Surveillance of Human Rotaviruses in Wuhan, China (2019–2022): Whole-Genome Analysis of Emerging DS-1-like G8P[8] Rotavirus

Group A rotaviruses (RVAs) are major etiologic agents of gastroenteritis in infants and young children worldwide. To study the prevalence and genetic characteristics of RVAs, a hospital-based surveillance study was conducted in Wuhan, China from June 2019 through May 2022. The detection rates of RVAs were 19.40% (142/732) and 3.51% (8/228) in children and adults, respectively. G9P[8] was the predominant genotype, followed by G8P[8] and G3P[8]. G8P[8] emerged and was dominant in the 2021–2022 epidemic season. The genome constellation of six G8P[8] strains was assigned to G8-P[8]-I2-R2-C2-M2-A2-N2-T2-E2-H2. Phylogenetic analysis revealed that the VP7, VP4, VP2, VP3, NSP1, NSP2, NSP3, and NSP5 genes of these G8P[8] strains clustered closely with those of the G8P[8] strains in Asia and were distant from those of the P[8] and G2P[4] strains simultaneously detected in Wuhan. In contrast, the VP1, VP6, and NSP4 genes were closely related to the typical G2P[4] rotavirus, including those of G2P[4] strains simultaneously detected in Wuhan. The detection rate of RVAs decreased in the COVID-19 pandemic era. It was deduced that the G8P[8] rotaviruses that emerged in China may be reassortants, carrying the VP6, VP1, and NSP4 genes derived from the G2P[4] rotavirus in the backbone of the neighboring DS-1-like G8P[8] strains represented by CAU17L-103.


