Pathogenesis of Korean Sapelovirus A in piglets and chicks

Sapelovirus A (SV-A), formerly known as porcine sapelovirus as a member of a new genus Sapelovirus, is known to cause enteritis, pneumonia, polioencephalomyelitis and reproductive disorders in pigs. We have recently identified α2,3-linked sialic acid on GD1a ganglioside as a functional SV-A receptor rich in the cells of pigs and chickens. However, the role of GD1a in viral pathogenesis remains elusive. Here, we demonstrated that a Korean SV-A strain could induce diarrhoea and intestinal pathology in piglets but not in chicks. Moreover, this Korean SV-A strain had mild extra-intestinal tropisms appearing as mild, non-suppurative myelitis, encephalitis and pneumonia in piglets, but not in chicks. By real-time reverse transcription (RT) PCR, higher viral RNA levels were detected in faecal samples than in sera or extra-intestinal organs from virus-inoculated piglets. Immunohistochemistry confirmed that high viral antigens were detected in the epithelial cells of intestines from virus-inoculated piglets but not from chicks. This Korean SV-A strain could bind the cultured cell lines originated from various species, but replication occurred only in cells of porcine origin. These data indicated that this Korean SV-A strain could replicate and induce pathology in piglets but not in chicks, suggesting that additional porcine-specific factors are required for virus entry and replication. In addition, this Korean SV-A strain is enteropathogenic, but could spread to the bloodstream from the gut and disseminate to extra-intestinal organs and tissues. These results will contribute to our understanding of SV-A pathogenesis so that efficient anti-sapelovirus drugs and vaccines could be developed in the future.

In comparison to other picornaviruses, the SV-A life cycle remains poorly characterized. We have recently demonstrated that SV-A can recognize a2,3-linked sialic acid (SA) on GD1a as a functional SV-A receptor (Kim et al., 2016). a2,3-linked SA is known to be highly expressed on cells of porcine and avian origin (de Graaf & Fouchier, 2014;Raman et al., 2014), indicating that SV-A has the potential to infect both pigs and chickens. Therefore, the objective of this study was to undertake a comparative analysis of the pathogenesis of a Korean SV-A strain in piglets and chicks.

SV-A strain caused diarrhoea and faecal viral shedding in piglets but not in chicks
Chicken cells are typically rich in a2,3-linked SA, whereas porcine cells are abundant in both a2,3-and a2,6-linked SAs (Raman et al., 2014). Our previous results have demonstrated that SV-A could recognize a2,3-linked SA as a receptor (Kim et al., 2016), suggesting that SV-A might be able to infect and induce pathology in both pigs and chicks. To determine whether SV-A could induce diarrhoea and faecal viral shedding in piglets and chicks, 3-day-old piglets obtained from sows by hysterectomy and 3-day-old specific pathogen-free (SPF) chicks were orally inoculated with 2Â10 9 or 5Â10 8 p.f.u. ml À1 of SV-A (KS04105 strain), respectively. Compared to mock inoculation, piglets inoculated with SV-A strain had continuous diarrhoea from 1 to 5 days post-inoculation (dpi) (data not shown). However, diarrhoea was not observed in mock-or SV-A-inoculated chicks during the entire experimental period.
To assess faecal viral shedding, one-step real-time quantitative reverse transcription PCR (qRT-PCR) assay was performed with faecal samples sequentially collected from mock-or SV-A-inoculated piglets and chicks (Chen et al., 2014). High viral RNA levels were detected in faecal samples collected from piglets at 1 dpi, reaching a peak at 3 dpi followed by decreasing viral loads from 5 dpi (Fig. 1a). However, SV-A RNA was only detected at 1 dpi from faeces of SV-A-inoculated chicks (Fig. 1b), most likely representing virus inoculum passing through the intestines. Collectively, these data indicated that SV-A could induce diarrhoea and faecal shedding in piglets but not in chicks.

