Molecular characterization and pathogenicity evaluation of enterovirus G isolated from diarrheic piglets

ABSTRACT Four enterovirus G (EV-G) strains were isolated from diarrheic piglets that were negative for common swine enteric viruses. The spherical enterovirus particles of roughly 30-nm diameter were observed under transmission electron microscopy by using plaque-purified enterovirus. The complete genome sequence analysis revealed that each of four enteroviruses contained a papain-like cysteine protease (PLCP) gene between the 2C and 3A junction regions of the viral genome. This insertion encoded a predicted protease similar to the PLCP of porcine torovirus. The phylogenetic analysis based on complete genome with and without PLCP gene revealed that the four isolated EV-G strains were grouped together with global enterovirus G1-PLCP strains, and more closely related to EV-G/PLCP strains previously detected in China, Japan, and Korea (90.3%–92.2% similarities based on nucleotides). The cell susceptibility test demonstrated that the isolated EV-G could infect and replicate in cell lines from various host species. Furthermore, pathogenicity evaluation showed that the isolated EV-Gs induced mild diarrhea, pyrexia, and reduced body weight in infected piglets. The epidemiological investigation revealed a high prevalence of EV-G in swine herds. Together, our findings demonstrate that the isolated EV-G is pathogenic in piglets and may be advantageous in providing more trustworthy data on the evolution and pathological properties of EV-G. IMPORTANCE Enterovirus G is a species of positive-sense single-stranded RNA viruses associated with several mammalian diseases. The porcine enterovirus strains isolated here were chimeric viruses with the PLCP gene of porcine torovirus, which grouped together with global EV-G1 strains. The isolated EV-G strain could infect various cell types from different species, suggesting its potential cross-species infection risk. Animal experiment showed the pathogenic ability of the isolated EV-G to piglets. Additionally, the EV-Gs were widely distributed in the swine herds. Our findings suggest that EV-G may have evolved a novel mechanism for broad tropism, which has important implications for disease control and prevention.

Swine are susceptible to a wide range of enteric viral and bacterial species, caus ing the global pork industry substantial economic losses.In majority of pig-produc ing countries, diarrhea is frequently caused by single or mixed infections of enteric coronaviruses, such as porcine epidemic diarrhea virus (PEDV), transmissible gastroen teritis virus (TGEV), and porcine deltacoronavirus (PDCoV) or rotaviruses, which are frequently identified in veterinary diagnostic laboratories using a reverse transcription PCR (RT-PCR)-based approach (29,30).In a few clinical cases, none of the above-men tioned viral pathogens were detected in diarrheic pigs, indicating that other viruses are involved in swine gastroenteritis.Since the advent of viral metagenomics (31,32), genomic detection and viral isolation methods were used to identify that those diarrheic pigs were positive for kobuvirus, astrovirus, sapelovirus, or EV-Gs but negative for enteric coronaviruses and rotavirus (33)(34)(35).In addition, the co-circulation of numerous EV-G genotypes, G1, G2, and G17, have been identified in fecal samples from diarrheic swine in the USA, Belgium, Japan, South Korea, and China.In contrast, studies regarding EV-G isolation are limited, and most of the whole-genome sequences of EV-G have been obtained by next-generation sequencing (36,37).In this study, we have reported the prevalence, characterization, and pathogenicity evaluation of swine enterovirus G.

Isolation and identification of EV-G strains
Marc145 cells were inoculated with four diarrheic fecal specimens that were positive for EV-G but negative for other common enteric viruses.After three blind passages on Marc145, EV-Gs developed cytopathic effect (CPE), characterized by cell rounding, shrinking, and detaching, compared to uninfected control cells (Fig. 1A).After four rounds of plaque purification, the purified virus developed a uniform and clear plaque under an agar overlay medium (Fig. 1B).After ultracentrifugation, the plaque-puri fied viruses were examined under transmission electron microscopy (TEM).Spherical, non-enveloped picornavirus-like particles of approximately 25-30 nm in diameter were observed (Fig. 1C).The RT-PCR and whole-genome sequencing confirmed that the isolated viruses were EV-G.To determine the growth kinetics of the EV-G isolates, Marc145 cells were inoculated with the purified virus at an multiplicity of infection (MOI) of 0.01, and the mean titers of three independent measurements at the indicated time points were measured.The results showed that the virus replication reached a peak of 10 6.5 50% tissue culture infectious dose (TCID 50 )/mL at 24 h, suggesting that the multiplication cycle of the isolated virus is completed within 24 h (Fig. 1D).

