Infectious Dose of a 2018 Senecavirus A Isolate in Neonatal Pigs

TL;DR

This study found that a 2018 Senecavirus A isolate had the same infectious dose as a 2011 isolate, suggesting increased infectivity is not the reason for the rise in SVA cases since 2015.

Contribution

The study determined the infectious dose of a post-2015 SVA isolate in neonatal pigs and compared it to a 2011 isolate.

Findings

The 2018 SVA isolate had a minimum infectious dose of 102.5 TCID50/mL, similar to the 2011 isolate.

Higher doses of the 2018 isolate caused clinical signs like diarrhea and lethargy in neonatal pigs.

The results suggest increased infectivity is not the cause of the post-2015 rise in SVA cases.

Abstract

Senecavirus A (SVA) was sporadically detected in swine samples in the United States starting in the late 1980s; however, in 2015 there was a sharp increase in the number of cases in the United States and around the globe. The purpose of this work was to determine if the infectious dose of a post-2015 SVA isolate was lower than previous work with a 2011 SVA isolate, which could help explain the sudden increase in SVA cases. The 2018 SVA isolate in this study had the same minimum infectious dose as prior work with a 2011 SVA isolate. This work provides support against the hypothesis that increased infectivity, and thus transmissibility, was the cause of the increase in SVA detections observed post-2015. Interestingly, animals receiving the highest doses of virus displayed clinical signs of diarrhea and lethargy, which is consistent with clinical signs reported in neonates during SVA outbreaks on sow farms. Previous experimental challenges have not been able to reproduce neonatal disease with SVA; therefore, this work can be used to help understand the pathogenesis of SVA in neonates.

Senecavirus A (SVA) is a picornavirus that causes vesicular disease in swine and has been associated with increased neonatal mortality. Although SVA had only been detected sporadically in the United States since the 1980s, there was a sharp increase in cases in the United States and around the world starting in 2015. The cause of this shift in SVA epidemiology remains unknown; however, changes in the virus that have resulted in enhanced infectivity may have contributed. The aim of this research was to establish the infectious dose of a post-2015 SVA isolate in neonatal pigs and compare its infectivity to previous work with a 2011 SVA isolate. A 2018 SVA isolate (SVA/KS/2018) was serially 10-fold diluted to generate six inoculums. Animals were individually housed with four pigs inoculated with 2 mL orally per dilution. Detection of SVA RNA in serum and swabs, as well as the presence of neutralizing antibodies, were used to classify the infection status of animals. The minimum infectious dose for SVA/KS/2018 in neonates was 102.5 TCID50/mL (2 × 102.5 or 632 TCID50/pig). This value is similar to the infectious dose determined for SVA/CAN/2011, thus providing evidence that the increase in SVA detections was not due to increases in infectivity of contemporary isolates. Neonatal mortality has not been experimentally reproduced; however, pigs inoculated with higher doses of SVA/KS/2018 developed diarrhea and mortality, suggesting increased virulence, which should be investigated further.

Full text

1. Introduction

Senecavirus A (SVA), recently changed to Senecavirus valles, is a single-stranded positive-sense RNA virus of the genus Senecavirus in the family Picornaviridae [1]. The virus had been found sporadically in swine samples since the late 1980s; however, in 2007 and 2011, there were two separate reports linking SVA to cases of vesicular disease in swine [2,3,4]. In 2015, vesicular disease cases in swine testing positive for SVA were reported in the United States and Brazil [5,6]. Experimental challenge studies in swine with SVA isolated from field cases of vesicular disease fulfilled Koch’s postulates and confirmed SVA as a causative agent for vesicular disease [7,8,9]. After 2015, field cases of SVA-induced vesicular disease have been reported in various countries across Asia and the Americas [10,11,12].

In addition to causing vesicular disease, increased neonatal mortality has been reported on sow farms during SVA outbreaks and the syndrome was termed epidemic transient neonatal losses (ETNLs) [13,14,15,16]. This increase in mortality varied considerably between farms with pre-weaning mortality reports ranging from 5–70% [13,14,15,17,18]. Farms experiencing ETNL saw clinical symptoms including lethargy, diarrhea, wasting, neurologic signs, and mortality [19]. Histologic lesions and immunohistochemistry and in situ hybridization assays have demonstrated a multi-systemic infection in neonates [17,20]. However, some reports did not identify any gross or histologic lesions in SVA-positive neonatal mortality cases [13,14]. In addition, neonatal mortality has not been experimentally reproduced after SVA challenge [21,22].

