Molecular characterization of a porcine sapelovirus strain isolated in China

Porcine sapelovirus (PSV) infections have been associated with a wide spectrum of symptoms, ranging from asymptomatic infection to clinical signs including diarrhoea, pneumonia, reproductive disorders, and polioencephalomyelitis. Although it has a global distribution, there have been relatively few studies on PSV in domestic animals. We isolated a PSV strain, SHCM2019, from faecal specimens from swine, using PK-15 cells. To investigate its molecular characteristics and pathogenicity, the genomic sequence of strain SHCM2019 was analysed, and clinical manifestations and pathological changes occurring after inoculation of neonatal piglets were observed. The virus isolated using PK-15 cells was identified as PSV using RT-PCR, transmission electron microscopy (TEM), and immunofluorescence assay (IFA). Sequencing results showed that the full-length genome of the SHCM2019 strain was 7,567 nucleotides (nt) in length, including a 27-nucleotide poly(A) tail. Phylogenetic analysis demonstrated that this virus was a PSV isolate belonging to the Chinese strain cluster. Recombination analysis indicated that there might be a recombination breakpoint upstream of the 3D region of the genome. Pathogenicity experiments demonstrated that the virus isolate could cause diarrhoea and pneumonia in piglets. In breif, a recombinant PSV strain, SHCM2019, was isolated and shown to be pathogenic. Our results may provide a reference for future research on the pathogenic mechanism and evolutionary characteristics of PSV.


Introduction
The genus Sapelovirus, family Picornaviridae, consists of two species: Sapelovirus A (porcine sapelovirus) and Sapelovirus B (simian sapelovirus) [1]. Porcine sapelovirus (PSV) is a non-enveloped, positive-sense, single-stranded RNA virus that was previously known as porcine enterovirus  and formerly belonged to the genus Enterovirus [2]. The genome of PSV is approximately 7500 nucleotides (nt) in length, consisting of a 5' untranslated region (UTR), a protein coding region, a 3' UTR, and a poly(A) tail. The coding region contains a single large open reading frame (ORF) that encodes a polyprotein precursor about 2331 amino acids that is cleaved by the viral protease into 12 mature functional proteins. The structure of the polyprotein is consistent with the L-4-3-4 structure of other picornaviruses, consisting of a leader peptide (L), four structural proteins (VP4, VP2, VP3, and VP1), and seven nonstructural proteins (2A, 2B, 2C, 3A, 3B, 3C, and 3D) [3][4][5].
Since PSV was first reported in the UK in 1958 [6], it has been identified in many countries, including Canada, Japan, Australia, Brazil, Spain, Korea, Italy, and China [5,[7][8][9][10][11][12][13]. PSV infection rates in pigs have been investigated globally, ranging from 7.1% in India to 71.0% in Hungary [14,15]. PSV infections have been associated with a wide spectrum of symptoms ranging from asymptomatic infection to acute fatal encephalomyelitis, reproductive disorders, diarrhoea, and pneumonia [12,[16][17][18]. Prodělalová et al. reported that PSV infection rates were higher in asymptomatic pigs than in diarrhoeal pigs [16]. However, Zhang et al. demonstrated a higher prevalence of PSV in diarrhoeal pigs by using metagenomic analysis of pig faeces [19]. Kim et al. reported that PSV causes intestinal lesions and colonizes villous epithelial cells in the small intestine [20]. Like porcine epidemic diarrhea virus (PEDV), PSV can be found in coinfections with porcine parvovirus, classical swine fever virus, porcine reproductive and respiratory disorder syndrome virus, porcine enterovirus, and other viruses, leading to atypical clinical signs in pigs [21][22][23]. The symptoms of PSV are often obscured or overlooked due to the mixed infections [16], which poses a considerable threat to the pig industry.
In this study, a PSV strain was isolated using PK-15 cells. Phylogenetic analysis showed that it is closely related to other Chinese isolates. Recombination analysis indicated a possible recombination breakpoint in the 3D region. Animal infection experiments demonstrated that this isolate is pathogenic to piglets. The study lays a foundation for understanding the pathogenic mechanism of PSV.