Phylogenetic analysis of isolated EV-G strains
The phylogenetic tree based on entire genome sequences was constructed using sequences of the isolated EV-G strains and representative EV-G strains from the NCBI database (Table S3).The results showed that EV-G strains were phylogenetically classified into four clades (clades I, II, III, and IV), and EV-G sequences retrieved in the current study were grouped together with EV-G1-PLCP strains (Fig. 3A).Moreover, EV-G strains isolated herein have a common ancestor within clade I that included EV-G3, EV-G9, and EV-G10 genotypes according to the results, but are phylogenetically distinct from prototypic G1 strains, which were classified under clade III.Interestingly, phylogenetic analysis of the entire genome, omitting the PLCP sequence from all recombinant strains, showed the same topological structure (Fig. 3B).These findings indicate that the insertion is the result of viral recombination, and substantial nucleotide differences between related EV-G genotypes have arisen throughout the whole genome.
The molecular classification of EV genotypes is solely based on capsid-encoding sequence (VP1) (42,43).Therefore, phylogenetic analysis based on EV-G VP1 sequence of isolated EV-G strains and reference sequences was carried out.The EV-G strains isolated during this study were grouped with the genotype (G1) and closely related with chimeric EV-G1 strains (Fig. 3C).Furthermore, phylogenetic analysis of the PLCP genes of EV-Gs and coronaviruses revealed that the PLCP gene of EV-G isolated in this study formed a cluster with PLCPs of other EV-Gs (Fig. 3D) and more closely related to the ToVs, but were distantly related to strain of swine, bovine, and equine nidoviruses, showing lower sequence similarities (46%-58.7%) in the aa sequence.

Cell susceptibility test of the EV-G isolate
To examine the host tropism of the isolated EV-G strains, 18 cell lines derived from various species were tested with the CH/HLJ-214/G1PLCP/2020 strain.The result revealed that 14 out of 18 cell lines, namely, porcine (PK15, ST, and IPEC-J2), human (Huh7, HepG2, and Hela), monkey (Marc145 and Vero E6), feline (CRFK), rabbit (RK13), hamster (BHK21 and CHO), duck (DEF), and chicken (DF-1) cell lines showed CPEs at 24-h post-infection (hpi), whereas the human (293T and A549), bovine (MDBK), and canine (MDCK) cell lines did not show CPEs (Table 4).Evidence of infection ability of the isolated EV-G strain in the susceptible cell lines was confirmed by the detection of viral VP1 protein expression by using immunofluorescence assay (IFA).As shown in Fig. 4, the VP1 protein was expressed in all tested cell lines except the human (293T and A549), bovine (MDBK), and canine (MDCK) cell lines, which was consistent with the CPE results.

Pathogenicity evaluation of the EV-G isolate
To determine whether the isolated EV-G is pathogenic in piglets, three 2-week-old specific-pathogen-free (SPF) piglets were inoculated with CH/HLJ-214/G1PLCP/2020 strain.After virus inoculation, all three EV-G-inoculated piglets showed temporary diarrhea.Two piglets exhibited pyrexia (39.5-40.5°C)at 2 days post-inoculation (dpi), which persisted for 2-7 days (Fig. 5A).Similarly, the same two piglets gained less weight at 4-7 dpi compared to the control piglet (Fig. 5B).Daily fecal swabs and tissue samples from all piglets were collected to investigate virus shedding and distribution using RT-PCR.The results demonstrated that EV-G was detected in fecal samples of all three infected piglets at 2-7 dpi but not in the control piglet (Fig. S2a).Furthermore, viral RNAs were detected in most tested tissues including blood, spinal fluid, tonsil, pancreas, ileum, colon, rectum, cerebrum, and cerebellum, but not in lung, liver, and kidney from the infected pigs (Fig. S2a and b).The necropsy results showed that the wall of small intestine of EV-G-infected piglets became thinner and translucent; the intestinal contents were mushy or watery; and there was obvious congestion compared to the control piglet (Fig. S3a); mild hyperemia was observed in the cerebrum of infected piglets (Fig. S3b).Tissues of infected animals were also histopathologically examined, and results showed edema, a small amount of degeneration, and necrosis of mucosal epithelial cells in the lamina propria and submucosa of the cecum and colon.Similarly, glial cell hyperplasia, neutrophil infiltration, and some neuronal cell degeneration and necrosis, which led to an increase in nearby glial cells, were observed in the cerebrum of infected piglets; however, there was no significant change in the control group (Fig. 6).Furthermore, the virus re-isolation from EV-G-infected intestinal contents showed obvious CPE on Marc145 cells at 24 hpi, which was confirmed by IFA with the specific antibody against VP1 of EV-G as shown in Fig. 7A and B.