SV-A strain caused intestinal and extra-intestinal lesions in piglets but not in chicks
We then assessed the histopathological changes in organs and tissues sequentially sampled from mock-or SV-Ainoculated piglets and chicks. SV-A infection resulted in histopathological changes in the small intestines of virus- inoculated piglets, including villous atrophy and crypt hyperplasia at 1 dpi (Table 1). These mucosal changes were gradually increased in all regions of the small intestine until 5 dpi followed by a decrease at 7 dpi ( Fig. 2, Table 1). Large intestinal lesions, including crypt fusion with epithelial cell hyperplasia, were observed at 2 dpi, increased until 5 dpi followed by a decrease at 7 dpi (Fig. 2). Lungs from infected piglets showed lymphoid cell infiltration in the peribronchiolar submucosa and perivascular space from 5 dpi to the end of the experiment (Fig. 3). As shown in Table 2, adaptive immune reactions to SV-A infection evident as perivascular cuffing of lymphocytes and gliosis were observed in both grey and white matters of spinal cord (myelitis) and brain (encephalitis) (Fig. 3). These typical host defence reactions were observed from 7 dpi until the termination of the experiment. Neuronophagia and chromatolysis were not frequently found in the spinal cord or the brain. However, spinal cords showed stronger inflammatory reactions than the brain (Table 2). No specific lesion was observed in other organs or tissues collected from piglets regardless of SV-A infection. However, SV-A did not induce any histopathological change in any organ or tissue sequentially sampled from SV-A-inoculated chicks during the entire experimental period (Fig. 4). These data indicated that SV-A could induce pathology in piglets but not in chicks.

Viral antigen was detected in the intestine of piglets but not chicks
To assess the distributions of SV-A antigen in the organs and tissues, immunohistochemical assay was performed with organs and tissues sampled sequentially from mock-or SV-A-infected piglets and chicks using monoclonal antibody (mAb) specific to SV-A capsid protein. SV-A antigen was only detected in the epithelial cells of villi from SV-Ainfected piglets at 1 to 5 dpi (Fig. 2, Table 1). Other organs and tissues collected from mock-or SV-A-inoculated piglets were negative for SV-A antigen (Fig. 3, Table 2). Consistent with clinical and histopathological observations, SV-A antigen was not detected in any organ or tissue collected from SV-A-inoculated chicks (Fig. 4).

SV-A caused viraemia and replicated in extraintestinal organs of piglets but not chicks
To assess whether SV-A induced viraemia and replicated in extra-intestinal organs of piglets and chicks, qRT-PCR assay was performed with sera and extra-intestinal organs and tissues collected from mock-or SV-A-inoculated piglets and chicks. SV-A RNA levels were relatively low in the serum, spinal cord, lung and brain in comparison with those in the faecal samples collected from SV-A-inoculated piglets ( Fig. 1a). However, SV-A RNA was not detected in the sera and any extra-intestinal organs and tissues collected from mock-or SV-A-inoculated chicks. These data indicated that SV-A induced viraemia, which was then disseminated to extra-intestinal organs and tissues in piglets but not chicks.

Binding and infection abilities of SV-A to various cells
To determine whether SV-A had a strict tropism for porcine cells in vitro, the binding and infection ability of SV-A was examined with various cell lines, including porcine, human,  ‡Mock inoculation with serum-free EMEM.
chicken embryo, canine, simian, hamster and feline. Alexa Fluor 594 (AF-594)-labelled SV-A was attached to all cell lines at various degrees ( Fig. 5a). Radio-labelled SV-A was able to bind to all cells examined at similar degrees (Fig. 5b). However, SV-A was only replicated in cells of porcine origin (LLC-PK and PK-15) ( Fig. 5c) with similar levels of cytopathic effect (data not shown). The SV-A genome copy numbers robustly increased in cells of porcine origin in a time-dependent manner, but not in cells of other species (Fig. 5d). Our results indicated that only porcine cells were available for SV-A infection.

