Characterization and Pathogenicity of the Porcine Deltacoronavirus Isolated in Southwest China

Porcine deltacoronavirus (PDCoV) is a newly emerging enteric pathogen in swine that causes diarrhea in neonatal piglets and creates an additional economic burden on porcine industries in Asia and North America. In this study, a PDCoV isolate, CHN-SC2015, was isolated from Sichuan Province in southwest China. The isolate was characterized by a cytopathic effect, immunofluorescence, and electron microscopy. CHN-SC2015 titers in LLC-PK cells ranged from 104.31 to 108.22 TCID50/mL during the first 30 passages. During serial passage, 11 nucleotide mutations occurred in the S gene, resulting in nine amino acid changes. A whole genome sequencing analysis demonstrated that CHN-SC2015 shares 97.5%–99.1% identity with 59 reference strains in GenBank. Furthermore, CHN-SC2015 contained 6-nt deletion and 9-nt insertion in the ORF1ab gene, 3-nt deletion in the S gene and 11-nt deletion in its 3′UTR compared with other reference strains available in GenBank. A phylogenetic analysis showed that CHN-SC2015 is more closely related to other PDCoV strains in China than to the strains from Southeast Asia, USA, Japan, and South Korea, indicating the diversity of genetic relationships and regional and epidemic characteristics among these strains. A recombination analysis indicated that CHN-SC2015 experienced recombination events between SHJS/SL/2016 and TT-1115. In vivo infection demonstrated that CHN-SC2015 is highly pathogenic to sucking piglets, causing diarrhea, vomiting, dehydration, and death. Virus was shed daily in the feces of infected piglets and upon necropsy, was found distributed in the gastrointestinal tract and in multiple organs. CHN-SC2015 is the first systematically characterized strain from southwest China hitherto reported. Our results enrich the body of information on the epidemiology, pathogenicity and molecular evolution associated with PDCoV.


Introduction
Porcine deltacoronavirus (PDCoV), belongs to the genus Deltacoronavirus within the family Coronaviridae [1]. It is an emerging swine enteric virus that causes diarrhea, vomiting, dehydration, and death in nursing piglets and the mortality rates are about 40%-80% [2]. The earliest identification proved to be PDCoV-positive by RT-PCR of the M gene. Prior to virus isolation, the samples were homogenized in 5 mL of Dulbecco's modified Eagle's medium (DMEM) (Gibco, Carlsbad, CA, USA) supplemented with 1% antibiotic-antimycotic (Solarbio, China), subjected to three cycles of freezing and thawing, then centrifuged at 5000 rpm for 20 min at 4 • C. Supernatants were filtered through a 0.45 µm sterile filter and stored at −80 • C.

Cells and Antibodies
LLC porcine kidney cells (LLC-PK) were maintained in DMEM supplemented with 10% fetal bovine serum (FBS) (PAN, Germany) and 1% antibiotic-antimycotic at 37 • C in a humidified atmosphere of 5% CO 2 . The rabbit anti-PDCoV N monoclonal antibody was prepared in our laboratory. FITC-conjugated goat anti-rabbit IgG was purchased from Bioss.

Virus Isolation and Propagation
For virus isolation, LLC-PK cells at 90% confluence in T25 flasks were washed three times with DMEM supplemented with 1% antibiotic-antimycotic and 7.5 µg/mL trypsin-EDTA (henceforth referred to as maintenance media) then inoculated with 1 mL of the filtered inoculums. The virus was allowed to adsorb 2 h at 37 • C, then 7 mL of maintenance medium was added to each flask. The cells were cultured continuously at 37 • C until a cytopathic effect (CPE) was observed. The infected cell suspensions were subjected to three cycles of freezing and thawing then centrifuged at 8000 rpm for 10 min at 4 • C. Supernatants were collected and stored at −80 • C for the subsequent passage.
For subsequent passage, LLC-PK cells at 90% confluence in T25 flasks were washed three times with maintenance media and then inoculated with 1 mL of CHN-SC2015. After incubation for 2 h at 37 • C, 7 mL of maintenance medium was added to each flask. Upon a >80% CPE being observed, the flasks were subjected to three cycles of and thawing then centrifuged at 8000 rpm for 10 min at 4 • C. Supernatants were collected and stored at −80 • C.