Epidemiological investigation of EV-Gs
To assess the frequency of EV-G infection among swine, a total of 232 porcine fecal samples, collected from 134 diarrheic and 98 non-diarrheic pigs, were examined by RT-PCR.As shown in Table 5, out of 232 samples, 37.5% (87 of 232) samples were positive for EV-G; of these EV-G-positive pigs, 44% (59 of 134) were diarrheic and 28.6% (28 of 98) were non-diarrheic.Furthermore, EV-G was identified in 21.4% (19 of 89) of the suckling pigs, 52.9% (45 of 85) of the nursery pigs, and 39.7% (23 of 58) of the fattening pigs.The EV-G-positive rates in the nursery and fattening groups were significantly higher than those in the suckling piglets.The data suggest that the prevalent rate of EV-G was clearly higher in diarrheic animals than in non-diarrheic animals.

DISCUSSION
Recently, the identification and characterization of a natural recombination EV-Gs with insertion of the PLCP gene in their genomes from stool samples of swine with diarrhea have been increasingly reported (26,(38)(39)(40)(41)44), indicating that EV-G-PLCP might be the potential causative agent of porcine diarrhea.The viral evolution and recombination events may have a significant impact on a variety of factors, including the emergence of novel virus variants, changes in host ranges and tissue tropism alterations, increases in virulence, and evasion of host immunity (45)(46)(47).A previous study has reported that ToV-PLCP might have an impact on EV-G pathogenesis by functioning as an innate immune response antagonist (39).Thus, the finding of PLCP gene insertions in EV-G may increase the potential public health risks.In this report, four recombinant EV-G strains, named CH/HLJ-141/G1PLCP/2020, CH/HLJ-214/G1PLCP/2020, CH/HLJ-312/G1PLCP/2020, and CH/HLJ-315/G1PLCP/2020, were isolated from diarrheic piglets.The genetic and biological characteristics of isolated viruses were determined by using several techniques including RT-PCR, sequencing, CPE, IFA, replication kinetics, and TEM.Furthermore, the pathogenicity of isolated EV-G strains in piglets was evaluated.
Since four isolated viruses have relatively similar characteristics, further experiments were carried out with the representative strain CH/HLJ-214/G1PLCP/2020.The epidemio logical data on EV-G infections are limited; thus, this study also provided novel informa tion about the prevalence of EV-G in diarrheic and non-diarrheic swine.Most studies reported that EV-Gs cause asymptomatic infections in swine; for example, in Germany, Hungary, Japan, and Vietnam, EV-G3, EV-G4, EV-G8, EV-G9, and EV-G10 were detected in healthy pigs (21,23,26,44,48).In contrast, our result revealed that the overall prevalence of EV-G was much higher in diarrheic than in non-diarrheic animals.In addition, the highest prevalence was found in nursery pigs followed by fattening pigs, which was consistent with a study from Thailand (27).Moreover, several studies have reported that EV-G is prevalent in swine populations, and infection is detected at higher frequencies in younger pigs compared to adults, which might be attributed to acquired immunity (23,24,49,50).The phylogenetic analysis based on structural protein sequences is widely used to determine picornavirus taxonomy (51,52).To understand the molecular character istics of the isolated EV-G strains, complete genomes were sequenced and phyloge netic relationships among representative EV-G strains were determined.The sequence analyses reveal that all four isolated EV-Gs contain a PLCP gene of 639 nt within the 2C/3A junction region of their genome and encode a protein similar to torovirus PLCP, which resembles the picornavirus leader protease.Consistent with our study, recombi nant EV-G strains (EV-G1, EV-G2, and EV-G17) carrying PLCP in the junction region 2C/3A of their genomes have been detected in pigs with diarrhea in the USA, Belgium, Japan, South Korea, and China (26, 38-41, 44, 53).Interestingly, Wang et al. have identified one recombinant EV-G strain carrying the torovirus PLCP gene, completely replacing the viral capsid protein gene region (VP4/VP2/VP3/VP1) in pigs in China (41).The previous study has reported that the 3C protease of EV-G cleaves the viral polyprotein precursor at the C-terminus of 2C protein and N-terminus of 3A proteins by using the cleavage sequence ALFQ↓GPPT (54).Moreover, Shang et al. have demonstrated that the chimeric EVG 08/NC_USA/2015 expressing PLCP using the reverse genetic technique can produce the exogenous PLCP protein at the cleavage sites of ALFQ↓GPPV and AEFQ↓GPPT in the virus-infected cells (39).According to the sequence similarity of the recombinant EV-G strains, the possible cleavage residues flanking PLCP includes GPPT↓ALFQ, GPPA↓ALFQ, and GPPE↓ALPQ (26,38,44).In this study, the PLCP in the isolated recombinant EV-G strains is flanked by the predicted viral 3C protease cleavage sequences, ALFQ↓GPPA and AVFQ↓GPPT, at its N-terminus and C-terminus, respectively, indicating that the viral 3C protease of the recombinant EV-G isolates can process the functional PLCP protein.
