Evolutionary analysis of 3 isolates of Seneca Valley virus from a single pig farm

Abstract

Vesicular disease caused by Seneca Valley virus (SVV) has recently emerged throughout China and caused certain industry losses. We used immunofluorescence and western blotting to confirm that 3 new SVV strains (CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3) were from 1 pig farm. Phylogenetic analysis revealed the following: i) all 3 strains belong to USA-GBI29-2015-like clades, ii) CH-GDSG-2018-3 might have diverged from CH-GDSG-2018-1 and CH-GDSG-2018-2, and iii) CH-GDSG-2018-3 is a recombinant of the CHhb17 and HeNKF-1 strains. Virus growth curves showed that CH-GDSG-2018-3 had stronger proliferation ability in vitro. Seneca Valley virus has evolved extensively within China and this study has furthered our understanding of SVV epidemiology.

Introduction

Seneca Valley virus (SVV) is a positive-sense RNA virus belonging to the family Picornaviridae and genus Senecavirus (1). The complete SVV genome is about 7.4 kb in length and includes a 5′-untranslated region (UTR), a polyadenylated 3′-UTR, and a large open reading frame (ORF) in the middle of the genome. The ORF encodes a large polyprotein that is cleaved into 12 proteins, including 4 structural proteins and 8 nonstructural proteins (2).

In 2002, Seneca Valley virus was identified as a contaminant in cell cultures in the United States (3) and the first isolate was named SVV-001 (2). Because of its oncolytic activity, research on SVV was mainly focused on cancer therapy (2), until 2008, when a researcher found that SVV was related to vesicular disease (4). Clinically, the highly similar symptoms of SVV were indistinguishable from foot-and-mouth disease virus (FMDV), vesicular stomatitis virus (VSV), swine vesicular disease virus (SVDV), and vesicular exanthema of swine virus (VESV). Since 2014, many countries have reported outbreaks of SVV, which have caused certain economic losses to the pig industry (5–9).

In 2015, an outbreak of vesicular disease in the Guangdong Province of China was shown to be caused by SVV; ever since, an increasing number of provinces have also found SVV infection in pigs with vesicular disease (5,10–14). Recently, researchers identified recombination of SVV in some areas, which suggests that the control of SVV transmission has become complex (15,16). Here, we report 3 novel SVV strains isolated from a single farm in Guangdong Province in November 2018. The genetic evolution of these strains was analyzed by constructing phylogenetic trees. These results will deepen our understanding of the SVV epidemic in Guangdong Province, China.

Materials and methods

Clinical signs of disease and sample collection

In November 2018, vesicular disease broke out among the sows of intensive pig farms in Guangdong Province. The sick sows had vesicular lesions on the snout and coronary bands, as well as depression and anorexia. Three vesicle fluid samples were collected for pathogen detection and virus isolation. All samples were collected according to the animal ethics regulations of the National Engineering Center for Swine Breeding Industry (NECSBI 2015-16).

RNA extraction and reverse transcription polymerase chain reaction (RT-PCR)

Total RNA from vesicle fluid samples was extracted using TRIzol reagent (TaKaRa); according to the manufacturer’s instructions and used to perform RT-PCR assays of genomic sequences of SVV, FMDV, SVDV, VSV, and VESV (5,17).

Virus isolation and identification

Porcine kidney (PK) 15 cells were maintained in Dulbecco’s Modified Eagle Medium containing 10% fetal bovine serum (Thermo Fisher Scientific). Vesicle fluid samples were filtered and added to cultured PK15 cells, which were then maintained at 37°C with 5% CO₂ and monitored daily for cytopathic effects (CPEs). When CPEs appeared in 80% of cells, the cells were fixed with anhydrous ethanol and assayed for immunofluorescence (IFA). Western blotting was used to identify the major capsid protein (VP1) band of SVV isolates. The sample for western blotting was cell lysate and the control was PK15 cells without SVV infection. Immunofluorescence and western blotting assays were performed with an anti-VP1 (rabbit) protein polyclonal antibody (anti-VP1 was prepared by our laboratory) diluted to 1:1000, with an Alexa Fluor 488 Affinipure conjugated goat anti-rabbit IgG secondary antibody (CKT0101; Proteintech, Rosemont, Illinois, USA) diluted to 1:400 for IFA and a peroxidase-conjugated goat anti-rabbit IgG (H+L) secondary antibody (ZB-2301; ZSGB-BIO, Beijing, China) diluted to 1:5000 for western blotting.

