MALDI-TOF Nucleic Acid Mass Spectrometry for Simultaneously Detection of Fourteen Porcine Viruses and its Application

Background: Porcine respiratory and digestive diseases pose signi�cant challenges in modern pig farming, often arising from mixed infections involving various pathogens. Current methods for detecting viral porcine pathogens have notable limitations in simultaneously identifying multiple pathogens. To address this issue, our study introduces a novel methodology that combines single-base extension PCR with matrix-assisted laser desorption/ionization time-of-ight nucleic acid mass spectrometry (MALDI-TOF NAMS). Results: Our approach accurately simultaneously identi�ed 14 critical porcine viruses, including porcine circovirus types 1 to 3, porcine bocaviruses groups 1 to 3, African swine fever virus, pseudorabies virus, porcine parvovirus, torque teno sus virus, swine in�uenza virus, porcine reproductive and respiratory syndrome virus, classical swine fever virus, and foot-and-mouth disease virus. The low limit of detection for multiplex identi�cation ranges from 13.54 to 1.59 copies/ μ L. Inter-and intra-assay stability was found to be ≥ 98.3%. In a comprehensive analysis of 108 samples, the assay exhibited an overall compliance with qPCR results of 97.88%. Conclusions: The developed MALDI-TOF NAMS assay exhibits high sensitivity, speci�city, and reliability in detecting and distinguishing a wide spectrum of porcine viruses in complex matrix samples. This underscores its potential as an e�cient diagnostic tool for disease surveillance and control in the pig industry.

Matrix-assisted laser desorption/ionization-time-of-ight mass spectrometry (MALDI-TOF MS) is an innovative technique used to detect nucleic acids, peptides, and proteins (19)(20)(21)(22)(23).The MALDI-TOF MS-based primer extension assay typically involves combining multiplex PCR with MALDI-TOF MS (19,24), this nucleic acid mass spectrometry (NAMS) uses speci c mass probe for single-nucleotide polymorphisms (SNP) sites.These mass probes selectively bind to the target in the single-base extension reaction.Consequently, variations at the SNP sites result in single-base extension products (SEP) of varying quality.The MALDI-TOF NAMS is adept at distinguishing these quality signals, which differ by tens of Daltons.MALDI-TOF NAMS has demonstrated its effectiveness in the clinical eld for detecting pathogenic microorganisms, offering novel solutions for discerning mixed infection samples and pathogen typing (25)(26)(27)(28).Since the emergence of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic (29), the MALDI-TOF NAMS was swiftly applied to these emerging pathogens.MALDI-TOF NAMS offers higher throughput than PCR, it has been used to detect and differentiate seven human coronaviruses (HCoV), with a limit of detection (LOD) ranging from 1 to 5 copies per reaction (30).Another study has used this method simultaneously detected 27 human respiratory syndrome-associated viruses (HRSSV) (31).
In this study, we developed a high-throughput MALDI-TOF NAMS method enabling the simultaneous identi cation of fourteen porcine viruses.These viruses including ASFV, CSFV, FMDV, PBoV-G1, PBoV-G2, PBoV-G3, PCV1, PCV2, PCV3, PRRSV, PRV, PPV, SIV, and TTSuV.The developed assay not only facilitates the monitoring of co-infections in pigs but also provides an effective means for surveilling and tracking the presence of these pathogens in porcine feed and pork products.

Simplex MALDI-TOF NAMS assay
Before multiplexing, all optimized primers, probes, and reagents were tested in the simplex MALDI-TOF NAMS system.The recombinant standard plasmids in Table 1 were used as PCR templates, and ddH 2 O was used as blank control.
After PCR and single-base extension, the products were analyzed by the MALDI-TOF MS system.As shown in Fig. 1, the mass spectrograms exhibited a sole peak for each standard plasmid at the predicted single-base extended products (SEP) mass (Table 2), without non-speci c stray peaks.The blank controls also displayed only a single peak at the un-extended probe (UEP) mass, it is easy to distinguish the mass spectrometry peaks of the corresponding SEPs from the UEPs.

Multiplex MALDI-TOF NAMS assay
The mixed positive plasmids (10 4 copies/µL) or ddH 2 O were subjected to ampli cation using the optimized multiplex primers.As shown in Fig. 2, all target SEPs (black peaks) were detected with no overlap with UEP (red peaks).Notably, PCV1 and PCV2 demonstrated distinct single peaks at 6103.9 and 6023.9Da, respectively, as valid a distinction can be made as in the case of the individual primer reaction (Fig. 1F & G).

