Emergence of Multiple Novel Inter-Genotype Recombinant Strains of Human Astroviruses Detected in Pediatric Patients With Acute Gastroenteritis in Thailand

Objective: Human astrovirus (HAstV) is recognized as an important cause of acute gastroenteritis in children. Recombination between different genotypes of HAstV can contribute to diversity and evolution of the virus. This study aimed to investigate the emergence of HAstV recombinant strains in pediatric patients hospitalized with acute gastroenteritis in Chiang Mai, Thailand, spanning 2011–2020. Methods: A total of 92 archival HAstV strains collected from pediatric patients with acute gastroenteritis during 2011–2020 were further characterized to identify the recombinant strains. The ORF1b and ORF2 junction region of each strain was amplified and sequenced. The obtained sequences were analyzed in comparison with the reference sequences retrieved from GenBank database. Their genotypes were assigned using MEGA X software based on the partial ORF1b (RdRp) and ORF2 (capsid) regions, and the recombination breakpoints of recombinant strains were determined by SimPlot and RDP4 analyses. Results: Five inter-genotype recombinant strains with three recombination patterns of ORF1b/ORF2 of classic HAstV, HAstV8/HAstV1, HAstV8/HAstV3, and HAstV3/HAstV2, were detected. The recombination breakpoints of all strains were located at the 3′-end region of ORF1b close to the ORF1b/ORF2 junction. Conclusion: Several novel inter-genotype recombinant strains of classic HAstV genotypes were detected in pediatric patients with acute gastroenteritis in Chiang Mai, Thailand, during the period of 10 years from 2011 to 2020.


INTRODUCTION
Human astrovirus (HAstV) is a non-enveloped, positive-sense, single-stranded RNA virus, which is associated with acute gastroenteritis and systemic diseases (Johnson et al., 2017). The genome of HAstV contains three open reading frames (ORFs), ORF1a, ORF1b, and ORF2 (Bosch et al., 2014). Currently, HAstVs are classified into three major groups, including classic HAstV (HAstV1-HAstV8), novel HAstV-MLB (MLB1-MLB3), and novel HAstV-VA/HMO (VA1-VA5) (De Benedictis et al., 2011;Bosch et al., 2014). The diversity of HAstV genotypes is plausibly associated with frequent interspecies transmission and recombination events (Wohlgemuth et al., 2019). The first evidence of recombination naturally occurring among HAstVs was documented in 2001 by demonstrating that the ORF1b and ORF2 of the new HAstV strain were closely related to those of HAstV3 and HAstV5, respectively, and a putative recombination site was demonstrated at the ORF1b/ORF2 junction . To date, many more recombinant strains have been reported even though little is known about the mechanism of astrovirus recombination (Bosch et al., 2014).
It is well documented that the recombination event is one of the mechanisms involved in the evolution of RNA viruses (Simon-Loriere and Holmes, 2011;Wohlgemuth et al., 2019). The epidemiology of HAstV has been studied in Thailand for decades (Malasao et al., 2012;Kumthip et al., 2018). However, no study on HAstV recombination has been carried out in Thailand. Therefore, the present study aimed to identify HAstV recombinant strains retrospectively in children hospitalized with acute gastroenteritis in Chiang Mai, Thailand, during the period of 2011-2020.

Fecal Samples and Human Astrovirus Strains
The fecal samples included in this study were collected from children under 5 years old who were admitted to five hospitals with acute gastroenteritis in Chiang Mai city, Thailand, during the period of 10 years from 2011 to 2020. The age of these inpatients ranged from neonate up to 5 years old. The criteria for acute gastroenteritis are defined as the sudden passage of watery stools (at least three times per day) for a duration of less than 2 weeks (Bassetto et al., 2019). A total of 3,534 fecal samples were collected and suspended in phosphate-buffered saline (PBS), pH 7.4, for viral genome extraction and screened for HAstV by the RT-PCR method using panastrovirus consensus primers (Finkbeiner et al., 2009a). Out of 3,534 fecal samples, 92 were positive for HAstV, and their genotypes, based on the ORF1b (RdRp) nucleotide sequence analysis, were assigned by comparison with those of the corresponding reference strains available in the GenBank database by performing the nucleotide sequence alignment and phylogenetic analyses as described previously (Kumthip et al., 2018). The ORF1b (RdRp) nucleotide sequences of these 92 HAstV strains have been deposited in the GenBank database under the access numbers of MH325213-MH325266, and MZ327095-MZ327133. Of 92 HAstV strains, 70 were classic HAstV; 19 and 3 were novel HAstV-MLB and novel HAstV-VA, respectively; and all of these HAstV strains were included in this study. More comprehensive information about the epidemiology, prevalence, seasonality, genotype distribution, and coinfection of HAstV with other diarrheal viruses of the stool samples included in this study are available in our published articles (Kumthip et al., 2018;Wei et al., 2021).