TCID 50 Assay
TCID 50 assay was performed using LLC-PK cells according to the method described by Dong et al. [13]. Briefly, the confluent cell monolayers seeded in 96-well plates were washed twice with maintenance medium and then inoculated with 100 µL of 10-fold serially diluted PDCoV; At each dilution, there were eight technical replicates. An amount of 150 µL of maintenance medium was added to each well after the cells and the virus had incubated for 1.5 h at 37 • C. The CPE was observed for 4 days and was analyzed by the methods of Reed & Muench [31].

Immunofluorescence Assay (IFA)
LLC-PK cells in 12-well plates were infected with PDCoV at a multiplicity of infection (MOI) of 0.04. After 24 h, the cells were rinsed then fixed with 4% paraformaldehyde for 30 min at room temperature. Cells were rinsed again and permeabilized with 0.5% Triton-X-100 for 30 min at room temperature, then washed three times with phosphate-buffered saline (PBS) for 5 min and blocked with 2% bovine serum albumin (BSA) for 1 h at 37 • C. Cells were then incubated for 16 h at 4 • C with rabbit anti-PDCoV N monoclonal antibody (1:200), washed again and incubated with FITC-conjugated goat anti-rabbit IgG (1:200) for 1 h at room temperature. Finally, cells were treated with 4 ,6-diamidino-2-phenylindole (DAPI) for 10 min, rinsed, mounted on glass slides and observed with the fluorescence microscope (IX73, Olympus, Japan).

Electron Microscopy
Cultures of PDCoV-infected LLC-PK cells were harvested when CPE was observed, subjected to three cycles of freezing and thawing, then centrifuged at 5000 rpm for 20 min at 4 • C. Supernatants were filtered through 0.45 µm sterile filters then centrifuged at 50,000 rpm for 4 h at 4 • C using the ultracentrifuge MTX150 (Thermo Fisher Scientific, Waltham, MA, USA). Viral particles were resuspended in PBS, negatively stained with 2% sodium phosphotungstic acid and examined with a transmission electron microscope (JEM-1200EX, Olympus, Japan).

RT-PCR/qRT-PCR
Total RNA was extracted from PDCoV-infected LLC-PK cell cultures using a UNlQ-10 Column TRIzol Total RNA Isolation Kit (Sangon, China), then was subjected to RT-PCR/qRT-PCR using primers specific to the S and M gene of the PDCoV. The primer sequences used for RT-PCR to amplify S gene are 5 -CACCAGGACGCCTTCTTGTGAGG-3 and 5 -CTACCATTCCTTAAACTTAAAGGACG-3 . PrimeSTAR Max Premix (Takara, China) was used for RT-PCR reactions and the amplification conditions were 98 • C for 5 min, then 32 cycles of 98 • C for 30 s, 56 • C for 30 s, and 72 • C for 45 s. The PCR products were sequenced by the Sangon company and aligned using DNAMAN 7.0 software. Primer sequences used for qRT-PCR are 5 -CCAATGGGTACATGGAGGT-3 and 5 -GTGGCGGATTTCTAACTGA-3 SYBR Green I Master Mix (Takara, China) was used for the qRT-PCR reactions and the amplification conditions were 95 • C for 30 s, then 40 cycles of 95 • C for 5 s, 55 • C for 30 s, and 72 • C for 30 s using the LightCycler 96 system (Roche, Germany). A standard curve was constructed using 10-fold serial dilutions of pMD18-T containing the PDCoV M gene sequence; the quantity of PDCoV RNA was calculated based on the standard curve and expressed as the copies per milliliter of cell culture. The primer sets were designed using the CH/Sichuan/S27/2012 sequence as a template (GenBank accession No. KT266822.1) and were synthesized by the Sangon company.

Viral Replication Kinetics
Confluent LLC-PK cells in T75 flasks were infected with PDCoV (MOI = 0.04). After adsorption for 1.5 h at 37 • C, the inoculum was removed and 25 mL of the maintenance medium was added to each flask. Culture supernatants were collected at 6, 12, 24, 36, 48, and 60 h post-infection and the quantity of viral RNA was determined using qRT-PCR.

Genetic Evolution and Mutation Analysis
The complete genome sequence of the PDCoV strain CHN-SC2015 from the passage 12 (10 6.35 TCID 50 /mL) was determined by next-generation sequencing (NGS) technology using an Illumina 1.9 platform. Sequences were mapped to the known porcine deltacoronavirus strain CH/Sichuan/S27/2012 and were assessed using the quality control tool FastQC (http://www. bioinformatics.babraham.ac.uk/projects/fastqc/). Sequence alignment, based on the complete genome of the PDCoV strain CHN-SC2015 and 59 available strains deposited in GenBank, was performed using ClustalW in MEGA X software. Phylogenetic analysis of CHN-SC2015 and the 59 reference strains were analyzed using MegAlign in DNAStar Lasergene Version 7. Phylogenetic trees were constructed based on the sequences of the complete genome and the S gene using the maximum likelihood method in MEGA X with a bootstrap analysis of 1000 replicates and visualized with iTOL (https://itol.embl.de/). The aligned sequences were also analyzed using Recombination Detection Program 4 (RDP4) [32] and Simplot 3.5.1 [33] to predict the possible recombination events in PDCoV strain CHN-SC2015.