Starting in 2015, there was a sharp increase in reports of SVA vesicular disease in the United States and around the globe. One hypothesis surrounding this increase in cases included greater pathogenicity and infectivity of the virus. Previous work determined the minimum infectious dose (MID) for a 2011 SVA isolate (SVA/CAN/2011) [21]. The primary aim of this work was to establish the minimum infectious dose of a 2018 SVA isolate (SVA/KS/2018) and compare the results with those of SVA/CAN/2011 to determine whether greater infectivity or pathogenicity could have played a role in increased disease incidence. SVA/KS/2018 was obtained from a sow farm experiencing increased neonatal mortality; thus, a secondary objective was to determine if neonatal mortality could be experimentally reproduced.

2. Materials and Methods

Minimum essential medium (MEM, MilliporeSigma, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (AtlantaBio, Flowery Way, GA, USA), 1% L-glutamine (Life Technologies, Carlsbad, CA, USA), and gentamicin was used to grow swine testicular (ST) cells provided by the National Veterinary Services Laboratory (Ames, Iowa) at 37 °C and 5% CO_2_. Cells were used to grow and titrate virus and in virus neutralization assays [21]. The SVA isolate utilized for animal inoculum was provided by Dr. Fabio Vannucci at the University of Minnesota College of Veterinary Medicine. It was isolated from a commercial sow farm in Kansas experiencing an SVA outbreak with an increase in neonatal mortality in 2018.

The viral isolate (SVA/KS/2018) was diluted to a titer of 1 × 10^6.5^ median tissue culture infectious dose (TCID_50_)/mL, which was termed the stock virus. The stock virus was serially 10-fold diluted in MEM to create six challenge inoculums ranging from 10^5.5^ to 10^0.5^ TCID_50_/mL. Virus titrations were performed on confluent 96-well plates of ST cells and TCID_50_ was calculated using the Reed–Muench method [23].

All animal procedures were approved by the National Animal Disease Center (NADC) Institutional Animal Care and Use Committee under the Animal Care and Use Protocol ARS-2018-750. When an animal met euthanasia criteria or when the study concluded, animals were humanely euthanized with an intravenous administration of a barbiturate (Fatal Plus, Vortech Pharmaceuticals, Dearborn, MI, USA) following the label dose. Euthanasia was performed if one or more of the following criteria were observed for two consecutive animal care checks: dehydrated appearance, lethargy, failure to avoid contact with people, failure to eat, and failing to compete with pen mates. Animals were observed by animal care technicians twice daily (morning and afternoon). If animals began to demonstrate clinical signs, a third observation was performed by scientific staff in the evening.

The animal study design and housing conditions were based on previously published work [21]. In brief, at 24–72 h after birth, thirty piglets were weaned and randomly assigned to study groups, with blocking by litter. Inoculums with estimated titers of 10^5.5^, 10^4.5^, 10^3.5^, 10^2.5^, 10^1.5^, and 10^0.5^ TCID_50_/mL were given to four pigs each. Six pigs were used as sentinels (n = 1/room) to control for the possibility of room contamination. At weaning, piglets were bled and swabbed. Twenty-four hours after placement in incubators, piglets were inoculated orally with 2 mL on 0 days post inoculation (dpi). On 6 dpi, piglets (inoculated and sentinel) were moved from individual isolators and placed in pens by their respective groups. Piglets were bled, oral swabbed, and rectal swabbed on 6, 10, and 14 dpi. Six piglets were euthanized or died prior to 10 dpi sampling, so data from 10 and 14 dpi are not available from those animals. All remaining animals were humanely euthanized on 14 dpi at the end of study.

Animal caretakers checked piglets daily for clinical signs including lethargy, inappetence, and diarrhea. Blood was collected via venipuncture in the jugular groove and serum was harvested from serum separator tubes (BD Vacutainer^®^, Franklin Lakes, NJ, USA) post centrifugation. Oral and rectal swabs were collected with a sterile polyester-tipped applicator (Puritan Medical Products, Guilford, ME, USA) and immersed in 3 mL of MEM. Serum and swabs were stored at −80 °C until future testing.

