Development of a Multiplex RT- PCR Assay for Simultaneous Detection of Four Potential Zoonotic Swine Enteric RNA Viruses

Swine enteric viruses like porcine Sapovirus (SaV), porcine encephalomyocarditis virus (EMCV), porcine rotavirus A (RVA) and porcine astroviruses (AstV) are potentially zoonotic viruses or suspected of potential zoonosis. These viruses have been detected in pigs with or without clinical signs and often occur as coinfections. Despite the potential public health risks, no assay for detecting them all at once has been developed. Hence, in this study, a multiplex RT-PCR (mRT-PCR) assay was developed for the simultaneous detection of SaV, EMCV, RVA and AstV from swine fecal samples. The PCR parameters were optimized using specic primers for each target virus. The assay's sensitivity, specicity, reproducibility, and application to eld samples have been evaluated. Using a pool of plasmids containing the respective viral target fragments as a template, the developed mRT-PCR successfully detected 2.5 × 10 3 copies of each target virus. The assay's specicity was tested using six other enteric viruses as a template and did not show any cross-reactivity. A total of 280 eld samples were tested with the mRT-PCR assay. Positive rates for SaV, EMCV, RVA, and AstV were found to be 24.6% (69/280), 5% (14/280), 4.3% (12/280), and 17.5% (49/280), respectively. Compared to performing separate assays for each virus, this mRT-PCR assay is a simple, rapid, and cost-effective method for detecting mixed or single infections of SaV, EMCV, RVA, and AstV.


Introduction
Swine enteric viruses cause a major economic loss in the swine industry [1][2][3][4][5]. So far, more than sixteen porcine enteric viruses have been identi ed [6][7][8][9]. From these swine enteric viruses, due to the asymptomatic nature of infections, unless there is a confounding factor or a mixed infection of another pathogenic virus, infections from porcine Sapovirus (SaV), encephalomyocarditis virus (EMCV), Porcine rotavirus A (RVA) and porcine astrovirus (AstV) might not be noticed [10][11][12][13]. Though the economic impact of SaV, EMCV, RVA and AstV viruses on the swine industry might not be signi cant enough to attract the attention of pig producers, their potential to become zoonotic diseases warrants proper attention [12]. More importantly, even asymptomatic pigs can shed the virus into the environment or can directly transmit it to people at risk [45]. Likewise, SaV, EMCV, RVA, and AstV were found genetically closely related to human viruses and hence have a zoonotic potential and/or are suspected of zoonosis [5,11,12,[14][15][16][17][18].
To develop vaccines against the prevalent viral strains, early and accurate virus detection is needed and hence, early virus detection is an integral part of disease control and prevention programs [19]. However, coinfection, which is common in swine enteric viruses, complicates early and accurate virus detection. In the presence of coinfection, it's often di cult to know the underlining cause of disease [20,21]. This is because, in the presence of coinfection, viruses that don't usually cause apparent clinical signs during a single infection could contribute to the severity of an existing infection, further complicating disease diagnosis [11,22]. Moreover, compared to other swine enteric viruses, these potential zoonotic enteric viruses, due to their high coinfection rate, higher virus shedding, fecal-oral transmission, asymptomatic nature, and potential to transmit to humans; demands special attention for early detection and accurate identi cation.
Virus isolation, electron microscopy, serological tests, and nucleic acid detection methods are the most commonly used methods for detecting viruses [21,23,24]. Recently nucleic acid-based and serological detection methods are being used widely for diagnosing enteric swine viral infections [23][24][25]. Nucleic acid-based detection methods are preferred over serological methods for determining current infection status [26]. The existing nucleic acid-based diagnostic assays for detecting potential zoonotic swine enteric viruses depend on a single PCR. These single PCR-based detection methods consume more time and resources to detect each virus at a one-time point. Whereas, using multiplex PCR assays, nucleic acid detection methods have another added advantage for the simultaneous detection of viruses [24,25].
Multiplex reverse transcriptase PCR (mRT-PCR) based virus diagnostic assays provide both the affordability and accuracy of virus detection, enabling faster and wider eld-based applications [23,26].
In this study, in an attempt to solve challenges associated with detecting potential zoonotic swine enteric viruses, a multiplex PCR assay that could simultaneously detect SaV, EMCV, RVA and AstV from porcine fecal samples has been developed.