Speci city of the MALDI-TOF NAMS
To verify the speci city of our method, the genomic DNA/cDNA of fourteen target viruses and four non-target controls including JEV, PDCoV, PEDV, and TGEV were used as templates individually.The mixed recombinant standard plasmids (10 4 copies/µL) were used as positive control and the ddH 2 O was added as negative control.PCR and single-base extension were performed using optimized primers and mass probes, then the products were analyzed by MALDI-TOF MS.As shown in Fig. 3, the mixed plasmid detected 14 target viruses as positive, while JEV, PDCoV, PEDV, TGEV, and ddH 2 O were negative.MALDI-TOF NAMS enables speci c detection of target genes from extracted nucleic acids in whole virus samples, without any cross-reactivity to the nucleic acid of JEV, PDCoV, PEDV, and TGEV.

Repeatability of the MALDI-TOF NAMS
A serial dilution of the mixed standard plasmids (10 6 , 10 4 , and 10 2 copies/µL) was assayed by this method, and 20 replicates were performed for each concentration.As shown in Table 4, among the three separate batches of duplicate experiments, only a single result did not detect PCV3 at 10 2 copies/µL, resulting in the positive rate of PCV3 is 98.3% (59/60), the results for the other targets were 100% positive (20/20, 60/60) at all three concentrations.This indicates that the method has high batch-to-batch reproducibility (Table 4).

MALDI-TOF NAMS enables the detection of single and multiple viruses in clinical samples of different matrices
To verify the application of MALDI-TOF NAMS to detect single and multiple viruses in clinical samples, we selected eight qPCR-veri ed positive samples from various types of samples (diseased tissue, feed, and pork), two samples was positive for a single virus (Fig. 4A & 4G), and six samples were positive for multiple viruses (Fig. 4B, 4C, 4D, 4E, 4F & 4H).These samples were tested by the MALDI-TOF NAMS, and the positive results were consistent with the previous qPCR assay.In the mass spectrograms, speci c sizes of the extended probes (red arrows) were detected in different matrix samples (Fig. 4), and they can be clearly distinguished from the UEP peaks.

Comparison between MALDI-TOF NAMS and qPCR
Three types of clinical samples were collected, 42 tissue samples from suspected infected pigs, 42 feed samples from pork sources, and 24 fresh pork samples were purchased from random markets.The DNA or cDNA templates of these samples were prepared by DNA/RNA extraction kit and reverse transcription kit.A total of 108 samples were tested by the multiplex MALDI-TOF NAMS developed in this study, and the qPCR results were used as the reference data.The results of the two methods are shown in Fig. 5. Based on the results of qPCR, the coincidence rate of NAMS results was analyzed.It was found that the MALDI-TOF NAMS results of seven viruses (ASFV, CSFV, FMDV, PBoV-G3, PRRSV, PRV, and SIV) were 100% consistent with the results of qPCR (Table 5).The MALDI-TOF NAMS detected more positive results for PBoV-G1 and PBoV-G2 in addition to the positive results of qPCR.The positive coincidence rates of PCV1 to 3, PPV, and TTSuV ranged from 88.6-97.8%.In total, 108 samples of different matrices were tested and found to have a high total coincidence rate (97.9%) between MALDI-TOF NAMS and qPCR, these results indicate the potential of MALDI-TOF NAMS to replace qPCR in clinic applications.