The phylogenetic analysis based on complete genome revealed that the EV-G strains isolated in this study grouped together with global G1-PLCP strains and more closely related to EV-G/PLCP strains previously detected in China, Japan, and Korea with (90.3%-92.2%nucleotide similarities) than other recombinant EV-Gs, implying that the isolated EV-Gs here may have originated from the common ancestor.However, all the recombi nant G1-PLCP strains, including four strains isolated here, were not phylogenetically close to the G1 strains but were grouped together with the large clade of enterovi ruses containing EV-G genotypes G3, G9, and G10.Additionally, phylogenetic analysis of all recombinant strains excluding the PLCP gene resulted in the same evolutionary tree.These data suggested that the PLCP insertion has occurred through cross-order recombination, and the nucleotide variations of EV-G entire genome have arisen from the accumulation of single-base changes or the recombination of different genotypes during viral propagation.Phylogenetic analysis of PLCP genes showed that the PLCP gene of EV-G isolated here are more closely related to that of the Japan/LC316778 strain (92.5% aa similarity), establishing a cluster with PLCPs from other EV-Gs.Despite the pathogenicity of EV-Gs in swine remaining debatable, all the isolated EV-G/PLCP strains here have been identified from fecal samples of swine with diarrhea, indicating that EV-G-PLCP is the potential causative agent of diarrhea in swine.
PLCP was previously shown to decrease the host cells' innate immune response when introduced into the EV-G genome (39), which, under specific conditions, may allow them to demonstrate their pathogenic potential.Furthermore, it has been noted that recombinant events in the EV-G2 and EV-G17 genotypes are infrequent, but they appear to occur more frequently in the EV-G1 genotype, which may play a key role in virus evolution (40).Given the fact that human EVs are frequently discovered in several mammalian species, it is plausible that EVs which naturally circulate in animal populations may potentially infect human populations as well.Due to strain diversity, a high rate of mutation, prolonged subclinical shedding, low infectious doses, and frequent genome recombination of enterovirus (55), there is a significant potential for the emergence of new strains that can infect and replicate in a wide range of hosts.To evaluate the potential cross-species infection ability of EV-G, 18 cell lines derived from various host species, including human, swine, monkey, hamster, bovine, dog, cat, rabbit, chicken, and duck, were subjected to a susceptibility study.Previous reports showed that EV-G can be propagated in BHK-21, Vero, ST, and Marc-145 cells in the presence of trypsin (13,56).Our results demonstrated that EV-G could infect and replicate in a wide range of cell lines including swine (PK15, ST, and IPEC-J2), human (Huh7, HepG2, and Hela), monkey (Marc145 and Vero E6), feline (CRFK), rabbit (RK13), hamster (BHK21 and CHO), duck (DEF), and chicken (DF-1).These various levels of susceptibility should help define the EV-G receptors.Additionally, the wide range of cells that the isolated EV-G strains can infect in vitro suggests that EV-Gs may pose a risk of cross-species infection.
Although the EV-Gs are frequently isolated from healthy pigs and have not been proved to cause disease, some reports indicate that EV-G infection in infected animals may induce clinical symptoms such as dermatitis, reproductive problems, neurological problems, and diarrhea (23,57,58).However, there is minimal evidence that EV-G infection causes clinical diarrhea.To date, the pathogenicity of EV-G has been poorly studied, especially the recombinant EV-G strains, and only two EV-G trains have been subjected to pathogenicity assessment (24,34).Since CH/HLJ-214/G1PLCP/2020 can infect broad types of cell lines from different species in vitro and the predicted precursor polyprotein of CH/HLJ-214/G1PLCP/2020 is relatively close to other EV-G isolates in China, therefore, further animal experiments were carried out with the representative strain CH/HLJ-214/G1PLCP/2020.To determine the pathogenicity of the recombinant EV-G strain isolated in the present study, four SPF piglets were infected.The results revealed that two piglets showed mild diarrhea, pyrexia, and lower body mass index.Additionally, most tissues of the infected piglets carried EV-G, which could be detected at 7 dpi.Even though no piglets died during the entire experimental infection process, it is important to note that the EV-G strains isolated herein could negatively impact the growth rate of piglets by reducing weight gain.The results of tissue histological examination revealed that isolated EV-G could cause pathological changes in the cecum, colon, and cerebrum, which is similar to the animal experiment by using EV-G stains without the PLCP gene (24,34).These findings support that the EV-G isolate carrying ToV PLCP gene is pathogenic, which might be one of the causative agents for swine diarrhea.
In summary, this research will provide trustworthy data on the biological properties, evolution, and pathogenicity of recombinant EV-G strain isolated from diarrheic piglets.The fact that isolated EV-Gs can infect cell lines from different species suggests that they may pose a risk of cross-species infection.Therefore, it is important to recognize that EV-Gs may have the potential to infect multiple species, and further research is needed to understand this risk.