Following the development of morbidity and mortality, fresh and fixed tissues of two euthanized piglets (pig IDs 157 and 175) and 15 rectal swabs were submitted to the Iowa State University Veterinary Diagnostic Laboratory for a diagnostic investigation that included gross examination, microscopic evaluation, bacterial culture of the small and large intestine, and rotavirus PCR (rectal swabs and feces from the large intestine of the submitted tissue).

Inoculum, serum and swabs were tested for SVA RNA by quantitative reverse transcription PCR (RT-qPCR) as previously described [24]. Briefly, RNA was extracted using the MagMAX™ Pathogen RNA/DNA kit (Applied Biosystems, Foster City, CA, USA). Next, 5 µL of extracted product was added to 20 µL of the AgPath-ID™ One step RT-PCR master mix (Applied Biosystems) for all samples except for rectal swabs that used 20 µL Path-ID™ Multiplex One-Step RT-PCR reaction master mix (Applied Biosystems). Inoculum samples were tested in triplicate, while all others were tested in duplicate with the average reported. The forward primer sequence was 5′-TGCCTTGGATACTGCCTGATAG-3′, the reverse primer sequence was 5′-GGTGCCAGAGGCTGTATCG-3′ and the probe sequence was 5′-FAM-CGACGGCCTAGTCG GTCGGT T-Iowa Black-3′. Real-time RT-PCR was performed on an ABI 7500 Fast instrument (Life Technologies) run in standard mode with the following conditions: 1 cycle at 48 °C for 10 min, followed by 1 cycle at 95 °C for 10 min, and 40 cycles of 95 °C for 15 sec and 60 °C for 45 sec. A plasmid containing the target region was used as a standard for quantification and samples were classified as negative using a cutoff of 35.

The virus neutralization (VN) assay was performed as previously described [24]. In brief, heat-inactivated serum samples were diluted from 1:4 to 1:4096 with four replicates. An equal volume of SVA/KS/2018 at approximately 200 TCID_50_ was added to each well for one hour and transferred to confluent ST cells. After four days, titers were recorded as the reciprocal of the highest dilution of serum at which cytopathic effect was present in 50% of wells. Titers ≤16 were considered negative.

3. Results

The stock virus (SVA/KS/2018) was serially 10-fold diluted to generate inoculums with theoretical titers between 10^5.5^ to 10^0.5^. All inoculums were back-titrated (Table 1). The 10^5.5^ and 10^3.5^ inoculums had titers higher than the theoretical titer. Inoculum Ct values ranged from 14.7 to 32.8, which corresponded to 8.64 × 10^8^ to 2.98 × 10^3^ genomic copies/mL. Unlike viral titers, PCR results include both viable and non-viable viral particles.

Piglets were observed daily, and any abnormal clinical signs were noted (Supplemental Table S1). Starting on 2 dpi, loose stool was noted in four piglets among the three lowest dilutions (highest quantity of virus), including the sentinel animal from the 10^4.5^ group. By 4 dpi, the majority of piglets in the 10^2.5^, 10^4.5^, 10^5.5^ inoculum groups had diarrhea, although all piglets continued to eat well. When piglets were removed from the isolators and placed in group pens, probiotics were added to the milk replacer to help resolve the diarrhea. Reduced consumption of milk replacer was noted for some piglets from the 10^4.5^ and 10^5.5^ groups.

On 8 dpi, two pigs in the 10^5.5^ inoculum met euthanasia criteria due to poor condition and were euthanized. On 9 dpi, two pigs were found dead in the 10^4.5^ group. Gross examination revealed an empty stomach in one animal and the other had thin-walled small intestines. Subsequently, two additional pigs in the 10^4.5^ group were euthanized due to poor condition and meeting humane endpoints and submitted to a veterinary diagnostic lab for pathology and bacterial culture to investigate other causes of neonatal diarrhea/morbidity. Gross pathology revealed mild fibrinous epicarditis in one animal; the cause of which was not identified. In addition, rectal swabs were submitted from a subset of animals from all inoculum groups to test for rotavirus A, B, and C and all swabs were PCR negative. No definitive cause of mortality or diarrhea was identified. The remaining animals recovered and had normal feces by 10 dpi.