Nucleic acid extraction and reverse transcription
Pooled swine fecal samples were collected, and 10% (w/v) stool suspensions were mixed with PBS (phosphate-buffered saline), centrifuged at 400g for 20 minutes at 4°C, and supernatants were collected.
Total RNA was extracted from fecal supernatants using the Tiangen virus RNA extraction kit (Tiangen Biotech, Beijing) according to manufacture instructions.
Complementary DNA (cDNA) was synthesized in a 20µL reverse transcription (RT) reaction mixture containing of 2µL of Golden MLV buffer, 1µL Golden MLV enzyme, 1µL Random hexamer primers and 1µL dNTP, 0.5µL Rnase inhibitor and 1µg of total RNA and RNase free double distilled water (ddH 2 O) and then incubated at 37°C for 15 minutes and at 85°C for 5 seconds (TakaRa, Dalian, China).
Primer design and Construction of SaV, EMCV, RVA and ASTV plasmids The primers (Table 1) were designed with primer3 [27] and used to amplify the fragments of SaV, EMCV, RVA, and AstV from swine fecal samples. Each amplicon was puri ed and cloned into the pMD18-T vector (TaKaRa, Dalian, China). The plasmids were transformed into competent Escherichia coli DH5α cells. The plasmids were extracted using a mini plasmid extraction kit (Tiangen, Beijing, China) from bacterial solution cultured at 37 ℃ for 14-16 h and quanti ed by a NanoDrop spectrophotometer (Thermo Fisher, USA). The plasmid constructs were con rmed by PCR and sequencing (CometBio, Jilin, China). The plasmids were used as templates for mRT-PCR optimization. The PCR products were visualized using gel electrophoresis on a 1.5 percent agarose gel in 1 TAE buffer. Following mRT-PCR condition optimization, the amplicons were sequenced for con rmation (CometBio, Jilin, China).
Assay sensitivity, speci city, and reproducibility The sensitivity of sRT-PCR and mRT-PCR was compared using a 10-fold serially diluted standard plasmids of known DNA copy number. The DNA copy number of each standard plasmid was calculated using the formula described in [24]. To serve as a template for the sRT-PCR and mRT-PCR, 10 6 copies/µL of each cloned virus fragment were mixed into a single tube and diluted 10-fold serially at a concentration gradient of 10 0 to 10 6 copies/µL. amplicons were puri ed and sequenced to con rm the speci city of the assay.
To evaluate the intra-assay reproducibility of the developed assay, mRT-PCR reactions were performed in triplicate on each optimized parameter. Furthermore, inter-assay reproducibility was tested by running four different PCR reactions under optimized conditions with freshly prepared templates.

Detecting target viruses from eld samples using mRT-PCR
The developed mRT-PCR assay was used to test a pool of 280 porcine fecal eld samples for SaV, EMCV, RVA, and AstV, and 130 of these samples were also tested using single PCR Primers for each virus.

Phylogenetic analysis of the detected viruses
Some of the porcine fecal eld samples which were mRT-PCR positive for SaV, EMCV, RVA and AstV were reampli ed using sRT-PCR. The ampli ed and puri ed PCR products were sequenced. MEGA-X program was used to align the nucleotide sequences. The phylogenetic trees were constructed with MEGA-X software using the neighbor-joining method [28], Kimura distances, and a bootstrap sampling technique with 1000 replicates [29].

Results
Optimized conditions of the mRT-PCR Before mRT-PCR reaction optimization, sRT-PCR standardization and optimum annealing temperature  Assay sensitivity The sensitivity of sRT-PCR and mRT-PCR was tested using ten-fold serial dilutions of the SaV, EMCV, RVA and AstV plasmid constructs. It was found that the developed mRT-PCR assay could simultaneously detect up to 2.5x10 3 copies of each template, while single PCR could detect 2.5x10 1 , 2.5x10 3 , 2.5x10 2 , and 2.5x10 2 copies of SaV, EMCV, RVA and AstV, respectively (Fig. 2).