Sample collection
In July 2020, 15 fecal samples of piglets with diarrhea were collected from swine farm in northeast of China.In addition, a total of 232 fecal samples (with or without clini cal diarrhea) were collected from different swine age groups (suckling, nursery, and fattening) in northeast of China (Table 6).All samples were stored at −80°C until use.

Isolation of EV-G
To identify the causative agents of swine diarrhea, fecal samples were diluted in phosphate-buffered saline (PBS) to prepare a 10% fecal suspension.The fecal suspen sions were clarified by centrifugation and filtered through a 0.22-µm filter (Merck Millipore, Burlington, MA, USA).The samples initially were examined for common swine enteropathogenic viruses, including TGEV, PEDV, PDCoV, and rotavirus genogroup A using RT-PCR.Five out of 15 samples were tested negative for these enteric viral pathogens.The negative samples were further tested for porcine astrovirus (AstV), porcine sapovirus (PSaV), porcine kobuvirus (PKV), porcine sapelovirus (PSV), and EV-G using virus-specific primers as previously described (59)(60)(61)(62).Only four samples were tested positive for EV-G alone.These four samples were then subjected to virus isolation.In brief, 1 mL of filtered fecal supernatants from each sample was pre-treated with trypsin (Gibco) at a final concentration of 20 µg/mL for 1 h, diluted at 1:1 ratio with complete DMEM (containing 1% penicillin-streptomycin), and then inoculated onto confluent Marc145 cells for 1 h.The inoculum was discarded; cell monolayers were washed twice with PBS and covered with fresh DMEM supplemented with 5 µg/mL trypsin; and observed daily for CPEs.Four days post-infection, cells were lysed by freeze-thawing three times and re-inoculated into respective Marc145 cells for three passages; at the same time, the growth of the EV-G in culture was further confirmed by RT-PCR with specific primers after every passage.

Viral plaque assay
To purify the isolated viruses, a plaque assay was performed as previously described with slight modifications (63).In brief, Marc145 cells were inoculated with serially diluted viruses and overlayed with agarose medium.After the overlaid medium was solidified, the plates were incubated at 37°C and 5% CO 2 .Plaques were allowed to develop for 5-7 days, and uniform plaques were picked and re-inoculated into cell monolayers to amplify the positive clones.After four rounds of plaque purification, the purified virus clones were successfully obtained.The monolayers were also fixed and stained with crystal violet to visualize the viral plaques.