Amplification and Nucleotide
Sequencing of Human Astrovirus ORF1b (RdRp) and ORF2 (capsid) Junction The viral RNA genomes of 92 archival HAstV strains were extracted using a Geneaid Viral Nucleic Acid Extraction Kit II (Geneaid, Taipei, Taiwan) according to the manufacturer's protocol. The cDNA was synthesized from the viral RNA genome using RevertAid Fist Stand cDNA Synthesis Kit (Thermo Fisher Scientific, United States). Amplification of the ORF1b and ORF2 junction encompassing the 3 end of ORF1b and 5 end of ORF2 was performed by PCR and semi-nested PCR using the primer sets listed in Table 1. For classic HAstV, a semi-nested PCR was performed using two different alternative forward primers, SF0073 and SF0076-F, in combination with a set of outer reverse primer AHAstVR1 and inner reverse primer AHAstVR2, which generated PCR product sizes of 1104 and 708 bp, respectively. For novel HAstV-MLB, a conventional PCR was performed using forward primer SF0073 and reverse primer AHMLBR1 to generate a PCR product size of 926 bp. For novel HAstV-VA, a semi-nested PCR was performed using forward primer SF0076-F in combination with a set of outer reverse primer AHVAR1 and inner reverse primer AHVAR2, which generated a PCR product size of 987 bp. The PCR products were purified by using the GenepHlow TM Gel/PCR kit (Geneaid Biotech, Taiwan), and then, the purified PCR products were sequenced by First Base Laboratory SDNBHN Selangor Darul Ehsan, Malaysia.

Phylogenetic and Recombination Analyses
The obtained nucleotide sequences were analyzed by comparing with those of the reference strains available in the GenBank database using the Basic Local Alignment Search Tool (BLAST) server. 1 The phylogenetic trees of the partial ORF1b (RdRp) and ORF2 (capsid) genes were constructed by using the MEGA X software (Tamura, 1992;Kumar et al., 2018) based on the maximum likelihood method and selected the best-fit evolutionary model for the data set via Tamura-3-parameter model with 1,000 replicates. The recombination breakpoint was determined by using SimPlot software v.3.5.1 (Lole et al., 1999) and Recombination Detection Program v.4.39 (RDP 4.39) (Wang et al., 2020). In addition, identification of the homologous position of the recombination breakpoint was also based on The mixed bases in the degenerate primer are as follows: Y, for C or T; W, for A or T; R, for A or G; D, for A, G, or T; S, for G or C; K, for T or G. The primer locations for classic HAstV, novel HAstV-MLB, and novel HAstV-VA strains were based on the nucleotide sequence positions of L23513, FJ402983, and GQ502193 reference strains from GenBank database, respectively.
nucleotide sequence alignment of our strains together with other reference strains.

Nucleotide Sequence Accession Number
The nucleotide sequences of HAstV recombinant strains described in this study have been deposited in the GenBank database under the accession numbers OK135150-OK135154.
For the classic HAstV3/HAstV2 recombination pattern, the nucleotide sequence of the ORF1b and ORF2 junction of CMH-S015-20 was analyzed in comparison with those of the classic HAstV3 (GU223905/HAstV3/RUS/2003) and classic HAstV2 (MK059950/HAstV2/USA) reference strains. The recombination breakpoint was observed at nt position 4322 based on the genome nt position of the classic HAstV2 reference strain (MK059950/HAstV2/USA), which is located at the 3 -end region of ORF1b close to ORF1b/ORF2 junction ( Figure 2E) with maximum chi-square value of 2.414 × 10 −6 ( Table 2).