Assay speci city
The speci city of each primer pairs was determined using sRT-PCR and mRT-PCR (

Assay reproducibility
Four different mRT-PCR reactions were performed at different times to assess inter-assay reproducibility using the same reaction conditions (Fig. 3). A freshly extracted plasmid was used as a template for each PCR reaction. Each of the four mRT-PCR reactions showed a similar result, indicating the reproducibility of the assay. Besides inter-assay reproducibility, intra-assay reproducibility was checked by carrying out triplicate mRT-PCR reactions (data not shown). The developed mRT-PCR assay's intra-assay and interassay reproducibility tests showed that it could be used to detect potentially zoonotic swine enteric viruses. respectively. The single infection positive rate for RVA was 4.3% (12/280) and the coinfection rate of RVA and AstV was 2.9% (8/280). Moreover, a single infection positive rate of AstV was 17.5% (49/280).

Detection of eld samples
For checking the reliability of the mRT-PCR assay, in addition to testing by mRT-PCR assay, 130 samples were selected and further tested using sRT-PCR for SaV, EMCV, RVA and AstV. Except for SaV (Table 2), samples positive for each virus using a single PCR test also showed similar positive results when tested by mRT-PCR. Some of the mRT-PCR positive samples were also checked by sequencing. The concordance of sequence results was then checked by subjecting to NCBI nucleotide blast and found no false-positive results. * The results were similar except for two samples that were positive for SaV by sRT-PCR but negative by mRT-PCR.
SaV infection is the most common of the four viruses tested, while RVA infection being the least common. Test results of mRT-PCR showed 98.5% agreement with the test results of sRT-PCR. This result indicated that, like sRT-PCR, the developed mRT-PCR is sensitive enough to detect eld samples.

Phylogenetic analysis
For each of the four viruses, phylogenetic trees were constructed using the neighbor-joining method based on the ampli ed fragments of the target region of some selected sequences from GenBank and nucleotide sequences obtained in this study. When constructing the phylogenetic trees, geographic location, host type, and virus genotype were all considered. The viral variants identi ed in this study are indicated by red circles (Fig.4), and all trees are drawn to scale, with branch lengths equal to the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were calculated using the Maximum Composite Likelihood method [30] and are expressed in base substitutions per site.
Within the Caliciviridae family, based on their complete VP1 nucleotide sequences, sapoviruses could be classi ed into 15 genogroups, eight of which have been detected in swine (GIII, GV-GXI) [31,32] but only the GI, GII, GIV, and GV genogroups are known to infect humans [31,33,34]. To investigate the genetic diversity of several of the sequences identi ed in this study, we performed a phylogenetic analysis using 32 reference nucleotide sequences from GenBank with and 224 positions from Sapovirus ORF1 region.
The SaV strains identi ed in this study named (Sapo 1, Sapo 2, Sapo3 and Sapo 5) clustered within the GIII genogroup and were found to be closely related to other GIII SaV strains detected in China (MW285642) and Mexico (MH490911). Whereas Sapo 4, which is also classi ed as a GIII genogroup member, formed a separate clade (Fig. 4A).
EMCV is a Picornaviridae virus that infects many animals, including pigs, mice, primates, and humans [35]. The EMCV strains isolated from various animals, for example, pigs and rats, demonstrated a high degree of homology [36]. Thus, to investigate the genetic variability of the detected EMCV variants, we constructed a phylogenetic tree for EMCV using 22 reference nucleotide sequences from GenBank and 3 EMCV detected in this study named (EMCV1, EMCV2, and EMCV3) with 211 positions per sequence.
According to the partial sequence of the 3D region, EMCV 3 detected in this study clustered with other EMCV strains from Japan (LC508268), whereas EMCV 1 and EMCV 2 clustered separately and more closely related to strains from China (Fig. 4B).
The RVA phylogenetic tree was constructed using a total of 33 RVA nucleotide sequences previously reported from various geographic locations and hosts, as well as strains detected in this study. The RVA variants identi ed in this study clustered into two distinct clades, with R1 and R2 belonging to one clade and R3 to another clade. However, all RVA variants identi ed in this study are more closely related to RVA previously described from China (MT874988, FJ617209 and KT82077), implying that they may have originated from the common ancestor. R1 and R2 clustered with MT874988 and FJ617209, previously described from China, whereas R3 clustered with KT820771, another Chinese RVA strain (Fig. 4C).
For AstV, phylogenetic analysis was performed using 25 nucleotides published sequences with 379 positions within the ORF1ab region, and seven AstV variants identi ed in this study named (Ast 1-7). All AstV variants detected during this study clustered under AstV4. Furthermore, Ast 1, 2, 3, 5 and Ast 7 clustered closely with AstV from China (MK460231) and Kenya (MT451918), whereas Ast 4 and Ast 6 clustered independently and close to strain from the USA (JX556692) (Fig. 4D).