TEM
The transmission electron microscopy was carried out according to the previous publication (63).The cell supernatants of plaque-purified viruses were harvested after three times freezing-thawing, and cell debris was clarified by centrifuging at 10,000 rpm for 30 min at 4°C.The supernatants were passed through a 0.22-µm filter and pelleted by centrifuging at 35,000 rpm for 4 h (Beckman SW32Ti rotor).The resulting pellet was re-suspended in DMEM and centrifuged through 20%-50% sucrose cushion at 35,000 rpm for 4 h.The white opalescent band at the interface formed by the virus particles were collected and re-centrifuged at 35,000 rpm for 3 h to get rid of sucrose and concentrate samples.The purified virus was examined under TEM (Hitachi Model H-7650).

Virus growth kinetic
To analyze the growth properties of isolated virus, the growth kinetics of isolated virus was evaluated in the same cell line used for virus isolation to avoid influences of adaptive mutations acquired through the virus isolation.In brief, Marc145 cells were cultured in 12-well plates to 90%-100% confluence; cells were washed twice by PBS after discarding the growth medium; the cells were infected in triplicate with EV-G at an MOI of 0.01 and incubated at 37°C and 5% CO 2 for 1 h.After the incubation period, the inoculum was discarded and cells were washed as above, then 1 mL of maintenance media was added and cells were incubated at 37°C and 5% CO 2 .Cell supernatants were harvested at 1, 6, 12, 24, 36, 48, 60, and 72 h post-inoculation and stored at −80°C until use.Virus titers were measured in Marc145 cells by TCID 50 according to the Reed-Muench method (64).

Whole-genome sequencing and analysis
The EV-G supernatants obtained from plaque-purified clones were clarified by centri fuging at 10,000 rpm for 30 min to remove cell debris.Total RNA was extracted using the TIANamp RNA Kit and cDNA was made using the PrimeScript Synthesis Kit (Takara, Dalian, China).The cDNA libraries were prepared from total RNA extracted from each isolate using the Agencourt AMPure XP-Medium kit (A63881, Beckman Coulter, USA) according to the manufacturer's instructions.The product was validated on the Agilent Technologies 2100 Bioanalyzer for quality control.The final qualified libraries were subjected to next-generation sequencing using BGISEQ-500 Sequenc ing System (BGI-Shenzhen, China).Based on the reference genome of Sus scrofa (Sscrofa11.1(GCA_000003025.6)), the high-quality non-host reads that matched to the candidate viruses were processed by eliminating host contamination reads using BWA and SOAP (65).The high-quality reads were then de novo assembled by using IDBA (66), SPAdes (67), and Edena (68) software.The assembled contigs were then analyzed by BLAST-based approach to identify the virus species (69).
To further understand the molecular characteristics of isolated EV-G strains, the complete ORF and cleavage sites of isolated EV-G were predicted by comparing with the genome of previously known EV-G strains.Furthermore, the nucleotide and aa sequences of the whole genome, VP1, and PLCP genes were aligned with reference EV-G strains using Clustal W software (70).MEGA v.6.0 software was used to analyze the phylogenetic trees of whole-genome, VP1, and PLCP sequences via the maximum-likeli hood method with the neighbor-joining method with Kimura 2-parameter and 1,000 bootstrap replicates (71,72).

Cell susceptibility test to the EV-G
To determine the viral tropism, 18 cell lines derived from different species were subjected to the isolated EV-G strain and observed for CPEs.An IFA was then performed to determine the production of viral VP1 protein as described previously (63).Briefly, 18 cell monolayers were inoculated with the EV-G for 12 h and then fixed with paraformalde hyde.The fixed cells were stained with mouse polyclonal antibody specific for EV-G-VP1 protein and FITC-conjugated goat anti-mouse IgG (Thermo Fisher Scientific).The cell monolayers were then stained with DAPI (Solarbio, China) and visualized under a digital inverted fluorescence microscope (Evos, FL, USA).