Pathogenicity of PDCoV Strain CHN-SC2015 in Piglets
Seven five-day-old piglets were randomly divided into two groups; five piglets in the experimental group and two in the negative control group. Fecal samples from all the piglets tested negative for PEDV, TGEV, classical swine fever virus (CSFV), porcine reproductive and respiratory syndrome virus (PRRSV), porcine pseudorabies (PRV), porcine group A rotavirus (PoRV), and porcine circivirus 2 (PCV2) prior to virus infection. The experimental piglets were inoculated orally with 10 mL of DMEM containing 10 6.35 TCID 50 /mL PDCoV CHN-SC2015 from passage 12, the control piglets received 10 mL of maintenance medium orally; experimental and control piglets were housed in separate rooms. All the piglets were monitored daily for clinical signs of infection, including diarrhea, vomiting, and lethargy and all observations were recorded. Three challenged piglets and one control piglet were necropsied at 5 days post-infection (dpi), the remaining piglets were necropsied at 8 dpi. After necropsy, jejuna were fixed in 4% formaldehyde and prepared for histological examination using hematoxylin and eosin (HE) staining.

Statistical Analysis
The experiments were performed in triplicate. Data are shown as the mean ± standard deviation (SD). A one-way ANOVA test was used to measure significant differences between groups. p values < 0.05 were considered statistically significant.

Ethics Statement
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. The animal experiment was approved by the Institutional Animal Care and Use Committee of Sichuan Agricultural University (IACUC#RW2016-090, approval date: 8 September 2016).

Virus Isolation and Identification
PDCoV was successfully isolated from one sample collected from sick piglet, namely PDCoV CHN-SC2015. The CPE of PDCoV CHN-SC2015 in LLC-PK was characterized by cell clustering and rounding, then death 24 h after infection ( Figure 1A). Mock-infected cells ( Figure 1B) remained flat and attached to the dish. IFA, at 24 h post-infection, was used to visualize the presence of the N protein in PDCoV CHN-SC2015. As shown in Figure 1C-H, green fluorescence, from FITC-anti-N antibodies, was observed in infected cells; none was observable in mock-infected cells.

Statistical Analysis
The experiments were performed in triplicate. Data are shown as the mean ± standard deviation (SD). A one-way ANOVA test was used to measure significant differences between groups. p values < 0.05 were considered statistically significant.

Ethics Statement
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. The animal experiment was approved by the Institutional Animal Care and Use Committee of Sichuan Agricultural University (IACUC#RW2016-090, approval date: September 8th, 2016).

Virus Isolation and Identification
PDCoV was successfully isolated from one sample collected from sick piglet, namely PDCoV CHN-SC2015. The CPE of PDCoV CHN-SC2015 in LLC-PK was characterized by cell clustering and rounding, then death 24 h after infection ( Figure 1A). Mock-infected cells ( Figure 1B) remained flat and attached to the dish. IFA, at 24 h post-infection, was used to visualize the presence of the N protein in PDCoV CHN-SC2015. As shown in Figure 1C-H, green fluorescence, from FITC-anti-N antibodies, was observed in infected cells; none was observable in mock-infected cells. Viral particles negatively stained with 2% sodium phosphotungstic acid were examined by EM. Visible in Figure 2 are the crown-shaped projections on the surface periphery, which is the typical topology of coronavirus. The particle diameter is about 60-120 nm. These results demonstrate that the PDCoV CHN-SC2015 was successfully isolated from the intestinal contents of the diarrheal piglet. Viral particles negatively stained with 2% sodium phosphotungstic acid were examined by EM. Visible in Figure 2 are the crown-shaped projections on the surface periphery, which is the typical topology of coronavirus. The particle diameter is about 60-120 nm. These results demonstrate that the PDCoV CHN-SC2015 was successfully isolated from the intestinal contents of the diarrheal piglet.