Discussion
According to viral metagenomics investigations, even phenotypically healthy pigs can harbor endemic strains (17,32), indicating widespread multiviral coinfection in pigs.Many of these viruses identi ed belonged to the linear and circular DNA virus families, including PCV2, PPV, PBoVs, and TTSuv (10,17,33).This is consistent with the ndings for pork products in this investigation (Table.4).
Moreover, in cases of syndromic-triggering viruses, it's important to consider the possibility of respiratory symptoms arising from a complex viral disease, wherein multiple pathogens may be involved.For instance, consider PCV2, which is the primary causative agent of porcine circovirus-associated diseases (PCVAD) (7), when present as a sole infection, its pathogenicity is relatively weak.However, some studies have shown that PCV2 can inhibit type I interferon (IFN-I) induction to promote other DNA virus infections (34,35).In cases of PCV2 infection, where antigenpresenting cells are infected, this can lead to a suppressed early antiviral response, potentially causing a signi cant impact on the host's ability to generate a speci c immune response (36).Experiments involving piglets showed that co-infections with PCV2 and viruses like PPV or PRRSV induced PMWS, a condition distinct from the outcome of a sole PCV2 infection.Co-infections of PCV2 with PRRSV, PPV, and PBoVs ampli ed the clinical signs of PCVAD (5,8,13).In addition to this, coinfections involving PCV2 have been observed with viruses such as CSFV, TTSuV, and other enteroviruses (10,12,15).Similarly, PRRSV also inhibits IFNs signaling by blocking STAT1/STAT2 nuclear translocation (37).This situation leads to a higher prevalence of co-infections, further underscoring the complex dynamics of virus-host interactions.
In our study, analysis of tissues or blood samples indicated that 82.93% (34/41) of the samples exhibited multiple pathogen detection, underscoring the severity of mixed infections in diseased pigs.Among PCV2 co-infections, the detection rate was 73.53% (25/34), involving PCV1, PCV3, PBoVs, PPV, PRRSV, and TTSuV.The in uence of PCV2 was slightly more pronounced in diseased material samples compared to pork, though not statistically signi cant.This may be because not all respiratory symptomatic pigs from diseased material sources had PMWS.Notably, besides PCV2, the pathogen with the highest incidence detected in this investigation was TTSuV.It was found in 52.38% (22/48) of the diseased material samples, 40.48% (17/48) of the feed samples, and 37.50% (9/24) of the pork samples.TTSuV exhibited a relatively elevated prevalence among healthy pig populations, a trend consistent with ndings from various other studies (38, 39).However, as of now, a de nitive causal relationship between TTSuV infection and speci c diseases in pigs remains unestablished (40)(41)(42).Despite its prevalent presence, further research is required to elucidate any potential pathogenic effects associated with TTSuV infection in pigs.Moreover, feed samples are susceptible to exogenous pathogen contamination.The mixed infection rate in our study reached 82.93%.However, it's important to note that this rate re ects the presence of diverse pathogen contamination in the samples and doesn't necessarily indicate a high rate of mixed infection in the originating diseased pigs.Contaminated feed samples may carry pathogens from various sources, and the slaughter, processing, and sales processes could further propagate these pathogens (18).This underscores the signi cance of accurately detecting multiple pathogens and understanding their role in swine herds for effective disease prevention and control measures.
Currently, a diverse array of virus detection methods are employed for animal quarantine purposes.In addition to traditional methods based on immunology, molecular biology-based methods like polymerase chain reaction (PCR), quantitative PCR, digital PCR, and others are utilized.Advancements in technology have also introduced secondgeneration sequencing (NGS), liquid-phase microarray, MALDI-TOF MS, and similar technologies into the realm of virus detection.Traditional methods can exhibit subjectivity in result assessment and may struggle to meet the demand for simultaneous detection of multiple pathogens within a short timeframe (43).PCR-based detection techniques, such as uorescence-based quantitative PCR (qPCR), and digital PCR (dPCR) (44) offer high sensitivity and ease of operation, making them widely used in pathogen detection.However, their application in scenarios necessitating the detection of multiple pathogens is often hindered by challenges such as primer cross-reactivity within the system, or limitations in the uorescence channels of the used uorescent dyes and probes.This limitation makes the simultaneous detection of six or more targets a complex endeavor.NGS technology permits highthroughput detection of unknown pathogens in samples, but library preparation and interpretation of results is complex.Third-generation sequencing (TGS) offers long reads but lacks deep sequencing depth and accuracy compared to NGS.Both sequencing methods are costly and may not be suitable for scenarios involving large sample sizes, such as disease surveillance, epidemiological monitoring, and entry-exit inspections.In contrast, MALDI-TOF NAMS presents speci c advantages when it comes to the concurrent detection of multiple pathogens within substantial sample sets.
MALDI-TOF NAMS methods offer high accuracy and speci city, making them suitable for medium to high throughput applications.The HAND (homo-tag assisted non-dimer) strategy ( 45) is used to design multiple primers.This tag serves to distinguish the mass range and effectively minimizes the occurrence of primer dimers during the PCR ampli cation process.Additionally, this approach enhances the ampli cation e ciency through a two-step ampli cation process, which is then followed by detection using mass spectrometry.The MALDI-TOF NAMS method identi es speci c nucleic acid fragments by leveraging differences in molecular weights.This enables the analysis of 30 to 50 or even more targets, with the detection results being constrained only by the LOD of system.This technology's strengths offer promising prospects for overcoming the limitations posed by other methods.While MALDI-TOF NAMS can detect various pathogen types, it's limited to known pathogens, and detection of unknown viruses requires non-speci c methods like NGS.