Animal experiments
To evaluate the pathogenicity of the isolated EV-G, four 2-week-old SPF piglets, negative for EV-G, ASFV, PRRSV, PCV2, PSV, PEDV, PDCoV, TGEV, rotavirus, PSaV, PKV, AstV and mammalian orthoreovirus in the fecal samples by RT-PCR, were housed in a HEPA-filtered level 2 biosecurity facility and divided into two groups.Piglets in group 1 (named G1, G2, and G3) were challenged orally with 5-mL 5 × 10 6 TCID 50 /mL of CH/HLJ-141/2020 /G1/ PLCP strain each, while piglets in group 2 (named C1) received 5-mL DMEM orally as a negative control.The piglets were monitored daily, and rectal temperature, clinical signs, and weight gain were recorded.In addition, fecal swabs and blood samples were collected at 0-7 dpi, and serum samples were collected for seroconversion tests at 1, 3, and 6 dpi.All piglets were euthanized at 7 dpi, and the tissues from the heart, liver, spleen, lung, kidney, tonsil, cerebellum, cerebrum, submaxillary nodes, inguinal lymph nodes, and alimentary tract, including the stomach, duodenum, jejunum, ileum, caecum, colon, rectum, and mesenteric lymph nodes as well as fecal, blood, and spinal fluids were collected.All samples taken from inoculated piglets were exposed to viral RNA detection to determine the virus shedding and distribution in the infected piglets.Additionally, certain tissue samples were immediately preserved in 10% neutral buffered formalin for histological investigation.

RT-PCR detection of EV-G
To assess the frequency of EV-G infection in pigs, a molecular survey was conducted on 232 stool samples, including feces and fecal swabs collected from 98 non-diarrheic and 134 diarrheic pigs.Fecal samples were diluted in PBS to prepare a suspension of 10% (wt/vol).Rectal swabs were processed by elution into 0.5-mL PBS and centrifuged at 14,000 × g for 10 min.The total viral RNA was extracted from supernatants using RNA extraction kit (Tiangen Biotech, Beijing) according to the manufacturer's instructions.The cDNA synthesis was carried out in a 20-µL reaction mixture using murine leukemia virus (MLV) reverse transcriptase kit (Takara), following the manufacturer's instructions.PCR was used to detect EV-G genome using particular primers based on a conserved sequence within the 5′-UTR as previously reported (61).

FIG 2
FIG 2 Schematic diagram of the genome organization of isolated EV-G.(A) The single ORF is flanked by 5′-UTR (813 nucleotides) and a short 3′-UTR (71 nucleotides), followed by a poly(A) tail.The torovirus (ToV)-PLCP gene is presented as a red box that locates at viral 2C/3A cleavage junction.The flanked sequences of putative 3C protease cleavage sites are shown in an enlarged red box.Vertical lines indicate the polyprotein processing site by the 3C protease.(B) Multiple alignment of the amino acid sequences of the PLCP regions of the recombinant EV-G and ToV strains.

FIG 3 8 FIG 4
FIG 3 Phylogenetic analyses based on nucleotide sequences of the complete genome (A), complete genome excluding insertion PLCP sequences of EV-G strains (B), VP1 gene sequences (C), and phylogenetic analyses based on PLCP genes of EV-G strains and nidoviruses including toroviruses (ToVs) and coronaviruses (CoVs) (D).Multiple sequence alignments were created using the ClustalX v.2.0 program, and the phylogenetic trees were constructed from the aligned nucleotide sequences using neighbor-joining.Hosts of origin, geographical origins, names of the strains, years of isolation, genotypes, and GenBank accession numbers are shown.The genotypes are indicated on the right-hand side.Solid diamonds denote the recombinant EV-G1-PLCP strains identified in this study.Scale bars indicate nucleotide substitutions per site.

FIG 5
FIG 5 Animal experiments.(A) Trends of rectal temperatures of four 21-day-old piglets inoculated with isolated EV-G.Normal rectal temperature was present in the mock (C), while the experimentally inoculated pigs (G1, G2, and G3) exhibited pyrexia (39.5-40.2°C)at 2 days post-inoculation, which persisted for 2-6 days.(B) Trends weight gain of piglets inoculated with EV-G.The weight gain rates in piglets from the infected group (G2 and G3) were less than those in control piglet (C).

TABLE 1
Homology comparison of the whole genome of the isolated EV-Gs with reference strains a

TABLE 1
Homology comparison of the whole genome of the isolated EV-Gs with reference strains a (Continued) a nt, nucleotide sequence identity; WGS, whole-genome sequence; -, not available.TABLE 2 Comparison of PLCP of the isolated strains with chimeric EV-G and ToV reference strains

TABLE 3
Comparison of full-length genomes of the isolated strains with EV-G reference strains

TABLE 4
Summary of cell lines and their susceptibility to porcine sapelovirus infection as determined by CPE and IFA a a N/A, not available; +, infection; −, no infection or obvious lesion.

TABLE 5
Prevalence of EV-G infection in diarrheic and asymptomatic animals

TABLE 6
Fecal sample background of EV-G