Viral Replication
To further characterize the growth of strain CHN-SC2015 in LLC-PK cells, TCID50 and qRT-PCR assays were performed every fifth passage for 30 passages (Table 1). Viral titers increased approximately 1000-fold between the fifth and 10 th passages, then increased 10-fold and held relatively steady through the 30 th passage. Viral RNA/ μL increased about 10-fold by the 25 th passage then fell off slightly. These data demonstrate that PDCoV strain CHN-SC2015 actively propagates in LLC-PK cells and that this cell line is suitable for PDCoV isolation and propagation. Note: + indicated that the CPEs were observed in LLC-PK cells. ND, not done.

Viral Replication Kinetics in LLC-PK Cells
qRT-PCR was used to characterize the replication kinetics of CHN-SC2015 passage 17 in LLC-PK cells. RNA was extracted from PDCoV CHN-SC2015 infected LLC-PK cells at different times during a 60-h infection time course. As shown in Figure 3, viral RNA copies gradually increased until peaking at 48 h post-infection, then fell off slightly at 60 h post-infection.

Viral Replication
To further characterize the growth of strain CHN-SC2015 in LLC-PK cells, TCID 50 and qRT-PCR assays were performed every fifth passage for 30 passages (Table 1). Viral titers increased approximately 1000-fold between the fifth and 10th passages, then increased 10-fold and held relatively steady through the 30th passage. Viral RNA/ µL increased about 10-fold by the 25th passage then fell off slightly. These data demonstrate that PDCoV strain CHN-SC2015 actively propagates in LLC-PK cells and that this cell line is suitable for PDCoV isolation and propagation. Note: + indicated that the CPEs were observed in LLC-PK cells. ND, not done.

Viral Replication Kinetics in LLC-PK Cells
qRT-PCR was used to characterize the replication kinetics of CHN-SC2015 passage 17 in LLC-PK cells. RNA was extracted from PDCoV CHN-SC2015 infected LLC-PK cells at different times during a 60-h infection time course. As shown in Figure 3, viral RNA copies gradually increased until peaking at 48 h post-infection, then fell off slightly at 60 h post-infection.

Sequence Variation in the S Gene During Serial Passage
To identify and analyze variations in the S gene of strain CH-SC2015 during serial passage, we amplified the S gene from the original isolate and at every fifth passage by RT-PCR. The nucleotide and amino acid sequences at every passage were compared against the parental sequences using DNAMAN 7.0 ( Table 2). Nucleotide changes were observed at positions 484, 500, 578, 787, 946, 1188, 1443, 1450, 1587, 1922, and 2636. All of these resulted in amino acid substitutions, except for the synonymous mutations at positions 1443 and 1587 (indicated in boldface). Four nucleotide mutations had occurred by passage 5, by passage 30, ten nucleotides were different than parental. Amino acid 162 (indicated in boldface) is a glycosylation site on the surface of the S protein.

Genomic Characterization of PDCoV Strain CHN-SC2015
The complete genome sequence of PDCoV strain CHN-SC2015 is 25,403 nt and was deposited at DDBJ/EMBL/GenBank under accession number MK355396. We compared CHN-SC2015 with 59 other PDCoV strains available in GenBank and found that the level of nucleotide identities ranged from 97.5% to 99.1%; the number of differences in the nucleotide sequences ranged from 606 to 230 ( Figure 4, Table 3). The sequence alignment of the CHN-SC2015 strain revealed a 3-nt deletion (TAA) in the S gene compared to the US, Japan, South Korea, and Southeastern Asia PDCoV strains. This deletion was also present in a number of other Chinese PDCoV strains, indicating that they may have evolved from a common ancestor ( Figure 5B). CHN-SC2015 also contained a 6-nt deletion (TATGAA)

Sequence Variation in the S Gene During Serial Passage
To identify and analyze variations in the S gene of strain CH-SC2015 during serial passage, we amplified the S gene from the original isolate and at every fifth passage by RT-PCR. The nucleotide and amino acid sequences at every passage were compared against the parental sequences using DNAMAN 7.0 ( Table 2). Nucleotide changes were observed at positions 484, 500, 578, 787, 946, 1188, 1443, 1450, 1587, 1922, and 2636. All of these resulted in amino acid substitutions, except for the synonymous mutations at positions 1443 and 1587 (indicated in boldface). Four nucleotide mutations had occurred by passage 5, by passage 30, ten nucleotides were different than parental. Amino acid 162 (indicated in boldface) is a glycosylation site on the surface of the S protein.

