Full Genome Characterization of Novel DS-1-Like G8P[8] Rotavirus Strains that Have Emerged in Thailand: Reassortment of Bovine and Human Rotavirus Gene Segments in Emerging DS-1-Like Intergenogroup Reassortant Strains

The emergence and rapid spread of unusual DS-1-like intergenogroup reassortant rotavirus strains have been recently reported in Asia, Australia, and Europe. During rotavirus surveillance in Thailand in 2013–2014, novel DS-1-like intergenogroup reassortant strains having G8P[8] genotypes (i.e., strains KKL-17, PCB-79, PCB-84, PCB-85, PCB-103, SKT-107, SWL-12, NP-130, PCB-656, SKT-457, SSKT-269, and SSL-55) were identified in stool samples from hospitalized children with severe diarrhea. In this study, we determined and characterized the complete genomes of these 12 strains (seven strains, KKL-17, PCB-79, PCB-84, PCB-85, PCB-103, SKT-107, and SWL-12, found in 2013 (2013 strains), and five, NP-130, PCB-656, SKT-457, SSKT-269, and SSL-55, in 2014 (2014 strains)). On full genomic analysis, all 12 strains showed a unique genotype constellation comprising a mixture of genogroup 1 and 2 genes: G8-P[8]-I2-R2-C2-M2-A2-N2-T2-E2-H2. With the exception of the G genotype, the unique genotype constellation of the 12 strains (P[8]-I2-R2-C2-M2-A2-N2-T2-E2-H2) was found to be shared with DS-1-like intergenogroup reassortant strains. On phylogenetic analysis, six of the 11 genes of the 2013 strains (VP4, VP2, VP3, NSP1, NSP3, and NSP5) appeared to have originated from DS-1-like intergenogroup reassortant strains, while the remaining four (VP7, VP6, VP1, and NSP2) and one (NSP4) gene appeared to be of bovine and human origin, respectively. Thus, the 2013 strains appeared to be reassortant strains as to DS-1-like intergenogroup reassortant, bovine, bovine-like human, and/or human rotaviruses. On the other hand, five of the 11 genes of the 2014 strains (VP4, VP2, VP3, NSP1, and NSP3) appeared to have originated from DS-1-like intergenogroup reassortant strains, while three (VP7, VP1, and NSP2) and one (NSP4) were assumed to be of bovine and human origin, respectively. Notably, the remaining two genes, VP6 and NSP5, of the 2014 strains appeared to have originated from locally circulating DS-1-like G2P[4] human rotaviruses. Thus, the 2014 strains were assumed to be multiple reassortment strains as to DS-1-like intergenogroup reassortant, bovine, bovine-like human, human, and/or locally circulating DS-1-like G2P[4] human rotaviruses. Overall, the great genomic diversity among the DS-1-like intergenogroup reassortant strains seemed to have been generated through additional reassortment events involving animal and human strains. Moreover, all the 11 genes of three of the 2014 strains, NP-130, PCB-656, and SSL-55, were very closely related to those of Vietnamese DS-1-like G8P[8] strains that emerged in 2014–2015, indicating the derivation of these DS-1-like G8P[8] strains from a common ancestor. To our knowledge, this is the first report on full genome-based characterization of DS-1-like G8P[8] strains that have emerged in Thailand. Our observations will add to our growing understanding of the evolutionary patterns of emerging DS-1-like intergenogroup reassortant strains.


Introduction
Group A rotaviruses (RVAs), members of the Reoviridae family, are the leading pathogens causing diarrhea in young children and many animal species worldwide. In humans, RVA disease is associated with high morbidity and mortality, being responsible for an estimated 453,000 deaths among children <5 years of age annually [1], with more than half of the deaths occurring in developing countries in Asia and Africa [1,2]. The RVA virion is a triple-layered, non-enveloped icosahedron with an 11-segment genome of double-stranded (ds)RNA, encoding six structural proteins (VP1-4, VP6, and VP7) and six non-structural proteins (NSP1-6) [3]. The segmented nature of the genome facilitates reassortment between/within human and animal strains, and the reassortment plays one of the major roles in the generation of the genomic diversity of RVAs [4].
Full genome-based analysis is a reliable method for obtaining precise information on the origin of an RVA strain, and for tracing its evolutionary dynamics [15,17,19,27]. To date, however, the full genome sequences of only two DS-1-like intergenogroup reassortant strains having G8P [8] genotypes, strain RVN1149 from Vietnam and strain DRC88 from the Democratic Republic of Congo, have been sequenced and characterized, indicating the presence of bovine-like segments on a DS-1-like genetic backbone [25,32]. However, because the overall genomic constellation and genomic diversity of DS-1-like G8P [8] strains remain to be elucidated, full genomic analysis of Thai DS-1-like G8P [8] strains might be useful for obtaining a more precise understanding of the evolutionary dynamics of the emerging DS-1-like intergenogroup reassortant strains involving DS-1-like G8P [8] HuRVAs. In the present study, we analyzed the full genomes of 12 DS-1-like G8P [8] strains that have emerged in Thailand. Furthermore, the full genomes of two locally circulating strains (DS-1-like G2P [4] HuRVAs) were also sequenced as references.