Introduction
Group A rotaviruses (RVAs) are the leading cause of severe diarrhea in children less than 5 years of age globally, being responsible for an estimated 128,500 deaths in 2016 alone [1]. More than 65% of these deaths are estimated to occur in 11 countries in Asia and Africa [2,3]. Vaccination is an essential strategy for the prevention and control of severe disease caused by rotavirus infection. Until now, a total of four vaccines have been introduced by WHO to the world [4].
According to the International Committee on the Taxonomy of Viruses (ICTV), rotaviruses are 11-segmented double-stranded RNA viruses that can be classified into nine species (from A to J) belonging to the genus Rotavirus in the family Sedoreoviridae. Rotaviruses possess six structural viral proteins (VP1-VP4, VP6, and VP7) and six nonstructural proteins (NSP1-NSP5/6) [5]. Each gene is differentiated into distinct genotypes according to a predefined nucleotide sequence identity. The outer layer protein, VP7, and spike protein, VP4, of rotaviruses are neutralization antigens, determining the virus 3 of 21 respectively. In phase 3, the detection rates of RVAs were 21.3% (85/400) and 6.5% (4/62) from children and adults, respectively.
The VP4 genes of the six G8 strains in the present study and four simultaneous cocirculating G1 and G9 strains were grouped into the P [8] genotype. The six G8 RVAs were closely related to those of the G8P [8] strains from China and neighboring countries, such as GZ-0005, SO1162, CAU17L-103, and PCB656, showing a minimum nucleotide identity of 98.1%, and they belonged to lineage "a" (Figure 2, P[8]a; Table 2) and were distant from the co-circulating G1P [8] and G9P [8] strains, with nucleotide identities of 95.7-97.2%. The VP4 genes of E6894 and E6896 were assigned to the P [4] genotype and were close to other G2P [4] strains found in Asia, Europe, Africa, and America during the years of 2005-2018. - All strains detected in the present study are indicated by a yellow background and bold, including their genotypes same as the emerging DS-1-like G8P [8] rotaviruses.
as GZ-0005, SO1162, CAU17L-103, and PCB656, showing a minimum nuc of 98.1%, and they belonged to lineage "a" (Figure 2, P[8]a; Table 2) and we the co-circulating G1P [8] and G9P [8] strains, with nucleotide identities of 9 VP4 genes of E6894 and E6896 were assigned to the P [4] genotype and wer G2P [4] strains found in Asia, Europe, Africa, and America during the year Figure 2. Phylogenetic dendrogram based on complete coding regions of the VP sentative RVAs. The best nucleotide substitution model was T92 + G. Bootstrap v are not shown. The G8P [8] strains are highlighted with a filled circle. The contemp G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" rep of G8P [8] within P [8].
The VP6 genes of the six G8 strains in the present study were assigne otype. These genes are closely related to two Chinese G8P [8] strains from G and the G8P[4] strain GER1H-09 from Germany, as well as the G2P [4] stra American strain LB2772 and the Chinese simultaneously co-circulating E6896, in the present study, with a minimum nucleotide identity of 97.7% The best nucleotide substitution model was T92 + G. Bootstrap values below 80% are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" represents a lineage of G8P [8] within P [8].
The VP6 genes of the six G8 strains in the present study were assigned to the I2 genotype. These genes are closely related to two Chinese G8P [8] strains from Guangzhou City and the G8P [4] strain GER1H-09 from Germany, as well as the G2P [4] strains, such as the American strain LB2772 and the Chinese simultaneous co-circulating G2P [4] strain E6896, in the present study, with a minimum nucleotide identity of 97.7%, while remaining distant from the G8 [8] strains found in Asia and Europe, with a maximum nucleotide identity of 96.6% (Figure 3, I2a and I2b; Table 2). The VP6 genes of the simultaneous co-circulating G1 and G9 strains were assigned to the I1 genotype ( Figure 3).  Table 2). The VP6 genes of the simultaneously cocirculating G1 and G9 strains were assigned to the I1 genotype ( Figure 3). are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represent lineages within I2 (G8P [8] strains, lineage "a").
The VP1 genes of all six strains in the present study were grouped into the R2 genotype. These VP1 genes clustered closely with those of two G8P [8] strains from Guangzhou City in China, the simultaneously circulating G2P [4] strains (E6896 and E6894) in Wuhan, and the G2P [4]   The best nucleotide substitution model was T92 + G. Bootstrap values below 80% are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represent lineages within I2 (G8P [8] strains, lineage "a").
The VP1 genes of all six strains in the present study were grouped into the R2 genotype. These VP1 genes clustered closely with those of two G8P [8] strains from Guangzhou City in China, the simultaneous circulating G2P [4] strains (E6896 and E6894) in Wuhan, and the G2P [4]   The best nucleotide substitution model was GTR + G + I. Bootstrap values below 80% are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represent lineages within R2 (G8P [8] strains, lineage "a") The VP2 and VP3 genes of all six G8P [8] strains in the present study were assigned to the C2 and M2 genotypes, respectively, and clustered closely with those of the G8P  Table 2). The VP2 and VP3 genes of the six strains mentioned above were distant from the simultaneously co-circulating Chinese G2P [4] strains E6894 and E6896 detected in Wuhan, with nucleotide identities of 87.1-97.5% ( Figure 5, C2, Figure 6, M2, Table 2). are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represent lineages within R2 (G8P [8] strains, lineage "a").
The VP2 and VP3 genes of all six G8P [8] strains in the present study were assigned to the C2 and M2 genotypes, respectively, and clustered closely with those of the G8P  Table 2). The VP2 and VP3 genes of the six strains mentioned above were distant from the simultaneous co-circulating Chinese G2P [4] strains E6894 and E6896 detected in Wuhan, with nucleotide identities of 87.1-97.5% ( Figure 5, C2, Figure 6, M2, Table 2).

Nonstructural Protein Genes
The NSP1, NSP3, and NSP5 genes of the six G8P [8] strains were assigned to the A2, T2, and H2 genotypes, respectively. The NSP1, NSP3, and NSP5 genes of these G8P [8] strains clustered with those of two G8P [8]   are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" represents a lineage of G8P [8] within M2.