Genomic Characterization of PDCoV Strain CHN-SC2015
The complete genome sequence of PDCoV strain CHN-SC2015 is 25,403 nt and was deposited at DDBJ/EMBL/GenBank under accession number MK355396. We compared CHN-SC2015 with 59 other PDCoV strains available in GenBank and found that the level of nucleotide identities ranged from 97.5% to 99.1%; the number of differences in the nucleotide sequences ranged from 606 to 230 ( Figure 4, Table 3). The sequence alignment of the CHN-SC2015 strain revealed a 3-nt deletion (TAA) in the S gene compared to the US, Japan, South Korea, and Southeastern Asia PDCoV strains. This deletion was also present in a number of other Chinese PDCoV strains, indicating that they may have evolved from a common ancestor ( Figure 5B). CHN-SC2015 also contained a 6-nt deletion (TATGAA) and a 9-nt insertion (GCCGGTCGG) in the ORF1ab gene compared with the PDCoV strains from Lao  Figure 5A). Additionally, a 11-nt deletion was found in the 3 UTR, not seen in 58 of the other PDCoV strain; PDCoV strain CHN-HG2017 from China also contained this 11-nt deletion ( Figure 5B). and a 9-nt insertion (GCCGGTCGG) in the ORF1ab gene compared with the PDCoV strains from Lao PDR, Thailand, and Vietnam, and a 3-nt deletion (GTT) compared with HKD/JPN/2016 ( Figure 5A). Additionally, a 11-nt deletion was found in the 3′UTR, not seen in 58 of the other PDCoV strain; PDCoV strain CHN-HG2017 from China also contained this 11-nt deletion ( Figure 5B).

Phylogenetic and Recombination Analysis of PDCoV Strain CHN-SC2015
The phylogenetic analysis was based on sequences of the complete genome and the S gene of CHN-SC2015 and the 59 reference strains. An analysis of the complete genome ( Figure 6A) indicated that the CHN-SC2015 isolates were more closely related to the other PDCoV strains in China than to the strains from Southeast Asia, USA, Japan, and South Korea, indicating the diversity of genetic relationships and regional and epidemic characteristics among these strains. CHN-SC2015 did not cluster with Southeast Asia strains. When compared with the Chinese strains, CHN-SC2015 was grouped into a separate novel subcluster. An analysis of S genes revealed that PDCoV CHN-SC2015

Phylogenetic and Recombination Analysis of PDCoV Strain CHN-SC2015
The phylogenetic analysis was based on sequences of the complete genome and the S gene of CHN-SC2015 and the 59 reference strains. An analysis of the complete genome ( Figure 6A) indicated that the CHN-SC2015 isolates were more closely related to the other PDCoV strains in China than to the strains from Southeast Asia, USA, Japan, and South Korea, indicating the diversity of genetic relationships and regional and epidemic characteristics among these strains. CHN-SC2015 did not cluster with Southeast Asia strains. When compared with the Chinese strains, CHN-SC2015 was grouped into a separate novel subcluster. An analysis of S genes revealed that PDCoV CHN-SC2015 clustered with CHN-HG-2017; there is a 3-bp deletion in CHN-HG-2017 and also present in CHN-SC2015 (Figures 5 and 6B). To evaluate the recombination events of PDCoV CHN-SC2015 during the process of evolution, its complete genome sequence was analyzed using RDP4 and Simplot 3.5.1. As shown Figure 7, CHN-SC2015 potentially underwent recombination with major parent SHJS/SL/2016 (GenBank accession number: MF041982) and minor parent TT_1115 (GenBank accession number: KU984334) (p < 0.01). The predicted recombination breakpoints in CHN-SC2015 are located between nucleotides 6020 and 7069, the region encoding nsp3 and nsp4. To evaluate the recombination events of PDCoV CHN-SC2015 during the process of evolution, its complete genome sequence was analyzed using RDP4 and Simplot 3.

Pathogenicity of PDCoV Strain CHN-SC2015 in Piglets
The pathogenicity of cell-cultured PDCoV strain CHN-SC2015 from passage 12 was demonstrated in the piglets. The daily clinical observations are shown in Table 4; one challenged piglet displayed mild diarrhea at 1 dpi and the other piglets developed mild or watery diarrhea at 3-5 dpi (Table 4, Figure 8 and Figure 9A). Lethargy and anorexia were developed in all the piglets by 2 dpi.