Ethics Statement
The study was approved by the Ethical Review Committee for Research on Human Subjects of the Ministry of Public Health, Thailand (Ref. no. 10/2555). In this study, written informed consent for the testing of stool specimens for rotaviruses and characterization of detected RVA strains was obtained from the children's parents/guardians.

Viral dsRNA Extraction
Extraction of viral dsRNAs was performed as described previously [14,16,33]. Briefly, viral dsRNAs were extracted from stool suspensions using a QIAamp Viral RNA Mini Kit (Qiagen). The extracted viral dsRNAs were subjected to Illumina MiSeq sequencing for full genomic analysis as described below.

cDNA Library Preparation and Illumina MiSeq Sequencing
Building of a cDNA library and Illumina MiSeq sequencing were performed as described previously [14,16,30,33,34]. Briefly, a 200 bp fragment library ligated with bar-coded adapters was constructed for the 14 strains using an NEBNext Ultra RNA Library Prep Kit for Illumina v1.2 (New England Biolabs), and NEBNext Multiplex Oligos for Illumina (New England Biolabs) according to the manufacturer's instructions. The cDNA library was purified using Agencourt AMPure XP magnetic beads (Beckman Coulter). After assessing the quality and quantity of the purified cDNA library, nucleotide sequencing was performed on an Illumina MiSeq sequencer (Illumina) using a MiSeq Reagent Kit v2 (Illumina) to generate 151 paired-end reads. Data analysis was performed using CLC Genomics Workbench v8.0.1 (CLC Bio). Contigs were assembled from the obtained sequence reads (trimmed) by de novo assembly. Using the assembled contigs as query sequences, the Basic Local Alignment Search Tool (BLAST) non-redundant nucleotide database was searched to obtain the full-length nucleotide sequence of each gene segment of the 14 strains. The nucleotide sequences were translated into amino acid sequences using GENETYX v11 (GENETYX).

Determination of RVA Genotypes
The genotype of each of the 11 gene segments of the 14 strains was determined using the RotaC v2.0 automated genotyping tool (http://rotac.regatools.be/) [35] according to the guidelines proposed by the Rotavirus Classification Working Group (RCWG).

Phylogenetic Analysis
Sequence comparisons were performed as described previously [14,16,33]. Briefly, multiple alignment of each viral segment was carried out using CLUSTAL W. Phylogenetic trees were constructed using the maximum likelihood method and the Tamura-Nei substitution model using MEGA6.06 [36]. The reliability of the branching order was estimated from 1000 bootstrap replicates. The results of phylogenetic analysis were validated using several other genetic distance models, such as the Hasegawa-Kishino-Yano, Jukes-Cantor, Kimura 2-parameter, and Tamura 3-parameter ones, ruling out any biases among different models concerning the study strains (data not shown).

Nucleotide Sequence Accession Numbers
The nucleotide sequence data presented in this manuscript have been deposited in the DDBJ and EMBL/GenBank data libraries. The accession numbers for the nucleotide sequences of the VP1-4, VP6, VP7, and NSP1  Table).

