The development of two field‐ready reverse transcription loop‐mediated isothermal amplification assays for the rapid detection of Seneca Valley virus 1

Abstract Seneca Valley virus 1 (SVV‐1) has been associated with vesicular disease in swine, with clinical signs indistinguishable from those of other notifiable vesicular diseases such as foot‐and‐mouth disease. Rapid and accurate detection of SVV‐1 is central to confirm the disease causing agent, and to initiate the implementation of control processes. The development of rapid, cost‐effective diagnostic assays that can be used at the point of sample collection has been identified as a gap in preparedness for the control of SVV‐1. This study describes the development and bench validation of two reverse transcription loop‐mediated amplification (RT‐LAMP) assays targeting the 5′‐untranslated region (5′‐UTR) and the VP3‐1 region for the detection of SVV‐1 that may be performed at the point of sample collection. Both assays were able to demonstrate amplification of all neat samples diluted 1/100 in negative pig epithelium tissue suspension within 8 min, when RNA was extracted prior to the RT‐LAMP assay, and no amplification was observed for the other viruses tested. Simple sample preparation methods using lyophilized reagents were investigated, to negate the requirement for RNA extraction. Only a small delay in the time to amplification was observed for these lyophilized reagents, with a time from sample receipt to amplification achieved within 12 min. Although diagnostic validation is recommended, these RT‐LAMP assays are highly sensitive and specific, with the potential to be a useful tool in the rapid diagnosis of SVV‐1 in the field.

Rapid and accurate detection of SVV-1 is necessary to confirm the disease causing agent, and to initiate the implementation of control processes. Virus isolation on cell cultures (Hales et al., 2008;Knowles et al., 2012), conventional and real-time RT-PCR (rRT-PCR) assays (Bracht, O'Hearn, Fabian, Barrette, & Sayed, 2016;Dall Agnol, Otonel, Leme, Alfieri, & Alfieri, 2017;Gimenez-Lirola et al., 2016) and full genome sequencing (Hales et al., 2008) have all been used to identify and investigate SVV-1 isolates. A number of accurate and sensitive rRT-PCR methods have been developed, targeting the viral polymerase 3D region , the VP1 coding region (Bracht et al., 2016), and the 5′ untranslated region (5′-UTR) (Gimenez-Lirola et al., 2016). However, diagnosis via these methods relies on the transport of samples under appropriate conditions from the point of collection to centralized laboratory settings, which may add a significant time delay and favour the spread of disease, particularly considering that modes of transmission have not yet been fully elucidated (Yoon, 2015).
The development of rapid, cost-effective diagnostic assays that can be used at the point of sample collection has been identified as a gap in preparedness for the control of SVV-1 (Yoon, 2015). Reverse transcription loop-mediated isothermal amplification (RT-LAMP) is able to rapidly amplify RNA with high specificity and efficiency under isothermal conditions at a single temperature, for example in a water bath (Notomi et al., 2000), and allows the simple, rapid and cost-effective detection of disease causing agents at the point of sample collection. A number of LAMP assays have been developed for veterinary pathogens such as foot-and-mouth disease virus (FMDV) (Dukes, King, & Alexandersen, 2006;Howson et al., 2017), African horse sickness virus (Fowler et al., 2016) and African swine fever (James et al., 2010), and some shown to be effective when deployed in field settings using simple sample preparation methods . This study describes the development of two RT-LAMP assays using lyophilized reagents, targeting the 5′-untranslated region (5′UTR) and virus protein (VP) 3-1 regions for the detection of SVV-1, and performed on a portable real-time fluorometer suitable for field use.

| Ethics
Samples used in this study (Table 1) were archival samples previously submitted to the World Reference Laboratory for FMD (WRLFMD; The Pirbright Institute, UK).