Pathogenicity of PDCoV Strain CHN-SC2015 in Piglets
The pathogenicity of cell-cultured PDCoV strain CHN-SC2015 from passage 12 was demonstrated in the piglets. The daily clinical observations are shown in Table 4; one challenged piglet displayed mild diarrhea at 1 dpi and the other piglets developed mild or watery diarrhea at 3-5 dpi (Table 4, Figures 8 and 9A). Lethargy and anorexia were developed in all the piglets by 2 dpi.   The gastrointestinal and non-gastrointestinal tissues of infected piglets were collected and analyzed. By gross observation, the infected piglets exhibited thin and transparent intestinal walls and had accumulated large amounts of yellow fluid in the intestinal lumen at 5 dpi ( Figure 9B). The stomachs contained coagulated milk ( Figure 9C,F) and mesenteric congestion was also observed ( Figure 9B). No obvious lesions were observed in the control piglets ( Figure 9D,E). Upon histopathological analysis of the gastric glandular and intestinal mucosa, gastric epithelial cells and villous enterocytes were necrotic and sloughing into the lumen. Numerous inflammatory cells infiltrates were absent in the intestinal cavity and lamina propria, which occasionally contained  The gastrointestinal and non-gastrointestinal tissues of infected piglets were collected and analyzed. By gross observation, the infected piglets exhibited thin and transparent intestinal walls and had accumulated large amounts of yellow fluid in the intestinal lumen at 5 dpi ( Figure 9B). The stomachs contained coagulated milk ( Figure 9C,F) and mesenteric congestion was also observed ( Figure 9B). No obvious lesions were observed in the control piglets ( Figure 9D,E). Upon histopathological analysis of the gastric glandular and intestinal mucosa, gastric epithelial cells and villous enterocytes were necrotic and sloughing into the lumen. Numerous inflammatory cells infiltrates were absent in the intestinal cavity and lamina propria, which occasionally contained The gastrointestinal and non-gastrointestinal tissues of infected piglets were collected and analyzed. By gross observation, the infected piglets exhibited thin and transparent intestinal walls and had accumulated large amounts of yellow fluid in the intestinal lumen at 5 dpi ( Figure 9B). The stomachs contained coagulated milk ( Figure 9C,F) and mesenteric congestion was also observed ( Figure 9B). No obvious lesions were observed in the control piglets ( Figure 9D,E). Upon histopathological analysis of the gastric glandular and intestinal mucosa, gastric epithelial cells and villous enterocytes were necrotic and sloughing into the lumen. Numerous inflammatory cells infiltrates were absent in the intestinal cavity and lamina propria, which occasionally contained congested blood vessels ( Figure 10). The PDCoV N protein was detected in the distal and proximal sections of the jejuna of infected piglets as can be seen in Figure 11. These results demonstrate that PDCoV strain CHN-SC2015 is highly pathogenic in piglets.
congested blood vessels ( Figure 10). The PDCoV N protein was detected in the distal and proximal sections of the jejuna of infected piglets as can be seen in Figure 11. These results demonstrate that PDCoV strain CHN-SC2015 is highly pathogenic in piglets.  Jejunum samples from five-old-day piglets infected with PDCoV CHN-SC2015 were fixed and then stained with the rabbit anti-PDCoV N protein monoclonal antibody and FITC-conjugated goat antirabbit IgG (green); the nucleus was counterstained with DAPI (blue). Magnification = 2×.

Viral Shedding and Distribution
To determine viral shedding during the days post-infection, qRT-PCR was used to quantify viral RNA from rectal swabs. Viral distribution in the intestines and internal organs was also determined using qRT-PCR. As shown Figure 12A, virus was shed in feces from 1 dpi to 5 dpi at about the same level, at 6 and 7 dpi the number of viral RNA copies rose about 100-fold to10 9.43 copies/mL of rectal swab sample. Viral RNA was detected in duodenums (average 10 3.94 copies/g), jejunums (average 10 3.77 copies/g), ileums (average 10 4.25 copies/g), colons (average 10 4.83 copies/g), and rectum (average 10 4.01 copies/g). It was also detected in hearts (average 10 4.38 copies/g), livers (average 10 3.89 copies/g), spleens (average 10 4.14 copies/g), lungs (average 10 4.04 copies/g), kidneys (average 10 3.81 copies/g), and stomachs (average 10 4.27 copies/g) ( Figure 12B), indicating that CHN-SC2015 has broad tissue tropism in five-day-old piglets. No PDCoV RNA was detected in the control swine.