DISCUSSION
All viruses initiate infection by binding to specific receptor (s) on the surface of susceptible host cells (Neu et al., 2011).
We have previously demonstrated that SV-A could utilize a2,3-linked SA on GD1a glycolipid as a receptor (Kim et al., 2016). Indeed, glycolipid-associated a2,3-terminal SA is abundant on the cell surface of both avian and porcine species (de Graaf & Fouchier, 2014;Raman et al., 2014). However, our results in this study revealed that SV-A infection was limited to cells of porcine origin. It could not infect cells of other origins including chickens and humans. Moreover, SV-A was replicated in piglets but not in chicks, confirming that SV-A could not cause interspecies transmission at least between pigs and chickens. Our results also suggested that other factors in addition to glycolipid-associated a2,3-terminal SA might be required for efficient SV-A replication in cells of non-porcine origin. We have previously observed that chymotrypsin or trypsin treatment of cells has no effect on virus infection (Kim et al., 2016), suggesting that cell surface-associated proteins do not play accessory roles in SV-A infection. The nature of post-binding block to infection in cells from non-porcine origin is as yet unknown, but it is possible the SV-A infection of avian cells induces the innate response leading to the restriction of viral replication.
Although previous studies have indicated that SV-A can cause enteritis, pneumonia, polioencephalomyelitis and reproductive disorders (Alexandersen et al., 2012;Huang et al., 1980;Lamont & Betts, 1960;Lan et al., 2011;Schock et al., 2014;Sibalin, 1963), the pathogenicity and/or host range restriction of SV-A have been poorly characterized. In the current study, the most significant lesions were found in the intestines during the early infection period, where severe villous atrophy was found to be associated with high viral RNA loads in the faecal samples and strong SV-A antigen reactivity in the intestinal epithelial cells. In contrast, the spinal cord and brain mainly showed signs of adaptive immune response (non-suppurative inflammation) and comparatively low viral RNA copy numbers without clear evidence of SV-A antigen-positive cells. These data indicated that the Korean SV-A strain (KS04105) used in this study was enteropathogenic. Contrary to our data, a recent study has reported that neuroinvasive English SV-A strain replication in the spinal cord could lead to severe adaptive immune responses (non-suppurative polioencephalomyelitis) without causing lesions in other tissues (Schock et al., 2014). This English SV-A strain (G5) is phylogenetically closer to the English V13 strain than to other SV-A strains circulating recently in Korea and China (Schock et al., 2014;Son et al., 2014a). The molecular basis involved in the differences in virulence and tropism among SV-A strains  remains unknown. However, there is clear evidence showing that variation in RNA structures can contribute to viral virulence (Son et al., 2014a). An additional explanation for discordance in tropism may be variation in the age of animals used in studies. The English SV-A study used adult pigs (Schock et al., 2014), whereas this study used neonatal piglets. Further studies are required to examine the predilections of age and inoculation routes, as well as antigenic and genomic differences, so that our understanding of variations in SV-A pathogenesis and/or host range restriction can be improved.
Similar to other enteric viruses (Blutt & Conner, 2007;Park et al., 2007), SV-A RNA loads were detected in the sera of piglets orally inoculated with SV-A strain in this study, suggesting that SV-A might be able to penetrate the gut barrier from the luminal side through destruction of enterocytes in the villi. This result also implied that SV-A could reach other organs and tissues via cell-free transmission. However, the mechanism by which SV-A reaches the blood and spreads to other organs and tissues remains to be determined.
In conclusion, this study demonstrated that the Korean SV-A strain could be replicated and induce pathology in piglets but not chicks. Our data also indicated that this Korean SV-A strain could reach the bloodstream from the gut, and be disseminated to extra-intestinal organs and tissues. These results will improve our understanding of the life cycle and pathogenesis of sapelovirus so that affordable, useful and efficient drugs can be developed for anti-sapelovirus therapy.

METHODS
Cells and viruses. LLC-PK, PK-15 and human cervical cancer HeLa cells [American Type Culture Collection (ATCC)] were maintained in Eagle's minimal essential medium (EMEM) supplemented with 10 % FBS, 100 U ml À1 penicillin and 100 µg ml À1 streptomycin. Chinese hamster ovary (CHO) cells were kindly provided by Dr Sun-Young Im, Chonnam National University, South Korea. Crandall Reese feline kidney (CRFK), Madin-Darby canine kidney (MDCK), African green *Histopathological changes in the central nervous system were scored according to the severity of gliosis and perivascular cuffing as follows: 0, normal; 1, one or two glioses or perivascular cuffing; 2, !2, 3 glioses and/or perivascular cuffing; 3, multiple glioses and/or perivascular cuffing with heavy infiltration of lymphocytes or glial cells. †Antigen distribution in the central nervous system was evaluated based on the number of antigen-positive cells as follows: 0, no positive cells; 1, one to two positive cells; 2, three to five positive cells; 3, many positive cells.

