Porcine rotavirus C in pigs with gastroenteritis on Thai swine farms, 2011–2016

Swine are economically important food animals, but highly contagious porcine epidemic diarrhea virus (PEDV) and rotavirus can afflict pig herds and contribute significantly to piglet morbidity and mortality. While there have been studies on rotavirus group A (RVA) in Thailand, reports of rotavirus group C (RVC) are limited. Here, we aimed to identify the prevalence of RVC circulating on Thai commercial swine farms. We analyzed 769 feces and intestine mucosal contents of pigs affected with diarrhea between 2011 and 2016 using RT-PCR specific for the PEDV spike (S), rotavirus glycoprotein (G) VP7, and protease-sensitive protein (P) VP4 genes. We found that 6.6% (51/769) of samples tested positive for RVC, of which 11 samples were co-infected with RVA and four samples were co-infected with PEDV. Three samples tested positive for all three viruses. Phylogenetic analysis of the VP7 gene showed that the most frequent RVC genotype was G1, which grouped with the prototypic RVC Cowden strain. While G6 and G9 were also common, G3 was relatively rare. Analysis of the VP4 gene revealed that the most common P type was P[5], followed by P[4], P[7], and P[1]. In all, there were six G/P combinations (G6P[5], G1P[1], G1P[4], G1P[5], G9P[4], and G9P[7]), of which G6P[5] was the most predominant.

The intestine mucosa were collected from dead animals by scraping the duodenum and upper part of the jejunum, particularly the thin walled area where gas accumulated inside the lumen. Approximately 10% (v/v) of mucosal or fecal suspensions in sterile phosphate-buffered saline (0.1 M, pH 7.2) were centrifuged at 3,000g for 20 min and the supernatants collected.

Viral nucleic acid detection
Nucleic acid was extracted using Ribospin vRD II viral RNA extraction kit (GeneAll, Seoul, Korea) according to the manufacturer's instructions. The partial S gene of PEDV, VP7 gene of RVA/RVC, and VP4 gene of RVC were amplified using SuperScript III One- Step RT-PCR System with Platinum Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA). Samples were reverse-transcribed at 48 C for 45 min. Cycling parameters were initial denaturation at 95 C for 2 min, followed by 35 cycles at 94 C for 30 s, 52 C or 55 C for 1 min, 72 C for 90 s, and final extension at 72 C for 5 min. Primer sequences are shown in Table 1. Amplicons were purified using agarose gel electrophoresis and sequenced. Nucleotide (nt) sequences were deposited in the GenBank database under the accession numbers KX911667-KX911708, MF139507-MF139509 and MF139516-MF139517 (VP7) and MG575522-MG575532 (VP4).
The nearly full-length VP7 sequence encompassing nt 112-952 from the Thai RVC strains encoded amino acid residues 38-316. This region spans the variable region 2 (VR2) to variable region 8 (VR8). Genotype G1 and G9 represented three variable sites  at residues 39, 53, and 57 (Table S2). Most G6 strains (9/10 strains) had four residue insertion between amino acid positions 245 and 248 (SSSV/SSTL/SSTM/SSSM) towards the carboxyl terminus of VR8. Potential N-linked glycosylation sites at residues 67-69 and The deduced amino acid sequences of several representative RVC strains from this study were compared with the amino acid sequences of the prototype strains (Fig. 3). The alignment region spanned residues 15-385 (based on Cowden numbering).

DISCUSSION
The viral etiology of diarrhea in pigs is not routinely investigated on Thai pig farms, which often contributes to the failure to prevent and contain disease transmission. Rotavirus infection including RVC disproportionately affects suckling and weaned piglets and often co-infects with other viruses (Saif et al., 1980;Amimo, Vlasova & Saif, 2013;Martella et al., 2007). Previous reports of porcine rotavirus prevalence in Thailand have only been RVA (approximately 10-23%), while epidemiological study of porcine RVC was limited (Chan-It et al., 2008;Khamrin et al., 2007;Maneekarn & Khamrin, 2014;Yodmeeklin et al., 2016). Our passive surveillance identified a lower prevalence of RVC (6.6%) than RVA (9.5%) in symptomatic piglets with diarrhea, both of which were detected as single and/or co-infections (Collins, Martella & O'Shea, 2008;Marthaler et al., 2014;Nagesha & Holmes, 1988;Theuns et al., 2016;Zhou et al., 2016). RVC appeared most frequently in pigs eight weeks old or younger, which was consistent with previous reports (Amimo, Vlasova & Saif, 2013;Jeong et al., 2009;Kim et al., 1999;Martella et al., 2007;Marthaler et al., 2013;Suzuki et al., 2015). It is believed that animals co-infected with more than one enteric virus experienced increased intestinal epithelium damage and/or viral replication, which results in more severe diarrhea (Amimo, Vlasova & Saif, 2013;Jeong et al., 2009;Ishimaru et al., 1991;Martella et al., 2007). In this study, dual infections between PEDV and rotavirus in younger piglets (<4 weeks old) often showed a higher morbidity rate. Younger piglets appear to be more susceptible to higher morbidity and mortality than older pigs Shibata et al., 2000;Steyer et al., 2008). There were instances of co-infection with PEDV and RVA in sows even though they are usually asymptomatic. This may explain the persistence of rotavirus within the herd and facilitate vertical transmission. Although rotavirus infection is frequent in the winter season on farms in the temperate climate, rotavirus infection occurred throughout the year on Thai swine farms. Some studies have suggested that rotavirus infection is not as seasonally dependent in the tropics because the relatively high humidity may facilitate increased rotavirus infection (Cook et al., 2004;Levy, Hubbard & Eisenberg, 2009).
The genetic relationship between VP4 sequences in this study and previous RVC isolates was also determined. The sequence identity among the Thai RVC strains and the prototypic strains was quite low. Comparison showed between 59.6% and 66% with the Cowden strain, 52.2% and 62.7% with the human strain Bristol, and 59.5% and 66.1% with the bovine strain Shintoku. Not surprisingly, most strains analyzed in this study possessed nt identities similar to previously reported Asian RVC strains (Korean and Japanese strains) (Jeong et al., 2015;Suzuki et al., 2015). Taken together, these data suggest that the same RVC genotypes are in circulation within several Asian countries.
This study was limited by the inability to successfully sequence all the VP4 gene from all 47 VP7-positive RVC, which suggests possible high sequence variability in the region we targeted for amplification. The fact that P[5] sequences were nearly identical reflected the single origin and time of infection. Indeed, these samples were derived from the same farm, which suggests an RVC outbreak and may not truly represent the variability of all the strains in circulation.

CONCLUSION
Rotavirus group C surveillance in Thailand and Southeast Asia will continue to be important in identifying the viral etiology of gastroenteritis in pigs and in managing the viral transmission on swine farms. This study identified several currently circulating RVC in swine, an economically important food animal and a potential source of zoonotic transmission of rotavirus.