DISCUSSION
The genetic recombination is commonly found in the RNA virus families (Lai, 1992;Worobey and Holmes, 1999). As with most other RNA viruses, HAstV recombination was initially reported for the first time in 2001 by Walter et al. (2001), and later, recombination was proposed as one of the mechanisms that plays a crucial role in the genetic diversity and evolution of HAstV (Wohlgemuth et al., 2019;Roach and Langlois, 2021). Since then, a number of studies have reported a wide variety of recombinant strains, and the ORF1b/ORF2 junction of the HAstV genome has been reported as the predominant location at which recombination events occur Wolfaardt et al., 2011;De Grazia et al., 2012;Martella et al., 2013;Babkin et al., 2014;Wohlgemuth et al., 2019). So far, several inter-genotypes of the ORF1b/ORF2 recombinant strains have been reported from many countries around the world. The recombination at the ORF1b/ORF2 region of HAstV3/HAstV5 was first reported from the United States and Mexico in 2001 . The recombinations at the ORF1b/ORF2 region of HAstV1/HAstV2, HAstV3/HAstV2, HAstV1/HAstV3, and HAstV1/HAstV4 were reported from Italy (De Grazia et al., 2012;Martella et al., 2013). In addition to Italy, the HAstV3/HAstV2 was also reported from Russia (Babkin et al., 2014) and Kenya (Wolfaardt et al., 2011). The HAstV4/HAstV5 was reported from South Africa (Taylor et al., 2001) and China (Zhou et al., 2016). In addition, the HAstV1/HAstV5, HAstV2/HAstV5, and HAstV1/HAstV2 were also reported from China (Zhou et al., 2016). Furthermore, the HAstV2/HAstV8 was reported from South Korea (Ha et al., 2016).
Our study reports the detection of five isolates of ORF1b/ORF2 recombinant strains of HAstV8/HAstV1 (n = 3), HAstV8/HAstV3 (n = 1), and HAstV3/HAstV2 (n = 1) in children with acute gastroenteritis (Figure 2). The HAstV8/HAstV1 was detected in three out of five (60.0%) of the ORF1b/ORF2 recombinant strains described in this study, suggesting that a recombination event between HAstV8 and HAstV1 likely occurs more commonly than between other genotypes. Moreover, it is interesting to point out that the recombinant strains of HAstV8/HAstV1 and HAstV8/HAstV3 at the ORF1b/ORF2 region, to the best of our knowledge, have not been reported previously from elsewhere and are being reported here in this study as the novel HAstV ORF1b/ORF2 recombinant strains. Nevertheless, the HAstV8/HAstV1 recombinant strain at the ORF1a/ORF2 region has been reported previously from India with high predominance of 67.7-76.9% (Verma et al., 2010;Pativada et al., 2011) compared to other inter-genotype recombinant strains detected in the same study. In the present study, the HAstV8/HAstV1 recombinant strains were detected in 2012 and 2015, whereas HAstV8/HAstV3 was detected in 2013 in Chiang Mai, Thailand ( Table 2). To ascertain that the recombination events had occurred in nature to generate these recombinant strains, HAstV8, HAstV1, and HAstV3 should have been demonstrated to circulate in this geographical area prior to or at the same period of time when the recombinant strains were detected. In fact, HAstV8 and HAstV1 were reported previously in children with acute gastroenteritis in this geographical area during the period of 2011-2016 (Kumthip et al., 2018) and HAstV3 in 2000 (Malasao et al., 2012).
The ORF1b/ORF2 region of the HAstV genome is demonstrated to be a predominant location at which recombination events frequently occur Wolfaardt et al., 2011;De Grazia et al., 2012;Martella et al., 2013;Babkin et al., 2014;Wohlgemuth et al., 2019). In our study, the recombination breakpoints of all five recombinant strains were located close to the 3 -end region of ORF1b (Figure 2). All three novel HAstV8/HAstV1 recombinant strains (CMH-N178-12, CMH-S059-15, and CMH-S062-15) contained recombination breakpoint at the same nt position 4358, suggesting that nt position 4358 is a hot spot for recombination events to occur between HAstV8 and HAstV1 genotypes. In addition, the recombination breakpoint of novel HAstV8/HAstV3 (CMH-N106-13) recombinant strain located at nt position 4292 of ORF1b region. Regarding the recombination breakpoint of CMH-S015-20 located at nt position 4322 of ORF1b, HAstV2 (MK059950/HAstV2/USA) was used as the reference strain, whereas the HAstV3/HAstV2 recombinant strain reported from Kenya (Wolfaardt et al., 2011) used HAstV1 (L23513/HAstV1/USA) as a reference strain to predict the recombination breakpoint at nt position 4328. We performed a nucleotide sequence alignment of these three reference strains and found the recombination breakpoint located at nt positions 4322, 4248, and 4325 based on HAstV2, HAstV3, HAstV1 as the reference strains, respectively (data not shown). Even though the recombination breakpoints are different between different HAstV genotypes, in fact, the recombination breakpoint occurs at the same nt position. The difference in the number of the nt position is based on the nt position in the full-length genome of each HAstV genotype used as the reference strain.
Besides the recombination events being shown in the classic HAstV, suggestive evidence of recombination between novel HAstV-MLB3 and HAstV-MLB1 or -MLB2 by using next generation sequencing the samples from environmental waters has been reported recently (Hata et al., 2018). Since then, no other reports are available in the literature. In the present study, we also investigated the recombinant strains among HAstV-MLB and HAstV-VA strains; no recombination event has been identified in this study. To date, no recombination of classic HAstV with either HAstV-MLB or HAstV-VA has been reported.

DATA AVAILABILITY STATEMENT
The data presented in the study are deposited in the GenBank database, accession number OK135150-OK135154 (https:// www.ncbi.nlm.nih.gov/genbank/).

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the ethical committee for human rights related to human experimentation, Faculty of Medicine, Chiang Mai University (MIC-2557-02710). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

AUTHOR CONTRIBUTIONS
HW performed the experiments, data collection and analysis, drafted the original and made a revised manuscript, and read and approved the final draft. PK and KK conceived and designed the study, funding acquisition, specimen collection, supervised the study procedures, data collection and analysis, and read and approved the final draft manuscript. AY supervised the study procedures, and performed the experiments, and read and approved the final draft manuscript. NM conceived and designed the study, funding acquisition, analyzed and interpreted the data, was responsible for the overall study, and critically revised and editing the manuscript. All authors contributed to the article and approved the submitted version.