All pigs were PCR negative for SVA RNA in samples collected at 0 dpi and neutralization titers were all ≤4. Serum and swabs collected at 6 dpi were tested for SVA RNA by PCR to classify infection status upon removal from individual isolators (Table 2). Neutralizing antibody response was evaluated on serum collected at 10 dpi. However, in the case of animals that were euthanized or died prior to 10 dpi, necropsy serum was utilized for those animals. This allowed animals time to develop an antibody response following challenge while limiting the development of antibodies in response to virus exposure from a pen mate. No animals in the 10^0.5^ and 10^1.5^ groups tested positive for SVA at 6 dpi or demonstrated seroconversion at 10 dpi. For the 10^2.5^ and 10^3.5^ groups, two animals and one animal tested PCR positive for SVA at 6 dpi and seroconverted by 10 dpi, respectively. For the 10^4.5^ and 10^5.5^ groups, all animals tested positive for SVA at 6 dpi; however, only 1/4 and 2/4 animals seroconverted, respectively. Of note, the animals in the 10^4.5^ and 10^5.5^ groups that did not seroconvert (titer ≤ 16) died or were euthanized prior to 10 dpi (8 and 9 dpi, as indicated). All pigs that were used as room sentinels were negative for SVA RNA in serum and swabs at 6 dpi and had VN titers ≤ 4 at 10 dpi, indicating no SVA room exposure.

Samples collected on 10 and 14 dpi were tested for SVA RNA to assess shedding and transmission dynamics once animals were placed in group pens by challenge dose. Primary and sentinel pigs in the 10^0.5^ and 10^1.5^ remained PCR negative. By 10 dpi, after five days of co-mingling, all pigs in the 10^2.5^, 10^3.5^, 10^4.5^, 10^5.5^ groups were PCR positive in either serum or swabs including those animals that served as room sentinels (Supplemental Table S2). Viremia was less consistently detected in animals at 14 dpi; however, most oral and rectal swabs were still PCR positive. Table 3 summarizes results from the inoculum and pig studies. Based on the results of this study, the minimum infectious dose (MID) and infectious dose 50 (ID50) of SVA/KS/2018 in neonatal pigs was 10^2.5^ TCID_50_/mL (2 × 10^2.5^ or 632 TCID_50_/pig).

4. Discussion

Although SVA had been detected in US swine samples prior to the vesicular disease outbreak in the United States in 2015, it was only found sporadically and was associated with a variety of clinical histories [2]. The sharp rise in SVA cases associated with vesicular disease and neonatal mortality in 2015 led to questions regarding changes in the virus that may be responsible for the shift in SVA epidemiology. One hypothesis was SVA had become more infectious, which could result in increased transmission and spread of the virus. The MID for SVA/KS/2018 in this study was 10^2.5^ TCID_50_/mL (2 × 10^2.5^ or 632 TCID_50_/pig), which was the same as that reported for SVA/CAN/2011 [21]. Results from these two studies suggest that increased infectivity of the virus does not explain the change in epidemiology and strengthen our knowledge surrounding the infectivity of SVA.

Experimental challenge studies comparing historical SVA isolates and more contemporary isolates have noted differences between strains. For example, one research group found that pigs challenged with the 2001 cell culture contaminant strain, SVV 001, did not develop vesicular lesions, while those challenged with a 2015 strain (SD15–26) did develop lesions [25]. In contrast, another study demonstrated that SVV 001 was able to induce vesicular lesions in pigs as well as SVA isolates from 2011 (SVA/CAN/2011), 2012 and 2015 [26]. Finally, a 2016 and 2017 isolate from China were compared with only animals challenged with the 2017 isolate, developing vesicular lesions [27]. Therefore, these studies demonstrate differences in clinical disease and replication kinetics of various isolates, but the differences between isolates do not uniformly demonstrate increased disease with more contemporary isolates or explain the global increase in reports of SVA starting around 2015.