Phylogenetic Analyses
We next constructed phylogenetic trees using the full-length gene sequence for each of the 11 genes because phylogenetic analysis of RVA nucleotide sequences makes it possible to obtain direct evidence of their relatedness to those of other RVA strains, even within the same genotype [17,19,26] (Figs 2-14).
The VP7 genes of all 12 strains exhibited the highest nucleotide sequence identities (99.8-100%) with that of Thai bovine-like human strain CU-B1715 (G8P [8]) [37] or Vietnamese DS-1-like G8P [8] strain RVN1149. On phylogenetic analysis, the 12 strains were shown to form a cluster with the above-mentioned bovine-like human G8P [8] strains from Southeast Asia in a common branch with several bovine and bovine-like human strains from Asia in G8 lineage-5, in which the majority of Asian bovine and bovine-like human strains cluster [38] (Fig 2).
The VP4 genes of all 12 strains exhibited the maximum nucleotide sequence identities (99.7-100%) with the cognate genes of Thai DS-1-like G2P [8] strain LS-04 and/or Vietnamese DS-1-like G8P [8] strain RVN1149. On phylogenetic analysis, the 12 strains were shown to form a cluster with these Southeast Asian DS-1-like intergenogroup reassortant strains and several locally circulating strains having a DS-1-like or Wa-like genetic backbone in humanlike P [8] lineage-3 (Fig 4).
The VP6 genes of the 2013 strains (KKL-17, PCB-79, PCB-84, PCB-85, PCB-103, SKT-107, and SWL-12) showed the maximum nucleotide sequence identities (94.8-94.9%) with the VP6 genes of Indian bovine-like porcine strains HP113 (G6P [13]) and HP140 (G6P [13]) [39,40]. Phylogenetically, the 2013 strains were found to form a cluster near the cluster comprising strains HP113 and HP140, and several bovine and bovine-like human strains within the bovine-like I2 subcluster (Fig 6). On the other hand, the VP6 genes of the 2014 strains (NP-130, PCB-656, SKT-457, SSKT-269, and SSL-55) exhibited the highest nucleotide sequence identities (99.6-99.8%) with Vietnamese DS-1-like G8P [8] strain RVN1149. Phylogenetically, LS-L7, reported in this and the preceding studies [19]. Gray shading indicates the gene segments with genotypes identical to those of the 12 strains. "−" indicates that no sequence data were available in the DDBJ and EMBL/ GenBank data libraries. a The gene segments that are most similar to those of strain  The gene segments that are most similar to those of strain PCB-79. c The gene segments that are most similar to those of strain PCB-84. d The gene segments that are most similar to those of strain PCB-85. e The gene segments that are most similar to those of strain PCB-103. f The gene segments that are most similar to those of strain SKT-107. g The gene segments that are most similar to those of strain SWL-12. h The gene segments that are most similar to those of strain NP-130. i The gene segments that are most similar to those of strain PCB-656. j The gene segments that are most similar to those of strain SKT-457. k The gene segments that are most similar to those of strain SSKT-269. l The gene segments that are most similar to those of strain SSL-55.  [19], and Thai DS-1-like G2P [8] strain LS-04 within the human-like I2 subcluster (Fig 6).
The NSP1 genes of all 12 strains showed the maximum nucleotide sequence identities (99.8-99.9%) with those of Thai DS-1-like G2P [8] strain LS-04 and/or Vietnamese DS-1-like G8P [8]   Emergence and Evolution of Novel DS-1-Like G8P [8] Human Rotaviruses in Thailand strain RVN1149. Phylogenetically, the 12 strains were shown to form a cluster with these Southeast Asian DS-1-like intergenogroup reassortant strains near the cluster with the other DS-1-like intergenogroup reassortant strains within the human-like A2 subcluster (Fig 10).
The NSP2 genes of all 12 strains showed the maximum nucleotide sequence identities (99.7-99.9%) with the cognate genes of Thai DS-1-like G2P [8] strain LS-04 and/or Vietnamese DS-1-like G8P [8] strain RVN1149. Phylogenetically, the 12 strains were found to form a distinct cluster with these Southeast Asian DS-1-like intergenogroup reassortant strains in a common branch with several bovine and bovine-like donkey strains from South Africa within the bovine-like N2 subcluster (Fig 11).