Conclusion
In this study, we developed and validated a MALDI-TOF NAMS assay capable of detecting 14 prevalent porcine respiratory and gastrointestinal viruses.Subsequently, we utilized this method to investigate the prevalence and coinfection rates of these pathogens in samples from diseased pigs.Additionally, we examined pork and swine-derived feeds to assess potential contamination with the exogenous viruses mentioned.To verify our ndings, we also conducted a comparative analysis with qPCR results.The scalability of this approach enables us to expand the range of viruses in future studies to include additional relevant epidemic pathogens as needed.The cost-effectiveness of MALDI-TOF NAMS complements qPCR and NGS methods, making it useful for quarantine, product testing, and pathogen investigation.In summary, the nucleic acid mass spectrometry method offers a powerful tool for accurate and sensitive pathogen detection, contributing to disease prevention and control under different scenarios.

Preparation of viral nucleic acid templates
The extraction and puri cation of nucleic acids from the strains were extracted by MiniBEST Viral RNA/DNA Extraction Kit (TaKaRa, Dalian, China).For RNA viruses, the eluted products were subjected to reverse transcribed into cDNA using random hexamers and PrimeScript™ IV 1st strand cDNA Synthesis Mix (TaKaRa, Dalian, China).The DNA or cDNA templates were stored at -20°C.

Construction of standard plasmids
We chose the highly conserved region of each virus gene, and fourteen reference sequences were synthesized and cloned separately into the backbone vector pUC-57 by Sangon Biotech (Shanghai, China), the standard recombinant plasmids used in this study are listed in Table 1.
The concentrations of plasmids were quanti ed, and the copy number was validated utilizing Qubit 4 (Thermo Fisher, Waltham, USA).The plasmids for 14 targets were diluted into 9 gradients, spanning 4.68×10 8 to 0.61×10 0 copies/µL in a 10-fold gradient progression.Subsequently, plasmids of equivalent concentration levels were amalgamated in equal proportions, resulting in the creation of mixed plasmids for each gradient (10 8 to 10 0 copies/µL).The mixed plasmids were used to evaluate the performance of the subsequent systems.

Primers and probes
To design the speci c PCR primer and extended mass probes for each target virus, we aligned reference genes of viruses in Table 1 using ClustalW (46) and MEGA-X by Mega (Auckland, New Zealand).Speci c PCR primers and probes were designed using Primer 3 Plus (47) and MassARRAYAssay® Design Suite (Agena Bioscience, San Diego, USA).A 10-base tag (ACGTTGGGATG, bold in Table 2) was appended to the 5' end of each PCR primer to avert multiple primers causing potential peak interference within mass spectra.The Primer-BLAST (48) and MFEprimer 3.1 (49) were employed to assess the speci city of primers and probes.The sequences of the primers and probes used in this study are listed in Table 2.All primers and probes were synthesized and puri ed by Sangon Biotech (Shanghai, China).

Method establishment and optimization
The recombinant plasmids in Table 1 were used for method establishment.After optimizing the concentrations of ampli cation primers and extension probes in the 14-target system, the nal optimal concentrations for each primer and UEP were listed in Table 3.For the PCR, 2 µL of recombinant plasmid (10 4 copies/µL) was mixed with 1.2 µL of PCR mix (Agena, San Diego, USA), 1 µL of PCR primer (0.5 µmol/L each), and 0.8 µL ddH 2 O for each target.PCR was performed at 95 ℃ for 2 min, followed by 35 cycles of 95 ℃ for 15 s, 60 ℃ for 30s, and 72 ℃ for 1 min, with a nal extension at 72 ℃ for 5 min.Subsequently, the ampli cation products underwent a dephosphorylation reaction by adding 2 µL of shrimp alkaline phosphatase (SAP) and reaction buffer (Agena, San Diego, USA).The reaction was incubated at 37 ℃ for 40 min, followed by enzyme inactivation at 85 ℃ for 5 min.For extension reaction, the products were mixed with 1 µL UEP mix (5 µmol/L each), 0.2 µL Inplex buffer, 0.04 µL Inplex enzyme, 0.