| Virus isolates
SVV-1 cell culture isolates were obtained from archival stocks held in WRLFMD repository (Table 1). For evaluation of direct detection, clinical samples were not available for this study, and therefore isolates were diluted 1/100 in negative pig epithelium tissue suspension (10% [w/v] diluted in M25 phosphate buffered saline: 35 mM Na 2 HPO 4 , 5.7 mM KH 2 PO 4 ; pH 7.6). This was to simulate an original suspension (OS) sample that would be prepared by homogenization of swine epithelium tissue either in the field, for example using the SVANO-DIP® Ag Extraction kit (Svanova), or in the laboratory, before testing.  three with GenBank accession DQ641257 as a reference: primer sets 1-3 (P1-P3), and three with the consensus sequence as a reference: primer sets 4-6 (P4-P6).
RNA dilution series was tested in triplicate using the rRT-PCR and RT-LAMP assays.
Five μl of RNA/diluted OS template was subsequently added to each reaction.

| Real-time reverse transcription PCR
Real-time reverse transcription PCR (rRT-PCR) assays were carried out as described previously , with primers and a probe targeting the conserved 3D region of SVV-1, using the Super-

| Direct detection by RT-LAMP
Twofold dilutions of cell culture isolates NC-88-23626 and LA-97-1278 already diluted 1/100 in negative pig epithelium tissue suspension (subsequently referred to as 'neat'), were prepared as template for the RT-LAMP reaction in the absence of RNA extraction, to evaluate simple sample preparation suitable for field use. Extracted RNA (as described above) from these 'neat' samples was used as a comparison.

| RT-LAMP optimization
Six primer sets targeting differing regions of the SVV-1 genome were initially investigated using extracted RNA from the seven SVV-1 isolates in Table 1

| Analytical sensitivity
A log 10 serial dilution series of RNA extracted from samples NC-88-23626 and LA-97-1278 was used to compare the analytical sensitivity of the SVV-1 RT-LAMP using P1 and P2, with the rRT-PCR (Table 2).
A higher analytical sensitivity was observed for P1, compared to P2 for both samples tested. The rRT-PCR showed higher analytical sensitivity than both primer sets of the RT-LAMP by at least one log 10 dilution (P1: LA-97-1278), and up to three log 10 dilutions (P2).

| Diagnostic sensitivity and specificity
RNA extracted from seven SVV-1 cell culture isolates diluted 1/100 in negative pig epithelium tissue suspension (Table 1)

| Evaluation of lyophilized RT-LAMP reagents
Diagnostic and analytical sensitivity, and direct detection methods were also evaluated using lyophilized reagents and compared with 'wet' reagents. For the seven SVV-1 samples available, the performance of both assays was comparable (Figure 3), and the limit of detection was equivalent when using P1; however for P2, one log 10 reduction in analytical sensitivity was observed using lyophilized reagents. For both primer sets, an increase in T p > 1 min for all dilutions was observed. When samples were added directly to lyophilized reagents, T p values were comparable at the higher dilutions (1/ 4-1/20); however when the sample was added either neat (P1 and P2) or at a 1/2 (P1) dilution, a reduction in inhibition was observed, with amplification occurring earlier, than when compared to using 'wet' reagents ( Figure 2).
T A B L E 2 Analytical sensitivity of the two SVV-1 RT-LAMP assays using either P1 or P2 and compared to the rRT-PCR