Virus isolation
Two faecal specimens screened by multiplex RT-PCR that were RNA-positive for PSV but RNA-negative for PEDV, TGEV, PKV, PAstV, PDCoV, PToV, PTV, PSaV, PoRV, and BVDV were further verified using separate PSV primers in multiplex RT-PCR and inoculated onto PK15 cells. The positive faecal samples were suspended in sterile phosphatebuffered saline (PBS) containing 1,000 units of penicillin and 1,000 μg of streptomycin per mL and centrifuged for 10 min at 3500×g. The supernatants were filtered through 0.22μm pore membrane filters (Merck Millipore Ltd., USA), and the filtrates were diluted 100-fold in serum-free DMEM. Subsequently, PK-15 cells growing in a six-well plate were inoculated with 1 mL of the diluted virus suspension per well. After incubation for 1 h at 37 °C, the inoculum was discarded, and the cells were washed three times with sterile PBS. Next, DMEM supplemented with 0.5 μg of TPCKtrypsin (Gibco, USA) per mL was added, and the cells were incubated at 37°C in 5% CO 2 , with daily observation for a cytopathic effect (CPE). When more than 80% of the cells showed CPE, the supernatant and cells were harvested for continuous subculture. The two virus isolates were cultured continuously for six generations, the cell supernatants were collected, and each generation was identified by the RT-PCR method. RNA-positive samples were then analyzed by TEM and IFA (see below).

RT-PCR
Total RNA from the first six generations of the two PSV isolates was extracted from the culture supernatant of infected cells using TRIzol Reagent (Takara, China) according to the manufacturer's instructions. The cDNAs were synthesized in a final volume of 20 μL containing 10 μL of RNA, 4 μL of 5 × RT buffer, 1 μL of 10 mM dNTPs, 2 μL of 25 μM random primer, 0.5 μL of Moloney murine leukaemia virus (M-MLV) reverse transcriptase (TaKaRa, China), 0.5 μL of RNase inhibitor (TaKaRa, China), and 2 μL of diethylpyrocarbonate (DEPC)-treated water. The reaction was incubated at 42°C for 1 h, followed by incubation at 70°C for 15 min. cDNA was used for PCR with the primers PSV-F (5′-GAT GTG GCG CAT GCT CTT -3′) and PSV-R (5′-TGC TGC CTC CTG TGT TGT TAT-3′). Amplified products were separated by electrophoresis in a 1.5% agarose gel and purified using a DNA gel extraction kit (Tiangen, China). The purified PCR products were subsequently cloned into the PEASY-Blunt Zero vector (TransGen, China) and sequenced (Biosune, China). Sequences were analysed using the BLAST search program (http:// blast. ncbi. nlm. nih. gov/).

Transmission electron microscopy (TEM)
PK-15 cells infected with the virus isolate identified by RT-PCR were frozen and thawed three times when more than 80% CPE was observed, followed by centrifugation at 2000×g for 30 min to remove cell debris. Crude virus was pelleted from the clarified supernatant by ultracentrifugation at 98,900 × g at 4 °C for 3 h in a type 70 Ti rotor (Beckman, USA). The resulting pellet was resuspended in a small amount of PBS and was subsequently layered onto a 20-60% (w/v) discontinuous iodixanol solution (OptiPrep™ Density Gradient Medium, Sigma-Aldrich, USA) by centrifugation at 124,400×g at 4 °C for 3 h in a SW-55 Ti rotor (Beckman, USA). The virus band at the interface was collected and placed onto a formvar grid (Electron Microscopy Sciences, USA) for 5 min, and excess liquid was subsequently removed using filter paper. The samples were observed using a transmission electron microscope (Hitachi, Japan) for the morphological identification of the virus.

Immunofluorescence assay (IFA)
IFA was performed using a polyclonal antibody specific for the PSV-VP1 protein (prepared in our laboratory). Briefly, PK-15 cells in 6-well plates were mock infected or infected with the isolate identified by RT-PCR and TEM, incubated for 24 h, fixed in 4% paraformaldehyde for 30 min at 4 °C, and subsequently permeabilized with 0.3% Triton X-100 (Biodee, China) for 10 min at room temperature (RT). The cells were then washed three times with PBS and blocked with 5% bovine serum albumin (BSA; Sigma, USA) at room temperature for 1 h. Mouse polyclonal antibody against PSV and Alexa Fluor TM 488-conjugated rabbit anti-mouse IgG (Invitrogen, USA) were used as the primary and secondary antibody, respectively. The cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; Beyotime, China) for 5 min at room temperature. The cells were then washed with PBS, and the stained cells were observed using a fluorescence microscope (Carl Zeiss, Germany).