Clinical signs in young neonates on farms experiencing ETNL during SVA outbreaks include diarrhea, lethargy, wasting and neurologic signs [14,17,18,19]. In this study, clinical signs of lethargy and diarrhea were observed in piglets. To eliminate pig-to-pig transmission of virus, piglets were placed in individual isolators for challenge. In addition, to look at the age group most susceptible, piglets were early weaned (1–3 days of age) [21]. During the time in the individual isolators, piglets were fed milk replacer as a sole feed source. Milk replacer has been shown to result in increased diarrhea in piglets compared to those that remain on the sow, making it difficult to conclude the role of SVA in the development of diarrhea [28,29]. In addition, other factors such as early weaning or housing in an isolator could also predispose neonatal pigs to enteric dysfunction [30]. Additionally, diarrhea was noted in one of the sentinel pigs (pig #45, 10^4.5^ group) prior to commingling. A recent study reported that 28% of SVA-positive cases analyzed were also positive for Rotavirus A [31]. To rule out other causes of neonatal diarrhea, further diagnostics were performed on rectal swabs and collected tissues. The diagnostic investigation did not identify the cause of the diarrhea. Rectal swabs collected tested negative for rotavirus A, B, and C and gross and microscopic evaluation did not identify lesions consistent with a bacterial etiology.

Diagnostics performed on cases of ETNL in the field primarily found only SVA; however, there were differences in gross and histologic lesions reported in neonates. Most reports of ETNL commonly report an absence of lesions [13,16,18]. In this study, there was mortality in the two groups with the highest challenge doses. There were no significant and consistent gross or histologic lesions observed in these neonates. Prior to this study, neonatal mortality had not been reproduced in an experimental setting where piglets have been challenged with SVA [21,32]. Mortality was observed in a dose-dependent manner in this study, which suggests causation; however, the role and possible mechanisms by which SVA may have resulted in the observed morbidity and mortality are unclear.

It was noted that pigs that were found dead or euthanized on 8 dpi or 9 dpi had low levels of neutralizing antibody against SVA in their serum. These pigs were PCR-positive at 6 dpi alongside their cohorts that did develop neutralizing antibody titers by 10 dpi. Although these pigs had less time to develop a neutralizing antibody response, previous SVA challenge studies have consistently demonstrated a neutralizing antibody response can develop as early as 5 dpi in the majority of experimentally challenged animals [8,33]. Therefore, it was unexpected that all animals that had to be euthanized or died before the 10 dpi sample had VN titers ≤ 16. The lack of neutralizing antibody response in the severely ill animals could be evidence of decreased immune function in the face of stress [34]. In addition, SVA has been detected in multiple lymphoid tissues; therefore, replication in these tissues could result in inflammation and tissue damage, which could impact the immune response [25,35].

There have been two studies evaluating the infectious dose of SVA. One assessed feed-based exposure and found that pigs given feed spiked with 10^5^ TCID_50_ over three days became infected with SVA and seroconverted [36]. The other study evaluated the MID of SVA/CAN/2011 with the same methods used in this work and resulted in the same MID, 10^2.5^ TCID_50_/mL [21]. This result was unexpected, considering the temporal gap between isolation; notably, SVA/KS/2018 exhibits a 95% nucleotide identity to SVA/CAN/2022. Interestingly, the MID of SVA is considerably higher than the published infectious doses for both porcine epidemic diarrhea virus (PEDV) and porcine reproductive and respiratory syndrome virus (PRRSV) [37,38]. However, the route of inoculation and age of animals were different for these studies and could have an impact on the infectious dose. Although the higher infectious dose of SVA may mean there is less risk of transmission compared to PEDV and PRRSV, biosecurity measures are still critical, especially considering the economic impact of foreign animal disease investigations performed when a vesicle is observed to rule out foot-and-mouth disease.

This work suggests that there have not been significant changes in the infectivity of SVA over time that would explain the SVA outbreaks across multiple countries beginning in 2015. Other explanations for the change in epidemiology should be explored, such as decreased cross-protection or lack of immunity, changes to the viral receptor, or changes in swine genetics over time increasing susceptibility. This work does strengthen our confidence in the infectivity of SVA as there have now been two studies with isolates with genetic, temporal, and geographic differences with similar results. In addition, we observed neonatal mortality, although the direct link between SVA and mortality remains unclear. This work highlights the need for further investigation into the impact of SVA on neonatal pigs to better define the role of SVA in neonatal mortality.