Electron microscopy

The PK15 cells inoculated with CH-GDZQ-2018-1, CH-GDZQ-2018-2, and CH-GDZQ-2018-3 were harvested when 80% CPE emerged. The PK15 cells uninfected with SVV were used as a negative control. After freeze-thawing, the cells were spun by centrifugation at 12 000 × g for 15 min at 4°C. The cell debris were discarded, and supernatants were collected and spun at 30 000 × g for 3 h at 4°C using a nonlinear 20 to 60% sucrose gradient. Purified viruses located in the interface of the sucrose layers were collected. The purified virus solution was adsorbed on the copper mesh for 5 min, while the filter paper absorbed the excess virus solution along the edge of the copper mesh. A 2% phosphotungstic acid with a pH value of 7.4 was used for negative staining for 5 min. Then, the excess negative stain solution was absorbed with filter paper and placed on a clean filter paper stored in a Petri dish away from light. After 24 h, virion particles were examined using a transmission electron microscope (H-7650; Hitachi, Tokyo, Japan).

Sequencing of complete genomes

Complete genomes of SVV isolates were sequenced using 7 pairs of overlapping primers, as described previously (1). Briefly, amplified PCR products were purified and sequenced by Sangon Biotech (Guangzhou, China). Sequence data were assembled and analyzed using DNASTAR Lasergene software (Madison, Wisconsin, USA).

Phylogenetic analyses

Representative strains of complete genome sequences available in GenBank were collected and used for phylogenetic analyses (18,19). Phylogenetic trees of whole genomes were constructed by the neighbor-joining method, with 1000 bootstrap replicates, using MEGA6.0 software (http://www.megasoftware.net/).

Recombinant analysis

To analyze the potential recombination events, the genomic sequence of the CH-GDSG-2018-3 strain was scanned with the SimPlot software package (v3.5.1; SCRoftware, https://sray.med.som.jhmi.edu/SCRoftware/SimPlot/) using CHhb-2017 and HeNKF-1 as the parent viruses and the CH-GDZQ-2018-3 genomic sequence as the query. A 200-base pair (bp) window and 20-bp step size were applied.

Titer detection of virus growth curve

Porcine kidney 15 cells cultured in 24-well plates were infected with isolated virus at a multiplicity-of-infection of 0.01. The cells and supernatants were collected at 3, 6, 9, 12, and 15 h post-infection and freeze-thawed 3 times. After centrifugation at 10 000 rpm at 4°C for 5 min, the culture supernatants were collected and the median tissue culture infectious dose (TCID₅₀) at each time point was determined using a microtitration infection assay.

Results and discussion

In November 2018, there was an outbreak of vesicular disease in the sows of intensive pig farms in Shaoguan, Guangdong Province. The sows displayed ulcerative lesions on the coronary bands and fluid-filled vesicles on the snout. We collected vesicle fluid samples from 3 of the infected pigs for RT-PCR testing, using primers specific for SVV, FMDV, SVDV, VSV, and VESV, all of which can cause vesicular disease, as described previously (1,20). Three samples were positive for SVV but negative for FMDV, SVDV, VSV, and VESV. The PK15 cells that were inoculated with vesicle fluid from the 3 SVV-positive pigs and incubated for 24 h showed typical CPEs characterized by rounding, shrinkage, and degeneration (Figure 1 A). Infected cells were fixed and assayed for IFA with anti-VP1 polyclonal antibodies. As shown in Figure 1 A, the SVV-infected group emitted specific green fluorescence, which indicated that SVV had been successfully isolated. Western blotting confirmed that SVV-infected cells, but not controls, contained a band specific for viral VP1 (Figure 1 B). Next, PK15 cells infected with each of the new strains were examined by electron microscopy. As shown in Figure 2, many spherical virus particles (diameter = 25 to 30 nm) could be clearly observed. Taken together, these results indicated that 3 new strains of SVV had been successfully isolated and identified. The 3 new strains were named CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3.

Figure 1.

Identification of isolated Seneca Valley virus (SVV) strains. A — The cytopathic effect and immunofluorescence assay of porcine kidney 15 cells infected with SVV strains CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3 at 8 h post-infection. B — Western blot identification of SVV VP1 protein in PK15 cells infected with SVV strains CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3 at 8 h post-infection.

Figure 2.

Electron microscopic images of purified Seneca Valley virus (SVV) particles. A — Isolated SVV strain CH-GDSG-2018-1. B — Isolated SVV strain CH-GDSG-2018-2. C — Isolated SVV strain CH-GDSG-2018-3.

Fragments comprising the full genome of SVV were amplified by RT-PCR using 7 pairs of primers and then sequenced. The complete genome sequences of CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3 were submitted to GenBank (accession numbers MN781982, MN781983, and MN781984, respectively).