Sensitivity and speci city of the MALDI-TOF NAMS
To assess the sensitivity of the method, the mixed standard plasmids were diluted 2-fold to 10 2 to 0.5 copies/µL.Each concentration gradient was subjected to detection using optimized primers and probes with 10 times replicates, the lowest concentration that consistently yielded a positive result was established as the low LOD of multiplex MALDI-TOF MS assay for the virus.
The speci city was evaluated using genomic nucleic acids obtained from cultures of target viruses, vaccines, or pseudoviruses.Additionally, the nucleic acids extracted from various non-target porcine viruses such as PEDV, TGEV, JEV, and PDCoV were used as the negative control.The mixed plasmid template (10 4 copies/µL) was employed as a positive control, while ddH 2 O served as a blank control.

Reproducibility of the MALDI-TOF NAMS
Furthermore, the intra-batch reproducibility of the MALDI-TOF MS method for each virus detection result was gauged by conducting 20 replicate experiments for each group of dilutions, employing mixed plasmids with high, medium, and low concentrations (10 6 , 10 4 , and 10 2 copies/µL) as templates.These conditions were replicated across three separate experiments conducted at different times.The results from the three independent experiments at each dilution were recorded to ascertain the batch-to-batch reproducibility of the method.

Evaluation of the MALDI-TOF NAMS on clinical samples
To further explore the method's capacity to simultaneously detect multiple pathogens in actual contaminated samples, a variety of sample types were assessed, including suspected diseased materials, feed, and commercially available pork samples.The overall sample count totaled 108, encompassing the following categories: tissue or blood samples from pigs with respiratory diseases (n = 42); feed samples containing pork ingredients (n = 42), and pork samples (n = 24).All the samples were sourced from the Hangzhou Customs Technology Center.
For different types of samples, nucleic acid extraction and puri cation using the MagNA Pure 24 Instrument in combination with the MagNA Pure 24 Total NA Isolation Kit (Roche, Basel, Switzerland).Following this, the total nucleic acid content of the resulting products was subjected to reverse transcription using the PrimeScript™ IV 1st strand cDNA Synthesis Mix (TaKaRa, Dalian, China).Subsequent PCR, SAP digestion, and UEP extension were carried out using the previously optimized conditions.The MALDI-TOF NAMS assay of multiple viral standard plasmids.
The fourteen viral standard plasmids mixed in equal parts with a nal concentration of 10 4 copies/µL, these mixed positive plasmids were used as template, H 2 O was served as blank control, then optimized multiplexed primers were added for MALDI-TOF NAMS detection.In the mass spectrograms, the black line represents the mass spectrometry result of mixed plasmids, and the red line represents the mass spectrometry result of blank control.

2 µL
Termination mix (Agena, San Diego, USA), and 1.56 µL ddH 2 O.The extension reaction program cycled 40 times at 95 ℃ for 30 s, 95 ℃ for 5 s, 52 ℃ for 5 s, and 80 ℃ for 5 s, with a nal extension at 72 ℃ for 3 s.were transferred to a 384-well plate and adjusted to a volume of 25 µL with ddH 2 O.After centrifugation at 8000 rpm for 2 s, the 384-well plate and SpectroCHIP Array for the inert matrix were placed into the DP-TOF (Zhejiang Digena Diagnostic Technology, Hangzhou, China).The mass spectrometry ight parameters were con gured post-sample setup and microchip plate preparation.Settings encompassed a value of 10 µL for the inlet volume of the puri cation resin, 30 laser pulses, and a maximum of 9 data acquisition rounds per sample.Subsequently, MassARRAY® Typer by Agena was employed for the processing and analysis of mass spectrometry signals.

Figure 4 Ability
Figure 4

Table 1
Plasmids used in this study

Table 4
Comparison of the intra-and inter-assay variations 2: Concentration of mixed plasmid (copies/µL)