Nonstructural Protein Genes
The NSP1, NSP3, and NSP5 genes of the six G8P [8] strains were assigned to the A2, T2, and H2 genotypes, respectively. The NSP1, NSP3, and NSP5 genes of these G8P [8] strains clustered with those of two G8P [8] Table 2). The NSP1, NSP3, and NSP5 genes of the six strains mentioned above were distant from the simultaneously co-circulating G2P [4] strains E6894 and E6896 detected in Wuhan (Figures 7-9, and Table 2).  Table 2). The NSP1, NSP3, and NSP5 genes of the six strains mentioned above were distant from the simultaneously cocirculating G2P [4] strains E6894 and E6896 detected in Wuhan (Figure 7, Figure 8, Figure  9, and Table 2).    The NSP2 genes of the six G8P [8] strains were assigned to the N2 genotype. These NSP2 genes were distant from those of the co-circulating G2P [4] strains from Wuhan, showing nucleotide identities of 96.5%, and were more distant from some G8P [8]   The NSP2 genes of the six G8P [8] strains were assigned to the N2 genotype. These NSP2 genes were distant from those of the co-circulating G2P [4] strains from Wuhan, showing nucleotide identities of 96.5%, and were more distant from some G8P [8]   are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represen lineages within N2 (G8P [8] strains, lineage "a").
The NSP4 genes of the six G8 strains were assigned to the E2 genotype. They ar closely related to two G8P [8] strains from Guangzhou City in the south of China, showing nucleotide identities of 99.4% and 99.6%, and they clustered together with the simultane ous G2P [4] strains (E6896 and E6894) in Wuhan and those of the G2P [4] strains in neigh boring countries, while remaining distant from the G8P [8] strains (15531, CAU17L-103 SKT-107, and PCB656) detected in neighboring countries, showing nucleotide identitie of 93.2-93.8% (Figure 11, H2a).  [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represent lineages within N2 (G8P [8] strains, lineage "a").
The NSP4 genes of the six G8 strains were assigned to the E2 genotype. They are closely related to two G8P [8] (Figure 11, H2a). Figure 11. Phylogenetic dendrogram based on complete coding regions of the NSP4 genes of repr sentative RVAs. The best nucleotide substitution model was T92 + I. Bootstrap values below 80% ar not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 an G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represen lineages within E2 (G8P [8] strains, lineage "a").

Discussion
Due to the COVID-19 pandemic, the lockdown from 23 January 2020 through 8 Apr 2020, and the subsequent zero-COVID-19 strategy, the collection of specimens was a fected. The number of specimens decreased compared to a previous surveillance in Wu han. In the previous surveillance of rotaviruses in Wuhan from 2011 through 2019, a tota of 6733 fecal specimens from 4409 children (under 15 years old) and 2324 adults (15-9 years old) were collected. RVAs were detected in 25.5% (1125/4409) and 12.3% (285/2324 of the specimens from children and adults, respectively [27]. The average annual sampl size from June 2011 through May 2019 was similar to that from phase 1 (pre-COVID-1 period). During phase 2 (outbreak and lockdown period) and phase 3 (zero-COVID-1 Figure 11. Phylogenetic dendrogram based on complete coding regions of the NSP4 genes of representative RVAs. The best nucleotide substitution model was T92 + I. Bootstrap values below 80% are not shown. The G8P [8] strains are highlighted with a filled circle. The contemporaneous G2 and G1/G9 strains are highlighted with filled squares and triangles, respectively. "a" and "b" represent lineages within E2 (G8P [8] strains, lineage "a").