Sequence and phylogenetic analysis
Third-generation viral RNA was extracted using TRIzol Reagent. Reverse transcription and PCR were performed using M-MLV reverse transcriptase and TransStart Fast-Pfu DNA polymerase (TransGen, China), respectively. Seven pairs of primers were used for PCR amplification of the complete ORF sequences of the PSV isolate. The 5'-UTR and 3'-UTR sequences were obtained using a 5'-3' RACE kit (Takara, China) according to the manufacturer's instructions. Primer sequences used to amplify the ORF region and GSP sequences required for 3' and 5'-UTR amplification were designed and stored in our laboratory (Supplementary Table S1). The PCR products were cloned using a pEASY-Blunt Cloning Kit, and the resulting plasmids were sequenced. The sequence fragments were assembled using DNAStar v. 7.1. The complete sequences of the polyprotein gene of all reference strains obtained from the GenBank database were used in sequence alignments and phylogenetic analysis. Phylogenetic trees were constructed using the maximum-likelihood method with 1000 bootstrap replicates in MEGA v. 7.0. Recombination analysis was performed using SimPlot v. 3.5.1 and Recombination Detection Program (RDP) v. 4.10.

Experimental infection of piglets
To assess the pathogenicity of isolate SHCM2019 in piglets, four 5-day-old piglets from the same sow were used. Before inoculation, we observed the piglets for 3 days and collected faecal samples from all of them to ensure that they were not infected with PSV. The four piglets were randomly divided into two groups (two piglets per group). Each piglet in group 1 (L1 and L2) was given 5 mL of PSV-SHCM (10 7.41 TCID 50 /mL) by the oral route, and those in group 2 (C1 and C2) received 5 mL of DMEM orally as a negative control. After inoculation, the piglets were monitored each day for clinical signs. Faecal samples were collected from each piglet for viral load determination by qRT-PCR at 0, 1, 2, 3, 4, 5 and 6 days post-inoculation (dpi), and serum samples were collected for seroconversion tests at 0, 1, 3 and 6 dpi. All of the piglets were killed at 6 dpi, and a portion of their tissue samples were fixed with 4% paraformaldehyde solution for histological slide preparation, and the other portion were collected for viral load determination by qRT-PCR.

Ethics statement
All animal experiments were performed under the guidance of the Institutional Animal Care and Use Committee at the Center for Disease Control and Prevention (CDC) and the Laboratory Animal Care International accredited facility.

Virus isolation and identification
PK-15 cells were inoculated with each of the isolates, but only one produced a distinct CPE after three passages, which was characterized by rounding and shrinking of the cells after 30 h and detachment of all of the cells by 70 h (Fig. 1A). The morphology of PK-15 cells inoculated with the other isolate did not change significantly after six continuous passages. In order to verify the isolation of the virus strain, RT-PCR was performed to detect viral RNA in the culture supernatant from continuous passages of each of the isolates. The results showed that the virus causing CPE remained positive by RT-PCR, producing a 624-bp amplicon (Fig. 1B), whereas the isolate without CPE remained negative by PCR at each passage. The RT-PCR-positive isolate was named SHCM2019. To determine its morphological characteristics, this virus was purified, negatively stained, and examined by transmission electron microscopy (TEM), which showed that the viral particles were nonenveloped, icosahedral, and approximately 25-30 nm in diameter (Fig. 1C). In addition, IFA confirmed that a polyclonal antibody against PSV reacted with infected cells producing specific green fluorescence in the cytoplasm (Fig. 1D). These data confirm that a PSV strain was successfully isolated from diarrheal faecal samples originating from a farm in Shanghai, China. C Transmission electron micrograph (TEM) of cultured PSV strain SHCM2019. Virus pelleted by ultracentrifugation was stained with uranyl acetate and sprayed onto a coated formvar grid. D Immunofluorescence analysis of PK-15 cells infected with PSV by laser confocal microscopy. Immunofluorescence assay with DAPI confirmed that strain SHCM2019 could be recognized by an anti-PSV polyclonal antibody