Viral Shedding and Distribution
To determine viral shedding during the days post-infection, qRT-PCR was used to quantify viral RNA from rectal swabs. Viral distribution in the intestines and internal organs was also determined using qRT-PCR. As shown Figure 12A, virus was shed in feces from 1 dpi to 5 dpi at about the same level, at 6 and 7 dpi the number of viral RNA copies rose about 100-fold to10 9.43 copies/mL of rectal swab sample. Viral RNA was detected in duodenums (average 10 3.94 copies/g), jejunums (average 10 3.77 copies/g), ileums (average 10 4.25 copies/g), colons (average 10 4.83 copies/g), and rectum (average 10 4.01 copies/g). It was also detected in hearts (average 10 4.38 copies/g), livers (average 10 3.89 copies/g), spleens (average 10 4.14 copies/g), lungs (average 10 4.04 copies/g), kidneys (average 10 3.81 copies/g), and stomachs (average 10 4.27 copies/g) ( Figure 12B), indicating that CHN-SC2015 has broad tissue tropism in five-day-old piglets. No PDCoV RNA was detected in the control swine.

Viral Shedding and Distribution
To determine viral shedding during the days post-infection, qRT-PCR was used to quantify viral RNA from rectal swabs. Viral distribution in the intestines and internal organs was also determined using qRT-PCR. As shown Figure 12A, virus was shed in feces from 1 dpi to 5 dpi at about the same level, at 6 and 7 dpi the number of viral RNA copies rose about 100-fold to10 9.43 copies/mL of rectal swab sample. Viral RNA was detected in duodenums (average 10 3.94 copies/g), jejunums (average 10 3.77 copies/g), ileums (average 10 4.25 copies/g), colons (average 10 4.83 copies/g), and rectum (average 10 4.01 copies/g). It was also detected in hearts (average 10 4.38 copies/g), livers (average 10 3.89 copies/g), spleens (average 10 4.14 copies/g), lungs (average 10 4.04 copies/g), kidneys (average 10 3.81 copies/g), and stomachs (average 10 4.27 copies/g) ( Figure 12B), indicating that CHN-SC2015 has broad tissue tropism in five-day-old piglets. No PDCoV RNA was detected in the control swine.