Discussion
Due to the COVID-19 pandemic, the lockdown from 23 January 2020 through 8 April 2020, and the subsequent zero-COVID-19 strategy, the collection of specimens was affected. The number of specimens decreased compared to a previous surveillance in Wuhan. In the previous surveillance of rotaviruses in Wuhan from 2011 through 2019, a total of 6733 fecal specimens from 4409 children (under 15 years old) and 2324 adults (15-96 years old) were collected. RVAs were detected in 25.5% (1125/4409) and 12.3% (285/2324) of the specimens from children and adults, respectively [27]. The average annual sample size from June 2011 through May 2019 was similar to that from phase 1 (pre-COVID-19 period). During phase 2 (outbreak and lockdown period) and phase 3 (zero-COVID-19 period), the average annual sample sizes were sharply reduced to 10.9% (60/551) and 36.3% (200/551) in children and 4.1% (12/291) and 10.7% (31/291) in adults, respectively.
Compared to the previous surveillance of rotaviruses in Wuhan from 2011 through 2019, the detection rate of RVAs in both children (χ 2 = 15.01, p < 0.01) and adults (χ 2 = 15.66, p < 0.01) declined in the present study. The public health measures taken during the COVID-19 era had a significant impact on people's lives. Restrictions on the mobility of people and enhanced hand hygiene measures were deduced to be responsible for the lower incidence of RVAs both in children (19.40%) and adults (3.51%) that was observed in Wuhan with the background of the zero-COVID-19 strategy. Considering that the highest incidence was in the 13-36-month age group, the early introduction of routine rotavirus vaccinations to children before the kindergarten period is desired.
The Lanzhou lamb rotavirus (LLR) vaccine (G10P [15]) was licensed in China in 2000. It was the only rotavirus vaccine available before 12 April 2018, when RotaTeq (G1-G4, G6 P [8], P [5]) was introduced to mainland China. The LLR has been available in Wuhan since 2005. Because rotavirus vaccines are not included in the National Expanded Program of Immunization, the coverage is relatively low [28]. As discussed in the previous study, the influence of the LLR on the prevalence and genotypes of RVAs in Wuhan was limited [27]. A total of 185,507 doses of the LLR and 228,466 doses of RotaTeq were used from June 2019 through May 2022, which means about 61,835 and 76,155 infants were vaccinated with the LLR and RotaTeq, respectively. It was deduced that the use of rotavirus vaccines also contributed to the decline of the detection rates of RVAs.
Phylogenetic analysis revealed that the VP7, VP4, VP2, VP3, NSP1, NSP2, NSP3, and NSP5 genes of these G8P [8] strains clustered closely with those of the G8P [8] strains in Asia and were distant from those of the P [8] and G2P [4] strains simultaneously detected in Wuhan, while the VP1, VP6, and NSP4 genes were closely related to the typical G2P [4] rotavirus, including the G2P [4] strains detected in Wuhan. The present study suggests that the G8P [8] rotavirus emerging in China was generated by multiple reassortment events between the locally co-circulating G2P [4] rotavirus and the neighboring DS-1-like G8P [8] strains represented by CAU17L-103. The Chinese G8P [8] rotavirus carries the VP6, VP1, and NSP4 genes derived from the local G2P [4] rotavirus in the backbone of the neighboring DS-1-like G8P [8] strains. According to the emerging time of G8 [8] strains in Guangzhou and Wuhan, the high nucleotide identity among these strains, and their closeness in phylogenetic trees, it was deduced that the G8P [8] strains in Guangzhou and Wuhan might have originated from the same rotavirus strain. The DS-1-like G8P [8] rotavirus might have emerged in coastal cities and then spread to the inland. These findings highlight the contribution of reassortment and interspecies transmission events to the high rotavirus diversity in mainland China.
Before the present study, the genomes of two G8P [8] strains from Guangzhou, China were analyzed [9]. However, due to a lack of data on the local co-circulating Wa and DS-1 genogroup strains, the reassortment between local G2P [4] strains and Asian DS-1-like G8P [8] strains could not be revealed.
With the accumulation of whole-genome data for rotaviruses in humans and animals, we can further understand the evolution of the DS-1-like G8P [8] rotavirus and the spread of G8 rotaviruses in humans. The G6, G8, G10, and G15 genotypes are mostly combined with the typical bovine DS-1-like genomic constellation [39][40][41]. The DS-1-like genomic constellation G10/G6-P [11]/P[5]-I2-R2-C2-M2-A3/A11/A13-N2-T6-E2-H3 is the most prevalent in cattle. The genotypes N2-T6-E2-H3 are constant in the genomic constellations of bovine rotaviruses. It was deduced that the evolution process of the DS-1-like G8P [8] rotavirus might have been accompanied by interspecies transmission and multiple genetic reassortment events between bovine/human and human RVAs. Finally, the VP7 gene from bovine strains and the VP4 gene from human strains are present on the backbone of the human DS-1-like rotavirus.
The present study is helpful for understanding the evolutionary dynamics of emerging DS-1-like G8P [8] strains, and it provides new evidence that the DS-1-like G8P [8] rotavirus adapted to become a predominant human rotavirus strain and an important pathogen in the world [34]. Neither LLR nor RotaTeq can cover the emerging G8P [8] rotavirus. Monitoring the emerging genotypes in humans will assist scientists in developing or importing new vaccines and adjust the components of existing vaccines to improve their protective effect. The continuous surveillance of rotaviruses in both animals and humans is necessary.

