
- 6. Xu M, Wang B, Deng F, Wang H, Wang M, Hu Z, et al. Establishment of a Reverse Genetic System of Severe Fever with Thrombocytopenia Syndrome Virus Based on a C4 Strain. Virol Sin. 2021;36(5):958–67. pmid:33721215; PMCID: PMC8558135
- 7. Luo N, Li M, Xu M, Shi C, Shi X, Ni R, et al. Research Progress of Fever with Thrombocytopenia Syndrome. Intensive Care Res. 2023;1–10. pmid:37360310; PMCID: PMC10033304
- 8. Yu X-J. Risk factors for death in severe fever with thrombocytopenia syndrome. Lancet Infect Dis. 2018;18(10):1056–7. pmid:30054189
- 9. Chen QL, Zhu MT, Chen N, Yang D, Yin WW, Mu D, et al. Epidemiological characteristics of severe fever with thtrombocytopenia syndrome in China, 2011-2021. Zhonghua Liu Xing Bing Xue Za Zhi. 2022;43(6):852–9. pmid:35725341
- 10. Kim K-H, Yi J, Kim G, Choi SJ, Jun KI, Kim N-H, et al. Severe fever with thrombocytopenia syndrome, South Korea, 2012. Emerg Infect Dis. 2013;19(11):1892–4. pmid:24206586; PMCID: PMC3837670
- 11. Takahashi T, Maeda K, Suzuki T, Ishido A, Shigeoka T, Tominaga T. The first identification and retrospective study of Severe Fever with Thrombocytopenia Syndrome in Japan. J Infect Dis. 2014;209(6):816–27. pmid:24231186; PMCID: PMC7107388
- 12. Saba Villarroel PM, Chaiphongpachara T, Nurtop E, Laojun S, Pangpoo-Nga T, Songhong T, et al. Seroprevalence study in humans and molecular detection in Rhipicephalus sanguineus ticks of severe fever with thrombocytopenia syndrome virus in Thailand. Sci Rep. 2024;14(1):13397. pmid:38862576; PMCID: PMC11167008
- 13. Wang Y, Pang B, Wang Z, Tian X, Xu X, Chong X, et al. Genomic diversity and evolution analysis of severe fever with thrombocytopenia syndrome in East Asia from 2010 to 2022. Front Microbiol. 2023;14:1233693. pmid:37670982; PMCID: PMC10476882
- 14. Zhang Z, Tan J, Jin W, Qian H, Wang L, Zhou H, et al. Severe fever with thrombocytopenia syndrome virus trends and hotspots in clinical research: A bibliometric analysis of global research. Front Public Health. 2023;11:1120462. pmid:36817929; PMCID: PMC9933999
- 15. Petruzziello A, Marigliano S, Loquercio G, Cozzolino A, Cacciapuoti C. Global epidemiology of hepatitis C virus infection: An up-date of the distribution and circulation of hepatitis C virus genotypes. World J Gastroenterol. 2016;22(34):7824–40. pmid:27678366; PMCID: PMC5016383
- 16. Yoshikawa T, Shimojima M, Fukushi S, Tani H, Fukuma A, Taniguchi S, et al. Phylogenetic and Geographic Relationships of Severe Fever With Thrombocytopenia Syndrome Virus in China, South Korea, and Japan. J Infect Dis. 2015;212(6):889–98. pmid:25762790
- 17. Liu J-W, Zhao L, Luo L-M, Liu M-M, Sun Y, Su X, et al. Molecular Evolution and Spatial Transmission of Severe Fever with Thrombocytopenia Syndrome Virus Based on Complete Genome Sequences. PLoS One. 2016;11(3):e0151677. pmid:26999664; PMCID: PMC4801363
- 18. Fu Y, Li S, Zhang Z, Man S, Li X, Zhang W. Phylogeographic analysis of severe fever with thrombocytopenia syndrome virus from Zhoushan Islands, China: implication for transmission across the ocean. Scientific Reports. 2016;6:19563. pmid:26806841; PMCID: PMC4726339
- 19. Liu L, Chen W, Yang Y, Jiang Y. Molecular evolution of fever, thrombocytopenia and leukocytopenia virus (FTLSV) based on whole-genome sequences. Infect Genet Evol. 2016;39:55–63. pmid:26748010
- 20. Liu B, Zhu J, He T, Zhang Z. Genetic variants of Dabie bandavirus: classification and biological/clinical implications. Virol J. 2023;20(1):68. pmid:37060090; PMCID: PMC10103499
- 21. Wu X, Li M, Zhang Y, Liang B, Shi J, Fang Y, et al. Novel SFTSV phylogeny reveals new reassortment events and migration routes. Virology Sinica. 2021;36(2):300–10. pmid:32960400; PMCID: PMC8087752
- 22. Shi J, Hu S, Liu X, Yang J, Liu D, Wu L, et al. Migration, recombination, and reassortment are involved in the evolution of severe fever with thrombocytopenia syndrome bunyavirus. Infect Genet Evol. 2017;47:109–17. pmid:27884653
- 23. Ding N-Z, Luo Z-F, Niu D-D, Ji W, Kang X-H, Cai S-S, et al. Identification of two severe fever with thrombocytopenia syndrome virus strains originating from reassortment. Virus Res. 2013;178(2):543–6. pmid:24055465
- 24. Lam TT-Y, Liu W, Bowden TA, Cui N, Zhuang L, Liu K, et al. Evolutionary and molecular analysis of the emergent severe fever with thrombocytopenia syndrome virus. Epidemics. 2013;5(1):1–10. pmid:23438426; PMCID: PMC4330987
- 25. Lv Q, Zhang H, Tian L, Zhang R, Zhang Z, Li J, et al. Novel sub-lineages, recombinants and reassortants of severe fever with thrombocytopenia syndrome virus. Ticks Tick Borne Dis. 2017;8(3):385–90. pmid:28117273
- 26. Gong X, Hu M, Wang B, Yang H, Jin Y, Liang L, et al. A Non-phylogeny-dependent Reassortment Detection Method for Influenza A Viruses. Frontiers in Virology.