H&E
SV-A strain KS04105 used in this study was isolated from faecal samples of diarrhoeic piglets from South Korea (Son et al., 2014a, b). This strain was passaged eight times in LLC-PK cells, including isolation, adaptation and triple plaque purification. Isolated viruses were confirmed as PSVs based on immunofluorescence assay (IFA), RT-PCR and transmission electron microscopy (Kim et al., 2016;Son et al., 2014a, b).
Reagents and antibodies. AF-594 succinimidyl ester, purchased from Molecular Probes (catalogue number, A-20004), was dissolved in DMSO. The mAb against SV-A capsid protein was kindly provided by Dr M. Dauber (Friedrich Loeffler Institute, Germany). FITC goat antimouse IgG antibody and peroxidase-conjugated goat anti-mouse IgG antibody were purchased from Santa Cruz Biotechnology. HRP-conjugated streptavidin was obtained from Jackson Immuno Research Lab. SlowFade Gold Antifade Reagent (Invitrogen) containing DAPI was purchased from Invitrogen.
Animal experiments. To evaluate the pathogenicity and host range restriction of SV-A, 3-day-old piglets (n=27) from sows by hysterectomy and 3-day-old SPF White Leghorn chicks (n=27) were used. Twenty-four piglets and 24 chicks were orally inoculated with 20 ml (1Â10 8 p.f.u. ml À1 ) or 5 ml (1Â10 8 p.f.u. ml À1 ) of the KS04105 strain individually. As negative controls, piglets and chicks were inoculated with the same volume of medium for mock infection. All animals were fed with sterilized commercialized milk or feed. After inoculation, clinical signs including diarrhoea, pneumonia and convulsions were evaluated daily as described previously (Park et al., 2013). Animals were euthanized at specified times (Table 1).
Necropsy was immediately performed after euthanasia. During necropsy, organs and tissues including each intestinal segment, spinal cords, brain, lung and liver were excised from piglets or chicks. They were immediately placed in 10 % buffered formalin for histological examination. Formalin-fixed samples were embedded in paraffin, sectioned, stained with Mayer's haematoxylin and eosin and examined microscopically (Park et al., 2013). All samples collected for qRT-PCR analysis were immediately snap-frozen in liquid nitrogen and kept at À80 C until use.
Immunohistochemistry. The distribution of SV-A antigens in tissues was evaluated by immunohistochemical examination using paraffinembedded sections and an mAb against SV-A capsid protein as described previously (Park et al., 2007). Briefly, paraffin-embedded sections of each organ and tissue were deparaffinized and rehydrated through a graded series of alcohol in 0.1 M PBS and then treated with 0.1 % trypsin/0.1 % calcium chloride in PBS for 1 h at 37 C. Trypsinized sections were first treated with 3 % H 2 O 2 to quench endogenous peroxidase and then incubated with an mAb against SV-A capsid protein at 4 C overnight. All sections were stained with peroxidase-labelled streptavidin-biotin (SAB-PO) using a Histofine SAB-PO kit for the mouse mAb. Antigen localization was visualized by incubating the sections with 3,3¢-diaminobenzidine/H 2 O 2 solution. The sections were then weakly counterstained with haematoxylin. To calculate the number of antigen-positive cells in the organs or tissues, 10 fields per section were analysed using a Â40 objective and a Â10 eyepiece, yielding a final magnification of Â400.
Virus purification by caesium chloride gradient centrifugation.
SV-A strain KS04105 grown in LLC-PK cells was purified using caesium chloride (CsCl) gradient centrifugation as described previously (Kim et al., 2016). Briefly, infected cell cultures were harvested at 72 h post inoculation and freeze-thawed three times. Cell debris was spun down at 2469 g for 10 min at 4 C. A total of 500 ml of virus-containing supernatants was concentrated by centrifugation at 245 853 g for 20 h at 4 C using a SW40 rotor (Beckman). The viruses in the pellets were resuspended in TNE buffer (50 mM Tris/HCl, 100 mM NaCl and 100 mM EDTA, pH 7.5). The suspension was then layered over discontinuous CsCl gradients. After ultracentrifugation, the virus band was collected by puncturing the side of the tube with a needle. The virus solution was then diluted in distilled water and further purified by ultracentrifugation. Purified viruses were dialysed in 0.1 M sodium bicarbonate buffer (pH 8.3) for fluorescence labelling or in TNE buffer for radioactivity assay overnight. Purified viruses were then stored in aliquots at -80 C.