Specimens
A hospital-based surveillance of sporadic diarrhea was conducted in Wuhan, China. The definition of diarrhea is loose/watery stools that occur three or more times within 24 h. The stool specimens were collected from inpatients and outpatients in four hospitals (Wuhan Commercial Staff Hospital, Wuhan Sixth Hospital, Wuhan Eleventh Hospital, and Wuhan Children's Hospital). All the specimens were stored at −80 • C. The concise epidemiological and clinical data were collected with a sampling form.

Detection of Rotavirus
Viral dsRNA was extracted from 200 microliters of a 10% stool suspension with a nucleic acid extraction kit by using the automatic nucleic acid extraction system NP968S (Suzhou Tianlong Science and Technology Co., Ltd., Suzhou, China) according to the manufacturer's instructions. Using real-time reverse transcription-polymerase chain reaction (RT-PCR) with a rotavirus kit (Zhijiang Science and Technology Co., Ltd., Shanghai, China), RVAs, RVBs, and RVCs were detected.

Genotyping of RVAs
The nearly full-length sequence of the VP7 gene and the partial VP4 gene were amplified using PrimeScript™ One Step RT-PCR Kit Ver.2 (Takara Biomedical Technology (Beijing) Co., Ltd. Beijing, China) according to the manufacturer's instructions. The nucleotide sequences of the primers are listed in Table S1. The genotypes of the VP7 and VP4 genes were determined by analyzing the sequences of the PCR products [42][43][44]. The PCR products were subjected to direct sequencing by the Sanger method at the Sangon Biotech (Shanghai) Co., Ltd. branch of Wuhan (Wuhan, China). The genotypes of RVAs were preliminarily assigned by the basic local alignment search tool (BLAST) and then confirmed by an analysis of phylogenetic trees.

Whole-Genome Sequencing
The complete eleven segmented genes of the whole genome were amplified by Prime-Script™ One Step RT-PCR Kit Ver.2 (Takara Biomedical Technology (Beijing) Co., Ltd. Beijing, China) according to the manufacturer's instructions. The nucleotide sequences of the primers are listed in Table S1 and include those reported in the previous studies [42][43][44][45][46]. Of these, the nearly full-length sequences of the VP4, VP6, VP1-3, and NSP1 genes were obtained by segmented RT-PCR and sequencing. The PCR products were subjected to direct sequencing by the Sanger method at the Sangon Biotech (Shanghai) Co., Ltd. branch of Wuhan (Wuhan, China).

Phylogenetic Analysis
The nucleotide sequence of each segment was assembled and edited using the DNA-MAN v10 software. The sequence identities were analyzed by the Lasergene bio-information v7 software (DNASTAR, Inc, Madison, WI, USA). Multiple alignments of the sequences were performed using MAFFT v7 (Kazutaka Katoh, Osaka, Japan) [47]. A phylogenetic analysis was conducted together with reference strain sequences obtained from the National Center for Biotechnology Information (NCBI) database by the MEGA program, version X [48]. The evolutionary history was inferred by using the maximum likelihood method, with 1000 bootstrap replicates, based on the best nucleotide substitution model with the lowest Bayesian information criterion (BIC) score in MEGA X [49]. The initial tree for the heuristic search was obtained automatically by applying neighbor-joining and BioNJ algorithms to a matrix of pairwise distances estimated using the maximum composite likelihood (MCL) approach, and then selecting the topology with a superior log likelihood value. The tree was drawn to scale, with branch lengths represented by the number of substitutions per site.

Statistical Analysis
Statistical analyses were performed using the Statistical Package for Social Science (SPSS, version 24, SPSS Inc., Chicago, IL, USA) software. Testing for the statistical significance was performed using the chi-squared test. A p-value less than 0.05 was considered significant. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the ethics committee of Wuhan Centers for Disease Control and Prevention (protocol code WHCDCIRB-K-20220012, approved on 8 April 2022). Fecal samples from patients with diarrhea were analyzed as part of a routine surveillance for infectious diseases.
Informed Consent Statement: Oral informed consent was obtained for all samples collected.
Data Availability Statement: The data will be provided upon request.