Genome organization and phylogenetic analysis
To investigate the molecular characteristics of the isolated virus, its complete genome sequence was determined and subsequently submitted to the GenBank database (GenBank ID: MN685785). The full-length genome of this virus strain is 7567 nt in length with a single 6996-nt ORF flanked by a 465-nt 5′-UTR and a 106-nt 3′-UTR. Although the genome of SHCM2019 differs in length from those of other PSV strains, its polyprotein gene is similar to those of most Chinese PSVs. Although the length of the polyprotein gene varies among PSV strains, it is relatively conserved among Chinese PSVs, exhibiting a consistent length of 6996 nt, except for the strains HuN4, HuN6, HuN21, and HuN32, which have some extra nucleotide insertions.
A phylogenetic tree was constructed by the maximumlikelihood method based on the sequence of the complete polyprotein gene of the SHCM2019 strain and those of other representative sapeloviruses from the NCBI database. The results showed that the SHCM2019 strain was most similar to the YC2011 strain, forming a subgroup with other PSV strains, which were clustered in the China group (Fig. 2). Further comparisons showed that SHCM2019 exhibited 90.8% nucleotide and 98% amino acid sequence identity to the YC2011 strain ( Supplementary Fig. S1). In addition, a high degree of genetic diversity was evident among the PSV isolates from different countries. The Chinese PSV strains were closely related to the Korean PSV strains but were distantly related to other members of the genus Sapelovirus.
Recombination analysis was carried out using SimPlot v. 3.5.1 and RDP v. 4.10 software. The SHCM2019 strain was used in separate queries to generate a standard similarity plot to identify possible recombination events using Simplot software. A high degree of nucleotide sequence similarity to HeB04 was found in the VP4, VP2, and VP3 regions, whereas the 3D region was highly similar to that of ISU-SHIC (Fig. 3A). RDP4 software was used to verify the recombinant nature of strain SHCM2019 and to identify potential breakpoints. Bootscanning analysis revealed a possible breakpoint in the 3D region (Fig. 3B), and separate phylogenetic trees based on the partial 3D region and the nonrecombinant region of the SHCM2019 strain provided further evidence of recombination (Fig. 3C). When predicting recombination events using RDP4 software, RDP, Chimaera, Bootscan, 3Seq, Maxchi, Phylpro, GENECONV, LARD and SISCAN algorithms were used, and six of these detected a recombination signal with a p-value <0.05 (Fig. 3D).

Experimental infection, necropsy, and histopathological examination
To test the virulence pathogenicity of SHCM2019 strain, two healthy 8-day-old piglets were inoculated with SHCM2019. Inappetence and lethargy were observed in PSV-infected piglets at 2 dpi, and watery diarrhoea occurred at 5 dpi (Fig. 4D). Necropsy results showed that lesions were mainly concentrated in the intestines and lungs. The intestines of PSVinfected piglets were thin and translucent, and the intestinal contents were pulpy or watery (Fig. 4E). There was obvious congestion in the lungs compared to uninfected pigs (Fig. 4F). The piglets in the control group were clean around the anus and had no obvious clinical symptoms (Fig. 4A). The intestines were thick and elastic, and the lungs were normal (Fig. 4B and C). Diseased tissues were also examined histopathologically. Histopathological results showed that villi in the small intestine were slightly atrophic, and submucosal oedema was obvious (Fig. 5C). The lung was damaged by thickened alveolar Fig. 2 Phylogenetic analysis based on the nucleotide sequence of the polyprotein coding region of PSV strains. The tree was constructed by the maximum-likelihood method using MEGA7.0 software. The numbers at the branches are bootstrap values (%) based on 1000 replicates. The PSV strain isolated in this study is indicated by a dot. Chinese PSV isolates are indicated by red boxes. Foreign PSV isolates are indicated by green boxes. All other members of the genus Sapelovirus are indicated by blue boxes walls, serous exudation, and inflammatory cell infiltration in the interalveolar septa (Fig. 5D). There was no significant change in the control group ( Fig. 5A and B).
To analyze the excretion and distribution of PSV, the viral load in faecal samples collected daily and tissue samples obtained after necropsy of piglets was determined using qRT-PCR. The results showed that there were high RNA copy numbers in faecal samples at 3-6 dpi (Fig. 6A). In addition, PSV was found to be widely distributed in lung and intestinal tissues, but it was not detected in the heart, liver, or kidney ( Fig. 6C and D). The results of enzyme-linked immunosorbent assay (ELISA) showed that the IgG antibody levels in PSVinfected pigs were significantly higher than in uninfected pigs, and they increased significantly at 6 dpi (Fig. 6B). These results confirmed the pathogenicity of SHCM2019 to piglets.