- 27. Lee K, Seok J, Kim H, Park S, Lee S, Bae J. Genome-informed investigation of the molecular evolution and genetic reassortment of severe fever with thrombocytopenia syndrome virus. PLoS Negl Trop Dis. 2023;17(9):e0011630. pmid:37713429; PMCID: PMC10529592
- 28.
Aqian L, Lin L, Wu W, Liu Y, Huang X, et al. “Molecular evolution and genetic diversity analysis of SFTS virus based on next-generation sequencing.” 2021.
- 29. Fu L, Niu B, Zhu Z, Wu S, Li W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics. 2012;28(23):3150–2. pmid:23060610; PMCID: PMC3516142
- 30. Alzohairy AM. BioEdit: An important software for molecular biology. Gerf Bulletin of Biosciences. 2011;2(1):60–1.
- 31. Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, von Haeseler A, et al. IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol. 2020;37(5):1530–4. pmid:32011700 Erratum in: IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol. 2020;37(8):2461.
- 32. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Molecular Biology and Evolution. 2018;35(6):1547–9. pmid:29722887; PMCID: PMC5967553
- 33. Nagarajan N, Kingsford C. GiRaF: robust, computational identification of influenza reassortments via graph mining. Nucleic Acids Res. 2011;39(6):e34. pmid:21177643; PMCID: PMC3064795
- 34. Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu C-H, Xie D, et al. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput Biol. 2014;10(4):e1003537. pmid:24722319; PMCID: PMC3985171
- 35. Bielejec F, Baele G, Vrancken B, Suchard MA, Rambaut A, Lemey P. SpreaD3: Interactive Visualization of Spatiotemporal History and Trait Evolutionary Processes. Mol Biol Evol. 2016;33(8):2167–9. pmid:27189542; PMCID: PMC6398721
- 36. Rattanakomol P, Khongwichit S, Chuchaona W, Vongpunsawad S, Poovorawan Y. Severe fever with thrombocytopenia syndrome virus genotype B in Thailand. Arch Virol. 2023;168(11):271. pmid:37833439
- 37. He C-Q, Ding N-Z. Discovery of severe fever with thrombocytopenia syndrome bunyavirus strains originating from intragenic recombination. J Virol. 2012;86(22):12426–30. pmid:22933273; PMCID: PMC3486477
- 38. McDonald S, Nelson M, Turner P, Patton J. Reassortment in segmented RNA viruses: mechanisms and outcomes. Nat Rev Microbiol. 2016;14(7):448–60. pmid:27211789; PMCID: PMC5119462
- 39. Han G-Z, Worobey M. Homologous recombination in negative sense RNA viruses. Viruses. 2011;3(8):1358–73. pmid:21994784; PMCID: PMC3185808
- 40. Wang LQ, Xiong HY, Wang WB. Application of phylogenetic analysis in the molecular epidemiological study of infectious diseases. Zhonghua Liu Xing Bing Xue Za Zhi. 2022;43(2):282–5. pmid:35184497
- 41. Kim E-H, Park S-J. Emerging Tick-Borne Dabie bandavirus: Virology, Epidemiology, and Prevention. Microorganisms. 2023;11(9):2309. pmid:37764153; PMCID: PMC10536723
- 42. Oh S-S, Chae J-B, Kang J-G, Kim H-C, Chong S-T, Shin J-H, et al. Detection of Severe Fever with Thrombocytopenia Syndrome Virus from Wild Animals and Ixodidae Ticks in the Republic of Korea. Vector Borne Zoonotic Dis. 2016;16(6):408–14. pmid:27043361
- 43. Li Z, Bao C, Hu J, Liu W, Wang X, Zhang L, et al. Ecology of the tick-borne phlebovirus causing severe fever with thrombocytopenia syndrome in an endemic area of China. PLoS Negl Trop Dis. 2016;10(4):e0004574. pmid:27035712; PMCID: PMC4818090
- 44. Shi X, Hu C, Soderholm J, Chapman J, Mao H, Cui K, et al. Prospects for monitoring bird migration along the East Asian‐Australasian Flyway using weather radar. Remote Sens Ecol Conserv. 2022;9(2):169–81.
- 45. Gu Z, Pan S, Lin Z, Hu L, Dai X, Chang J, et al. Climate-driven flyway changes and memory-based long-distance migration. Nature. 2021;591(7849):259–64. pmid:33658718
- 46. Cornell Lab of Ornithology. [website]. Available at: https://www.birds.cornell.edu/home/ [cited 2024 Dec 28. ]
- 47. Zhang X, Zhao C, Cheng C, Zhang G, Yu T, Lawrence K, et al. Rapid Spread of Severe Fever with Thrombocytopenia Syndrome Virus by Parthenogenetic Asian Longhorned Ticks. Emerg Infect Dis. 2022;28(2):363–72. pmid:35075994; PMCID: PMC8798674