Discussion
Coronaviruses are enveloped RNA virus in the Coronaviridae family [34] and to date, they have been classified into four genera (α, β, γ and δ). Several alphacoronaviruses, such as PEDV and TGEV cause diarrhea and vomiting in piglets, leading to enormous economic losses in the swine industry [35,36]. Swine enteric coronavirus (SeCoV), a recombinant between PEDV and TGEV reported in 2016, also causes diarrhea in swine [37]. Additionally, some novel alphacoronaviruses related to the bat coronavirus HKU2 have recently been reported in China; these viruses, porcine enteric alphacoronavirus (PEAV) [38], swine enteric alphacoronavirus (SeACoV), [39] and swine acute diarrhea syndrome coronavirus (SADS-CoV) [40] are also major pathogens of diarrheal disease in swine. The deltacoronavirus, PDCoV, also causes diarrhea, vomiting, and dehydration in nursing piglets, and it is clinically indistinguishable from other enteric diseases caused by these porcine enteropathogenic coronavirus. Moreover, coexistence of pathogens, such as PDCoV and PEDV, PDCoV and PoRV, increase the mortality rate of infected piglets, which incurs massive economic losses to porcine industries [41]. Taken together, the pathogens associated with diarrhea in piglets have attracted much needed attention to the necessity for expanded preventative, diagnostic, and treatment measures.
In this study, PDCoV was purified from fecal samples from 10 diarrheal, PDCoV-positive piglets then passed in LLC-PK cells. From the 10 samples, only the PDCoV strain CHN-SC2015 was successfully isolated. Viral identification was done by RT-PCR/qRT-PCR, IFA, and EM. Sample quality is a crucial factor for the successful isolation of PDCoV. Clinical samples may contain unknown viruses or other pathogens or very low amounts of infectious virus, all making successful isolation difficult [18]. The addition of exogenous trypsin in vitro enhances virus entry into the host cells by proteolytic cleavage of the S protein, which is necessary for coronavirus infection [42]. Therefore, the concentration of trypsin in the LLC-PK cell culture media may contribute to the isolation of PDCoV [43]. In our study, the maintenance medium was supplemented with 7.5 µg/mL trypsin, which is higher than the concentrations used in the studies that demonstrated the effectiveness of trypsin in in vitro PDCoV infections [44]. In addition, the time at which the sample is actually collected may also contribute to the success of PDCoV isolation.
Mutations in the coronavirus S protein are considered the key factors responsible for PDCoV virus pathogenesis and host tropism change [47,48]. A sequence analysis based on the complete genome of the PDCoV strain CHN-SC2015 showed a 3-nt deletion in the S gene; this deletion is found in other Chinese isolates but not in isolates from the USA, Japan, South Korea, or Southeast Asia, suggesting a mechanism for differences in virulence among these PDCoV strains. However, whether this particular deletion contributes to the efficiency of viral replication and/or the virulence of these strains is in question. After passage 30, we found six nucleotide changes in the S gene compared with the original sample CHN-SC2015, which induced corresponding amino acid changes, excluding the synonymous mutations at the position 1443 and 1587. The amino acid mutation at position 162 is a glycosylation site located on the surface of amino-terminal domain within the S protein. The lack of glycosylation at this site may hinder the virus' ability to evade the host's immune response because of the lack of glycan shielding of S protein epitopes [34,49]. The 11-bp deletion in 3 UTR in this strain was not found in 58 of the reference strains; the exception was PDCoV strain CHN-HG-2017, indicating that CHN-SC2015 may be a novel strain to the province of Sichuan. According to the phylogenetic results, based on the complete genome and S gene, PDCoV CHN-SC2015 falls into a different clade compared with many other PDCoV strains currently circulating in China. This difference indicates that variations, for the most part, existed in other genes of the PDCoV during evolution and transmission [50]. Moreover, CHN-SC2015 belongs to a different clade than PDCoV isolates from Lao PDR, Thailand, and Vietnam.
Recombination events have been reported in animal and human coronaviruses, such as PEDV and severe acute respiratory syndrome (SARS), these viruses have a history of recombination with other lineages of coronavirus [51,52]. Su S et al. posit that the large genome and multiple host species were major factors for the high rate of recombination events, which can alter tissue tropism, transmission routes, and host specificity [53]. Like most coronaviruses, deltacoronavirus also possess the potential to undergo some recombination events that can facilitate interspecies transmission and jump to the new hosts. Lau et al. reported that porcine coronavirus HKU15 from recombination between sparrow coronavirus HKU17 and bulbul coronavirus HKU11 in the S gene region [17]. In this study, a recombination analysis predicted that CHN-SC2015 had undergone recombination with strain SHJS/SL/2016 from the China and TT_1115 from Vietnam. The predicted breakpoint is located in the nsp3 gene, which is similar to the strain CHN-HG-2017, which may have been derived from the major parent strain SXD1 from China and minor parent strain HaNoi6 from Vietnam [54]. However, the breakpoint of the recombinant virus between the two different strains may be distinct.
In recent years, the pathogenicity of PDCoV to piglets has become a focus of investigation [16,55,56]. The clinical profile of PDCoV infection, diarrhea, vomiting, dehydration, and death, are indistinguishable from other enteropathogenic porcine coronavirus infections such as PEDV and TGEV [19]. In our study, the pathogenicity of PDCoV strain CHN-SC2015 was demonstrated in five-day-old pigs orally inoculated with 10 6.35 TCID 50 /mL cell-culture-adapted virus. Gastrointestinal tissues from a challenged piglet at 5 dpi showed evident lesions indicating that the intestinal tract is the main site for virus replication. Recent reports have found that PDCoV is transmissible by indirect contact with an infected piglet, indicating that the respiratory tract may be the minor target for PDCoV infection [57]; Moreover, in our study, we also found that the control piglets had developed mild diarrhea along with PDCoV CHN-SC2015 infection, leading to a low score of diarrhea from 6 DPI, which further supports the notion that PDCoV can transmit via the respiratory tract. This route of infection is also true for PEDV, which can disseminate from the nasal cavity to the intestinal mucosa, providing further evidence for the airborne transmission of gastrointestinal coronaviruses [58]. Analysis of daily fecal samples and post-mortem tissues and organs by qRT-PCR demonstrated that PDCoV RNA continually and gradually increased from 1 to 7 dpi in feces, PDCoV RNA was distributed in multiple tissues and organs, indicating extensive tissue tropism; these results are consistent with other studies [18,59].
In summary, CHN-SC2015 is the first PDCoV strain isolated from southwest China, the country's largest pig-producing area. The phylogenetic and recombination analyses indicate that CHN-SC2015 is closely related to the other PDCoV strains in China than to the strains from Southeast Asia, USA, Japan, and South Korea. CHN-SC2015 contained deletions and/or insertions in ORF1ab, S, and the 3 UTR compared with the strains available in GenBank. Infection of five-day-old piglets with PDCoV CHN-SC2015 demonstrated its pathogenicity. Our results enrich the body of information on the molecular epidemiology and pathogenic of PDCoV.