Discussion
PSV is widespread in the pig population in China. In this study, a PSV isolate from a pig farm in Shanghai, SHCM2019, was identified and sequenced. Its genome is 7,567 nucleotides in length, containing a 5'-UTR of 465 nucleotides, a 3'-UTR of 106 nucleotides, and a single ORF of 6,996 nucleotides that encodes a polyprotein precursor of 2,331 amino acids. The length of the 5'-UTR is variable among PSV strains. According to previous research, the 5′-terminal residues of a complete picornavirus genome should be UU [5], but the two 5′ terminal residues were AC for the SHCM2019 strain, which is not accordance with those sequence features. The antigenic Fig. 3 Identification of a possible recombination event in the polyprotein gene of the SHCM2019 strain. A Similarity analysis of the polyprotein gene of potential recombinants using the SimPlot program with a 200-bp window size, 20-bp step size, and 100 bootstrap replicates. SHCM2019 was used as the query sequence, and HeB04 (red line) and ISU-SHIC (blue line) were identified as potential parental strains. B Genome origin and recombination breakpoint identification in the SHCM2019 genome. The cutoff value in the bootstrap test (> 70%) is indicated by a dotted line, and the arrows indicate puta-tive recombination breakpoints. The red area represents the fragment derived from ISU-SHIC, the blue area represents the fragment derived from HeB04, and the white area is of unknown origin. C Evolutionary relationships between the SHCM2019 strain and the putative parental strains based on the polyprotein gene. Phylogenetic trees were generated by the distance-based neighbor-joining method in the RDP4 program. D The average p-values of the recombination events detected by the nine methods in RDP4 epitope of the VP1 protein of SHCM2019 showed high sequence similarity to those of other PSV strains (data not shown), indicating slow evolution, which is useful information for disease prevention.
Phylogenetic analysis based on the polyprotein gene revealed that Chinese PSV isolates are closely related to the South Korean strains, with 87.2%-89.8% nucleotide sequence identity, but they are distantly related to PSV isolates from Europe, with 78.7%-86.4% nucleotide sequence identity, suggesting a degree of territoriality of PSV strains ( Supplementary Fig. S1). When we constructed an evolutionary tree based on the VP1 gene (Supplementary Fig. S2), the topological structure and phylogenetic branching were slightly different from those of the ORF tree and a VP1 tree reported previously [24], indicating that PSV could not be divided into genotypes. At present, many picornaviruses are genotyped by VP1 sequence alignment [25]. The classification criterion is that the VP1 sequence difference between isolates is greater than 25% [26]. Among PSVs, the nucleotide and amino acid sequence identity values for the VP1 region were >70.8% and >75.1%, respectively, and these values were higher than those of porcine enteroviruses (>53.9% and >53.3%, respectively) [24]. This might contribute to the difficulty in genotyping PSV strains.
The driving forces of viral evolution include gene mutation, natural selection, genetic drift, and viral gene recombination. The results of recombination analysis suggested that recombination had occurred among PSV strains; however, the recombination site was located in the 3D region in the SHCM2019 strain, which is not consistent with the results of previous studies [24,[27][28][29]. The 3D protein is a virusspecific RNA polymerase [3], and recombination in this gene would be expected to affect viral replication. In view of the diversity of recombination sites reported to date [24,[27][28][29], we must accelerate the process of research on the molecular mechanisms of PSV evolution.
The SHCM2019 strain was found to replicate in the intestinal tracts of piglets and cause diarrhoea and, to a degree, pneumonia, in pigs, suggesting that the SHCM2019 strain is pathogenic. Clinically, like that of porcine circovirus type 2 (PCV2), the rate of infection with PSV is high in both domestic and wild pigs [9], but in most cases, pigs infected with PSV do not exhibit clinical signs [8,9,14,20], suggesting that this virus is not consistently pathogenic. For this reason, PCV2 and PEDV are usually the main focus of most group were thick and elastic. E In PSV-infected group, the intestines thinned and its contents were pulpy. C The lung performance of piglets in the healthy control group was normal. F Local bleeding points appeared in the lungs of piglets in the PSV-infected group  Pathogenicity of the SHCM2019 strain in neonatal piglets. A High levels of viral RNAs were shed in faecal samples of PSV-infected pigs (L1 and L2) at 3-6 dpi, and faecal swabs from DMEM-inoculated piglets (C1 and C2) were negative for PSV viral RNA. B IgG antibody levels were measured in infected and control pigs, and antibody levels increased significantly after 6 dpi. C-D Viral RNA was detected in some organs. PSV was detected in duodenum, jejunum, ileum, caecum, colon, rectum, lung, brain, and intestinal lymph, but not in the heart, liver, or kidney researchers when large-scale diarrhea occurs in pig populations, although PSV, which can cause similar symptoms, does not attract as much attention [30,31]. It remains to be investigated whether PSV requires currently unidentified factors to achieve full pathogenicity in pigs.

Conclusion
In this study, a swine sapelovirus strain was isolated and named SHCM2019. The isolated strain was confirmed as PSV by RT-PCR, IFA, and TEM assays. The molecular characteristics and pathogenicity of the SHCM2019 strain were studied. Phylogenetic analysis showed that it belongs to the China cluster of PSV. Recombination analysis indicated that there may be a recombination breakpoint upstream of the 3D region of the genome. SHCM2019 was confirmed to be pathogenic. To better assess the adverse effects of PSV on pig populations, we need to further explore the epidemiological characteristics and pathogenic mechanisms of PSV.