The genome length of all 3 strains, excluding the poly A tail, was 7284 nucleotides. The 3 strains shared 97.2 to 99.8% nucleotide identity with each other, 98.1 to 98.3% nucleotide identity with strain USA-GBI-29-2015, 95.7 to 96.2% nucleotide identity with strain CH-01-2015, and only 93.2 to 93.6% nucleotide identity with SVV-001, the first SVV strain to be isolated. The polyproteins encoded by the 3 strains shared 98.9 to 99.9% amino acid identity with each other, 99.1 to 99.5% amino acid identity with USA-GBI-29-2015, and 97.2 to 97.7% amino acid similarity with SVV-001. These results indicated that our strains were highly homologous to the United States strain USA-GBI-29-2015. To further explore the genetic evolution of the newly isolated strains, phylogenetic trees were constructed using representative SVV strain sequences from China in GenBank. Phylogenetic analysis showed that CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3 belonged to a clade with the USA-GBI-29-2015 and USA-GBI-29-2015-like strains previously isolated in China. The USA-GBI-29-2015-like strain was first reported in Guangdong Province of China (19). Soon after, the USA-GBI-29-2015-like strain was found in the neighboring Fujian Province (11). At present, most USA-GBI-29-2015-like strains are isolated from Guangdong Province. Our phylogenetic tree (Figure 3) revealed that USA-GBI-29-2015-like strains formed 2 clades. CH-GDSG-2018-1 and CH-GDSG-2018-2 belonged to clade 2 and had higher identity (98.3%) with USA-GBI-29-2015 than with CH-GDSG-2018-3 (98.1%). CH-GDSG-2018-3 belonged to clade 1 and was more closely homologous to other USA-GBI29-2015-like strains isolated in Guangdong Province in 2017 (CH-GDLZ01-2017, CH-GDLZ02-2017, CH-GDQC-2017, CH-GDYD-2017, and CH-GDYS01-2017). Considering that the 3 new strains were isolated from the same pig farm, these data indicated that SVV had mutated widely. Therefore, it is unknown whether the CH-GDSG-2018-1 and CH-GDSG-2018-2 strains were present in Guangdong Province even earlier. Combined with the current research, it is possible that Chinese USA-GBI29-2015-like strains might have begun to evolve from CH-GDSG-2018-1 and CH-GDSG-2018-2. Recombination analysis indicated that CH-GDSG-2018-3 was a recombinant strain of CHhb2017 and HeNKF-1. Four recombination breakpoints were found at nucleotides 1555, 2028, 2417, and 2678 of the HeNKF-1 genome (Figure 4), with the resulting recombinant fragments located in the regions of VP2, VP3, and VP1.

Figure 3.

Phylogenetic analysis of Seneca Valley virus (SVV) strains CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3. Phylogenetic trees were constructed using the neighbor-joining method, with 1000 bootstrap replicates, using MEGA6.0 software. The newly isolated SVV strains CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3 are marked as black circles.

Figure 4.

Recombination analysis of Seneca Valley virus CH-GDSG-2018-3. Reference strains, HeNKF-1 (green) and CHhb17 (blue) were used as putative parental strains. The X-axis indicates the location of the query sequence, and the Y-axis indicates the percentage of identity.

To evaluate the proliferation dynamics of these strains, virus titers were measured at different time points. As shown in Figure 5, the titers of all 3 strains reached a peak at 12 h post-infection. However, CH-GDSG-2018-3 had a higher titer (10^8.78 TCID₅₀/mL) than the other 2 strains (10^7.22 TCID₅₀/mL) at 12 and 15 h post-infection, which indicates that CH-GDSG-2018-3 replicated faster. These data implied that, although the new SVV strains had mutated on the same farm, the in vitro viral proliferation rate of CH-GDSG-2018-3 had become enhanced.

Figure 5.

Proliferation dynamics of Seneca Valley virus strains CH-GDSG-2018-1, CH-GDSG-2018-2, and CH-GDSG-2018-3. Porcine kidney 15 cells cultured in 24-well cell culture plates were infected with isolated virus of MOI 0.01. The cells and supernatants were collected at 3, 6, 9, 12, and 15 h post-infection, and TCID₅₀ at each time point was determined using a microtitration infection assay.

Seneca Valley virus prototype strain SVV-001 was not pathogenic to pigs when it was first isolated from the human cell line PER.C6 (1). In 2008 and 2012, researchers in Canada and the United States found that SVV-001 was related to vesicular disease (4). Since 2014, many countries have begun to isolate SVV and demonstrate its increasing pathogenicity to pigs. Meanwhile, recombinant SVV strains were also identified and shown to have pathogenicity in pigs (15,16). The results of this study combined with previous reports suggest that the region of the SVV genome that encodes the structural proteins might have a relatively high rate of recombination. The varying pathogenicity of SVV might be related to genome mutation. Studies have shown that SVV can evade the host innate immune system (21). For example, SVV proteins 2C and 3C were found to induce cell apoptosis as well as the degradation of the cytoplasmic sensor retinoic acid-inducible gene I to inhibit type I interferon production (22,23). The functions of other SVV proteins, including those that play a vital role in virulence, remain unclear and require further exploration.

In conclusion, we identified 3 new SVV strains that were coexisting on the same farm. The proliferation ability of 1 strain (CH-GDSG-2018-3) became stronger than the others during cultivation, which might be because SVV was evolving quickly under in vitro conditions. It is urgently necessary to research SVV pathogenesis in vivo and thereby improve the approaches to disease prevention and control of this virus.