Labelling of viruses with AF-594. Labelling of viruses with AF-594 was performed as described previously (Kim et al., 2016). Briefly, purified virus (10 mg at 1 mg ml À1 ) in 0.1 M sodium bicarbonate buffer (pH 8.3) was labelled with one-tenth-fold molar concentration of AF-594 succinimidyl ester (1 mg at 1 mg ml À1 in DMSO). Each reaction was mixed thoroughly by vortexing for 30 s and incubated at room temperature for 1 h with continuous stirring. Labelled virus was repurified with CsCl as described above, dialysed against virion buffer and stored in 2 µg aliquots at -20 C.
Dye-labelled binding assay. Dye-labelled binding assay was performed with purified AF-594-labelled viruses as described previously (Kim et al., 2016). Briefly, mock-infected or treated cells were inoculated with m.o.i. of 1000 of AF-594-labelled virus and incubated on ice for 5 min followed by incubation at room temperature for 10 min. Cells were washed extensively with cold PBS, fixed with 4 % formaldehyde and washed three times with cold PBS. Dishes were mounted with Slow-Fade Gold Antifade Reagent containing DAPI solution for nucleus staining. Infected cells were observed under an LSM 510 confocal microscope and analysed using LSM software (Carl Zeiss). Immunofluorescence assay. IFA was performed as previously reported (Kim et al., 2016). Briefly, fixed cells on a confocal dish were permeabilized by 0.2 % Triton X-100, incubated at room temperature for 10 min and washed with PBS containing 0.1 % newborn calf serum (PBS-NCS). The mAb against SV-A capsid protein (1 : 40 dilution) was added and incubated at 4 C overnight. Cells were then washed three times with PBS-NCS. FITC-conjugated secondary antibody (diluted to 1 : 100) was then added. Nuclei were stained with propidium iodide. Cells were then examined by confocal microscopy.
SV-A qRT-PCR. SV-A genome RNA levels in the faeces, sera and each organ or tissue were quantified by qRT-PCR as described previously with slight modifications (Chen et al., 2014;Park et al., 2013). Briefly, all tissues and fluid samples collected from experimental animals were individually weighed, homogenized or vortexed at a 1 : 10 dilution in 0.01 M PBS and centrifuged (tissues 13 000 g for 3 min; faecal samples 5000 g for 10 min). The supernatants, along with the remaining bulk samples, were collected and stored at À80 C for analysis. To quantitate SV-A genome copy numbers, cells were infected without or with SV-A strain KS04105 at an m.o.i. of 1 p.f.u. cell À1 and incubated at 37 C for 4, 8, 15 and 72 h post infection as described above. Each infected cell culture was freeze-thawed three times, and cell debris was spun down at 2469 g for 10 min at 4 C. After extracting total RNA from supernatants, each realtime RT-PCR was performed using a Rotor-Gene real-time amplification system (Corbett Research) and SensiFAST SYBR Lo-ROX One-Step mix (Enzynomics) in a final volume of 20 µl containing 10 µl of Sensi-FAST SYBR Lo-ROX One-Step mix (Enzynomics), 0.2 µl of reverse transcriptase, 0.4 µl of RiboSafe RNase Inhibitor, 0.8 µl of PSV1 primer (GGCAGTAGCGTGGCGAGC at positions between 153 and 170 of the 5¢-UTR), 0.8 µl of PSV2 primer (CTACTCTCCTGTAACCAGT at positions between 242 and 260 of the 5¢-UTR), 4 µl of template and 3.8 µl RNase-free dH 2 O. Reverse transcription was carried out at 42 C for 15 min followed by the activation of Hot Start DNA polymerase at 95 C for 2 min and 45 cycles of 95 C for 10 s, 60 C for 14 s and 72 C for 10 s. Quantitation of virus RNA copies was carried out using a standard curve derived from 10-fold serial dilutions of in vitro transcripted cRNA amplified in separate PCR tubes. Rotorgene 6000 ® (Corbett Research) software was used to calculate the amount of SV-A-specific RNA in the samples. The threshold was defined automatically in the initial exponential phase, reflecting the highest amplification rate. With regard to the crossing points resulting from the amplification curves and threshold, a direct relation between cycle number and log concentration of RNA molecules initially present in the RT-PCR was evident. By linear regression analysis of these data, Rotorgene 6000 ® software was used to set up a standard curve to determine the concentration of RNA present in the samples.
Ethics statement. All animals were handled in strict accordance with good animal practices as described in the NIH Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, 1985, revised 1996. Our experiment protocol was approved by the Committee on Ethics of Animal Experiments, CNU with permit number of CNU No. 2012-87. Statistical analysis. All statistical analyses were performed using SPSS version 11.5.1 for Windows (SPSS). One-way ANOVA was used to determine the statistical significance (P<0.05).