| DISCUSSION
Rapid detection of SVV-1 is important to identify the infectious agent, and to differentiate between clinically indistinguishable notifiable diseases such as FMD. An incorrect diagnosis may have severe consequences, including the type of control strategies implemented and financial implications (Anderson, 2002;Ferris, King, Reid, Shaw, & Hutchings, 2006). A number of sensitive molecular assays for the detection of SVV-1 have been previously reported (Bracht et  cation was observed for the other viruses tested. Diagnostic sensitivity was 100% for both assays when compared to a recently developed rRT-PCR , using the seven samples that were available for this study. As this is a small sample size, it is recommended that these assays be further evaluated using more samples of different types, taken from a wider geographical distribution. The analytical sensitivity of the SVV-1 RT-LAMP using primer set P1 was found to be at least one-log 10 higher than the analytical sensitivity of the SVV-1 RT-LAMP when using primer set P2, for the two samples tested, and at least one-log 10 lower than the rRT-PCR . However, for the dilutions that were not detected in all replicates by RT-LAMP, for example 10 −7 and 10 −8 , high C T values were observed when tested with the rRT-PCR (>37 C T average) suggesting a low level of virus was present. Although F I G U R E 2 Comparison of 'wet' and lyophilized reagents using direct detection by RT-LAMP with primer set 1 (a) and primer set 2 (b). Black bars represent 'wet' reagents and grey bars represent lyophilized reagents. Neat: SVV-1 sample NC-88-23626 diluted 1/100 in negative pig epithelium tissue suspension to simulate a natural original suspension sample. This 'neat' sample was then diluted ½, ¼, 1/8, 1/10, 1/16 and 1/20 in nuclease-free water (NFW) and compared to extracted RNA from the 'neat' sample as a positive control F I G U R E 3 A box-plot to compare Tp values of 'wet' and lyophilized reagents for primer sets 1 (P1) and 2 (P2) using extracted RNA from the seven SVV-1 samples. SVV-1 samples were diluted 1/ 100 in negative pig epithelium tissue suspension prior to RNA extraction the SVV-1 RT-LAMP demonstrates a slightly lower sensitivity than the rRT-PCR, samples from clinical cases are likely to contain high viral loads, and therefore RT-LAMP has the capacity to be a useful tool in the rapid diagnosis of SVV-1.
To enable the potential of these assays to be employed for rapid detection in the field, simple sample preparation methods were investigated to negate the requirement for RNA extraction, which may be difficult to perform in field conditions. For the rapid detection of FMDV from clinical samples, previous studies demonstrated that a 1/5 dilution of epithelium tissue suspension or serum, and a 1/ 10 dilution of oesophageal-pharyngeal fluid, in NFW, was sufficient to reduce the inhibitory effect observed by the addition of a neat sample to the RT-LAMP Waters et al., 2014).
This study therefore investigated whether this methodology could also be applied to the SVV-1 RT-LAMP assays, using a twofold dilution series. As clinical samples were not available for this study, isolates were diluted 1/100 in negative pig epithelium tissue suspension to simulate an original suspension (OS) sample that would be prepared by homogenization of swine epithelium tissue either in the field, for example using the SVANODIP® Ag Extraction kit (Svanova), or in the laboratory, before testing. When these samples were added directly to the RT-LAMP, no amplification was observed for P1, and there was delayed amplification when using P2, likely due to contaminants present in the sample causing reaction inhibition. Further dilution of the sample enabled amplification, with an optimum dilution of 1/16 for P1 and 1/8 for P2. Although a slight delay to amplification was evident using these dilutions when compared to using RNA extraction coupled with RT-LAMP, time from sample receipt to amplification was achieved within 12 min, highlighting the potential of these assays for rapid field diagnosis. However, further validation is required using a variety of field samples, including epithelial tissue samples, serum and vesicular swabs, to check for inhibitory effects from contaminants such as soil and faeces.
To overcome the difficulties of using temperature-sensitive 'wet' reagents in molecular assays employed in field settings, many studies have evaluated the use of thermostable lyophilized reagents that do not require the maintenance of a cold chain Goller et al., 2018;Howson et al., 2018;Semper et al., 2016). Lyophilized and 'wet' reagents demonstrated comparable performance to one another when the seven available SVV-1 samples were tested, and when diluted samples (>1/4) were added directly to the RT-LAMP. Additionally, when a sample was added either neat or diluted 1/2 in nuclease-free water, the amplification inhibition observed with 'wet' reagents was reduced when replaced with lyophilized reagents. This provides an indication that a lyophilized SVV-1 RT-LAMP could be utilized as an efficient and rapid diagnostic tool. However, it is recommended that these lyophilized assays are validated on a variety of sample types and viral loads in field settings.
In conclusion, this study describes the development of RT-LAMP assays for the rapid detection of SVV-1, suitable for employment in field settings. These assays could be performed alongside field tests for FMD (Ambagala et al., 2016;Howson et al., 2017Howson et al., , 2018Madi et al., 2012;Paixão et al., 2008), providing a rapid alternative diagnosis when FMD is negated. RT-LAMP can be performed on a portable real-time fluorometer, with results achieved in under 12 min, removing the requirement for RNA extraction. Furthermore, the use of lyophilized reagents enables rapid and simple methodology.
Deployment of these RT-LAMP assays into in situ settings could assist in disease control by enabling simple, rapid and highly sensitive detection of SVV-1.

ACKNOWLEDG EMENTS
The authors thank colleagues in the Vesicular Disease Reference