Discussion
In the present study, we analyzed the full genomes of 12 DS-1-like intergenogroup reassortant strains having G8P [8] genotypes (seven strains in 2013 and five strains in 2014) identified in stool specimens from hospitalized children with severe diarrhea in Thailand. All 12 strains exhibited a unique genotype constellation comprising a mixture of genogroup 1 and 2 genes: G8-P[8]-I2-R2-C2-M2-A2-N2-T2-E2-H2. With the exception of the G genotype, the unique genotype constellation of the 12 strains (P[8]-I2-R2-C2-M2-A2-N2-T2-E2-H2) is commonly shared with DS-1-like intergenogroup reassortant strains. On phylogenetic analysis, six of the 11 genes of the 2013 strains (VP4, VP2, VP3, NSP1, NSP3, and NSP5) appeared to have originated from DS-1-like intergenogroup reassortant strains, while the remaining four (VP7, VP6, VP1, and NSP2) and one (NSP4) were assumed to be of bovine and human origin, respectively. Therefore, the 2013 strains seemed to have been derived through reassortment involving DS-1-like intergenogroup reassortant, bovine, bovine-like human, and/or human rotaviruses. On the other hand, five of the 11 genes of the 2014 strains (VP4, VP2, VP3, NSP1, and NSP3) were assumed to have originated from DS-1-like intergenogroup reassortant strains, while three (VP7, VP1, and NSP2) and one (NSP4) appeared to be of bovine and human origin, respectively. Of note is that the remaining VP6 and NSP5 genes of the 2014 strains were assumed to have originated from co-circulating DS-1-like G2P [4] HuR-VAs. Therefore, the 2014 strains appeared to be multiple reassortment strains involving DS-1-like intergenogroup reassortant, bovine, bovine-like human, human, and/or locally circulating DS-1-like G2P [4] rotaviruses. These results might imply the occurrence of reassortment between the 2013 strains and locally circulating DS-1-like G2P [4] strains to form the 2014 strains having the VP6 and NSP5 genes derived from co-circulating DS-1-like G2P [4] rotaviruses. Overall, the great genomic diversity among the DS-1-like intergenogroup reassortant strains seemed to have been generated through additional reassortment involving animal and human strains. However, the exact origins of the bovine-like VP7, VP6, VP1, and/or NSP2 genes, and human-like NSP4 genes of the 12 strains could not be ascertained due to a lack of a sufficient number of representative bovine-like human, bovine, and human strains as references. In any case, our findings add the increasing evidence supporting animal-to-human interspecies transmission through reassortment [7]. The bovine origin of the 12 strains suggests interspecies transmission due to the close proximity between humans and livestock, especially in developing countries in Asia, where there is intimate contact between humans and livestock such as cattle in daily life [14].  Emergence and Evolution of Novel DS-1-Like G8P [8] Human Rotaviruses in Thailand Emergence and Evolution of Novel DS-1-Like G8P [8]

Human Rotaviruses in Thailand
Of note is that all the 11 segments of three of the 2014 strains, NP-130, PCB-656, and SSL-55, were very closely related to those of Vietnamese DS-1-like G8P [8] strains that emerged in 2014-2015, indicating the derivation of these DS-1-like G8P [8] strains from a common ancestor. In contrast, the low genomic correlation between Asian DS-1-like G8P [8] strains and African ones indicates the distinct evolution of DS-1-like G8P [8] strains in Asia and Africa. However, a global rotavirus strain collection is required to determine the exact evolutionary dynamics of the emerging DS-1-like G8P [8] strains in Asia. To our knowledge, this is the first report on full genome-based analysis of DS-1-like G8P [8] strains that have emerged in Thailand. It is essential to continue monitoring of the emergence of any uncommon RVA variants such as DS-1-like intergenogroup reassortant strains in order to correlate the relationship with the current introduction of HuRVA vaccines [19].
The emergence of DS-1-like G8P [8] strains in Thailand may imply the constant circulation of DS-1-like intergenogroup reassortant strains and the occurrence of reassortment involving them in Asia. Because DS-1-like intergenogroup reassortant strains have successfully spread in multiple locations in Asia, Australia, and Europe in a short period [19,[21][22][23][24][25][26][27], continued monitoring of DS-1-like intergenogroup reassortant strains is required. Although most investigations on RVA genotype distributions have been focused on only G/P defining genes, PCRbased genotyping for non-G/P defining gene(s) or PAGE analysis will assist the recognition of novel RVA variants such as DS-1-like intergenogroup reassortant strains, as described in the preceding study [19]. Furthermore, full genome-based analysis is a reliable tool to understand the emergence and evolution of DS-1-like intergenogroup reassortant strains [19].
Supporting Information S1 Table. The