Discussion
Despite the presence of a plethora of different types of diagnostic assays, nucleic acid-based assays offer an added advantage of high speci city and sensitivity with the possibility of a lower detection limit [37]. Recently, mRT-PCR is being used as one of the most important diagnostic methods for rapidly and simultaneously detecting viral pathogens.  [43,44] swine enteric viruses have been developed. Due to the asymptomatic nature of SaV, EMCV, RVA, and AstV infections, these viruses have received insu cient attention, and no single diagnostic assay capable of simultaneously detecting these viruses from porcine fecal samples has been developed.
A mRT-PCR assay capable of simultaneously detecting SaV, EMCV, RVA, and AstV RNA from porcine fecal samples was developed in this study. Our assay and other previously developed assays [23,26]for other viruses suggest that sRT-PCR may be more speci c than mRT-PCR, but the improvement in turnaround time and cost-effectiveness would compensate for this minor reduction in sensitivity. Yet, compared to sRT-PCR, mRT-PCR is more economical and rapid. The sensitivity of the developed mRT-PCR assay using plasmids containing the speci c viral target fragments was 2.5 X 10 3 (Table 2) to the test results of sRT-PCR. Besides sensitivity, test results of speci city and reproducibility of this assay indicate that similar to the sRT-PCR, the developed mRT-PCR could be employed to detect SaV, EMCV, RVA, and AstV from porcine fecal samples. Similarly, previous reports [25,38] and others suggested that mRT-PCR could be a highly sensitive and speci c assay that could be used for rapid detection of enteric swine viruses.
A phylogenetic analysis was also made to explore further the epidemiologic characteristics of the detected viruses. The phylogenetic analysis revealed that the SaV detected in this study belongs to the GIII genogroup. Similar to this study, previous research [50,51] indicated that from the eight SaV genogroups detected in swine [31,32], the GIII genogroup is the most prevalent in China. Phylogenetic analysis of the partial sequence of the 3D region of EMCV, similar to previous studies from China [52,53], revealed that the EMCV variants detected in this study clustered into two groups. Yuan, et al. [53] assigned to the notion that coinfection of these viruses is common in swine. Through recombination and mutations, the coinfection of viruses accelerates the evolution of coinfected viruses [59,60], allowing more virulent virus strains to emerge. As a result, the developed mRT-PCR assay could play a critical role in controlling and preventing SaV, EMCV, RVA, and AstV through early and accurate detection.

Conclusion
The assay's sensitivity, speci city, and repeatability demonstrated that SaV, EMCV, RVA, and AstV could be detected simultaneously in swine feces. Potential zoonotic swine enteric viruses such as SaV, EMCV, RVA and AstV are prevalent in China. These viruses frequently exist in pigs as a coinfection, necessitating the employment of specialized diagnostic techniques for quick diagnosis. The current mRT-PCR may help to reduce the cost and time associated with sample processing and testing during large-scale eld sample screening for these viruses. Thus, the assay may help control and prevent potential zoonotic swine enteric viruses such as SaV, EMCV, RVA, and AstV by detecting them early and accurately.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.